Official JVC RS20 / HD750 Calibration and CMS thread (NEW FIRMWARE V1.1) - Page 49 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1441 of 1634 Old 04-29-2010, 01:56 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Nathan - also as a side note, make sure you are not using ANY offset above 0 (negative values are OK if that's what your grayscale calibration calls for). Doing so can completely destroy the black level and on/off CR.
lovingdvd is offline  
Sponsored Links
Advertisement
 
post #1442 of 1634 Old 04-30-2010, 12:33 AM
Senior Member
 
karrih's Avatar
 
Join Date: Jul 2007
Posts: 254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by nathan_h View Post

And you end of up with a good grayscale this way, too, right? Do you adjust the individual color gamma controls, or just the W?

Yes and all colors have their own gamma. Doing grayscale separately and only white gamma might be less work, I guess I have to try it some day.
karrih is offline  
post #1443 of 1634 Old 05-01-2010, 12:20 PM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Did another run today. The before snapshot of color temp:



and the gamma measurement (this is the default 2.2 curve in the custom settings):



Which means this is the spreadsheet to use for correction:



I was able to get all the points on the W curve either exactly on the recommended corrections, or 1 off, at the extremes.

But the resulting gamma curve is not really hitting it quite right still. It looks better with real material, though it's a little tough to see shadow detail still -- but not unbearably so:



For some reason, it also screwed a bit with the color balance at the low end:


____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1444 of 1634 Old 05-01-2010, 02:59 PM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
I need to think through how to do a second run, and not always start from a zero-ed out gamma at one of the pre-sets, though this is how I am reading the instructions...

Of course, I got the instructions wrong before -- and having the process pointed out to me made dialing in the spreadsheet values 100% easier. So maybe I'm missing a bit again...

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1445 of 1634 Old 05-01-2010, 03:05 PM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
My wife's out of town this weekend, so it's a good time to play a bit with calibration,etc. Here below are some results I've measured fairly carefully (light meter on a tripod, so no variation with position,etc.) that are related to 'setting contrast', a topic that has been under active discussion in the 'Setting Contrast with the Munsil&spears calibration disc ...' in the Display Calibration thread.

I use 'Expanded' HDMI mode (as recommended by Greg Rogers, and a number of others), am in Normal (i.e., low) lamp, with Iris at 0 (fully open). Brightness is set to -5; this is unambiguous, coming out to this value from a variety of calibration discs (M&S, GetGray, AVSHD709, DVE BluRay, AVIA, etc.) as do a number of others.

The question comes with setting Contrast. If I set it as M&S recommend, to show the 'white bars' all the way up to ~251 (i.e., the WTW region), then Cst comes out to +6 (as other respected pro's--e.g., ChrisWiggles--also find). OTOH, if one sets Cst to show only the region up to 235 ('reference white'), as the highly respected GetGray disc recommends, I come up with Cst = +13 or +14. The discussion over in the calibration thread is whether or not there is any 'useful video info' in the WTW region, or whether setting Cst to show WTW gives less 'ringing' (or other artifacts) in the 'allowed' region (</= 235). So the possible range I have to choose for Cst is from +6 (showing all the WTW region) to +14 (showing none of it).

What I have measured is the effect this range of Cst has on the lumen output of my RS20. A higher Cst setting gives more lumens, so other things being equal would be preferred, unless lower Cst has some pq advantages. Here are the measurements:

Contrast Lumens ftL Ratio

+6 287 14.6 1.00
+8 300 15.3 1.05
+10 314 16.0 1.09
+12 327 16.7 1.14
+14 341 17.4 1.19

(The ftL #'s assume a screen gain of 2.4 for my 126" diag 16x9 HP screen.) The ratio is taken relative to the Cst = +6 value.

So going to the max Cst, i.e., eliminating all WTW, gives ~19% more brightness (lumens), and one sees very linear behavior over this region. If one is 'ftL poor', then one may very well wish to use these higher Cst settings. It will also be extremely useful to see if the 'Setting Contrast' thread comes to any conclusion about the advantages, or not, of covering the WTW region. (Again, there are very knowledgeable persons who seem to come down on opposite sides of this question.)

