My computer is giving me fits. It is resetting itself, sometimes maybe once/day, and others 4x-5x/day. My computer is as follows:

ASUS Sabertooth X58 MB
Intel Core i7-990X CPU
Samsung 2TB SSD (primary)
Hytachi 1TB HD (backup)
Kingston 16GB (4GBx4) DDR3 1066 DRAM
EVGA 2060 GPU
EVGA 650W G1 Gold 80+ PSU

The motherboard I believe is not compatible with Windows 10 since it is not listed in the ASUS website as such, though I have been using it for Windows 10 for years and just recently my computer has been acting up. I only have 2 SATA ports that have 6GB/sec which is what my SSD is plugged into, and my DRAM slots (of which I have 6) has only 4 working, the other 2 slots do not work anymore. When the computer resets it does not freeze, it just reboots unexpectedly.

I am reluctant to buy another "used" motherboard (though it is my cheapest option) at $200+ for the same one. If I buy a new motherboard it will cost me way more for the motherboard, CPU and DRAM (the CPU + DRAM are not compatible with my old one).

Any suggestions on how I can determine whether my motherboard is bad or not?

---

George

unexpected reboots are usually PSU problems, something hitting OCP (over current protection). even a reputable brand PSU can have issues. you could try to replace it with a different one and see if the problem goes away.

but can also be the motherboard, if the VRMs are overheating or overworked. are you crunching on the CPU also? is it overclocked? those old chips are power hungry.

if overclocked, remove the OC and go back to stock settings. if that doesnt work, try stopping CPU crunching completely (or reduce it), GPU only.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

I'll try stopping seti first and see what happens. Yes, I can swap PSU if needed.

Thanks!

---

George

Just an FYI update...

I have reset my CPU to the original speed level (3.48GHz), down from 4.0GHz figuring that it may have been causing my computer to reset itself. It has been over one day since I changed to speed and still no resetting of my computer. Per CPU-Z's Hardware Monitor my "package" CPU power draw has dropped from a 146W which it was at before the change down to the original 130W for the CPU. I think the CPU power draw (which I had set up a few months ago) was a bit too much for this old motherboard. I had stopped BOINC all together thinking perchance that was it with nothing else running in the foreground or background, but it still reset itself several times. I'll wait and see if it still holds now.

---

George

Just an FYI update....

+1

---

A proud member of the OFA (Old Farts Association).

Just an FYI update.... #2

My computer has been up and running constantly for 4 days and 15 hours as of this writing and it is still functioning well.

What I have done is step up my CPU again 100MHz to 3.58GHz from 3.48GHz with a core voltage of 1.216v, this gives me a max power package of 143.19W. I also upped my DDR3 memory speed from 1066MHz to over 1300MHz @ 1.5v. My CPU usage is hovering around 90-95% with a max of 100%, and I'm still only using 10 of 12 threads. My GPU is using .744v (max at 1.050v) and 86-93% usage with a max of 100% while using only 16% memory. And yes, I am using both CPU and GPU for BOINC calculations.

I'd like to know your opinions my temps.

Intel I7-990X CPU: max temps averaging around 43-46C, peak on one core only at 49C, with average temps around 34C. TDP is ~68C. I'm still an air cooler guy using a Noctua NH-D15 with fans running at a mild, less than 1200 RPM.
EVGA RTX 2060 GPU: max temps hitting 59C with actual (current) temp around 39-41C. It's only fan running at 1,000RPM.

All above while running Windows 10, Version 1903 (OS Build 18362.10024).

Thoughts?

---

George

George, re-run the Lunatics installer, scroll down and select the SoG option for your GPU as that ancient Cuda 3.2 app that your using ATM just doesn't cut the mustard on modern hardware. Also free up 1 more CPU core before you do that. ;-)

Cheers.

George, re-run the Lunatics installer, scroll down and select the SoG option for your GPU as that ancient Cuda 3.2 app that your using ATM just doesn't cut the mustard on modern hardware. Also free up 1 more CPU core before you do that. ;-)

Thanks Wiggo, but I'm having a bit of trouble locating the SoG option. Where might I find it?

---

George

George, re-run the Lunatics installer, scroll down and select the SoG option for your GPU as that ancient Cuda 3.2 app that your using ATM just doesn't cut the mustard on modern hardware. Also free up 1 more CPU core before you do that. ;-)

Thanks Wiggo, but I'm having a bit of trouble locating the SoG option. Where might I find it?

