Using Boinc And Other Tweaks Thread

V_R

¯\_(ツ)_/¯
Moderator
Joined
Jan 31, 2005
Messages
13,572
Reaction score
1,888
1. Machine Specs (CPU (inc. any overclocked speed) & RAM)
1) AMD Athlon 3200XP @ 2.36Ghz & 1GB RAM

2. Date started using Boinc
2) 11/01/06

3. Any Tweak/Enhancement used & which one
None, Installed Guntec.de tweak last night. Will let you know.

4. Average daily points/results returned PRIOR to installing boinc
About 1000/4

5. Average daily points/results returned AFTER installing boinc
About 1600-2000/6

6. Average Machine runtime (ie 24/7 or otherwise)
24/7

Definatly makes a difference here, Heres the graph of my stats. BOINC installed on 11/01. :)

EDIT: On the graph, my other pc returned 1 result on the 22, 15, 13, 7, 5, 3 jan and 31st Dec.

attachment.php
 

Attachments

  • WCG Graph After Tweaking.jpg
    WCG Graph After Tweaking.jpg
    36.4 KB · Views: 255
Last edited:

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
muckshifter said:
The only other advantage I can see is if you have an AMD dual-core CPU ... you in effect now have 2 PCs if using Boinc.

Sorry ... this post is not intended to throw Ady's great work into disrepute, in fact I greatly appreciate all his work done on this, nay, I'm just not convinced that using Boinc over UD will make a scap of difference to your points. However, don't let me put you off trying and if you is happy, so am I.

:D
Obviously I can't speak for others, but in my case I was using UD with a tweak so I was utilising both cores already - and as you can see since using boinc + enhancement it has doubled my points on the machine, so in my case there has been a significant improvement.

I think boink is better for high performing machines rather than lower ones due to the way each benchmarks - with UD there is only a certain level you can get to so it comes to a point where it doesn't matter how much RAM you have or type of CPU you can't improve your benchmark any further.
With boinc it gives ever higher benchmark scores to higher performing CPU's which is obviously an advantage.

This is why I am seeking this information, so we can assess the relative merits of each program according to machine spec, and advise accordingly.

:thumb:
 
Joined
Apr 19, 2005
Messages
6,175
Reaction score
2
Mine is showing quite an increase:cool:
 

Attachments

  • untitled.JPG
    untitled.JPG
    7.5 KB · Views: 143

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
itsme said:
Mine is showing quite an increase:cool:
Thats great itsme :thumb:

Can you list the specs as per my earlier post (#37) so we can evaluate the information? cheers!:)
 
Last edited:
Joined
Apr 19, 2005
Messages
6,175
Reaction score
2
Ady>
Spec 1
AMD Athlon XP 2600+ running 2.06 MHz

512 RAM

Join BOINC 13/01/06

Spec 2
AMD Athlon XP 2800+ running 2.00 MHz
1GIG RAM

Join BOINC 16/01/06

Any Tweak/Enhancement used= NONE
Average Machine runtime 12/24

Points have definitely increased :thumb:
 

V_R

¯\_(ツ)_/¯
Moderator
Joined
Jan 31, 2005
Messages
13,572
Reaction score
1,888
Thought i'd share a quick update with you guys. :)

Heres how my Device Stats are reading on the WCG.

I started crunching with the UD Agent on the 13th Nov '05 up to the 11th Jan '06.
Total Run Time - 56 days, 19 hours and 13 minutes.
Total points - 50,868
Results returned - 211


Now BOINC. 12th Jan to 9th Feb (Today)
Total Run Time - 26 days, 1 hour and 36 minutes
Total points - 50,398
Results Returned - 146


This is the same pc no change in components or overclock.

I installed the optimized BOINC on 28th Jan.

