Official Sony VPL-HW50ES Owners Thread - Page 43

It is not untrue. If one could align the pixel mechanically or actually control the electron beam lighting the phosphers in a CRT FP, your statement would be true. Unfortunately we have three fixed panels. And there is nothing that one can do to align them correctly after less than perfect alignment by the factory when building the light engine. also with cheap lenses of narrow diameter, using lens shift cause chromatic aberration of the red grean and blue colors appearing as the same effect of misconvergence but causing the amount of non alignment of the pixels to vary accross the picture.

what the manufsactures do is to provide you toos to hide you from seeing it. By shifting colors where they should be located on tyhe grid. This is OK when done in full pixel increments causing only a loss of one line from the direction shifted from. using subpixel adjustments this causes using two pixels to do the job of obe with the worst case being a shift of 1/2 pixel. You pick up the space between pixels causing a loss of sharpness as well as a loss of resolution. The Sony manual warns about these affects and at least when the feature was first introduced, not to use it things really get messy when multiple zones are independently adjusted. Generally, the professional reviewers like WGR think a net benefit may be obtained by subpixel shifts done globally of no more than say a 1/4 pixel. you mauy dissagree but you are wrong.

what it doers it shut dissastified customers up. see its fixed.

For one crts use dots or pixels just like any device. The pannel itself does not care what pixel of the picture it displays. whether its pixel no 90 down and 30 over . the pixel at 91 down and 30 over is just as sharp as the one at 90. the deviced does not care.
What is important and noticable is the fact that there are 3 different colored pictures that you are watching at once, Red ,Green,Blue.
If these 3 different pictures are not alined exactly then you loose sharpness and detail.
If pixel no 90 down and 30 over is red and it has blue on top of it(misconverengce) there is no way it is as sharp with the (incorrect) blue on top of it being out of convergence. Its even just a common sense fact.
But it dont matter at the end of the day cause u can test this with the 50. and I am right,,,, been there done that for many years.
The choice, watch with one or more of the pictures out of alinment obsquring parts of the image.Tell me how thats sharper

And what difference does it make what (WGR) thinks. Thats just what that is he thinks. I would rather deal with facts. Every green pixel will never aline with every red and blue pixel. they are already corrected by the factory period. Zone correction is already aplied. What few clicks the owner makes has no effect but to make the picture sharper clearer cleaner with more detail. Pixels are not muddied by improper colors being displayed where they should not be. ie on top of an improper pixel.

The panels are aligned with a fairly wide tolerance at the various factories. Some manufacturers preset the subpixel alignments globally, but none do it zonally. Sony does not preset. And what I say is absolutely true, using subpixel destroys sharpness and resolution. Do you think you know something and Sony just put the warning in their manual for the hell of it. Subpixel alignment sucks and you obviously don't understand what it is doing and how it is done. But its a free country, if you like it and it makes you happy use it.
Edited by mark haflich - 2/5/13 at 6:06am

Just tested the panel alignment and to my eyes there were only plus sides doing it. It didn't "destroy" sharpness when the projector is not even sharp anyways. Digital sharpness is not real but fools the eye pretty good. After doing the panel alignment, I could see big improvement making white more whiter when looking at white text. Didn't affect full 100% white picture or greyscale not a bit. I also tried to find how it affects the sharpness by watching countless test picture and I didn't find any signs of degrading the sharpness. Even the nec test pic looked as sharp with the panel alignment or without.

When done the adjustments, I could see that the nec test picture transformed from "old white" (yes, like a really old room whit white walls that have turned to reddish yellow) to almost perfect white. No matter how much I tried to tweak by placing the lines on top each other, I could see some halo from either one or two of the main colors. Like blue line was wider than the red. Hard to explain.

But overall people are cough up with the convergence, from seating distance(12 feet away) I could only see difference in couple of the test pictures, which made the picture from reddish yellow to more natural white. Yes, the red panel was most out of line. Eventually, in movies I could not tell the difference and whit out the slight color shift in white text I wouldn't make any difference from that too.

Those are my initial findings, need to test some more, but so far so good.

I had something strange happen with my projector tonight. The picture flashed on and then back on and when it came back on there were lines of dots all over the image. The were perfectly spaced apart and in rows and columns. They looked like large dark pixels all perfectly spaced. After a few minutes the projector clicked and shut down by itself yet the fan was still running. The only way I could get it to turn back on was to unplug it and plug it back in. Strange. Hopefully this is a one time problem and not the start of something big. I really am loving the projector.

