WMC losing one pixel column on right

[barnabas1969]

If Media Center is scaling the video to 1919x1080, then that would definitely explain why there is more moire. I do see it in things with thin stripes and checkers, like striped neck ties.

[tzr916]

For the WOW disc.... then ripped the bluray using MakeMKV... but, of course the LAV splitter is still there to enable MKV...

Sorry I don't rip. I tried LAV as the DECODER in WMC (disabled MS mpeg2 decoder) and it did fix my moire/jaggies/shimmering/de-interlacing issues but had major side effects- no protected content, live Tv intermittent freezing, video stutter, audio sync...

Looks like I can copy indiviual m2ts files into my Videos directory and WMC will play them native. I just need to figure out which file is the overscan file on the AVCHD disc???

[tzr916]

If Media Center is scaling the video to 1919x1080, then that would definitely explain why there is more moire. I do see it in things with thin stripes and checkers, like striped neck ties.

I knew I wasn't crazy

[tzr916]

Sharpness & Overscan mp4 here- http://ge.tt/30qOwZY/v/0

It is 00076.m2ts on the AVCHD disc.

Played both files on my Win7 Pro 64bit HTPC to 55" LCD Tv verifies that the right most line is MISSING.

[STC]

If Media Center is scaling the video to 1919x1080, then that would definitely explain why there is more moire. I do see it in things with thin stripes and checkers, like striped neck ties.

I knew I wasn't crazy

Comparing an identical recording from a TiVo HD and WMC where moire is evident, they are pretty much similar in the artifacts they display.
I'm probably being subjective only able to compare my compressed iffy quality cable feed between devices, but the problem really isn't noticeable for me.

It would naturally be great to resolve but I think MS left it like that for a reason (read: unresolvable bug / prevents some other problem in MC).

[tzr916]

...Comparing an identical recording from a TiVo HD and WMC where moire is evident, they are pretty much similar in the artifacts they display.
I'm probably being subjective only able to compare my compressed iffy quality cable feed between devices, but the problem really isn't noticeable for me...

I had Tivo HD for a year and a half with the same cable card, same provider, same feed, same display, same AVR, same hdmi cables. The Tivo NEVER had these picture issues. I notice it in almost every Tv show I have played in WMC. It's annoying and distracting in most (75%), hard to spot in some (20%), and horrendously unwatchable in a few (5%).

[Marc_G]

The "funny" thing is I got a nice TV specifically so I could do pixel perfect playback of my (ripped) content. Previously with an old 1080i set, none of this mattered. It all looked a bit slushy, all the time, as a system limitation.

Now things are perfect EXCEPT through WMC. Sigh.

Hopefully there will be a fix we will learn about. I've got posts here and on AVS Forum. Where else should I cast my net, to make sure we get as many contributors involved as possible? Maybe the Windows Experts Exchange (as in, the official MS board)?

Marc

[STC]

...I notice it in almost every Tv show I have played in WMC...

As a test try upping the sharpness of your GPU by around 15% It has a big effect on moire and seems to flatten out the jaggies.
This 'matches' STB sharpness outputs as GPU sharpness set at zero is actually too low.
With ATI, 15% is the sweet spot., with my TV set at 0.

It's weird I know but filtering sharpness correctly at the GPU is far more beneficial then doing it at the TV.

This info may help.

[tzr916]

As a test try upping the sharpness of your GPU by around 15% It has a big effect on moire and seems to flatten out the jaggies.
This 'matches' STB sharpness outputs as GPU sharpness set at zero is actually too low.
With ATI, 15% is the sweet spot., with my TV set at 0.

It's weird I know but filtering sharpness correctly at the GPU is far more beneficial then doing it at the TV.

This info may help.

Nvidia driver has Edge Enhancement and Noise Reduction. I have them both set to Use Player (disabled) because when I try playing with them, it just makes the picture worse (in other ways).

[Marc_G]

...I notice it in almost every Tv show I have played in WMC...

As a test try upping the sharpness of your GPU by around 15% It has a big effect on moire and seems to flatten out the jaggies.
This 'matches' STB sharpness outputs as GPU sharpness set at zero is actually too low.
With ATI, 15% is the sweet spot., with my TV set at 0.

It's weird I know but filtering sharpness correctly at the GPU is far more beneficial then doing it at the TV.

This info may help.

I will play with these later. However, if I can't get the pixels to match up to the screen, sharpness adjustments are not going to have the desired effects. My first goal is to eliminate any overscan, even a single pixel width. Clearly something is funky in WMC. Hopefully we can find a solution, because though many who test for it find the missing pixel width, not ALL do, so there must be some path to correcting it.

