I7-2600K Intel HD Graphics (igdkmd64) Crash Win 7 x64 when using HDMI

[idata]

Hi,

Getting frustrated - help please. Every time I try to use an LCD TV (using mobo HDMI-out to TV HDMI-in) it results in a BSOD with STOP 0x16—telling me that the igdkmd64 did not recover. Usually before the BSOD I get something like "Display driver igdkmd64.sys stopped responding and has recovered" several times, and them–KABOOM.

The system works fine with a VGA monitor without the HDMI attached.

This is a clean Win 7x64 (Home Premium) install on a minimal system:

ASUS P8H67 B3 (Realtec HD Audio, and Realtec Gbit Network);

i7 2600K with Intel HD3000 on processor graphics

4 Gb RAM (2x 2Gb) DDR3 1333

HD - Crucial C300 128Gb SSD AHCPI.

Supposedly (?) I have the newest HW drivers

The culprit TV's are an older Olevia 42"-1080i, and a newer 26" VIZIO 1080p. Both are HDCP compliant, and both work fine with HDMI from a DVD, A/V Receiver or cable box.

So far I have tried swapping HDMI cables, tried using a DVI-out to HDMI adapter (TVs don't have DVI ports) – same result. Tried removing one memory stick (leaving only 2Gb on a single slot) same problem. Also tried with various BIOS settings -- no help. The same things happens if I booth in Safe Mode with networking, but not in Safe Mode. So, I disabled the network drivers and devices, and rebooted normally --but didn't help, same BSOD.

Have been searching the web and forums, but nothing seems to be helping.

Any suggestions are much appreciated!

[idata]

After the BIOS Upgrade ive got VBIOS Version 2080 but i have to report some hdmi problems but it needs more testing to confirm that.

[idata]

Vbios not upgraded, still 2080 with newest Bios.

[idata]

do you have asus P8H67 mb? if so is that means latest bios did not have any affect/update to the latest vbios? btw, do you notice anything the lates bios provide but not before. thanks

[idata]

yes 2080 vbios here too

[idata]

well, the driver should work correctly no matter what vbios version is in effect..

just my opinion

[idata]

You can also run GPU-Z (although, it says that I have Intel HD 1000 instead of HD 2000 with the i3 2100 T !?!).

The new driver seems to be ok, but not for all frequencies.

About the vbios. You say that the MoBo provider must provide the new vbios. But is it Intel who informs all the MoBo vendors that it would be good to include the latest vBios into their bios upgrades ? How does this work ? I'm just curious

[idata]

GPU-Z is just software, not an almighty allknowing entity..

driver is obviously not ok for quite a large amount of people here, not even for you; would be nice if everybody interested who's following this thread tried to address ALL reported issues during his/her testing (e.g. - you're not stating what's "wrong" with other than "not all" frequencies; this can be valuable info for Robert_U)

vbios resides on some chip (same as bios) on motherboard; you can flash this chip via motherboard's vendor provided tool; so you need the tool & something to feed the tool; that something has vendor dependendent format, for you - it's black box; that's why it's not easy to change just the vbios part via DIY

[idata]

GPU-Z is only software, but fine to show you the vBios version.

I already stated that 60 Hz is ok for me. And as I don't need to run my monitor in another frequency, the new driver is an acceptable workaround for me.

But for your information, when you asked me to test at 59 Hz and 23 Hz, I already answered that 23 Hz is not allowed by my monitor and 59 Hz did only work at 1024x768 resolution.

I know how to flash my bios with @bios and my motherboard has the latest installed.

I only wanted to know in which case does the MotherBoard vendor add a new vbios version in a bios upgrade. Does Intel provide a new one to the different vendors or does the MoBo vendor develop it itself ? Because in my case, the new bios (F4, Gigabyte GA-H67A-UD3H-B3) did not upgrade the vbios version (still 2080) and for Asus, it seems to be the same.

[MSchm21]

Bios file are included within the driver package. It's up to Gigabyte, Asus, Asrock, MSI etc. to update the system bios with a newer vbios. Asrock did it and MSI as far as I know too. I did not see Asrock and MSI users with BSOD issues here.

[idata]

Chris_G:

ad GPU-Z: fine to show some things, not so fine for others, as YOU stated; are you gonna believe hw manufacturer or some "notoriously used sw" (tm)?

ad testing: why bother when you did it; syntactically Chris_G <> Chris btw.

ad vbios: it's out of your action radius while Robert_U is not

[MSchm21]

Asus posted a new Bios for P8H67M-Evo with version 1850. Someone can check this if vbios still old or updated.

ftp://ftp.asus.com.tw/pub/ASUS/mb/LGA1155/P8H67-M_EVO/ ftp://ftp.asus.com.tw/pub/ASUS/mb/LGA1155/P8H67-M_EVO/

[idata]

Hi Robert_U

Can u say me, why do you ignore Intel DH67GD user? I know, there are only 3 people in this thread, but we havn't a Manufactor as Asus. We only have this forum for support.

