Free calibration soft (non-excel) [Text View] - Page 6

lovingdvd

01-08-07, 11:56 AM

> If I was to stop at this point and not measure the grayscale,
> then is this essentially emulating how Colorfacts does the training?

Yes, except that you have to either do have to do the gray scale measure if you want/need white. (you can measure white with the continous reading, and then type it in the 100 IRE column, but that is fastidious). The grayScale measure recomendation is only to get white.

That is exactly what I did. As mentioned, I did not want to do a full grayscale because EyeOne is inaccurate in low IREs and I didn't want to be training the Spyder meter based on that bad data.

However I was wondering if it would be a benefit to measure other IREs and hand edit them into the table. For instance, let's say I did what you said not just for 100 IRE but for say 30-100 IRE. Is there any benefit of doing this? Or is this the same as just doing it for 100 IRE as far as HCFR is concerned and how it uses this information for training/offsets?

In other words, I'm wondering if I get better training to let HCFR use as a reference 30-100 IRE instead of just 100 IRE. ?

If so, I was thinking I would great the reference set by doing a full grayscale, then hand edit the results and just delete 0-20 IRE results from the table. Then I'd have the results from 30-100 IRE.

Please let me know what you think. Thanks!

--Patrice[/QUOTE]

lovingdvd

01-08-07, 11:57 AM

Laric and team - I spend several hours with HCFR last night and let me just say what a great program it is. I got very accurate results and it is nice to use and offers a more intuitive way to calibrate over CF.

First I followed Laric's instructions for training my SpyderTV meter to the EyeOne. I then measured grayscale and primaries/secondaries and recorded all results. Then I repeated using Colorfacts and compared all results. I use external/DVD patterns via GetGray calibration disc.

As a side note, I set HCFR to take 5000ms to do its readings and checked the option to average many samples at lower IREs. In CF I use 8000ms, but HCFR would not let me use > 5000. I do not mind the extra reading time, and this actually saves me time in the long run by giving me more accurate measurements to work with from the start.

A few observations:

- HCFR seemed to do a much better job reading lower light levels with the EyeOne than CF.

- all readings from 0-100 IRE taken with HCFR were within 100-150K of the readings taken with Colorfacts (with HCFR readings consistently higher by that amount). This was excellent! I do not know which is system was more accurate, but it doesn't really matter.

Issues:

About the only issue I saw is that it seems to take HCFR about 7 seconds to prepare for a measurement when working with the Spyder meter. I did not notice this with the EyeOne at all.

For instance let's say I tell it to take a single reading. My screen goes black for 7 seconds before it then puts up a message box that says "put up pattern X and hit OK". Then the reading takes x seconds which I understand. However it is this 7 second delay that is slowing things down. Do you know what is causing this?

Same thing if I tell it to measure primaries. It goes to sleep for about 7 seconds, then prompts me for the red pattern. Then as it moves from pattern to pattern (blue green etc) it works just fine. Its the initial loading part that takes the additional time.

It almost seems as if HCFR is apply the meter reading ms time and taking that delay prior to even prompting to hit OK to beging the measurement... ? I am suspicious of that as the culprit because I left the default ms timing when using the EyeOne and I did not see this issue with that meter.

Feature requests:

Please consider the following feature requests / improvements for your future versions. These are in no particular order of priority.

a) Is there a way to plot the dE on the Color Temperature results chart? If not please add this as an option like is available on the continuous/free measurement graph. dE is the most important thing for me when calibrating. Although I can see it with single or continuous readings, I couldn't find any way to see these results after performing a grayscale reading without exporting the data to Excel.

b) I like the continuous/free reading graph very much. It would be great if this graph could also chart what is being done in the other measurements that are not the "free" ones.

c) I use single readings a LOT. I would much prefer it if HCFR would not make me say OK to the "prepare your pattern and hit OK box" with each single reading. This really slows me down. Ideally HCFR would just assume that I know what I am doing and that at the time I hit the option to take a single measurement that I already have the pattern displayed. This is the way CF works. Just press F5 and it takes a single reading and finishes.

d) Please add an option so HCFR will play a tone/sound when a reading begins and when it ends. CF does this and it is very helpful. Although HCFR blanks the screen during a measurement (VERY NICE BTW!), I still need to keep my laptop shut to keep its stray light out (a blank screen is still bright with my LCD laptop). Without this tone I don't know when the measurement is completed and have to "peek" under the laptop cover.

e) Please replace the plotted "bubbles" on the CIE chart with something much more precise like a cross hair. Even with the zoom function the bubbles obfuscate the target points and its hard to judge where the exact center of the plotted bubble really is.

f) Please add a way to have HCFR show the dE of a measurement based on any specified xy instead of just the target. For example when I am dialing in green I want the dE of my reading as it pertains to 0.6,0.3, not D65.

g) Please add a way to display the measured luminance larger. I use this when aiming the meter. With CF I make the readout very large on the screen. Then when I am 8 feet away aiming the probe I can watch the reading and figure out where my alignment is getting the brightest reading. The way I do this know is take continuous readings and the View Info box which is like 10pt font and impossible to read from a distance.

h) I think you could streamline things quite a bit by eliminating the separate "secondaries" measure, and just add an option to the primaries to also measure the secondaries and combine all this into one measure. Personally I always measure primaries and secondaries.

i) Please add a Luminance calibrator function for calibrating the correct Y value in xyY measurements when measuring primaries/secondaries. CF does not have this and it would be a great addition. Let me explain further.

As you know when setting primaries/secondaries not only do you have to get xy right but also Y. Currently I have to do this manually. First I measure white and note its Y value. Then I know by standards that red Y is x% of white Y. So I then measure Y for red and have to adjust its "Lightness" until it is at the right % of white. I do this by using only raw measurements and keeping track in my head or with pen/paper. Instead it would be great if you had a table or something that would update with each single reading that showed the % of Y for the reading vs. its target of % of white by spec!

j) Please add a hot key that would cause a single reading to be taken, regardless of what the "continuous reading option" was set at in the Preferences. Sometimes I need continuous readings, other times single readings. It is very awkward to have to manually switch the preferences back and forth to control whether I am taking a continuous vs single reading.

k) Regarding on/off and ANSI CR measurements - please add an option to select which one of these measurements you are taking. For instance I often want to just measure on/off but not ANSI at all. I work around this by just faking the ansi cr patterns when prompted by HCFR but it would be nice if I could tell it I'm only interested in doing one or the other.

I hope you find these suggestions helpful. Although I've been using CF for years I am a HCFR newbie so if there are ways to use the software to accomplish what I've mentioned above please let me know.

Thanks to the HCFR team for this truly excellent program and all the hard work that goes into it!!

HT-Naimee

01-08-07, 12:09 PM

Puh, that's quite a list of requests :)

So do you use both EyeOne AND Spyder2? Or have you replaced your Spyder2 with the EyeOne because your measurements with the Spyder2 were no good?

HDholic

01-08-07, 12:22 PM

Don't know :)
I just read so much about the Spyder2. Do you have any links to those two products for me to check them out? And what are the differences?
Here's info about Display2 (http://www.xritephoto.com/product/eye1display.asp) . The DTP94 (created by X-rite) is being phased out and I think replaced with the Display2 after Eye-One merged with X-rite. You could find some info about DTP94 by searching the threads. These days it's somewhat hard to find but it's a great probe.

Differences between them and the S2 is that they are MUCH better at reading from 0IRE to 30IRE, and somewhat more accurate overall by what users have seen. Readings on the low end with the S2 are useless.

lovingdvd

01-08-07, 12:26 PM

Yes, I use both the EyeOne and Spyder meters together (Spyder trained to the EyeOne). This gives me the accuracy of the EyeOne but with the Spyder's ability to read very low light levels directly from the pj lens (I can read down to 5 IRE with excellent and consistent results). Without the EyeOne the SpdyerTV does not provide the accuracy, and without the SpyderTV the EyeOne does not provide accurate low level readings. Together it is the best of both worlds.

As an example, at 100 IRE I purposely push my CT to 7000K with a dE of about 5-6. It is definitely noticeable when you switch from a 90 to 100 IRE pattern that the color has shifted. According to the Spyder meter with no training it says I am at 6700 with a dE of 3 which is definitely wrong. But with the training it says I am at 7100K with a dE of 6 which is correct.

laric

01-08-07, 01:10 PM

Thanks for your encouragement and feedback... Let us go a bit in depth on you long post before answering ;)

--Patrice

HDholic

01-08-07, 01:20 PM

a) Is there a way to plot the dE on the Color Temperature results chart? If not please add this as an option like is available on the continuous/free measurement graph. dE is the most important thing for me when calibrating. Although I can see it with single or continuous readings, I couldn't find any way to see these results after performing a grayscale reading without exporting the data to Excel.
dE is plotted on the RGB chart, why would you need it on the color temp chart as well?

lovingdvd

01-08-07, 01:32 PM

dE is plotted on the RGB chart, why would you need it on the color temp chart as well?

From what I recall, dE was not plotted on the RGB chart. If I right-clicked it, it had an option that said something like "dE curve" and when I added that to the chart it opened up a new panel underneath the RGB plot which just had a single point/dot. Instead what I am looking for is dE at each IRE just like what gets shown in the free measurements graph. This is all from memory but I'll double check this later today.

HDholic

01-08-07, 01:40 PM

Instead what I am looking for is dE at each IRE just like what gets shown in the free measurements graph.
Yes, that's exactly how it works. You probably saw a single point because the margin of error in the other points was out of the "predetermined scale" (ex. show dE from 0-5). If you select "Auto-adjust" it will scale itself to show all points.

j) Please add a hot key that would cause a single reading to be taken, regardless of what the "continuous reading option" was set at in the Preferences. Sometimes I need continuous readings, other times single readings. It is very awkward to have to manually switch the preferences back and forth to control whether I am taking a continuous vs single reading.
That would work nicely.

Hughman

01-08-07, 01:47 PM

First off, thanks to all who have contributed to this software and thanks to lovingDVD for the CF comparison.

I have an Eyeone colorimeter on it's way and have a couple questions regarding it's use with the HCFR software for a DLP projector.

1- I understand that using the diffusor on the eye, for PJ lens measurements, does not diffuse in a linear fashion and, as yet, HCFR does not account for this. Can the training fuction be utilized to provide the offset to account for this?

2- I note in the HCFR software that there are two calibration modes for the eyeone, CRT and LCD (I was sure the proj option was there but obviously was mistaken). Did I error in thinking this meter would work with DLP's or will either CRT or LCD selection work.

3- I'm confused as to the accuracy of the eyeone at lower luminance levels, HDholic says it very accurate, lovingdvd says not. Which is it and and what levels of luminance is it not an issue? Obviously the diffusor, if attached, would significant lower the light levels entering the meter, is the accuracy at lower levels therrefore primarily a diffusor on/off issue?

Thanks to HCFR for supplying this software for free, I suspect an impending knock on your door will change this sooner rather than later though.

HDholic

01-08-07, 01:51 PM

I'm confused as to the accuracy of the eyeone at lower luminance levels, HDholic says it very accurate, lovingdvd says not.
I'm referring to the Display2 against S2. Lovingdvd uses the Eye-One Beamer, totally different.

They all have +'s and -'s ;) .

Hughman

01-08-07, 01:56 PM

OK thanks, I missed that important piece of info.

laric

01-08-07, 03:11 PM

In other words, I'm wondering if I get better training to let HCFR use as a reference 30-100 IRE instead of just 100 IRE. ?
No you won't have better results (at least in v1.21). You only need 100 IRE to have correct info on dE in the small window after applying matrix...

In a later version, white will be included with Primaries measures, that will remove this need (although not mandatory).

--Patrice

lovingdvd

01-08-07, 03:29 PM

Hugh2 - I have not tried it but heard from others that the EyeOne with the diffuser pointed into the lens does a much better job reading lower levels than the EyeOne beamer. That being said I don't know how good it is, but only that it is much improved.

Ki_

01-08-07, 03:31 PM

From what I recall, dE was not plotted on the RGB chart. If I right-clicked it, it had an option that said something like "dE curve" and when I added that to the chart it opened up a new panel underneath the RGB plot which just had a single point/dot. Instead what I am looking for is dE at each IRE just like what gets shown in the free measurements graph. This is all from memory but I'll double check this later today.

May be you had only one dot because curve was out of scale but I confirm that you should have DeltaE at each IRE level plotted with RGB levels.

lovingdvd

01-08-07, 03:34 PM

May be you had only one dot because curve was out of scale but I confirm that you should have DeltaE at each IRE level plotted with RGB levels.

Thanks Ki_ and HD - I will load my chc file tonight and change the scale - sounds like this will do the trick. My black measurement was something like 22 dE so I'm thinking that trashed the scale and caused this.

noizemaker

01-08-07, 03:36 PM

lovingdvd

01-08-07, 04:04 PM

OK guys i'm going to be trying out this wonderful software tonight on my Sony LCD RPTV. I would just like to know, if possible, if there is anything i need to do before i start taking readings. For example, this is my set-up, I'll be using Get-Gray as my pattern disc so i set "DVD MANUAL" as my generator, I select "EYE-ONE" as my meter since i have an EYE-ONE DISPLAY2. Is there anything else I MUST do before starting my grayscale readings?

Any info would be great guys.
Carmine.

If your EyeOne requires a dark reading (I know for certain the EyeOne beamer does but not familiar with yours), then you must do so before proceeding. Otherwise your measurements will be quite inaccurate. The software provides you with an opportunity to do so but you must know to click on the button that says something like "calibrate internal meter settings" or something like this. I missed this the first time through and it really throw me for a loop.

laric

01-08-07, 04:10 PM

laric

01-08-07, 04:29 PM

lovingdvd

01-08-07, 04:32 PM

I'd have a quick question to any of you, why don't you use the build in patterns (by plugin your PC directly to projector) ?

I understand there could be some practical reasons, but trust me you'll see all the benefit if you ever try ones ;)

--Patrice

Hi Patrice - in my case I do not use a HTPC. I do have a PC that has a DVI port and can (and do) connect that to my projector for special projects. However it is a major pain for me to do so (have to pull the PC apart from another room, lug it over, etc).

Its much faster for me to just use my DVD player and the excellent GetGray calibration disc.

