Interesting. I just purchased an S9150 on ebay for $68 plus shipping, and was wondering what the power draw for that particular unit was. I expect to put it in play in a few weeks here. The hope is to 2x my sapphire toxic r9 280x, which chunks out 1 every 30 seconds, with an average power of 175 watts total. It looks like yours gets 1 every 28 seconds. Have I got that right? If so, can it be pushed harder? Say to 1 every 15 seconds?

Prices have really dropped. I thought $80 (USD) was good, now I see $68 !

This fan can be made to fit with Aluminum tape and using a rear bracket guide or blade. It is an exact fit to the plastic case and with enough metal tape the guide is not needed. If you use a guide there are 3 screws onto the back (front?) but (depending on guide shape) only 2 of the 3 holes can be used due to the pair of power connectors. I had to turn my dell guide upside down and use two screws.

I used the following to step the voltage from 12 to 16 so I could run during the summer in my garage. You may not need it.

All devices take about 200 watts each.

Yeah, external cooling fan required. Disadvantage when compared to the non-server type GPUs. But the price was good. I'll be using this fan. NEW Dell Optiplex 3050 5050 SFF AVC CPU Fan Heatsink BAZB0925R2U TKR4X 7D86K. I plan on using nylon screws to hold it to the case. It's a pretty good fit, except for the power cables, but nothing a box cutter and some foil tape can't fix. Good find on the voltage step up. I just may need that in the summer. Our garage gets up to 100 some days in the summer. I do need to find the mounting pci mounting bracket, or repurpose the ones that came out since the r9 280x had a built in one.

Do you think I can get 2x performance over the existing r9 280x?

As a side note, I looked at the top computers listing and did not see any radeon pro VII(6,528 Gflops), only the radeon VII(3,360 Gflops. Is the pro not usable here?

this is my build desktop crunching milkyway@home cpu + gpu 5950x rx6800xt 128gb ram Monster liquid cooled x5 copper radiator runs super cooled at 4.4ghz all cores at full load always below 50 degrees celcius heres the link https://youtu.be/sugAwW3cRQk

---

Yeah, external cooling fan required. Disadvantage when compared to the non-server type GPUs. But the price was good. I'll be using this fan. NEW Dell Optiplex 3050 5050 SFF AVC CPU Fan Heatsink BAZB0925R2U TKR4X 7D86K. I plan on using nylon screws to hold it to the case. It's a pretty good fit, except for the power cables, but nothing a box cutter and some foil tape can't fix. Good find on the voltage step up. I just may need that in the summer. Our garage gets up to 100 some days in the summer. I do need to find the mounting pci mounting bracket, or repurpose the ones that came out since the r9 280x had a built in one.

Do you think I can get 2x performance over the existing r9 280x?

As a side note, I looked at the top computers listing and did not see any radeon pro VII(6,528 Gflops), only the radeon VII(3,360 Gflops. Is the pro not usable here?

Sorry, should have replied earlier.

I have not taken the 9150 or 9100 apart but I suspect the processor chip is mounted like the s9050, s9000 as they are all "hawaii"
The chip is recessed, and a standard heat sink will need a copper shim

The 30 second you mention for the r9 is extremely good. You might get 24 for the 9150 running 4 concurrent work units.

If you get a blower fan make sure it takes at least 2 amps minimum. There is a gap between the bottom of the case and the cooling fins. I had to block the gap to force more air through the fins.

Sorry, should have replied earlier.

I have not taken the 9150 or 9100 apart but I suspect the processor chip is mounted like the s9050, s9000 as they are all "hawaii"
The chip is recessed, and a standard heat sink will need a copper shim

The 30 second you mention for the r9 is extremely good. You might get 24 for the 9150 running 4 concurrent work units.

If you get a blower fan make sure it takes at least 2 amps minimum. There is a gap between the bottom of the case and the cooling fins. I had to block the gap to force more air through the fins.

I wondered where you went! I promise, I brushed my teeth that day!

Regarding the Dell cooling fan, I removed the heat sink and set is aside. Wont be using it. The blower is only 1 amp, but it is 24 CFM, so hopefully that will be good enough. With a bit of surgery on the fan housing, especially around the 8 and 6 pin power connectors,I was able to mount it to the GPU cover using nylon fasteners. With the fan powered on, right away I did notice the leaks from the gap you mentioned. It will be interesting to see how much that impact really is, but at any rate, it can easily be taped up if needed. When this is all done, I will post pictures and characterization data.

You might want to read this discussion

https://www.techpowerup.com/forums/threads/s9150-flash-problem-bios-size-inconsistency.303055/

One of the $80 boards I bought is not reporting sensor data like the newer firmware does and I was unable to upgrade that board's firmware. It otherwise works fine.
I do not have a fan yet for the other board I bought.

