Posts by Uioped1
|
41)
Message boards :
Number crunching :
Configuring dual core dual processor
(Message 248852)
Posted 16 Feb 2006 by  Uioped1
Post:
2 physical processors, each with 2 cores, which each run two threads (Hyperthreading -- should show up as logical processors) Thus, if everything is correct, you should show 8 processors. . . I checked on the windows license, and it checks the cpu limit based on physical processors, not on the number of cores, as I was worried about, so that's not the problem. Edit: Didn't see your last post. 2. If the results aren't identical to the 'normal' application, doesn't this kind of pollute the stream? Yes, it does. Crunch3r's optimized version should give results that are strongly similar to the regular seti results though. Also, the system is built to handle a certain amount of garbage.
Yes, they do. Much of the optimizations are specific to the characteristics of your specifc processor, (How much cache, which instruction set) that they do not currently have the ability (I think) to check in boinc. Also it would be a lot more work on the testing side to have to test and maintain 4 or 5 app versions for each OS. They are working on incorporating the non processor-specific optimizations into the next generation science app -- Seti Enhanced. |
|
42)
Message boards :
Number crunching :
Configuring dual core dual processor
(Message 248701)
Posted 15 Feb 2006 by Uioped1
Post:
Also, he's using an optimized seti science app. I highly recommend getting one! Currently the fastest client for windows can be found at Crunch3r's site. Instructions can be found on the boinc wiki. Note that running an optimized seti app will _reduce_ the credit requested per Workunit. I'll leave it up to you whether you install an optimized boinc client such as crunch3rs or the truxoft one mentioned in another thread here. I'm confused as to whether you have one physical dual-core chip, or 2. If you have two, shouldn't you show 8 processors? Maybe you should enable hyperthreading as well? Or do you not have access to a version of windows that will allow 8 procs? |
|
43)
Message boards :
Number crunching :
Pulse Power
(Message 247632)
Posted 14 Feb 2006 by Uioped1
Post: How high do the pulse power has to be interesting, i have one at 7,42. never seen one this high before. I may be wrong, but as I understand it it's not so much the power of the pulse as it is the combination of reported factors. RFI may cause a pulse that is more powerful, but otherwise distinguishable from E.T. |
|
44)
Message boards :
Number crunching :
Getting the most out of a Pentium D Chip
(Message 246425)
Posted 11 Feb 2006 by Uioped1
Post: You could start with the latest optimized client from Crunch3r |
|
45)
Message boards :
SETI@home Science :
Just Mucking Around with Euler
(Message 174239)
Posted 4 Oct 2005 by Uioped1
Post: Well, I just can't get past the symantics of your equation; need a lot more punctuation, or perhaps a nice graphic. Also, maybe you could provide a refresher on the concept of ininity, for those of us who haven't had math courses for a while ;) Here's my favorite math problem that's been bugging me since middle-school geometry: Give a nice, succinct explanation why the limit of the length of the path of an approximation of an diagonal line using a number of horizontal+vertical steps as the number of steps approaches infinity does not approach the length of the original line. |
|
46)
Message boards :
Number crunching :
POLL for power users
(Message 167521)
Posted 14 Sep 2005 by Uioped1
Post: I have many classic machines that will never be migrated. (Ex girlfriends, old jobs...) They will just keep on requesting new work until someone decides to rebuild them. As for at home, my newer two computers are running boinc, and two older computers run classic (when they're on that is...) One laptop doesn't have anything, just because I don't like to expose them to such unremitting heat. |
|
47)
Message boards :
Number crunching :
Sparc Vs. Intel
(Message 161668)
Posted 1 Sep 2005 by Uioped1
Post: Would it not be better to have single processes making more data request in sequence than having 4 making request for data at the same time? Actually you've got it exactly backwards. Even on a 400 mhz cpu, computation and memory access will be _much_ faster than disk IO, even on local drives. So, if you have one process emitting IO requests very fast, your results will be more time spent blocked on those IO requests. Even if had some fancy non-blocking scheme that allowed the process to continue working until it needed the results of that IO, you would find that you'll be better off with multiple requests from different processes. Furthermore, raid is a technology that only delivers performance increases if you have parallel data requests, by splitting volumes across multiple drives and optimizing the request order. |
|
48)
Message boards :
Number crunching :
Back in service!
(Message 160372)
Posted 31 Aug 2005 by Uioped1
Post: Still none for me, but bandwidth usage can be seen here -> http://fragment1.berkeley.edu/~cricket/inr-668-interfaces.html There is a strange downturn on the graph now as well. Maybe something upstream can't handle the heat? Or perhaps somethings been turned off that doesn't show up on the status page. Whatever it is, my uploads started to go, but about half way through the first one it died. Tried resending several times in the last half hour, to no avail :( |
|
49)
Message boards :
Number crunching :
5th: Compiling faster Windows client with open source programs...
(Message 158503)
Posted 28 Aug 2005 by Uioped1
Post: Metod, Sorry, my question wasn't clear, so now I don't know what you meant. Did you mean the official binaries or the optimized versions? I wanted to know what the official binaries were compiled with. Thanks! |
|
50)
Message boards :
Number crunching :
5th: Compiling faster Windows client with open source programs...
(Message 157993)
Posted 27 Aug 2005 by Uioped1
Post: Does anyone know which compiler has been used to make the posted windows binaries? |
Previous 20
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.