| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 21
|
|
| Author |
|
|
mdxi
Advanced Cruncher Joined: Dec 6, 2017 Post Count: 109 Status: Offline Project Badges:
|
- is the measured power when undervolted also after subtracting 24.5 W idle power? (I guess so?) Yes. - was any GPU installed? Or was this a headless system? Normally, headless. For testing purposes I needed to be able to see the BIOS, frequently, so I had an old video card plugged in. Of course that was accounted for in the power measurements. - how are points per hour of run time for completed MIP units for the different CPUs? This might also be interesting. For me, completion times for MIP fluctuate wildly, from below 1 hour to well above 3 hours Yeah, there's a whole section of the doc that talks about that specific thing. You're seeing wild variances because Rosetta (the underlying bioinformatics package) wants to use enormous amounts of L3 as a hot data cache. So on anything other than an actual server-class CPU (or, now, the high end Ryzen 3ks), it's very very easy to oversubscribe the L3 cache when running MIP1 WUs. You can read more about this in the MIP thread, here on the forums. That's where I learned about it. ![]() |
||
|
|
|