But it looks like in Beemer Bikers case, moving the GPU to another motherboard allowed the sensors to be read correctly. Which means the card is good, at least as far as sensors are concerned. And it looks like the windows driver is older, he says. It would be nice to have a bunch of hardware like he has to aid in troubleshooting issues. 15 motherboards and at least 10 GPU cards listed. That's got to be one he!! of a power bill. Hope he's on solar!!

edit: and btw, at 100% GPU load his temp is only 58C. that's really good. my r9 is 66 at 100%

Anybody running a Radeon Pro VII? Gflops rated at 6,528 and 250W. Ebay prices are $450 to $800.

---

Here are some more WU times.


BOINC Version: 7.4.36

BOINC Version: 7.4.36
.

You have some OLD versions of Boinc running there, I thought all the Projects stopped accepting tasks from any version below 7.10.? but obviously that was wrong. The latest release version of Boinc is 7.20.2.

Here are some more WU times.


BOINC Version: 7.4.36

BOINC Version: 7.4.36
.

You have some OLD versions of Boinc running there, I thought all the Projects stopped accepting tasks from any version below 7.10.? but obviously that was wrong. The latest release version of Boinc is 7.20.2.

Thanks for providing that info and especially the message url
Just above that referenced message is a nice table of GPU elapsed times.

Some observations ---
The 21.29 to 23.7 seconds is based on overclocking the HD7970 to 1200 gpu clock and 1550 memory clock. I suspect that board is pulling the full 250 watts (or more) that it is spec'ed at. It normally runs at 925 / 1375

The 7970 is an excellent board and cost a lot more than the 7950 that I could afford years ago.

The top boards listed in that table message 63317 (dated 2015) are the 28nm fab sizes and based on "Southern Isles". The 79xx , s9000 and s9050 are Tahiti and the s9100, s9150 are Hawaii. The prices have dropped for s9150 and they are $80 USD and lower on eBay. Those "S" boards have no moving parts (fans) and except for the s90x0 they have no video out. They cannot be overclocked that I know of. The "s" boards do not have UEFI bios so the motherboard, if newer, needs on-board video or a temp video card to set the system up for remote access.

Some statistics on S9150: Temp average 71c, Power draw (estimate using a wattmeter): 175, Anything less or greater than 4 concurrent work units seems to run slower. Throughput seems to be 24 seconds, best I can do using 1x risers. [edit] 24 was average for 4 different video boards. I am seeing under 20 seconds but if I raise the concurrent WU count up then the s9000 boards to not have enough memory. Going to re-structure the board assignment to get the better boards on one system.

GPUz shows 133-222 with average of 162 watts drawn through AMD VII but the s9xxx series do not report power.

Basically, the s9150 is a really cheap HD7970 that runs cooler, and one can easily DIY a blower fan onto. MY 0.02c opinion

Here are some more WU times.


BOINC Version: 7.4.36

BOINC Version: 7.4.36
.

You have some OLD versions of Boinc running there, I thought all the Projects stopped accepting tasks from any version below 7.10.? but obviously that was wrong. The latest release version of Boinc is 7.20.2.

Thanks for providing that info and especially the message url

Thank you for doing that, it was a totally useless message otherwise wasn't it!!

So, bearing in mind that I am someone who normally is fine with detailed instruction like this for overclocking and benchmarking and such...

These are restrictions that in allot of cases dont make much sense on modern HW any more...

That said, I will add some results:

//Windows 10, R9-5950x (PBO On, Core Optimizer off, AiO), Radeon RX-6900xt (stock)

Machine is run with full load on GPU and stock 0.05 CPU thread per GPU, with full CPU loading including on SMT.

Milkyway@home Separation v1.46 (opencl_ati_101) windows_x86_64 times (GPU) : ~25-35s per work unit with 1 unit on GPU, but GPU had lots of idle time and bursts of active time. ~90 seconds per work unit with 4WU on the GPU at once.
Milkyway@home Separation v1.46 windows_x86_64 (CPU) - 2,600 Seconds both real and CPU time for compute, seems to run single threaded with up to 32 simultaneous possible at once.
Milkyway@home N-Body Simulation v1.82 (mt) windows_x86_64 - all over the place as these tend to get paused and such. But from the runs where it gets to go solid for awhile I am seeing ~3-5k seconds real world time on 16 compute threads for 50-70k seconds CPU time.

// Have two more rigs, will post later, lazy now haha, just off work an all.

