For projection CRT how do you position/use the Spyder sensor?

I've read posts that take some measurements facing the screen, some up close facing the lenes, and others far away at the screen facing the lenes.

And what about the honeycomb baffle, On or Off? :confused:

What is the best practice to get accurate and consistent readings for this consummer grade colorimeter. I have an 8" LC machine and am playing around with the HCFR software currently.

Good questions!!! I have a Spyder "in shipment" so these Q's will soon be pertinent to me.

However, it doesn't seem like you could use the sensor "up close facing the lenses". But beyond that, I'm all ears.

It seems like "facing the screen" would be the most accurate, as that is what we are looking at with our eyes.

There was a pretty long Spyder thread at one point, and IIRC the method described was at 30 or 45 degrees off-axis, pointing at the screen.

IF you face directly at the screen you WILL cause a shaddow of the unit and tripod - so thats OUT.

at 30-40 deg to the screen WOULD work round the shaddow issue.

My Sencore is setup to look at the projector from the SCREEN distance - to me that seems the best option, you are calibrating the PJ then.

YES I see that calibrating the reflection from the screen IS what our eyes see. But screens have different angle issues, and you would usually view directly at the screen NOT at 30-40 deg ???

I initially aimed at the screen assuming the matte white 1.3 gain would impact the readings. Because my PJ is ceiling mount down ~12 degrees I aimed up ~12 degrees at screen center (angle of incidence & reflection) instead of 45 degrees. In theory this should have provided the most intense light to the sensor. Unfortunately the readings were unusable.

I'm able to get readings positioned at the screen facing the PJ with and without the baffle. Are there potential issues with this method?

Hi,

Consider the PJ as level...........

Tubes angled at 12 deg.

light hits screen and is reflected.

HOWEVER the light is reflected back directly from the screen NOT angled down at 12 deg.

SO you colourimeter should have been in line with dead center of the screen.

This WILL create a shaddow.......

SO middle of the screen facing the PJ is my best suggestion , I dont have a spyer so cant be 100% sure - Is there a MANUAL ????

Sit it at the screen and point it at the projector.

I'm wondering if it would be possible to write software to attach the spyder directly to each lens in turn, as if it was a TV and adjust each colour accordingly? But I don't really understand how this stuff works, so probably not?

Or I should say to each tube face, one at a time, but I guess not if using colour corrected lenses.

Tinkered a little between bowl games. :D

If I place the Spyder ~10 inches from the screen facing the screen I get seemingly stable readings. I must have been too far away with my last setup.

MadMrH,

Nothing in the manual for FP, I have the SpyderTV version although Id be curious what the Spyder2 Pro owners have in their manuals for FP.

sorry for the slightly OT question:

I have beenout of the "Spyder for calibrating FP" loop for quite some time after I read somewhere that the readings are not reliable.

Has it gotten better? Can you now really calibrate a front projector using a Spyder?

I read somewhere that the readings are not reliable.


I've read that also. My understanding is that the current retail Spyder is a revised version of the old model.

Hi There,

You certainly can calibrate a FP CRT pj with a spyder2....Check out CalMan. I find it very good in the latest incarnation.

There are settings for emissive and Front Projection CRT which then means you can aim the spyder2 at the screen at a 45 degree angle and get good results....low light levels of CRTs do prove difficult but i can certainly get repeatable results at 20%, more often than not at 15% too....less than that and a slightly blue or red black isn't particularly noticeable anyway.

cya,
Lee

I just order the spyder2 pro because the software is compatible with my Macintosh operating system (requires Mac OSX 10.3 and higher). I am looking forward to using this for my pj (pg xtra).

http://archive2.avsforum.com/avs-vb/showthread.php?t=672531

I purchased Spyder2 Pro last week, so I'm just starting to learn how to calibrate my projector with it. The instructions are pretty clear on how to position the Spyder relative to the screen. They say you should point the sensor directly at the screen, centered and perpendicular, about 30 centimeters or 12 inches away from the screen. See the screen shot attached to this post.

The instruction say you shouldn't be concerned with the shadow being cast on the screen by the sensor.