My laptop (on which I run my calibration software) does not have a DVI or HDMI output. Last I looked a year or so ago such things were rare...

Does anyone know of any notebooks that have build in DVI or HDMI, or maybe a USB or firewire adapter that adds a DVI/HDMI port?

HT-Naimee

01-08-07, 04:53 PM

Oh, I mistookt that also that HDholic was not talking about the EyeOne colorimeter.

And was is this training a Spyder2 business? Do you need a ExeOne in order to correct the values the Spyder2 gives?
Like telling the Spyder2 what to add/substract from its measurements at low IRE in order to get proper measurements below 30 IRE and leaving it as it is for values above 30 IRE?

If so, can I use a profile from someone who has done this or must I buy EyeOne DigitalLT AND a Spyder2?

This whole business is starting to confuse me. I had hoped to buy a Spyder2 for 85quid, use the free software and get D65 within a few hours of playing and fiddling and maybe get better contrast and the best black/white levels.

Am I mistaken?

lovingdvd

01-08-07, 04:56 PM

lovingdvd, on your 7sec delay issue, can you double check in "preference" menu "general" tab, that you don't have anything (0) in the "Dynamic Iris Latency timeout" ?

I will double check and let you know, but I'm pretty sure it is at 0. I remember because I took note of this setting as I was very impressed to set it there (I used to have a Ruby and can certainly appreciate the need for that feature!).

Other than that, I dunno and will pass this to other team members.
(I admit I almost never use "DVD Manual" mode, so that may be this).

--Patrice

I don't think its the DVD Manual mode itself. As I recall this issue did not occur with the EyeOne, only the SpyderTV. I'll try and play with it some more to confirm a few things. In the meantime if your team has any further thoughts please let me know.

richlo

01-08-07, 08:33 PM

d) Please add an option so HCFR will play a tone/sound when a reading begins and when it ends. CF does this and it is very helpful. Although HCFR blanks the screen during a measurement (VERY NICE BTW!), I still need to keep my laptop shut to keep its stray light out (a blank screen is still bright with my LCD laptop). Without this tone I don't know when the measurement is completed and have to "peek" under the laptop cover.

Turn up your volume on your laptop..it always beeps..

Ungermann

01-08-07, 08:33 PM

Hi Ungermann, could you post your .chc file ? Without this file, it's impossible to make a precise diagnostic.
Attached is the CHF file. I hope it's ok to use data read with 1.2.0 version. It does not have primaries though. Umm... If it does not have primaries, how did it plot four big dots per each primary color on CIE diagram? I guess I do not understand something again...

lovingdvd

01-08-07, 08:36 PM

Laric - I double checked and the latency time under the dynamic iris is indeed set to 0. Please let me know what additional info I can provide to help troubleshoot this 7 second delay and what ideas your team may have regarding this.

Thanks guys for the tip about the dE in the RGB - the scaling as was pointed out was the issue.

lovingdvd

01-08-07, 08:39 PM

I noticed that the Calibration Mode options for the SpyderTV meter have an entry for LCD and PROJ. How does this setting impact HCFR and the measurements it provides?

I did not notice this setting and had it set to LCD as I measured my front pj directly out of its lens using this meter. Are my measurements still to be trusted?

Additionally, why is it that when the EyeOne meter is used there is no PROJ option? Certainly this meter can be used to measure projectors so I must not be following what the purpose or meaning of this setting is.

Thanks!

richlo

01-08-07, 08:51 PM

Ungermann

PRIMARIES are actually below your greyscale readings to the left, then your SECONDARIES are next to it. Those are part of the original release

btw...choose another gamma, whatever you have selected is not a good one..

vnitto

01-08-07, 08:53 PM

Here are the resultsof my Optoma HD70 and philips 5960/12 (firmware upgrade)

Before and after are in the attached file. I used the Spyder 2 and the HCFR DVD

Once you get the hang of this calibration, it's a snap.

I did my third TV in like 15 minutes!

I think it turned out pretty well, what do you think?

HDholic

01-08-07, 09:03 PM

lovingdvd, on your 7sec delay issue, can you double check in "preference" menu "general" tab, that you don't have anything (0) in the "Dynamic Iris Latency timeout" ?

Other than that, I dunno and will pass this to other team members.
(I admit I almost never use "DVD Manual" mode, so that may be this).

--Patrice
Patrice,

I also have the S2 and the delay happens whenever starting any measures, be it grayscale, primary or secondary. It always does this with any generator type (DVD or View images).

why don't you use the build in patterns (by plugin your PC directly to projector) ?
I like to calibrate the entire chain. It's possible that PC won't display patterns the same as a DVD player. Never tried it so can't say for sure :D .

I noticed that the Calibration Mode options for the SpyderTV meter have an entry for LCD and PROJ. How does this setting impact HCFR and the measurements it provides?
The Sypder2 is most accurate as many would know with the "filter/baffle" ON. LCD mode seems to be the most accurate "mode/offset" (if that's how it works) with the filter on. I'm not sure if PROJ is accurate taking into account the filter being ON. There used to be a CRT mode that I tried but that wasn't accurate either, so w/ S2 LCD is the only accurate option AFAIK.

You could compare results with both methods and post your results.

jayzfelon

01-08-07, 10:38 PM

What is the difference beetween calibrating gamma and black measures with hcfr and just calibrating with the dvd getgray?

laric

01-09-07, 02:47 AM

Its much faster for me to just use my DVD player and the excellent GetGray calibration disc.
of course and getGray is very nice... My point was just to highlight the software can run all the patterns directly from the very PC you use to run software and that makes things a LOT faster as all is done without having to go back and forth with DVD commands...

i disgress a bit, but with our DIY probe, we include build in IRs diode that allow ColorHCFR to direct send command to DVD player... hence nearly revamp the ease of direct PC plug as it chose the patterne, etc... (works only with our probe and our PatternDVD).

My laptop (on which I run my calibration software) does not have a DVI or HDMI output. Last I looked a year or so ago such things were rare...
cannot you use VGA port ? ;)

I did pass you suggestion to other team members, many thanks for those, they make a lot of sense.
Some were allready on our list (Ansi contraste removal, single reading confirmation removal, etc...)
I especially like your point i) Very good.

We will answer on delay (hoping we can reproduce the behaviour)

--Patrice

laric

01-09-07, 04:55 AM

lovingdvd

01-09-07, 09:25 AM

I did pass you suggestion to other team members, many thanks for those, they make a lot of sense.
Some were allready on our list (Ansi contraste removal, single reading confirmation removal, etc...)
I especially like your point i) Very good.
--Patrice

Glad you liked the suggestions and its my pleasure to contribute. I am glad you like the suggestion for Y calibration in xyY (suggestion "i"). I'm not sure if ANY calibration software offers this, and it will make a great and welcome addition.

One other feature I wanted to mention was to please consider adding the dE to the grayscale measures table (not the graph as I know now its already on the RGB measures, but I am referring to the table of results with 0-100 in it). So in addition to showing XYZ or xyZ or whatever the user has selected, it would also show dE as a 4th row in that table. You could put a little check box under the format choices (the selection of xyY, RGB, XYZ etc) that said something like "show dE?" that when checked adds the data there.

I think it is amazing that your software is already so refined and stable with lots of features and functionality. And the best part is how you and your team are hungry to keep making improvements and enhancements. Keep up the good work!

lovingdvd

01-09-07, 09:30 AM

Hello,

As highlighted by lovingDVD and confirmed by HDholic, the Spyder2 seems to require some initiallisation time (5 - 8sec)...

It looks like this was a fact users allready know and none report it (including team members :D).

This is because we re-init the probe each time you do a measure (and release it thereafter).

We are doing some tests at the moment to see if we can change this to init it once (on first measure or may be when starting document)...

Will update you.

--Patrice

Patrice - I'm glad you are able to easily reproduce this. The good news is that Colorfacts does not have this issue. It takes the measurement instantly when you hit the Single Reading command and so forth. Therefore I am confident that it is not necessary/mandatory to initialize the probe as is being done now. So I am hopeful this will be a quick fix once the developers dig in a bit. Please keep us posted. Thanks!

laric

01-09-07, 09:49 AM

lovingdvd

01-09-07, 10:10 AM

Yes, this is now sorted, we have an internal beta with some fixes (including Display probe problems sorted), and some suggestions included (yours and others)...

Too early to release it, but, as I expect, there will be a bugs/small features 1.22 fix soon.

--Patrice

Excellent news! I look forward to trying it.

Also related to 1.3, you mentioned wanting to perhaps get some feedback on the early meter training feature functionality put into 1.21. Is there any feedback I can provide that on this that would be helpful for your team? If so just let me know what questions you have or what you'd like me to comment on.

laric

01-09-07, 10:18 AM

lovingdvd

01-09-07, 10:29 AM

Thnaks for your proposal, I'll check with team, but so far I don't see anything.

We want to have probes remaining issues/problems/details to be sorted first, plus some small additions/changes (basicaly all that did not imply help files changes) to rollout a fix version instead of having everybody waiting for 1.3 (as we initially tought)

Then we may be a bit less present, working on v1.3 on bigger changes...

So, we may ask for more feedback later on when we put out probes fixes/small features v1.22.

--Patrice

Sounds good. I look forward to trying 1.22 hopefully soon.

Just thinking ahead - what is the upgrade process once a fix like this is available? Just install the new version and it'll remove the old one, or first use control panel add/remove software to remove the old one before installing the new?

audioholicJeffL

01-09-07, 10:30 AM

My denon 1920 has two settings for the HDMI output, RGB and YCbCr. Which should I be using when I calibrate. There is a difference when I switch between them. One of them has more red in the low end. My sat dish is also connected via hdmi, should I be using YCbCR or not?

lovingdvd

01-09-07, 10:39 AM

Laric - I do not recall seeing any info about this and wanted to make sure your team saw this question. Please see this post: http://www.avsforum.com/avs-vb/showthread.php?p=9397324&&#post9397324 Thanks!

Georges G

01-09-07, 01:35 PM

Hi lovingdvd

Well, about "proj" option in Spyder II calibration mode, it's useless... We heard about a possibility to pass this parameter to the SDK, but experimentally, it seems to return same values than LCD mode. It's experimental, and you should just ignore it. There is no equivalent with other probes, basically tri-stimulus probes have two modes: LCD and CRT, no more.

And actually, 1.21 has a bug with CRT mode with Eye One Display (we are actually working on it, and it's the main reason we have to plan a 1.22 soon instead of a brand new 1.3 in a longer future)

Regards
Georges

lovingdvd

01-09-07, 02:36 PM

jvincent

01-09-07, 02:43 PM

I've been following this thread with interest and it appears that I've just found a good source for a Display LT, so here's my question.

I don't remember seeing a specific answer for what mode (CRT vs LCD) to use when calibrating a plamsa. Since it's phospor based I'd be tempted to use CRT but it's not really the same.

Thanks and good work guys.

Whisper2

01-09-07, 05:09 PM

Lyckman!!!! Your the MAN!!! Problem solved!!!!

For ANYONE having trouble with this meter locking the HCFR software use the .dll I have attached. It is version 1.2.0.0 and completely solved my lockup issues with the DTP94!

Hi! Yesterday, my last sonde, DTP94 (yes, I've got a HCFR sonde - I did it - and a Spyder2 too), arrived.

I installed them in my PC, and I didn't have any problems ( XP Pro, SP2),

Today I installed the HFCR soft and the probes in my notebook. The Monaco Optix hang up the system (XP Home, SP1). I read this (upper) post from Adam and I used his dll and it fixed my problem..

Now DTP94 works properly

Has the HCFR soft got any problems with XP Home edition and DTP94?

Regards..

PD: Is it mandatory to upgrade the firmware of HCFR sonde from 5.1 to 5.2 for simultaneous measures ? I made a modification (yes .. my green LED is in the other side...I think it's better :D

laric

01-09-07, 05:47 PM

Has the HCFR soft got any problems with XP Home edition and DTP94?
Not to my knowledge...

Is it mandatory to upgrade the firmware of HCFR sonde from 5.1 to 5.2 for simultaneous measures ? I made a modification (yes .. my green LED is in the other side...I think it's better :D
It is not required for simultaneous measures... Just a small "+" ;)
The only purpose of 5.2 is to shutdown the led on second probe when doing measure with first one... not mandatory unless led light really annoys you...
(a sticker on it is fine tought ;))

Side question, where did you eared about Probe Firmware 5.2 ?? :D

--Patrice

zancid1

01-09-07, 08:07 PM

Any updates on this "issue"...I posted a little while back my calibration attempt:

http://www.avsforum.com/avs-vb/showthread.php?p=9326388&highlight=plv#post9326388

I seemed to get the same thing with my Spyder2. Anytime I put RBG in line the picture was clearly too red. The result was I purposely lined up RED off target even across the range but at a point only based upon my eye.

Any comments from you on this or otherwise from folks regarding the .chc file in the attached post.

HDholic

01-09-07, 08:24 PM

Have you tried reading off the projector to see if it reads more accurately?

richlo

01-09-07, 08:46 PM

ZANCID1

your RED looks to be your clipping color, so when doing your greyscale, avoid using Red, cut Green to bring up Red, using blue and green should help bring RED into place..On my mits300U, Red is my clipping color, so this is the way I go. Bring a 100IRE, set everything back to default, then hit Continuous Read, then go highlight 100IRE (make sure you pause this pattern), then click on Display Measurement icon, start adjust your blue and see how it effect the other color, adjust green and see how effects other color, you should be able to work with those as to avoid using RED.

In addition, you still has some work to do according to this file in the greyscale..Try concentrating in the strong areas of the spyder2, 30IRE and up..make sure you have selected the right mode..I personally like to get as close to the screen with the pod to get as much light into the sensor, some say 12", I say NO WAY..get closer..and aim at a 45degree angle to point away from the shadow of the pod..

Goodluck..

HDholic

01-09-07, 09:07 PM

You will immediately see that Red is off - this is on purpose - no matter what I try if I bring RGB all to the same level (low DE) in the RGB historgram the picture is obviously too red, especially in the higher IRE's. HCFR reading seem to think everything is perfect in this too red situation.

sidenote: I've looked around a bit and found some references to calibrated Z1's without issue, but more interestingly is the cine4home article that stated that they never ever could bring the run up to the levels of Green and Blue and that a CC filter was needed to bring balance.