Also, (general Q, not at OP) where are the GPU WU's at? They seem to have dried up!? lol

I've been running a RX 580 and a HD 7970 GHz edition. I recently bought a couple of second-hand GPUs to muck around with - a Quadro K6000 and a RX 5700XT.

The 7970 is a star, it was doing a task every 45-48 secs, slightly overclocked at 1125MHz. The 580 does a unit in about 100 sec at stock. The 5700 XT was a disappointment, slower than both, and drawing 90%+ load, so I'm not going to try running multiple tasks.

The Quadro - basically the same as a Titan Black, slightly down-clocked - takes 250 sec per task using about 15% GPU load. I've now got it happily running 6 tasks at a time, using 85-90% GPU in the same amount of time, so in the same league as the 7970. Alas, I thought it'd be better. I could probably get 7 units without a significant slowdown, but I'll stick with it as is. Unfortunately I can't increase the core speed, it seems locked at 900 MHz, though I can speed up the vram.

I've changed the 7970 to processing 2 units, and it is now doing 2 in 70-ish seconds, instead of 1 in 48.

I've been running a RX 580 and a HD 7970 GHz edition. I recently bought a couple of second-hand GPUs to muck around with - a Quadro K6000 and a RX 5700XT.

The 7970 is a star, it was doing a task every 45-48 secs, slightly overclocked at 1125MHz. The 580 does a unit in about 100 sec at stock. The 5700 XT was a disappointment, slower than both, and drawing 90%+ load, so I'm not going to try running multiple tasks.

The Quadro - basically the same as a Titan Black, slightly down-clocked - takes 250 sec per task using about 15% GPU load. I've now got it happily running 6 tasks at a time, using 85-90% GPU in the same amount of time, so in the same league as the 7970. Alas, I thought it'd be better. I could probably get 7 units without a significant slowdown, but I'll stick with it as is. Unfortunately I can't increase the core speed, it seems locked at 900 MHz, though I can speed up the vram.

I've changed the 7970 to processing 2 units, and it is now doing 2 in 70-ish seconds, instead of 1 in 48.

For max points you should be running 3 or 4 concurrent tasks on that 7970 if you haven't tried already. On the K6000 you will most likely get errors if you go for more than the 6 concurrent you're doing now.

For max points you should be running 3 or 4 concurrent tasks on that 7970 if you haven't tried already. On the K6000 you will most likely get errors if you go for more than the 6 concurrent you're doing now.

Thanks for the advice. I'll leave the Quadro as is. I've changed the 7970 to 3 workunits - it is running stable, but hot enough that I've placed a fan blowing directly on the back of the card. Probably time to pull it out, check dust, paste, etc, etc. <sigh>

A question: I've just bought a 6GB version of the 7970, currently waiting for it to arrive. The extra RAM isn't going to make any difference in performance on Milky Way, but does anyone think there's any other projects it might make a difference for, versus the normal 3GB card?

For max points you should be running 3 or 4 concurrent tasks on that 7970 if you haven't tried already. On the K6000 you will most likely get errors if you go for more than the 6 concurrent you're doing now.

Thanks for the advice. I'll leave the Quadro as is. I've changed the 7970 to 3 workunits - it is running stable, but hot enough that I've placed a fan blowing directly on the back of the card. Probably time to pull it out, check dust, paste, etc, etc. <sigh>

A question: I've just bought a 6GB version of the 7970, currently waiting for it to arrive. The extra RAM isn't going to make any difference in performance on Milky Way, but does anyone think there's any other projects it might make a difference for, versus the normal 3GB card?

You will be able to run all of the current gpu tasks at Einstein as several require a minimum of 4gb onboard ram now, but I'm wondering if the extra ram will mean you can run even more tasks at the same time here?

For max points you should be running 3 or 4 concurrent tasks on that 7970 if you haven't tried already. On the K6000 you will most likely get errors if you go for more than the 6 concurrent you're doing now.

Thanks for the advice. I'll leave the Quadro as is. I've changed the 7970 to 3 workunits - it is running stable, but hot enough that I've placed a fan blowing directly on the back of the card. Probably time to pull it out, check dust, paste, etc, etc. <sigh>

A question: I've just bought a 6GB version of the 7970, currently waiting for it to arrive. The extra RAM isn't going to make any difference in performance on Milky Way, but does anyone think there's any other projects it might make a difference for, versus the normal 3GB card?

You will be able to run all of the current gpu tasks at Einstein as several require a minimum of 4gb onboard ram now, but I'm wondering if the extra ram will mean you can run even more tasks at the same time here?

