| Author | Message |
|---|
|
|
|
A month ago I made optimized clients for Windows using an evaluation copy of Intel C++ compiler and fftw library, but as I mentioned in another thread, I received money from a person to buy the license...so anyone can use the binary the binaries legally. Thank you, Mr.X!! (He wants me not to mention his name.)
Links are as follows: I added seti-sse2.zip, which I hope works for Athlon64.
seti-plainvanilla.zip For P2 or earlier....but this uses MMX instructions.
seti-p3.zip for P3 (and Athlon XP?)
seti-sse2.zip for Athlon64(?) This requires special caution. No test has been done yet.
seti-p4-no-sse3.zip for P4 w/o SSE3. According to Intel, this is slightly different from the above one. So this doesn't work for Athlon64.
seti-p4-sse3.zip for P4 with SSE3
seti-pm.zip for Pentium M.
referenceunits.zip a work unit and a result unit for benchmark test.
Most of them are the ones I uploaded before, except for seti-sse2.zip. All of them are based on Apr-12 nightly build of boinc/seti and fftw-3.0. If you want to compile by yourself (you need Intel compiler and VC++7.1), sources are
seti-source-1.zip, seti-source-2.zip, and seti-source-3.zip.
CAUTION
These binaries are provided as is. No guarantees about proper operation can be given. Only you will be to blame if binaries cash your computer, damage data or hardware, kick your dog or insult your mother. (from Metod's page and I concur.)
I strongly recommend to make a backup copy of the whole boinc folder before installing these clients in case it crashes....it may lose your precious data due to incompatibility (or by my mistake?)
Happy crunching!! You can copy or distribute them for free :)
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
 |
|
|
|
|
|
You are the best! Respect!
Only one a problem - I can't download :(
____________
|
|
|
|
|
You are the best! Respect!
Only one a problem - I can't download :(
Works fine for me, thanks Tetsuji
____________
Raithmir's SPARC64/UltraSPARC Linux Builds
http://www.kulthea.net/boinc/ |
|
|
|
|
You are the best! Respect!
Only one a problem - I can't download :(
I know that problem :) I edited and fixed the links :) SORRY!!!!! and thanks!!!! But it isn't me that boost the client, but Intel compiler and fftw library :)
Good luck all!!!
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
Thats really kool.
Thank you, Tetsuji Maverick Rai.
Thank you, Mr. X.
For all your works and efforts.
____________
http://www.cyclistlog.com/ |
|
|
|
|
|
Yes, a really _big_ thankyou Tetsuji.
Brilliant.
Genki desu. Arigato gozaimasu.
(I hope those are appripriate Japanese Thank yous' - there seem to be so many ways for different situtions!)
____________
 |
|
|
|
|
|
Many thanks Maverick and also to the anonymous Mr "X" you are both champions!
____________
 |
|
|
|
|
I know that problem :) I edited and fixed the links :)
All OK!
Tnx!
____________
|
|
|
|
|
|
I know that problem :) I edited and fixed the links :) SORRY!!!!! and thanks!!!! But it isn't me that boost the client, but Intel compiler and fftw library :)
Good luck all!!![/quote]
But without you both, there would be no optimised client. So please accept the admiration and gratitude of us all. Thank you Tetsuji and Mr. X.
Kindest regards and respect
Tony
____________
I may be getting older but I refuse to grow up!
|
|
|
|
|
|
Thank you very much, Tetsuji and Mr. X! :)
Dig |
|
|
|
|
|
Thanks everyone who made this possible. I downloaded two of the zips and I found a small problem. You still have the old app_info.xml file that does not tell boinc what to do in case it runs into a 4.09 seti work unit which is to use the 4.11 executable. Not going to be a problem for those of us that have been running optimized for the last month. However, if someone new downloads the files it may cause problems that we have seen before. I forget who posted the "correct" version of the app info file but I have a copy of it.
HokieForever
____________
 |
|
|
|
|
|
Horaaaaay!
:D
Hugs,
____________
 |
|
|
|
|
|
Thanks :)
Now also available at http://www.esea.dk/esea/optimized.asp
____________
 |
|
|
|
|
|
thank you TMR and Mr X!
happy crunching! (and faster)
Micah
____________
 S@h Berkeley's Staff Friends Club |
|
|
|
|
|
You are welcome all :) and thanks Mr.X!
These days I've been wondering how to get more credits granted, and found a long "work_buf_min_days" in global_prefs.xml seems to work. The longest valid value sounds 10 days (as far as I read in this message board. I may be wrong.) Optimized clients calculates more with fewer instructions/clocks than the official ones. Benchmark tests in the core client isn't so optimized as seti client. So usually optimized clients claim less credits than the official ones.....so I guess many people are annoying...(and that it affects the other 3 hosts also)
Here is how it works. There are 4 clients crunching one work unit. And the granted credit is determined when 3 of them finish crunching. The 2nd highest (or lowest) claimed credit is granted. And the claimed credit of the 4th client is ignored. So if you are the 4th, you will (almost) always get "usual" or "ordinary" credits....
May the "4th" be with you!!!
sorry, I just wanted to write this ;) It works when there is only one optimized client among the first 3 clients....but to make it certain, you'd better be the 4th.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
Happy crunching!! You can copy or distribute them for free :)
Thanks a lot!
Of course again available on www.marisan.nl/seti/
Greetings,
Speedy67
____________
 |
|
|
|
|
|
Hi All,
I downloaded seti-p3-sse2.zip
I followed the directions
and Restarted BOINC all seems to be running with no problems
E@H is running because of the schedualer.
This was in the messages is this correct?
5/31/2005 12:56:09 PM|SETI@home|Found app_info.xml; using anonymous platform
Should I suspend E@H and Run S@H for a run?
What improvements should I see?
Thanks for all your hardwork!
Tim
____________
Account frozen... |
|
|
|
|
Hi All,
I downloaded seti-p3-sse2.zip
I followed the directions
and Restarted BOINC all seems to be running with no problems
E@H is running because of the schedualer.
This was in the messages is this correct?
5/31/2005 12:56:09 PM|SETI@home|Found app_info.xml; using anonymous platform
It's normal :) See lower part of this page.
Should I suspend E@H and Run S@H for a run?
What improvements should I see?
Thanks for all your hardwork!
Tim
No you don't have to suspend E@H. Just let it go as it likes (core client will run S@H according to the preference values you have set.) Since optimized core client gives more benchmark scores, so your E@H will claim more credits than before. And the optimized Seti@Home will be faster. If you want to let SETI work more, then change the preference values as you like. If you want to test S@H sooner, you may suspend E@H for a while (but usually deadline of E@H is closer than that of S@H) If your E@H returns the result behind its deadline, your result will be rejected as "validation error."
And it's not so hard (compiling procedures have been established on Linux).....the harder part was to write how to build it in other threads :) But it's not so hard.
regards, and happy crunching, everyone..........
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
And it's not so hard (compiling procedures have been established on Linux).....the harder part was to write how to build it in other threads :) But it's not so hard.
regards,[/quote]
Thanks Tetsuji,
I am going to try it on and amd machine next....
____________
Account frozen...
|
|
|
|
|
|
Thanks again to you, Tetsuji and thanks to Mr. X for the licensing!
____________
![](http://boinc.mundayweb.com/one/stats.php?userID=498&trans=off">) |
|
|
|
|
|
Hello Tetsuji. I am using your optimized 4.11 for several weeks now. Didn't have any probs during install etc. This client is enormously faster then the official one. So I want to thank you and the unknown Maecenas for this substantial improvement.
|
|
|
|
|
Hello Tetsuji. I am using your optimized 4.11 for several weeks now. Didn't have any probs during install etc. This client is enormously faster then the official one. So I want to thank you and the unknown Maecenas for this substantial improvement.
YYYYY...You are welcome. Hhhhhappy ttttto hhhearrrr that
....tttthanks from Darth Vader...??? :) (oh my father :) )
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
Is there any thoughts on creating optimized clients (both the CC and SETI) of the latest versions, ie 4.44 CC?? Just asking because the schedular works sweet for me and would like to see an optimised client encompasing that.
I've only just started reading about these so dont know alot about them yet, but I do understand that trying to keep up with each new release would be difficult if not near on impossable hehe.
____________
Cheers
Gav
Nothing is 'fool-proof', someone will always invent a better fool! |
|
|
|
|
|
Tetsuji Maverick Rai,
Thanks again to you and the mysterious Mr. X for keeping us all legal and out of jail! LOL.
I'm sorry that there were some folks who took a lot of the fun out of this by making such a big stink about who was responsible for the legalities of it. Rest assured that the majority of us were just happy to have some fun and see how far we could take this thing in order to optimize our SETI throughput. :)
I want to also thank Fuzzy Logic (or Pascal) who purchased a version of the compiler with his own money so that we could all continue using the optimized clients. I'm not sure if his license is the one that is actually being used right now, but nonetheless he forked out the cash to buy one and deserves some recognition.
Thanks to all of you!
Digger |
|
|
|
|
|
TMR,
Couple of questions:
1.) Is the new seti-sse2.zip meant to replace the old seti-generic-sse2.zip or is this something new entirely?
2.) With regards to the Pentium M processors, I was under the impression that the older ones should use seti-p3-sse2.zip and the newer ones based on the P4 processor should use seti-pm.zip. Please advise.
Thanks much,
Digger |
|
|
|
|
Is there any thoughts on creating optimized clients (both the CC and SETI) of the latest versions, ie 4.44 CC?? Just asking because the schedular works sweet for me and would like to see an optimised client encompasing that.
I've only just started reading about these so dont know alot about them yet, but I do understand that trying to keep up with each new release would be difficult if not near on impossable hehe.
Gav,
You don't have to install (copy in) the BOINC.EXE that comes with the optimized package - it is only the setiathome4.11....exe and the app_info.XML that you need.
BOINCMgr.EXE and BOINC.EXE comprise the scheduling and user interface bit currently at version 4.44 and these you can leave alone.
Quite why BOINC.EXE is included in the package I'm not sure - probably because compiling the seti app requires some by-products of compiling/linking BOINC.EXE.
It is likely that the seti app will continue to run with any future 4.xx releases, but action might be needed if/when Berkeley move to V5
____________
|
|
|
|
|
Thanks everyone who made this possible. I downloaded two of the zips and I found a small problem. You still have the old app_info.xml file that does not tell boinc what to do in case it runs into a 4.09 seti work unit which is to use the 4.11 executable. Not going to be a problem for those of us that have been running optimized for the last month. However, if someone new downloads the files it may cause problems that we have seen before. I forget who posted the "correct" version of the app info file but I have a copy of it.
HokieForever
Like HokieForever says make sure you use AndyK's app info file if you still have 4.09 work units. You can find it here the fourth post down.
2nd: Compiling faster
____________
|
|
|
|
|
|
btw Thanks Tetsuji Maverick Rai
and Mr x
Greetings from Germany NRW
Ulli  S@h Berkeley's Staff Friends Club m7 ©
|
|
|
|
|
|
These clients are also available on my site:
seti-p4-sse3.zip
seti-p4-no-sse3.zip
seti-pm.zip
Pentium 4 M owners should use this one.
seti-p3-sse2.zip
Pentium M owners should use this one.
seti-p3.zip
seti-sse2.zip
seti-generic-sse2.zip
seti-plainvanilla.zip
Updated App_Info.xml
referenceunits.zip
These all include the updated app_info.xml file from AndyK to crunch your remaining 4.09 units.