In my case I can in fact get the red up to that level but as I said it's clearly wrong.

From what I understand, you have "too much red" not a "lack off" correct? I have a S2 as well and never gave me such results. It's either a bad probe or something being done wrong.

If your projector "Lacks" red, then do what Rich suggested, otherwise that won't help if your projector DOESN'T lack Red. Are you using a red filter?

zancid1

01-09-07, 09:07 PM

What makes the determination that red is the clipping color? Was it my statement of no matter what I get too much red if I try to aim for 100% with Blue and Red Adjustments? or is the fact that my Red is low across my greyscale. Keep in mind that the latter was on purpose to in the short term deal with my inability to remove red overtones. I'll take heed of your advice and try again (with fancy 1.22 ;-)).

I had to take a break as this was my first calibration attempt and my wife was beginning to wonder if I had gone a bit strange spending hours in a darkened theater room stairing at "those grey squares". :p

Cheers and Thanks.

ZC

richlo

01-09-07, 09:16 PM

No..look at your gamma curve, and you see red kink on the curve gamma at the top, if you right click the pattern and select logarthic, and select RED, GREEN,BLUE to view, you will see that Red takes a deep turn down on the contrast side, meaning that RED is clipping - your using too much red, you can get away by cutting red but not by increasing it...

My wife just rolls her eyes at me, but of course, I do it when she goes to bed..I AM ONE STRANGE DUDE, but my DLP is calibrated nice :0)..

What are you calibrating here

zancid1

01-09-07, 09:22 PM

HDh,

Yes too much red, sorry I get my nomenclature backwards sometimes. I'll try again with this suggestion and also do a basic S2 calibaration of my laptop LCD to see if I can repeat the excess red issue.

I have my sensor just below center, parallel to the screen from a top/bottom view and tilted up 45 degrees towards center from a side view.

thx

HDholic

01-09-07, 09:26 PM

Gotcha. It's very strange.

zancid1

01-09-07, 09:33 PM

Boxlight HD1 (which is really a Sanyo PLV - Z1) which is a LCD front project about 4 years old. Haven't been able to affort as of yet to upgrade so decided I'd wait until a) I could afford and b) prices decrease on the 1080 natives such as the Mits5000, JVC etc.

I think I've confused myself again but I'll give it a shot.

From a RGB histogram perspective is where I'm getting the idea or too much or too little red. If I try to push up my red from my current 90%ish to 100% targeting the proper De then I get a visually too red picture. Your gamma curve observation kinda makes sense. Essentially I'm trying to push out too much red so the PJ is clipping it affecting the ability of the sensor to properly sense these levels. So try as a basic basis point to zero everything out to out of box, leave RED where it is at 100IRE and move Blue and Green down to it instead of the other way around.

Obliged.

richlo

01-09-07, 09:34 PM

Also, with the spyder..select extend reading time on the lowend and average read time on the dark measurments, yes, it will take longer but it will give you better results (I HOPE)

lovingdvd

01-09-07, 09:39 PM

Also, with the spyder..select extend reading time on the lowend and average read time on the dark measurments, yes, it will take longer but it will give you better results (I HOPE)

I've been wondering about this option myself. I set my probe for the maximum 5000ms AND checked the box to tell it to take the average of multiple measurements (takes about 60 seconds for very low IREs).

I tried to ALSO check the "extended testing time" box too, but when I did it unchecked the "take multiple averages".

Note to developers - I'm not sure why these settings need to be mutually exclusive. Please consider a) allowing up to 10,000ms max testing time (ColorFacts allows up to 8000ms which is what I've always used), and b) allowing both boxes to be checked. It takes way longer this way, but I've found I get done overall much faster if I get accurate, consistent results to begin with...

Adam Gutierrez

01-09-07, 09:47 PM

Can someone list the do's and dont's of the CIE triangle when adjusting primarys and secondaries. Should I keep all the secondaries on the same "line" with the primaries? My Benq 8720 has separate adjustments for all 6 colors and I can do some pretty wild things like getting almost all colors at reference except the green, which pushes cyan off the "line" of my triangle. After doing this color and tint are out of whack and I have to readjust with color and tint controls.

Also, should I select 601 or 709 for color space reference? I am using GetGray patterns from my HD-A1 to my 8720. I am using 0-255 for the CIE Triangle (color/tint) and 16-235 for the gray scale measurements.

And I would just like to say thanks again to the HCFR Team, what a great piece of software. I'm having a ball (screwing up my very expensive PJ!) Thank God for "Factory Reset"!

richlo

01-09-07, 10:16 PM

Use709 Rec
Use 16-235 for ALL measurements
Do not use 0-255 for the CIE triangle (I know at one time I told people this- not correct)- this is for the computer world..

Try to get your color and tint as close as possible, there is also compromises to be made,most projectors do not have the perfect triangle unfortunately..Read up more on the 8720 and its color decoder adjustments and see what people are saying (if you havent already)

lovingdvd

01-09-07, 10:37 PM

Adam - do make sure your secondaries are on the same line that connects the primaries which make them up. Also check the Lightness (luminance) of each color and tweak that to standards assuming your pj offers this adjustment on a per color basis. You have to do this by hand but Laric and team may considering adding this calibration as a feature in a future release.

primetimeguy

01-10-07, 12:05 AM

Use709 Rec
Use 16-235 for ALL measurements
Do not use 0-255 for the CIE triangle (I know at one time I told people this- not correct)- this is for the computer world..

Just curious as to what changed your mind or how you arrived at the new conclusion. And what effect will I see since I had done my colors using 0-255.

For me, this is what my GetGray guide says...

"This disc assumes the display, DVD, and the entire video chain is set to display "video levels". Some devices are set to interpret digital colors at "PC levels" where digital RGB levels extend from 0 to 255 with 0 being black. Powerpoint presentations from a PC are an example of this. DVD's are not encoded at PC levels; they are encoded using video levels. "PC Levels", “encoded YCbCr”, and "Video Levels" are subjects beyond the scope of this document. In short "Video levels" means the display interprets digital 16 as black, and digital 235 as white. Additionally, many but not all displays and/or DVD devices can display levels called "below black" or "above white". Some devices "clip" any video information in those ranges (below digital level 16 or above digital level 235). These devices may have control settings to allow display of “below black” and/or "above white" (read your device manuals). Although DVD's are created with "video levels" (levels 16-235), some DVD's may contain video information that
falls outside the video level limits. The GetGray DVD is encoded to intentionally contain below black and above white information, to enable the calibrator to have full visual information when calibrating. Some calibrators prefer to adjust so a display shows some "below black" information and/or some "above white" information. The author recommends no below black, and possibly 1 step above white, but this is just a personal calibrators preference. Showing some above white information for example may reveal some bright white details (i.e. in bright sky clouds that may have otherwise been clipped). Strictly technically speaking, not showing information outside of 16-235 is correct."

kylek23

01-10-07, 12:08 AM

Has anyone had any luck with the Eye One Display 2 sensor and a front projector? I've really been struggling with mine.

Should I be calibrating it with the Black level button? Do I just place it on a dark surface for the calibration? (In Eye One Match you only have a chance to calibrate the Display2 if you are going to take ambient light measurements, Babelcolor has you first place the sensor on a dark surface and then take a reading from a 100% white square on whatever screen you're calibrating.)

If I place the sensor 12 in. from the screen without its ambient light diffusor I get crazy measurements (red isn't even on the map) from previous calibration settings done with the Eye One Pro and Spyder2 that I know are at least close to D65. If I attach the ambient diffusor the Display2 gets in the ballpark but I don't know if I can trust it as HCFR doesn't support the diffusor. Very frustrating (and this was after struggling with the latest Parallels beta and trying to get Eye One Match and HCFR to even detect the Display2, I had to downgrade to a retail build of Parallels to get it to work).

HCFR developers: What Eye One devices have you tested? Both the Display2 and Eye One Pro?

Thanks

Markz2k

01-10-07, 12:17 AM

I got a SpyderII meter, and played around with the HCFR software tonight for a few hours. I think I did a decent job with greyscale on my Sony KDS-R60XBR2 RPTV. Wasn't really able to do much with the CIE chart, don't fully understand it. See attached file, and let me know how I did, please.

I only used the White Balance setting in the user menu. I used the DVD from the HCFR site, played in my HTPC connected to the TV via component. (The Toshiba HD-A1 I previously had refused to play the disc, complaining it was in PAL format. I have an HD-A2 due to be delivered tomorrow, so I'll try it in that.)

If anyone has an .ISO of the HD-DVD patterns disc, I'd appreciate a link to it. I don't have the needed software to make my own disc from the pattern files on HCFR's forum.

My thanks to the HCFR team, this is excellent software at any price.

HDholic

01-10-07, 12:26 AM

Markz2k

01-10-07, 01:12 AM

See my post here (http://www.avsforum.com/avs-vb/showthread.php?p=9348191&&#post9348191) for HD DVD patterns. Your blue needs more adjustment. I'd suggest doing an 11 step grayscale reading for gamma calculation.

Could you explain how to do an 11-step? I see options for 4, 10, or 20 steps in the drop-down box when doing a grey-scale read. How do I do 11? Thanks for the link to the HD-DVD pattern, I'm downloading them now. I also have copies of Avia and DVE if those would help.

HDholic

01-10-07, 01:15 AM

Pick the 10 steps option, although technically it's 11 since you count 0IRE ;) .

nate358

01-10-07, 02:17 AM

hey anyone else getting crazy contrast readings?.... I'm getting something like 11312:1. Sometimes I get 1770:1, but mostly the bigger number... Now I know my projector isn't that sweet! It's rated at 2500:1 and I know it won't hit that. What gives? The new release is great guys!

laric

01-10-07, 03:27 AM

The CR is simply the ratio between your black (0%) and white (100%) reading...
Obviously black reading have a huge impact on this and likely explain variations.
(as you likely have variations on reading on black)

--Patrice

Georges G

01-10-07, 04:52 AM

Hi Lovingdvd
Note to developers - I'm not sure why these settings need to be mutually exclusive. Please consider a) allowing up to 10,000ms max testing time (ColorFacts allows up to 8000ms which is what I've always used), and b) allowing both boxes to be checked. It takes way longer this way, but I've found I get done overall much faster if I get accurate, consistent results to begin with...
Well, I made it mutually exclusive because I considered it's useless to add both. But you know this probe much more than me, so you're certainly right. I can change it very easily in future 1.22.

Thanks for all your feedback ;)

Regards
Georges

retrof

01-10-07, 05:17 AM

Hi folks,

I got my greyscale into pretty good shape I think, and decided to start looking at the gamma settings on my Z3 to see how they affected things. I am using an HD-DVD player with the HCFR HD test patterns, and HCFR set to gamma 2.2 and Rec.709

I was looking to see 2.2 or close on the readings. In setting gamma to -2 I get an overall of 2.3 according to HCFR - if I set it to -1, I get 2.18. So, I have a couple of questions (I have attached both the gamma-1 and gamma-2 chc files) - my tracking is better on gamma -1 than on -2, but I have a light controlled room. Should I perhaps leave it at gamma-2 and re-adjust the offsets/gains to get the deltas back in line? or should I just stick with -1?
The reason I ask is because I have read here in places that higher gamma might be better for dark rooms, so thought maybe I should be shooting for 2.3 or 2.4 instead?
Anyone have any comments on my current greyscale tracking on both? - I think I can get the deltas down to almost 0 from 30 to 100 if I really try, but don't want to spend too much time on it until I get the gamma sorted out...
What difference does it make changing the reference gamma in HCFR?
I have had no luck whatsoever in getting any movement on green in the primaries.

Thanks again for everyone's help, and to the HCFR team for such an excellent application. It can be frustrating to a newbie, but it is also very rewarding.

-rf

Whisper2

01-10-07, 06:16 AM

Not to my knowledge...

It is not required for simultaneous measures... Just a small "+" ;)
The only purpose of 5.2 is to shutdown the led on second probe when doing measure with first one... not mandatory unless led light really annoys you...
(a sticker on it is fine tought ;))

For me, The green led is not a problem, because is in opposite side of your design.. ( I like it more, because I can see when the HCFR Probe is working.. )

Side question, where did you eared about Probe Firmware 5.2 ?? :D

--Patrice

Well, actually, I didn't ear , I've got it (the 5.2 firmware....) Where?? it's in your soft..

I was looking for help to make "simultaneous measures " and there, it said "I must to upgrade the firmware"... There is a link.... :D and now I've got the firm and a program to load the firm..but my french is, well isn't... I haven't done nothing for the moment

Regards

richlo

01-10-07, 07:01 AM

Just curious as to what changed your mind or how you arrived at the new conclusion. And what effect will I see since I had done my colors using 0-255.

F

previous old released when selecting 16-235 the primaries where not coming up like with 0-255, that has been corrected since, and at the end, this did not effect anything. So just stick with 16-235, no need to switch..

0-255 is for PC levels and not DVD levels, which is 16-235..

primetimeguy

01-10-07, 08:40 AM

previous old released when selecting 16-235 the primaries where not coming up like with 0-255, that has been corrected since, and at the end, this did not effect anything. So just stick with 16-235, no need to switch..

0-255 is for PC levels and not DVD levels, which is 16-235..

Thanks, I played around with both settings and didn't see much of a difference anyway.

audioholicJeffL

01-10-07, 09:53 AM

Rich,

Why would we be using rec 709? That's were I am confused. I have an HD sat box and an upconverting DVD player. Which color space should I use? Or does it even matter much?

richlo

01-10-07, 10:11 AM

Rich,

Why would we be using rec 709? That's were I am confused. I have an HD sat box and an upconverting DVD player. Which color space should I use? Or does it even matter much?

because its a true HD player...

an upconverting player can either have a colorspace of 601 or 709, its dependent. Like the Oppo is an upconverting player (the 971), its 601 colorspace.

greeno

01-10-07, 12:01 PM

Hi Richlo,
I don't know specifically about the 971, but the 970 and the 981, when upconverting to 720p/1080i, output Rec 709, *eventhough* the original DVD is Rec 601.

