AVS Forum banner
Status
Not open for further replies.
1 - 20 of 53 Posts

·
Registered
Joined
·
103 Posts
Discussion Starter · #1 ·
I had my new HS51 calibrated last week by Eliab Alvarez de la Campa of AviCal. Eliab calibrated my previous set, a second generation Toshiba CRT rear projector, the TW56X81. After a TON of work on that set he was able to achieve what I always considered a spectacular picture so I trust Eliab's ability and opinion.


He was able to achieve an excellent gray scale calibration but he was strongly against the use of the auto iris feature. We put one of the grayscale patterns from DVE on the screen (I believe it was average picture level), the one that has the 4 stacked gray boxes and the pluge patterns to either site of those. With the auto iris off the gray boxes were practically perfect. When you turned auto iris on and then especially set to "auto" there was a DRAMATIC difference. The top box turned cyan and the next box down was definitely pushing magenta hard. Also the black level detail was heavily effected, with the darkest pluge pattern (-4 db I think they are) disappearing. Eliab considered this to be proof positive that the auto iris was doing more harm than good and was clipping data.


I have done some viewing since the calibration, though only a few hours at this point. My early observations are that the color balance is outstanding, but with the auto iris disabled blacks are predictably dark gray and not black, though detail again is excellent. Also I find that the picture seems much more bright (generally a good thing with this low light level projector) but that I'm far more aware of pixel structure/screen door. I don't consider any of these observations to be final as watching a few DVD scenes, a couple of sporting events and a movie hardly qualify as critical viewing. I will spend more time with the calibration before making any decisions about the results.


Most things I've read about this projector indicate that the auto iris does make calibrating grayscale quite difficult but that overall it's benefits outweigh the drawbacks. What I wonder about is that with this calibration I may have a technically perfect display, but one that could produce a subjectively "better" picture if the auto iris were engaged, and the calibration performed with the necessary compromises.


I'm curious as to the opinions and experiences others here have with this projector and grayscale calibrations. Eliab told me he'd be happy to respond to any thread I might post here and comments it may draw. I'm all ears!!
 

·
Premium Member
Joined
·
12,075 Posts
It is indeed very hard to calibrate for the iris on this machine, but it can be done. The thing is the contrast is more then doubled with the Iris set to Auto. You are probably getting at best 2K:1 whereas I am getting just a tad under 6K:1!! That is a HUGE difference.
 

·
Registered
Joined
·
103 Posts
Discussion Starter · #3 ·
Quote:
Originally Posted by Kris Deering
It is indeed very hard to calibrate for the iris on this machine, but it can be done. The thing is the contrast is more then doubled with the Iris set to Auto. You are probably getting at best 2K:1 whereas I am getting just a tad under 6K:1!! That is a HUGE difference.
So what is it you did in order to achieve that difference?
 

·
Registered
Joined
·
1,757 Posts
komoman, You need to get your money back for that calibration or have him come back and do it right. You just lost the main reason to own this pj otherwise you could have bought a z3 and saved a bundle. You should write all the settings down and then compare it to Kris's settings posted in the HS(50)51 thread. Good luck.
 

·
Registered
Joined
·
7,222 Posts
Quote:
Originally Posted by komoman
I had my new HS51 calibrated last week by Eliab Alvarez de la Campa of AviCal. Eliab calibrated my previous set, a second generation Toshiba CRT rear projector, the TW56X81. After a TON of work on that set he was able to achieve what I always considered a spectacular picture so I trust Eliab's ability and opinion.
Any chance you might post your Iris-OFF calibration settings? I have a setup I use with low & high lamp iris off mode, and it's very differnet from my iris ON settings.


I just did some tweaking last night, with the 3910. I now have (3) settings I really like, which are as follows.


Contrast 80

brightness 55

color 45

hue 50

sharp off

RG 128, GG 51, BG 62, RB 129, GB 128, BB 125

RG 140, GG 65, BG 85, RB 128, GB 128, BB 128

RG 135, GG 75, BG 105, RB 128, GB 128, BB 128


The 2nd two setting of numbers of these RGB Gain & Bias settings make nice, natural shift toward blue from the 6500K LOW setting of the 1st number.


My Iris OFF settings are:


contrast 90

brightness 50

color 45

hue 50

sharp OFF

RG 110, GG 60, GB 80, RB 128, GB 128, BB 128
 

·
Registered
Joined
·
23,188 Posts
It doesn't really surprise me that somebody with measuring tools would decide that iris off is the way to go, but if you put 20 people in a room and compared an iris off machine to an iris auto machine (with both calibrated correctly for that) with the same material, I think way more than half would choose the iris auto machine. The reason is that the "black" level is so important that even a little color shift would be less important to the vast majority of people. Basically, it is possible to get too caught up in standards and not see that something is a visual improvement. So far I've heard people say that they would prefer iris off based on theory, but don't know of anybody who actually has ended up prefering it to iris auto when comparing with real material.