Please let me know if you have any problems.
Thanks again to those who worked on these to provide them for all of us. :)
Dig |
|
|
|
|
|
Hi,
TMR,
Couple of questions:
1.) Is the new seti-sse2.zip meant to replace the old seti-generic-sse2.zip or is this something new entirely?
They are quite different. New seti-sse2.zip is compiled with /QxW option, while seti-generic-sse2.zip was with /QxN option. According to Intel's manual, the former uses SSE2, while the latter "also performs some new Pentium 4 processor-specific optimizations not enabled with W" in addition to SSE2. The old (now eliminated) seti-generic-sse2.zip was created in my struggle to make a client for Athlon64, but in vain. I hope seti-sse2.zip will work for Athlon64 (but have not confirmed yet), and P4 prefers seti-p4-no-sse3.zip or seti-p4-sse2.zip.
To tell the truth, this /QxW option is hidden in the default VC++ IDE menu and I didn't notice it. When jojo4u made a client for Athlon64, I found in his page he used /QxW instead of /QxN and so this time I added this one for Athlon64.
2.) With regards to the Pentium M processors, I was under the impression that the older ones should use seti-p3-sse2.zip and the newer ones based on the P4 processor should use seti-pm.zip. Please advise.
Thanks much,
Digger
Sorry...I don't know much about Pentium M. All I can say is the compiler has an independent option for Pentium M /QxB (while normal P4 uses /QxN). So I created seti-pm.zip with this option. I don't know the real difference, but Intel manual says /QxB option "Generate instructions and optimize for the Pentium M and compatible processors based on Intel Centrino mobile technology, which includes SSE and SSE2."
regards,
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
Thank you for your response TMR...
I'll simply provide all of the optimized clients on my site and let the user decide which they should ultimately use. As long as they back-up their Boinc directory beforehand, as instructed, there should be no problem.
My understanding is that older Pentium M processors were based on the P3 with SSE2 and should use that client (seti-p3-sse2.zip), whereas the new Pentium M processors based on the P4 should use that one (seti-pm.zip).
As far as the others, they seem pretty self-explanatory.
Thanks again :)
Dig
Once again, all of the optimized clients with the updated AndyK app_info.xml files are available at:
seti-p4-sse3.zip
seti-p4-no-sse3.zip
seti-pm.zip
Pentium 4 M owners should use this one.
seti-p3-sse2.zip
Pentium M owners should use this one.
seti-p3.zip
seti-sse2.zip
seti-generic-sse2.zip
seti-plainvanilla.zip
Updated App_Info.xml
referenceunits.zip
boinc-0531.zip
Digger |
|
|
|
|
As far as the others, they seem pretty self-explanatory.
Thanks again :)
Dig
Dig, I am a dummy I cant figure out wich one to use for this
Unit...Sorry.
Can you help?
AMD Sempron 3100+
MMX, Extended MMX, 3DNow!, Extended 3DNow!, SSE, SSE2
____________
Account frozen... |
|
|
|
|
|
hey Digger, I need help also.
Athlon XP2200+ winxpsp1
and
P4 1.8 winxpsp2 |
|
|
|
|
hey Digger, I need help also.
Athlon XP2200+ winxpsp1
and
P4 1.8 winxpsp2
I have exactly the smae processor. He has seti-p3.zip for P3 (and Athlon XP?)
And thats the one I'm using right now, and its working fine, but It says that the seti version is still 4.09, I dont know if its a bug or he just forgot to update it.
____________
|
|
|
|
|
|
New addition:
boinc-0531.zip
boinc core client based on 05-31 nightly build (ver. 4.44). Optimized with /QaxPBNWK option which means this uses "runtime dispatch" which selects the proper optimized code for each processor on runtime. This can reduce the speed by 2-3% so I don't want to use in seti, but I don't think core won't harm.
But I haven't tested enough..please use this with care and with your own responsibility...
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
And thats the one I'm using right now, and its working fine, but It says that the seti version is still 4.09, I dont know if its a bug or he just forgot to update it.
You will work through your existing 4.09 WUs. When you download new work, they will process with 4.11.
____________
 |
|
|
|
|
|
Hi Tetsuji Maverick Rai and all.
I am interested in the optimized clients as my laptop is quite old and is slow at processing WUs.
Alas I can not get any client to work, the processor is a Pentium III-M with Streaming SIMD extensions (but no SIMD extensions 2).
I have tried
seti-p3-sse2.zip, seti-p3 and seti-pm, all crash I guess because the processor I have does not match up. Any ideas if any optimized client will work on my laptop?
Many thanks
____________
|
|
|
|
|
Hi Tetsuji Maverick Rai and all.
I am interested in the optimized clients as my laptop is quite old and is slow at processing WUs.
Alas I can not get any client to work, the processor is a Pentium III-M with Streaming SIMD extensions (but no SIMD extensions 2).
I have tried
seti-p3-sse2.zip, seti-p3 and seti-pm, all crash I guess because the processor I have does not match up. Any ideas if any optimized client will work on my laptop?
Many thanks
It sounds strange.. I don't know much about Pentium III-M, but if it has SSE, seti-p3.zip should work. seti-pm.zip requires SSE2, so it won't work. If you want to make sure the capability of your cpu, try this utility or any other one to confirm it really supports SSE..I guess it does. Sorry I cannot help more than that....all SSE/2/3 combinations are already out (K,W,N,B,P options and no options).
BTW I even tried 3dnow in fftw, but it failed in inline assembly code, as a matter of course :)
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
Just looking at Neds page (http://mkx.feridot.com/boinc/) and his optimized Athlon-XP ver 4.07.3 results and for an Athlon 2600+ theres a %55 increase in performance, sounds good I havent tried it yet I wanted to confirm something first;
Is there any problem using Neds Athlon 4.07.3 as opposed to the 4.09 or optimized 4.11? I couldnt find an specific Athlon-XP 3200+ client.
____________
Cheers
Gav
Nothing is 'fool-proof', someone will always invent a better fool! |
|
|
|
|
Is there any problem using Neds Athlon 4.07.3 as opposed to the 4.09 or optimized 4.11? I couldnt find an specific Athlon-XP 3200+ client.
Neds clients are for Linux.
____________
|
|
|
|
|
Hi Tetsuji Maverick Rai and all.
I am interested in the optimized clients as my laptop is quite old and is slow at processing WUs.
Alas I can not get any client to work, the processor is a Pentium III-M with Streaming SIMD extensions (but no SIMD extensions 2).
I have tried
seti-p3-sse2.zip, seti-p3 and seti-pm, all crash I guess because the processor I have does not match up. Any ideas if any optimized client will work on my laptop?
Many thanks
It sounds strange.. I don't know much about Pentium III-M, but if it has SSE, seti-p3.zip should work. seti-pm.zip requires SSE2, so it won't work. If you want to make sure the capability of your cpu, try this utility or any other one to confirm it really supports SSE..I guess it does. Sorry I cannot help more than that....all SSE/2/3 combinations are already out (K,W,N,B,P options and no options).
BTW I even tried 3dnow in fftw, but it failed in inline assembly code, as a matter of course :)
Many thanks for the prompt reply. I have tried the Intel Processor Frequency ID utility it reports Pentium III-M, Streaming SIMD extensions: yes (but no SSE2). I have tried seti-p3.zip so I guess I am out of luck :-(, is there anything special for the PIII-M chip?
____________
|
|
|
|
|
|
Doh! Silly me, should have read it a bit more closely... Ok then I guess I have the latest for an Athlon-XP then.... seti-p3.zip ??
____________
Cheers
Gav
Nothing is 'fool-proof', someone will always invent a better fool! |
|
|
|
|
Doh! Silly me, should have read it a bit more closely... Ok then I guess I have the latest for an Athlon-XP then.... seti-p3.zip ??
Yeah, that should be the one. That's what I was running on my AMD64 FX-53 before using Jojos compiled version targeted specifically at the AMD64s.
____________
|
|
|
|
|
|
AMD64.... Yeah Yeah skite why dont ya, lol Its on my wish list, its a long list!
____________
Cheers
Gav
Nothing is 'fool-proof', someone will always invent a better fool! |
|
|
|
|
...
Many thanks for the prompt reply. I have tried the Intel Processor Frequency ID utility it reports Pentium III-M, Streaming SIMD extensions: yes (but no SSE2). I have tried seti-p3.zip so I guess I am out of luck :-(, is there anything special for the PIII-M chip?
I have seti-p3.zip running fine on my notebook with Pentium III-M processor 1 GHz. It's a Tualatin core with 512 kB 2nd level cache. It corresponds to the Pentium III-S processors for servers used 3-4 years ago. It is perfoming extremely well and i got a boost of 20-30% from using Tetsuji's cruncher.
____________
Aloha, Uli
 |
|
|
 Mike Volunteer tester
Send message
Joined: Feb 17 01
Posts: 8826
Credit: 3,263,964
RAC: 2,493
|
|
Hi
Has anyone tried the new client from TMR on Athlon XP it won´t run on my XP 2800+.
I get only connection failed error.
greetz Mike
____________
 
|
|
|
|
|
...
Many thanks for the prompt reply. I have tried the Intel Processor Frequency ID utility it reports Pentium III-M, Streaming SIMD extensions: yes (but no SSE2). I have tried seti-p3.zip so I guess I am out of luck :-(, is there anything special for the PIII-M chip?
I have seti-p3.zip running fine on my notebook with Pentium III-M processor 1 GHz. It's a Tualatin core with 512 kB 2nd level cache. It corresponds to the Pentium III-S processors for servers used 3-4 years ago. It is perfoming extremely well and i got a boost of 20-30% from using Tetsuji's cruncher.
Hi
I have had another go with seti-p3.zip. The problems with this are different, the other optimized clients crashed which is not a surprise as the CPU architecture is wrong. With seti-p3 when I put the client exe and app_info.xml files into the setiathome project folder and start boinc it goes and deletes all WUs and the pdb file (what does this file actually do?). It then justs sit there and does not download any new work etc. It is just idle.
I have a backup of my original boinc folder so I can revert back OK. Any ideas why boinc just sits idle with the optimized client? I must have missed something.
____________
|
|
|
|
|
...
Many thanks for the prompt reply. I have tried the Intel Processor Frequency ID utility it reports Pentium III-M, Streaming SIMD extensions: yes (but no SSE2). I have tried seti-p3.zip so I guess I am out of luck :-(, is there anything special for the PIII-M chip?
I have seti-p3.zip running fine on my notebook with Pentium III-M processor 1 GHz. It's a Tualatin core with 512 kB 2nd level cache. It corresponds to the Pentium III-S processors for servers used 3-4 years ago. It is perfoming extremely well and i got a boost of 20-30% from using Tetsuji's cruncher.
Hi
I have had another go with seti-p3.zip. The problems with this are different, the other optimized clients crashed which is not a surprise as the CPU architecture is wrong. With seti-p3 when I put the client exe and app_info.xml files into the setiathome project folder and start boinc it goes and deletes all WUs and the pdb file (what does this file actually do?). It then justs sit there and does not download any new work etc. It is just idle.
I have a backup of my original boinc folder so I can revert back OK. Any ideas why boinc just sits idle with the optimized client? I must have missed something.