I regret that I slowly bought an pure Intel product.

[ROBERT_U_Intel]

Snake566977 , We are not ignoring the issues you are seeing with the Intel Desktop Boards however, we have not been able to duplicate the issue on the DH67GD board. We have been able to easily duplicate the issue on the ASUS board and the majority of the posts here are on the ASUS board so the focus is there.

That said,, we just posted a new BIOS for the DH67GD that includes the 2111 VBIOS. The BIOS can be downloaded here:

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=20107&lang=eng&OSVersion=&DownloadType=BIOS http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=20107&lang=eng&OSVersion=&DownloadType=BIOS

[idata]

Ah ok, thanks.

Then I had a wrong impression. I thought you've overlooked us, sry.

I'll testing that bios this weekend.

[idata]

Robert_U schrieb:

Snake566977 , We are not ignoring the issues you are seeing with the Intel Desktop Boards however, we have not been able to duplicate the issue on the DH67GD board. We have been able to easily duplicate the issue on the ASUS board and the majority of the posts here are on the ASUS board so the focus is there.

That said,, we just posted a new BIOS for the DH67GD that includes the 2111 VBIOS. The BIOS can be downloaded here:

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=20107&lang=eng&OSVersion=&DownloadType=BIOS http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=20107&lang=eng&OSVersion=&DownloadType=BIOS

We've tested that BIOS with newest 15223, but unfortunately it the same and a new bug (with BIOS).

With 15223 on DH67GD we have less irql_drivers BSOD and more igdkmd64.sys BSOD, again. And now with newest BIOS we have irregular, rarely BSOD without driver!! (remember: without driver the system was still stabile)

sorry still failed. 😞

[idata]

Installed an EVGA GT430 last night... works great, fixes the memory bleed, doesn't crash my computer, etc.

[idata]

As I said next year it will be fixed maybe. But then we have Ivy Bridge, so probably it won't be fixed. Or they gave up? Last official driver is already two months old and counting. It's so poor, many people here tried to help in order to locate the problem and nothing happens from such a big company. Intel can't get their drivers right. I have to buy a small dedicated card now. Bye

[idata]

What is strange is that no reviewer of the sandy bridge is speaking about this error. All are saying how wonderful the new sandy bridge are !

Did they not connect their monitor with hdmi or aren't they as impartial as they should be?

[idata]

Dear Chris_G,

I suppose most of those "reviewer" just test them on the fly and we are the "real" day to day end users and live with such problem daily.

Dear SB users,

This can be another big PR opportunity for Intel as a follow up SB recall. Anyone know anyone in WSJ?

This might be the only way make the upper management aware of such problem "still" going on and prioritize the proper engineering resources to work on our driver faster than we can complaining here. Otherwise, we will be nothing but 2nd class citizen praying and waiting here... I promise you I am not joking and consider seriously to send some info to some media news outlets that I know of and Renée James.

Dear Intel engineers,

We all spent between $100 to $300 1155 CPU plus $100 to $200 MB, $200 to $500 investmen and believe that you guys will deliver the iGPU promises. In short, I hope at the very least you agree with me that we are not getting what we bought today (after 6 months from launched). We all need some commitments and respect here. We need a RTW date now. You guys clearly checking this thread and answer questions selectively. Please let us know when the new working driver release, otherwise I will start from Renée James and goes from there.

John Wong

[MSchm21]

It will take more time I bet. Let's say thex fixed it end of June, which is quite optimistic to be honest, then you can add at least another ~4 weeks for internal testing and for the WHQL certification from Microsoft. This means we have the fixed driver by end of July in our hands. But what if they can't get it right until end this month? August, September, October I call you. Last WHQL driver is from mid-April.

[idata]

Dear Yups,

That is not an acceptable time line IMO. That's why I suggested we need to have upper management involve soon and allocate more engineering resources for such project. I also understand Robert_U and developers hands might be tie perhaps they are not getting enough resources internally. Renne leading the software and service group and I met her 6 months ago. To be honest, I afraid in the reality it is not going to be priority properly until some media follow up regarding SB recall. I welcome any ideas and hopefully we can work together to fix such problem.

Make no mistake, I am not trying to make any one life miserable I am just trying to see what we can do together and move this from open to close case as soon as we can.

John Wong