Posts by hoppisaur

1) Message boards : Number crunching : Future of Milkyway@Home (Message 75084) Posted 26 Feb 2023 by hoppisaur Post: It is. ahem will be. I assume Linux + CUDA which means only Nvidia cards can use the "mod". Currently, I cannot run any of my (older) hi performance AMD cards under Ubuntu 20.04.5. Every month or two I take another stab at getting them to work. Under 18.04 my S9xx0 and HD-79xx cards worked fine but not since a disk crash and upgrade to 20.04.5 Advice from AMD forum was not helpful: use the exact release listed and no other. The release, dated 2021 Q2, is for Ubuntu 20.04.2 https://www.amd.com/en/support/professional-graphics/firepro/firepro-s-series/firepro-s9000 The instructions come with a warning that if one upgrades to kernel 4.15 then the driver from the year 2018 needs to be used instead of the 2021 drivers. This does not make a lot of sense considering that the 20.04.5 kernel is way past 4.15 OS: Linux Ubuntu: Ubuntu 20.04.5 LTS [5.4.0-139-generic\|libc 2.31] If I follow AMD's exact installation instructions for the Firepro s9000 card, only my RX-570 card works. I discovered this by plugging in the RX-570, powering the system back on, and doing nothing else. The 570 is significantly slower than the s9000 or s9050 cards. OpenCL: AMD/ATI GPU 0: Radeon RX 570 Series (driver version 3224.4, device version OpenCL 1.2 AMD-APP (3224.4), 4082MB, 4082MB available, 5095 GFLOPS peak) The only, quasi-stable, config I have been able to do for linux for firepros (and Bristol Ridge R7): Ubuntu 20.04 (either .1 or .2 (I think). Install and pull the internet during installation as even if you say don't upgrade: it will helpfully update you to kernel 5.15.XX which I haven't been able to get amdgpu to work on. period. Uninstall snap. Disable unattended upgrades. Turn off crash reporting. Never allow upgrading to go above kernel 5.4.0.XXX otherwise amdgpu 20.50 will break, at least in my experience. I use amdgpu v20.50. Nothing else works. I only install --opencl=legacy. I don't bother with vega or later if the firepro is on the system. Since I have dedicated crunchers, I will ssh into them as amdgpu tends to make video broken, but the firepro is seen and works. boinctui makes life bearable via cmdline. Hope this helps.
2) Message boards : Number crunching : Validation Pending too many tasks (Message 74522) Posted 21 Oct 2022 by hoppisaur Post: My tasks status as of today (10.21). Server migration can't come soon enough... In progress (1158) Â· Validation pending (25563) Â· Validation inconclusive (1850) Â· Valid (11077) Â· Invalid (95) Â· Error (1) Edit: this is 99.99+% separation.
3) Message boards : Number crunching : Problem with ntoskrnl errors with AMD A8-7800 APU (Message 70389) Posted 18 Jan 2021 by hoppisaur Post: You can also create a separate "location" in Milkyway@home preferences under your account. For that separate location (work, school, doesn't matter) you would assign the problem machine to that location. You can specify only CPU tasks and not have it affect the rest of your computers.
4) Questions and Answers : Unix/Linux : GPU tasks error out on Linux (Message 70306) Posted 4 Jan 2021 by hoppisaur Post: I did not have many errors on my RX 570 until I upgraded to Ubuntu 20.04.1, which requires the latest (20.45) drivers. Now I get many zero-second fails, while the others complete normally. https://milkyway.cs.rpi.edu/milkyway/results.php?hostid=872175 The same thing happens on Einstein, so it is not just MW. It seems to be a driver problem, but maybe MW can work around it? Don't know if this helps: I have been able to use amdgpu 20.20 or 20.30 with Ubuntu 20.04.1 Had a problem with getting a stable desktop and found that the trick is that 5.4.0-56 or -58 breaks the amdgpu drivers. See here: https://community.amd.com/t5/drivers-software/can-t-install-amdgpu-drivers-on-ubuntu-20-04-1-5-4-0-56-generic/td-p/426676. Once I uninstalled kernels greater than 54 and reinstalled the amd drivers, it has been stable. If your computer is your daily driver, you will have to come up with a strategy to prevent Ubuntu from nagging you to death to upgrade.