Did you put updated App_Info.xml in your projects\\seti folder?
Happy crunching,
____________
|
|
|
|
|
Hi
Has anyone tried the new client from TMR on Athlon XP it won´t run on my XP 2800+.
I get only connection failed error.
greetz Mike
Mike, Have you copied in both the seti application and BOINC.EXE?
I presume that you are referring to the Connection Failed message that pops-up when you run BOINCMgr?
If so, connection fail is usually related to the communication between the BOINC Daemon (BOINC.EXE) and the BOINC Manager (BOINCmgr.EXE). In turn, this is often due to a software firewall issue. If you have a s/w firewall, and copied in BOINC.EXE it may not be allowing BOINC.EXE server rights due to the version difference.
____________
|
|
|
|
|
I have had another go with seti-p3.zip. The problems with this are different, the other optimized clients crashed which is not a surprise as the CPU architecture is wrong. With seti-p3 when I put the client exe and app_info.xml files into the setiathome project folder and start boinc it goes and deletes all WUs and the pdb file (what does this file actually do?). It then justs sit there and does not download any new work etc. It is just idle.
I have a backup of my original boinc folder so I can revert back OK. Any ideas why boinc just sits idle with the optimized client? I must have missed something.
When the version number changes, it may delete WU's and other files in the project directory. If you see the message with boincmgr, I guess it says "connection deffering xx minutes xx seconds" or something. Now your WU cache is empty. That's why it sits quiet. Wait until boinc.exe coonects to the server and see what happens next. It will download new WU's and begin crunching.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
...
Many thanks for the prompt reply. I have tried the Intel Processor Frequency ID utility it reports Pentium III-M, Streaming SIMD extensions: yes (but no SSE2). I have tried seti-p3.zip so I guess I am out of luck :-(, is there anything special for the PIII-M chip?
I have seti-p3.zip running fine on my notebook with Pentium III-M processor 1 GHz. It's a Tualatin core with 512 kB 2nd level cache. It corresponds to the Pentium III-S processors for servers used 3-4 years ago. It is perfoming extremely well and i got a boost of 20-30% from using Tetsuji's cruncher.
Hi
I have had another go with seti-p3.zip. The problems with this are different, the other optimized clients crashed which is not a surprise as the CPU architecture is wrong. With seti-p3 when I put the client exe and app_info.xml files into the setiathome project folder and start boinc it goes and deletes all WUs and the pdb file (what does this file actually do?). It then justs sit there and does not download any new work etc. It is just idle.
I have a backup of my original boinc folder so I can revert back OK. Any ideas why boinc just sits idle with the optimized client? I must have missed something.
Did you put updated App_Info.xml in your projects\seti folder?
Happy crunching,
Ha ha, that is the issue. The updated app_info.xml you point out has a reference to 4.09 but the old one in the zip does not which is why I guess Boinc went and deleted the old WUs. No big issue as I waited until a WU had finished and I am sure seti@home will resend the other WUs if they are needed. After 10 minutes Boinc did download some more work and started processing so it will be interesting to see what speed improvement I get on my old laptop.
But it is worth others using the updated app_info.xml with the optimized clients.
____________
|
|
|
|
|
I have had another go with seti-p3.zip. The problems with this are different, the other optimized clients crashed which is not a surprise as the CPU architecture is wrong. With seti-p3 when I put the client exe and app_info.xml files into the setiathome project folder and start boinc it goes and deletes all WUs and the pdb file (what does this file actually do?). It then justs sit there and does not download any new work etc. It is just idle.
I have a backup of my original boinc folder so I can revert back OK. Any ideas why boinc just sits idle with the optimized client? I must have missed something.
When the version number changes, it may delete WU's and other files in the project directory. If you see the message with boincmgr, I guess it says "connection deffering xx minutes xx seconds" or something. Now your WU cache is empty. That's why it sits quiet. Wait until boinc.exe coonects to the server and see what happens next. It will download new WU's and begin crunching.
Hi Tetsuji Maverick Rai
Our posts crossed, appologies I have got the p3 optimized client running now. Thanks for pointing me at the correct client to use.
____________
|
|
|
MikeVolunteer tester
Send message
Joined: Feb 17 01
Posts: 8826
Credit: 3,263,964
RAC: 2,493
|
Hi
Has anyone tried the new client from TMR on Athlon XP it won´t run on my XP 2800+.
I get only connection failed error.
greetz Mike
Mike, Have you copied in both the seti application and BOINC.EXE?
I presume that you are referring to the Connection Failed message that pops-up when you run BOINCMgr?
If so, connection fail is usually related to the communication between the BOINC Daemon (BOINC.EXE) and the BOINC Manager (BOINCmgr.EXE). In turn, this is often due to a software firewall issue. If you have a s/w firewall, and copied in BOINC.EXE it may not be allowing BOINC.EXE server rights due to the version difference.
Hi
I did it exactly so when i changed to tetsuis first p3 client with 4.27 and had no probs with my firewall.
I´ve noticed with the original 4.44 version that there are problems with my firewall.
Getting no shedulers responded and was not able to up and download til i deactivated the firewall.
Has the ports changed or what should i do???
Thanks i advance
Mike
____________
|
|
|
|
|
|
Do you have a compiled version of the 4.43 Boinc?
____________
Red Bull Air Racing
Gas price by zip at Seti
 |
|
|
|
|
Hi
Has anyone tried the new client from TMR on Athlon XP it won´t run on my XP 2800+.
I get only connection failed error.
greetz Mike
Hi Mike
On our XP Maschines the same Problem ---> no connection
____________
<img src="http://www.boincsynergy.com/images/stats/comb-1619.jpg"> |
|
|
|
|
Hi
Has anyone tried the new client from TMR on Athlon XP it won´t run on my XP 2800+.
I get only connection failed error.
greetz Mike
Hi Mike
On our XP Maschines the same Problem ---> no connection
My best guess is that you have XP with SP2 installed. If so check the exceptions settings in the firewall to allow the optimised client access to the Internet.
See this web page for details on how to allow a specific program access.
http://www.microsoft.com/windowsxp/using/security/internet/sp2_wfexceptions.mspx
Follow the instructions under the heading: "If the program (or service) that you want to allow is not listed"
____________
|
|
|
|
|
My best guess is that you have XP with SP2 installed. If so check the exceptions settings in the firewall to allow the optimised client access to the Internet.
See this web page for details on how to allow a specific program access.
http://www.microsoft.com/windowsxp/using/security/internet/sp2_wfexceptions.mspx
Follow the instructions under the heading: "If the program (or service) that you want to allow is not listed"
the internet connection is o.k. , the boincmanager announces a connecting error with the core client
____________
<img src="http://www.boincsynergy.com/images/stats/comb-1619.jpg"> |
|
|
MikeVolunteer tester
Send message
Joined: Feb 17 01
Posts: 8826
Credit: 3,263,964
RAC: 2,493
|
My best guess is that you have XP with SP2 installed. If so check the exceptions settings in the firewall to allow the optimised client access to the Internet.
See this web page for details on how to allow a specific program access.
http://www.microsoft.com/windowsxp/using/security/internet/sp2_wfexceptions.mspx
Follow the instructions under the heading: "If the program (or service) that you want to allow is not listed"
the internet connection is o.k. , the boincmanager announces a connecting error with the core client
Hi
Servus Heiko
Same problem here, no probs so far with original 4.44.
greetz Mike
____________
|
|
|
|
|
Do you have a compiled version of the 4.43 Boinc?
Look at Tetsujis "new addition" post in this thread.
____________
|
|
|
MikeVolunteer tester
Send message
Joined: Feb 17 01
Posts: 8826
Credit: 3,263,964
RAC: 2,493
|
Do you have a compiled version of the 4.43 Boinc?
Look at Tetsujis "new addition" post in this thread.
Hi
This is the one with the connection error.
greetz Mike
____________
|
|
|
|
|
Do you have a compiled version of the 4.43 Boinc?
Look at Tetsujis "new addition" post in this thread.
Hi
This is the one with the connection error.
I never had the connection error, but isn't it fixed in 4.44? That is what Tetsujis compiled version is from.
____________
|
|
|
|
|
|
Thanks Maverick :)
Files mirrored at my site too (including updated app_info.xml file):
http://www.pperry.f2s.com/downloads.htm
Can anyone confirm if the seti-sse2.zip file works on an AMD64 yet?
Ned
____________
*** My Guide to Compiling Optimised BOINC and SETI Clients ***
*** Download Optimised BOINC and SETI Clients for Linux Here *** |
|
|
|
|
Thanks Maverick :)
Files mirrored at my site too (including updated app_info.xml file):
http://www.pperry.f2s.com/downloads.htm
Can anyone confirm if the seti-sse2.zip file works on an AMD64 yet?
Ned
i will test this tomorrow, on my Athlon64 it is to late for this at this Day, 22:45 here in Germany..
Greetings from Germany NRW
Ulli S@h Berkeley's Staff Friends Club m7 ©
|
|
|
|
|
Can anyone confirm if the seti-sse2.zip file works on an AMD64 yet?
I haven't done a full run and don't know if the results will verify, but I did do a quick test just to see if it was compatible. It does start running in standalone.
I'm in the process of some WUs using Jojo's AMD64 version. I'll finish them and give this version a test.
____________
|
|
|
|
|
hey Digger, I need help also.
Athlon XP2200+ winxpsp1
and
P4 1.8 winxpsp2
Tony, I'm sorry I didn't see your post until just now. I hope someone much more experienced than me was able to help you. I've been following all of the various threads regarding the optimized clients, but have to admit I'm a dummy when it comes to this stuff.
Dig |
|
|
|
|
Can anyone confirm if the seti-sse2.zip file works on an AMD64 yet?
I haven't done a full run and don't know if the results will verify, but I did do a quick test just to see if it was compatible. It does start running in standalone.
I'm in the process of some WUs using Jojo's AMD64 version. I'll finish them and give this version a test.
It would be nice if you could tell us which is faster on the benchmark WU if you're running it - JoJo's vs Maverick's :)
Ned
____________
*** My Guide to Compiling Optimised BOINC and SETI Clients ***
*** Download Optimised BOINC and SETI Clients for Linux Here *** |
|
|
|
|
It would be nice if you could tell us which is faster on the benchmark WU if you're running it - JoJo's vs Maverick's :)
Good call. Running the reference unit right now with Jojo's build. It'll be later this evening until I have the results. I'll have to leave while Tetsuji's is running. I'll post the results as soon as possible. |
|
|
|
|
Thanks Maverick :)
Can anyone confirm if the seti-sse2.zip file works on an AMD64 yet?
Ned
Working fine so far on my two AMD 64 3200's. 25-30 predictor work units per machine without error and most have validated, the rest are pending. Haven't gotten any SETI units back yet, (low resource share) but should have something in the next hour or two. It does run. So far no issues.
Benchmarks are strange though. Integer (drystone) went down but the floating point (whetsone) went up.
OLD:
Measured floating point speed 1811.32 million ops/sec
Measured integer speed 4496.24 million ops/sec
NEW
Measured floating point speed 2344.51 million ops/sec
Measured integer speed 3637.37 million ops/sec
And thanks from me as well Maverick. Your work and effort is greatly appreciated.
____________
Team MacNN - The best Macintosh team ever. |
|
|
|
|
Do you have a compiled version of the 4.43 Boinc?
