or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › ruby convergence / registration specification
New Posts  All Forums:Forum Nav:

ruby convergence / registration specification

post #1 of 12
Thread Starter 
I just received a mostly new (~50 hrs demo) Ruby and set it up for the first time. While I'm generally quite pleased with what I'm seeing (compared to the current state of my VPH-1272Q), I definitely can see panel mis-registration in HD video. I was just watching some of the U.S. Open tennis, and looking at the court lines the horizontal ones are white and crisp, while the vertical ones are actually 3 colors... red on the left, white in the center, and aqua on the right.

Looking closely at the built-in test patterns in the lens adjustment, the horizontal lines look to be exactly 2 pixels wide with only the slightest hint of blue on the bottom. The vertical lines, though, look to be almost 4 pixels wide, with the same red / aqua boundaries. This pattern is pretty much the same across the screen (as opposed to being only at the edges).

My question is this... does anybody know what the specification is for panel registration to be deemed acceptable by Sony? I have read of another that sent their vw100 in for a 'convergence check' and that it was adjusted to 'below spec'... but without mentioning exactly what that 'spec' was. I hope that I'm entitled to have this covered under warranty, but don't want to have them send it back saying that it is already within spec.

This is my first digital... as a CRT owner for many years now I'm used to watching for these type of 'halos' as a sign that it's time to tweak the convergence. I certainly love the fact that I was able to set this up (in a temporary position right now) and get the basic adjustments (zoom, lens shifts, focus) done quite quickly (compared to the hours and hours it took to focus and converge the CRT the first time.) In the end, though, I'm going to be less than happy if this amount of error is something that I'll just have to accept with this technology (or until I can afford to upgrade to the vw200).
post #2 of 12
Thread Starter 
Ok... so lots of reads and nobody seems to have any answer on what constitutes out-of-spec registration. Maybe if I ask the question in a different way...

Those of you that have had the convergence / registration of a Ruby adjusted by Sony... how far out was the convergence before adjustment? A Full pixel? More? Less?

Alternatively... what is the level of mis-convergence that Ruby owners are accepting as normal? .5 pixel? 1 pixel?

I've been reading with interest the speculation about how the VW200 is going to do the convergence adjustment. My set would be almost perfect if I could just move the entire red field one pixel to the right, which seems like it wouldn't be a problem at all to accomplish in the software. (Heck, maybe this possibility is even hiding somewhere in the current maintenance settings.) I presume that the real challenge is getting sub-pixel adjustment.
post #3 of 12
Sent mine out today to have misconvergence checked/fixed. Will keep you posted
post #4 of 12
Thread Starter 
Thanks. How far was it out?
post #5 of 12
Quote:
Originally Posted by GKevinK View Post

Ok... so lots of reads and nobody seems to have any answer on what constitutes out-of-spec registration. Maybe if I ask the question in a different way...

Those of you that have had the convergence / registration of a Ruby adjusted by Sony... how far out was the convergence before adjustment? A Full pixel? More? Less?

Alternatively... what is the level of mis-convergence that Ruby owners are accepting as normal? .5 pixel? 1 pixel?

I've been reading with interest the speculation about how the VW200 is going to do the convergence adjustment. My set would be almost perfect if I could just move the entire red field one pixel to the right, which seems like it wouldn't be a problem at all to accomplish in the software. (Heck, maybe this possibility is even hiding somewhere in the current maintenance settings.) I presume that the real challenge is getting sub-pixel adjustment.


From past posts when the Ruby was much in the news here, I think the spec is at least 1pixel of not 2 pixels, if memory serves me.
post #6 of 12
There is no real published spec...not that Sony will admit to. It is more of a case by case basis really. Though annoying, that actually isn't too bad considering some I have seen. You could send it to Sony and "demand" the optical block be swapped, but it would be a crapshoot as to whether your new one would be better, worse, indifferent, etc...
post #7 of 12
Well I put "demand" in quotations as I know how they operate. The first thing they tell you is that they don't guarantee perfection and it may end up no better. So by "demand" I meant more insist I suppose. That's all.
post #8 of 12
Thread Starter 
Quote:
Originally Posted by CharlesJ View Post

From past posts when the Ruby was much in the news here, I think the spec is at least 1 pixel of not 2 pixels, if memory serves me.