I actually think that Kris was being a little generous, as I would expect an HS51 with the iris set to off to only have about 1000:1. With the iris on auto I would expect anywhere from 3000:1 to 6000:1 depending on machine (some seem to be better than others).


--Darin
 

·
Registered
Joined
·
103 Posts
Discussion Starter · #7 ·
Quote:
Originally Posted by darinp2
It doesn't really surprise me that somebody with measuring tools would decide that iris off is the way to go, but if you put 20 people in a room and compared an iris off machine to an iris auto machine (with both calibrated correctly for that) with the same material, I think way more than half would choose the iris auto machine. The reason is that the "black" level is so important that even a little color shift would be less important to the vast majority of people. Basically, it is possible to get too caught up in standards and not see that something is a visual improvement. So far I've heard people say that they would prefer iris off based on theory, but don't know of anybody who actually has ended up prefering it to iris auto when comparing with real material.


I actually think that Kris was being a little generous, as I would expect an HS51 with the iris set to off to only have about 1000:1. With the iris on auto I would expect anywhere from 3000:1 to 6000:1 depending on machine (some seem to be better than others).


--Darin
This technical vs subjective was my greatest concern and why I posted this thread. I did some watching last night and some playing around with that DVE pattern. Increased contrast (therefore Iris on/auto) on that projector REALLY messes with the color balance. As calibrated the contrast was on 80. In order to remove the blue from the top bar of the pattern I had to pull it down to nearly 20. This was using the numbers Kris had posted, though really it was the same relative result with the calibrated settings. I'm going to continue messing with the auto iris on and off and the settings and see what real world viewing shows.


Bytehoven:


I will post the calibrated numbers later, I don't have them with me. The Red numbers are very close to Kris's but the Blue & Green are very different.
 

·
Registered
Joined
·
3,679 Posts
The trick to calibrating the HS51with Iris ON is to put up a single screen with all the gray level patches at once.


Then move your sensor between say 20 and 80 IRE patches. With the single display, the Iris will be on, but fixed during the cal, thus avoiding the "whack the mole" game.


Ken
 

·
Registered
Joined
·
1,390 Posts
I just calibrated one of these and went through the same analysis with the customer. Shane Buettner mentioned this issue on his review of the HS-51. A 100% luminance stripe measured 6825k with iris off and 8200k with iris on. However, he also noted that there is no free lunch in video, and the ADVANCED IRIS decided helps far more than it hurts. He measured the contrast ratio at 370:1 vs. 1346:1 with iris on a significant difference.


Obviously, a lot of this is a matter of personal opinion, but with the iris on, we were still able to get a reasonably flat grayscale (see attached), consistent with what Buettner reports in his review of the HS51.
 

·
Registered
Joined
·
23,188 Posts
Quote:
Originally Posted by Kevin R. Anderson
He measured the contrast ratio at 370:1 vs. 1346:1 with iris on – a significant difference.
It sounds like he has the brightness set way too high or a faulty machine. I think those numbers should be minimums of about 800:1 and 2500:1 with a proper machine and calibration. Also, people can add a color filter to balance the colors out optically (I think red is pretty weak on most of these) and then calibrate with that for the best CRs.


--Darin
 

·
Registered
Joined
·
1,390 Posts
Here are quotes from Buettner's review


"Measurements were taken at the screen plane with a Sencore CP5000

using an Accupel HDG-3000 SD/HD signal generator and an Ayre DX-

7 DVD player as sources. Measurements for contrast ratio were taken "real

world," with the projector dialed in for home-theater use (i.e., no crushed

white or blacks or color tint)."


"With the ADVANCED IRIS set at "auto" and the lamp at LOW, a 100 IRE

window measured 12.12fL, with a black frame measuring an excellent

.009fL. While the top end number is not blazing, the correspondingly low

black level makes for a real-world contrast ratio of 1346:1!"
 

·
Registered
Joined
·
1,234 Posts
Dear forum members,


I have to respectfully disagree with a number of things that have been stated in regards to optimal imagery on this particular display.