Look at Tetsujis "new addition" post in this thread.
That is the development version 4.44 not the released version 4.43.
I would like the released version.
____________
Red Bull Air Racing
Gas price by zip at Seti
|
|
|
|
|
|
Got my first two SETI work units running SETI-SSE2.zip on Athlon 64 3200+ (two different machines). No problems encountered.
http://setiweb.ssl.berkeley.edu/workunit.php?wuid=16681378
1 hour 35.2 minutes. Claimed 19.6 credits. still pending
http://setiweb.ssl.berkeley.edu/workunit.php?wuid=16681343
1 hour 29.2 minutes. Claimed 18.53 credits. Validated and granted 21.66 credits.
No problems with either work unit. Everything worked fine.
____________
Team MacNN - The best Macintosh team ever. |
|
|
|
|
|
The "new addition" - the boinc.exe compiled based on 4.44 it is crashing when I try to restart the boinc service. I have AthlonXP and a PIII processors.
I will try it on a P4 (no sse3) processor too... tommorow.
Message is "boinc daemon couldn't start". I am running the official 4.44 package with the optimised clients.
PS: The new boinc.exe is way bigger than original, but I think that's ok.
PPS: The biggest gain was for my P4@1.3GHz - the cruncing time dropped less than half - now it is competing with any AMD Athlon 3000+...
____________
|
|
|
|
|
It would be nice if you could tell us which is faster on the benchmark WU if you're running it - JoJo's vs Maverick's :)
Looks like Tetsuji edges out JoJo by a little bit on the reference WU. :)
setiathome_4.11_windows_intelx86.exe (Tetsuji) = wu_cpu_time 5527.390625
setiathome_4.11_windows_jo-athlon64-sse2_1105-1.exe (JoJo) = wu_cpu_time 5586.562500
I'm also guessing that Tetsuji's compilation will verify, because when comparing the state.sah files, the few numbers that were different were the same to several decimal places.
Now to run out the WUs I have with JoJo's build and install Tetsuji's. Every second counts right? ;)
|
|
|
|
|
|
I just tried running Tetsuji's optimzed CC based on ver 4.44 and it wouldnt even start, saying the service couldnt start... Is that because I'm trying to run it as a service? The optimzed app runs fine, although I've only done a couple of WU's so far. I've noticed I'm claiming less credit tho, will keep an eye on that.
____________
Cheers
Gav
Nothing is 'fool-proof', someone will always invent a better fool! |
|
|
|
|
|
As for my "new addition" boinc core client,
It works fine on my P4, and the binary contains several types of codes for P4, Pen-M, P3, Athlon64, P2 using runtime dispatcher. That's why it's slightly larger. I suspect this runtime dispatcher doesn't work on some types of processors. So I'm building type-specific 4.43 cores and will post them in another thread soon (this thread has become too long.)
Sorry for the trouble....:(
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
Ack!!
Not sure why the seti-sse2.zip would process the reference WU fine and not actual WUs, but when I try to do normal WUs using this .exe, I get this:
Incorrect function. (0x1) - exit code 1 (0x1) :(
The WUs don't even start...
Going back to JoJo's build.
|
|
|
|
|
|
This should be stickied.
____________
 |
|
|
|
|
|
Stickied or not, there is soooo much 3rd party development on optimized clients/apps that the SETI development team should make a stand either one way or another.... Offically support optimized client/apps and develope/release them on their site... or prohibit there use.
____________
Cheers
Gav
Nothing is 'fool-proof', someone will always invent a better fool! |
|
|
|
|
|
Core clients are here. Somehow there are not so much differences between these clients.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
|
Why is my seti application still 4.09 and not .11 ?? I laready finished one and downloaded a new one...
____________
|
|
|
|
|
|
On my seti folder (in the "projects" folder) I found 3 different versions; 4.10, 4.09, and the 4.11 which isd the one I'm trying to use.
____________
|
|
|
|
|
Why is my seti application still 4.09 and not .11 ?? I laready finished one and downloaded a new one...
Did you put the app_info.xml file in your S@H projects folder? |
|
|
|
|
On my seti folder (in the "projects" folder) I found 3 different versions; 4.10, 4.09, and the 4.11 which isd the one I'm trying to use.
I think that is the way my folder was when I started. Remember when we switched to 4.10 for a while but had trouble with it and switched back? Well anyways, I sort of remember that those extra files (4.09 4.10 and banner)disappeared on their own. This time I removed them. If you have AndyK's app info file in there the new 4.11 client will run the old 4.09 work units and continue running when it gets to the new 4.11's it is making.
____________
|
|
|
|
|
|
Yya! I finally got it working, I just added the app_info.xml like Heffed told me. Though I lost about 5 WU that I had already finished last night... but oh well.
____________
|
|
|
|
|
|
I updated the page at www.marisan.nl/seti,one of the links at which Tetsuji's optimized clients are mirrored.
Hope this little bit more info helps everyone choosing the right client.
Thanks to anyone who's helped with this info (i have cut and paste from the threads in this message board) :-)
I'll try keeping this link up-to-date as new clients appear.
Greetings,
Speedy67
____________
|
|
|
|
|
4.43 (May 23 build)
There are 3 versions
boinc-443-mmx.zip
boinc-443-sse.zip
boinc-443-sse2.zip
On my P4 box, they don't make so much differences in benchmark tests...but better than the official one.
Thank you! Now all the boinc.exe are working on AthlonXP, PIII (SSE one) and P4 with no SSE3 (SSE2 one). Great job!
____________
|
|
|
|
|
|
Just installed the AMD 64 Client, going to let it run for a couple of days and see how it goes .. Guess i am the "tester" for this one :)
I updated the page at www.marisan.nl/seti,one of the links at which Tetsuji's optimized clients are mirrored.
Hope this little bit more info helps everyone choosing the right client.
Thanks to anyone who's helped with this info (i have cut and paste from the threads in this message board) :-)
I'll try keeping this link up-to-date as new clients appear.
Greetings,
Speedy67
____________
Wiki de BOINC
<img src="http://www.boincstats.com/stats/banner.php?cpid=51a3eaaef8df276544f56140a8a65413"> |
|
|
truxVolunteer tester
Send message
Joined: Feb 6 01
Posts: 344
Credit: 1,127,051
RAC: 0
|
I updated the page at www.marisan.nl/seti,one of the links at which Tetsuji's optimized clients are mirrored.
Hope this little bit more info helps everyone choosing the right client.
Thanks to anyone who's helped with this info (i have cut and paste from the threads in this message board) :-)
Some of the CPU assignments are not quite exact on your website. So for example Pentium 4 M (Dothan) runs much better with "seti-p4-no-sse3" than with pm or p3-sse2. I am not sure how it is with other cores, but most likely every P4 M supporting SSE2 will do the same. Download the freeware CPU-Z to see what exactly your processor suports.
As for AMD's - I have tested Athlon MP (Palomino) and Sempron - currently both run best with p3.
trux
|
|
|
|
|
Download the freeware CPU-Z to see what exactly your processor suports.
Or you can get Intel's own Processor Identification Utilities: here (Intel only of course)
____________
|
|
|
|
|
I updated the page at www.marisan.nl/seti,one of the links at which Tetsuji's optimized clients are mirrored.
Hope this little bit more info helps everyone choosing the right client.
Thanks to anyone who's helped with this info (i have cut and paste from the threads in this message board) :-)
Some of the CPU assignments are not quite exact on your website. So for example Pentium 4 M (Dothan) runs much better with "seti-p4-no-sse3" than with pm or p3-sse2. I am not sure how it is with other cores, but most likely every P4 M supporting SSE2 will do the same. Download the freeware CPU-Z to see what exactly your processor suports.
As for AMD's - I have tested Athlon MP (Palomino) and Sempron - currently both run best with p3.
trux
Yes, seti-pm is for Pentium M, as in Centrino, NOT Pentium 4 Mobile's. Pentium 4 Mobiles should use either of the P4 clients depending on whether or not they support SSE3.
Ned
____________
*** My Guide to Compiling Optimised BOINC and SETI Clients ***
*** Download Optimised BOINC and SETI Clients for Linux Here *** |
|
|
|
|
Just installed the AMD 64 Client, going to let it run for a couple of days and see how it goes .. Guess i am the "tester" for this one :)
If you are referring to the seti-sse2.zip version, you could look at my posts regarding it in this thread. ;)
|
|
|
|
|
Some of the CPU assignments are not quite exact on your website. So for example Pentium 4 M (Dothan) runs much better with "seti-p4-no-sse3" than with pm or p3-sse2. I am not sure how it is with other cores, but most likely every P4 M supporting SSE2 will do the same. Download the freeware CPU-Z to see what exactly your processor suports.
As for AMD's - I have tested Athlon MP (Palomino) and Sempron - currently both run best with p3.
trux
Yes, seti-pm is for Pentium M, as in Centrino, NOT Pentium 4 Mobile's. Pentium 4 Mobiles should use either of the P4 clients depending on whether or not they support SSE3.
Ned
Thanks Trux and Ned, I updated the file descriptions accordingly.
Of course when anything else is not quite right on my site, I invite everyone to please report it, to make sure everyone can easily pick the right client for his/her system.
Greetings,
Speedy67
www.marisan.nl/seti/
____________
|
|
|
|
|
Some of the CPU assignments are not quite exact on your website. So for example Pentium 4 M (Dothan) runs much better with "seti-p4-no-sse3" than with pm or p3-sse2. I am not sure how it is with other cores, but most likely every P4 M supporting SSE2 will do the same. Download the freeware CPU-Z to see what exactly your processor suports.
As for AMD's - I have tested Athlon MP (Palomino) and Sempron - currently both run best with p3.
trux
Yes, seti-pm is for Pentium M, as in Centrino, NOT Pentium 4 Mobile's. Pentium 4 Mobiles should use either of the P4 clients depending on whether or not they support SSE3.
Ned
Thanks Trux and Ned, I updated the file descriptions accordingly.
Of course when anything else is not quite right on my site, I invite everyone to please report it, to make sure everyone can easily pick the right client for his/her system.
Greetings,
Speedy67
www.marisan.nl/seti/
Nice job Speedy :)
The seti-P3 version works on some Duron's too - those (later ones) based on XP cores I guess.
____________
*** My Guide to Compiling Optimised BOINC and SETI Clients ***
*** Download Optimised BOINC and SETI Clients for Linux Here *** |
|
|
|
|
Nice job Speedy :)
Thanks. :)
The seti-P3 version works on some Duron's too - those (later ones) based on XP cores I guess.
Thanks for the info. Website is updated. Additional info is always welcome.
Greetings,
Speedy67
www.marisan.nl/seti/
____________
|
|
|
|
|
|
I'm really impressed with the difference in processing time between these optimized clients and the standard one.
My sample size is small for the optimized client (2 days), but I checked the other crunching hosts to make sure I didn't have a "short" unit. My numbers are eyeballed from a list, but should be close enough for my point. I've received credit for each host so the optimized clients seem to be working.
Performance (1 SETI WU):
866MHz P3
old: about 6.2 hours new: about 5.5 hours
1.3GHz P3 Celeron
old: about 6.5 hours new: about 5.5 hours
3.4GHz P4 HT
old: about 3.0 hours new: about 1.25 hours
One of my hosts scares me tho :-) On the two slower computers the gain is about 1.2x. The 3.4GHz computer shows a 2.4x gain. WOW! It has done 11 WU consistently at this rate. I don't think it's an HT thing since my numbers are per WU from BOINCView.
