How To: Enable SLI on pre-i7/i5 hardware

[anatolymik]

2600k 5ghz
Gigabyte P67a-UD3P
Ripjaws 2133mhz
Corsair tx850
Evga Gtx 460's
Water

I recently fixed 2 bugs. but i didnt update a the link. I'm waiting for a result of testing on server board. If there would be success then i'll test on all windows version and post a link, if not i have one more idea to find out.

 

[Jonesey I7]

I recently fixed 2 bugs. but i didnt update a the link. I'm waiting for a result of testing on server board. If there would be success then i'll test on all windows version and post a link, if not i have one more idea to find out.

Any eta on the results? I've been pulling my hair out trying to get this to work. And I can't just go buy an sli capable board as they have all been pulled for about a month, and won't be around again for at LEAST another one. My supplier won't even ship me one.

I really appreciate all your hard work and dedication. I will be watching this thread like a hawk (and my inbox).

On a side note, I'd be more than happy to "help" you test out your next beta if you think it may work with p67.

 

[Nick [D]vB]

Hi anatolymik, great to see you are still working on this, I have not checked-in here for a long time, remember you helped me with a strange nForce4 problem a few years ago. SLI is working great on my new M3A770DE motherboard, I like the new installer, much easier than the old system!

[A small tip for people with my board, and maybe others with PCIE4+16, to make SLI stable when overclocking I had to disable PCIE Asynchronous mode in the BIOS and lock PCIE speed in ratio with FSB (FSB250 / 2 = PCIE125) after I did this SLI was 100% stable]

I just wanted to check with you about installing Service Pack 1 for windows 7, has the HAL.dll file been updated to a new build, will the SLI patch need updating for it? I wonder if now Nvidia's chipset bussiness is dead if they will remove all checks and protections from their driver in the future? Until that time it is great to know the SLI patch is still alive, keep up the good work Anatolymik!

 

[anatolymik]

Hi anatolymik, great to see you are still working on this, I have not checked-in here for a long time, remember you helped me with a strange nForce4 problem a few years ago. SLI is working great on my new M3A770DE motherboard, I like the new installer, much easier than the old system!

[A small tip for people with my board, and maybe others with PCIE4+16, to make SLI stable when overclocking I had to disable PCIE Asynchronous mode in the BIOS and lock PCIE speed in ratio with FSB (FSB250 / 2 = PCIE125) after I did this SLI was 100% stable]

I just wanted to check with you about installing Service Pack 1 for windows 7, has the HAL.dll file been updated to a new build, will the SLI patch need updating for it? I wonder if now Nvidia's chipset bussiness is dead if they will remove all checks and protections from their driver in the future? Until that time it is great to know the SLI patch is still alive, keep up the good work Anatolymik!

When you update windows and hal or kernel was updated it's enough to run slipatch and push update button. Installer will make build with new kernel and hal automaticlly. Before installer i has made patch for each version windows, but with installer it's not needed anymore. Installer contains mini PE editor and contains patching rules(separate files which contains my code and rules of injecting of it). When you push install/update button installer will take latest hal and kernel from the system and according to these rules injects code that emulates x58 platforms.
But installer has static files also. It's hacked windows loader. To use patched kernel or hal without test signing mode you need loader which ignores signing. Hacked loader do the same.

 

[Nick [D]vB]

I was wondering why we did not need to use test mode any more, thanks for explaining it. I did not realise the new installer could patch any file version, that is very cool!

 

[soundood]

2600k 5ghz
Gigabyte P67a-UD3P
Ripjaws 2133mhz
Corsair tx850
Evga Gtx 460's
Windows 7 Pro x64
Water

and you have def enabled ACPI in bios?

 

[Nick [D]vB]

Don't these new intel CPUs only have one PCIEx16 controller, on the Northbridge intergrated into the CPU die? If the southbridge chipset controls all the other PCIE lanes maybe that will be a problem? I wouldn't know, I'm an AMD man... lol

 

[Jonesey I7]

Bought an Asus p8p67 pro....... no more issues lol.

 

[anatolymik]

I just updated the link. Fixed two very old bugs. need test on x32. Big thanks to RVC. I hope he'll help me sort out one more problem.

 

[Nick [D]vB]

Great! I can test on XP32, or do you need to test Win7x32?

There were a few small glitches in 3Dmark06 feature tests last time I tested on XP32,

and I got some BSOD, but it was a long time ago so hopefully it is fixed now.

ps- did you get my PM?

 

[anatolymik]

Great! I can test on XP32, or do you need to test Win7x32?