I'll see what Einstein does once the card arrives. Previously I've only run single E@H tasks. Ah well, it is good to play around, it can keep one entertained for hours and hours.
vRam doesn't seem to be a significant issue with MW: The current 3GB 7970 is using less than 500 MB according to both task manager & GPU-Z while running 3 tasks & 95% GPU usage. The K6000 is using about 1.5 GB running 6 tasks & a web browser open with multiple tabs, so practically nothing of its 12GB capacity. (One machine is a FX 6300 w/ 24GB RAM, the other a Ryzen 1700 w/ 32GB) Hence my thought it wouldn't affect things.

For max points you should be running 3 or 4 concurrent tasks on that 7970 if you haven't tried already. On the K6000 you will most likely get errors if you go for more than the 6 concurrent you're doing now.

Thanks for the advice. I'll leave the Quadro as is. I've changed the 7970 to 3 workunits - it is running stable, but hot enough that I've placed a fan blowing directly on the back of the card. Probably time to pull it out, check dust, paste, etc, etc. <sigh>

A question: I've just bought a 6GB version of the 7970, currently waiting for it to arrive. The extra RAM isn't going to make any difference in performance on Milky Way, but does anyone think there's any other projects it might make a difference for, versus the normal 3GB card?

You will be able to run all of the current gpu tasks at Einstein as several require a minimum of 4gb onboard ram now, but I'm wondering if the extra ram will mean you can run even more tasks at the same time here?

I'll see what Einstein does once the card arrives. Previously I've only run single E@H tasks. Ah well, it is good to play around, it can keep one entertained for hours and hours.
vRam doesn't seem to be a significant issue with MW: The current 3GB 7970 is using less than 500 MB according to both task manager & GPU-Z while running 3 tasks & 95% GPU usage. The K6000 is using about 1.5 GB running 6 tasks & a web browser open with multiple tabs, so practically nothing of its 12GB capacity. (One machine is a FX 6300 w/ 24GB RAM, the other a Ryzen 1700 w/ 32GB) Hence my thought it wouldn't affect things.

Sounds good!!

New poster here - getting back into DC with my rigs. Started on Seti back in 2002, and joined MW in 2009, but haven't been terribly active while growing a business and family.

Main rig:
- 5950x, PBO:4.2-4.7o/c, 360AIO (runs all MH threads at 60°C)
- 128Gb memory, 3666/16 w/timings tightened
- 3x Samsung 980Pro NVMe drives (4Tb)
- 4090 w/240 AIO + 1050ti (4090 o/c @2925Mhz core and memory at 11250Mhz @ 36°C)
- 3x 60" 4k monitors + 60" short throw projection on to my desk + 30" widescreen above it all
- Win 10 Pro

I'm seeing that the mighty 4090 is not worthy in the realm of AMD GPUs for MH@H!

Milkyway@home Separation v1.46 (opencl_nvidia_101):
- 1WU: 28-34s, avg 32s
- 2WU: avg 70s or 35 per
- 3WU: 114s or 38 per
- 4WU: 152s or 38 per

Milkyway@home Separation v1.46 windows_x86_64 (1x core):
- Runtime: 3x135s avg
- CPU time: 2,943s avg

Milkyway@home N-Body Simulation v1.82 (mt) (16x cores)
- Runtime: 1,938s avg
- CPU time: 18,335s avg


Second rig running MH:
- 5900x PBO o/c
- 32Gb memory
- 2x 980Pro NVMe drives
- NVS 310 video card
- 22" monitor
- all within a Pelican 1535Air travel case
(normally an onsite file server, so no graphix really)

Milkyway@home N-Body Simulation v1.82 (mt) (16x cores) at a time
- Runtime: 2,164s avg
- CPU time: 20,445s avg

I'm almost ashamed to post this -- seeing/reading about all these beautiful and high performing GPUs !!
Anyway, might be of interest for someone ?

My Oldie ASUS Nvidia Strix Geforce GTX 750Ti 4GB:

Microsoft Windows XP Professional x64 Edition, Service Pack 2, (05.02.3790.00)
Intel(R) Core(TM) i5-3570 CPU @ 3.40GHz [Family 6 Model 58 Stepping 9]
BOINC version 7.16.20
Coprocessors [5] NVIDIA GeForce GTX 750 Ti (4095MB) driver: 368.81 OpenCL: 1.2
Completed and validated 657.20 157.02 227.11 Milkyway@home Separation v1.46 (opencl_nvidia_101) windows_x86_64
app config .... gpu usage 1.00 cpu usage 0.50 ...

According to GPU-Z readings for GPU (all max over several hours):

Temp 54C
Memory 148MB
Load 100%
Board Power Draw 17W
CPU Temp 52C
Sys Mem Used 3460MB

Each GPU around 60$ (of course used).
No tuning what-soever.

I'm very happy with the setup!

S-F-V