5) Message boards : Number crunching : Benchmark thread 1-2019 on - GPU & CPU times wanted for new WUs, old & new hardware! (Message 68953) Posted 5 Aug 2019 by hoppisaur Post: A12-9800 #1 on Ubuntu 18.04 using amdgpu-pro 18.50 opencl=legacy 227.xx = 96.3 seconds from last 13 valids A12-9800 #2 (latest purchase already killed a motherboard after 3 days!) on Ubuntu 16.04.6 using amdgpu-pro 17.50 opencl=legacy 227.xx = 96.6 seconds for last 20 valids Both motherboards are set to C6 = disabled, Core performance boost = disabled, Cool'nQuiet = enabled. Both mobos have 2 x 4gb DDR4 2400. Both BOINC clients are set to using 50% of CPUs. Watching clockspeeds in terminal, clock speed will sit at 1396 across all "cores" and jump up to the 2-3 GHz range briefly when it is offloading from the GPU. Difficult to tell peak clockspeed as time resolution of terminal window is not that fine.
6) Message boards : Number crunching : New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new! (Message 68098) Posted 6 Feb 2019 by hoppisaur Post: hoppisaur Ok I'm confused now lol, so previously your AMD A12-9800 APU @4.2 GHz managed a time of 120s, but now it's at 3.8 GHz & managing 101s? You mentioned you tweaked the bios, did you by chance increase the iGPU clock? (if that's possible?). Sorry to hear about the dead m/brd :(. Sorry for late response. Normal freq of chip is 3.8 with capability to boost to 4.2. Main change I made in was to disable C6 as it was freezing when left alone for a while. If I watch clockspeeds via lscpu, the most it boosts is ~4.0 so call it whatever. AMD CnQ is on because that allows it to drop frequency on the cpu cores to give power to the graphics cores. It is on a A320 chipset mobo so iGPU is not possible to OC. The 227s have been in the 100-101 range and the 243s are in the 107-108 range. My computer is visible so verification is available to anyone who wants to see.
7) Message boards : Number crunching : New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new! (Message 68013) Posted 15 Jan 2019 by hoppisaur Post: Of latest 14 validated 227.62 WUs, avg is 100.65 secs. Chip is at 3.8GHz. I've tweaked BIOS on another board as much as I can. First board died complete with magic smoke. A12-9800 with dual channel DDR4-2400. Is the 135s time with the CPU at 3.8 GHz with 227.62 WUs? (I'll add that time too if so). Will link the AT thread again in a moment.
8) Message boards : Number crunching : New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new! (Message 67962) Posted 23 Dec 2018 by hoppisaur Post: It is trending slower today as I turned off motherboard core boost (3.8 -> 4.2) to see if I am hitting a wall with what the motherboard is able to deliver. 10WU avg is now 135.06. I think my powersupply isn't up to the task of boosting the cpu while the iGPU units are being used. Or it could just be driver or mobo related. I am going to try other hardware as it was frozen this morning (locked up). Yes, agreed, I think the proximity of the gpu and cpu units is advantageous in an otherwise mediocre chip. Or maybe they share the cache? Avg (of 5) = 120.33 seconds from the 227.62 WUs AMD A12-9800 stock; ddr4-2400, onboard (iGPU) R7 graphics Boinc 7.9.3, Ubuntu 18.04.1 LTS, OpenCL 1.2, amdgpu-pro 18.50 drivers. Only MW@Home running. No CPU tasks. 1 WU at a time. That seems very fast to me for an APU, since you are getting times comparable to a lot of GPUs (both AMD and Nvidia) that must have much more graphics power, I would think. So maybe the APU offloads the dual-precision workload to the CPU more efficiently? I really don't know anything about APUs, so maybe someone who does can explain it.
9) Message boards : Number crunching : New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new! (Message 67960) Posted 23 Dec 2018 by hoppisaur Post: Avg (of 5) = 120.33 seconds from the 227.62 WUs AMD A12-9800 stock; ddr4-2400, onboard (iGPU) R7 graphics Boinc 7.9.3, Ubuntu 18.04.1 LTS, OpenCL 1.2, amdgpu-pro 18.50 drivers. Only MW@Home running. No CPU tasks. 1 WU at a time. Still not 100% stable. Will investigate if power delivery issue or driver issue. Locks up when asked to do too much. Stable if left to crunch. Average from at least 5 WU times using 227.62 credit WUs only (not cherry picked please! ;)). A dedicated physical CPU core for each GPU (for optimal MW WU times). If only using BOINC for CPU tasks, & you have an HT capable CPU, then the only way to be certain of this (bar disabling HT) is to set the BOINC computing preferences (in advanced mode>options) so that you have 1 less CPU thread running then you do physical cores. Don't panic too much about lost CPU ppd, it doesn't take long to run MW GPU WUs ;) (see previous table). Please state what speed & type CPU you have, as it now has a significant affect on GPU WU times! Please state GPU clock speeds if overclocked(including factory overclocks) or state 'stock'. Please state whether only crunching 1 WU at a time per GPU, or state how many are run concurrently (I will create a separate table for that). For CPU times please state whether Hyper Threading (or equivalent) is enabled or not, times for both states welcomed :). It would also be useful if you could state your BOINC, driver version, & OS, incase it does make any difference.