Best,
jeff

lovingdvd

01-10-07, 12:25 PM

Gamma Calculation Methods

1) Display gamma

The gamma is compute in light to human eyes perception.
After approximation brightness (L) is expressed according to the signal (V) using the equation:
[1] L = V^gamma

Standards recommend to us a 2.5 Gamma target (2.45 after math) in this mode.

2) Display Gamma with Black Compensation (Default)

In practice one can never achieve an absolute black and to properly setup a projector (or screen) it is necessary to take in account of the black you can really obtain.

This is usually done by computing Gamma after removing the "absolute" black from every measure.

3) Camera Gamma (Standard Offset)

To have an accurate reproduction, some interpretations say that you must apply a symmetrical treatment to the encoding one. The reference equation being: [2] Y = ((V+offset)/(1+offset))^gamma

The "offset" is a parameter use at black encoding stage (and reproduction using this method). It's value is 0.099 on all video standard. (0.055 in sRGB)

4) Optimized (Regression)

This calculation method is based on the fact absolute black recorded by camera is the absolute black projector can produce.
But offset in standard is roughly for 200:1 contrast ratio, a lot lower than current ration projector can achieve.
The optimized method tries to find the Offset AND Gamma parameters that best match the measures, black included.

Pictures as contrasted as in method 2 can be achieved using this method with even a better black details level.
But the compute method -- and so the resulting projector settings -- are more dependant of proper and accurate low level probes readings.

--ColorHCFR Team

A contrast calculator suggests I should set my gamma to 2.5 for achieve the best settings. Since there are so many definitions for gamma, I want to make sure I set HCFR to the correct gamma formula so that my 2.5 result matches what they had in mind when recommended a setting of 2.5...

Can you take a quick look at their formula for gamma calculation here: http://home1.gte.net/res18h39/curve.htm and let me know which of your gamma measurement settings would match theirs? It looks to me the best match would be your first one "1) Display gamma" but just wanted to make sure.

lovingdvd

01-10-07, 12:30 PM

Hi Lovingdvd

Well, I made it mutually exclusive because I considered it's useless to add both. But you know this probe much more than me, so you're certainly right. I can change it very easily in future 1.22.

Thanks for all your feedback ;)

Regards
Georges

Thanks Georges. Yes if you can bump up the value for reading timems to allow up to 10,000ms that would be great.

When you check the box that says to average multiple readings, how many readings does it take? My guess is about 5?

Also when you check the box to tell it to read longer at lower light levels, how much longer does it then read (%-wise?) and what is the IRE point that is considered low light levels (i.e. defined as 20 IRE and below, 30 IRE and below?)?

If its not too much trouble, it would be nice to make all those parameters editable. For instance one could tell it to read 30% longer below 30 IRE and to take the average of 8 readings. With this one could experiment to find the combination of settings that yielded consistent results with minimum read time. Seems like all these params are hard coded so it may be just a matter of exposing the data in the dialog box for editing. Thanks!

HDholic

01-10-07, 01:06 PM

Can you take a quick look at their formula for gamma calculation here: http://home1.gte.net/res18h39/curve.htm and let me know which of your gamma measurement settings would match theirs? It looks to me the best match would be your first one "1) Display gamma" but just wanted to make sure.
"Camera Gamma" uses the 0.099 offset.

lovingdvd

01-10-07, 01:21 PM

"Camera Gamma" uses the 0.099 offset.

Thanks. That's the option I was thinking of but listed the wrong one in my post. OK I'll give that a try. Right now I measure 2.19 with option #2 - it'll be interesting to see what I get with option #3 with the same calibration settings.

richlo

01-10-07, 01:29 PM

Thanks. That's the option I was thinking of but listed the wrong one in my post. OK I'll give that a try. Right now I measure 2.19 with option #2 - it'll be interesting to see what I get with option #3 with the same calibration settings.

my bet..it would be something like 2.59ish..I think Camera Gamma is dependent on the display device. I know on my Mits 3000 DLP, there is no way I can even get that..without signficantly reducing brightness and contrast..with OPtion 2, I get exactly what I am looking for, I get a linear gamma (Logarthic and curve are very tight for all color and luminance). Some people fail to realize that although at the top of the greyscale it states AVERAGES of gamma, this is a bit misleading as you can have a very low end and a very highend and the average of the two will mislead you. Its best to look at Logarthic to see how each and every one of the colors are running, at it will also show quickly if your clipping any of the colors.

richlo

01-10-07, 01:31 PM

Hi Richlo,
I don't know specifically about the 971, but the 970 and the 981, when upconverting to 720p/1080i, output Rec 709, *eventhough* the original DVD is Rec 601.

Best,
jeff

That is why I mentioned that they can have either one. I know for a fact that the 971 is Rec 601, since I posted earlier in this forum that even Oppo stated this.

lovingdvd

01-10-07, 01:37 PM

my bet..it would be something like 2.59ish..I think Camera Gamma is dependent on the display device. I know on my Mits 3000 DLP, there is no way I can even get that..without signficantly reducing brightness and contrast..with OPtion 2, I get exactly what I am looking for, I get a linear gamma (Logarthic and curve are very tight for all color and luminance). Some people fail to realize that although at the top of the greyscale it states AVERAGES of gamma, this is a bit misleading as you can have a very low end and a very highend and the average of the two will mislead you. Its best to look at Logarthic to see how each and every one of the colors are running, at it will also show quickly if your clipping any of the colors.

Thanks Rich. I am on the pre-order for the JVC RS1 that has some impressive specs, mainly at >12,000:1 native CR @ D65 with plenty of brightness (as I'm sure you've read about).

We've been studying intra-scene contrast and ways to measure it. Our models suggest that the RS1 with a gamma of 2.5 (measured via option #3) would work best, and the pj just may have the specs and capabilities to pull this off in a light controlled room (i.e. no loss of detail in the blacks etc).

retrof

01-10-07, 01:39 PM

One additional question for anyone that may be able to answer: if I change the gamma chart to log scale, should it look like this (see my attached file), or should all colors be in a straight horizontal line close to the 2.2 reference?

Cheers,
-rf

Hi folks,

I got my greyscale into pretty good shape I think, and decided to start looking at the gamma settings on my Z3 to see how they affected things. I am using an HD-DVD player with the HCFR HD test patterns, and HCFR set to gamma 2.2 and Rec.709

I was looking to see 2.2 or close on the readings. In setting gamma to -2 I get an overall of 2.3 according to HCFR - if I set it to -1, I get 2.18. So, I have a couple of questions (I have attached both the gamma-1 and gamma-2 chc files) - my tracking is better on gamma -1 than on -2, but I have a light controlled room. Should I perhaps leave it at gamma-2 and re-adjust the offsets/gains to get the deltas back in line? or should I just stick with -1?
The reason I ask is because I have read here in places that higher gamma might be better for dark rooms, so thought maybe I should be shooting for 2.3 or 2.4 instead?
Anyone have any comments on my current greyscale tracking on both? - I think I can get the deltas down to almost 0 from 30 to 100 if I really try, but don't want to spend too much time on it until I get the gamma sorted out...
What difference does it make changing the reference gamma in HCFR?
I have had no luck whatsoever in getting any movement on green in the primaries.

Thanks again for everyone's help, and to the HCFR team for such an excellent application. It can be frustrating to a newbie, but it is also very rewarding.

-rf

richlo

01-10-07, 01:45 PM

One additional question for anyone that may be able to answer: if I change the gamma chart to log scale, should it look like this (see my attached file), or should all colors be in a straight horizontal line close to the 2.2 reference?

Cheers,
-rf

you want them straight as possible. I'll post my latest results for my Mits3000U today and you will see what it should TRY to look like. WHat you dont want is a huge shift going up on one end and then down on another..so getting it linear is very important..not easy, that is why I have about 17 calibrations done on my Mits3000

audioholicJeffL

01-10-07, 02:35 PM

because its a true HD player...

an upconverting player can either have a colorspace of 601 or 709, its dependent. Like the Oppo is an upconverting player (the 971), its 601 colorspace.

Any idea what Denon is using in their upconverting players?

richlo

01-10-07, 02:52 PM

Any idea what Denon is using in their upconverting players?

best way is to probably see which CIE triangle seems the closest.

laric

01-10-07, 03:05 PM

Can you take a quick look at their formula for gamma calculation here: http://home1.gte.net/res18h39/curve.htm and let me know which of your gamma measurement settings would match theirs?
That is exactely Option 3 (Camera Gamma / Std offset)
As a side note, it is not what ColorFacts uses ;)

My post on this, on purpose, a bit short, but we can go further if need be... We assume this Gamma question will anyway be raised often...

--Patrice

benven

01-10-07, 03:24 PM

Ok gurus...help please. BenQ PB6200, LG upconverter DVD player and SpyderII. Did the projector first and then turned the probe towards the screen. Gray scale is OK, not spectacular. Colours OK, again not spectacular. I haven't really monkeyed around in the service menu. Only changed the PW gamma (whatever that is) to YPbPr. Much better looking picture and measurement at this gamma setting. Problem is...why is the gamma so ugly?? And how in the heck can I go about resolving that issue? Any comments would be appreciated.

And thanks to laric et al for a great piece of software!!

richlo

01-10-07, 03:47 PM

Instead of running a 20 step scale, run a 10 step scale..see how it looks then, work on your greyscale some more, you still have some work there..then we will see how your gamma looks. Try to find any reviews on your projector to find out which is the best out of the box gamma setting, generally there is one called something like Theather, or Cinema...PW?? wonder what that actually means..

laric

01-10-07, 04:01 PM

I would double wath richlo says... ad 10 (11 level) is way enough to start with...
You are on good track, need a bit more tunning...

And thanks to laric et al for a great piece of software!!
I'm far far from being alone ;) !

--Patrice

audioholicJeffL

01-10-07, 04:45 PM

I just want to say thanks again guys. I am going back to school at age 41 and there is no way I could affoard to get an ISF calibration at this time. This software has allowed me to get great results at a fraction of what an ISF cal would cost. I thank you one and all. :D

benven

01-10-07, 04:54 PM

Thanks for your quick response guys!

Will do the 10 (11) step IREs. Yeah, PW Gamma..that is something I have been scratching my head on. Even sent BenQ tech support an e-mail and they said they couldn't help me?? I have to call a tech support number for help? Oh well.... I'll post the 10 step run when I get a chance to perform the testing.

Ungermann

01-10-07, 05:01 PM

I am still in the dark which target gamma curve to use. I have Pioneer plasma, I run HCFR DVD with patterns on my OPPO-970 connected via HDMI. I use 480p resolution, therefore I guess colorspace is Rec 601.

If I use display gamma as target, then my quite bright settings are still below target line. If I use Display Gamma with Black Compensation, then my gamma curve appears higher than target gamma, I find this more believable. I would appreciate a word of wisdom before I post my chc files (don't have them right now).

HDholic

01-10-07, 05:32 PM

Ungermann

You should use window patterns with a Plasma otherwise your calibration most likely won't be as accurate as it can be. HCFR DVD only has full screen patterns and that's the reason you're questioning the gamma.

laric

01-10-07, 06:21 PM

HCFR DVD only has full screen patterns and that's the reason you're questioning the gamma.
Nop ;) The pattern DVD has full and windowed patterns ;)

Obviously the build in (PC Generator) also have and u can even decide the percentage of window...

--Patrice

HDholic

01-10-07, 06:24 PM

I was never able to find the window patterns. I have DVD version .15 I think. Where can I find those?

laric

01-10-07, 06:33 PM

I must admit I have to look and come back with the info (I never use that ;))
But I know for sure they are on DVD, we add them in 1.15 version, they are Title 21 and up...

Edit : Just checked, they are, just choose "TV / Plasma / RPTV", then pattern are 10% windows... (they are full if you chose "Video Projecteur")

--Patrice

HDholic

01-10-07, 06:35 PM

I just finished looking at the DVD and again I couldn't find them.

Ki_

01-10-07, 06:47 PM

It should be on the first menu of the DVD.
You have choice between TV/Plasma and projector.
If you choose TV/Plasma, all patterns should be 10 %.

Ungermann

01-10-07, 06:51 PM

Ungermann, you should use window patterns with a Plasma otherwise your calibration most likely won't be as accurate as it can be. HCFR DVD only has full screen patterns and that's the reason you're questioning the gamma.HCFR DVD has both full-screen for projection TVs and window patterns for plasmas and LCDs (I think it actually does not matter which one to use for LCD, but I have plasma so I don't have questions).

By the way, I like that some patterns on the DVD have IRE numbers on the bottom of the screen, this works great when I cover up the meter with the top-mid portion of the screen to protect meter from sunlight. Not all screens are labelled though, I would like they all to be labelled on the bottom like some of them are already labelled now.

Also, is it possible to retain my choice RP/plasma when I choose another pattern? Right now I have to go back to the main menu to choose plasma, and then to choose different pattern set. Or is this a limitation of DVD format?

How complex is it to recode the DVD into NTSC? My player supports PAL-to-NTSC conversion, but most players in the stores do not.

In regards to window patterns. My laptop does not have HDMI/DVI output, all it has is composite, but this is ok as a first approximation when I go shopping and check the TVs out right at a store. Anyway, running patterns from a computer complicates my "target gamma" question even more, I don't know if my laptop outputs in TV colorspace or in computer colorspace.

The last note. I noticed that HCFR program now blanks the screen. This makes sense for computer-generated patches, but absolutely no sense in case when patterns are used from DVD :)

HDholic

01-10-07, 06:52 PM

Dbower

01-10-07, 06:58 PM

Can someone list the do's and dont's of the CIE triangle when adjusting primarys and secondaries. Should I keep all the secondaries on the same "line" with the primaries?

Me too wants to know! Is it more important to line everything on the reference triangle, or is it most important to get the secondaries to line up with the opposing primary and the white point (even if the primary is outside CIE)?

I have found I can move green to close to the reference, but at the expense of the secondaries (cyan and yellow in this case) lying outside the triangle all alone. I suspect I need to not move the primary so much - to keep everything roughly in line. At least the TV gives me the full 6-color adjustment (but, more to mess up I guess).

Also: I've got a SpyderTV - it's not clear in this thread if the SpyderII setting will work. Last, the TV is a 50" plasma - should I use the CRT setting?