Show us how it destroys sharpness and resolution! Test patterns don't show any loss of resolution. Real viewing material shows no loss of sharpness or resolution. Maybe Sony's scaling is no good, because JVC has no such warning in their manual against the use of fine convergence controls. I've spoken to a JVC visual products engineer and was told point-blank that the fine convergence adjustment is provided for a reason and when used properly improves the picture in a way that you can see.

All I'm asking for is concrete evidence that sub pixel adjustments hurt the picture more than they harm it. Blowing us off by using logic like "you obviously don't understand what it is doing and how it is done" is insulting, childish, condescending, and avoids the issue. This isn't rocket science. I know how the sub pixel adjustments work and, depending on the specific implementation, they could work very well or be awful. On my JVC they've been great. Maybe they suck on Sony projectors. But if so, show me. And then explain why what's true for the Sony is true for every manufacturer's sub pixel convergence adjustment implementation.

I could see, in theory, how extreme use of sub pixel adjustments could soften the picture. But so could the misconvergence that the sub pixel adjustments aim to correct. So show us why the latter is preferable to the former. I've also seen a few reviewers' claims that sub pixel adjutments kill sharpness and resolution. But I've never seen these claims backed up by concrete evidence (screenshots, measurements, etc.). Any evidence Ive ever seen has (sometimes surprisingly) shown the opposite. I've also personally yet to see a real viewing case where the sub pixel adjustments have harmed the picture more than the presence of slight misconvergence. Seriously, show us why we're all wet. And don't condescend...this "nanny nanny boo boo, I'm right and you're wrong" stuff gets old. I'm not letting you and your aggressive, abrasive stance on practically everything projector-related off the hook easily this time. :-p This time we want real evidence and not anecdotes.
Edited by Schwa - 2/4/13 at 5:36pm

hometheater.com review is out. Highest ever score for a projector.

Schwa. I am not anti projectors and have owned probably 40 or 50 over my life time including FP CRTs. I don't have an exact count.


I am negative on several aspects of projection. Huge lens shift ranges with cheap ass small lenses. This is not a discussion of lens shift and the bad things it does. Electronic adjustment of improper projector set up. Keystone adjustment. And subpixel hiding of convergence and CA errors. I am sure there are a few more. The category is far from perfect and ladden with cost compromises. BTW I am an engineer as well as a lawyer and I have consulted for various companies in the high end audio and HT industries.

I suggest you do some research right here on AV Science forum and look up a thread called Projector Convergence and an earlier thread called Thoughts about Projector Convergence and Effects on Resolution. You should also read Greg Rogers Review in WSR where he first discussed electronic adjustment of misconverhence. I am sorry that I don't remember which projector but I think its was a Sony Ruby, not sure.

If you put up an on off pixel grid you can see how badly using global subpixel and zonal subpixel screw the patterns up. However, this pattern is obviously an extreme test and not really relevant to real life viewing Say square wave test as an analog test comparison. Real life is bad but not that bad

Zonal subpixel use scaling. Sony uses 1/10 pixel adjustments. JVC uses 1/16. The methods of implementaion are similar I have been told under source confidentuality.

If you can't see misconvergence from your viewing position one should not worry about it. But customers bitch. They go up close and see it and bitch. and return and return projectors trying to cherry pick.
Of course they probaly end up with a worse in other regards sae model but with good convergence.The point is THEY can see it. They need no sophistication to see it up close and bitch.Many post pictures showing blue misconvergence and bitch to high hell.Blue misconvergence is mostly irrelevant to picture quality. Ones eyes are very very insentive to blue as compsared to red or green and misconvergence of blue is usually not visable a short distance away from the screen. Blue also is only a small fractional compent. omething like 8% of white. I hopoe I am saying this correctly. In the old days of CRT FPs we usually defocused blue because of limitations in the output of blue tubes at higher IRE levels or drive levels, defocusing improved the gray scale by allowing for a straight color temperature liner. The defocus was not visable a small distance away from the screen. But idiotic consumers and many many calibrators, who didn't know what they were doiung, would refocus the blue and tell their customers, seee I made your projector better, but in reality all they did was screw up gray scale performance.

I frequently dis cuss with industry professionals the bitch about blue misconvergence here on AVS. And they just laugh at how it makes no difference. It doesn't need to be addressed. The solution is to just let the customers hide it. like an ostrach hiding itshead in the sand. Whatever.