If you can't find it on the Nvidia page after scrolling down the listed apps then you're using an old version of the installer so go here and download the latest Lunatics Win64 0.45 Beta6 installer and run that. ;-)

Cheers.

Considering the title. It occurs to me if the MB is 16 years or younger maybe its acting up because it is still a (human) kid?

Tom

---

A proud member of the OFA (Old Farts Association).

Oh yeah, my MB is only 10 yrs old and it has at least 6 more yrs to give me fits.

---

George

Hi Wiggo,

I've done as you suggested and I think it is doing fine. Milkway is running ~2:45 minutes/seconds for a CPU & GPU calculation. I'll see if it holds up for the next day or two. This is my log:

11/3/2019 10:19:22 AM |  | cc_config.xml not found - using defaults
11/3/2019 10:19:22 AM |  | Starting BOINC client version 7.14.2 for windows_x86_64
11/3/2019 10:19:22 AM |  | log flags: file_xfer, sched_ops, task
11/3/2019 10:19:22 AM |  | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
11/3/2019 10:19:22 AM |  | Data directory: C:\ProgramData\BOINC
11/3/2019 10:19:22 AM |  | Running under account George
11/3/2019 10:19:22 AM |  | CUDA: NVIDIA GPU 0: GeForce RTX 2060 (driver version 441.08, CUDA version 10.2, compute capability 7.5, 4096MB, 3554MB available, 6451 GFLOPS peak)
11/3/2019 10:19:22 AM |  | OpenCL: NVIDIA GPU 0: GeForce RTX 2060 (driver version 441.08, device version OpenCL 1.2 CUDA, 6144MB, 3554MB available, 6451 GFLOPS peak)
11/3/2019 10:19:23 AM |  | Host name: GWG-PC
11/3/2019 10:19:23 AM |  | Processor: 12 GenuineIntel Intel(R) Core(TM) i7 CPU X 990 @ 3.47GHz [Family 6 Model 44 Stepping 2]
11/3/2019 10:19:23 AM |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt aes syscall nx lm tm2 pbe
11/3/2019 10:19:23 AM |  | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.18362.00)
11/3/2019 10:19:23 AM |  | Memory: 15.99 GB physical, 23.80 GB virtual
11/3/2019 10:19:23 AM |  | Disk: 1.82 TB total, 1.57 TB free
11/3/2019 10:19:23 AM |  | Local time is UTC -6 hours
11/3/2019 10:19:23 AM |  | No WSL found.
11/3/2019 10:19:23 AM | SETI@home | Found app_config.xml
11/3/2019 10:19:23 AM | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12623039; resource share 125
11/3/2019 10:19:23 AM | Milkyway@Home | URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 789643; resource share 125
11/3/2019 10:19:23 AM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 8625038; resource share 1000
11/3/2019 10:19:28 AM | SETI@home | General prefs: from SETI@home (last modified 03-Nov-2019 10:05:23)
11/3/2019 10:19:28 AM | SETI@home | Host location: none
11/3/2019 10:19:28 AM | SETI@home | General prefs: using your defaults
11/3/2019 10:19:28 AM |  | Preferences:
11/3/2019 10:19:28 AM |  | max memory usage when active: 8187.55 MB
11/3/2019 10:19:28 AM |  | max memory usage when idle: 16375.11 MB
11/3/2019 10:19:28 AM |  | max disk usage: 186.30 GB
11/3/2019 10:19:28 AM |  | max CPUs used: 10
11/3/2019 10:19:28 AM |  | suspend work if non-BOINC CPU load exceeds 35%
11/3/2019 10:19:28 AM |  | (to change preferences, visit a project web site or select Preferences in the Manager)
11/3/2019 10:19:28 AM |  | Setting up project and slot directories
11/3/2019 10:19:28 AM |  | Checking active tasks
11/3/2019 10:19:28 AM |  | Setting up GUI RPC socket
11/3/2019 10:19:28 AM |  | Checking presence of 222 project files
11/3/2019 10:22:07 AM | Milkyway@Home | Computation for task de_modfit_83_bundle4_4s_south4s_bgset_2_1570151403_14587392_0 finished
11/3/2019 10:22:07 AM | Milkyway@Home | Starting task de_modfit_14_bundle4_testing_3s4f_1_1570151403_14578808_1
11/3/2019 10:24:45 AM | Milkyway@Home | Computation for task de_modfit_14_bundle4_testing_3s4f_1_1570151403_14578808_1 finished
11/3/2019 10:24:46 AM | Milkyway@Home | Starting task de_modfit_14_bundle4_testing_3s4f_1_1570151403_14588195_0
11/3/2019 10:27:24 AM | Milkyway@Home | Computation for task de_modfit_14_bundle4_testing_3s4f_1_1570151403_14588195_0 finished
11/3/2019 10:27:24 AM | Milkyway@Home | Starting task de_modfit_80_bundle4_4s_south4s_bgset_2_1570151403_14588787_0

