1) Message boards : Graphics cards (GPUs) : GTX 1080 (Message 43832)
Posted 7 hours ago by Profile Retvari Zoltan
Today I've ordered a Gainward GTX 1080 Phoenix, so I hope that things will rev up here, as the Pascals fill the market thus the computers of the potential crunchers in the following weeks.
2) Message boards : Graphics cards (GPUs) : PCI-e Bandwidth Usage (Message 43801)
Posted 4 days ago by Profile Retvari Zoltan
I was afraid of that. Really wanted to drop a good $10K on a dedicated F@H/BOINC rig and use PCI-e splitters to multiply the GPU capacity. Would have saved me a lot of money buying extra mobos and CPUs.
You don't have to buy a very expensive MB and CPU for GPU crunching, provided that you want to put only 1 GPU in every MB, and you don't crunch for CPU projects on that host. Even a recent Celeron could feed a GTX980Ti in a cheap m-ATX MB (however, I would recommend an i3 at least). You can gain 10-15% performance by using a non-WDDM OS like Linux, or Windows XP.

You think every other GPU project on BOINC would be the same?
Surely they are for some extent. Any calculation which is modelling an N-body process is much more complex (as it could need some double precision calculation, or applying extra "forces" depending on the state of the given system) than a hashing algorithm, thus it's need to be controlled by the CPU, and it needs PCIe bandwidth. There's a variation in the PCIe bandwidth requirement between different workunit batches for the GPUGrid app, it could be the same for other projects. The algorithm of "purely mathematical" projects (like primegrid, Collatz or maybe SETI@home) is more like a hashing algorithm, thus they could need less PCIe bandwidth than GPUGrid, Einstein@home or MilkyWay@home, but it could change over time. This situation is the result of that the GPUs we use for these projects are made for gaming, thus their computing capabilities are "crippled" (disabled or non-present double precision FPUs in the cores), but even the "professional" GPUs are still just co-processors, they can't do everything on their own (however their development is going to achieve this).
3) Message boards : Number crunching : Ghost workunits (Message 43797)
Posted 5 days ago by Profile Retvari Zoltan
Is there a way to cancel these ghost WUs?
No, there is not.
4) Message boards : Number crunching : Gridcoin (Message 43776)
Posted 11 days ago by Profile Retvari Zoltan
Any secrets to your earning speed, Retvari?

There's no secrets. My magnitude is 1200, and the GridCoin client is always running on my laptop.
You have to make sure that you have the same CPID across all of your projects.
5) Message boards : Graphics cards (GPUs) : nVidia Pascal X80 (Message 43770)
Posted 13 days ago by Profile Retvari Zoltan
Just tried my nice shiny new Gigabyte GTX1080 on GPUGRID and I just got Computation Errors on the two wu's I downloaded. :-(
Astroids was the same but Collatz works OK, so I know it's not the card.
I've turned off the project now (Allow no new taks) until you say it's OK for me to try again, I don't want to waste wu's or time, as I presume you need to adapt for Pascal?
Exactly. See this post:
GDF wrote:
HI,
we expect great performance from the GTX1080 but at the moment we don't have any.
As soon as we have them, we need to recompile the code for them and check it.

At the moment, the app will crash on any new pascal gpu.

gdf
6) Message boards : Graphics cards (GPUs) : GTX 1080 (Message 43762)
Posted 14 days ago by Profile Retvari Zoltan
we expect great performance from the GTX1080 but at the moment we don't have any.
As soon as we have them, we need to recompile the code for them and check it.
When will this happen?
7) Message boards : GPUGRID CAFE : The milestone thread (Message 43743)
Posted 17 days ago by Profile Retvari Zoltan
I'm passed over 7 billion credits overnight.
8) Message boards : Number crunching : cant get any new tasks... why? (Message 43742)
Posted 17 days ago by Profile Retvari Zoltan
Anyone have a solution for this? Maybe an admin reset the quota? I am having this problem and would love for it to just go away.
If you have this problem on host 327304, then you should put back the GTX980Ti, as GPUGrid is not running on GTX 1080 yet.
Your host produced too much errors, and therefore the server stopped sending work to this host.
9) Message boards : Graphics cards (GPUs) : What Happened To The GPU Short Runs? (Message 43739)
Posted 18 days ago by Profile Retvari Zoltan
Perhaps when the new PhDs begin their research at the lab there will be short runs again.
(I guess it will be at the start of the next semester e.g. september)
10) Message boards : Graphics cards (GPUs) : GTX 1080 (Message 43737)
Posted 18 days ago by Profile Retvari Zoltan
Read through the Einstein thread. Why anyone would buy NVidia cards for running Einstein is beyond me. Maybe the new ones will perform better but I'm skeptical. I run mostly NVidia but that's because I primarily crunch GPUgrid. Have tried them on other projects and generally they're poor performers compared with AMD. Maybe I just haven't found the right project. Are there any other projects in which NV cards perform well?
I don't know the AMD cards' credit/watt ratio, but my GTX980Ti can process a Binary Radio Pulsar Search (Parkes PMPS XT) v1.57 (BRP6-Beta-cuda55) task in 1860~2000 sec. By a quick browse on the top hosts list I could not find any GPU (including AMD and NVidia) which can process a workunit from this batch faster than this. Maybe I'm a record holder at Einstein@home too?
EDIT: ok, I know that the user can set the number of simultaneous Einstein@home tasks in their profile for maximizing GPU usage/throughput which hurts runtimes, but I'm not sure if it's worth doing on Linux hosts, and I've checked a couple of such hosts too.


Next 10