I've NEVER crunched APs on my video cards. Do they create more heat than MBs?
Also since I run Lunatics would I have to reinstall to do so?

---

I Desire Peace and Justice, Jim Scott (Mod-Ret.)

In my experience APs are less demanding of the GPU, but more demanding of the CPU.
As for the Lunatics installer, I've had luck and installed the AP apps without having to re-run the whole installation, and I've had to re-run the whole installation. I think it is possible to just install the AP app without running the installer just by downloading the app and moving it into the BOINC/projects/setiathome folder, but I've not tried that myself.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

In my experience APs are less demanding of the GPU, but more demanding of the CPU.
As for the Lunatics installer, I've had luck and installed the AP apps without having to re-run the whole installation, and I've had to re-run the whole installation. I think it is possible to just install the AP app without running the installer just by downloading the app and moving it into the BOINC/projects/setiathome folder, but I've not tried that myself.

You would need at least one extra step - to merge the extra 'app_info' information for the additional application into the single app_info.xml file which is all that BOINC allows. It's easy to do, but requires a little care and some understanding of what is going on.

It's easier to re-run the whole installer, re-selecting your existing applications and adding the AP choices as well.

AstroPulse tasks may use a bit more CPU than MultiBeam but well-tuned AstroPulse applications will also push the GPU a lot harder and often run them at higher temperatures compared with MultiBeam.

---

Soli Deo Gloria

AstroPulse tasks may use a bit more CPU than MultiBeam but well-tuned AstroPulse applications will also push the GPU a lot harder and often run them at higher temperatures compared with MultiBeam.

That pretty much answered my question...I don't want to over stress my OLD stuff until I can afford to rebuild.(About 20 more months.) I'm on a very tight budget. Thanks folks.

---

I Desire Peace and Justice, Jim Scott (Mod-Ret.)

AstroPulse tasks may use a bit more CPU than MultiBeam but well-tuned AstroPulse applications will also push the GPU a lot harder and often run them at higher temperatures compared with MultiBeam.

That pretty much answered my question...I don't want to over stress my OLD stuff until I can afford to rebuild.(About 20 more months.) I'm on a very tight budget. Thanks folks.

Actually, running APs on AMD video cards don't stress your CPUs much more than MBs on AMD cards. However, the difference in RAC is about 4x times as much when running APs on AMD cards. The difference isn't nearly as much with nVidia cards. I don't see why there would be any problem running APs on your AMD 7770, which really isn't an old card. Just make sure you leave one CPU core free to feed the GPU.

AstroPulse tasks may use a bit more CPU than MultiBeam but well-tuned AstroPulse applications will also push the GPU a lot harder and often run them at higher temperatures compared with MultiBeam.

That pretty much answered my question...I don't want to over stress my OLD stuff until I can afford to rebuild.(About 20 more months.) I'm on a very tight budget. Thanks folks.

Actually, running APs on AMD video cards don't stress your CPUs much more than MBs on AMD cards. However, the difference in RAC is about 4x times as much when running APs on AMD cards. The difference isn't nearly as much with nVidia cards. I don't see why there would be any problem running APs on your AMD 7770, which really isn't an old card. Just make sure you leave one CPU core free to feed the GPU.

I do not have a problem with feeding my GPU. I am letting it run on it's own. I run four cores and 0.126 CPUs + 1 AMD/ATI GPU.

---

AstroPulse tasks may use a bit more CPU than MultiBeam but well-tuned AstroPulse applications will also push the GPU a lot harder and often run them at higher temperatures compared with MultiBeam.

That pretty much answered my question...I don't want to over stress my OLD stuff until I can afford to rebuild.(About 20 more months.) I'm on a very tight budget. Thanks folks.

Actually, running APs on AMD video cards don't stress your CPUs much more than MBs on AMD cards. However, the difference in RAC is about 4x times as much when running APs on AMD cards. The difference isn't nearly as much with nVidia cards. I don't see why there would be any problem running APs on your AMD 7770, which really isn't an old card. Just make sure you leave one CPU core free to feed the GPU.

I do not have a problem with feeding my GPU. I am letting it run on it's own. I run four cores and 0.126 CPUs + 1 AMD/ATI GPU.