For the moment I have suspended SETI to let Milkway and Einstein play catch up. I will reinstate SETI in a day, and then I'll suspend Milkway and Einstein both and see how much SETI improves.

Should I also make the same <app_config.xml> and <mb_cmdline_win_x86_SSE3_OpenCL_NV_SoG.txt> for Milkyway and Einstein? Grant had suggested the <mb_cmdline_win_x86_SSE3_OpenCL_NV_SoG.txt> file for SETI.

Thoughts?

---

George

Should I also make the same <app_config.xml> and <mb_cmdline_win_x86_SSE3_OpenCL_NV_SoG.txt> for Milkyway and Einstein? Grant had suggested the <mb_cmdline_win_x86_SSE3_OpenCL_NV_SoG.txt> file for SETI.

No- other than limiting the number of CPU threads being used using the Computing preferences settings, all of the changes we have suggest are for Seti processing only & will have no effect on your other projects.
You need to talk to the people on those projects to see what performance improvements can be gained, and how to go about them for those particular projects.

Edit-

11/3/2019 10:19:28 AM | | suspend work if non-BOINC CPU load exceeds 35%

Seems rather extreme to me. Something along the lines of 75% (or higher) would allow you to continue to process BOINC projects while the system does other things, and not impact on the other things the system is doing at the time.

---

Grant
Darwin NT

The fact that machine hard resets would point to a power problem, either the PSU, or the voltage regulators on the motherboard. Age of the hardware also makes this likely because electrolytic capacitors do loose some of their "capacity" as they age. Heat and high current just speed up the process.

Now that you have backed off the overclock and reduced the number of threads, tends to confirm this. The CPU is now drawing less power, so it's below the threshold where the power dips out.

Only real way to isolate the fault further is to swap parts out for known good ones and see what happens. There isn't really any simple diagnostic that can test this because one second the machine is humming along nice, then something dips out for a millisecond, and the system has crashed. Other faults in the system (RAM / GPU / Hard Disk etc) tend to throw some sort of errors up on the screen. The CPU is still running, so it can try and report an error, although it may be a bit random depending on what it was doing when it noticed the RAM or disk error.

The system temps look like nothing that's concerning, and now the machine is running OK at a mild overclock, that might just the new "normal" for your system. If you have reduced the number of threads ( to allocate them to the GPU tasks) the it's probably OK to leave the CPU at 100%. This still leaves threads pretty much free for foreground tasks. If it's a gaming machine, I'd set it to suspend completely in use, and only run when it's not being used.

Thanks Grant, you have been very helpful. I feel that my learning curve is taking a much steeper path, but none the less I'm trying to make the climb.

---

George

The fact that machine hard resets would point to a power problem, either the PSU, or the voltage regulators on the motherboard. Age of the hardware also makes this likely because electrolytic capacitors do loose some of their "capacity" as they age. Heat and high current just speed up the process.

Now that you have backed off the overclock and reduced the number of threads, tends to confirm this. The CPU is now drawing less power, so it's below the threshold where the power dips out.

Only real way to isolate the fault further is to swap parts out for known good ones and see what happens. There isn't really any simple diagnostic that can test this because one second the machine is humming along nice, then something dips out for a millisecond, and the system has crashed. Other faults in the system (RAM / GPU / Hard Disk etc) tend to throw some sort of errors up on the screen. The CPU is still running, so it can try and report an error, although it may be a bit random depending on what it was doing when it noticed the RAM or disk error.

The system temps look like nothing that's concerning, and now the machine is running OK at a mild overclock, that might just the new "normal" for your system. If you have reduced the number of threads ( to allocate them to the GPU tasks) the it's probably OK to leave the CPU at 100%. This still leaves threads pretty much free for foreground tasks. If it's a gaming machine, I'd set it to suspend completely in use, and only run when it's not being used.

