Syntax Olevia 532H BAD PQ on 480i/p sources

UPDATE 11/8/2006: Syntax WILL NOT FIX THIS BUG! See end of thread for more info.

I wanted to break off this discussion from the rest. I am presuming that this defect will apply to all ATI based 5XX series Syntax Olevia, but I have no way of confirming at this time. The original conversation started here.

I believe this will include 527, and 532 models.

http://www.avsforum.com/avs-vb/showt...&&#post8230142


My setup is as follows right now.

Sony DVR for ATSC
Pannasonic RP-56 Progressive scan DVD
XBox
All connected through analog component ( only 1 port so I have to switch around).

532H Firmware A30-2.A19.A10.A01.A19

I have owned a Benq 6200 XGA DLP, Toshiba 42HXX ctr RPTV, and a Sony 1251 so I have some idea of what I'm looking for and the defect to which I am referring to.

Disclaimer: I like the HDTV PQ, it's great and I believe that Syntax will be able to correct in firmware this error

What I am seeing affects all inputs of 480p/i sources no matter how clean or input ( tried component and s-video )

1) Blurring / Noise reduction: This only affects motion and not static scenes. It's atrocious since it's low-grade and I can even see ghosting between scene cuts. Any NR worth it's salt should at least reset on scene changes!

2) Jumpy / motion weirdness: This is obvious enough that even my wife recognized it right away. It's probably more objectionable than the blurring since it really pulls you out of the image.

None of these problems affect up-scaling/HDTV on analog or digital inputs. Feeding the DVD into XBMC and playing at 720p produced a stellar picture that I could in no way complain about.

I tried to capture the effect with a dSLR on ISO1600 ~1/60'th of a second. You can clearly see in some shots up to 5 frames are being blurred together. Looking at the surrounding you can see it's not from camera shake, and from what I can see this is an accurate portrayal of what the motion blurring looks like.

Panasonic RP-56 Progressive scan through Component




Interference pattern seen on some of these is not seen on the TV, it's a problem of photographing a LCD with a CMOS camera.




I'll post some HDTV/up-scaling shots soon


NOTE: If someone wants access to the high resolution originals PM me and I'll send you the link, I don't want to trash the connection that they are on by posting them here. I don't think the defect is any less visible on these shots.

Added shots from nemo taken at 1/100'th of a second.

537 - Dunno, probably not since it's a different chipset

Interference.... no, the interference pattern is between the camera and the LCD, that red/green patter is not visible when I took the picture

Update

Called and spoke to techincal support. They will bring these images to engeneering.

Uploaded HDTV images, as you can see HDTV sources are unaffected and they are razor sharp.

What firmware version, what resolution, what "mode" are you using?

No, just 527 and 532 as far as we know.

I would be willing to work with anyone from the company to reprodce the problem and test any solutions.

I have every confidence that this can be fixed in firmware and the HDTV quality is great on this set. If I were to venture a guess I would say that the highest filtering setting are somehow being applied to all 480 sources rather than to just VCR or other low quality sources.

I can now confirm that the effect is not limited to analog sources. I have a Sony DVR and got a HDMI cable. If I set it up to 480p I can clearly see the effect on all type of video, but if I hardcode it to something higher like 720p or 1080i it appears to go away.

I think the same filtering is being applied to HD, but because the pixels are so much smaller the effect is not as bad on the picture. I can not get any pictures of the effect on HD.

I hope to start a dialog with syntax over the next week to get this issue resolved.

Ok here is what I know so far..

1) The severity of the effect is related to the input resolution. 480 lines is the most affected and 1080 is the least affected. While easily visible at 480 I have to strain to see it at 1080. It does not matter the input used. I have tried component, VGA, s-video and HDMI at this point.

2) It's not a problem with the panel itself. Running my PC through the VGA port at native resolution I was able to run synthetic benchmarks that show the panel was working perfectly with another gorgeous picture. I was unable to make the display do anything that was worthy of complaint.

As best I can figure it's some sort of default filtering that the ATI chip is doing, running the unit at 1080 or at native resolution will reduce the effect.

Ditto, I need to know by the end of the week or I'll have to bite the bullet and eat $150 on returning the set. It's really agravating since the set looks great on many sources, it's just this "clayface" filtering really bothers me.

I have 1 or 2 more things to try, but I think this will come down to Syntax releasing a firmware that correct the problem or exposes the probles setting to the users.