I'll try to test it somewhere, my NAS has PowerPC processor, so I can't test it there.

I'd be glad to be of some help.

I've been helping out at Yoyo@home to get a couple of applications running on ARMv5 too, as you can see starting here.

TIA

---

Again, I'd be glad to help you out. I could compile to application on my ARM NAS if you'd like it.

---

In spite of the ARMv5 lacking HW FP, it's clocked higher than the ARMv6 on the RaspberryPi. Again, I reiterate my willingness to help out creating the app for this platform, arm-unknown-linux-gnueabi.

TIA

---

Guys, I could not help noticing that you mentioned that you got running times of 40-80 hours for an Asteroids@Home task on your Raspberry Pis (even if this was 1-1.5 years ago). My Asteroids@Home tasks run for 210-230 hours on both my Pis involved in this project.

My Pis are not overclocked at all, but still it seems that there's a huge running time difference between what you have measured in the past and I am measuring now. Have A@H tasks become larger lately? Or could it be that my Pis are not set up well? They run the latest Raspbian and BOINC 7.027.

My running times of 210-230 hours are a big problem because they can barely make the deadline. In fact it looks like the currently running tasks won't make it.

Any ideas?
Thanks!

---

http://iqjar.com

The WUs are now longer than before. I will adjust the deadline.

I took all my Pi's over to E@H after the last size increase

I did post about an updated FFT library for the Pi here that might yield a speed increase, however the project would need to revise their app to make use of it. Of course it all depends on if the app uses FFT's.

---

BOINC blog