Thanks Ianab, you have given me very useful information. As I've said before many times, I have much to learn... but I'm trying.

---

George

Hi Wiggo,

I've done as you suggested and I think it is doing fine. Milkway is running ~2:45 minutes/seconds for a CPU & GPU calculation. I'll see if it holds up for the next day or two.

I'm still waiting for you reinstate SETI (or at least do a days worth of SETI work) to see how your rig performs now and if it's setup correctly before we start on the next phase. ;-)

Cheers.

I'm still waiting for you reinstate SETI (or at least do a days worth of SETI work) to see how your rig performs now and if it's setup correctly before we start on the next phase. ;-)

OK, I have been running SETI (only) for at least a day with apparently no problems, and then I've noticed something odd. When my computer is running the combination CPU/GPU task, setiathome_v8 8.20 (opencl_nvidia_SoG), it shows an elapsed time of up to ~5 minutes with the same time remaining time and this time remaining does not change until the two are the same. When the elapsed time matches the time remaining, the elapsed time jumps upwards of ~100+ days, and yet my progress is only 0.004%. So I'm a bit confused. I have stopped BOINC altogether for two minutes and then relaunched it, waited for the 5+ minutes with no change in CPU/GPU tasks. I then rebooted, waited again, and still no change. I then aborted the CPU/GPU task thinking something was amiss with it. Stopped BOINC, relaunched BOINC, it received another task, and the same thing happens again.

My event log shows this:

11/13/2019 9:47:19 AM |  | cc_config.xml not found - using defaults
11/13/2019 9:47:19 AM |  | Starting BOINC client version 7.14.2 for windows_x86_64
11/13/2019 9:47:19 AM |  | log flags: file_xfer, sched_ops, task
11/13/2019 9:47:19 AM |  | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
11/13/2019 9:47:19 AM |  | Data directory: C:\ProgramData\BOINC
11/13/2019 9:47:19 AM |  | Running under account George
11/13/2019 9:47:19 AM |  | CUDA: NVIDIA GPU 0: GeForce RTX 2060 (driver version 441.08, CUDA version 10.2, compute capability 7.5, 4096MB, 3554MB available, 6451 GFLOPS peak)
11/13/2019 9:47:19 AM |  | OpenCL: NVIDIA GPU 0: GeForce RTX 2060 (driver version 441.08, device version OpenCL 1.2 CUDA, 6144MB, 3554MB available, 6451 GFLOPS peak)
11/13/2019 9:47:19 AM | SETI@home | Found app_info.xml; using anonymous platform
11/13/2019 9:47:20 AM |  | Host name: GWG-PC
11/13/2019 9:47:20 AM |  | Processor: 12 GenuineIntel Intel(R) Core(TM) i7 CPU X 990 @ 3.47GHz [Family 6 Model 44 Stepping 2]
11/13/2019 9:47:20 AM |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt aes syscall nx lm tm2 pbe
11/13/2019 9:47:20 AM |  | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.18362.00)
11/13/2019 9:47:20 AM |  | Memory: 15.99 GB physical, 23.80 GB virtual
11/13/2019 9:47:20 AM |  | Disk: 1.82 TB total, 1.57 TB free
11/13/2019 9:47:20 AM |  | Local time is UTC -6 hours
11/13/2019 9:47:20 AM |  | No WSL found.
11/13/2019 9:47:20 AM | SETI@home | Found app_config.xml
11/13/2019 9:47:20 AM | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12623039; resource share 125
11/13/2019 9:47:20 AM | Milkyway@Home | URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 789643; resource share 125
11/13/2019 9:47:20 AM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 8625038; resource share 1000
11/13/2019 9:47:25 AM | SETI@home | General prefs: from SETI@home (last modified 04-Nov-2019 18:06:31)
11/13/2019 9:47:25 AM | SETI@home | Host location: none
11/13/2019 9:47:25 AM | SETI@home | General prefs: using your defaults
11/13/2019 9:47:25 AM |  | Preferences:
11/13/2019 9:47:25 AM |  | max memory usage when active: 8187.55 MB
11/13/2019 9:47:25 AM |  | max memory usage when idle: 16375.11 MB
11/13/2019 9:47:25 AM |  | max disk usage: 186.30 GB
11/13/2019 9:47:25 AM |  | max CPUs used: 9
11/13/2019 9:47:25 AM |  | suspend work if non-BOINC CPU load exceeds 70%
11/13/2019 9:47:25 AM |  | (to change preferences, visit a project web site or select Preferences in the Manager)
11/13/2019 9:47:25 AM |  | Setting up project and slot directories
11/13/2019 9:47:25 AM |  | Checking active tasks
11/13/2019 9:47:25 AM |  | Setting up GUI RPC socket
11/13/2019 9:47:25 AM |  | Checking presence of 480 project files
11/13/2019 9:47:25 AM | SETI@home | Sending scheduler request: To report completed tasks.
11/13/2019 9:47:25 AM | SETI@home | Reporting 7 completed tasks
11/13/2019 9:47:25 AM | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
11/13/2019 9:47:27 AM | SETI@home | Scheduler request completed: got 9 new tasks
11/13/2019 9:47:29 AM | SETI@home | Started download of blc14_2bit_guppi_58691_72750_HIP40671_0066.24082.0.22.45.90.vlar
11/13/2019 9:47:29 AM | SETI@home | Started download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23326.818.22.45.13.vlar
11/13/2019 9:47:33 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_72750_HIP40671_0066.24082.0.22.45.90.vlar
11/13/2019 9:47:33 AM | SETI@home | Started download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23348.818.21.44.40.vlar
11/13/2019 9:47:35 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23326.818.22.45.13.vlar
11/13/2019 9:47:35 AM | SETI@home | Started download of blc14_2bit_guppi_58691_68031_HIP40774_0051.23893.409.22.45.3.vlar
11/13/2019 9:47:36 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23348.818.21.44.40.vlar
11/13/2019 9:47:36 AM | SETI@home | Started download of blc13_2bit_guppi_58691_75016_HIP39767_0073.24018.0.22.45.161
11/13/2019 9:47:39 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_68031_HIP40774_0051.23893.409.22.45.3.vlar
11/13/2019 9:47:39 AM | SETI@home | Finished download of blc13_2bit_guppi_58691_75016_HIP39767_0073.24018.0.22.45.161
11/13/2019 9:47:39 AM | SETI@home | Started download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23348.818.21.44.64.vlar
11/13/2019 9:47:39 AM | SETI@home | Started download of 17my10aa.5640.1724.3.30.40
11/13/2019 9:47:44 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23348.818.21.44.64.vlar
11/13/2019 9:47:44 AM | SETI@home | Finished download of 17my10aa.5640.1724.3.30.40
11/13/2019 9:47:44 AM | SETI@home | Started download of blc14_2bit_guppi_58691_67714_HIP40938_0050.24052.818.22.45.117.vlar
11/13/2019 9:47:44 AM | SETI@home | Started download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23326.818.22.45.19.vlar
11/13/2019 9:47:47 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_67714_HIP40938_0050.24052.818.22.45.117.vlar
11/13/2019 9:47:48 AM | SETI@home | Finished download of blc14_2bit_guppi_58691_67714_HIP40938_0050.23326.818.22.45.19.vlar

Any help would be greatly appreciated.

---

George

That log isn't very helpful for the issue you are describing.

However, looking at your errored tasks from your system here: https://setiathome.berkeley.edu/results.php?hostid=8625038&offset=0&show_names=0&state=6&appid=

It's clear you have the "Windows 10, SoG, nvidia driver, Arecibo VHAR" issue.

Basically, when using the SoG app (which you are), on Windows 10 (which you are), with drivers newer than 431 (which you are), when processing Arecibo VHAR tasks (meaning tasks with a WU angle range greater than 1.0) it exhibits problems. on Turing cards this is usually the task not processing at all until it times out.

you can remedy this by removing your 441 driver and installing the 431 driver. don't install the newest drivers until you get confirmation that the problem has been resolved.

you could also get around this problem by running a different app (the sah app seems to work), or by running Windows 7 (or any other OS).

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Nothing wrong with your host other than you are running Windows10 with drivers incompatible with VHAR tasks. The runtime observation is typical in your configuration. The issue has/is being discussed ad nauseum in its own thread.
https://setiathome.berkeley.edu/forum_thread.php?id=84694

Either change to Windows7, backlevel to driver version 431 or change to the 8.16 SAH application.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)