I'm just providing my observations on my early experience with the optimized client. However, I am curious if the P4 HT's gain is typical. Could this difference be because of SSE3? |
|
|
|
|
I'm just providing my observations on my early experience with the optimized client. However, I am curious if the P4 HT's gain is typical. Could this be a difference because of SSE3?
I've got a Northwood P4 with Hyper-threading and it went from about
3 Hrs. 20 Min. down to 1 Hr. 40 Min. No SSE3 though. The difference is Tetsuji's great work. Kind of a gas isn't it.
____________
|
|
|
|
|
866MHz P3
old: about 6.2 hours new: about 5.5 hours
My 450 MHz P3 went from 12h 45m to 11h 25m, a 1.15x gain
1.3GHz P3 Celeron
old: about 6.5 hours new: about 5.5 hours
3.4GHz P4 HT
old: about 3.0 hours new: about 1.25 hours
My 3.0 GHz P4 HT (Prescott, SSE3) went from 3h 30m to 1h 50m, a 1.9x gain
The 3.4GHz computer shows a 2.4x gain. WOW! It has done 11 WU consistently at this rate. I don't think it's an HT thing since my numbers are per WU from BOINCView.
I'm just providing my observations on my early experience with the optimized client. However, I am curious if the P4 HT's gain is typical. Could this difference be because of SSE3?
I think your experiences are quite right.. lots of gain with the new optimized clients, especially on the faster machines..
Greetings,
Speedy67
www.marisan.nl/seti/
____________
|
|
|
|
|
The difference is Tetsuji's great work. Kind of a gas isn't it.
It sure is! Thanks to Tetsuji for the improvement. |
|
|
|
|
|
Time for an Update
Installed the AMD64 sse2 client last night here are the results so far
OLD - 8000 - 9000 sec (2 - 2.5 hrs) per WU
NEW - 4000 - 5000 sec (1.5 hrs) per WU
Claimed Credit is about 30% lower with the optimized client, so far it seems that these results are the "low" ones that are eliminated, in the credit recived area it seems to be working well.
My unit has been completely stable no issues at all, I was returning about 10 results a day but with the new client it looks like it will be around 15 results a day... Now if i could just figure out more ways to tweek it get evan more results :).
Just installed the AMD 64 Client, going to let it run for a couple of days and see how it goes .. Guess i am the "tester" for this one :)
I updated the page at www.marisan.nl/seti,one of the links at which Tetsuji's optimized clients are mirrored.
Hope this little bit more info helps everyone choosing the right client.
Thanks to anyone who's helped with this info (i have cut and paste from the threads in this message board) :-)
____________
Wiki de BOINC
<img src="http://www.boincstats.com/stats/banner.php?cpid=51a3eaaef8df276544f56140a8a65413"> |
|
|
|
|
|
I'm using the optimized client, but what happens when SETI sends out a successor to 4.09? Does BOINC download and install it? Will I have to do anything? Will WUs crunched with 4.11 still be valid?
____________
Help and BOINC documentation is available here.
 |
|
|
|
|
Nice job Speedy :)
Thanks. :)
The seti-P3 version works on some Duron's too - those (later ones) based on XP cores I guess.
Thanks for the info. Website is updated. Additional info is always welcome.
Greetings,
Speedy67
www.marisan.nl/seti/
For AMD CPUs, the seti-P3 client should run on any AMD CPU that is i686 or Pentium Pro compatible AND has MMX and full SSE compatability too (so Athlon XP, Athlon MP, Athlon4, and later Duron's). It _probably_ wont work on older (original) Athlon, Athlon thunderbird and older comparable Duron's as these only have SSE prefetch instructions support, not full SSE support, nor K6-2/3 as these only have MMX and no SSE at all. These older CPUs should presumably use the plain vanilla i686 client, although I don't have access to any of these platforms to test.
Personally, I also relabelled the seti-sse2 client to AMD64 on my site as presumably this is the only class of CPU that should be using this client (hopefully makes things clearer).
Ned
____________
*** My Guide to Compiling Optimised BOINC and SETI Clients ***
*** Download Optimised BOINC and SETI Clients for Linux Here *** |
|
|
|
|
K6-2/3 as these only have MMX and no SSE at all. These older CPUs should presumably use the plain vanilla i686 client, although I don't have access to any of these platforms to test.
i have confirmed metod's i686 client works on both of my k6-2 systems (300@ 300, 400@504), although with no major improvement over my own lightly optimized non-fftw3 compiles. the generic i686 client i got from your site didn't work on either. |
|
|
|
|
For AMD CPUs, the seti-P3 client should run on any AMD CPU that is i686 or Pentium Pro compatible AND has MMX and full SSE compatability too (so Athlon XP, Athlon MP, Athlon4, and later Duron's). It _probably_ wont work on older (original) Athlon, Athlon thunderbird and older comparable Duron's as these only have SSE prefetch instructions support, not full SSE support, nor K6-2/3 as these only have MMX and no SSE at all. These older CPUs should presumably use the plain vanilla i686 client, although I don't have access to any of these platforms to test.
Personally, I also relabelled the seti-sse2 client to AMD64 on my site as presumably this is the only class of CPU that should be using this client (hopefully makes things clearer).
Thanks for the info.. I updated the site. Also added a link to the CPU-Z utility as this is probably the easiest way to figure out what client one needs to download.
Greetings,
Speedy67
____________
|
|
|
|
|
|
I've returned the first WU crunched by this optimized SETI client, and the processing time is greatly reduced, but so is the claimed credit. As long as I am the only member of the quorum processing the WU with the optimized client it looks like I will benefit (as my low number will be discarded in the averaging process); but if all members of the quorum were using this client, then we would all claim and be granted a smaller credit.
Of course, this would be balanced by the fact that we would all process more WUs per unit of time.
So it looks like, for now, I have an "advantage".
Is my reasoning correct here?
____________
|
|
|
|
|
I've returned the first WU crunched by this optimized SETI client, and the processing time is greatly reduced, but so is the claimed credit. As long as I am the only member of the quorum processing the WU with the optimized client it looks like I will benefit (as my low number will be discarded in the averaging process); but if all members of the quorum were using this client, then we would all claim and be granted a smaller credit.
Of course, this would be balanced by the fact that we would all process more WUs per unit of time.
So it looks like, for now, I have an "advantage".
Is my reasoning correct here?
I think you're quite right. Using an optimized client you will probably claim the least credit of the results returned, and as credit is granted with the median value of the results returned, you will benefit.
As long as you don't use an optimized boinc-client, you will claim just as much credit for every second your computer has been crunching as you would have when not using an optimized seti-client. So I guess people can't accuse you of 'cheating'. Seen a lot of discussion about that lately, unfortunately.
Are optimised clients affecting everyone?
Credit for Optimised Clients
Greetings,
Speedy67
____________
|
|
|
|
|
|
Which optimization would work best for a AMD Turion 64 notebook?
Before I start playing with them ;)
____________
 |
|
|
|
|
Which optimization would work best for a AMD Turion 64 notebook?
Before I start playing with them ;)
My guess would be the seti-sse2.zip for Athlon64 (SSE2) as to my knowledge there is no AMD SSE3 optimized client available (yet).
Greetings,
Speedy67
____________
|
|
|
BenherVolunteer developer
Volunteer tester
Send message
Joined: Jul 25 99
Posts: 517
Credit: 465,152
RAC: 0
|
|
Lets (somebody who knows which is which) make a table
1. List all the CPUs and Cores
2. Correspond each to the best optimized client.
A nice site for listing of cores.
[b]Intel[/b]
Pentium
Pentium Pro
Pentium MMX
Pentium II
Celeron II
Pentium III
Celeron III
Xeon III
Pentium III-M (=Mobile)
Pentium IV (Willamette, Foster, Northwood, Northwood HT, Prescott...)
XEON IV (Prestonia, Gallatin...)
Pentium M "Centrino" (Banias, Dothan...)
[b]AMD[/b]
K6
K6-2
K6-III
Athlon (Thunderbird, Spitfire...)
Athlon XP (Palomino, Thourougbred, Morgan, Barton...)
Athlon 64 (Newcastle, Sledgehammer...)
|
|
|
|
|
|
First I want to say thank you to all who's contributed to this project! Many fine download sites now exists for the different clients. Congratulations also to Speedy for his new site. It's great! :)
I recently switched from 4.19 to 4.44 and was wondering if anyone has managed to get seti-p3 and 4.44 work together? Have tried both with and without Tetsui's boinc.exe 4.44 in boinc-0531.zip.
Thanks in advance.
|
|
|
|
|
I recently switched from 4.19 to 4.44 and was wondering if anyone has managed to get seti-p3 and 4.44 work together? Have tried both with and without Tetsui's boinc.exe 4.44 in boinc-0531.zip.
Thanks in advance.
I'm using the original seti-p3 optimized client (shows as 4.11) with boinc 4.44, on an Athlon XP 2600+ under W2K Pro. Works fine.
____________
 |
|
|
|
|
|
Some tests with optimized code.
It sure is a difference in some cases; Dhrystone down and Whetstone up. Is this normal? I have to ask since I don't know what should be expected in this case. I've been running around between different threads but haven't been able to figure it out.
Original figures from 4.19
--- - 2005-06-07 08:44:01 - 916 double precision MIPS (Whetstone) per CPU
--- - 2005-06-07 08:44:01 - 2827 integer MIPS (Dhrystone) per CPU
with optimized
2005-06-07 12:39:22 [---] 1218 double precision MIPS (Whetstone) per CPU
2005-06-07 12:39:22 [---] 2144 integer MIPS (Dhrystone) per CPU
with optimized Win client
--- - 2005-06-07 12:52:08 - 916 double precision MIPS (Whetstone) per CPU
--- - 2005-06-07 12:52:08 - 2793 integer MIPS (Dhrystone) per CPU
I found a link to following site Optimized Seti Clients to be a very informative site. I really recommend it for further reading - especially a tool called CPU-Z that tells you about the processor you're using in your box - then it's easy to find out what optimized package you should use.
Last but not least: You guys working on the optimization, putting new sites on Internet for the same purpose... you're doing a great contribution to this community! Thanks!!!
____________
Happy crunching |
|
|
|
|
|
Great work, and great support with the license
So far I have today put the plain P3 on my intel PIII 801 mhz, and the plainvanilla on my good old and trusted amd tbird 1.2 ghz.
Bot machines seems to be running very very fast. No result produced from here yet though.
Now only my Prescott needs to be optimized. Will await the return of the first optimized result and then do this last 1.
Cheers
____________
Best regards / B52 |
|
|
|
|
|
What optimized client should i download and use for my AMD64 3000+? I've been reading a bunch of threads and stuff and a lot of it is over my head, plus there is so much of it now i dont know where to begin looking anymore. Should i get seti-sse2 or seti-p3? I'm assuming i should download the optimized boinc client as well? Any help would be greatly appreciated.
Email = joeknowsall@gmail.com
AIM = JoeNM06
____________
Join the team, SETI.USA We are growing and could use your help to overcome SETI.Germany...www.setiusa.net
 |
|
|
|
|
What optimized client should i download and use for my AMD64 3000+? I've been reading a bunch of threads and stuff and a lot of it is over my head, plus there is so much of it now i dont know where to begin looking anymore. Should i get seti-sse2 or seti-p3? I'm assuming i should download the optimized boinc client as well? Any help would be greatly appreciated.
