Reduced hash rate over PCI-E x1 | Bitcointalksearch.org

the_joey_o

newbie

Activity: 40

Merit: 0

Quote from: cicada on July 12, 2011, 01:41:43 PM

Overheating could definitely cause it, if you aren't monitoring your temps you should start. The GPU will automatically throttle to reduce temperature.

Burning anything smell is usually a bad indication Wink

Yeah. Thanks. It was overheating.

cicada

full member

Activity: 196

Merit: 100

Overheating could definitely cause it, if you aren't monitoring your temps you should start. The GPU will automatically throttle to reduce temperature.

Burning anything smell is usually a bad indication Wink

the_joey_o

newbie

Activity: 40

Merit: 0

Quote from: cicada on July 12, 2011, 09:43:39 AM

The only difference between PCIe 1x and PCIe 16x is bandwidth, and mining requires very little. Some of the FPGA miners I've read about use simple serial links for getwork calls, and that's significantly less available bandwidth than a PCIe 1x slot on the mobo.

I'd look for other differences between the cards - make sure GPUz is showing the proper shader counts, gpu/mem speeds are equal, and that you're using the same miner flags.

Power issues wouldn't usually show themselves as reduced hashrates - likelier you'd see the card not responding at all in that case. Or in a dual-GPU card, maybe one of the GPUs would be inactive, effectively halving your hashrate.

Another good test would be to swap that card into the mobo and move another out to the extender; if you still get the same 140mhash directly connected, and everything else is equal, something is up with your card.

I switched out the card outside the case with one that was inside. The new out-of the-case card performed much slower than the old one, while the old one performed at the full 180 MHash/s rate while inside the case. This problem doesn't show up right away. All three cards start at a full 180 MHash/s, and then the card outside the case slows down. Could this be caused by overheating? There was a burning plastic smell this time.

cicada

full member

Activity: 196

Merit: 100

The only difference between PCIe 1x and PCIe 16x is bandwidth, and mining requires very little. Some of the FPGA miners I've read about use simple serial links for getwork calls, and that's significantly less available bandwidth than a PCIe 1x slot on the mobo.

I'd look for other differences between the cards - make sure GPUz is showing the proper shader counts, gpu/mem speeds are equal, and that you're using the same miner flags.

Power issues wouldn't usually show themselves as reduced hashrates - likelier you'd see the card not responding at all in that case. Or in a dual-GPU card, maybe one of the GPUs would be inactive, effectively halving your hashrate.

Another good test would be to swap that card into the mobo and move another out to the extender; if you still get the same 140mhash directly connected, and everything else is equal, something is up with your card.

Jabba

member

Activity: 109

Merit: 11

Definitely no performance reduction. I'm running 4 x 5870@950 getting 405 MHashes out of them. All cards connected via x1 (hardware wise).

Jabba

the_joey_o

newbie

Activity: 40

Merit: 0

I've got 3 5770s in my machine. One of them uses a PCI-E x16 to PCI-E x1 extender cable adapter that I got from Cablesaurus and is sitting outside of my machine. The two cards in my machine get right about 180 MHash/s. The one out of my machine, running across the PCI-E x1 adapter cable only gets about 140 MHash/s. I have a 1080W power supply that, according to newegg.com's PSU calculator, should be enough for 4 cards, plus all kinds of other craziness that I don't have going on in my machine. I'm using guiminer on Windows 7.

So, is anybody else seeing reduced performance over PCI-E x1?

Topic: Reduced hash rate over PCI-E x1 (Read 2207 times)