Skip,

Keith is right. You should be able to run atleast 5 units on your RX 580. I think I am running 8 or 10 on mine and I only have 8 cpu cores and you have 16. Just depends on how long you want to limit your run time to. I just kept on increasing the number of units I was running until it was not that advantageous to run anymore.

Good Luck,

Allen

Just changed app_config from .5 gpu to .333
Hope this goes well.

Thanx again, Skip

I would like to be able to run a Ryzen 5600G, an R9 280X and an S9150, all on one mother board. The PSU is beefy enough to do this. A query into AMD yielded this result:

Thank you for the e-mail.
Based on the inquiry, i understand that you are using multiple GPU on the same system and want to know if the driver work simultaneously.
No, the driver won't work together as all the 3 components requires different driver sets. If the system components meet the same driver set the there is possibility to work.
Thank you for contacting AMD.

For the 5600G I was using this driver: Adrenalin 22.11.2 Recommended (WHQL) (Win 10)
For the R9 280X, I proposed using this: Adrenalin 22.6.1 Recommended (WHQL) (Win 10)
For the Firepro S9150, I proposed using this: 21.Q1.1 (Win 7)

I'm not so sure they understood the question. I did say I wanted to use these 3 driver sets, and they did agree that 3 different driver sets were needed.

The motherboard is an ASUS TUF Gaming B550-PRO AMD AM4. The 280X is from 2013, the S9150 is from 2014, so not that far apart, time wise. The CPU is from 2021, so quite a bit newer.

I guess the question is, Is it possible to run these 3 on one Motherboard???

(I have since replaced the 5600G with a Ryzen 3 4100, and am running the 280x. But it sure would be nice to get more horsepower here!)




Anybody have any ideas?

TIA

Mikey,

Thanks for the comeback.
Whether it's worth running 4 tasks or not, I'll know in a while.
I've just left it alone so it can settle down and determine just how long it will
take to run a WU on average. I'll let you know when I check in on it later.
I was just very surprised that I didn't have to do anything special to set it up.
I thought I would have to fix something in the bios, but there wasn't anything
in there that I could change.

Love getting feedback on my follies!

Allen

No problem, but there could be a problem if the Project can't supply you with enough tasks to keep that thing running consistently on MW tasks, then it might just be an exercise in hot and cold episodes
for the gpu. As long as you have sufficient cooling for that thing then getting it revved up shouldn't be a bad thing but if you have to wait for tasks, I know you are running the special MW version of Boinc, then the heat up and cool down cycling could kill that thing.

The difference IS huge as compared to the compute cores for each one
AMD 560 896
AMD 580 2304

There is a recurring theme posted here at Milkyway and throughout the BOINC forums that you don't want to use a Nvidia card at MW and you should always purchase a AMD because of their superior FP64 calculation rate.

Best AMD card at MW has been the Radeon VII. The 3080 Ti matches and beats that card.

No arguments about the heat production from Nvidia cards though as I look at my power bill every month from the A/C usage to try and remediate the heat output from the computers with 13 cards.

I would guess it will take time to shut down everything needed to remove and the install the new Server and that it could affect other parts of the University as well, they do share the internet connection at least, and then make sure they have the right people on hand in case something goes wrong either in hardware mismatches and in the software area as the Server does it's first boot up and gets ready to do it's thing for us. Tom did not go into details about whether it was plug and play from his end or what was involved so there's alot of guessing going on over this. There have also been suggestions to keep the old Server and repurpose it for something Boinc related, ie run the NBody tasks while the new Server handles the Separation tasks but Tom didn't say if that was even a possibility, remember Seti literally had a repurposed closet they had to fit their stuff into before they finally shut down.

can confirm - as of 25.X.22 - regularly getting the error internet access ok ut project servers are unreachable from BOINC logs on the Milkyway@Home project
did some drive crash again?

Server status shows everything running at 7pm East Coast US time

Yes, it's obvious that the computers at Milkyway are not able to keep up with the WU's we are producing. I don't understand why they are not able to keep up since we are seemingly not finishing anymore now than when it suddenly stopped being able to keep up. The System always says everything is running and I've heard very little from Tom on the subject.
Wish we knew what is wrong.

The workunit generators don't generate tasks (wingman tasks or initial tasks) if the WU pools have more tasks than they should. So when the nbody pool had like 100k tasks in it, any tasks that you sent back were essentially put on hold because the WU generator wouldn't make any wingman tasks for validation until the pool was cleared.

It's not so much that the "computer can't keep up" as it is that the WU pool being overfilled is a nasty bug. I want to migrate to the new server hardware ASAP but other people have to get things compiled and sorted out before that can happen. Currently, the server is slow becaus eit is short on memory, so in that regard the computer can't keep up.

There is a recurring theme posted here at Milkyway and throughout the BOINC forums that you don't want to use a Nvidia card at MW and you should always purchase a AMD because of their superior FP64 calculation rate.

Best AMD card at MW has been the Radeon VII. The 3080 Ti matches and beats that card.

No arguments about the heat production from Nvidia cards though as I look at my power bill every month from the A/C usage to try and remediate the heat output from the computers with 13 cards.

I don't know what times a single task on a AMD can run because I am sure most AMD users run at least 2x, 3X, 4X or more. I can run a single Separation task in under a minute. I run 2X on all my cards. Certainly better for the 3000 series compared to previous generations. So blanket statement that Nvidia cards are no good at Milkyway does not hold water.

Keith,

Another question:

I have been wondering if I can put two of my RX 560's in one machine. I have a RX 570 that I could put in the other machine. Also, what about putting two 1050ti's in one machine? Non of them require the connecting wires between the cards. Just thinking......

Thanks,

Allen

Sure, why not? As long as the cards can fit on the motherboard, preferably directly in the PCIe slots, go for it. And you have enough PCIE power cables to feed them and sufficient power capacity from the power supply.

If you two or more of the same type card in the host, BOINC will crunch on both of them as long as they are detected at startup. If you put dissimilar cards into the host, then BOINC will only crunch on the most capable card. In your example, then only the 570 will get detected and crunched with and the 560 will be ignored.

To get around that problem you need to set a cc_config.xml parameter in the <Options> section. <use_all_gpus>1</use_all_gpus> and re-read config files in the Manager or restart the client and Manager to get BOINC to crunch on both the 570 and 560.

I think AMD GPUs are better suited for MilkyWay because of their high native double precision capability compared to NVIDIA cards. This is probably also the reason that their CPU usage is much less. I recently got a R9 280X (a 9 year old card) and was amazed at its capabilities in this project compared to the 3060Ti (latest gen. card) that came with my PC. After some testing I found that for most productivity I can run 4 concurrent GPU tasks (.25 GPU, .24 CPU settings). The average time per tasks is about the same as one task on 3060Ti running one at a time. Running more than 1 at a time on 3060Ti is counterproductive as the average time per task slows down too much. It'd be interesting to try NVIDIA TITAN V here as its rated double precision capability is about 7 times that of an R9 280X. Even at 5 years old, its price is about 16 times that of a R9 280X though. I think a good modern consumer grade card that'd run well on this as well as other projects is AMD RX 6800 XT. Much cheaper than 6900 XT and only ~10% less capable, with excellent single precision and very good double precision ratings.