I suppose what one considers a problem would be the difference. If I had an AMD 7790 and it was running slower than My 7750 I would consider that a problem. A CPU starved GPU would generally produce slower and inconsistent run times. AMD CPUs, especially old AMD CPUs, are generally less capable at feeding a GPU than Intel CPUs. The OP has an Old Dual Core AMD CPU. It would be advisable if the OP freed at least one CPU core to feed his much newer GPU while running APs.

I have a similar Old Dual core machine as the OP, it is an Intel but still has One CPU core free to feed an Older ATI 6770. It is producing times similar to your much faster 7790. Have you tried freeing a CPU core? You might also add the CMDlines I'm using on your card to see if that would help. In your setiathome.berkeley.edu folder you should find the AstroPulse ReadMe file, it has CMDline options listed. It also suggests using CPU_lock instead of freeing a CPU core, you could try that as well although I've found CPU_lock to be host dependent. The CMDline option oclFFT_plan seems to work on most hosts and gives a noticeable speed increase.

I just don't track it anymore.
The killawatt meters on my top rigs serve more to give me a heads up when something crashes...
If I know that rig 'x' usually draws 800 watts and I see it down at 400, I know a GPU or two has crashed and needs to be rebooted.
I believe that running all MB actually taxes the systems more than all AP, but it is so close that I don't have to keep tabs on it.

I CAN tell you that running GPUGRID is a significantly bigger GPU taxer than Seti by 10-20%.
My electric bill went up accordingly the month I was running almost all GPUGRID.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I just don't track it anymore.
The killawatt meters on my top rigs serve more to give me a heads up when something crashes...
If I know that rig 'x' usually draws 800 watts and I see it down at 400, I know a GPU or two has crashed and needs to be rebooted.
I believe that running all MB actually taxes the systems more than all AP, but it is so close that I don't have to keep tabs on it.

I CAN tell you that running GPUGRID is a significantly bigger GPU taxer than Seti by 10-20%.
My electric bill went up accordingly the month I was running almost all GPUGRID.

I run about 210 watts and if it drops to about 120, the GPU ran out of work. If it drops to about 40 the system ran dry.

---

I just don't track it anymore.
The killawatt meters on my top rigs serve more to give me a heads up when something crashes...
If I know that rig 'x' usually draws 800 watts and I see it down at 400, I know a GPU or two has crashed and needs to be rebooted.
I believe that running all MB actually taxes the systems more than all AP, but it is so close that I don't have to keep tabs on it.

I CAN tell you that running GPUGRID is a significantly bigger GPU taxer than Seti by 10-20%.
My electric bill went up accordingly the month I was running almost all GPUGRID.

I run about 210 watts and if it drops to about 120, the GPU ran out of work. If it drops to about 40 the system ran dry.

All of my rigs have at least 2 GPUs.
So, if GPU work runs out, the drop in power consumption is very dramatic.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I just don't track it anymore.
The killawatt meters on my top rigs serve more to give me a heads up when something crashes...
If I know that rig 'x' usually draws 800 watts and I see it down at 400, I know a GPU or two has crashed and needs to be rebooted.
I believe that running all MB actually taxes the systems more than all AP, but it is so close that I don't have to keep tabs on it.

I CAN tell you that running GPUGRID is a significantly bigger GPU taxer than Seti by 10-20%.
My electric bill went up accordingly the month I was running almost all GPUGRID.

I run about 210 watts and if it drops to about 120, the GPU ran out of work. If it drops to about 40 the system ran dry.

All of my rigs have at least 2 GPUs.
So, if GPU work runs out, the drop in power consumption is very dramatic.

If I were running both GPUs and the CPUs, the power demand would be about 300 watts which causes the cip temperatures to reach boiling. To make the system last far longer, one GPU crunches and the other drives the display.

---

If I were running both GPUs and the CPUs, the power demand would be about 300 watts which causes the cip temperatures to reach boiling. To make the system last far longer, one GPU crunches and the other drives the display.

I run all rigs side-off. With a few 120mm fans blowing ambient air at the GPUs, so the effects of having two of them beasties one atop the other are minimized a bit.
All also have air cooled CPUs. Most have Thermaltake Big Typhoon units, and they have worked flawlessly over the years. I have upgraded most of their fans though.
2 of my rigs are sporting Cooler Master V8's. The are working OK as well.
The thing I do NOT like about them is that the fan is situated between the finned heat sinks, and are not readily removed for a simple cleaning unless you tear the whole thing down.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster