tictoc

I'm guessing you sorted this out. We are moving right up the ranks thanks to that K5000.

Brake and clutch pedal pads for the old beast.

If you want to keep it on the inexpensive side, just use barb fittings with clamps, or even zip ties, with how low the pressure is on PC water cooling loops. Many modern day water cooling components are just pretty garbage that's priced high, since it's a "premium" hobby. Personally, I like monsoon free center fittings, since I have found the barb side to be slightly overbore and/or the collar to be underbore. They also have a double barb which helps to ensure that the tube is properly compressed against the fitting. This makes them harder to snug all the way down on most EPDM or ZMT tubing, but in the many fittings I've installed (many hundreds at this stage) I've never had one leak.

Glad you didn't wash away in the floods. No worries on the downtime. I know how it goes with mother nature. You can pretty much count me out for at least a few weeks every winter, since I'll inevitably lose internet and/or power for extended periods of time. Good to see those GPUs up and running.

Totally off topic, but have you tried running your powershell scripts natively in Linux? I messed around with powershell on Linux a year or two ago. It broke my brain a little bit to run PS in my terminal. Forget that we're already parsing ls and look at the slash horror. PS /home/tictoc> dir \var\log | ls -al /var/log | grep "log" > /home\tictoc/logfiles.txt PS /home/tictoc> type \home/tictoc\logfiles.txt -rw-rw-r-- 1 root utmp 292292 Jul 30 13:44 lastlog -rw-r--r-- 1 root root 1071994 Aug 18 17:08 pacman.log -rw------- 1 root root 64064 Aug 19 2020 tallylog -rw-r--r-- 1 root root 63010 Aug 18 17:08 Xorg.0.log -rw-r--r-- 1 root root 158720 Aug 17 16:30 Xorg.0.log.old -rw-r--r-- 1 root root 6704 Feb 8 2021 Xorg.1.log -rw-r--r-- 1 root root 6704 Feb 8 2021 Xorg.1.log.old Despite the fact that this is utter madness, it does work. I never tried running any scripts.

Not sure if anyone else runs HFM in Linux, but it once again works on Linux through mono. I haven't messed around trying to rebuild it so it will run "natively" in Linux, but running the latest release through mono works without a hitch. It can also be ran in WINE, but I don't think the Web Generation will work when running it in WINE.

I don't even mess with FAHControl. They are doing a rewrite to get rid of the python2 dependencies, and I imagine it will be ready by the next client release.

Sorry, I skipped over the log section before replying. The PTX error is indicative of an older CUDA version. For Windows I think you need 456.xx or newer drivers for compatibility with CUDA 11, and the latest fahcore.

First thing I would do is ditch the Quadro drivers, and see if the regular drivers pick up both cards. I'm pretty sure that you can only initialize one NVIDIA driver at a time without creating conflicts. There is not really anything in the Quadro drivers that will accelerate F@H, compared to the regular GeForce drivers.

Thanks. I am must have been pulling all my work from that one server, and I didin't notice points go missing for anyone else so assumed it was something on my end.

I seem to have come out the other side of whatever black hole I was folding into. Earlier today I had a 27.6M (3.3M actual) update on the 980, or about 3+ days worth of folding on that GPU. Not sure what actually happened, since it doesn't look like anyone else had the same thing going on. My logs showed zero upload errors, and other than no stats registering, everything looked fine.

Nice to see you around these parts nova. In Q_Cruncher

It is Compute 3.0 capable, but do you know which version of CUDA is installed? If it's the latest driver (R470 U3 - 471.68), then it should be CUDA 11.4, and that should work with the new core that was released yesterday.

I seem to be folding into a black hole. I have completed and uploaded many WUs over the last few days, but my stats have stayed the same. This is not an issue with the ETF stats, my stats are not updating anywhere. I've never had this issue before. I was running an older version of the client, so I just updated to 7.6.21. Downloading my first WU with this client, so we'll see if this fixes the issue.