-Dave

richlo

01-10-07, 07:21 PM

The last note. I noticed that HCFR program now blanks the screen. This makes sense for computer-generated patches, but absolutely no sense in case when patterns are used from DVD :)

you can uncheck that under GENERATOR- then CONFIGURE.

richlo

01-10-07, 07:39 PM

10% ?! Wow that's small :eek: ! I guess I did find them then...but I thought there was an error since they are so small!

Here's is the interesting thing though. If I go to advanced patterns and then return to "10% window patterns" they become Full Screen.

On the other hand, I think they should be about 50% and labeled, otherwise users won't know what they are displaying.

I did a comparison on the HCFR pattern full screen and the GetGray window, nothing changed in my settings, but I actually like GetGray much better for overall use - I love their 20%/80% window and their 30%/80% for starting your greyscale..but it makes no sense other than that to not to use the HCFR patterns (well its good for contrast and brightness settings)..I wish though that HCFR window patterns were 50% also...

btw..here is other one of my famous mits calibration..hahah...see my Gamma results...I used 2nd option since there is no way I can even get close with using Camera Gamma (3rd option), although I will be using that for my Mits55807CRT

jvincent

01-10-07, 07:59 PM

but it makes no sense other than that to use the HCFR patterns (well its good for contrast and brightness settings).

Don't forget that you need to calibrate the entire display chain since it's not necessarily true that the DVD player will be "neutral".

That being the case, using a DVD in a player is still a requirement IMHO.

richlo

01-10-07, 08:03 PM

Don't forget that you need to calibrate the entire display chain since it's not necessarily true that the DVD player will be "neutral".

That being the case, using a DVD in a player is still a requirement IMHO.

the HCFR pattern I am speaking about is the one you burn on a DVD disk, not the built-in-one in the program (VIEW IMAGES)

jvincent

01-10-07, 08:08 PM

DOH! Never mind.

primetimeguy

01-10-07, 08:09 PM

HCFR DVD has both full-screen for projection TVs and window patterns for plasmas and LCDs (I think it actually does not matter which one to use for LCD, but I have plasma so I don't have questions).

I thought it was recommended to use window patterns for all TVs and the full-screen for projectors. I have a CRT RPTV and have been using the 75%. Is this correct?

richlo

01-10-07, 08:12 PM

Primetimeguy

some say to use windows ALL the time..but for CRT, for sure, use windows..with front projectors I find it to be okay to use FULL..

Ungermann

01-10-07, 08:22 PM

I did a comparison on the HCFR pattern full screen and the GetGray window, nothing changed in my settings, but I actually like GetGray much better for overall use - I love their 20%/80% window and their 30%/80% for starting your greyscale..but it makes no sense other than that to use the HCFR patterns (well its good for contrast and brightness settings)..I wish though that HCFR window patterns were 50% also...
HCFR windowed patterns are just the right size on my 42-inch plasma, maybe they will be too small on smaller screens. 50% will be too large, plasma screens have problems with brightness of large areas.

HDholic

01-10-07, 11:36 PM

I did a comparison on the HCFR pattern full screen and the GetGray window, nothing changed in my settings...
DLP's have no problem outputting full brightness since they use lamps, but try it on your CRT Mits and you're guaranteed to see a big difference.

50% will be too large, plasma screens have problems with brightness of large areas.
That's the norm including "Pro" discs like Avia, and of course Getgrey(created specifically for digital displays) :).

primetimeguy

01-10-07, 11:55 PM

Interesting comment on gamma. Previously I was using an older Sony DVD player and a gamma of 2.2 in Camera mode was a dead match for my CRT RPTV and seemed right. Using 2.5 was way too dark. I just got an Oppo 970 and did an initial pass without changing anything on the TV and using the same input and now my TV tracks dead on at 2.5 on Camera mode and does not look too dark. Must have been an offset in my old player.

Georges G

01-11-07, 03:33 AM

Hello
Thanks Georges. Yes if you can bump up the value for reading timems to allow up to 10,000ms that would be great.

Ok, I do it in next version, and I let you free to check both checkboxes.

When you check the box that says to average multiple readings, how many readings does it take? My guess is about 5?

By default, the number of readings depends on two parameters: a minimum value for the sum of all Y, which is 10 by default. It can be achieved with only one measure, but the software will add more measure until the sum of Y achieve 10. The X, Y and Z values are then averaged between all measures. There is a limit to the number of measures. Even if the value of 10 cannot be achieved, the software will stop measuring after 8 identical measures. Usually, this limit or eight is used when measuring black.

Indeed, those parameters (Y to achieve before averaging and maximum number of loops) can be changed, but you must put them by hand in the ColorHCFR.ini file (using notepad for example). Under section [SpyderII] you can add those lines:
MinRequestedY=10
MaxReadLoops=8
You can put any values instead of those default ones, even ones which will make your measuring times particularly long ;)

Those two undocumented values exist also for sections [EyeOne] and [DTPSensor]. This is still experimental, we plan to use them internally to define experimentally the best choice for each probe, to know precisely which is the best balance between read time and accuracy. But those experiments take really much time, and we did not have enough time to test it...

Also when you check the box to tell it to read longer at lower light levels, how much longer does it then read (%-wise?) and what is the IRE point that is considered low light levels (i.e. defined as 20 IRE and below, 30 IRE and below?)?

I do not take an IRE point as a limit: I check the minimum RGB component. This way, primary and secondary colors, which are difficult to measure accurately, also benefit of time extension, exactly like black. The inflexion point is 80 (on a 0-255 scale), which is is little upper 30% gray. At 30% gray, there is a very small time increase (about 5%), and at 0% the time increase is 300% (usual time multiplied by 4). For the next version, since you request to use both time increase/time adjust at the same time, I will add an absolute limit... 32000 or so, to avoid potential trouble if this value is a signed 16 bit in S2 driver ;) I confess I just want to avoid boring testing around it ;)

If its not too much trouble, it would be nice to make all those parameters editable. For instance one could tell it to read 30% longer below 30 IRE and to take the average of 8 readings. With this one could experiment to find the combination of settings that yielded consistent results with minimum read time. Seems like all these params are hard coded so it may be just a matter of exposing the data in the dialog box for editing. Thanks!
Well, some are editable... Time adjustment is not, but I don't think it should be... but if you find good arguments about that, I'm ok to add user parameters ;)

Regards
Georges

Georges G

01-11-07, 03:44 AM

Hello Ungermann
The last note. I noticed that HCFR program now blanks the screen. This makes sense for computer-generated patches, but absolutely no sense in case when patterns are used from DVD :)
This is an option. If you go to the generator properties, on the first tab you will find a blanking checkbox. You can uncheck it if your PC screen light cannot disturb measures.

Regards
Georges

oomka

01-11-07, 09:30 AM

Hi everyone,

Can you take a look at my latest run and make suggestions (I can seem to get the gamma flat enough, also the primaries seem to be off). Any input would be appreciated.
Sony KDS-60A2000 with Spyder II.

Also am I using a correct gamma reference?(Display Gamma with Black Compensation (Default))

belmore

01-11-07, 09:59 AM

Love your software! I'm a bit of a nube, but I also own umr's AccuCal software. Both applications are simple to use and have very similar features and functions, but I'm preferring HCFR polished interface at this point. I guess the rule of "you get what you pay for" does not apply to HCFR!

At any rate, I have an Eye-One Design sensor. As I'm sure you know, the Eye-One cannot accurately read blacks below 30 IRE. Which gamma should I be using with this sensor? Using the default of Gamma with Black Level Compensation doesn't sound like the right one, given the Eye One probably cannot accurately measure the lowest black my TV can produce. Note that I am trying to calibrate both a Sony CRT and a Panasonic Plasma.

I've already done a calibration of the Sony CRT. It looks great with lots up "pop", however, it is a little dark. Any recommedations? Thanks.

HDholic

01-11-07, 10:01 AM

If it looks dark try the Camera Gamma. It should give better results.

laric

01-11-07, 10:51 AM

I guess the rule of "you get what you pay for" does not apply to HCFR!
That is really crazy thinking ! Do you really think Apache, Firefox, Linux and thousands of others... are dumb softwares ? :D

As for your gamma, as suggested by HDholic, try the Camera Gamma.

--Patrice ;)

Georges G

01-11-07, 11:00 AM

Patrice,

In your free software list, you forgot DScaler ;) Since one of DScaler authors is part of HCFR development team, you are unforgivable ! :D

Georges

laric

01-11-07, 11:21 AM

oh dear !! Sorry Laurent :D !

--Patrice ;)

Ungermann

01-11-07, 11:36 AM

If it looks dark try the Camera Gamma. It should give better results.What is the point of instrumental testing if at the end of the day you still verify it by eye? I can check the smoothness of the Gamma curve without targeting any specific curve. I thought the point is to hit the very specific curve, this is what this program and the probes are for. The question is: which curve is the one? The gamma question is getting mooter and mooter (so to speak).

richlo

01-11-07, 12:02 PM

What is the point of instrumental testing if at the end of the day you still verify it by eye? I can check the smoothness of the Gamma curve without targeting any specific curve. I thought the point is to hit the very specific curve, this is what this program and the probes are for. The question is: which curve is the one? The gamma question is getting mooter and mooter (so to speak).

Its dependent on display, I am believing right now that Camera Gamma is good for CRT displays, and that with DLPs (at least with mine), then the 2nd option.

HDholic

01-11-07, 12:54 PM

What is the point of instrumental testing if at the end of the day you still verify it by eye? I can check the smoothness of the Gamma curve without targeting any specific curve.
You know very well that Gamma can't be calibrated by eye. You need to measure against a reference.

I thought the point is to hit the very specific curve, this is what this program and the probes are for. The question is: which curve is the one?
Now this is the "right" statement ;) .

In all measurements I've done including my Direct View CRT, RP CRT and LCD monitor, I find "Camera Gamma" best to my eyes. Using "Disp Gamma w/ BL Comp." results in bringing down brightness and contrast too much, giving me a dull picture.

Ungermann

01-11-07, 01:58 PM

What is the point of instrumental testing if at the end of the day you still verify it by eye? I can check the smoothness of the Gamma curve without targeting any specific curve.Now this is the "right" statement ;) .

In all measurements I've done including my Direct View CRT, RP CRT and LCD monitor, I find "Camera Gamma" best to my eyes. Using "Disp Gamma w/ BL Comp." results in bringing down brightness and contrast too much, giving me a dull picture.Aren't you saying the same thing you said several postings before? Meaning that your measurements look differently if you compare them to different curves, and that you arbitrarily chose the curve which seems better for you. If you are still relying on your eyes (you said that you find "Camera Gamma" best to your eyes), what is the point of having several target curves at all?

Currently there is choice of four target gamma curves, all corresponding to the same gamma value. You are choosing between them and are deciding which of the four is best for you. Now increase number four to infinity, and choose any curve you like, it just has to be smooth enough, without clipping or having a weird shape like an S-curve, and then simply assess with your eyes whether it looks good enough.

Hence my question, again: if you are assessing the result with your eyes, what is the point of targeting a specific curve?

Update: Now, after thinking about it, will it make sense to display a target area instead of one target curve? HCFR program could draw an area between the "straightest" and "brightest" curve and up to the most "curvaceous" one, for the same Gamma value. Then, if an actual measurement fits inside this area, it is considered acceptable, and the decision on whether the curve is "good" or not will be strictly in the eye of beholder so to speak.

richlo

01-11-07, 02:39 PM

assessing the result with your eyes, what is the point of targeting a specific curve?

Update: Now, after thinking about it, will it make sense to display a target area instead of one target curve? HCFR program could draw an area between the "straightest" and "brightest" curve and up to the most "curvaceous" one, for the same Gamma value. Then, if an actual measurement fits inside this area, it is considered acceptable, and the decision on whether the curve is "good" or not will be strictly in the eye of beholder so to speak.

well said - shorter version - NUMBERS ARENT EVERYTHING.

Most pros use their eyes for final tweaking..especially many carry Optical Comparators that they make for final tweaking of greyscale.

laric

01-11-07, 02:45 PM

I'll try (again) to explain a bit later, but as (shortly) explain with the 1.21 introduction, we propose different method to compute and "display" the gamma...

Obvisouly (well, we thought it was obvious) the target should not be the sames, chosing a method implies you know why and you chose your target accordingly.
As said, some methods are highly dependant on the quality of samples (measures) especially in dark area... And so depends on probes (they are not equals) and also projectors...

--Patrice

Whisper2

01-11-07, 03:12 PM

I'll try (again) to explain a bit later, but as (shortly) explain with the 1.21 introduction, we propose different method to compute and "display" the gamma...

Obvisouly (well, we thought it was obvious) the target should not be the sames, chosing a method implies you know why and you chose your target accordingly.

--Patrice

I think that it's well as this it's now. Each one choose the option that he prefers, knowing the advantages and disadvantages that they have

Regards

rsuber

01-11-07, 03:14 PM

Knowing that CRT's normally have a gamma of 2.2, which setting corresponds to the 2.2 gamma curve? This would seem to be the correct choice for a RPTV.

Thanks
Rick

laric

01-11-07, 03:47 PM

There is two things, first the target Gamma, that you set in the box, we set it to 2.22 by default, but you can change it. Then the displayed curves, they all use a different method of computing the gamma...

BTW, CRTs are not 2.2, they are 2.5 from factory (Phosphore), then electronic may tune them differently.

--Patrice

Xander

01-11-07, 03:49 PM

lovingdvd

01-11-07, 04:10 PM

I'm going to borrow a Spyder 2 Express edition from a friend and try to do a calibration on my Epson TW700 using HCFR.

From what I've read so far, it appears that I need to follow the following steps:
1) Set colour mode to Theater Black 1 or 2
2) Set Iris to ON
3) set brigthness and contrast using Avia
4) make multiple runs using HCFR

Please let me know if I have it wrong. I don't suppose there is a simple english list of steps on how to run the HCFR software yet?