I would be interested hearing what value others of you choose for setting Cst (assuming that you use 'Expanded' HDMI mode).
millerwill is offline  
post #1446 of 1634 Old 05-01-2010, 09:57 PM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
I'm a fan of not clipping whiter than white, and have mine set at +8.

(Of course, I am fascinated by your brightness setting. I'm using expanded on the projector, and normal on my Oppo, and get +2 ... though my aperture is stopped fully closed.)

EDIT: just rechecked with the Spears disc and I must have made some change while working on the gamma. Now with the projector set to enhaced and the OPPO set to video levels both brightness ans contrast look most correct at 0.

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1447 of 1634 Old 05-01-2010, 11:07 PM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
Quote:
Originally Posted by nathan_h View Post

I'm a fan of not clipping whiter than white, and have mine set at +8.

(Of course, I am fascinated by your brightness setting. I'm using expanded on the projector, and normal on my Oppo, and get +2 ... though my aperture is stopped fully closed.)

Interesting. In 'Enhanced' mode I've only heard of Br settings of -5 or -6. And I just did a quick check and find that changing the Iris has no effect on the correct Br and Cst settings on my unit. Would be very interested to hear from others.
millerwill is offline  
post #1448 of 1634 Old 05-01-2010, 11:23 PM
Senior Member
 
karrih's Avatar
 
Join Date: Jul 2007
Posts: 254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by nathan_h View Post

I need to think through how to do a second run, and not always start from a zero-ed out gamma at one of the pre-sets, though this is how I am reading the instructions...

Just set the values you dialed in to the projectors gamma curve to 'Current Gamma Curve' and your measurements using those to 'values on the gamma graph' and you will get new, smaller corrections.

I wonder if the trick of using enhanced levels is somehow related to these gamma results.
karrih is offline  
post #1449 of 1634 Old 05-01-2010, 11:31 PM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Quote:
Originally Posted by millerwill View Post

Interesting. In 'Enhanced' mode I've only heard of Br settings of -5 or -6. And I just did a quick check and find that changing the Iris has no effect on the correct Br and Cst settings on my unit. Would be very interested to hear from others.

I just rechecked with the Spears disc and I must have made some change while working on the gamma. Now with the projector set to enhanced and the OPPO set to video levels both brightness and contrast look most correct at 0

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1450 of 1634 Old 05-01-2010, 11:35 PM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Quote:
Originally Posted by karrih View Post

Just set the values you dialed in to the projectors gamma curve to 'Current Gamma Curve' and your measurements using those to 'values on the gamma graph' and you will get new, smaller corrections.

I wonder if the trick of using enhanced levels is somehow related to these gamma results.

I would not be surprised but I appreciate the tip and will give it a try.

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1451 of 1634 Old 05-02-2010, 07:42 AM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by millerwill View Post

My wife's out of town this weekend, so it's a good time to play a bit with calibration,etc. Here below are some results I've measured fairly carefully (light meter on a tripod, so no variation with position,etc.) that are related to 'setting contrast', a topic that has been under active discussion in the 'Setting Contrast with the Munsil&spears calibration disc ...' in the Display Calibration thread.

I use 'Expanded' HDMI mode (as recommended by Greg Rogers, and a number of others), am in Normal (i.e., low) lamp, with Iris at 0 (fully open). Brightness is set to -5; this is unambiguous, coming out to this value from a variety of calibration discs (M&S, GetGray, AVSHD709, DVE BluRay, AVIA, etc.) as do a number of others.