During the calibration procedure, a Photo Research PR650 SpectroRadiometer was used, which is an appreciably more accurate instrument than most analyzers used in the field. Patterns from DVE Pro were generated via an Oppo OPDV971H DVD player outputting at 720p. I took a display with horrendous gamma issues, crushed blacks and whites and was able to render a grayscale that was as linear as the best Sony direct-view monitors while affording the most dynamic image that the projector was capable of producing. If light output had been increased, the display would have been operating outside of its linear operating range and if it had been darkened, shadow detail would have been clipped


I was hired to calibrate a VPL-HS51 to professional standards as much can be had on it and its associated video gear. The fact that the display and screen size combination didn't produce "ink-like" blacks is not a function of the calibration but rather a function of the system's limitations. It would have been easy to increase the light output and darken the picture but this would have proven impossible without adversely affecting black-level, white-level, gamma, grayscale, color, tint, etc. In short, the display would no longer be calibrated according to system standards.


Sincerely,


Eliab
 

·
Registered
Joined
·
7,222 Posts
Quote:
Originally Posted by Eliab
Dear forum members,


In short, the display would no longer be calibrated according to system standards.
Eliab...


Keep in mind the HS-51 is the 1st to try the auto iris and the associated video processing, to expand the on/off contrast range of an LCD projector.


While the attempt may have left a great deal to be desired, aesthically it can be made to perform very well under auto iris.


I look forward to seeing some of your calibration info under iris OFF.


I would like to hear how well you were able to achieve a balanced calibration under auto iris. We understand a perfect calibration is not possible, but feedback on how close you were able to get would be appreciated.


If you follow some of the discussion in the main HS-51 thread, you will see the feedback of many folks and their calibration efforts.


I'm still looking forward to any feedback William Phelps might offer, as I understand he has been playing with the HS-51.
 

·
Registered
Joined
·
103 Posts
Discussion Starter · #14 ·
One thing I do want to make clear here folks...


I am not complaining about the job Eliab did. He has always done an excellent job on my displays and I enthusiastically recommend him to people. I also have no doubt he achieved the best possible grayscale tracking for my HS-51. The only thing I'm trying to decide is if techinical perfection is what I'm really after, and that's why I requested feedback.


Eliab if you took my posting as an "attack" or even a complaint I do apologize. I look forward to discussing the calibration with you.



Bytehoven.... sorry, didn't make it to the "cave" last night to grab those numbers.
 

·
Registered
Joined
·
1,234 Posts
Bytehoven,


Typically, a calibration is performed within the mode that is closest to the desired D6500 color temperature reference standard so that the display isn't taxed too much during the procedure. For example, if three color temperatures were offered and the "low" color temperature were to hover around 8000k and the "high" at around 20,000K, a technician would likely choose the low color temperature to perform the calibration in as it is already much closer to where it needs to be.


Such was the case with the HS-51 that I worked on. The auto iris setting wasn't even remotely close to where it needed to be from a color temperature perspective. Worse, however, was the fact that it introduced wildly discolored segments throughout the grayscale that wouldn't have been correctable. Therefore, it simply didn't make any sense whatsoever to perform the calibration within that mode.


The bottom line is this, with the auto iris on, a balanced calibration would have been an impossibility unless by balanced calibration you're referring to a subjectively balanced image that doesn't abide by system standards which is something that I don't do.

Komoman,


No offense taken... :)


Eliab
 

·
Premium Member
Joined
·
12,075 Posts
Eliab


I completely agree with your assesment that doing a calibration with the Iris set to Auto is extremelly difficult and will produce non-linear results, but with a lot of patience and trial and error it can be calibrated very close. I invite you anytime you are in the Seattle area to come by and see mine. I am not clipping head or toe room and color and gamma are pretty much dead on. Using fixed window patterns for the HS-51 is difficult with the Iris for gray scale because of the processing going on inside. We did our best with that. Afterward we had to start doing a little more subjective tweaks. Using a ramp will produce completely different results.


I think the HS51 and any other PJ that uses a dynamic iris will change a lot of the standard practices that come with calibrating. I am pretty familiar with the process and when my 51 was calibrated I had DarinP there who is also very familiar with the proper techniques for calibrating a PJ. Stacey Spears was also there who is ISF certified and has worked directly with Joel Silver and Joe Kane in this respect. Yes there are some compromises, but the pluses far outweigh any minuses.
 

·
Registered
Joined
·
23,188 Posts
Quote:
Originally Posted by Kris Deering
I think the HS51 and any other PJ that uses a dynamic iris will change a lot of the standard practices that come with calibrating. I am pretty familiar with the process and when my 51 was calibrated I had DarinP there who is also very familiar with the proper techniques for calibrating a PJ. Stacey Spears was also there who is ISF certified and has worked directly with Joel Silver and Joe Kane in this respect. Yes there are some compromises, but the pluses far outweigh any minuses.
As I basically told Kris while I was there (only partially joking), all the calibrating we have done has been to get us ready for a projector like the HS51. One which requires a certain amount of art instead of just science. In other words, I think the goal is to make the projector looks its best and the standards are just there to help with that. But when adhering to one standard means that the images look worse than ideal, I am picking the better looking images and I think a poll of viewers would prove that out.


