CurtPalme.com Home Theater Forum

[mp20748]

Register to remove this ad. It's free!

It was also done in many other threads. I had a Mike Parker Mods thread and Mike Parker Mods Forum here. and so did Nash have a very long mods and tweaks thread, where most of the test results were posted. In those threads many joined in also posting and commenting.

And with bandwidth being the biggest goal, not only were scope pattern posted, but test patterns were also posted, showing the SMPTE pattern properly resolving over 300mhz based on what the pattern generator was showing to be the bandwidth.

In fact, there were so much going on with pattern and scope readings, that the argument back then was "post screenshots", because the test patterns prove nothing. And I agree and have been saying, screenshots CANNOT be used to determine what a setup looks like. "they can only be used as an idea"

It is impossible to properly judge a setup from a screenshot, and be able to determine things like black crush, etc, because there is too many variables in between what's on the screen and what's being posted. Camera, camera skills, experience using the camera, hosting service, room condition and I can go on. And as been said so many times, "what I've seen on the screen itself is very different from what the shots show"

And in the nearly 15 years of me doing mods, there has been a very few who said they did not SEE a difference. And I've provided mods for Reference Imaging and two other operations in (one in Arizona) that also sold the Chris Stevens mods.




But now that any new mods are being offered by Gjaky, he can link you to his test results. I've burnt out on trying to prove image quality using test patterns, because you can't use patterns to show actual end results. Only the screen can show that.

.

[mp20748]

Removed my green CRT to try and tap out particles that causes flickering and blurring. Reinstalled the tube and then got further into dialing in the magnetic's, really for the first time myself. But still have not touched the convergence, so the bottom half of the screen is off. Was really waiting to change out the tube before doing this, but last night seemed good for this.

[racerxnet]

I will have the Giaky mods installed on my projector soon. Should be interesting to revisit some of the shots posted here. Looking forward to see what the picture looks like.

MAK

[mp20748]

Took some time and fixed the remote, and was able to converge things a bit

[mp20748]

[mp20748]

[racerxnet]

Blacks are crushed, details muted, and colors are off. What was it we are suppose to be looking at? What fL are you outputting?

MAK

[mp20748]

[racerxnet]

The true essence (as you put it) of whats on your screen is not even close to the directors intent. It's known that you don't have the tools to properly calibrate your projector for displaying movie content. But hey, if that's what you like (uncalibrated) and you feel good about it, that's all that matters. I've been in the printing industry for 40 yrs plus and know color very well. I was wondering if you calculated your fL or other display properties to see if you had ANY idea of what you were doing; D65, rec 601, Rec 709. Guess not. Oh well, have fun.

MAK

[mp20748]

[mp20748]

Things could be a lot better if I had a better camera and spent more time dialing in things, but for now, no Darby, peaking and cheap camera does not have sharpness

[mp20748]

[nidi]

[quote="mp20748"]

[mp20748]

[mp20748]

Oh and while we're at ti. let me clear up one other thing on this issue of Light Output, and the incredible claims of phenomenal light output.

I've been saying this over the years and using the Blend set up and Stacking as reference for large screen projected images from CRT technology.

I'll start out first by mentioning when I was at VDC some years ago on a visit as was some other who hang out here. There on the shelf Scott showed me batches of VERY HIGH RESOLUTION 8" CRT's in all three colors. And the first thing that came to mind was why not put them in a Marquee, and I did mention that to Scott. He responded with something I already knew, and that was that those tubes were not designed for the light output we're so accustomed to. They are low rive CRT's, as I've been saying all along that in order to achieve full high Resolution from CRT technology, you would need to derive the tubes using a test pattern to determine the proper light output. Not the method the calibrators use, but a method that confirms ON-SCREEN that you are actually resolving the resolution. and that is the LINE TEST. There are several of them, that are similar to the common SMPTE, but work best for resolution only.