To get accurate reading with CRT FP, you need to use offset.
Spyder pod is not sensitive enough to get Low IRE reading when facing screen. mine start being insensitive to red from 60IRE!
What u need to do is follow MadMrH instruction and place the POD facing the screen with a 100IRE white field pattern. Adjust white for D65. Then place the pod facing the tubes, about 1m away. Get reading and use it as your new target for your Sub 100IRE calibration. That should provide you accurate reading down to 25IRE. If you want to get accurate result down to 5IRE, use a loupe to gather more light into sensor and readjust offset.
These are complicated task but Spyder was not designed with CRT FP in mind,

A loop?

Ah a loupe!

I am unaccountably dense tonight.

a loupe :D.

To get accurate reading with CRT FP, you need to use offset.
Spyder pod is not sensitive enough to get Low IRE reading when facing screen. mine start being insensitive to red from 60IRE!
What u need to do is follow MadMrH instruction and place the POD facing the screen with a 100IRE white field pattern. Adjust white for D65. Then place the pod facing the tubes, about 1m away. Get reading and use it as your new target for your Sub 100IRE calibration. That should provide you accurate reading down to 25IRE. If you want to get accurate result down to 5IRE, use a loupe to gather more light into sensor and readjust offset.
These are complicated task but Spyder was not designed with CRT FP in mind,


Sorry, but :
1. what is 'Offset'?
2. About "place the pod facing the tubes, about 1m away"....The projected image wil neither be converged nor focussed at 1 m. Or is it that those two things dont matter for taking readings?

Sorry, but :
1. what is 'Offset'?

Offset is when you apply different values as target to your reading in order to compensate for measurements shifts. The correct offset should allow you to map pod readings back to the correct value.


2. About "place the pod facing the tubes, about 1m away"....The projected image wil neither be converged nor focussed at 1 m. Or is it that those two things dont matter for taking readings?

With a spyder, focus is not required @position in order to get reading in ambient mode. Unlike some other devices that require diffuser. Misconvergence is again not an issue as the test window is covering all the screen and offset will compensate for the difference in position of the RGB guns.

In short, what you need to do is:

1- Place pod front screen at angle as already mentioned. Display 100IRE white Window. Adjust PJ bias/gain till you get white @d65.
2- Place spyder 1m in front of PJ facing the tubes. Leave the white window @100IRE and take readings again. Write the coordinates down. You will notice those numbers have changed. If the pod is right centered in front of your PJ, you ll notice your xy coordinates are closer to green primary and Y is much higher then before. No worries here, this is normal since primaries color source are distant and your pod is now closer to the green tubes relatively to the other Blue/Red. Apply the offset method in order correct those reading. That is, the newly written coordinates are your offset value and they will map your readings back to D65 @screen. Thus they should be set as your new target for this pod position. Now you can proceed with your calibration for <=100IRE values. at this point, it is very important not to touch the pod because any position or angle shift will tremendously change your readings.

Hope This is clearer. :)

The reason I suggest you do all this is that Spyder is not linear at dark reading range. Bringing the pod closer to the light source will help the sensor to capture more light.

Have been thinking about using an ekstra HD8b lens with the lens-CRT side connected with the spyder 2 sensor to get a higer reading aimed directly at the screen with a longer distance to eliminate the RGB shadows. Can this be a way to overcome the lower IRE readings and get a better color readings from the screen?

Have been thinking about using an ekstra HD8b lens with the lens-CRT side connected with the spyder 2 sensor to get a higer reading aimed directly at the screen with a longer distance to eliminate the RGB shadows. Can this be a way to overcome the lower IRE readings and get a better color readings from the screen?
Only if you intend to watch movies through that "ekstra" (sic) lens, but it would get heavy holding it up all the way through a 2-2.5 hour movie.

I think the sensor in front of the screen (perhaps not as much as 30 CM or 1 foot, maybe closer) is the best. And,.. you can use the shadow the sensor makes on the screen to judge the reletive angles between the screen, sensor, and pj (you know look at the shadow and move the sensor till it seem to be angled good, it's hard to explain.)

If you're using DVE one tip is to use the patterns with just a smaller IRE window instead of the full screen patterns that tend to light up the room to much.

I'm really looking forward to trying HCFR with my Spyder2 and G70 tomorrow. (the rest of the family will be out of the house for a few hours.)

. . . I'm really looking forward to trying HCFR with my Spyder2 and G70 tomorrow. (the rest of the family will be out of the house for a few hours.)
Brian,

Please do update us. I've got my Spyder2 and HCFR software, but I'm heading out of town for the weekend in a little bit.

I'll be interested to hear your results.

I'm really looking forward to trying HCFR with my Spyder2 and G70 tomorrow.

Brian,

I have the exact same setup G70/SpyderII/HCFR.

I've taken a couple readings, but I've been struggling trying to make sense of what I see and the next steps.

If you wouldn't mind can you post your *.chc file. I'd like to compare settings.

Shows what I know,.. I have no idea what a CHC file is.

I worked on Focus today and was planning to do Geometry/Convergence before grey scale but I think I'll do some greyscale work tomorrow just to report back.

I'm not sure how it will go but I'm betting I work quite a lot with 100% and 40% and then just graph the color temp and see what I get. I'm pretty sure I won't get into looking at Gamma or anything very complex.

-Brian

The *.chc file is the saved file with the settings and measured results.

I've taken readings and adjusted gain and bias but the color reading didn't appear to change on the CIE chart.

So I was thinking I might not have the HCFR software setup right.

The CIE chart primarily shows the gamut (total range) of the primary colors (R, G, B) and the placement of the secondaries (cyan, magenta and yellow).
The primaries angles can't be shifted, they are what they are. They are limited by the phosphors in Your projector. However, the collective range of RGB will change if You lower color saturation in Your setup somehow - i.e. the gamut will shrink. There is however no way of enlargening the gamut.
The secondaries can to some extent be manipulated by changing HUE in Your setup.
None of these however are affected very much by manipulating the white balance/grey scale. The secondaries might move a bit, but not much usually.
To see the results of grey scale calibration You should examine the Luminance Histogram (for gamma) and the RGB Levels Histogram.
In the RGB Levels Histogram window You will also see a delta E (dE) chart. The closer You are to dE=1 at any given luminance level, the better. dE<4 is always acceptable, since the human vision is incapable of discerning such small differences.

Ive been out of the colorimeter scene for a while as well, but when I had my sencore I was under the belief that it was the projector you wanted to calibrate not the screen. I positioned the sensor at the screen facing the PJ and aimed it up to the center of the green lense (pj ceiling mounted) to maximized the light input.

Hi,

I got time to do a simple calibration last night.

I was using Digital Video Essentials. The first thing I did was measure the primaries and secondaries.

The next thing I did was setup the sensor to take constant measurements (there's a little window for that.) Using that, I set RGB outputs equal to the best of my ability at 40 and 100 IRE.

I did a 5 point measurement of greyscale using video 20,40,60,80, and 100IRE fields. The resulting greyscale look-ed really good (with the exception of the 20 IRE reading which I would actually prefer not to have tried to measure. I later replaced that measurement with 0,0,0 ) I had 6500 to plus or minus 300k or so and it was pretty flat.

The way the sensor responded to the changes I made on the pj leads me to believe it's functioning fairly accurately and the image afterwords was at least as accurate as I could have made it by eye and perhaps even a bit better.

All and all I think it works great. I love this HFCR software and using it is a snap.

(requested file is attached.)

-Brian

Brian,

Thanks! Just curios did you measure with or without baffle and toward screen or PJ?

Arioch,

Thanks also for clearing up CIE graphs and primary colors.

I measure facing the projector just in front of the screen.

I remove the suction cup (used for direct view CRT) and don't use the baffle designed for LCD.

-Brian

Hi,

I got time to do a simple calibration last night.

I was using Digital Video Essentials. The first thing I did was measure the primaries and secondaries.