The question comes with setting Contrast. If I set it as M&S recommend, to show the 'white bars' all the way up to ~251 (i.e., the WTW region), then Cst comes out to +6 (as other respected pro's--e.g., ChrisWiggles--also find). OTOH, if one sets Cst to show only the region up to 235 ('reference white'), as the highly respected GetGray disc recommends, I come up with Cst = +13 or +14. The discussion over in the calibration thread is whether or not there is any 'useful video info' in the WTW region, or whether setting Cst to show WTW gives less 'ringing' (or other artifacts) in the 'allowed' region (</= 235). So the possible range I have to choose for Cst is from +6 (showing all the WTW region) to +14 (showing none of it).

What I have measured is the effect this range of Cst has on the lumen output of my RS20. A higher Cst setting gives more lumens, so other things being equal would be preferred, unless lower Cst has some pq advantages. Here are the measurements:

Contrast Lumens ftL Ratio

+6 287 14.6 1.00
+8 300 15.3 1.05
+10 314 16.0 1.09
+12 327 16.7 1.14
+14 341 17.4 1.19

(The ftL #'s assume a screen gain of 2.4 for my 126" diag 16x9 HP screen.) The ratio is taken relative to the Cst = +6 value.

So going to the max Cst, i.e., eliminating all WTW, gives ~19% more brightness (lumens), and one sees very linear behavior over this region. If one is 'ftL poor', then one may very well wish to use these higher Cst settings. It will also be extremely useful to see if the 'Setting Contrast' thread comes to any conclusion about the advantages, or not, of covering the WTW region. (Again, there are very knowledgeable persons who seem to come down on opposite sides of this question.)

I would be interested hearing what value others of you choose for setting Cst (assuming that you use 'Expanded' HDMI mode).

Thanks for the report. I run with HDMI type Normal. I keep contrast and brightness both at 0. Technically speaking for my setup brightness should be at 1 and contrast at -1 or -2.

I use brightness at 0, however, because just a single click significantly raises the absolute black level significantly. So I am willing to trade a little bit of black crush (notably I cannot really see bar 17 with brightness at 0) for inky blacks.

Contrast at 0 for me makes the 234 bar barely visible. It is more defined at contrast -1 and a bit more so at -2. However I have not found material yet with very bright whites where I could notice a difference between -2, -1 and 0, so I keep it at 0 to maximize brightness and CR.

As far as Enhanced HDMI goes, I do not run it because I have no interest in running BTB or WTW. This is a personal decision. The purists out there have good reason to want to have WTW because as I understand it there is sometimes material in the 235+ range that if not calibrated for will result in some white crush.

However, from what I understand such scenes are very rare, and may only be on particular films. It sounds like we are talking about a very rare occurrence and that when it happens it may even go unnoticed to the trained eye (i.e. not stick out like a sore thumb).

To me personally I would much rather run calibrated up to 235 and pick up all this extra brightness and on/off CR, then to be calibrated for all scenarios, when such scenarios (235+) are rare. Again this is a personal taste type of thing so YMMV.

As an experiment I did try running in Enhanced mode with things set to clip at BTB and WTW. The settings for this were contrast at 15 and brightness at -7. Such settings produced identical results to running in Normal mode with contrast/brightness at 0 with various material I tested. For reference, in my case contrast of 13 with brightness of -6 in Enhanced mode was equivalent to contrast of -2, brightness of 1 in Normal mode.

millerwill - I highly encourage you to ensure that you really need brightness at -5. That seems a bit high to me. Forget about your contrast-setting patterns for now and try this experiment...

With brightness at -5, put up a 0% pattern (full field/screen absolute black). Bring up the Brightness control and click on it so the slider is shown on the bottom of the screen and rest of the RS20 menu is hidden. Now watch the overall black level on the screen and reduce the slider to -6. Notice how much the blacks go down? It should be significant if your set up is like mine. Now bright it down one click again to -7. You should notice the same dropoff and you are now at inky blacks (comparatively). You will also notice that if you try to go any lower the black level will remain constant.

This experiment has real world implications because your lowest black levels at -5 will look that much brightness in your scenes (especially very dark ones) compared to the -7. My guess is that by running at -5 you are running with a on/off CR about 30% lower than it would be with -7. Of course there are trade offs because with -7 you will have slight black crush. But to me this is a worthwhile trade off.