TIA

Regarding "set iris to on" - I'm not sure if your pj has a dynamically adjusting iris based on the APL of the scene, but if so your gray scale calibration may be very tricky. If thats the case the easiest thing to do is to just use full fields (not windows) for all your gryascale measurements. Once you get that set best as possible check it again with the windows and it will likely have moved, possibly quite a bit. You'll need to go back and forth like this until you find a balance where the grayscale works well with both the full field and windows.

rsuber

01-11-07, 04:14 PM

lovingdvd

01-11-07, 04:29 PM

HCFR team - would it be possible to make the offset used in your new Gamma #3 calculation configurable by the end user? For instance, some people like to use an offset that is derrived back on the on/off contrast capabilities of the projector. For example the author of the contrast calculator suggests I use an offset of 0.02 instead of the hard coded 0.099 to calibrate a pj with 12,000:1 on/off CR.

Is this offset used configurable via the ini file? If not could it be added?

Also while we are discussing features related to gamma - it would be a nice touch if not too difficult to add some small text in the gamma graph graphic (perhaps centered at the top) which shows what the target gamma and actual measured gamma was, i.e. "Target gamma: 2.20, Actual 2.18". Thanks.

Xander

01-11-07, 04:36 PM

Regarding "set iris to on" - I'm not sure if your pj has a dynamically adjusting iris based on the APL of the scene, but if so your gray scale calibration may be very tricky. If thats the case the easiest thing to do is to just use full fields (not windows) for all your gryascale measurements. Once you get that set best as possible check it again with the windows and it will likely have moved, possibly quite a bit. You'll need to go back and forth like this until you find a balance where the grayscale works well with both the full field and windows.Yes, the pj has a dynamic iris, so I will switch it on and follow your advice above. Thanks for the reply.

laric

01-11-07, 05:14 PM

lovingdvd

01-11-07, 05:20 PM

lovingdvd, what you suggest (programmable offset) was in v1.21 betas (with a lot more gamma options) but we decided it was too advanced option.
Btw, it was removed in final... But could re-appear at some stage...

--Patrice

Thanks for letting me know. Would you kindly consider adding it back in? Perhaps it could just be a hidden option that only those with the advanced needs to control this would ask about and prevent others without the need from messing things up. If you'd like I can point you to several references and discussions within AVS that talk about the importance of this. In my case I need to use an offset of 0.02 so you can see how a fixed offset of 0.099 would be way off from the one I need. Thank you for the consideration.

HDholic

01-11-07, 06:03 PM

Aren't you saying the same thing you said several postings before? Meaning that your measurements look differently if you compare them to different curves, and that you arbitrarily chose the curve which seems better for you. If you are still relying on your eyes (you said that you find "Camera Gamma" best to your eyes), what is the point of having several target curves at all?
I think you are missing the point. Yes we visually judge the end result and determine if we like what we see, BUT we NEED to calibrate to a "reference", otherwise is like driving at night with your eyes closed. Why did I choose a particular reference? My choice was based on endless reading of references about the subject. I don't think it's pure coincident that my choice worked best. Let's not forget that "Video Professionals" are the ones that created the reference that when used should bring out all detail and accuracy there should be. The problem comes when more references are created to "improve :confused: " on the original reference, ending up w/ countless of them(not referring to a particular program). We would live in a better world if the industry came up with an Universally recognized reference, avoiding confusion ;) . Will that ever happen? Don't count on it.

Currently there is choice of four target gamma curves, all corresponding to the same gamma value. You are choosing between them and are deciding which of the four is best for you. Now increase number four to infinity, and choose any curve you like, it just has to be smooth enough, without clipping or having a weird shape like an S-curve, and then simply assess with your eyes whether it looks good enough.
I don't agree w/ that. First, all choices use different calculation methods to come up with a gamma value even though they all target the same "number". If you don't know how/which to use, you won't have a proper calibration. Second, you could have a perfect curve matching any of "infinity" references and "think" that everything looks good but that doesn't mean that the display is performing the best it can, again w/out proper reference.

Hence my question, again: if you are assessing the result with your eyes, what is the point of targeting a specific curve?
I think I already answered that :).

Ungermann

01-11-07, 06:29 PM

I think you are missing the point. Yes we visually judge the end result and determine if we like what we see, BUT we NEED to calibrate to a "reference", otherwise is like driving at night with your eyes closed.What is the point of calibrating to a "reference" if you might not like the result? Such "reference" is worthless.
If you don't know how/which to use, you won't have a proper calibration.Does this mean that if you do know which one to use, you will get a proper calibration? Then I will ask again: why judging the result visually if calibration is supposed to be scientifically proper?
Why did I choose a particular reference? My choice was based on endless reading of references about the subject.I can accept this approach. Actually, this is what I hoped for. Here is how my TV works, here is how my DVD player works, this is why I chose this target Gamma with this particular corrections, and this is my result. If I don't like it - I just should suck it up and go visit an eye doctor. If I start questioning my measurements by visually judging the resulting picture then these measurements and adjustments become worthless.

Please correct me if I am not getting something important yet simple.

HDholic

01-11-07, 06:35 PM

The biggest problem is that no two persons see one thing the same way. Maybe we should all go to the doctor....Nahh, let's all have a beer :D !!!

richlo

01-11-07, 07:29 PM

The references are simply guidelines to follow and you have the ability to change what you like and dont (most calibrators sway in their own way from them). Calibrations are always redefining themselves as we go, what more, is that with the new technology, newer and newer ideas are coming, even old ideas are being thrown out for something new and more advance...Calibrations, if it was that easy just to use the eye, someone will just simply PLUG in the same numbers for the same set, but as we know - THAT DOESNT WORK..wish it did though..

No carpenter works the same way, they may do the BASICS the same but some approach the more difficult work in a different manner. One is not always better than another, the end results is getting what you FEEL is right. Heck there is a thread right now going on someone who paid a calibrator, and has placed stop payment on his check because he didnt like what he saw...AFTER the calibrator left, not while he was there...so it goes to show that even if the numbers are right, they really dont mean squat if you dont like it, so you tailor it to your liking - in this case to the customers desire...

..2.2 - 2nd option, 2.2 3rd option, heck - 1st option too...whatever you calibrated, see what you like - its your display, and HCFR gives you the flexibility as a guideline, if you want to follow, if not, modify, experiment, learn...I can argue that 2.2 3rd option is wrong, but clearly several here think its right...and it is right for their given display, dont let the program fool you - let your eyes be a guide also..Dont worry too much about the references...they are their to get you close to standards that you can change to your likings.

good luck with this..it should be more fun than worrying about which one is the SUPPOSE correct setting...give them all a try...

laric

01-11-07, 07:45 PM

...here is how my DVD player works, this is why I chose this target Gamma with this particular corrections, and this is my result. If I don't like it - I just should suck it up and go visit an eye doctor.You don't have to go to the doctor ;)

I think the important point you may miss is that you can perfectly calibrate a set to Gamma 2.22 and D65 and have a poor image !

This for two main reasons, first the tools can fool the drawings (in a certain way), especially low levels, and as you may understand can way impact gamma.
Second, and main important one, what you try to acheive is get the best of your set, especially the best Contrast Ratio... This could imply you distord some settings... What the point to have all fine but a 200:1 CR ? ;)

--Patrice

Athlonstein

01-11-07, 10:09 PM

Hughman

01-11-07, 11:49 PM

I'm very tired and a inebriated so I'll keep this short and very simple. I spent the last two nights calibrating my sp4805 and D1 player with the HCFR program and an Eye-one LT meter. After going through more adjustments than I care to mention I've ended up with the following.

1- eyeone display LT relatively close to lens facing lens with diffusor = good, no diffusor facing lens= bad.

2- eyeone displayLT close to screen facing screen sans diffusor (similar results as facing lens wth diffuser) = good, with diffuser = bad

3- My 4805 with a UHP lamp, contrary to popular experience, is blue deficient not red deficient.

4- My 4805 measured closer to 5500k with default PC offset and gain settings using default 6500k temp, (28.5/58)

6- After way too much trial and error I arrived at the best D65 grayscale beginning with the 9500K default temp and adjusting from there.

7- Primaries are typical of what i expected for the 4805 (similar to cinema4home published results) but the secondaries measure off and appear as they measure but the D1 has no adjustments for color or tint nor does the 4805 when using DIVI input so I'm stuck with what it is????

Edit: I just realized I believe I measured the primaries with 100% fields and the secondaries 75%, not sure if this would cause my color issues but will try again tonight using 75%/75%.

7- Using DVE and thier 0, 20, 40, 60, 80, 100% fields the D65 result are damn near perfectly stacked the larget Delta E being .5. :D

8- I'm placing considerable trust in the accurracy of the equipment but subjectively the calibrated result looks outstanding, but how would I really know and how does anyone really know! I presume the more you spend coupled with increasing experience the better the measured, therefore, perceived result.

I'm happy for a few more days. :)

Lyckman

01-12-07, 07:48 AM

Uh guys..

What happened to the PAL colorspace reference? I'm pretty sure it was there in a previous version...

http://www.immortal.se/pics/HCFR-References.jpg

According to wikipedia (http://en.wikipedia.org/wiki/RGB_color_space) the green coordinates differs from REC709/sRGB...

// Lyckman

zancid1

01-12-07, 09:23 AM

Questions:

I'm connected via VGA port to my projector from my HTPC. What color space should I use? HD709 or sRBG.

and another follow-on similiar in a way to Hughs, when connected via VGA I only have RGB controls in the user menu no color and tint. Extremes of the RGB controls make negligable changes to the gamut...am I stuck with the gamut I get (not that it's bad just peaks outside the reference gamut on green corner.

I see the occasional post for pointing at the projector lens versus the screen. I thought the screen was what you needed to us as it in a sense is part of the display system.

Cheers

Whisper2

01-12-07, 11:21 AM

Lyckman

01-12-07, 12:38 PM

This makes no sense? (http://www.avsforum.com/avs-vb/showthread.php?p=9242382&&#post9242382)

The coordinates for PAL primaries are far from REC601....

// Lyckman

belmore

01-12-07, 12:48 PM

This question should probably go in its own thread, since I noticed this using both HCFR and AccuCal.

Does anyone else prefer D75 over D65? I'm finding calibrating to D65 results in an image with a yellowish cast (too warm). I've calibrated to D65 over and over, changing the gamma settings, etc. and I always end up with a warm image that leaves me thinking "that doesn't look quite right". Note that I'm calibrating a Sony CRT with an Eye One sensor. Any thoughts?

greeno

01-12-07, 01:11 PM

Belmore,
Not all folks like D65 compared to some other white point. But D65 is what DVD's and NTSC content is referenced to.

Is your i1 still in calibration spec? they do drift over time and might this be the case?

Best,
jeff

laric

01-12-07, 01:18 PM

Lyckman, PAL/SECAM are more than close to REC709 (look at your link) ;)

But we may bring them back just to avoid confusions...

--Patrice

Whisper2

01-12-07, 01:30 PM

This makes no sense? (http://www.avsforum.com/avs-vb/showthread.php?p=9242382&&#post9242382)

The coordinates for PAL primaries are far from REC601....

// Lyckman

Yes, you're right ( they are very near from REC 709, but green is different and secondary are different too)

Laric, I've got a question about REC709 and EBU 3213 (PAL) primary/secondary colors.

What are the coordinates of the primary colors in your DVD ( version 1.1.15) ? Are they Rec709 or EBU 3213 standard?

and If I use the option "generator" (PC)?

and what about secondary?

Thanks in advanced

belmore

01-12-07, 01:50 PM

Is your i1 still in calibration spec? they do drift over time and might this be the case?

I suspected this and contacted GretaMacBeth's support. They told me that as long as my i1 passed the diagnostics, it should still be within spec. There's a calibration "plate" that comes with the design package that I guess contains a reference white. The diagnostic software reads this and supposedly tells you if the i1 is in need of re-calibration. My i1's certification did elaspe in Oct. 2006.

It is odd and I'm still not convinced there isn't something wrong here. I failed to mention that I'm calibrating the TV thru component from an XBox 360's DVD player using the HCFR DVD. Could the fact that this DVD is PAL encoded have anything to do with my problem? Could using the 360 introduce problems?

Also note that I am primarily watching recorded HDTV using the XBox 360's Media Center Extender capabilities. Does ATSC content also use D65?

Lyckman

01-12-07, 01:54 PM

Lyckman, PAL/SECAM are more than close to REC709 (look at your link) ;)

But we may bring them back just to avoid confusions...

--Patrice
hehe, yes I know but that's not the point. Please bring them back! :D
For myself, I don't give a damn, but for people comparing HCFR to other soft a small error like this might make the whole difference.... I'm just guarding your back. ;)

Ungermann

01-12-07, 01:55 PM

output level and colorspace are two different things, hence they are independantly set...
I know that as a programmer you would like to have everything as flexible and configurable as possible, I like my software the same. On the other hand The Guide to Digital Television (http://digitaltelevision.com/publish/dtvbook/ch2f.shtml) states:

The official sampling standard for standard definition television is ITU-R 601 (short for ITU-R BT.601-2, also known as "601"). ... Using 8-bit samples, ITU-R 601 places black at value 16 and white at value 235. This permits values below black and above white; these values do not exist in RGB signal representations that place black at value 0 and white at value 255. ITU-R 601 also permits luminance sampling with 10-bit values (1,024 levels); this improves dynamic range and conversions to 8-bit RGB values for processing and display. In 10-bit, black is at value 64 and white is at value 940.
My understanding is that the colorspace and output level are related by default. SD DVDs (both PAL and NTSC) seem to be 601-compliant and as such should use 16 for black and 235 for white by default.

What I mean is that the ability to override the default is nice, but if there are sensible defaults it would be great if they could be chosen as a package. Say, I would have a "packaged" setting like "Calibrate with DVD patterns, SD resolution". In this case application would default to ITU-R 601 with 16/235 settings. If someone needs to choose different output levels, that should be allowed as advanced option. Another "packaged" setting might be "Calibrate with DVD patterns, HD resolution", this would default to 709.

This quote (http://www.techweb.com/encyclopedia/defineterm.jhtml?term=ITU-RBT.601) states: "ITU-R 601 is a standard for component digital video (YCbCr)." As I understand this works for HDMI. What about analog component (YPbPr)? Is compliant with ITU-R 601 as well? More generally, all outputs besides RGB (VGA only?) are YUV-type outputs, and as such carry either YPbPr (analog) or YCbCr (digital) signal. Does this mean, that ALL connections (composite, S-video, analog component, HDMI) are ITU-R 601-compliant for SD, and 709-compliant for HD?