The next thing I did was setup the sensor to take constant measurements (there's a little window for that.) Using that, I set RGB outputs equal to the best of my ability at 40 and 100 IRE.

I did a 5 point measurement of greyscale using video 20,40,60,80, and 100IRE fields. The resulting greyscale look-ed really good (with the exception of the 20 IRE reading which I would actually prefer not to have tried to measure. I later replaced that measurement with 0,0,0 ) I had 6500 to plus or minus 300k or so and it was pretty flat.

The way the sensor responded to the changes I made on the pj leads me to believe it's functioning fairly accurately and the image afterwords was at least as accurate as I could have made it by eye and perhaps even a bit better.

All and all I think it works great. I love this HFCR software and using it is a snap.

(requested file is attached.)

-Brian


Hi Brian -

There are ALOT of different ways to hit 6500K, what you want is to hit a target of x=.313 y=.329, and what you want to hit is D65 (which is those x,y coordinates), and keep your range Delta E's anywhere for 4 or less, right now, yours are at 3,6,7,7 (not including 0IRE). Also, I know you did a short read, but try sticking with 10 measure reading. In addition, your Gamma is off, you need to hit, I believe for CRT, a 2.5 gamma, so you need to increase that, this can be obtain by parameters for gamma in the CRT in addition to contrast and brightness settings.

USE the RGB tracking and pretty much ignore the greyscale - it useless - and actually pro calibrators no longer even like to mention this - since - as mentioned - its misleading..Let me give you an example, if your Green is reading 100%, then your red is 80%, but your blue is 120%, guess what?? you have a perfect 6500K reading..and we know this is not correct, it should be 100% for all three, therefore using greyscale is not the way to go, use RGB tracking to hit .313 .329 across the range, the closer you get to this, the better your Delta E/D65 are.

What I like to do and how I see the measures is if you highlight 50IRE and hit the Display Measurement icon, you will see the information I provided above, then you will see how, when using the repeat read function how your color bars changed, keep an eye on the delta E..generally with CRT, Green is either locked out when adjusting cuts and gains, or minor tweaking as this is your luminance , so you do not want to reduce that and rather adjust using red and blue to bring all bars to 100% (or as close as you can get)..

Your CIE is about as weird as Ive seen, not sure whats going on, but even when you dont have a perfect CIE, you still want a triangle where the points for primary and the secondaries fall within the same line..your secondaries (tint) are not in line with your primaries..

Consider doing longer extended reading for lower IRE with the Spyder2..it may take longer but it might give you a more accurate reading..

Good luck...get to know your projector..

last thing:

KEEP YOUR BAFFLE ON - ALL THE TIME..only for directview do you remove the baffle..this may be why your CIE is reading incorreclty..Always select LCD mode..dont bother ever using CRT mode with baffle off (unless your doing a CRT computer type monitor)

Hi Rich,

Thanks a lot for the tips.

Clearly,.. I'm just getting started with this software and I certainly do have quite a bit to learn and I don't mind learning if it means getting better performance out of the whole process.

I rejected the baffle at first because my manuals said it was for LCD (yuck.) but I've had a close look at it and it seems it serves to aim the sensor in a way. It looks like it would cut down on reading reflected light and I guess I'll try it again with it and see how that goes.



-Brian

Hi Rich,

Thanks a lot for the tips.

Clearly,.. I'm just getting started with this software and I certainly do have quite a bit to learn and I don't mind learning if it means getting better performance out of the whole process.

I rejected the baffle at first because my manuals said it was for LCD (yuck.) but I've had a close look at it and it seems it serves to aim the sensor in a way. It looks like it would cut down on reading reflected light and I guess I'll try it again with it and see how that goes.



-Brian

THE BAFFLE ON BAFFLE OFF HAS BEEN BEATEN UP ALREADY..

finally it has come to the conclusion. to keep it on...see for yourself

USE the RGB tracking and pretty much ignore the greyscale

..generally with CRT, Green is either locked out when adjusting cuts and gains, or minor tweaking as this is your luminance , so you do not want to reduce that and rather adjust using red and blue to bring all bars to 100% (or as close as you can get)..


richlo,

For the G70 and D65 according to the manual the Blue gain is set at MAX and is the reference. The RED and Green gains are adjusted. All three have bias control.