Lastly, all this assumes that your projector is calibrated near perfectly including extensive gamma tweaking aka leDahu's spreadsheet so that the 5% (and all other levels) are dialed in correctly.
lovingdvd is offline  
post #1452 of 1634 Old 05-02-2010, 07:49 AM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by nathan_h View Post

I'm a fan of not clipping whiter than white, and have mine set at +8.

Hi Nathan,

I am curious why you are a fan of this. Is it because you like the follow the theory of having that upside protection, just in case? Or have you found particular scenes where this setting has prevented some white crush you would have had otherwise? And if so, how often do you come across such scenes in your overall viewing?

I assume you are aware that with the approach you are taking you are trading a significant amount of brightness and on/off CR (your on/off contrast is "compressed" as a result of this setting) in exchange for this upside protection. This is why (as I just wrote to millerwill above) that I personally choose not to run this way - because unless it is well known that there are often scenes where this upside protection is necessary I'd much rather maximize my brightness and on/off CR.
lovingdvd is offline  
post #1453 of 1634 Old 05-02-2010, 09:12 AM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
I've got a couple of factors that may be worth mentioning:

1) I don't need more brightness. Although I don't trust what the Spyder2+HCFR are telling me about brightness output, my eyes tell me I have more than enough. I use a Stewart 2.0 screen, 2.35:1, with the zoom method. The screen is about 115 inches wide horizontally. My room is the proverbial bat cave.

2) My Oppo is set to RGB-video output for the HDMI setting (ie, "normal"), and my projector is set to "enhanced", yet I am still getting a setting of 0 now for both brightness and contrast when using the Spears blu ray. Something is strange about that, based on what most other people are seeing. So take this with a grain of salt. (Perhaps something I've done with the grayscale setting or the gamma settings have put me outside the realm of what one would expect. I should also explicitly ensure that I really am still watching with the enhanced HDMI setting enabled on the projector.)

3) On/off contrast matters to me, but mostly only in two specific areas: absolute black level, and detail in dark scenes -- but always without any hint of haze, which drives me batty. In fact, this was the reason for upgrading from a Sony Pearl: increased sharpness is nice, more accurate color is nice, higher light output is nice, getting rid of any possible dynamic iris artifacts is nice. But I watch 99% movies, and getting a quality viewing experience in low light situations was my goal. So that is what informs all my other decisions.

Those things said, I haven't noticed anything negative about allowing WTW to pass through. (I cannot say the same thing about BTB, which I like to be able to see when setting brightness, but which I set brightness to barely crush -- because I, like you, find that even one step and definitely two steps higher brightness, allowing a little BTB to pass through, really washes out the picture. Did I mention that that kind of haze is my #1 bugaboo?)

I like to know that when I see a blown out highlight on the screen, that is the way it is intended, and there really is no overshoot or additional video detail accidentally placed there by the disc author but clipped by me (because I am adhering absolutely to the spec). I have not systematically checked certain movies or scenes for detail in the WTW area, nor know whether it is something that only exists in 1% or 50% of the discs I watch.

---

I've got less than 100 hours on my bulb, since the bulb that was in the projector when I bought it was damaged and I went the safe route and replaced it, even though it was more than bright enough for me to have the iris stopped down below 50% still. So some of my results will likely be very different once this bulb stabilizes and I get a professional in to set things up. I should have the cash by the time of UMR's California tour in June, and I am on his list. I will be very curious to see whether I am in the ballpark or not, with my own calibration efforts. I am sure I'll learn a lot about what I don't know, then

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1454 of 1634 Old 05-02-2010, 10:02 AM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
Quote:
Originally Posted by lovingdvd View Post


millerwill - I highly encourage you to ensure that you really need brightness at -5. That seems a bit high to me. Forget about your contrast-setting patterns for now and try this experiment...

