I can confirm the same happening on an i5-3570K (HD 4000) in an Asus P8Z77-M. And I'm also using Windows 7 64 bit, and Windows Media Center for Live TV and recordings. It happens on a few of the HD channels. The moving parts of the picure show heavy corruption for about 40 seconds and then it stabilizes. When I skip back in the time shift buffer, the corruption starts again and stabilizes after 40 seconds.
The problem still exists with driver version 2761.
Is this issue currently under attention?
Let's try to narrow it down:
I currently have installed the generic x64 driver 2761 from intel downloadcenter. The issue happens when switching from SD to HD (720p50 or 1080i50. I'm living in 50Hz land, i.e. outside US).
It also happens when watching a HD video in full screen and then switching to windowed mode of media center.
All this without any additional codecs, just stock Windows (up to date). Especially not TMT or anything else to watch BDs (that will come later, after verifying the basics). TMT does install it's codescs, I'm just not aware that get used for life tv playback esp. in WMC.
Can someone from Intel comment on the 'known issue' in the release note, if it matches this symptom?
No solutions in this post, just more problems. I have the following specs:
MOBO: GA-Z77X-D3H (Gigabyte)
CPU: Intel Core 7 i3770k (Intel HD 4000)
TUNER: Ceton InfiniTV 4 PCIe
OS: Windows 7 x64 SP1
I have similar issues with both Live TV and Recorded TV in Media Center and Windows Media Player. Some recordings have the initial corrupted frames for a few seconds before stabilizing, as you mentioned. Visually to me it looks like macroblocking and pixelation. But even after it stabilizes I have "tearing" between between screen or camera transitions, some stuttering, and flickering in the back ground.
Some streams (live and recorded) play back flawlessly. Additionally, these problematic recordings play back flawlessly on my other HTPC (a Core 2 Duo build w/ Geforce 9400).
I am assuming this is Intel's flavor of the 29/59 bug, well written about on the web. Unfortunately, most literature discusses this issue as it pertains to various ATI and Nvidia graphics cards. I have found no information about how to deal with this issue w/ this brand new i7 setup. The reason I feel this is the 29/59 bug is because I spent a month trying to resolve it on my Nvidia based PC, and did so successfully. That is why I think it plays back fine on that HTPC.
I don't know anything for sure about the known issue. However, the wording may be a little misleading. For instance, I know my playback resolution is not changing (fixed at 720p). But the resolution of the incoming media stream is probably changing (between interlaced and progressive). That condition IS what triggers the 29/59 bug.
Sorry for the long post. This is the first forum posting that seemed similar to what I am seeing, so I dumped all the info I could think of.
I think that's something else. My corruption is not just for a few seconds, it's more like 2 mins.
And yes, it happens any time you change 'the resolution', which can be a switch from SD to HD, or from HD 720p to HD 1080i, or if you resize the playback window.
BTW, I'm currently not talking about the 29/59 issue, since all my (current) content is either 25p, 50p or 50i and my screen refresh rate is fixed to either 60(when using a desktop) or 50 (when using the TV).
The issue isn't there with an ATI Card, nor is it there if I use any older intel system, e.g. a dg45ID. It is purely IB related.
I'm also using a clean Win7 install without codecs, just Media Center for Live TV. Because I have a different digital TV provider I think we see slightly different symptoms. Most of my HD channels are OK, only some show the video corruption and then it lasts 40 seconds. But still basically the same problem I guess. First we see corrupted video, and after quite some time the channel is displayed correctly.
Can you confirm that corruption is seen in moving parts of the picture? Static parts are OK.
Can you also confirm than the corruption starts again after rewind / fast forward / skip back / skip forward?
When pressing pause and then play again, the picture stays OK.
I have tested on Intel Graphics HD 1000 and there all TV channels are displayed flawless.
Yes, that is the same symptom.
Skipping/rewinding and resizing the window (e.g full screen to window and back) will cause it. Also switching from e.g. an SD channel to a HD channel.
Switching between HD channels of the same type (e.g. 1080i) won't show corrunption, only switching between HD of different types: 720p to 1080i or so.
Depending on your TV provider, you might want to check if you actually have different types of HD, else just try switching from sd to hd.
-the use case is windows7sp1 x64 media center
-the corruption exists for live-tv 1080i50 content when using MS-DTV video decoder (which uses DXVA)
-same corruption when playing back this type of content
-the corruption lasts for 40s, whenever the video is started, skipped, or window resized.
-720p50 content seems OK with the driver 2761
The DH77EB is perfect (besides the 1080i jitter).
I use it with a Core i7 3770T - passively cooled with the "HFX mini Metal" case and a Digital Devices DuoFlexS2 TV card.
I read that only protected channels with 1080i (SKY, HD+) have the problem. FTA channels are fine (source: http://intelopenport.hosted.jivesoftware.com/message/71271#71271).
I also read that disabling DXVA solves the problem, well, ok, its a workarround until Intel fixed the drivers (source: http://intelopenport.hosted.jivesoftware.com/message/99355#99355).
Unfortunately I don't have much time to test this at the moment. But playing recorded TV shows with VLC player is fine.
Nice to hear that the board is otherwise OK, incl. support for the DD card. Did you have to patch the FW of the DD card? I heard there were issues with their core for the 7 Series chipsets.
Anyway the other posts you refer to:
This Jitter has nothing to do with crypted or not, since the content is always decrypted by the time it reaches WMC.
The other post is the well known old ATi issue, which Intel never had, but now, months after Ati solved it, has found a new home in the IB DXVA driver. Interesting.
Might actually be that MS is using some feature of DXVA that was software emulated in earlier Intel drivers/chips. Now maybe Intel decided to actually try to do it in their new HW but are experiencing the same problems ATi had months/years ago. Let's see if they care to fix it. If not, their media series boards (the concept of which I really like) are just another waste.