Does this change anything?

Non-aligned primary and secondary color readings could be the result of not having actually measured one of them if I remember correctly. I.e. if I measure primaries but not secondaries, the application will still draw them in the diagram, but in weird places... Not sure about this though...

arioch,

I'm confused on the LCD vs CRT setting.

If I use RGB tracking and adjust the primaries in LCD mode the color shifts extremely blue/green visually obvious.

In CRT mode the grey/white was visually better once I got the R,G,B at the same level.

The only settings I've adjusted are the sensor (Spyder) and display (CRT & LCD) are there other default settings that need to change to properly read a FP CRT?

You can get increased sensitivity for low light levels with a magnifier lens. The bigger in diameter the lower light level you can use. Get a lens from something like one of those lighted magnifiers or even one of those Sherlock Holmes jobs. Position it to focus the light spot on the light sensor like you are frying ants with the sun. See pic.

Scott

I spent a few hours going over the HCFR thread and decided to bite the bullet and ordered a sypder2. Its been a few years since I calibrated my PJ, but for a good colorimeter going for less than $65 and a great free proggie i couldnt resist.

arioch,

I'm confused on the LCD vs CRT setting.

If I use RGB tracking and adjust the primaries in LCD mode the color shifts extremely blue/green visually obvious.

In CRT mode the grey/white was visually better once I got the R,G,B at the same level.

The only settings I've adjusted are the sensor (Spyder) and display (CRT & LCD) are there other default settings that need to change to properly read a FP CRT?

I'm not sure what You mean... And I can't dl Your zip:s, have they been removed?
As people have said before here, You should use LCD mode and baffle on for everything except direct view CRT-monitors.

My meter came today and I spent a hour or so fooling around with it basically just taking readings. Looks like I need to work on my greyscale, but its not too bad....


http://xs511.xs.to/xs511/07044/xg852cie.jpg.xs.jpg (http://xs.to/xs.php?h=xs511&d=07044&f=xg852cie.jpg)
http://xs511.xs.to/xs511/07044/xg852rgb.jpg.xs.jpg (http://xs.to/xs.php?h=xs511&d=07044&f=xg852rgb.jpg)

Fernando... I'm sure with a bit more time you should be able to level your RGB tracking a bit more and get the dE<4 across the IRE range

fs,

Nice. Isn't it great to have such an ecomonical solution? Someone in the Blu Ray forums made files for Blu Ray out of high quality HD ts test patterns and also made MP4s of them for ps3. I'll be using those with my next round of grey scale tweaking. But,.. cheap sensor, free high quality software, and free high quality HD test patterns sure do make me smile.

-Brian

Yeah being crt guys I know a big thing for many of us is cost vs performance, which is the reason I sold my Sencore (ca-1) colorimeter (paid around $450) a few years back. Now that you can buy a good colorimeter for around $60 (sure theres a better one for 2x the price but we are crt guys!) and free software that is amazingly simple to use there is no reason why everyone shouldnt go out and get one.

Brian,

Have you taken any other measurements?

What HFCR setting for display type are you currently using, LCD or CRT setting?

I'am noticing with the LCD setting and adjusting the RGB levels the all white is more blue.

How about you ?

I don't have a Spider , but I own a CA-1 and here is a method I use to get the sensor as close to the screen as possible with the sensor facing the PJ.

I use two of the spring clips that have magnets , the kind you find at Target and are used to put notes on the fridge. I put one clip behind the screen and a second on the front. The magnets hold the clips in place. I then use the spring on the front clip to hold the cable and position the sensor with its back to the screen. If you have strong enough magnets you can slide them around to put the sensor where ever you want.

I did take more measurements but then when I tried to see some graphs the laptop crashed. That's the second time that happen-ed and I guess I'm determined that I have to measure with the laptop (because it's somewhat portable) and then look at the graphs on my main pc. (I can transfer the file easily.) I don't want to troubleshoot the laptop as it's an old celeron 300 windows 98 machine.. but it does let me take the measurements easily.