Subpixel was added to shut customers up. To fix it not by fixing it but by hiding it and most viewers simply aren't knowledgeable or sophisticated enough to see how the picture has been degraded. All they see is no more misconvergence. Hikus pocus the manufacturers have done a sneaky trick and hidden. You research it. Read about. Call your JVC engineer. Did he design it. Ask him to explain how it works and whether it say err duh makes the converged lines less sharp. Ask him. Ask hime if it cost you pixels. Ask him.

THE COMPANIES ADDED IT FOR ONE REASON. To stop complaints and returns. And it has done that very much.

Go do the reading. I am simply not going to say more. You can have the last word. This is not a new subject. Like I say enjoy using it if you think it provides a net improvement. And go to bed thinking that you are right. Hug that JVC engineer Teddy bear. (I hug my dog and my wife). Whatever makes you happy. What you like is all that counts, really. Being wrong hurts no one. But in my opinion you are wrong and nothing will change the facts on which my opinion is based.After you do the reading, I would be happy to discuss this with you by phone but I am not going to post more information when there has been full discussion of this issue before.


I get tired of people not doing research and starting discussion over as if it is something new. I am talking in general and not about you. you are clearly offering your observations and relatig an what i would call a not complete answer given to you by JVC. As one gets oilder, it is hard to keep answering the same questions over and over and arguing points when ones position is correct and nonarguable and I choose not to engage further With respect to adjusting red misconvergence or CA to green as the reference, ther may be instances where one can see on say let crawl on ESPN bottom info line red misconvergence and using subpixel in specfic screen areas, score boxes, etc may indeed be net benificial focus and resolution are not important there. the letters are big. Red misconvergence there can be very annoying. Just like some of my posts.
Edited by mark haflich - 2/5/13 at 3:22pm

I'm a formally educated engineer as well but I'm not sure how that matters in the context of this discussion. And I have done the on/off pixel test and, with my projector, fine convergence adjustments do NOT cause a loss of resolution nor sharpness on this pattern.

On one hand you say that "if you can't see misconvergence from your viewing position don't worry about it." However, on the other hand, you say that customers "aren't knowledgeable or sophisticated enough to see how the picture has been degraded" when fine convergence adjustments are used. So, in essence, you're saying that we should avoid using fine convergence adjustments (to fix image flaws we can see) so that we won't suffer from image degradation that we can't see because we're not "knowledgeable or sophisticated enough?" Even you have to admit that doesn't make any sense. "Don't fix that projblem you can see because you might run into a problem that you can't see."

I've done the reading. I think Greg Rogers mentioned fine convergence adjustments in his review of the VPL-VW60 because I have that projector and I remember reading the review. However, regardless of whether his comments related to the Ruby or the Black Pearl, much has changed since then and his opinions and observations of those projectors' performance certainly can't be applied to today's projectors. Regading the other two threads you reference, the first makes not a single mention of the effects of fine convergence adjustments, and the second only has one mention.

I understand the theoretical reasons why performance should be degraded. But I've yet to see any degradation in practice. In fact, I just read the HomeTheater.com review of the VPL-HW50ES where the author said much the same thing. So stop quoting the reviewers and SHOW ME. Show me how the image is degraded. Show me a screenshot with and without the fine convergence adjustment activated and explain how the latter is better than the former. All the credentials and quotes and citations of other reviewers are not substitues for proof. And resorting to name calling and putdowns is disingenous and immature. Honestly, I want to learn. Maybe I'm missing something and I just haven't noticed it. So show me. And then I can finally make an educated decision as to whether my fine convergence adjustments really do result in artifacts that are worse that the misconvergence itself.

To my "unknowledgeable, unsophisticated" eyes, I see no image degradation using the fine convergence adjustments. If it's there, I don't know why it's so hard to get someone to show it to me...or even describe it! I'm certainly not the only one to say that the net benefit of the fine adjustments outweigh the downsides. But for some reason we get branded as unsophisticated, uneducated loons when we say this. I'd just like to see why. And without all of the insults and name calling.

I am not insulting you. I never called you anything. I was talking in generalities. I apologize if I worded something so as you thought it applicable to you specifically. And I respect any decision by you to use this feature and your conclusion to yourself that it improves things. Others have posted what the on off patterns look like after using subpixel zonal.