With brightness at -5, put up a 0% pattern (full field/screen absolute black). Bring up the Brightness control and click on it so the slider is shown on the bottom of the screen and rest of the RS20 menu is hidden. Now watch the overall black level on the screen and reduce the slider to -6. Notice how much the blacks go down? It should be significant if your set up is like mine. Now bright it down one click again to -7. You should notice the same dropoff and you are now at inky blacks (comparatively). You will also notice that if you try to go any lower the black level will remain constant.

This experiment has real world implications because your lowest black levels at -5 will look that much brightness in your scenes (especially very dark ones) compared to the -7. My guess is that by running at -5 you are running with a on/off CR about 30% lower than it would be with -7. Of course there are trade offs because with -7 you will have slight black crush. But to me this is a worthwhile trade off.

Lastly, all this assumes that your projector is calibrated near perfectly including extensive gamma tweaking aka leDahu's spreadsheet so that the 5% (and all other levels) are dialed in correctly.

Thanks for the detailed feedback. Many good points, showing that there's more than one way to get an excellent pic.

Re Brightness, the 'correct' setting re the Pluge pattern is -6 or -5 for my setup. And this is also what I get using a procedure similar to what you describe, i.e., turning Br way down and then gradually increasing it until one sees the first sign of the background lighten. I choose -5 rather than -6 in order to avoid ANY black crush, something I really detest. But you are right: -6 is perhaps more technically correct.

Re Contrast, I find the 'white bars' (e.g., the Contrast test pattern on the M&S disc or AVSHD709 one) to be MUCH more sharply defined with 'Enhanced' rather than 'Normal', even when I set Cst up to +12 or +14, to clip down close to 235. At present I am using Cst at +8 or +10 in order to get a bit of headroom, which gives up only ~ 10% of lumens; this shows the 235 white bars fully resolved, with the ones up to 242 or so just barely visible.

Re calibration, I use the ChromaPure package (with an I1Pro and I1LT) and feel that I get gray scale and CMS spot on. I haven't done leDahu's gamma procedure, but with the RS20 set to 2.4, and the other calibration done, it comes out to 2.30 +/- 0.05 over the whole range. Is there much more to be gained in this regard?

Thanks again for the the thoughts (on this and many other issues!).
millerwill is offline  
post #1455 of 1634 Old 05-02-2010, 10:39 AM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Quote:
Originally Posted by millerwill View Post

I haven't done leDahu's gamma procedure, but with the RS20 set to 2.4, and the other calibration done, it comes out to 2.30 +/- 0.05 over the whole range.

Wow, that's a great result. Even with grayscale and CMS dialed in closely, my RS20's gamma, when using the correction curve set for a high 2.6, actually measured about 1.9.

If I was within .05 of 2.3 for the entire range, and preferred to avoid any black crush at all as you do, I'd probably leave well enough alone and not adjust gamma further.

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1456 of 1634 Old 05-02-2010, 11:11 AM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by nathan_h View Post

Wow, that's a great result. Even with grayscale and CMS dialed in closely, my RS20's gamma, when using the correction curve set for a high 2.6, actually measured about 1.9.

If I was within .05 of 2.3 for the entire range, and preferred to avoid any black crush at all as you do, I'd probably leave well enough alone and not adjust gamma further.

There seems to be something very odd about many of the JVC units as far as gamma drift goes. I was surprised to learn that millerwill's unit does not have this issue. Unfortunately what you are experiencing I and many others have experienced as well. We have not yet been able to determine the root cause. But we do know the fix (leDahu's spreadsheet).
lovingdvd is offline  
post #1457 of 1634 Old 05-02-2010, 11:24 AM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Quote:
Originally Posted by lovingdvd View Post

There seems to be something very odd about many of the JVC units as far as gamma drift goes. I was surprised to learn that millerwill's unit does not have this issue. Unfortunately what you are experiencing I and many others have experienced as well. We have not yet been able to determine the root cause. But we do know the fix (leDahu's spreadsheet).

