Posts by HA-SOFT, s.r.o.
1)
(Message 6040)
Posted 13 Nov 2018 by HA-SOFT, s.r.o. Post: Beta comparsion RTX2080 vs 1050 Ti: http://asteroidsathome.net/boinc/workunit.php?wuid=96455209 |
2)
(Message 6037)
Posted 12 Nov 2018 by HA-SOFT, s.r.o. Post: Hello Jacob, I know about this problem, some workunits hang when calculating 10th pole and hang with never ending cycle (100% CPU usage). Now I'm waiting for Visual Studio 2013 to debug nvidia code. I have one workunit prepared for debug with this problem. This bug is very often on RTX cards. Zdenek |
3)
(Message 6015)
Posted 26 Oct 2018 by HA-SOFT, s.r.o. Post: My new RTX2080 is somewhere between me and eshop. I will inspect optimal coverage and setup for blocks/cores next week. |
4)
Posted 16 Mar 2017 by HA-SOFT, s.r.o. Post: AMD powered miracles, nothing less :-D So it's ok :-) Compiler never brings any speedup except 32->64 bit version without SSE,AVX,FMA. |
5)
Posted 15 Mar 2017 by HA-SOFT, s.r.o. Post: http://asteroidsathome.net/boinc/results.php?hostid=312297 And what do you expect in terms of speed of an app? |
6)
(Message 4940)
Posted 13 Jun 2016 by HA-SOFT, s.r.o. Post: New NVidia card brings new Compute capability (CC=6.1) so we need upgrade our application. I will look into this problem today. Zdenek |
7)
(Message 4848)
Posted 12 Apr 2016 by HA-SOFT, s.r.o. Post: Go to your accoount home on website, then click on Asteroids@home preferences, click on Edit preferences and uncheck Use NVIDIA GPU - click Update preferences. Wait or do project update in boinc manager on your pc. |
8)
(Message 4820)
Posted 4 Mar 2016 by HA-SOFT, s.r.o. Post: I have focused on FMA4 app for new intel CPU's. I will look on GPU apps with 9xx cards, but I have no timeline for now. Happy crunching. |
9)
Posted 8 Jan 2016 by HA-SOFT, s.r.o. Post: Not now. I will check grid dimensions for 9xx cards. Actual CUDA app was built before 9xxx had been released. |
10)
(Message 4617)
Posted 4 Sep 2015 by HA-SOFT, s.r.o. Post: Sorry guys. No free time. |
11)
Posted 11 Jun 2015 by HA-SOFT, s.r.o. Post: There is absolutely no difference in speed between sse2 and ss3 apps. |
12)
(Message 4471)
Posted 28 May 2015 by HA-SOFT, s.r.o. Post: Could be correct say that highly optimized (double precision) apps do not benefit by the HT, while less optimized benefit more by the HT ? For double precision: yes. |
13)
(Message 4464)
Posted 25 May 2015 by HA-SOFT, s.r.o. Post: Did you restart boinc client? or Did you place app_info in correct location? or Is the real name of file app_info.xml or app_info.xml.txt (win nightmare)? |
14)
(Message 4430)
Posted 6 May 2015 by HA-SOFT, s.r.o. Post:
Because number of double precision units match number of physical cores. |
15)
(Message 3888)
Posted 15 Dec 2014 by HA-SOFT, s.r.o. Post: No problem with periodic outage every day, I think. Work buffer in boinc should be enough to live with it. Zdenek |
16)
(Message 3723)
Posted 28 Oct 2014 by HA-SOFT, s.r.o. Post:
We need double precision floating point instructions and CC 1.1 do not support them. |
17)
(Message 3669)
Posted 9 Oct 2014 by HA-SOFT, s.r.o. Post: As I know our database is constantly overloaded. Kyong is planing second server for mysql db, but this may take a time (and money). Stay tuned and happy crunching Zdenek |
18)
(Message 3586)
Posted 26 Aug 2014 by HA-SOFT, s.r.o. Post: I think it's one raid5 volume with 7 disks. |
19)
(Message 3527)
Posted 15 Aug 2014 by HA-SOFT, s.r.o. Post: First, we are very sorry for this situation. Admin of Asteroids@home (Kyong) went on vacation. He will be back on Sunday 17.8. Big thanks for your patience. Zdenek |
20)
(Message 3476)
Posted 11 Aug 2014 by HA-SOFT, s.r.o. Post: We have now two volunteers who try to make OpenCL app for AMD cards. Next I'm so busy at work for now. If both of them failed to do OpenCL app, I will do one. Sorry guys ... I'm still very busy at work. Meanwhile I made success at court of law and I can finally spent my free time with my son. |
Next 20