The last thing I want to do is spoil someone's enjoyment of their projector and I do not want to teach people how to see bad things they do not see now. Forums specializing in making people see ghosting and rainbows etc. I won't do it .Others can chirp in and do it.
Edited by mark haflich - 2/5/13 at 3:30pm

I will do it...

Because if we never did, nothing would ned to improve.

I will pick a part any projector regardless... Give me a VW1000, I'll find its faults.

SOWK. I guess they need you on that wall and want you on that wall. Go find some of those posted test patterns showing the effects of subpixel manipulation and post them
Edited by mark haflich - 2/5/13 at 3:31pm

If I use alignment, on my hw50 my red and blue are 5 click off, so does that means that those two panels are ½ pixel off ? Is that considered good, average or bad ?

Thank you

Fair enough. We'll agree to disagree and move on. And there's no need for anyone to post pictures of test patterns. I've looked at single pixel test patterns and convergence grids, and while I've seen minor color shift, I've never seen a lack of sharpness or loss of resolution. Besides, test patterns aren't what I spend my time watching, and I've never seen any type of color shift in actual viewing material even when A/Bing a freeze frame with the convergence adjust function both on and off. What I was interested in seeing is honest-to-goodness real viewing material (not test patterns) that show the negative effects of zone fine convergence adjustments. But at this point don't strain yourselves looking for it and/or posting pictures of it. It's easier to just agree to disagree and leave it at that.

I still feel that minor convergence tweaks are preferred to minor inconvergence and outweigh any resulting image degradation (mainly because I've never been able to spot such degradation). Mark disagrees. So be it. And he has a point...maybe I don't want to see the degradation. Sometimes ignorance is bliss.

There is some motion test you can do where you can see the effects, but to really see them you need two of the same model of the same projector side-by-side, one adjusted and one not.

That is fully within the specs of all manufacturers though some will not publish an official spec.

Most manufacturers provide for full pixel shifts and there are no bad consequences of shifting the timing of an image on one or more chips by full pixels. Say you are off by 2.3 pixels, simplky shift by two pixels and your error is reduced to .3. Let's suppose you are off by 1.5. Shift one pixel and now you are left with a .5 error. X.5 is the worst error. X.4 is the second worse. After full pixel shifts. the worst error you can have is a .5. and that is BAD. Its the worstr one can do. It can be hidden by using 5 clicks but that is bad. Tou have used scaling and now use two pixels to do the job of one plus now have incorporated the spece between pixerls into your pixel. So you have lost resolution (using two pixels to do the job of one) and you have lost sharpness.

If you put up a non moving pattern of single on off pixel lines and use subpixel shifting you will make the pattern go bizirk. There is no way this won't happen unless you shut off subpixel. You are no longer sending a signal to the panel that represents on off pixel lines and I would humbly suggest that any one who says he has used subpixel shifting without affecting such pattern has perhaps unknowingly shut the shifting off while testing or is not using on off patterns in the direction of the shifting.


This doesn't mean you have a bad projector. It may be great in all other aspects and I would say don't use any 5 click adjustments. Just watch it the way it is and if you don't see any misconvergence from your seat, don't worry abouit it. If you see misconvergence from there, try two clicks and watch it there.`Try it and see what looks best to you from your seat. forget what you see with your nose pressed to the screen. Try it subpixel off. Try it fully subpixel adjusted, 5 clicks in your case, and two clicks. tell us what looks best to you when watching some content.
Edited by mark haflich - 2/6/13 at 1:09pm

Mark, I'm just gonna have to show you someday. Let me know the next time you're in Cincinnati. On my RS4810 there is some slight color shifting but the pattern doesn't go berserk. If your convergence is off by 1/2 pixel the single pixel on/off lines are gonna be a mess anyway. Say red's off by 1/2 pixel and you don't use the subpixel adjustments to minimize/hide/correct it. You're gonna have red lines all through your black/white single pixel line pattern. Granted, the subpixel adjust won't make the red fringing disappear, but it will reduce its intensity by dimming the offending pixels.

I'm not trying to be argumentative because, when talking about the theory, I agree with you 100%. Those single pixel on/off lines should be a mess when using subpixel adjustments. But in reality the test pattern doesn't look as bad as you'd expect. It really looks no worse IMHO than the native misconvergence (I think all in all it looks considerably better, color shift nowithstanding). That pattern also doesn't look less sharp (again, from my experience). Maybe it's because I don't have to shift a full 1/2 pixel to fix my convergence; I don't know. But seriously, I'd love to show you sometime. Maybe this weekend I'll try to get a picture...but by then, we'll be on to something else.

