Folding Forum

rwh202 wrote:

Aurum wrote:

Applefreak wrote:Note: Even though the GTX 970 is on a x1 slot it almost always has the highest Estimated PPD. Is Maxwell better at folding than Pascal
11:42:05: GPU 0: Bus:6 Slot:0 NVIDIA:5 GM204 [GeForce GTX 970] (Bus runs at x1)
11:42:05: GPU 1: Bus:4 Slot:0 NVIDIA:5 GP104 [GeForce GTX 1070] (Bus runs at x8)
11:42:05: GPU 2: Bus:4 Slot:0 NVIDIA:5 GP104 [GeForce GTX 1070] (Bus runs at x8)
11:42:05: GPU 3: Bus:4 Slot:0 NVIDIA:5 GP104 [GeForce GTX 1070] (Bus runs at x16)

More likely is that the slot numbering / naming is wrong. GTX 970 will be ~300k and 1070 at ~550k. What happens if you start each GPU slot up one at a time and see which one loads and heats up? The latest beta clients are meant to be an improvement in this regard, but the previous client was rarely (never?) correct.

I agree it's odd. I moved it to a x16 slot with 16x riser. Then all the bus speeds switched. Each GTX 1070 is a different brand so that helps sort them out: Asus, EVGA & Gigabyte.
I got the bus speed from GPU-Z. It should match since I only have one GTX 970 on the MB.
I lost ATT internet and everything's a mess. I'll look at it again later after all GPUs have new WUs.

Slowfood wrote:

SombraGuerrero wrote:I tend to air on the conservative side, but honestly, if you're able to fold with the driver you currently have, I wouldn't mess with it. If you're experiencing the bug with 0x21 though, I would recommend pinning your kernel and installing the hotfix driver if you're comfortable with manual driver installation. In my experience, pinning the kernel is necessary, especially if you want to keep your system stable, because every time the kernel upgrades, manually installed Nvidia drivers tend to break.

I'm asking, because i experience the 0x21 bug. I'm using the NVIDIA-Linux-x86_64-375.26 driver. Manually installed. The 1050 Ti is offically supported since the 375.10

I figured. We just haven't had many *nix users chime into this conversation, so I didn't want to make any assumptions. I just realized that everything I've said up to this point is in error because I had it in my head that Nvidia had a Linux package available of the hotfix. This does not appear to be the case. They did release a beta driver just four days ago that, by revision number, I would assume includes the hotfix code. Perhaps you'd have luck with that.
Beta 378.09. From the looks of it, Linux support for the other cards in the 10x series has followed the same release cycle, so the older drivers that worked on the Windows side aren't available either. This beta is probably one of the only potential paths forward for now.

SombraGuerrero wrote:

Slowfood wrote:

SombraGuerrero wrote:I tend to air on the conservative side, but honestly, if you're able to fold with the driver you currently have, I wouldn't mess with it. If you're experiencing the bug with 0x21 though, I would recommend pinning your kernel and installing the hotfix driver if you're comfortable with manual driver installation. In my experience, pinning the kernel is necessary, especially if you want to keep your system stable, because every time the kernel upgrades, manually installed Nvidia drivers tend to break.

I'm asking, because i experience the 0x21 bug. I'm using the NVIDIA-Linux-x86_64-375.26 driver. Manually installed. The 1050 Ti is offically supported since the 375.10

I figured. We just haven't had many *nix users chime into this conversation, so I didn't want to make any assumptions. I just realized that everything I've said up to this point is in error because I had it in my head that Nvidia had a Linux package available of the hotfix. This does not appear to be the case. They did release a beta driver just four days ago that, by revision number, I would assume includes the hotfix code. Perhaps you'd have luck with that.
Beta 378.09. From the looks of it, Linux support for the other cards in the 10x series has followed the same release cycle, so the older drivers that worked on the Windows side aren't available either. This beta is probably one of the only potential paths forward for now.

I tried 378.09 but it wouldn't work for me. Fell back to 370.28.

Yeah, unfortunately that's not an option for 1050 Ti users.

SombraGuerrero wrote:Yeah, unfortunately that's not an option for 1050 Ti users.

Yes, and no. It is possible to edit the INF file to make it work. Mentioned in this thread somewhere.

Yeah, except we're talking about Linux here, and it doesn't look like the old drivers are packaged for Linux.

I just downloaded 370.28 driver off of Nvidia website. You have to be on Linux when you go to the site. hen I tried Windows 7, I could not find them. Good luck.

Windows driver WHQL 378.49 released. Fixed: [Folding @ Home] Work unit errors occur
Linux driver 378.09 - BETA fix still missing! Should nvidia be pushed somehow to also include the fix in the linux driver 378.xx?

Core_21 v0.0.18 released to beta for testing!

There's hope. FAH has been working on a new version of FAHCore_21 which avoids the bug in nVidia's OpenCL driver. Once that revision has had some time for beta testing, it will be released to everyone. Hopefully it will also avoid the performance degradation introduced in the hot-fix driver.

foldy wrote:Windows driver WHQL 378.49 released. Fixed: [Folding @ Home] Work unit errors occur
Linux driver 378.09 - BETA fix still missing! Should nvidia be pushed somehow to also include the fix in the linux driver 378.xx?

Does this mean I can run install the latest driver on my Windows machine for use with F@H? Not sure what this is saying.

foldy wrote:Windows driver WHQL 378.49 released. Fixed: [Folding @ Home] Work unit errors occur
Linux driver 378.09 - BETA fix still missing! Should nvidia be pushed somehow to also include the fix in the linux driver 378.xx?

I've been filing bug reports since the 375.xx drivers released. The only acknowledgement of the issue I've seen, was a response to my post on the GEForce forums that said "We'll look into this".

Hannibal13 wrote:

foldy wrote:Windows driver WHQL 378.49 released. Fixed: [Folding @ Home] Work unit errors occur
Linux driver 378.09 - BETA fix still missing! Should nvidia be pushed somehow to also include the fix in the linux driver 378.xx?

Does this mean I can run install the latest driver on my Windows machine for use with F@H? Not sure what this is saying.

No, the conversation has gotten a little criss-crossed here because we had someone ask a couple posts back about Linux driver suggestions. The latest Linux beta does not contain the workaround that Nvidia put into the Windows hotfixe drivers, which some of us are griping is pretty unfair. Most Linux folders should be fine, except if they own a 1050Ti. If they do, then the only option currently would be to find a way to "hack" an older Linux driver like how we earlier discovered how to "hack" the older Windows drivers.

Hannibal13 wrote:Does this mean I can run install the latest driver on my Windows machine for use with F@H? Not sure what this is saying.

Not yet.

Wait until you see an announcement of the release of the new revision of FAHCore_21 which contains a work-around for the problems with every driver version since 375.57 except 376.48. My previous post was just do show that FAH-Development is working on a better solution that the one contained in 376.48 and that work-around is being beta tested.

Quick update: i installed the 370.28. Works with the 1050 Ti under Debian 8 Jessie. (Works = it is folding wihtout errors)

i think this works because the Pascal Architecture is supported since Version 367.35?

So you're getting 0x21 WUs and folding cleanly? If so, awesome! This is exactly why we've needed more input from Linux users in this thread.