In the commercial industry. the only way to be able to get both high light output and resolution on large projected screens, is to go with a Blend Setup, or a multi-projector blended Setup. And when high light output is the goal and a matching High Resolution is not that important, then Stacking would be the best solution. Because stacking gives twice the light output, but also lowers the resolution capability.

So for a TRUE 1920X1080P image from a CRT, you may find using the line test that your light output may be somewhere around 7 flt. Or even lower, or somewhere near that or maybe a little higher. But you'll never be able to maintain tight lines at high brightness, especially with LCP tubes. The LUG's have a smaller Pixel Dot, so they would be better when driving higher. On large screen setups, the resolution of 1280X1024 was the best for both light output and resolution, and is why most of the commercial setups used that resolution. And that was also the case when doing STEREO, because a STEREO setup would have a higher refresh rate, that would also soften the image and affect the overall resolution. But STEREO required a higher refresh.

So any and all boasting on high Light Output especially using standard or LCP tubes, would also mean an inability to past the Line Test at 1920X1080P 60 or 72hz. Simply because of CRT's inability to do both HIGH Resolution and High Gain at the HD rate of 1920X1080P. There is an exception at 1280X720P, which would allow a much higher gain drive over 1920X1080P using the line test.

So though my SETUP can max out (contrast 100) and still look sharp, it will not pass the line test. So at that point, I would need to decide which is more important to me, the gain or the resolution. And when the resolution is REALLY TRUE, you can get away which a lower drive because the image looks more fuller because there is more detail on the screen. This can be confirmed using a camera test, and is also why Ive been posting some shots on the forum, because the camera will reveal the highest resolution from the file size. Dial the contrast down, you could get up to 5mp. Lick it up higher, it will go below the 5mp.

[nidi]

So by stating the above you finally are saying that you have been using a lower light output to improve
sharpness on your screenshots?

how much lower?

some people claim that they are able to drive LUG tubes at 11-12 fl and still get all the resolution for 1080P.

how much lower does the light output need to be for LCP AND LUG to still get that resolution at 1080P?

Numbers please?

G2
Drive
and Cotrast setting

Michael

[mp20748]

[jbmeyer13]

Some things to add to this discussion:

1) No one is talking about the screen itself and that will have significant impact on light output. I've been using Stewart's ST130 material for the past 4-yrs and have used both perforated and non-perforated types. The non-perf screen pumps out 1.3 gain while the perf was 1.10-1.15. We are talking about 15-30% more light output than a 1.0 gain screen which is what I believe most people are using. Furthermore, the Stewart material is noticeable sharper than the other materials I have tested and that will have an effect when using various test patterns.

2) The testing I have done on my set up in various incarnations has shown me that the only time phosphor bleeding is really an issue is at 100IRE which is not common when watching normal content and usually when such transitions take place your eyes are temporarily blinded by the light shift so it's a bit of a moot point. Stock marquees would suffer from this issue well before 100IRE at light output levels > 10ftl because of their penchant for blooming. With stock boards you really had to dial the gain back and this is one of the reasons why guys like Craig Rounds weren't particularly fond of a stock marquee when compared to a G90.

3) My own experience tells me that to get better performance coming out of black requires light output north of 10ftl. Making out low level details AND having a decent dynamic range require sufficient light output. I think most would agree that one without the other is not an optimal viewing experience.

4) Finding the balance between gain and resolution is equal parts of art, science and preference. Something can pass various academic tests but not appear ideal during real world viewing. Kurt and I experienced this first hand while playing with the Lumagen's CMS over the past couple of years. Having the right tools, knowing how to get the most out of them and an experienced eye are the reasons guys like Kurt, Craig Rounds and Ken Whitcomb are able to really optimize total video chain performance from modified Marquee CRT displays.

[mp20748]

[racerxnet]

I thought you said you had your own probe and calibrated many customer systems! Hmm. Something smells a bit fishy..

MAK