Go Back   hwbot.org > Overclocking & benchmarking > Benchmark software

Benchmark software Discuss benchmarking software.

Reply
 
Thread Tools
  #11  
Old 05-05-2009, 17:52
allegratorial allegratorial is offline
maintenance bot
 
Join Date: Jan 2009
Location: UNITED STATES
Posts: 15
Default

Quote:
Originally Posted by Tharamis View Post

again: maxxpi does *not claim to be the fastest PI application.
it does not need to be.

fast enough to bench without sleeping *and long enough to show clearly differences between different setups. speed doesnt matter at all. it's the comparative between pc's
that makes a benchmark a benchmark.
true, being more of a software benchmarker, I've never really thought about this.

Quote:
Originally Posted by Tharamis View Post

hmm good question...!?!, here a screen from an very early alpha, but as i said
i don't think that this will be used in maxxpi (q6600 at 4ghz):
Assuming clock speed scales roughly linear, 36 seconds for 32M will beat both quickpi and y-cruncher in single-threaded mode.
It will also beat quickpi in multi-threaded mode... But it scales too poorly to even compare with y-cruncher.


Interesting question to ask:

Of the 3 multithreaded pi programs that exist now:

Why do MaxxPi and QuickPi's implementations for Chudnovsky's formula scale so poorly with multiple cores? Whereas y-cruncher achieves near linear scaling.

Interesting thing to notice is that the author y-cruncher is merely a junior in college. His purpose for writing the program was to smash a few size records (and he did). (and as you'd mentioned: for record breaking, speed matters)

Also, when I compared the speeds of the other constants that y-cruncher can compute with QuickPi (all of which y-cruncher currently holds the world record for), y-cruncher beats QuickPi hands down even in single-threaded mode. It's only with Pi is y-cruncher slower than QuickPi - which leads me to think that because there's no "attainable" record at stake for Pi, this kid never even bothered to optimize his implementation for Pi.

So from a software benchmarker's standpoint, this has me wondering what y-cruncher could turn into given that it's already a killer program in terms of pure-multithreaded speed. I also wonder what will happen when this kid gets older and becomes more experienced.

Of course crunching pi itself is pretty useless, but it's the underlying arithmetic engine in y-cruncher that is valuable as it currently has the only multithreaded multiplication in the world that will beat even GMP - and it does so single-threaded and without assembly optimizations.


Quote:
Originally Posted by Tharamis View Post

i think you have to read this:
http://en.wikipedia.org/wiki/Benchmark_(computing)
to understand.
if you willing to get a worldrecord by calculation PI
with xxxxM then your are right=speed matters.



well as i said *binary splitting*, that means multicore(thread) for one calculation.

chudnovsky *is the fastest formula at current,
but it will give not that consistently cpu load as gauss do.



surely do, look at MaxxPI :-)

but anyways, if your favorite is y-cruncher then use it!
it's a pice of wonderfull and incedible fast software.



that's the point!

cu
Yes I know what benchmarking is. I'm more of a software benchmarker than a hardware benchmarker.

I know that Chudnovsky's formula is currently the fastest known algorithm, but all that binary splitting stuff is over my head. too much math for me.


And by "fast", I mean something comparable to QuickPi at the least. If you can get him to release a GUI version of his Chudnovsky implementation, then it will satisfy both "fast" and "pretty".
__________________

Last edited by allegratorial; 05-05-2009 at 17:55.
Reply With Quote
  #12  
Old 05-05-2009, 19:20
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

Quote:
Originally Posted by allegratorial View Post
Why do MaxxPi and QuickPi's implementations for Chudnovsky's formula scale so poorly with multiple cores? Whereas y-cruncher achieves near linear scaling.
nearly linear scaling with cores/threads is not possible.
i personally think, that y-cruncher also uses binary splitting and a strong
usage of gmp. so if this is true (or near by truth) linear scaling is im-possible.
this is one of the reasons why MaxxPI² will calculate parallel PI results for each core.
to keep much load as possible on the cores.

Quote:
Originally Posted by allegratorial View Post
Interesting thing to notice is that the author y-cruncher is merely a junior in college. His purpose for writing the program was to smash a few size records (and he did). (and as you'd mentioned: for record breaking, speed matters)
only one thing matters here: TIME, if you had time, everything is possible
and as i was in college... i had time. much time.

Quote:
Originally Posted by allegratorial View Post
So from a software benchmarker's standpoint, this has me wondering what y-cruncher could turn into given that it's already a killer program in terms of pure-multithreaded speed. I also wonder what will happen when this kid gets older and becomes more experienced.
hmm... GMP... i think, but anyways i wish him luck!

Quote:
Originally Posted by allegratorial View Post
And by "fast", I mean something comparable to QuickPi at the least. If you can get him to release a GUI version of his Chudnovsky implementation, then it will satisfy both "fast" and "pretty".
and again: MaxxPI² is only comparable (PI and all other calc.benchs) to it self.

well, for me personally, MaxxPI is fast enough, it provides reliable consistent results.
this is the most important thing.