Email = joeknowsall@gmail.com
AIM = JoeNM06
Try to use the link provided by Peder just 2 posts above. Use the cpu utility as descriped. It will inform you about your cpu's capabilities.
____________
Best regards / B52 |
|
|
|
|
|
I showed a little improvement with my setup. I'm going to run this for a while before I put it on my other systems. I'm running 4.44.
System
Intel Pentium 4, 2600 MHz (13 x 200)
Northwood HyperThreading
MSI 865GM2-S (MS-6743)
1536 MB PC3200 DDR SDRAM
Old
[---] Running CPU benchmarks
[---] Benchmark results:
[---] Number of CPUs: 2
[---] 1128 double precision MIPS (Whetstone) per CPU
[---] 947 integer MIPS (Dhrystone) per CPU
New
||Running CPU benchmarks
||Benchmark results:
|| Number of CPUs: 2
|| 1370 double precision MIPS (Whetstone) per CPU
|| 1194 integer MIPS (Dhrystone) per CPU
One thing I noticed was my old WU waiting for crunching may have been lost. Are these lost or will they be picked back up?
Here are just a few examples...
6/7/2005 12:16:07 AM||Starting BOINC client version 4.27 for windows_intelx86
6/7/2005 12:16:07 AM|SETI@home|Host location: home
6/7/2005 12:16:07 AM|SETI@home|Using your default project prefs
6/7/2005 12:16:07 AM|SETI@home|Found app_info.xml; using anonymous platform
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
6/7/2005 12:16:07 AM|SETI@home|Can't link workunit in state file
6/7/2005 12:16:07 AM|SETI@home|WU refers to nonexistent app_version: setiathome 409
>Fred
____________
http://www.teamstarfire.org/
 |
|
|
|
|
One thing I noticed was my old WU waiting for crunching may have been lost. Are these lost or will they be picked back up?
Here are just a few examples...
6/7/2005 12:16:07 AM||Starting BOINC client version 4.27 for windows_intelx86
6/7/2005 12:16:07 AM|SETI@home|Host location: home
..........snip.............
sorry...you've lost them all...because of my app_info.xml. With improved app_info.xml you could have kept them all.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
One thing I noticed was my old WU waiting for crunching may have been lost. Are these lost or will they be picked back up?
Here are just a few examples...
6/7/2005 12:16:07 AM||Starting BOINC client version 4.27 for windows_intelx86
6/7/2005 12:16:07 AM|SETI@home|Host location: home
..........snip.............
sorry...you've lost them all...because of my app_info.xml. With improved app_info.xml you could have kept them all.
Hi Tetsuji
Could'nt the above situation be avoded by simply putting more entries into to first part of the xml, meaning assigning 4.09, 4.10 etc etc to your optimized bin, which in turn thn could be named 9.01 or some crap like that
____________
Best regards / B52 |
|
|
|
|
Hi Tetsuji
Could'nt the above situation be avoded by simply putting more entries into to first part of the xml, meaning assigning 4.09, 4.10 etc etc to your optimized bin, which in turn thn could be named 9.01 or some crap like that
Hi,
I don't know the real solution either....IIRC on linux I edited the version info in client_state.xml and it worked perfectly, but I forgot it (at that time I was a real newbie and didn't know what client_state.xml was). Once you've run a new cruncher with app_info.xml (and lost your WU's) you'll have no further trouble. And that, if you want to change the cruncher of another name in the future, you can replace the old one by editing the cruncher field of app_info.xml. But I forgot what happened when the version info changed in app_info.xml....although I did it once.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
One thing I noticed was my old WU waiting for crunching may have been lost. Are these lost or will they be picked back up?
Here are just a few examples...
6/7/2005 12:16:07 AM||Starting BOINC client version 4.27 for windows_intelx86
6/7/2005 12:16:07 AM|SETI@home|Host location: home
..........snip.............
sorry...you've lost them all...because of my app_info.xml. With improved app_info.xml you could have kept them all.
Hi Tetsuji
Could'nt the above situation be avoded by simply putting more entries into to first part of the xml, meaning assigning 4.09, 4.10 etc etc to your optimized bin, which in turn thn could be named 9.01 or some crap like that
I was wondering if the version could be 409o instead of 411 (where o is the letter 'Oh' :). Does app_info.xml's version_num have to be a number? If not, maybe just a 'o' or 'opt' suffix of the seti client's current version might be sufficient?
____________
|
|
|
|
|
With improved app_info.xml you could have kept them all.
Is there an improved app_info.xml so I won't loose anything on my other systems. According to the other post it's not out there yet.
____________
http://www.teamstarfire.org/
|
|
|
|
|
Is there an improved app_info.xml so I won't loose anything on my other systems. According to the other post it's not out there yet.
It's widely available, unfortunately it was not included in most of the .zip files with the optimized binaries.
app_info.xml (right click and save as)
Put it in the same folder as your optimized seti app.
(edit: thanks guys for the compliments on the website :) and thanks again Tetsuji for all the hard work)
Greetings,
Speedy67
____________
|
|
|
|
|
Is there an improved app_info.xml so I won't loose anything on my other systems. According to the other post it's not out there yet.
It's widely available, unfortunately it was not included in most of the .zip files with the optimized binaries.
app_info.xml (right click and save as)
Put it in the same folder as your optimized seti app.
(edit: thanks guys for the compliments on the website :) and thanks again Tetsuji for all the hard work)
Greetings,
Speedy67
Spot on Speedy, that was the solution I was referring to.
____________
Best regards / B52 |
|
|
|
|
I recently switched from 4.19 to 4.44 and was wondering if anyone has managed to get seti-p3 and 4.44 work together? Have tried both with and without Tetsui's boinc.exe 4.44 in boinc-0531.zip.
Thanks in advance.
I'm using the original seti-p3 optimized client (shows as 4.11) with boinc 4.44, on an Athlon XP 2600+ under W2K Pro. Works fine.
Thanks Angus, nice to know someone with a similar setup as myself has gotten it to work. After I wrote that question, I fiddled a little more and it seems to work for me also now. At least the new WU's show up at the end of the list as beings scheduled for v4.11. It'll be very interesting to see how the optimized 4.11 performs on my machine. :) After I upgraded from 4.19 to 4.44, my recent average score has dropped dramatically from about 350 to now just above 315. So I need to crunch more WU's per day! LOL
I think I know a couple of possible fall pits I was experiencing before:
- BOINC didn't like to find both 4.09 and 4.11 in "C:/Program Files/Boinc/Projects/setiathome.berkeley.edu/"
I guessed that 4.09 wasn't supposed to be there anymore and deleted it. That seems to have helped.
- BOINC didn't like to find the new app_info.xml in "C:/Program Files/Boinc/" as well as in C:/Program Files/Boinc/Projects/setiathome.berkeley.edu/
It seems to have helped to delete the former instance.
- I couldn't get the optimized boinc.exe in boinc-0531.zip to work, so I'm just using the one that came with the original 4.44.
All this mess comes from my own fiddling of course. But no harm done when I did take backup every time before trying something new. However, It would perhaps help others too if there were a bit more verbose and non-ambigutious install comments. I would be glad to contribute to a readme.txt or something, if the need arises.
|
|
|
|
|
- BOINC didn't like to find both 4.09 and 4.11 in "C:/Program Files/Boinc/Projects/setiathome.berkeley.edu/"
I guessed that 4.09 wasn't supposed to be there anymore and deleted it. That seems to have helped.
- BOINC didn't like to find the new app_info.xml in "C:/Program Files/Boinc/" as well as in C:/Program Files/Boinc/Projects/setiathome.berkeley.edu/
It seems to have helped to delete the former instance.
- I couldn't get the optimized boinc.exe in boinc-0531.zip to work, so I'm just using the one that came with the original 4.44.
All this mess comes from my own fiddling of course. But no harm done when I did take backup every time before trying something new. However, It would perhaps help others too if there were a bit more verbose and non-ambigutious install comments. I would be glad to contribute to a readme.txt or something, if the need arises.
It would be nice to have a little readme file in the zip archive.
I figured out quickly that I had to delete or rename the 4.09 client to get it to use the 4.11 version, but I could never figure out where to put the app_info.xml so I just copied it to every place that seems like a possibility.
____________
|
|
|
|
|
|
NOTICE
I made a new thread. Let's continue there. I'm tired of scrolling down.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
It would be nice to have a little readme file in the zip archive.
And what did you learn from the readme that came with Seti classic back in 1999?
|
|
|
|
|
It would be nice to have a little readme file in the zip archive.
And what did you learn from the readme that came with Seti classic back in 1999?
Say what? If I recall, there was a good readme or help file for the Unix cli client. What's your point?
____________
|
|
|
|
|
|
____________
|
|
|
|
|
If I recall, there was a good readme or help file for the Unix cli client. What's your point?
Show it to me.
|
|
|
|
|
If I recall, there was a good readme or help file for the Unix cli client. What's your point?
Show it to me.
Even if I still had it on disk, which I don't, I fail to see the relevance.
You obviously have some issue with the SETI-Classic distro, but this discussion is about the user-built optimized clients for SETI-BOINC.
edit - it's on the classic site:
Here you go, Sparky - read away.
____________
|
|
|
|
|
I fail to see the relevance.
What percentage read them in the first place?
If you know what you're doing you don't need instructions. If you don't know what you're doing, it won't help much if you do read them. Example follows -
==============================================================
= = = = = = = = = = Mozilla Read Me = = = = = = = = = = =
==============================================================
Mozilla is subject to the terms detailed in the license
agreement accompanying it.
This Read Me file contains information about system
requirements and installation instructions for the Windows,
Mac OS, and Linux builds of Mozilla.
For more info on Mozilla, see www.mozilla.org. To submit bugs
or other feedback, see the Navigator QA menu and check out
Bugzilla at http://bugzilla.mozilla.org for links to known
bugs, bug-writing guidelines, and more. You can also get help
with Bugzilla by pointing your IRC client to #mozillazine
at irc.mozilla.org.
==============================================================
Getting Mozilla
==============================================================
You can download nightly builds of Mozilla from the
Mozilla.org FTP site at
ftp://ftp.mozilla.org/pub/mozilla.org/mozilla/nightly/
For the very latest builds, see
ftp://ftp.mozilla.org/pub/mozilla.org/mozilla/nightly/latest-trunk
Keep in mind that nightly builds, which are used by
Mozilla.org developers for testing, may be buggy. If you are
looking for a more polished version of Mozilla, Mozilla.org
releases Milestone builds of Mozilla every six weeks or so
that you can download from
http://www.mozilla.org/releases
Be sure to read the Mozilla release notes for information
on known problems and installation issues with Mozilla.
The release notes can be found at the preceding URL along
with the milestone releases themselves.
Note: Please use Talkback builds whenever possible. These
builds allow transmission of crash data back to Mozilla
developers, improved crash analysis, and posting of crash
information to our crash-data newsgroup.
==============================================================
System Requirements
==============================================================
*All Platforms
To view and use the new streamlined "Modern" theme,
your display monitor should be set to display
thousands of colors. For users who cannot set their
displays to use more than 256 colors, Mozilla.org
recommends using the "Classic" theme for Mozilla.