Please note the difference in the points and run time. :)

BOINC is King!! :D
 

Ian

Administrator
Joined
Feb 23, 2002
Messages
19,873
Reaction score
1,499
I've had similar results here too - I think BOINC is the way forward for most :D
 

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
Nice Stats V_R - Boinc is definitely producing the goods :thumb:
 

Quadophile

Hon. Acoustical Engineer
Moderator
Joined
Mar 16, 2002
Messages
6,643
Reaction score
566
I have read the entire thread today as I am contemplating switching to Boinc! :D However, most of you guys know that I had teething problems in the begining and issue about system heating up. I am a bit skeptical about the whole issue but I am willing to give it a try if it would help the cause. I am also willing to put in time to monitor my PC very carefully for any issues and also update you guys with any feedback you ask for.

Now please, someone tell me if I should give it a try or not. My specs are as follows:

IBM T43 Laptop
Pentium M 2.0Ghz mobile CPU with 2 MB cache built-in
1 GB of PC 4200 DDR Ram

They only tweak that I had to do when started crunching was to undervolt the CPU to address the issue of over heating, other than that no changes are made.

Look forward to your opinions/suggestions.
 

Ian

Administrator
Joined
Feb 23, 2002
Messages
19,873
Reaction score
1,499
I don't see why it should cause any other issues as opposed to the other application - so I think the only way to find out will be to try it :thumb:

Might be worth running it for 2 days and then see how it goes?

I've just had a thought - take look at this page:

http://einstein.phys.uwm.edu/download_network.php

There is a plugin called "ThreadMaster" that limits the CPU to a percentage - you could run BOINC and set this at say 80%? That way you'd get an overall result increase (as BOINC seems to almost double results) but you would keep your laptop nice and cool?
 

Quadophile

Hon. Acoustical Engineer
Moderator
Joined
Mar 16, 2002
Messages
6,643
Reaction score
566
Ian,

Thanks for the advise and great link. I will get on with it and prepare everything for the coming Friday when I switch. Weekend is better for such kind of things and if there are issues I can deal with them without worrying about my work getting affected. :thumb:

Cheers! :)
 

muckshifter

I'm not weird, I'm a limited edition.
Moderator
Joined
Mar 5, 2002
Messages
25,739
Reaction score
1,204
Another fine day ... here's another for you Ady;

Before ...

03/04/2006 6:05:20 pm||Running CPU benchmarks

03/04/2006 6:06:19 pm||Benchmark results:

03/04/2006 6:06:19 pm|| Number of CPUs: 1

03/04/2006 6:06:19 pm|| 1863 double precision MIPS (Whetstone) per CPU

03/04/2006 6:06:19 pm|| 3465 integer MIPS (Dhrystone) per CPU

03/04/2006 6:06:19 pm||Finished CPU benchmarks




After ...

03/04/2006 6:23:29 pm||Benchmark results:

03/04/2006 6:23:29 pm|| Number of CPUs: 1

03/04/2006 6:23:29 pm|| 2775 double precision MIPS (Whetstone) per CPU

03/04/2006 6:23:29 pm|| 8019 integer MIPS (Dhrystone) per CPU

03/04/2006 6:23:29 pm||Finished CPU benchmarks


... rock on Tommy.
nod.gif
 

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
Thanks Mucks - I'm assuming thats the before and after tweaking benchmarks - just shows the improvement when using the optimiser :thumb:

Can you remind me of the CPU again please....

Does this mean your fully up and running with XP now then :D
 

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
Quadophile said:
Ian,

Thanks for the advise and great link. I will get on with it and prepare everything for the coming Friday when I switch. Weekend is better for such kind of things and if there are issues I can deal with them without worrying about my work getting affected. :thumb:

Cheers! :)
Sorry Quad, I missed this due my week away!

Using Boinc should put no more strain on your lappy than the WCG UD agent - in fact it may be less due to the lack of flashy graphics.

Give it a go at 100% CPU usage, and see what the temps are like compared to UD, if they are higher then maybe try Threadmaster....