I haven't quite got it dialed in but yes, it was pretty far out of whack, and has required quite a bit of adjustment to get in the right ballpark.

And this is only about 500 hours after UMR dialed in in perfectly (albeit almost a year ago, and in a different, room, for a different owner, on the original bulb).

Is the drift related to actual time used? Bulb life? Some other degradable component (I hope not!)?

Has anyone ever drifted so far that the gamma cannot be corrected through the user menu adjustments?

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1458 of 1634 Old 05-02-2010, 11:24 AM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by nathan_h View Post

...1) I don't need more brightness...

This is fine. However then the right way to adjust for too much brightness is to use a ND filter. This way your bright level goes down, but your dark level goes down an equal amount.

What you are doing currently by setting the contrast control low is that you are compressing the range that can be displayed. IMO is unnecessarily diminishing the range and quality of almost everything you watch, in exchange for that rare moment (and more likely close to never) where a few scenes have a small amount of white crush eliminated.
lovingdvd is offline  
post #1459 of 1634 Old 05-02-2010, 11:27 AM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Fair enough. Since I am still in the figuring-this-projector-out phase, I'll give the clipping WTW settings a try. I have some ND filters but I'm not sure I'd ready to go there, yet.

But the thing that I really want to dial in better is the gamma, and then next time a fire it up with a few hours to spare, I'll be taking another run at refining that

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1460 of 1634 Old 05-02-2010, 11:54 AM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
Quote:
Originally Posted by lovingdvd View Post

This is fine. However then the right way to adjust for too much brightness is to use a ND filter. This way your bright level goes down, but your dark level goes down an equal amount.

What you are doing currently by setting the contrast control low is that you are compressing the range that can be displayed. IMO is unnecessarily diminishing the range and quality of almost everything you watch, in exchange for that rare moment (and more likely close to never) where a few scenes have a small amount of white crush eliminated.

Two points/questions:

1. Isn't cranking down the Iris a better way to reduce brightness (if you have too much) than using an ND filter?

2. Even if one doesn't want WTW, I found the 'white bars' in the Contrast patterns in 'Enhanced' mode to be much better defined--even with Cst set to clip down to ~235--than in 'Standard' (the correct term, as I just checked) mode. When you tried 'Enhanced' mode, and set Cst and Br to be ~equivalent to 'Standard' mode, did you notice any difference in how well the 'white bars' were resolved?
millerwill is offline  
post #1461 of 1634 Old 05-02-2010, 04:37 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by millerwill View Post

Two points/questions:

1. Isn't cranking down the Iris a better way to reduce brightness (if you have too much) than using an ND filter?

Doh! Right, of course. Forgot about the 'ole variable iris. So used to the old days where there was no such convenience.

Quote:

2. Even if one doesn't want WTW, I found the 'white bars' in the Contrast patterns in 'Enhanced' mode to be much better defined--even with Cst set to clip down to ~235--than in 'Standard' (the correct term, as I just checked) mode. When you tried 'Enhanced' mode, and set Cst and Br to be ~equivalent to 'Standard' mode, did you notice any difference in how well the 'white bars' were resolved?

I looked hard for any difference. I could not discern anything difference between Normal and Enhanced when the contrast and brightness was set according.
lovingdvd is offline  
post #1462 of 1634 Old 05-02-2010, 05:35 PM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
[quote=lovingdvd;18573764
I looked hard for any difference. I could not discern anything difference between Normal and Enhanced when the contrast and brightness was set according.[/QUOTE]

OK, this is useful info. I certainly see the diff on calibration discs, but the more relevant question, which you address, is what does one see in real scenes. And I'm more inclined to trust your judgement on this than mine.

FWIW, I note that the developers of Calman, in a post today over in the Display Calibration section of the forum, recommend calibrating to show material all the way up to 254.
millerwill is offline  
post #1463 of 1634 Old 05-02-2010, 10:16 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by millerwill View Post

...FWIW, I note that the developers of Calman, in a post today over in the Display Calibration section of the forum, recommend calibrating to show material all the way up to 254.