To select the Modern theme after you have installed
Mozilla, from the Navigator browser, open the View
menu, and then open then open the Apply Theme submenu
and choose Modern.
*Mac OS
-Mac OS X or later
-PowerPC processor (266 MHz or faster recommended)
-64 MB RAM
-36 MB of free hard disk space
*Windows
-Windows 95, 98, Me, NT4, 2000 or XP
-Intel Pentium class processor (233 MHz or faster
recommended)
-64 MB RAM
-26 MB free hard disk space
*Linux
-The following library versions (or compatible) are
required: glibc 2.1, XFree86 3.3.x, GTK 1.2.x, Glib
1.2.x, Libstdc++ 2.9.0. Red Hat Linux 6.0,
Debian 2.1, and SuSE 6.2 (or later) installations
should work.
-Red Hat 6.x users who want to install the Mozilla
RPM must have at least version 4.0.2 of rpm
installed.
-Intel Pentium class processor (233 MHz or faster
recommended)
-64MB RAM
-26MB free hard disk space
==============================================================
Installation Instructions
==============================================================
For Mac OS and Windows users, it is strongly recommended that
you exit all programs before running the setup program. Also,
you should temporarily disable virus-detection software.
For Linux users, note that the installation instructions use
the bash shell. If you're not using bash, adjust the commands
accordingly.
For all platforms, install into a clean (new) directory.
Installing on top of previously released builds may cause
problems.
Note: These instructions do not tell you how to build Mozilla.
For info on building the Mozilla source, see
http://www.mozilla.org/source.html
Windows Installation Instructions
---------------------------------
Note: For Windows NT/2000/XP systems, you need Administrator
privileges to install Mozilla. If you see an "Error 5" message
during installation, make sure you're running the installation
with Administrator privileges.
To install Mozilla by downloading the Mozilla installer,
follow these steps:
1. Click the the mozilla-win32-installer.exe link on
the site you're downloading Mozilla from to download
the installer file to your machine.
2. Navigate to where you downloaded the file and
double-click the Mozilla program icon on your machine
to begin the Setup program.
3. Follow the on-screen instructions in the setup
program. The program starts automatically the first
time.
To install Mozilla by downloading the .zip file and
installing manually, follow these steps:
1. Click the mozilla-win32-talkback.zip link or the
mozilla-win32.zip link on the site you're down-
loading Mozilla from to download the .zip file to
your machine.
2. Navigate to where you downloaded the file and
double-click the compressed file.
Note: This step assumes you already have a recent
version of WinZip installed, and that you know how to
use it. If not, you can get WinZip and information
about the program at www.winzip.com.
3. Extract the .zip file to a directory such as
C:\Program Files\mozilla.org\Mozilla.
4. To start Mozilla, navigate to the directory you
extracted Mozilla to and double-click the Mozilla.exe
icon.
Mac OS X Installation Instructions
----------------------------------
To install Mozilla by downloading the Mozilla disk image,
follow these steps:
1. Click the mozilla-mac-MachO.dmg.gz link to download
it to your machine. By default, the download file is
downloaded to your desktop.
2. Once you have downloaded the .dmg.gz file, drag it
onto Stuffit Expander to decompress it. If the disk
image doesn't mount automatically, double-click on the
.dmg file to mount it. If that fails, and the file
does not look like a disk image file, do a "Show Info"
on the file, and, in the "Open with application"
category, choose Disk Copy. In Mac OS 10.2, you can
use "Open with" from the context menu.
3. Once the disk image mounts, open it, and drag the
Mozilla icon onto your hard disk.
4. We recommend that you copy it to the Applications
folder.
5. Now Eject the disk image.
6. If you like, you can drag Mozilla to your dock to
have it easily accessible at all times. You might also
wish to select Mozilla as your default browser in the
Internet system preferences pane (under the Web tab).
Linux Installation Instructions
-------------------------------
Note: If you install in the default directory (which is
usually /usr/local/mozilla), or any other directory where
only the root user normally has write-access, you must
start Mozilla first as root before other users can start
the program. Doing so generates a set of files required
for later use by other users.
To install Mozilla by downloading the Mozilla installer,
follow these steps:
1. Create a directory named mozilla (mkdir mozilla)
and change to that directory (cd mozilla).
2. Click the link on the site you're downloading
Mozilla from to download the installer file
(called mozilla-1686-pc-linux-gnu-installer.tar.gz)
to your machine.
3. Change to the mozilla directory (cd mozilla) and
decompress the archive with the following command:
tar zxvf moz*.tar.gz
The installer is now located in a subdirectory of
Mozilla named mozilla-installer.
4. Change to the mozilla-installer directory
(cd mozilla-installer) and run the installer with the
./mozilla-installer command.
5. Follow the instructions in the install wizard for
installing Mozilla.
Note: If you have a slower machine, be aware that the
installation may take some time. In this case, the
installation progress may appear to hang indefinitely,
even though the installation is still in process.
6. To start Mozilla, change to the directory where you
installed it and run the ./mozilla command.
To install Mozilla by downloading the tar.gz file:
1. Create a directory named "mozilla" (mkdir mozilla)
and change to that directory (cd mozilla).
2. Click the link on the site you're downloading
Mozilla from to download the non-installer
(mozilla*.tar.gz) file into the mozilla directory.
3. Change to the mozilla directory (cd mozilla) and
decompress the file with the following command:
tar zxvf moz*.tar.gz
This creates a "mozilla" directory under your mozilla
directory.
4. Change to the mozilla directory (cd mozilla).
5. Run Mozilla with the following run script:
./mozilla
To hook up Mozilla complete with icon to the GNOME Panel,
follow these steps:
1. Click the GNOME Main Menu button, open the Panel menu,
and then open the Add to Panel submenu and choose Launcher.
2. Right-click the icon for Mozilla on the Panel and
enter the following command:
directory_name./mozilla
where directory_name is the name of the directory
you downloaded mozilla to. For example, the default
directory that Mozilla suggests is /usr/local/mozilla.
3. Type in a name for the icon, and type in a comment
if you wish.
4. Click the icon button and type in the following as
the icon's location:
directory_name/icons/mozicon50.xpm
where directory name is the directory where you
installed Mozilla. For example, the default directory
is /usr/local/mozilla/icons/mozicon50.xpm.
____________
|
|
|
|
|
- BOINC didn't like to find both 4.09 and 4.11 in "C:/Program Files/Boinc/Projects/setiathome.berkeley.edu/"
I guessed that 4.09 wasn't supposed to be there anymore and deleted it. That seems to have helped.
- BOINC didn't like to find the new app_info.xml in "C:/Program Files/Boinc/" as well as in C:/Program Files/Boinc/Projects/setiathome.berkeley.edu/
It seems to have helped to delete the former instance.
- I couldn't get the optimized boinc.exe in boinc-0531.zip to work, so I'm just using the one that came with the original 4.44.
All this mess comes from my own fiddling of course. But no harm done when I did take backup every time before trying something new. However, It would perhaps help others too if there were a bit more verbose and non-ambigutious install comments. I would be glad to contribute to a readme.txt or something, if the need arises.
It would be nice to have a little readme file in the zip archive.
I figured out quickly that I had to delete or rename the 4.09 client to get it to use the 4.11 version, but I could never figure out where to put the app_info.xml so I just copied it to every place that seems like a possibility.
http://www.shadyshack.dyndns.org/boinc.html
____________
<img src='http://www.boincsynergy.com/images/stats/comb-1527.jpg'> |
|
|
|
|
http://www.shadyshack.dyndns.org/boinc.html
Thanks very much, Shady! :)
<blockquote>Tetsuji said:
NOTICE
I made a new thread. Let's continue there. I'm tired of scrolling down.</blockquote>
OK. :)
____________
 |
|
|
|
|
|
Hiya!
Could someone help me out? I swear once upon a time I could figure this stuff out but my stint as a SAHM has completely corroded my computer skills.
How do I figure out what kind of P4 that I have? I know its multi-threading and 3.something GHz.
Thank you for the hand holding.
- T.
____________
 |
|
|
|
|
|
Go here and get 1.29. It should tell you all you need to know.
http://www.cpuid.org/cpuz.php
|
|
|
|
|
Go here and get 1.29. It should tell you all you need to know.
http://www.cpuid.org/cpuz.php
Thank you SO much!
____________
|
|
|
|
|
|
Ok so I checked out CPU-Z program and it told me what my ancient tape, chewing-gum and bailing-wire lap-top is using. Surprise! It has MMX. Sweet. So I d/l'd the appropriate vanilla optimized program and after several hiccups and misunderstandings have the thing running. This is, of course, why I did it to my ancient laptop, first. My "big machine" is my daily workhorse and I only mess with it after I'm SURE I know what I'm doing.
(BIG SHOUT OUT TO SHADY! Thank you for the detailed "How To".)
SO then I went to the "my computers" list and *erp* that listing is now showing "remoteigmu" (ancient laptop) as never having done any work before!
Is this normal? How does this happen? Does this mean that the work "remoteigmu" has done has been wiped from my credits?
____________
|
|
|
|
|
|
{EmilyLatilla} Nevermind! {/EL}
____________
|
|
|
|
|
|
Just to clarify Re: Pentium M, Pentium 4-M
The Pentium M (Banias, Dothan) IS based on the Pentium III (with "Pentium 4 FSB, Pentium III Execution Units")
The Pentium 4-M is a MOBILE version of the Pentium 4 (as seen in some laptops! eg Tosh Sat Pro 1.8ghz P4-M) - it is _NOT_ the Dothan.
The 1.8ghz Pentium M (Dothan, 2mb cache) is drastically quicker than the 1.8ghz Pentium 4-M. (PM 1.8ghz around 7-8000 seconds, Pentium 4-M 1.8ghz around 18000 seconds)
Sorry to sound a bit shouty - it's just I wanted to clarify the difference, and try and clear up some confusion.
You can read all about the Pentium M here:
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2129
____________
|
|
|
PappaForum moderator
Volunteer tester
Send message
Joined: Jan 9 00
Posts: 2115
Credit: 5,106,949
RAC: 9,201
|
|
I know that you have did a lot of work to make clients that actually perform as they should, according to the processor type that run SETI... It would be very nice that Seti UCB Accredits your work and others...
A small thing, You are donating more than most Users! This goes along with Ned and others that are making the basic flawed "CORE BOINC" and the "Client" work properly...
R/
Al
A month ago I made optimized clients for Windows using an evaluation copy of Intel C++ compiler and fftw library, but as I mentioned in another thread, I received money from a person to buy the license...so anyone can use the binary the binaries legally. Thank you, Mr.X!! (He wants me not to mention his name.)
Links are as follows: I added seti-sse2.zip, which I hope works for Athlon64.
seti-plainvanilla.zip For P2 or earlier....but this uses MMX instructions.
seti-p3.zip for P3 (and Athlon XP?)
seti-sse2.zip for Athlon64(?) This requires special caution. No test has been done yet.
seti-p4-no-sse3.zip for P4 w/o SSE3. According to Intel, this is slightly different from the above one. So this doesn't work for Athlon64.
seti-p4-sse3.zip for P4 with SSE3
seti-pm.zip for Pentium M.
referenceunits.zip a work unit and a result unit for benchmark test.
