41)
Message boards :
Number crunching :
credit comparison to other projects
(Message 9374)
Posted 30 Jan 2009 by Gavin Shaw Post: Arion. Arion asked for his numbers to be checked. That's what I did. As for credit high or low I didn't comment. I just pointed out that DA wants Seti stock as the standard, not Einstein stock. But if we are going to have a standard why don't those that are lower get asked to bring up their 'pay rate'. Rosetta pays less than Seti and I think Rosetta is more valuable. I like to help out Rosetta, but with a low pay rate it is a killer on crunchers RAC and that puts people off. I do intend to get my Rosetta contribution up to 100,000 but that will take a lot of time. |
42)
Message boards :
Number crunching :
credit comparison to other projects
(Message 9349)
Posted 29 Jan 2009 by Gavin Shaw Post: Arion. Based on your numbers: MW - 31.72 credits / hour. EH - 15.50 credits / hour. However, remember DA wants Seti to be the standard. So how does MW compare to Seti stock, or how does Einstein stock compare to Seti stock? It is possible Einstein is a lower paying project than Seti. |
43)
Message boards :
Application Code Discussion :
optimized app suggestion
(Message 9328)
Posted 29 Jan 2009 by Gavin Shaw Post: I'm also wondering how the APP_NAME/APP_VERSION stuff is supposed to work. What is the GMLE_BOINC option for? Also I have fixed the srand issue using Travis's code. But now I get 6 errors of can not open include file 'boinc_win.h' Before I had only one. Anyone else avoid this? Or is it just my library paths are incomplete? |
44)
Message boards :
Application Code Discussion :
optimized app suggestion
(Message 9295)
Posted 28 Jan 2009 by Gavin Shaw Post: Well I get it to kind of get close to compiling. I have two errors. 1. It can not open include file boinc_win.h but I wonder if that error is more related to boinc_win.cpp being an 'empty' file. 2. srand48 variable in probability.c (right near the end of the file). I think it is suppose to be drand48? void prob_ok_init() { srand48(time(NULL)); } What version / revision of the boinc source are you using (optimizers and project)? |
45)
Message boards :
Application Code Discussion :
optimized app suggestion
(Message 9252)
Posted 27 Jan 2009 by Gavin Shaw Post: I don't think you guys have read any of my posts. Unless someone makes a post with a link to their optimized application(s) (preferably in the application code discussion), there's nothing I can do about it. And I'm probably going to catch flak for this. But several people here have been asking for information on what is used to compile the win apps. We know you don't do that, Dave does. But he hasn't been around on the forum for a while. Can you get him or get him to tell you and post that information somewhere. There are a few members of the team I'm on (including me) who are discussing recompiling win apps for MW, but we would like to know what Dave uses (compilers, libraries, scripts etc). As I said linux and mac have make files which allow someone to pretty much compile the app from the word go (they only need the source and boinc api). Win has nothing like this given in the code release. Somehow I don't think Dave give multiple commands to recompile the app ever time it is needed, but rather uses a batch or script file or VS solution file. If so, can we see it. |
46)
Message boards :
Application Code Discussion :
What's needed to compile the app?
(Message 9205)
Posted 26 Jan 2009 by Gavin Shaw Post: Travis, Can you get Dave to tell us what he uses to build the win apps? If there is some kind of script, batch or make like file can that please be included in the code release. We have make files for linux and mac to give people a starting point to work from but there is nothing for win versions. Does he use Visual Studio, cygwin, mingw, plain g++ or something else? It would really be appreciated... |
47)
Message boards :
Number crunching :
v0.17 for windows
(Message 9190)
Posted 26 Jan 2009 by Gavin Shaw Post: Is the 0.17 version going into the code release dir? Or do we use 0.16 for optimising? |
48)
Message boards :
Number crunching :
Jan 24 credit change complaints/comments here
(Message 9083)
Posted 25 Jan 2009 by Gavin Shaw Post: Also, arguments that we're just trying to keep Dave Anderson happy are probably directed at the wrong person, because he's on my PhD thesis committee :P My university would consider that a conflict of interest. |
49)
Message boards :
Number crunching :
Jan 24 credit change complaints/comments here
(Message 9053)
Posted 25 Jan 2009 by Gavin Shaw Post: Soon there'll be an opt. appl., which will calculate the WU for windows in less than 6 Minutes again. And you know this for a fact? The linux people got their act together and have come good. The Windoze people, well I have no idea. Does anyone know what is required to compile for Win? |
50)
Message boards :
Number crunching :
Jan 24 credit change complaints/comments here
(Message 8993)
Posted 25 Jan 2009 by Gavin Shaw Post: I want to know what happened to all those 'optimized' win apps. Ever since the feedback that some are wrong and then the tag changes all the apps have gone. Linux guys are lucky. And I remember asking back in Dec (when I had some free time to play around) over in the app code forum what was needed to compile for windows (environment, flags etc). Never got an answer from the project. |
51)
Message boards :
Number crunching :
Reverting to Stock
(Message 8894)
Posted 23 Jan 2009 by Gavin Shaw Post: I known a stock version needs to be the one that runs on any architecture supported by the project but... Probably too much work to maintain all the versions and bug check them for each update. Would be nice as I have C2 processors capable of functions up to SSSE3 that could speed up the process. But then that would just make me a credit whore wouldn't it? |
52)
Message boards :
Number crunching :
What applications will be valid?