Recently I loaded my PS3 hard drive with HD test patterns and have started to use them (the ones for geometry for example.) So,.. I'm sure I'll try this again very soon. Hopefully this weekend sometime.

-Brian

My meter came today and I spent a hour or so fooling around with it basically just taking readings. Looks like I need to work on my greyscale, but its not too bad....


http://xs511.xs.to/xs511/07044/xg852cie.jpg.xs.jpg (http://xs.to/xs.php?h=xs511&d=07044&f=xg852cie.jpg)
http://xs511.xs.to/xs511/07044/xg852rgb.jpg.xs.jpg (http://xs.to/xs.php?h=xs511&d=07044&f=xg852rgb.jpg)

I would adjust the bias/cut around 30% to bring that point within tolerance. :)
What PJ is it?

^ I have an XG852. I did a quick adjustment last night bringing the red bias up a bit and lowering the blue and it all evened out nicely, but I still have a small blue hump at 60.

I'm not sure what You mean... And I can't dl Your zip:s, have they been removed?
As people have said before here, You should use LCD mode and baffle on for everything except direct view CRT-monitors.

arioch,

Files are still there.

I'm trying to figure out why the color looks terrible using LCD mode. :confused:

fs123,

Could I trouble you for a *chc file? Are you using LCD or CRT mode in the software?

I have it LCD mode, with that baffle thing on, facing the projector from the center of my screen, and I am calibrating it to a HDTV signal with HDTV test patterns. To be honest I have no idea if this is remotely the right way to do any of this, but from my first tests it seemed(?) to look right. Im pretty much just following the directions over at Curt's site for g2/ref white balance adjustments.

I played around with the spyder2 pro on my crt yesterday. The new spyder comes with a tripod attachement that makes it easy to use a tripod. So, I attached it to my tripod and placed it, probably about 6" from the screen center and turned the sensor towards the CRT, not the screen. I initially specified in the software that came with the spyder that the monitor is a projector, since that is what I have. But, and this is what I don't like about the software, I saw that would not work for me because I currently can't attach my Mac to the CRT. I prefer to use the VE and DVE discs. So, I changed the monitor specification to LCD, selected gamma 2.2 - 6500, and proceeded. The software puts up four consecutive fields; red, green, blue, and white and the sensor takes readings at each. So, to match that, I selected the appropriate title/chapters on VE when the software changed among the colors so the sensor would take the same readings. Then, the software comes up with an R/G/B slider that reflects the relative levels of each in the white field. Then, I changed VE to the small window chapters and took readings at 20, 50, and 100 IRE. I also adjusted B/W at each to try and attain 6500. That seemed to work fairly well for me. I don't know if this is an appropriate method to use or not. But I at least felt I was doing something correctly, for better or worse.

Also, the sensor seemed sensitive enough since single digit changes of B or W at 20 IRE were detected.

^ You should take a look over that the HCFR Thread (http://www.avsforum.com/avs-vb/showthread.php?t=737550) in the display calibration forum. Its a free program thats is much better than the ones that come with the spdyer2 sensor. Also if its not too late you should take back that spyder2 pro and just get the basic spyder2 express package since the sensor that it comes with are exactly the same and you can than just use that free calibration program.

What is the advantage of the HCFR software? I downloaded it to my wife's work computer since it is not compatable with Mac, but have not tried to use it. I'm hestitant to be reliant on the HCFR software since it is not compatable with the computer that I own. I just learned that Calman and accucal are also not compatable with Macs. Do people just not realize that there is a significant portion of the population out there that own Macs?

Id hardly call less than 6% (http://marketshare.hitslink.com/report.aspx?qprid=2)significant

:p

That is only because, for a long time, PCs were less expensive than Macs. That is starting to change though. I also don't see them at the big box stores like BB and CC.

Anyway, I'm not going to go out and buy a PC for the express purpose of those programs.

So, what is the advantage of HCFR, or Calman, or Accucal?