What version of CUDA do you have installed, and what WU failed to build and reverted to OpenCL?

You just need to speed run your thermal paste checks. It can be a comp within a comp, to see who can swap out pads and paste the quickest.

My ppd has taken a big hit over the last few days. My monitoring/alert server is down since I'm working on getting my network put back together and transitioning everything over to my new all-in-one home server. I had been periodically looking at HFM, and everything seemed fine at 875k to 1M ppd. Just looked at my EOC stats and the last few days ppd is way down. Checking my HFM config I had it set to report ppd based on the last 3 frames. I must have changed it while I was testing some other cards and never switched it back. Looking through my FAH log, it appears that my rock solid OC on the 980 is not so rock solid. I dialed the clocks back a bit, and hopefully it will stay stable.

With that GPU PCI-e 2.0 x8 shouldn't be any sort of a bottleneck. I think the GTX 680 only saw a 1-2% performance reduction going from PCI-e 3.0 x16 to PCI-e 2.0 x8. The K4000 is basically a cut down GTX 660 so you should be good to go.

I also remember that there was a certain driver that was better on Linux. If you dig around the OCN TC forum, or maybe in the Infinty thread, you should find a post from me about it. I feel like I did some testing on that, and at some point in time, a current driver caught up with the older driver. Not positive on any of the above, since that was a few years ago.

The wee little units are not good for more powerful GPUs. There is some development going on for the next client that is looking to give more fine grained controls to the work server when distributing WUs.

Looking for a sanity check on my overkill home server drive/FS layout. Below is what I've been testing, and so far, I haven't ran into any issues. /boot (esp) - 1GB FAT32 partition mirrored (sort of) on 2 nvme SSDs. Primary partition on one SSD with the secondary rsync'd to the primary on changes. /, /home, etc - Btrfs raid1 on 2x 1TB nvme SSDs /var - ext4 on 480GB Optane 905p. Containers, running VM images, logs, etc. /data - Btrfs raid10 on 6x 8TB HDDs. Room to expand to 8 total drives on current HBA /backups - Btrfs raid1 on 2x 4TB HDDs. Critical backups of other machines/VMs along with snapshots of the host OS. /victim - md Raid-0 on 4x 1TB nvme SSDs. Steamcache and squid. /offline - ??? FS on 2x 12TB HDDs (only mounted in hotswap bay while running backups) for offline/cold storage backups. Leaning towards Btrfs dup, but I'm not super thrilled with the idea of using a single file system for everything. Probably too paranoid about losing data...

Native res won't happen on a 290x, I thought you were talking about your RX 580. I think the native max single display resolution with a little hackery is something like 4096x1440. Anything larger will need eyefinity, and it's been almost a decade since I ran a de-bezeled eyefinity setup. If you are going to run the 290X in linux, you will want to force the amdgpu (kernel) driver. That card will default to the radeon driver, and when I was running my 290s the kernel driver was more performant (and still is). If you go down that rabbit hole, maybe start a thread, or just post to the thread I started in the OS form.

You should be able to at least force the resolution if it's not picked up automatically. If you want to give it a go, I can help you out. I currently run the following triple monitor setup on my main machine: 1440x2560-3440x1440-2560x1440(165Hz).

OP is updated with your proper card. My 980 has been cruising along at about 900k ppd. I'm not quite stable with the CUDA WUs at the clocks that I used to run back in the day for TC. I might be able to eek out a few for MHz if I put it back under water.

I haven't used OPNSense with 10G routing, so I can't say for sure, but a 4790k should be able to get good throughput. You might need to do some tuning, but a 4790k should be a pretty good all-around performer. If you need a little more juice, didn't ASUS unlock OC'ing on the H81 boards? Either way that CPU should be able to handle what you want to do. I might know a guy, that knows a guy, with a delidded 4790k sitting on a shelf.