I had no problems watching 1080p streamed content while gpu folding. Only problems I had was when I forgot to stop gpu folding before toggling crossfire for gaming.
I'm not trying to argue that folding with nvidia isn't better - just trying to make it clear that folding on amd is pretty painless (and quite productive).
That was the opposite of my experience, but it was several years ago. I'm very happy they have improved the support since then. More folding is good for everyone!
Yeah, the v7 client is pretty sweet - easy to set up cpu/gpu folding
Indeed. Once they get the weird point calculation quirks and bad GPU identification stuff out of the way, it will be quite glorious. I'm also looking forward to HFM getting updated to work with v7 client (simply because I imagine the quality will be better than v7s internal points estimation and seeing totals or tracking a lot clients will still be simpler for HFM.
On my AMD quad, it was definitely beneficial to drop one core (SMP v6) to give the GPU client on my 5770 more CPU time (although surely with smaller margins than Shwaip saw with a 5870).
Interestingly though (perhaps only to me with my vintage hardware), by moving my 5770 to my home rig and turning off SMP folding there, I increased my production slightly. That's with one less client running, but the SMP client on the Athlon gets all 4 cores. The SMP client on the Celeron was only making about 1000 PPD, I think. Now the Celeron appears 100% used by the two GPU clients, but I can tell it's not as fully used as the SMP client or any benchmark would use it. The CPU runs far cooler than if it were still running SMP. I thought about throwing on a single core client to get a bit more PPD, but it's not worth the extra maintenance.
While we're throwing around anecdotes, when I was folding with an nvidia 8800 (over a year ago), my display had much more lag. I had to pause it to get anything done, but no problems with my AMD cards.
Maybe part of the guide should be making the most of your hardware. When to use the regular client, SMP, GPU, how many cores, etc. That overall guide could link to more depth on GPUs and such. I guess v7 will outdate a lot of that advice, though.
While we're throwing around anecdotes, when I was folding with an nvidia 8800 (over a year ago), my display had much more lag. I had to pause it to get anything done, but no problems with my AMD cards.
Maybe part of the guide should be making the most of your hardware. When to use the regular client, SMP, GPU, how many cores, etc. That overall guide could link to more depth on GPUs and such. I guess v7 will outdate a lot of that advice, though.
I have Nvidia cards exclusively now, and only experience that problem when I set the priority to low rather than idle (negligible lag when on idle). Who knows if it is a client change or that I'm running faster cards.
I wish I had enough time and access to equipment, but the looming changes invalidate a lot of the advice I'd be giving anyways. I am doing a separate v7 guide, but am kinda holding off to let it get closer to gold.
I never had any desktop lag issues with my 88GT G92 while folding but that was back when NVidia needed a core to fold and it was worth it still running an affinity changer to control the clients better.
Comments
That was the opposite of my experience, but it was several years ago. I'm very happy they have improved the support since then. More folding is good for everyone!
Indeed. Once they get the weird point calculation quirks and bad GPU identification stuff out of the way, it will be quite glorious. I'm also looking forward to HFM getting updated to work with v7 client (simply because I imagine the quality will be better than v7s internal points estimation and seeing totals or tracking a lot clients will still be simpler for HFM.
Interestingly though (perhaps only to me with my vintage hardware), by moving my 5770 to my home rig and turning off SMP folding there, I increased my production slightly. That's with one less client running, but the SMP client on the Athlon gets all 4 cores. The SMP client on the Celeron was only making about 1000 PPD, I think. Now the Celeron appears 100% used by the two GPU clients, but I can tell it's not as fully used as the SMP client or any benchmark would use it. The CPU runs far cooler than if it were still running SMP. I thought about throwing on a single core client to get a bit more PPD, but it's not worth the extra maintenance.
While we're throwing around anecdotes, when I was folding with an nvidia 8800 (over a year ago), my display had much more lag. I had to pause it to get anything done, but no problems with my AMD cards.
Maybe part of the guide should be making the most of your hardware. When to use the regular client, SMP, GPU, how many cores, etc. That overall guide could link to more depth on GPUs and such. I guess v7 will outdate a lot of that advice, though.
I have Nvidia cards exclusively now, and only experience that problem when I set the priority to low rather than idle (negligible lag when on idle). Who knows if it is a client change or that I'm running faster cards.
I wish I had enough time and access to equipment, but the looming changes invalidate a lot of the advice I'd be giving anyways. I am doing a separate v7 guide, but am kinda holding off to let it get closer to gold.