I can get one for a decent price and was just curious if both GPU's on the board work as expected with seti. Doesn't seem like many are being used by folks around here...

Thanks,

Chris

---

I can get one for a decent price and was just curious if both GPU's on the board work as expected with seti. Doesn't seem like many are being used by folks around here...

Thanks,

Chris

They will work, Chris.

The reason not many people here run those is because the applications are more "optimized" for the NVIDIA cards, so cheaper cards will do more work.

The AMD cards do some things better than the NVIDIA cards, that's for-sure, but not the **current** applications.

BUT - if you can get a good price on the card - a really good price - there's nothing wrong with using that card to crunch numbers here and it should work fine.


I highlighted "current" because rumors are flying about new applications for new work units. I'm not holding my breath. But if I were, I wouldn't know if the NVIDIA cards would have their current advantage.

I would think there's no reason they shouldn't.

Based on my experience with Einstein concerning PCIe bandwidth usage and not exactly knowing how bandwidth-limited SETI is exactly...

I'd just make sure to use it on a PCIe 3.0 platform and reserve a good part of the avilable CPU power just to feed it to the max. (i.e. check the GPU loads with GPU-Z or a similar tool and experiment with the BOINC maximum CPU usage to see how many running CPU cores still allow for maximum GPU loading).

Based on my experience with PCIe 2.0 and AMD platforms, I found out at least at Einstein, the maximum PCIe bandwidth of the CPU/Chipset was already limiting a single HD7970 GPU somewhat (albeit running 2 or 3 tasks on the GPU in parallel).
So seeing the HD7990 has to feed 2 potent GPUs over a single PCIe slot, I'd definitely prefer a PCIe 3.0 platform to make sure you don't run into an unexpected bottleneck.

Of course, you'll have to own a power supply capable of safely handling the beast and a potent cooling setup to keep it from overheating. Plus, I do hope the energy cost in your region is a low one. That setup will sure draw alot of juice from the plug.

---

Thanks for the input. Its $500 for a new one, so on par with a 780 but hopefully around 50% on astropulse. For astropulse at least the 32 compute until Tahiti's seem to curb stomp all the current nvidia cards. Grants multi beam doesn't seem to be ATI's forte. I've always run nvidia except for one mac laptop so I'm curious what other folks were seeing.

And yeah, I have a PCIe 3 board and pretty much always dedicate a full core to any astropulse wu's.

Thanks again for the input.

Chris

---

I'm not interested in starting yet another brand war, but my understanding of things is that AMD's current batch of GCN-based GPUs fare very well against NV's Kepler GPUs, especially when you consider that AMD cards are generally cheaper than comparable NV ones. At least, this was the case before the cryptocurrency gold rush that spiked demand/prices on AMD cards in the US a few months back.

Of course, NV GPUs have the advantage in MultiBeam where the applications use its native CUDA while AMD GPUs use OpenCL.

With regards to PCIe bandwidth, I don't think SETI applications are severely constrained by it. It seems fine on PCIe 1.1 x8, although Einstein definitely suffers. There shouldn't be any noticeable degradation with PCIe 2.0.

Edit: Your reply came before mine - yes, Tahiti does very well on AstroPulse.

---

Soli Deo Gloria

Well I got the card installed, but for some reason Boinc only sees one of the GPU's, not both of them.

Any thoughts on how to get it to see both? I am slightly worried now that ATI's internal crossfire may be confusing boinc, but thus far I have not found a way to turn it "off" exposing both GPUs...

Anyone dealt with this before?

Thanks,

Chris

---

Nevermind, seemed to have magically resolved itself...ha

Chris

---