Most of them are the ones I uploaded before, except for seti-sse2.zip. All of them are based on Apr-12 nightly build of boinc/seti and fftw-3.0. If you want to compile by yourself (you need Intel compiler and VC++7.1), sources are
seti-source-1.zip, seti-source-2.zip, and seti-source-3.zip.
CAUTION
These binaries are provided as is. No guarantees about proper operation can be given. Only you will be to blame if binaries cash your computer, damage data or hardware, kick your dog or insult your mother. (from Metod's page and I concur.)
I strongly recommend to make a backup copy of the whole boinc folder before installing these clients in case it crashes....it may lose your precious data due to incompatibility (or by my mistake?)
Happy crunching!! You can copy or distribute them for free :)
____________
Please consider a Donation to the Seti Project.
 |
|
|
|
|
|
Well, I finally decided to take the plunge and went 'optimized' this weekend, with some interesting results:
2.4 P4, 512MB - This fell from 4 hours and change per WU to about 2.5
1.8 P4-M, 512MB - This fell from 3.5 hours to about 2.5
AMD Duron 1.8, 512MB - This went from just under 4 hours to 2.25
2.5 Celeron, 1GB - This was my 'problem child' that would bench at 6-7 hours and crunch in 9.5/wu. This now benches at 6h 20 and crunches in 2h 30 - 2h 40.
My WU output jumped from about 20 WU/day to almost 40.
All machines are running BOINC 4.72 w/ YAOSCW (8.1).
____________
Stewie: So, is there any tread left on the tires? Or at this point would it be like throwing a hot dog down a hallway?
Fox Sunday (US) at 9PM ET/PT |
|
|
|
|
I've returned the first WU crunched by this optimized SETI client, and the processing time is greatly reduced, but so is the claimed credit. As long as I am the only member of the quorum processing the WU with the optimized client it looks like I will benefit (as my low number will be discarded in the averaging process); but if all members of the quorum were using this client, then we would all claim and be granted a smaller credit.
Of course, this would be balanced by the fact that we would all process more WUs per unit of time.
So it looks like, for now, I have an "advantage".
Is my reasoning correct here?
I have an Excel file compairing the times and credit for Non Optomized and Optomized Clients at:
http://www.rays-place.com/seti-times.xls
This was not scientific, just my first 20 Optomized and the last 20 non optomized. You will notice that even thou the claimed credit goes down per unit the granted credit per hour goes up quite a bit.
When I did that about 6 weeks ago I was using the standard BOINC core, which I later updated to a Optomized one, this increased the Claimed credit quite a bit and increased the granted credit on about half the units, with the same time used as with the Optomized SETI unit on the Excel file.
You can save a copy of the file and fill in your own info inplace of mine to see how much better you do.
All figures for credit on the file are for Granted Credit, the important line is at the bottom whare it figures the average gredit per hour. Mine went from 6.2742 up to 13.1082 (a little more than dubble per hour).
Cheers
Ray
____________
Pizza@Home Rays Place Rays place Forums |
|
|
|
|
|
Thanks all. I'm happy to hear the clients help.
____________
Luckiest in the world. WMD = Weapon of Mass Distraction.
Click this table.
|
|
|
|
|
Thanks all. I'm happy to hear the clients help.
Your optimized clients are fantastic! :)
____________
Join BOINC Synergy!
misfit@boincsynergy.com |
|
|
|
|
|
one q will the XP program work with the MP chips? okay 2 qs with MP you have mult chips (2 for me) will these programs work for me?
____________
|
|
|
|
|
one q will the XP program work with the MP chips? okay 2 qs with MP you have mult chips (2 for me) will these programs work for me?
Yes, Athlon MPs are to all intensive purposes just Athlon XP chips that work in multiprocessor formats.
Ned
____________
*** My Guide to Compiling Optimised BOINC and SETI Clients ***
*** Download Optimised BOINC and SETI Clients for Linux Here *** |
|
|
|
|
|
Good lord!
I've been using the 4.18 Official client on my system. Win2k with 2 gigs, SATA drives, and AMD 64 3500+.
Usually crunch per unit is 2h 25m.
DL/d and installed the 4.11 Optimized SSE2 client and Boinc, and let it fly.
46m per unit. <CRASH!!> (picks jaw up off floor)
For those that can do the math in their head, YES. Thats a 3x increas in speed.
Needless to say, I was NOT expecting that kind of acceleration.
Only did this cause Seti servers were down for the 'catchup/compression', elsewise, I might not have tried it. :)
____________
Computers:
AMD 64 3500+ w/Win2k <dead>
AMD 1.4ghz w/Win98SE <dead>
AMD 1ghz Duron w/Win98SE <dead>
AMD 1ghz w/Win98SE <dead>
AMD 64 X2 4000+ w/Vista Home Premium |
|
|
|
|
Good lord!
46m per unit. <CRASH!!> (picks jaw up off floor)
I hope it didn't hurt to much while your jaw hit the floor :)
P.S. you can speedup you Athlon too. There's a "PI/II and compatible" version available for download at my page ;)
____________
Have a bad Day!
 |
|
|
|
|
|
Um... JakeSnake, if you'd picked a thread newer than 6 months old to look at, you might have seen the fact that Crunch3r's version is TWICE AGAIN as fast as the one you just installed... so instead of 3x, you would see more like a 6x improvement. :-)
|
|
|
|
|
Um... JakeSnake, if you'd picked a thread newer than 6 months old to look at, you might have seen the fact that Crunch3r's version is TWICE AGAIN as fast as the one you just installed... so instead of 3x, you would see more like a 6x improvement. :-)
Tsk Tsk, for ASSUMING I didn't install Crunch3rs version, you get a neg :)
____________
Computers:
AMD 64 3500+ w/Win2k <dead>
AMD 1.4ghz w/Win98SE <dead>
AMD 1ghz Duron w/Win98SE <dead>
AMD 1ghz w/Win98SE <dead>
AMD 64 X2 4000+ w/Vista Home Premium |
|
|
|
|
I hope it didn't hurt to much while your jaw hit the floor :)
P.S. you can speedup you Athlon too. There's a "PI/II and compatible" version available for download at my page ;)
Not sure if it'd work. My other 2 machines are Athlons (Thunderbirds), but their running Win98se on them.
____________
Computers:
AMD 64 3500+ w/Win2k <dead>
AMD 1.4ghz w/Win98SE <dead>
AMD 1ghz Duron w/Win98SE <dead>
AMD 1ghz w/Win98SE <dead>
AMD 64 X2 4000+ w/Vista Home Premium |
|
|
|
|
I hope it didn't hurt to much while your jaw hit the floor :)
P.S. you can speedup you Athlon too. There's a "PI/II and compatible" version available for download at my page ;)
Not sure if it'd work. My other 2 machines are Athlons (Thunderbirds), but their running Win98se on them.
I am using Crunch3r's app on my xp2500+ under win98se. Not a single problem using it if your system is healthy. Crunch time is 1hr 10min to 1hr 20min.
:)
____________
98SE XP2500+ @ 2.1 GHz Boinc v5.8.8
And God said"Let there be light."But then the program crashed because he was trying to access the 'light' property of a NULL universe pointer. |
|
|
|
|
Good lord! --- AMD 64 3500+. 46m per unit.
Is that just the current estimate or the actual run times?
If actual run times, you might have something wrong. My AMD Athlon 64 3400+ is doing 38m to 39m WU's with Crunch3r's client and Host. The 3400+ has the 1MB L2 cache and your 3500+ may only have the 512KB L2 cache, but it should not make that much difference. I have an AMD Athlon 64 3000+ with 512MB L2 cache that is doing 45m to 46m WU's. Are you running PC3200 memory? I have CL=2.5 which now makes over a minute difference in WU times as compared to only a few 10 seconds with the standard client.
The only thing I know of that would be causing your machine to run that slow is if you have PC2700 memory or if you system board is running your PC3200 at DDR333 speed instead of DDR400. This happens on the older socket single channel boards that have three memory sockets (most just have two) and you try to use two double sided PC3200 memory modules. The board is forced to drop them back to DDR333 speed, and does not warn you. You can see this when you run a current version of MEMTEST32 that shows your actual FSB speed, not what it is set for and that you think it is running. On those boards, it will run at DDR400 speed if you have one double sided and one single sided DIMM but you lose the 4-way interleave.
If you have two double sided DIMM in a three memory socket system board, pull one of them out and give it a try. Or at least boot with and run MEMTEST32 and see what it tells you your FSB is running. Hum.. 2GB, you almost have to have double sided memory, unless that is the newer socket and a dual channel FSB board. Then I really don't know why your system is so slow.
|
|
|
|
|
|
4 double sided ddr400s, clocked down to ddr333 speeds to they'll run dualchannel and not futz up. nasty little hairball that, but it works.
and an extra 5-7 mins isnt going to tweak me. its already 3x faster with Crunch3r's clients :)
____________
Computers:
AMD 64 3500+ w/Win2k <dead>
AMD 1.4ghz w/Win98SE <dead>
AMD 1ghz Duron w/Win98SE <dead>
AMD 1ghz w/Win98SE <dead>
AMD 64 X2 4000+ w/Vista Home Premium |
|
|
|
|
4 double sided ddr400s, clocked down to ddr333 speeds
OK, if that is what it takes to keep the hardware happy.
I would have tried playing with the other memory timing settings to get around any problem. Have not played with a dual channel board (yet), and don't find BOINC interesting enough to spend that much.
In another thread several days ago, I pointed out that on the Athlon 64 processors, memory running at DDR333 speed only had a very small effect on the standard client and a small but noticeable one on the older optimized client. But with Crunch3r's newest, the slower memory makes a big difference. With your processor and FSB at DDR400 with the dual channel and 4-way interleave, it should do 37m WU's and below 36m WU's with CL=2.5 memory. But then that would get expensive. |
|
|
|
|
|
Bios entries for the Crucial memories I'm using are optimized for their capabilities. Stepping the mem speed down so the dualchannel wouldn't bark was a necessity.
Trouble is, nobody had the single sided's when I built this beast. :/
____________
Computers:
AMD 64 3500+ w/Win2k <dead>
AMD 1.4ghz w/Win98SE <dead>
AMD 1ghz Duron w/Win98SE <dead>
AMD 1ghz w/Win98SE <dead>
AMD 64 X2 4000+ w/Vista Home Premium |
|
|
|
|
|
Anyone have any input on which one of Crunch3r's clients should be used on a sempron? I have 3 sempron systems between work and home...I'm guessing the Athlon XP rather than the Athlon 64, even though one of them is a 64bit 3100+...thanks
____________
|
|
|
|
|
Anyone have any input on which one of Crunch3r's clients should be used on a sempron? I have 3 sempron systems between work and home...I'm guessing the Athlon XP rather than the Athlon 64, even though one of them is a 64bit 3100+...thanks
Use the CPU-Z program from the www.cpuid.org web site. It'll tell you what your CPU supports. Look specifically for the Instructions supported. (MMX, 3DNow, SSE, SSE2, SSE3, x86-64) and choose the client accordingly.
My AMD 64 3500+ supports up to SSE2, so thats the client I used.
____________
Computers:
AMD 64 3500+ w/Win2k <dead>
AMD 1.4ghz w/Win98SE <dead>
AMD 1ghz Duron w/Win98SE <dead>
AMD 1ghz w/Win98SE <dead>
AMD 64 X2 4000+ w/Vista Home Premium |
|
|