I can use MPC-HC to get around it, but would rather not...

Marc

[tzr916]

...Hopefully we can find a solution, because though many who test for it find the missing pixel width, not ALL do, so there must be some path to correcting it...

If you are talking about my post:

I played this 1920x1080 mpeg2 ts in WMC and it shows all pixels, none missing-
http://www.w6rz.net/overscancrop.zip

Then I must correct myself because using that test pattern it is too difficult to see the one pixel line. Using the AVCHD test pattern reveals the flaw in WMC.

Was there another person NOT seeing this problem? Where's the post?

[Marc_G]

I'll need to review the posts here and on AVS, but over on the Media Browser community, a kind user (JeremyFr79) has been trying to help me out, and he doesn't see it.

http://mediabrowser3.com/community/inde ... entry31548

I can't imagine that in the entire history of WMC7 (&8 apparently) I'm the first person to run a test pattern!

[STC]

For the record I do see it.

[Marc_G]

For the record I do see it.

You do see the rightmost column, or you do see the symptom I'm reporting?

And, I'm curious to know if you are set up as a flat panel, or set up as a TV.

Thank you for your participation!

[STC]

I see the line of pixels missing on the right.
59.9Hz.
It's setup as TV. As flat panel you will get overscan.

[tzr916]

I'll need to review the posts here and on AVS, but over on the Media Browser community, a kind user (JeremyFr79) has been trying to help me out, and he doesn't see it.

http://mediabrowser3.com/community/inde ... entry31548

I can't imagine that in the entire history of WMC7 (&8 apparently) I'm the first person to run a test pattern!

I would respectfully ask JeremyFr79 to download this exact test pattern http://ge.tt/30qOwZY/v/0 , save it to his My Videos folder, play it using WMC, and walk up to his screen and look very closely at the right hand side. Then play the same test pattern in third party software, walk up to the screen, look at the right side, and compare that to WMC. It's easy to say he "hasn't run into that issue, all my stuff seems to play pixel for pixel and it would be more than obvious on a 106" screen" until he actually plays a real test pattern and walks up to the screen.

I am seeing it on my laptop and my HTPC. I am not surprised that this went under the radar for so long. I've come to the opinion that the average people who use WMC for playback are not as picky as those who have experienced the superior quality of third party players/decoders/hardware boxes (like Tivo HD). They watch their WMC Tv shows and are happy with a good HD picture and leave it at that.

I've only had WMC for less than 6 months and this picture problem along with daily guide/record schedule babysitting is gnawing at me. I can live with it because 6 tuners and no monthly box fees for 4 Tv's. But I am clambering for an easy to setup third party solution that has LAV decoder, cable card encrypted channels, live Tv, free & better guide, smarter recording scheduler, good show info/organization, extender support (xbox & echo), and commercial skip.

[STC]

I did some more testing.-

For those with the AVS test patterns, a good set to try are the horizontal resolution patterns in miscellaneous, especially single pixel.
Comparing the single pixel pattern with TMT, WMC shows the pixels blending into the middle of the screen and becoming lined up 1x1 both far left and far right.
If you try the two pixel patterns, the results are less obvious but still there.

It's still very hard to see the problem for TV source material. I'm not going to worry too much As long as my blu ray playback is 1x1 through TMT, I'm okay with that.
It shows the 'issue' to be certainly hard coded into MC. I very much doubt there is anything anyone can do. It's not a case of adding a line of pixels in the registry unfortunately.

[Marc_G]

I concur with what STC reports.

I live in hope that there is a reg key or something that would undo whatever is doing this.

Anybody got a line to someone in MS who might bring inside knowledge to bear?

I know MC hasn't been on the hot development list for a while (understatement of the year award), but maybe someone somewhere has a line on this...

[STC]

Hmm, perhaps I had a brain fart with my last set of comments:

Thinking about it, the visual artifacts you see in the 1 pixel pattern are to be expected if the whole picture left to right was 1 pixel 'short'. You'd see the transposed shift toward the center of the screen (half pixel over half pixel) and it blend back out again to 1x1 (only these pixels have now moved 1 pixel to the left).

/edited

[STC]

Playing some DVRMS files in TMT (I wasn't aware that TMT played them until today - I am fortunate to be copy free) and also Media Player, I roughly queued up the same files in Media Center and compared one to one.

I could not see a difference between them especially from the seating position.
Up close to the TV, I may have been able to tell with the F1 Qualy session timer top center screen (black icon). It had a slightly fuzzier edge but only very slight. Moving away a couple of feet I could not tell.

Maybe a good test that others can try.