Ultimately what it comes down to is: 1) How much material is out there where WTW is present, and 2) In those cases where WTW is present, to what extent is the white crush, and in how many frames/scenes?

I do not doubt that clipping WTW can create some issues in certain circumstances. However from what I have heard, we are talking about very rare occurrences. And even in those cases, just for a particular shot.

It would be helpful if people in favor of WTW calibration could elaborate on how many films they saw issues on without WTW calibration, and to what extent was the picture impacted.

Again, my understanding is limited, but from what I have heard in such discussions, these are rare occurrences. IMO calibrating for WTW up to 254 has significant consequences in terms of overall brightness and on/off CR. I would only be willing to make such a sacrifice if WTW issues were quite prevalent. And my understand is, its far from it. Again this is a personal taste and decision.
lovingdvd is offline  
post #1464 of 1634 Old 05-02-2010, 10:29 PM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Quote:
Originally Posted by karrih View Post

Just set the values you dialed in to the projectors gamma curve to 'Current Gamma Curve' and your measurements using those to 'values on the gamma graph' and you will get new, smaller corrections.

I gave this a try. Again, just adjusting the W curve. Ran out of adjustment room in the lower ire.

So I decided to look at the RBG gamma readings, and not just W. This data is from BEFORE I did this additional run.



I haven't done a new set of readings but in addition to adjusting the W curve, I worked on the lower IRE red curve -- and the results were an improvement with the the program material I was watching (The Matrix).

I'm beginning to think I should start over and ignore the W curve, and address each individual color first. They clearly don't all track together, but getting them to track more closely, helps. And as the readings above show, instead of raising W in the lower IRE, I should have first lowered the green and increased the red -- and then addressed all three with W.

____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1465 of 1634 Old 05-03-2010, 02:04 AM
Senior Member
 
karrih's Avatar
 
Join Date: Jul 2007
Posts: 254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by nathan_h View Post

I'm beginning to think I should start over and ignore the W curve, and address each individual color first. They clearly don't all track together, but getting them to track more closely, helps. And as the readings above show, instead of raising W in the lower IRE, I should have first lowered the green and increased the red -- and then addressed all three with W.

You could ignore white, but starting with white gives good initial curve for others.
karrih is offline  
post #1466 of 1634 Old 05-03-2010, 07:12 AM
AVS Special Member
 
nathan_h's Avatar
 
Join Date: Mar 2001
Location: SF Bay Area
Posts: 5,004
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 87
Hmmm.... That makes sense. And I need to measure again before I know where I am at, now. But since I ran out of room at the bottom end to make adjustments, and I suspect part of that is that red needed lots more adjustment than the others, so I was adjusting some wrong in order to make red right, I'm hoping to avoid that "running out of room to make an adjustment" problem. Here is the set of corrections I input yesterday. As you can see, there is no way I'd ever get where I need to go, at10 ire (-10) and 5 ire (even further into the negative) since the adjustment range on the projector ends at a value of 5.


____________________

Build Thread:
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
nathan_h is offline  
post #1467 of 1634 Old 05-03-2010, 07:49 AM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
Quote:
Originally Posted by lovingdvd View Post

Ultimately what it comes down to is: 1) How much material is out there where WTW is present, and 2) In those cases where WTW is present, to what extent is the white crush, and in how many frames/scenes?

A lot of the discussion in the 'Setting Contrast ...' thread in the Display Calibration section deals with trying to answer these questions. I certainly don't know the answers.

I do find it interesting that even the most respected and experienced guru's don't have a consensus on this issue. Many professional calibrators use the GetGray calibration disc, and its Contrast procedure clips down to 235. The new Spears&Munsil disc is highly regarded, and it recommends that one calibrate to show white up to 251 or so (or they point out that it's 'OK' to clip down somewhat if lumen output of the display is critical).

