| log in |
|
21)
Message boards :
Number crunching :
Badges?
(Message 450)
Posted 2 Jan 2015 by frankhagen
looks pretty fine. |
|
22)
Message boards :
Number crunching :
Raspberry Pi
(Message 432)
Posted 2 Jan 2015 by frankhagen The LLR application uses "gwnum code". This code is written for INTEL processors. Therefore you cannot compile the LLR application for ARM based OS's ..... until there is "gwnum code" available for ARM processors. this will hardly ever happen simpy because afaik there is no arm-design that is IEEE 754 compatible. there are commercial libraries for newer arm-design which supoort this, but since they need to emulate on that limited arm hardware.... |
|
23)
Message boards :
Number crunching :
Raspberry Pi
(Message 403)
Posted 31 Dec 2014 by frankhagen So I don't really know, but I'll hazard a guess? Is it because the "R" in "ARM" stands for "RISC," which stands for Reduced Instruction Set Computing? Maybe these processors don't have the instructions needed for efficiently computing Fast Fourier Transforms? bingo! there are VFP coprocessor designs by arm, but you will not find such things on cheap gadgets. and for the worst they are NOT IEEE 754 compliant even on latest arm socs. plain simple: forget it. |
|
24)
Message boards :
Number crunching :
Raspberry Pi
(Message 393)
Posted 31 Dec 2014 by frankhagen first you show up wining about "large" deadlines. the next bright idea you bring up is suggesting to run FFT-math on 400Mhz 32bit INTEGER cores. funny fraggles indeed you are... |
|
25)
Message boards :
Number crunching :
Raspberry Pi
(Message 388)
Posted 30 Dec 2014 by frankhagen Do enlighten me.... finish prescool, learn just a little bit about processor architectures and you will figure out with no problem why LLR on ARM is just - well - not that bright of an idea. |
|
26)
Message boards :
Number crunching :
Raspberry Pi
(Message 385)
Posted 30 Dec 2014 by frankhagen Lol. The rat pack has lots of them(or at least I do). But rebirther is correct; they are very slow :( I was just hoping to get lucky and find a good use for them other than attracting dust. nice side step! thought you would come up with an llr-app for arm hardware. i bet you know what ARM stands for and why... |
|
27)
Message boards :
Number crunching :
Raspberry Pi
(Message 378)
Posted 30 Dec 2014 by frankhagen There is no LLR version for ARM. well, the rat pack just voluntered to provide one - did they? |
|
28)
Message boards :
Number crunching :
TOP5000 entries
(Message 349)
Posted 28 Dec 2014 by frankhagen They are using another boinc system than me. I will contact the finder if there is a TOP5000 entry. fair enough. automatic like PG does would be a nice extra. ;) |
|
29)
Message boards :
Number crunching :
New Tasks?
(Message 220)
Posted 20 Dec 2014 by frankhagen Waiting for more tasks to crunch. well - it's them slow crunchers we have to wait for.... |
|
30)
Message boards :
Number crunching :
New Tasks?
(Message 207)
Posted 18 Dec 2014 by frankhagen I am waiting for the rest to remove canditates from the sieve file. reduce the deadline to 6 hours - basta. |
|
31)
Message boards :
News :
cable provider interruptions
(Message 201)
Posted 15 Dec 2014 by frankhagen unless murphy strikes and something else happens. |
|
32)
Message boards :
Number crunching :
HTTP File not found
(Message 178)
Posted 13 Dec 2014 by frankhagen <core_client_version>7.2.7</core_client_version> <![CDATA[ <message> process exited with code 195 (0xc3, -61) </message> <stderr_txt> 13:29:29 (3706): wrapper (7.2.26012): starting 13:29:29 (3706): wrapper: running llr64 ( -d -oPgenInputFile=input.prp -oDiskWriteTime=10 -oOutputIterations=10000 -oResultsFileIterations=99999999) execv() failed: : Text file busy 13:29:30 (3706): llr64 exited; CPU time 0.000000 13:29:30 (3706): app exit status: 0x6c00 13:29:30 (3706): called boinc_finish |
|
33)
Message boards :
News :
400 test WUs for all apps loaded
(Message 159)
Posted 12 Dec 2014 by frankhagen well - bummer. no linux-32 at hand... |
|
34)
Message boards :
Number crunching :
no checkpoints
(Message 64)
Posted 30 Nov 2014 by frankhagen yes the app does write checkpoints and picks them up when restarted. and no - the silly wrapper of course does not pick up the time elapsed already. SNAFU. |
|
35)
Message boards :
News :
Next plans
(Message 63)
Posted 30 Nov 2014 by frankhagen since this is LLR, why not adopt the credit formula they use over at PG? |
|
36)
Message boards :
News :
Next plans
(Message 39)
Posted 29 Nov 2014 by frankhagen I got an answer from primegrid about the validation. Its much more complicated than I thought and need all the time testing to catch up the correct new values so credit new is the only way I can go here, unfortunately. OMFG! |