(Message 8803)
Posted 21 Jan 2009 by Gavin Shaw Post: Could you add something extra in the stock app so that it would be noticed as the true stock. And they probably also would like a working opt app too? |
53)
Message boards :
Application Code Discussion :
Recompiled Linux 32/64 apps
(Message 8722)
Posted 20 Jan 2009 by Gavin Shaw Post: I suppose the question then has to be asked. Why are the linux opt apps giving good results, but the Win ones are causing a load of trouble? Is it a compiler flag difference or code change? And I'm sure there are a lot of Win users (and maybe MAC users too) who would like to have similarly optimized apps (that return good results). |
54)
Message boards :
Number crunching :
New faster application?
(Message 8440)
Posted 16 Jan 2009 by Gavin Shaw Post: And can anyone confirm what Logan brought up? Then perhaps the project can specify what the accepted 'error margin' is. Or is that the no difference greater than 10^-10ish. |
55)
Message boards :
Number crunching :
New faster application?
(Message 8436)
Posted 16 Jan 2009 by Gavin Shaw Post: So where do we stand? As Labbie said can people who have tested specify what download they used. Also can anyone get in contact with the guys/gals (or they themselves) who made the apps and find out if they did any testing before release. For your information I have been using the SSE3 version from the MW.rar file downloaded from Ice's site. For the moment I have set the machine to NNW and I will let it crunch out the remaining units. Then I guess I'll take it elsewhere while this is all sorted out. And can anyone confirm what Logan brought up?
If this is the case does that not mean there could be an issue with the stock app too? |
56)
Message boards :
Number crunching :
Vote for the MilkyWay favicon!
(Message 8176)
Posted 4 Jan 2009 by Gavin Shaw Post: I go for number 7. |
57)
Message boards :
Number crunching :
So whats DA gonna do about CUDA
(Message 7914)
Posted 22 Dec 2008 by Gavin Shaw Post: But CUDA uses the GPU to do work right? And GPU is different to CPU. So it would not be fair let alone right to compare the two versions to each other. Maybe comparing CUDA at project A to CUDA at project B and making them pay roughly the same is what he could do. Note: This post contains no support either way (for or against) on the issue of project parity. That debate can be done by others. I have my view but I'll keep it to myself for now. :) |
58)
Message boards :
Application Code Discussion :
What's needed to compile the app?
(Message 7910)
Posted 22 Dec 2008 by Gavin Shaw Post: Well I do have Visual Studio 2005 already (the joy of being a student here is you can get free software from Microsoft) and it does include C/C++. So I guess I fire it up and bring the source into it and try to compile? |
59)
Message boards :
Application Code Discussion :
What's needed to compile the app?
(Message 7908)
Posted 22 Dec 2008 by Gavin Shaw Post: Kind of a silly question, but since it is Christmas and there is a slow down at uni here. Would be interesting for me to try (especially since I am waiting for several instances of my research work to run on the HPC). How would you go about compiling a 32bit Win (XP) version? I have the 0.7 source (which has make files for linux and osx) and gcc-g++-4.3.2. I assume I would have to get the libraries as well. But with no make file for Win (probably because it doesn't use one) I don't really know the procedure. Looking at the linux make file I see references to the operating environment I would assume they would have to be changed to match the Win 32bit OS? I mainly want to add in sse? support/usage since my systems have it and the current 32bit app doesn't use it. |
60)
Message boards :
Number crunching :
credit comparison to other projects
(Message 7803)
Posted 17 Dec 2008 by Gavin Shaw Post: Whats about girls? Every milion u get one for free, deal? I don't think my wife would let me go for that deal... Then again, why would I need more? Doesn't having one girl cause enough trouble? Lol... |
©2024 Astroinformatics Group