But yes, one does pay a price in overall brightness if one chooses to show WTW up to ~250; I found (above) the lumen output to be about 20% less if one shows all the WTW region, compared to showing none of it.
millerwill is offline  
post #1468 of 1634 Old 05-03-2010, 01:41 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by millerwill View Post

...But yes, one does pay a price in overall brightness if one chooses to show WTW up to ~250;...

The key point isn't so much the lose is brightness. Rather, it is the loss in CR. As I'm sure you know, on/off CR measures the amount of "space" between the brightness white and darkest black. When CR is not maximized, one "squishes" the CR so that it effectively lowers the resolution of the range (not resolution as in pixels, but rather is in the "space" between the steps in shades of gray).

Unless there is proof that significant errors can be seen by not adjusting for WTW, and that such errors can be seen with enough regularity to make it worthwhile, I cannot see how folks can justify giving up the brightness and on/off CR. But to each his own.
lovingdvd is offline  
post #1469 of 1634 Old 05-03-2010, 02:18 PM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,416
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 38
Quote:
Originally Posted by lovingdvd View Post

The key point isn't so much the lose is brightness. Rather, it is the loss in CR. As I'm sure you know, on/off CR measures the amount of "space" between the brightness white and darkest black. When CR is not maximized, one "squishes" the CR so that it effectively lowers the resolution of the range (not resolution as in pixels, but rather is in the "space" between the steps in shades of gray).

Unless there is proof that significant errors can be seen by not adjusting for WTW, and that such errors can be seen with enough regularity to make it worthwhile, I cannot see how folks can justify giving up the brightness and on/off CR. But to each his own.

Very valid points. Yet note this post below in response to a post I made in the 'Setting Contrast ...' thread, asking why there wasn't consensus on where to set Contrast:

For what it's worth I believe there actually is concensus*

Gregg and Michael at THX both said, disclose all WTW.
Charles Poynton just said disclose all WTW.
Spear and Munsil say disclose all WTW.
I didn't have a chance to talk to Joel Silver about it when I met him.

The leading industry experts all seem to agree.

*Of course they all also recognize that the real world comes with tradeoffs and compromises, and that certain attributes take priority in different cases.
__________________
I do not speak officially in any sense for
SpectraCal, but I do work there.
millerwill is offline  
post #1470 of 1634 Old 05-03-2010, 04:59 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,875
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 32
Quote:
Originally Posted by millerwill View Post

Very valid points. Yet note this post below in response to a post I made in the 'Setting Contrast ...' thread, asking why there wasn't consensus on where to set Contrast:

For what it's worth I believe there actually is concensus*

Gregg and Michael at THX both said, disclose all WTW.
Charles Poynton just said disclose all WTW.
Spear and Munsil say disclose all WTW.
I didn't have a chance to talk to Joel Silver about it when I met him.

The leading industry experts all seem to agree.

*Of course they all also recognize that the real world comes with tradeoffs and compromises, and that certain attributes take priority in different cases.
__________________
I do not speak officially in any sense for
SpectraCal, but I do work there.

Thanks for the information. It would be great to see a list of all known instances of real films/material where it is known that clipping WTW causes an issue, along with some info how many scenes are affected and how noticeable it is. My understanding is that it is very rare. And while it may be considered a great best-practice, IMO its not worth it unless the issues are numerous and substantial.

Another way I look at it is that many on the forums are very particular about their PQ of course, myself included. Yet I don't recall in years and years of hanging here a single reported instance about a problematic scene that was traced back to be a WTW issue. I'm sure some of these exist and have been discussed. But I haven't come across it. And when I figure that, in combination with the fact that most folks I think are running with WTW clipped, it just doesn't seem to be much of an issue.

Now, if someone presents a list of dozens and dozens of modern films with scenes which have noticeable artifacts in numbers, then maybe I would concern myself with it.
lovingdvd is offline  
Reply Digital Hi-End Projectors - $3,000+ USD MSRP

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off