There is no need to hassle about 512M in 1sec. this is useless
and the author of MaxxPI² shares this opinion with me (i strongly think).

and don't forget this thread is written for MaxxPI²,
not for y-cruncher and comparing against them.

cu
__________________
Reply With Quote
  #13  
Old 05-05-2009, 20:19
allegratorial allegratorial is offline
maintenance bot
 
Join Date: Jan 2009
Location: UNITED STATES
Posts: 15
Default

Quote:
Originally Posted by Tharamis View Post
nearly linear scaling with cores/threads is not possible.
i personally think, that y-cruncher also uses binary splitting and a strong
usage of gmp. so if this is true (or near by truth) linear scaling is im-possible.
this is one of the reasons why MaxxPI² will calculate parallel PI results for each core.
to keep much load as possible on the cores.
Maxxpi uses GMP? So that Pi chudnovsky implementation that it uses is merely this?

http://gmplib.org/pi-with-gmp.html


Whatever y-cruncher uses, it gets more than 4x scaling on Core i7 (with HT) and 7x on his dual-harpers (according to his website) - which by "my" judgement is nearly linear.


But anyways, I'll let this thread get back to Maxxpi. Sorry I interrupted.
__________________
Reply With Quote
  #14  
Old 05-06-2009, 10:10
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

Quote:
Originally Posted by allegratorial View Post
Maxxpi uses GMP? So that Pi chudnovsky implementation that it uses is merely this?
no, i mean that y-cruncher's characteristics match in wide areas with GMP...

for MaxxPI i don't know this at all, but i don't think so because it will use the gauss algo.

Quote:
Originally Posted by allegratorial View Post
But anyways, I'll let this thread get back to Maxxpi. Sorry I interrupted.
no problem, fine

cu
__________________

Last edited by Tharamis; 05-06-2009 at 16:27.
Reply With Quote
  #15  
Old 05-26-2009, 06:40
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi all,

some little news, now i'm *authorised to post this:

1, MaxxPI² MultiCore ( Pre Alpha ):



screen with an Q6600 at 4100mhz, first 1core below 4cores (scaling)

will support 2,3,4 and 8cores (for now), calculate up to 256M (for now)
chudnovsky used, incl. splitting.

put about >78%! constant load (PMU),
on *all cores, so be carefull. has no CPU-specific optimizations for any
CPU manufaturers. uses mmx/sse

main problem was, load balancing (especially with chudnovsky) and not
to prefer any CPU manufacturer.
this both slow down the calc. speed, but i think at CPU/PC -benchmarking,
this doesn't matter at all, because comparability is the key.

optimized for an major CPU manufacturer a performance
gain from +12% to +18% is possible.

@allegratorial, i know it's important for you:

MaxxPI² MultiCore ( Pre Alpha all x86):
256M, with i7 at 4ghz, with 4cores/4threads (not 8): 473sec.
256M, same machine, with 4cores/4threads (not 8): QPI: 402sec.
256M, same machine, with 4cores/??threads, (not 8??): y-chruncher: 229sec.

2, much more intressing I think, is this:

MaxxMEM²


this will released soon

cu
__________________

Last edited by Tharamis; 05-26-2009 at 08:47.
Reply With Quote
  #16  
Old 05-26-2009, 08:31
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

Quote:
Originally Posted by jmke View Post
MaxxMem is very interesting
yes

Quote:
Originally Posted by jmke View Post
do you think the author could come up with a total score for MaxMem, something like MaxMem-Total = (MemCopy+MemRead+memWrite)*(1/MemLatency)

this in combination maybe with CPu-Z memory tab algorithm detect would allow for auto submit to HWbot.
for now, the *memory score is the arithmetic average between:
"read" and "write", same as the big brother MaxxPI² does.

Memory copy is not part of the memory score, because big MaxxPI²
doesn't use memory copy at all (no need for).

Reaced memory / latency score's will be comparable to MaxxPI².

For further suggestions concerning hwbot, you should contact him directly, via:

http://www.maxxpi.net/pages/contact.php

Regards

Tharamis
__________________

Last edited by Tharamis; 05-26-2009 at 08:51.
Reply With Quote
  #17  
Old 05-28-2009, 19:02
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

little update to v1.40:



* v1.40, application name change, to MaxxPI² - PreView - Single (28/05/2009)
* v1.40, change in OS name detection (28/05/2009)
* v1.40, added 256M calculation option for x64 (28/05/2009) NEW!

cu
__________________
Reply With Quote
  #18  
Old 06-03-2009, 06:52
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

now new!

as a part of MaxxPI²'s memory benchmark, as a preview-version:
MaxxMEM² (Memory/Latency, v1.05):



www.maxxpi.net

cu
__________________
Reply With Quote
  #19  
Old 06-11-2009, 13:12
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

Review/preview of:

upcomming "MaxxPI² - PreView - Multi" (unpublished until now)

Review done by Frank Hempel, at: www.radeon3d.org (review in german)

Direkt link to review: www.radeon3d.org/artikel/sonstiges/maxxpi_multi

cu
__________________

Last edited by Tharamis; 06-11-2009 at 16:05.
Reply With Quote
  #20  
Old 06-12-2009, 19:17
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

here: http://www.maxxpi.net/pages/reviews/rev.php

Review/preview of: "MaxxPI² - PreView - Multi" (english translation)

it scales *very well, on any kind of cpu (cpu/type/manufacturer)

cu
__________________
Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 07:19.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright COLARDYN IT GCV 2004 - 2013