Also remember the optimiser available in my first post which increases the benchmarking significantly (see Mucks's figures above) - they list one for Pentium 3 (SSE) and Pentium 4 (SSE2) - not sure what the Pentium M CPU comes under, but I would have thought it would be part of the Pentium 4 family.

I will try to find out for you, unless you already know, in which case let me know for future reference! :D
 

muckshifter

I'm not weird, I'm a limited edition.
Moderator
Joined
Mar 5, 2002
Messages
25,739
Reaction score
1,204
Adywebb said:
Thanks Mucks - I'm assuming thats the before and after tweaking benchmarks - just shows the improvement when using the optimiser :thumb:

Can you remind me of the CPU again please....

Does this mean your fully up and running with XP now then :D
Yep, fully running XP :rolleyes:

... with Linux. :D


The CPU is an AMD 3200 64bit on an Asus A8N-E SLI 1024mb ram with a 6600 vid

It will be interesting to see what the returns are like compared to Linux ... I never got around to Tweaking the Linux Boinc ... so I'll leave Windows running 24/7 for awhile and if it is better, then I won't bother re-installing Boinc on the Linux side. :thumb:
 

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
Thanks for the link Quad :thumb:

I see you have already optimised boinc and it has increased the benchmarking considerably.

What this does basically is make the WU's complete faster therefore you crunch more results which = more points - so it is a win-win situation :cool:

Can you tell me exactly which Guntec optimiser you used for your Pentium M, the Pentium 3 (SSE) or the Pentium 4 (SSE2)? (I'm assuming your using Windows)

Also, so we can keep the benchmarking stats in one place, can you list your PC specs together with the before and after benchmarks in a post here please :)

Cheers :D
 

Quadophile

Hon. Acoustical Engineer
Moderator
Joined
Mar 16, 2002
Messages
6,643
Reaction score
566
Adywebb said:
Thanks for the link Quad :thumb:

I see you have already optimised boinc and it has increased the benchmarking considerably.

What this does basically is make the WU's complete faster therefore you crunch more results which = more points - so it is a win-win situation :cool:

Can you tell me exactly which Guntec optimiser you used for your Pentium M, the Pentium 3 (SSE) or the Pentium 4 (SSE2)? (I'm assuming your using Windows)

Also, so we can keep the benchmarking stats in one place, can you list your PC specs together with the before and after benchmarks in a post here please :)

Cheers :D

No problems Ady, here are the details you asked for;)


Before the Guntec.de tweak

28/03/2006 22:01:53|| 1722 double precision MIPS (Whetstone) per CPU

28/03/2006 22:01:53|| 3513 integer MIPS (Dhrystone) per CPU

After the Guntec.de tweak

28/03/2006 22:17:12|| 2384 double precision MIPS (Whetstone) per CPU

28/03/2006 22:17:12|| 7880 integer MIPS (Dhrystone) per CPU

The specs for my Laptop are as follows:

IBM T43
Pentium M 760 2.0Ghz with 2 MB cache
Ram 1GB
HD 80 GB
Windows XP


I used the Boinc Client version 5.2.13 Windows SSE2

Just updated the Notebook Hardware Contol software to 1.10 Beta 3 for monitoring purposes.

If you need anything more just ask :)
 

Adywebb

Growing old....
Moderator
Joined
Jan 1, 2005
Messages
5,459
Reaction score
21
Excellent Quad - many thanks :thumb:

From what I see for the last few days points returns, it certainly seems to be making a difference :D
 
Last edited:

Quadophile

Hon. Acoustical Engineer
Moderator
Joined
Mar 16, 2002
Messages
6,643
Reaction score
566
Adywebb said:
Excellent Quad - many thanks :thumb:

From what I see for the last few days points returns, it certainly seems to be making a difference :D

It sure is going like a rocket :D I am really surprised at what it was capable of but the best tweak was to lower the voltage in order to get the temperatures to come down to realistic levels without which I am afraid all this would not be possible ;)
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top