What does this quote mean in that case: HD is REC709, PAL is EBU 3213 and NTSC Rec601.Are ITU-R 601 and Rec601 the same things or not? if they are the same things, how Rec601 is NTSC only?

Sorry for stupid questions :)

audioholicJeffL

01-12-07, 02:27 PM

richlo

01-12-07, 02:44 PM

Jeff -

when was the last time you actually seen a calibrator using a filter..;0)...NEVER have I..they use the right instruments..plus Im kinda color blind..so that doesnt help..

audioholicJeffL

01-12-07, 02:49 PM

Jeff -

when was the last time you actually seen a calibrator using a filter..;0)...NEVER have I..they use the right instruments..plus Im kinda color blind..so that doesnt help..

That's what I was thinking but I was surprised to see just how different the settings were. Personly, with what I know now, I think I could set color and tint better just eyeballing a known scene like the chip chart lady on DVE rather than using those filters.

laric

01-12-07, 02:57 PM

For myself, I don't give a damn, but for people comparing HCFR to other soft a small error like this might make the whole difference.... I'm just guarding your back. ;)
Thanks for caring for our back ;)

That is likely to be back, so no more "noise" ;)

--Patrice

Ungermann

01-12-07, 02:59 PM

I find filters helpful and they seem to work, but... After I set saturation and tint using filters, I always tone down the color because it is usually way too oversaturated to my tastes. This ruins colors somewhat, but I prefer having slightly incorrect colors instead of tolerating almost blooming colors jumping on me.

Ungermann

01-12-07, 03:10 PM

Returning to gamma choice: W3C sRGB paper on camera gamma (http://www.w3.org/Graphics/Color/sRGB)
viewing gamma - the overall system gamma that we want to obtain and is typically computed by multiplying the camera gamma by the display gamma as shown below.

viewing gamma = camera gamma × display gamma,
where display gamma = CRT gamma × LUT gamma

Viewing gamma value of 1.125 is used instead of 1.0 is to compensate for the viewing environment conditions, including ambient illumination and flare.
...
ITU-R BT.709 ... contains camera gamma of 1.0/1.956. The actual exponent factor in the 709 transfer function is 1.0/2.222. Despite the fact that the exponent of the 709 function is 1.0/2.222, the actual 709 encoding transfer function is closer to a CRT gamma of 1.0/1.956 than 1.0/2.222. This is due to the large offset of 0.099 in the transfer function equation. This is well matched to the eye's own non-linearity and it helps minimize transmission noise in the dark areas.

Broadcast television camera gamma standards and the ITU-R BT.709 standard in particular defines the transformation ... [as] not an exact appearance match of the original scene, but instead is a preferred reproduction of the original scene that is consistent with the limitations of a monitor.

CJO

01-12-07, 03:11 PM

What I mean is that the ability to override the default is nice, but if there are sensible defaults it would be great if they could be chosen as a package. Say, I would have a "packaged" setting like "Calibrate with DVD patterns, SD resolution". In this case application would default to ITU-R 601 with 16/235 settings. If someone needs to choose different output levels, that should be allowed as advanced option. Another "packaged" setting might be "Calibrate with DVD patterns, HD resolution", this would default to 709.

I think that is a great idea- especially for those of us just starting out. You could have a startup screen where you can pick your type of display, the sensor you are using, whether you are using PAL or NTSC, and what your source is for the patterns. The software then would pick the appropriate settings. Down the road, it might then walk you through modifying the settings, similar to what the SpyderTV software does.

CJ

laric

01-12-07, 03:20 PM

...but if there are sensible defaults it would be great if they could be chosen as a package. Say, I would have a "packaged" setting like "Calibrate with DVD patterns, SD resolution". In this case application would default to ITU-R 601 with 16/235 settings. If someone needs to choose different output levels, that should be allowed as advanced option. Another "packaged" setting might be "Calibrate with DVD patterns, HD resolution", this would default to 709.
Sure that make sense, and is exactely what is done by default in ColorHCFR.
We assume that people that chose say REC601 and 0-255 levels knows what they are doing... :D
But you"re right, some may don't know, hence future experts/beginners mode we have planned.

What about analog component (YPbPr)? Is compliant with ITU-R 601 as well? More generally, all outputs besides RGB (VGA only?) are YUV-type outputs, and as such carry either YPbPr (analog) or YCbCr (digital) signal. Does this mean, that ALL connections (composite, S-video, analog component, HDMI) are ITU-R 601-compliant for SD, and 709-compliant for HD?
Basically no, composite/s-video is a whole different story as PAL/NTSC stuff apply there... But practically yes, nowadays we are all using component or digital connections, aren't we ?
Also, don't forget take in account what your device(s) outputs and how your projector interpret it... May sounds silly, but do you that some projectors automatically default to REC709 if you send 720p (even if the sources is sending REC601 in 720p) !!

What does this quote mean in that case: Are ITU-R 601 and Rec601 the same things or not? if they are the same things, how Rec601 is NTSC only?
Well, dunno where you want we go there, but as you probably know we all tend to use PAL or NTSC terms to define what is not PAL or NTSC (except it may use the same color spaces)...
I assume what you call NTSC in our context is 480i/p (or other) 60Hz sent to projector via YUV/DVI/HDMi, then yes it is REC601.

That's why there is no point in putting back PAL/NTSC nothing there, it is "SD (REC601)" or "HD (REC709)" nothing else...

Your questions are certainly not stupids :)

--Patrice

Lyckman

01-12-07, 03:40 PM

That is likely to be back, so no more "noise" ;)lol, that sais it all! Keep up the good work! :D

// Lyckman

jvincent

01-12-07, 05:38 PM

Woot! EyeOne Display LT was waiting at home when I got back from work.

Looks like I'll be busy tonight. :)

Mist8rioso

01-12-07, 05:41 PM

Information I found about gamma:
In all broadcast video systems, gamma correction is done in the camera. This choice was made in the days when television electronics were all analog, and a good gamma-correction circuit was expensive to build. The original NTSC video standard required cameras to have a transfer function with a gamma of 1/2.2, or about 0.45. Recently, a more complex two-part transfer function has been adopted [SMPTE-170M], but its behavior can be well approximated by a power function with a gamma of 0.5. When the resulting image is displayed on a CRT with a gamma of 2.5, the image on screen ends up with a gamma of about 1.25 relative to the original scene, which is appropriate for "dim surround" viewing.
So wouldn't "Camera Gamma" setting be the more accurate?

carillon

01-12-07, 06:31 PM

I have a Spyder 2 and would like to know how close to the screen should it be when taking measurements. My TV is a Sony KDS50A2000 SXRD which is their version of LCoS rear projection. I started experimenting last night with HCFR and would like to try to get my set calibrated this weekend.

Thanks!

richlo

01-12-07, 07:44 PM

primetimeguy

01-12-07, 08:02 PM

To show what the different devices do with gamma, here are two of my files. The first (initial) is with my old DVD player and the second file (oppo) is my new Oppo 970 where I just disconnected the old player and plugged in the new one. Both connected via component at 480p. So adjusting for one component in your system may not be good for all.

HDholic

01-12-07, 08:15 PM

Looking at the "Y" value shows that overall brightness has decreased. Have you checked brightness/contrast levels with pattern?

primetimeguy

01-12-07, 10:46 PM

Looking at the "Y" value shows that overall brightness has decreased. Have you checked brightness/contrast levels with pattern?

Yes, brightness levels were much different between the two looking at test patterns. I have since adjusted accordingly. When you say "Y", what are you looking at, the xy"Y"?

HDholic

01-12-07, 10:54 PM

You look at Y(luminance) in the IRE table. Capital Y if xyY is selected. What is your new gamma value after adjustments?

primetimeguy

01-12-07, 11:13 PM

You look at Y(luminance) in the IRE table. Capital Y if xyY is selected. What is your new gamma value after adjustments?

Here is what I ended up with (I think, so darn many files playing around with this). Gamma around 2.5. Looking at the log curve I still have a slope from left to right. Is this pretty normal or should I be able to do better? I do have a gamma setting in the service mode but all that does is make the slope steeper from left to right and my gamma changes from 2.5 to 2.35 to 2.2 or so with the other settings.

HDholic

01-12-07, 11:26 PM

You want it as flat as possible. Make it steeper and you'll loose black detail. You want to calibrate to a setting that gives you deep blacks w/ good shadow detail, and high contrast as possible. What are you calibrating?

primetimeguy

01-12-07, 11:39 PM

primetimeguy

01-12-07, 11:44 PM

And is this brightness/gamma issue all a compromise? My gamma is now just above 2.5 and by increasing brightness a couple clicks I can bring this down closer to 2.5 but lose contrast ratio. And currently the brightness is set about right using test patterns, maybe even a couple clicks too bright. Just wondering how I get that final number. By eye? Like you said, adjust so not too tak, still have black detail and maintain good CR.

HDholic

01-13-07, 12:06 AM

Working on RP CRT. And I agree as flat as possible but in the real world how flat is that? I have a drop of about .2 from left to right and wondering how typical that is? Pretty good? Or should I be able to get better.? Also wondering how accurate those lower measurements are 0-20IRE using the DTP-94 probe.
I think it's actually very good. Much better than what my set can do :) . See if you can adjust R,B contrast to come closer to the reference line and adjust green as needed. The DTP94 is quite good at low levels, increase the read time and see if it improves.

And is this brightness/gamma issue all a compromise? My gamma is now just above 2.5 and by increasing brightness a couple clicks I can bring this down closer to 2.5 but lose contrast ratio. And currently the brightness is set about right using test patterns, maybe even a couple clicks too bright. Just wondering how I get that final number. By eye? Like you said, adjust so not too tak, still have black detail and maintain good CR.
Sometimes it can come down to compromising something. For example, some sets may give good black detail at expense of black level or vice versa, it comes down to preference. Technically CRT have a gamma of 2.5, unless circuitry has been built in to correct to 2.2. In your case, it looks like calibrating to 2.5 is your best choice. Set your contrast/brightness w/ pattern and adjust gamma if necessary as close to 2.5 as possible.

In my case, Hitachi RP CRT, I have to calibrate to 2.2 and even then my S2 can't even get a reading at 0IRE most of the time. I have the 971 thru HDMI.

primetimeguy

01-13-07, 12:13 AM

Thanks for the help HDholic. I forgot about the longer read times at lower levels so I'll try that first and see if I see anything different.

carillon

01-13-07, 12:19 AM

HDholic

01-13-07, 12:20 AM

HDholic

01-13-07, 12:53 AM

Well, I just finished my first run with HCFR. I really feel like I don't know what I'm doing but here's what I ended up with. Based on what you see, what adjustments do you suggest?

I used a Spyder 2 as my probe.
You did a good job on grayscale ;) . If your set has color decoder controls then you'll benefit from that. Also, your gamma needs a lot of work :( . It's a lot harder when using dynamic contrast. Look at Logarithmic Gamma mode (right click on gamma window and you'll see it) and adjust gamma settings to get as close as possible to a 2.2 reference.

*Make sure you select "Extend Read Time or Average many reads on dark measurements", whichever one gives better results.

carillon

01-13-07, 01:03 AM

You did a good job on grayscale ;) . If your set has color decoder controls then you'll benefit from that. Also, your gamma needs a lot of work :( . It's a lot harder when using dynamic contrast. Look at Logarithmic Gamma mode (right click on gamma window and you'll see it) and adjust gamma settings to get as close as possible to a 2.2 reference.

Thanks for the reply... I don't believe my set has color decoder controls in the user menus. Now about gamma, I couldn't seem to find a way to get closer to 2.2 without really losing a lot of contrast. I was clicking the brightness setting up to try and get closer to 2.2 and did, but lost a lot of contrast in the process. My brightness is set with Avia and correct according to the pattern.

Which window is the gamma window... I don't see one labeled as such. What do you mean by dynamic contrast. Oh so much to learn but FUN!

richlo

01-13-07, 01:49 AM

JariT

01-13-07, 05:10 AM

How do you set DTP-94 for longer read time for low ire levels?

ssj2

01-13-07, 08:41 AM

I set up a near perfect CIE chart on my panny projector. Picture now looks incredible. I was finaly able to calibrate the normal mode which has higher lumen output. Just for giggles I popped in DVE and checked the colors with the red, green and blue filters. Not even close to being the same. I trust the software and I suspect that the color filters are just not accurate on a front projector or maybe not accurate at all. Has anyone else seen this?

The CIE chart is measuring the saturation of each color. Inside th triangle and they're under saturated, outside and they're over saturated.. This is different than whether or not intensity (set with the color control) is to low or high.

I can get a nearly perfect CIE result on my Hitachi CRT RPTV using just HCFR. However, color intensity is too low and it shows on screen. If I use the filters (DVE, NOT AVIA), I get perfectly balanced color and tint. How do I know? By covering the red and green CRTs and viewing the blue color decoder bars. Blue only is the proper way to align color and tint. The CIE chart is slightly off on green now, but overall the image is much better.

I'm about to re-do my Z4, for which I had also previously relied only on the CIE chart, and will compare my results.

ssj2

01-13-07, 08:41 AM

How do you set DTP-94 for longer read time for low ire levels?

Set to "average read times at lower levels" and you'll get the best results.

primetimeguy

01-13-07, 09:25 AM

The CIE chart is measuring the saturation of each color. Inside th triangle and they're under saturated, outside and they're over saturated.. This is different than whether or not intensity (set with the color control) is to low or high.

So what you are saying is the color control should not be used to make changes on the CIE chart since that affects intensity? Other adjustments should be used to try and adjust the CIE triangle points?

primetimeguy

01-13-07, 09:29 AM

primetime-

I'm also wandering why your contrast is so low. CRTs are capable of 10,000+. Since the Spyder2 has a black limit of 0.02 cd/m^2 and can't read my 0IRE level, my set is close or exceeding that.

Contrast has been difficult for me to set using DVE test patterns. I don't see any blooming and I always see more than white signals. I feel I get it some what close and that look and various test patterns and adjust so white "looks" white as well as make some adjustments while watching programming. I'd agree it might be a couple clicks low at this point. Any ideas for me to set it better.

lovingdvd

01-13-07, 10:35 AM

Contrast has been difficult for me to set using DVE test patterns. I don't see any blooming and I always see more than white signals. I feel I get it some what close and that look and various test patterns and adjust so white "looks" white as well as make some adjustments while watching programming. I'd agree it might be a couple clicks low at this point. Any ideas for me to set it better.