In reading the Ruby calibration thread, I did happen to follow a link to some archived posts which mostly agree with your memory. It sounded like Sony tried initially to say 4 or 5 pixels was fine, but then maybe backed off to 2 or 3.

It boggles my mind that a level of misconvergence that is deemed acceptable would result in the built in test patterns producing non-overlapping lines for different colors (which would be the case if it is over 2 pixels, I think). If my ~1 pixel shift for red is representative of the average level of misconvergence in the Ruby, then I guess it is what it is. There are only some specific video conditions that cause me to see it (like the HD tennis I mentioned), so I guess I can deal.

Thanks
post #9 of 12
Quote:
Originally Posted by GKevinK View Post

In reading the Ruby calibration thread, I did happen to follow a link to some archived posts which mostly agree with your memory. It sounded like Sony tried initially to say 4 or 5 pixels was fine, but then maybe backed off to 2 or 3.

It boggles my mind that a level of misconvergence that is deemed acceptable would result in the built in test patterns producing non-overlapping lines for different colors (which would be the case if it is over 2 pixels, I think). If my ~1 pixel shift for red is representative of the average level of misconvergence in the Ruby, then I guess it is what it is. There are only some specific video conditions that cause me to see it (like the HD tennis I mentioned), so I guess I can deal.

Thanks


How large is your screen and how far away do you sit?

It's been a while since I put of the cross hatches, 114" and about 13ft away, but I don't think I could see, at least not well, those 1 pixel misconvergences; yes, up close, another story.
post #10 of 12
Thread Starter 
Quote:
Originally Posted by CharlesJ View Post

How large is your screen and how far away do you sit?

It's been a while since I put of the cross hatches, 114" and about 13ft away, but I don't think I could see, at least not well, those 1 pixel misconvergences; yes, up close, another story.

My current screen is 1.85:1 91' diagonal 80" wide, and the 'eye position' of the first row is right about 10'... so I'm at 1.5 times the screen width. With the built-in test pattern I can easily see the red on the left and what looks like cyan on the right. Looking more closely I can see it is almost a full pixel. In video I don't see it most of the time, except I did see it clearly and consistently while watching tennis in HD (the court lines provided the same sort of high contrast edge that the test pattern has.)

I guess I'm also thinking ahead to a CIH future, where the horizontal elongation is going to emphasize this visibility.
post #11 of 12
Test patterns are always easier to see this on. The real test is whether it is visible during the video/movies (which it sounds like it is not). I have tested lots and lots of units and this isn't all that uncommon.

If it bothers you though, I would just call Sony and see what can be done (doesn't really matter what any of us say...it's your projector!).
post #12 of 12
Thread Starter 
I believe I've successfully identified a workaround!

I'm just getting around today to going through the configuration of the Lumagen VisionHDP that I got from Clarence earlier this week. Since I'm planning on continuing to use the component switching of my Lexicon MC12, this brings all of the adjustments of the HDP which pertain to the component signal into consideration.

My particular convergence issue with the Ruby is that the entire red field is nearly a full pixel offset to the left, as viewed in the Ruby internal test patterns. What I've done is utilize the YC delay offset capabilities of the HDP to delay the red channel... and now test patterns from my source components (SD DVD at 480i or 480p and 720p or 1080i from my HD Tivo at the moment) show essentially perfect alignment of the vertical lines as well!!!

Obviously, I've really lucked out in that the particular way in which my Ruby was misconverged is adjustable with this particular parameter. I'm SOL with 1080p sources... but the HDP doesn't handle those anyway. I'm guessing this discovery will essentially delay the pressure I'll feel for quite awhile to upgrade to one of the new convergence-adjustable projectors... maybe even for years.

Though the exclamation is getting tired now... WOOHOO!!!

Kevin
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › ruby convergence / registration specification