To the OP's questions -- I wouldn't worry about 1/2 pixel misconvergence. That's certainly within the manufacturer's specs and probably next to invisible from your viewing position. Give the fine convergence adjustments a shot and see what you think...worst case is you like the picture better with the fine convergence off. Obviously there are different opinions about this as you can see by Mark and my bantering.

Thanks for the offer. I use to fly to Cincy twice a year and then drive to Lexington but my friends want to drive the whole trip now and we don't come through Cincy and all they care about are horses running at keeneland.

I could understand how using global subpixel might not screw the pattern up, but adding zonal would just make it a mess. What are on off pixel patterns used for. You can use them to detrmine brightness uniformity, for testing lens and screen resolution, some screens will blur the pattern through bleed, and it ha ha can be used to see if you have global and switched on. If they are doing they are designed to do, the pattern will show it. It has too.
Edited by mark haflich - 2/6/13 at 10:04am

I'm using zonal and, honestly, it's not as bad as you might think. The resolution/sharpness is fine but there's color shifting. I'd just have to show you.

I am not talking about huge losses in sharpness or drastically visable loses in resolution. Obviously you lose resoltion, maybe you cut it roughly in half or a littele more but it is not all over the screen Likewise you lose some sharpness in spots. If you sit at normal distances you will not notice it but it still occurs. BUT if you sit at normal distances you will not see the misconvergence eith. So whay hide it with the accompanying losses if you can't see it when viewing at normal distances. It makes no sense except to pacify mad customers who can see it when they put their faces a few inches away from the screen?t

I am new to the thread and really have enjoyed the detail information it provides to owners. I received my new VPL-HW50ES last Friday. After playing the Epson game and returning 21 projectors the Sony unit arrived and has met my expectations in every area of performance but one. The one area is motion artifact. The motion artifact on this projector is terrible even with interpolation turned high. I know motion artifact is some what weaker on SXRD type projectors but I was not expecting this. As bad as Epson projectors can be in many areas motion artifact was not one. Slow pans, fast moving action and slight slow head motions or movements cause a tearing effect. I have played with the entire settings Film mode etc. and I cannot reduce this to an acceptable level. Is this typical performance for this product? Have I missed something in my setup? My sources are an OPPO BDP-95 and the most current DIRECTV HD-DVR. I am using the older MONOPRICE certified silver HDMI cable. Is something wrong with the projector?
Please advice.

No, not normal. Sony is great with motion.

Can you define "tearing"? Are you saying some portion of the image at the top is horizontally shifted from the lower half?

Are you going directly from oppo to Sony? Or is something in the middle?

Exactly, it's ridiculous, since the hs50 (despite inferior dynamic iris at the time), it was a crime to choose "off".

People ignoring all the best features, FI, rc, etc etc etc just because it's not "cinema" or not "chomapure approved", pushing brightness to see last THX black spot (you don''t have to see), etc are chosing to transform a 2012 beast into a 2009 one XD (you can see the same funny set ups in TV forums).

Returning an Epson 21 times! That's not playing a game, that's a career.

Re the iris, you should turn it off when calibrating and if you notice pumping in a rare scene that trips the iris up. otherwise enjoy it.

The iris action on Epson's projectors is pretty decent. You shouldn't see it working too often. For me, their dynamic gamma was more apparent than the iris action.

The oppo and the directv hd turner are going through a mono price 4 in 2 out 3d hdmi switcher. I have had a epson 8500, 8700, 9700 and the new panasonic ae-8000 projector running through this with no issues. I will run the sources directly to the projector to ensure compatibility beause I kmow all products vary somwhat. I have the oppo set to 1080 with the 24 frame option set to auto. I selected color space to rgb because on the epson projectors this gave me the best resolution, I will try auto. Any othe suggestions.

Small world, I am in Cincinnati as well and have the HW50.

I have a similar effect on mine, but i took it to my buddies house, and it does not do it in his setup. I found i was also using older non high speed hdmi cables so i got all new ones from mono and am going to hook them up today.

I would suggest try it in a friends setup to see.


to see if we are talking about the same thing... what i notice on mine, is say i am watching cable TV... I have motion enhancement turned off because that is just what i like. It will be normal motion, then best i can describe it gets jittery for a moment, ( sound stays fluid ) then the picture catches up, and it flows again.

Is that sort of what you were seeing too ? Or different...