Yes. When white does not clip on a digital display, the proper way to set it is to use a stair step type of pattern that has white at different levels and goes beyond 100 IRE to brighter than white.

There is a pattern like this on DVE (can't recall where or which) but also on GetGray (I think its on the very first pattern under brightness and contrast).

The way you work it is that you start with the contrast down a bit, and start raising it up. As you raise it, watch for a color shift in the blocks that are past 100 IRE white (digital 135). When the setting gets too high, that value will cause a shift in color > 135. That is how you know its too high. Back it off until there is no color shift in white and that is your ideal point.

primetimeguy

01-13-07, 10:41 AM

Yes. When white does not clip on a digital display, the proper way to set it is to use a stair step type of pattern that has white at different levels and goes beyond 100 IRE to brighter than white.

There is a pattern like this on DVE (can't recall where or which) but also on GetGray (I think its on the very first pattern under brightness and contrast).

The way you work it is that you start with the contrast down a bit, and start raising it up. As you raise it, watch for a color shift in the blocks that are past 100 IRE white (digital 135). When the setting gets too high, that value will cause a shift in color > 135. That is how you know its too high. Back it off until there is no color shift in white and that is your ideal point.

Right, but I have CRT RP - analog.

primetimeguy

01-13-07, 10:44 AM

So what you are saying is the color control should not be used to make changes on the CIE chart since that affects intensity? Other adjustments should be used to try and adjust the CIE triangle points?

I think I found the answer to my own question. Best way to set color/tint is use this type of pattern on GetGray and turning off red and green CRTs. I did this and it brought my color setting up about 6 clicks from where it was when I adjusted using the CIE triangle. Now I'll go back and check everything with HCFR. But then are my color points just where they are and that is what I get on the CIE?

HDholic

01-13-07, 10:44 AM

Contrast has been difficult for me to set using DVE test patterns. I don't see any blooming and I always see more than white signals. I feel I get it some what close and that look and various test patterns and adjust so white "looks" white as well as make some adjustments while watching programming. I'd agree it might be a couple clicks low at this point. Any ideas for me to set it better.
It's almost impossible to get rid of the WTW signals, unless increasing Contrast too much. But it's better to have them anyways for extra detail on white scenes, ect.

I'm also referring to on/off contrast. Your 0IRE level reads 0.100 when CRTs are capable of absolute black. That's why you're getting a low on/off contrast. Make sure your brightness setting is right.

I think I found the answer to my own question. Best way to set color/tint is the this type of pattern on GetGray and turning off red and green CRTs. I did this and it brought my color setting up about 6 clicks from where it was when I adjusted using the CIE triangle. Now I'll go back and check everything with HCFR.
More often than not, color/tint controls are often mistanken as controls to adjust decoder errors. Wrong, you need DECODER controls to adjust CIE gamut.

primetimeguy

01-13-07, 10:47 AM

I'm also referring to on/off contrast. Your 0IRE level reads 0.100 when CRTs are capable of absolute black. That's why you're getting a low on/off contrast. Make sure your brightness setting is right.

So brightness is a bit high is what you are saying? If so, I agree it is because I set it using test patterns and then went up a few from there for daytime viewing.

HDholic

01-13-07, 10:52 AM

So brightness is a bit high is what you are saying? If so, I agree it is because I set it using test patterns and then went up a few from there for daytime viewing.
If your set allows for day and night settings then you should adjust for both. Adjust correctly with pattern again and see if your on/off improves.

richlo

01-13-07, 10:54 AM

Right, but I have CRT RP - analog.

For CRT, Id used AVIA or DVE to set your contrast, then you will increase contrast until you see BLOOMING, then back down before that point.

You can also use THX Optimizer that is in the Pixar movies, supposedly there are some differences, but nevertheless its a very good start.

lovingdvd

01-13-07, 10:56 AM

What are your guys thoughts on splitting the discussions we've been having here into two threads?

I'm thinking it could be beneficial to keep this thread focused on the HCFR software itself, for discussions about how to use the software, bug reports, feature requests, installation matters, notices/updates from the developers, and "what does this switch do" type of questions.

Then we can use another thread for discussion about how-tos and matters more related to calibration techniques and findings themselves which are independent of which calibration software is being used.

Just wanted to float this out there.

carillon

01-13-07, 10:57 AM

carillon-

what did you select for gamma on your set?? is it actually numbers or is it like something has a name for each gamma, ie, Movie, Sports, Video, etc... Right now you have a S type of gamma, there is probably another one that will give you a nicer curve..

btw..1st attempt looks good, really good. Your color decoder is not close to being accurate, read up on your model to see if they have adjustments in the service menu, if not your stuck with what you got in the user menu, just eyeball it to taste, if you havent tried to use some calibration disk..

btw..what I think he means by DYNAMIC CONTRAST, its basically your setting...

Like gamma has modes, contrast settings sometimes have HIGH, MID, LOW, or VIVID, DYNAMIC, NIGHT modes..these are in the USER menu...it is always best to set your modes to a more realistic mode - so stick with MID, theater mode, etc...or whatever they call it on your user menu..

Hi richlo... thanks for the reply.

On my set there is a STANDARD, VIVID and CUSTOM mode. I chose the CUSTOM mode since it gives me access to all available controls.

Here are the settings I am using for each.

Display type: Sony KDS50A2000 SXRD rear projection LCoS

Setup

Picture
Picture Mode = Custom
Advanced Iris = Auto 1
Picture = Max
Brightness = 38
Color = 50
Hue = 0
Color Temp = Warm 2
Sharpness = 35
Noise Reduction = Off

Advanced Settings
Black Corrector = Off
Gamma = Off
Clear White = Off
Live Color = Off
Detail Enhancer = Off
Edge Enhancer = Off

White Balance
R-Gain = 0
G-Gain = 0
B-Gain = -8
R-Bias = 0
G-Bias = 1
B-Bias = -1

Power Saving = Off

What chart are you looking at to see an S shaped gamma curve? As for color decoder, what are you looking at to see that it is not accurate? When I check it using Avia's patterns and filters it looks dead on. Hmmm. I do have one setting that is labeled Color Matrix. It is set to custom which means that 480 content uses ITU601 and HD content uses ITU709. This feature is described as allowing you to select the method for reproducing color-difference signals.

Anyway, I'm using Avia to generate all of the patterns I'm using. Perhaps I'm not using the correct ones or in the correct way. For example, when I was measuring primaries and secondaries, I chose the color fields pattern which filled the screen completely.

Any other A2000 owners please chime in with your results. Please include your HCFR file and settings used so I can compare.

Thanks!!

primetimeguy

01-13-07, 10:57 AM

If your set allows for day and night settings then you should adjust for both. Adjust correctly with pattern again and see if your on/off improves.

It definitely improves as I have seen this over the many iterations I have done. Once I feel I really have a grasp for all of these tools/settings I may try setting up day/night settings using Movie/Pro since I do not have the day/night option.

Thanks for all of the help guys. This is one of the most informative and helpful threads around here.

primetimeguy

01-13-07, 10:59 AM

For CRT, Id used AVIA or DVE to set your contrast, then you will increase contrast until you see BLOOMING, then back down before that point.

I've been using DVE and maybe my eyes just don't see it, but I don't see any real blooming. If I do it is a such a gradual thing that I have a range of about 10 clicks I'd say is probably good. This puts it around the 30-40% of full scale setting.

primetimeguy

01-13-07, 11:02 AM

richlo

01-13-07, 11:04 AM

Hi richlo... thanks for the reply.

On my set there is a STANDARD, VIVID and CUSTOM mode. I chose the CUSTOM mode since it gives me access to all available controls.

Here are the settings I am using for each.

Display type: Sony KDS50A2000 SXRD rear projection LCoS

Setup

Picture
Picture Mode = Custom
Advanced Iris = Auto 1
Picture = Max
Brightness = 38
Color = 50
Hue = 0
Color Temp = Warm 2
Sharpness = 35
Noise Reduction = Off

Advanced Settings
Black Corrector = Off
Gamma = Off
Clear White = Off
Live Color = Off
Detail Enhancer = Off
Edge Enhancer = Off

White Balance
R-Gain = 0
G-Gain = 0
B-Gain = -8
R-Bias = 0
G-Bias = 1
B-Bias = -1

Power Saving = Off

What chart are you looking at to see an S shaped gamma curve? As for color decoder, what are you looking at to see that it is not accurate? When I check it using Avia's patterns and filters it looks dead on. Hmmm. I do have one setting that is labeled Color Matrix. It is set to custom which means that 480 content uses ITU601 and HD content uses ITU709. This feature is described as allowing you to select the method for reproducing color-difference signals.

Anyway, I'm using Avia to generate all of the patterns I'm using. Perhaps I'm not using the correct ones or in the correct way. For example, when I was measuring primaries and secondaries, I chose the color fields pattern which filled the screen completely.

Any other A2000 owners please chime in with your results. Please include your HCFR file and settings used so I can compare.

Thanks!!

Picture = MAX!!! :eek: ..this is your Contrast and it should never even come close to MAX setting , bring that baby down (start at 50%)..
Avia is not the greatest to set digital displays but use the typical contrast setting, then look at the bottom, you should see two moving white bars, if you dont, lower your contrast setting until you see them both, bring up to the one on the right dissapears, bring down back to it barely reappears.

Not sure why you have selected GAMMA off, but turn it on to see what your response is..

for the GAMMA, look for an ICON that looks like a curve of 1/2 a football shape - its call LUMINANCE HISTOGRAM..you curve should look like the icon curve..

richlo

01-13-07, 11:09 AM

HappyFunBoater

01-13-07, 11:16 AM

HDholic

01-13-07, 11:20 AM

richlo

01-13-07, 11:35 AM

Whoa. I thought the "never set contrast to max" warning was only for CRT sets. I thought DLP, LCOS, plasma and LCD were immune to this. When I use Spyder or DVE to calibrate my TV, I get a final contrast value very close to max. Anything else doesn't look white enough. (And I have a Sony LCOS.)

Yes, for CRT you want to bring it down because of blooming and burning out your CRT..but still when did you ever hear to MAX out contrast in any display??Never..CONTRAST set high will show colorshifting in digital displays..a good pattern for brightness is with GetGray where they have 2 pattern over 235 for white, if maxed out, you will see color shifting, you bring that down until colorshifting is no longer there and/or the two bars dissapear in the background...Most people setup digital contrast high because they never see blooming which is not the way to setup digital displays..that is why you probably have yours close to maxed out..

DVD and AVIA, as stated before where geared for CRT displays, so even those patterns you have to modify on how to use it with Digital displays.

noizemaker

01-13-07, 12:08 PM

Hey guys. Ok , so i did my first run with my new Display 2 & HCFR. Grayscale & Gamma are really good, but does anyone know of certain settings i can use on a Sony LCD RPTV that can tame my way oversaturated Primary colors?

Thanks so much guys.
Carmine.

noizemaker

01-13-07, 12:17 PM

How can i attach my calibration file for you guys to check out? When i try to upload it keeps telling me invalid file type.

Thanks guys!
Carmine.

carillon

01-13-07, 12:36 PM

Picture = MAX!!! :eek: ..this is your Contrast and it should never even come close to MAX setting , bring that baby down (start at 50%)..
Avia is not the greatest to set digital displays but use the typical contrast setting, then look at the bottom, you should see two moving white bars, if you dont, lower your contrast setting until you see them both, bring up to the one on the right dissapears, bring down back to it barely reappears.

Not sure why you have selected GAMMA off, but turn it on to see what your response is..

for the GAMMA, look for an ICON that looks like a curve of 1/2 a football shape - its call LUMINANCE HISTOGRAM..you curve should look like the icon curve..

Well I guess I'm back to the drawing board. When I use the Avia DVD to set Picture I see both moving white bars at MAX. I read your post about color shifting. Is there a pattern in Avia that would be better to use for setting Picture?

I see what you mean now about my gamma curve compared to the reference line it crosses at about 65 IRE so compared to reference it is shaped slightly as an "S".

I will try setting the gamma to something other than OFF. Settings for gamma are OFF, LOW, MEDIUM, HIGH and MAX. Which one should I start with given my current reading?

I'm still at a loss as to why my colors are so oversaturated. Perhaps when I get the gamma down to 2.2, they will fall in line.

When I first started last night it was obvious that I would have to compromise contrast a great deal in order to get anywhere close to 2.2

Thanks for everyone's help. How about it A2000 owners, post your .chc files and settings!

richlo

01-13-07, 12:47 PM

Noizemaker
Zip the file...

richlo

01-13-07, 12:59 PM

Well I guess I'm back to the drawing board. When I use the Avia DVD to set Picture I see both moving white bars at MAX. I read your post about color shifting. Is there a pattern in Avia that would be better to use for setting Picture?

I see what you mean now about my gamma curve compared to the reference line it crosses at about 65 IRE so compared to reference it is shaped slightly as an "S".

I will try setting the gamma to something other than OFF. Settings for gamma are OFF, LOW, MEDIUM, HIGH and MAX. Which one should I start with given my current reading?

I'm still at a loss as to why my colors are so oversaturated. Perhaps when I get the gamma down to 2.2, they will fall in line.

When I first started last night it was obvious that I would have to compromise contrast a great deal in order to get anywhere close to 2.2

Thanks for everyone's help. How about it A2000 owners, post your .chc files and settings!

Start at Medium for Gamma..and this is just an guesstimate..

As far as what to use in AVIA, use a greyscale ramp, lower your contrast down to about 20%, then bring up and see if the white part of the ramp starts to display color shifting, if it does, then reduce until it does not. Other than that, you might need to get a little creative in figuring it out..I think GetGrey actually is an awesome disk to have for digital displays.

First thing to, just change your Gamma setting and rerun and see what you get, it is possible (not really sure) that you Max contrast is the correct setting..I doubt it but nevertheless make the easiest change and work from there

btw..when you see your LUMINANCE HISTOGRAM, right click the pattern and ask for LOGARTHIC, which is another way of viewing it, in addition, click red, blue, and green to view...

Log view will show a more precise look (almost like a zoom feature)