There were a few small glitches in 3Dmark06 feature tests last time I tested on XP32,

and I got some BSOD, but it was a long time ago so hopefully it is fixed now.

ps- did you get my PM?

no matter which windows it would

 

[Nick [D]vB]

Cool I will test on XP32 for you soon, first I must re-image that drive but it will not take long.

 

[juanchacin]

I just updated the link. Fixed two very old bugs. need test on x32. Big thanks to RVC. I hope he'll help me sort out one more problem.

Hi anatolymik, can you describe the 2 bugs that you fixed, it is always important to know what changes are make to the SLIPatch.

Thanks and keep up the good work my friend.

 

[Nick [D]vB]

I get a BSOD in windows XP32, just before the login screen (page fault in non-paged area)

But I get this with both 1.0 final patch builds 20.14 (30/12/10) and 20.15 (23/02/11)

Patch 0.91, build 1.2 / 20.13 (01/12/10), is working ok for me in XP32,

my HAL version is 5.1.2600.5512 (xpsp.080413-2111)

I can upload a memory-dump if it helps?

.

 

[johnspack]

Updated to new patch, seems to be more stable. I can overclock my cards in sli now, before games would crash unless at stock clocks. Still testing, but interesting!

 

[Nick [D]vB]

I have done some more testing with 0.91 on XP32 and all my games run ok. The only problem I can find is with one of the 3Dmark06 feature tests, the "Vertex Shader - Simple" test runs very slowly (the test with four spinning white dragon models). The fps reports as 400fps but the models spin very slowly, the same test runs at normal speed (50fps) on Win7x64. But I think this might be an problem even on "real" SLI systems:

http://forums.guru3d.com/showthread.php?t=266457

I have overclocked GPU, CPU, and PCIE but it is 100% stable and if this was the problem it should also happen in windows 7, maybe its just an XP thing??? All my games are running great with SLI and I have not had any BSOD yet so it is not a problem for me, I just thought it might help to know about it. I will try and test the new 1.0-final build 20.15 with Win7 x64 later tonight.

.

 

[elbubi]

I keep geting this bsod very frecuently while loging in into windows 7x64 :x

 

[Nick [D]vB]

Just tested Win7 x64 with the new 1.0 final patch (build 20.15) and it is working ok for me, tested with new 267.05 drivers.

@elbubi - I used to get freezes / BSOD at login but disabling the Afterburner / Rivatuner services fixed it for me, maybe you can try it?

 

[l33tlinuxh4x0r]

I am having problems with the latest slipatch driver and AVAST antivirus... With AVAST 5 it would boot and work pretty well but then playing games it would BSOD. Now with AVAST 6 I can't even get the computer to boot with the patch. Is anyone working on a fix? Is there any way that I can help? (Probably by alpha/beta testing a fix.)

P.S. could the patch also be updated to windows 7 SP1??? Maybe the patch could take the current files and modify them instead of being released with already modified binaries.

 

[anatolymik]

I get a BSOD in windows XP32, just before the login screen (page fault in non-paged area)

But I get this with both 1.0 final patch builds 20.14 (30/12/10) and 20.15 (23/02/11)

Patch 0.91, build 1.2 / 20.13 (01/12/10), is working ok for me in XP32,

my HAL version is 5.1.2600.5512 (xpsp.080413-2111)

I can upload a memory-dump if it helps?

.

try to install with additional boot

 

[anatolymik]

I keep geting this bsod very frecuently while loging in into windows 7x64 :x

http://img684.imageshack.us/img684/4127/bsodo.jpg

under patch? can you try to boot into safe mode under it to see what will happen

 

[anatolymik]

Just tested Win7 x64 with the new 1.0 final patch (build 20.15) and it is working ok for me, tested with new 267.05 drivers.

@elbubi - I used to get freezes / BSOD at login but disabling the Afterburner / Rivatuner services fixed it for me, maybe you can try it?

i keep it in my mind.
so driver works ok or not?

 

[Nick [D]vB]

The 267.05 driver is still good, it works well for me in Win7x64 with 1.0 final (builds 20.14 and new 20.15)

In XP32 I get BSOD with any driver using patch 1.0 final (builds 20.14 and new 20.15) I always use additional boot method.

But 0.91 is working great in XP32.

 

[anatolymik]

The 267.05 driver is still good, it works well for me in Win7x64 with 1.0 final (builds 20.14 and new 20.15)

In XP32 I get BSOD with any driver using patch 1.0 final (builds 20.14 and new 20.15) I always use additional boot method.

But 0.91 is working great in XP32.

did you try with general? in x64 you have problems or you didnt try?

 

[Nick [D]vB]

I have no problems in Windows 7 x64 (I don't have XP 64bit to test)

In XP 32bit I get BSOD with 1.0 final using both General and Additional boot mode.