I'll put it a different way. Let's say that you have a chair that has super smooth leather, a great look and everything you want except that there is a pin sticking up that will bite you on one buttock. Then you have another chair that has a fabric seat and look reasonable, but there is no pin sticking up. That is how I see the raised black level of not using the auto iris mode. It is a pin that will bite in a lot of dark scenes. You can say that black level (with a particular white level which is what on/off CR really represents) is only one image parameter and you can find more than one image parameter that you can't get to 100% accuracy to standards if you make the black level better. But that is like the chair. I would prefer grinding that pin down to the point that is is barely felt even if it means making the leather a little less perfect. In this case, I and everybody who I've heard from who evaluated the images for how much they like them and not for how they measure, prefers improving that black level significantly even if some other things are only 80% or 90% or whatever compared to a standard.


Remember that the standard for video black is black, as in no light. So, putting the projector in a mode other than auto iris means deviating further from that standard and so an argument that it needs to be calibrated to standards should take this into account.


Also, in short, we buy these things for our enjoyment and the ultimate goal should be to set them up the way we will enjoy them most.


That is my take and no offense meant to anybody who sees it differently.


--Darin
 

·
Registered
Joined
·
11,213 Posts
So where do you draw the line Darin?


Should we calibrate like Digital cinema is doing - for a more green operating point that is more efficient with the lamps used because that is the technology and D65 just limits it - and someone did a study that said most people don't notice that people are green?


Or should we calibrate for a Cyan white because that is even more efficient and always the highest contrast with the HT projectors. Whiter whites to achieve higher contrast of course means blacker blacks because it is assumed that you are setting up screen/projector for 12ftL white.


Should we abuse the gamma curve with things like white peaking for very Cyan at 100% - because again most people will not notice that pin sticking in their butt?


In the film world absolute black is not the standard for black - you would find the standard for commercial theatre rather pathetic actually- as most Home Theater projectors passed that standard some time ago. What you are chasing is a contrast ideal achieved with CRT that has nothing to do with the commercial reality of film.


What most people forget is if you do not get the greyscale right - then you are compromising everything but the primary colors. Some displays have such blue greyscale that you cannot even tell that the Monsters, Inc. characters actually have very detailed colour texture in their skins. Should one destroy that detail by optimizing for maximum black?


If you start allowing for subjectivity in calibrators - then you will have one artist that thinks the deepest blacks should be achieved at all cost, another artist that thinks the whitest whites should be achieved at all cost, and yet another artist that thinks bananas should be blue. That is why these standards exist - to constrain artists that get out of control with their paintbrushes.


This argument is no different than that of the last decade that said you should not calibrate your TV for ISF because it gets dimmer - and the manufacturer designed it to be bright.
 

·
Registered
Joined
·
23,188 Posts
Quote:
Originally Posted by krasmuzik
So where do you draw the line Darin?
I'm not just talking about "most people" here like with blowing lumens out and people preferring that at first impression. Do you know anybody who has watched an HS51 in iris off mode and in iris auto mode (set up as best as possible for each) and prefered the images from the iris off mode with their eyes (not instruments)? Or is your argument just theoretical?


And an HS51 in iris off mode is not even close to proper film in black level. I could ask where you draw the line of ignoring one standard in favor of another. My best answer is that if I knew that somebody was going to prefer one mode over another in the long run, I would set it up the way that they will prefer it if the display is for them, as it is their money that paid for the entertainment device. For instance, if somebody was colorblind or just preferred reds that were outside of the standards, I knew they would prefer this in the long run and it was something the display could reasonably do, then I would tell them to set it up that way (or do it myself if I was helping them).


So, in short, I basically draw the line at not making the images look worse in the long run just to meet some standard. We aren't talking about a public display here, we are talking about somebody's personal property.


--Darin
 

·
Registered
Joined
·
12,980 Posts
Greg Rogers did very well between 30IRE and 80IRE, I would concentrate on these, take a look at movies with good flesh tones and if it looks good end it there. I sure the customer would be more than happy with what he had before. When I see one of these out there for $1500 I'd love to buy it, sounds like a great machine.
 
1 - 20 of 53 Posts
Status
Not open for further replies.
Top