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

Benchmark software Discuss benchmarking software.

Reply
 
Thread Tools
  #1  
Old 05-04-2009, 14:23
TheKarmakazi TheKarmakazi is offline
robo cop
 
Join Date: Feb 2008
Location: Buffalo, NY USA
Posts: 209
Default MaxxPI² - Pi/CPU/System Benchmark

I am going to be beta testing this program soon. I have notified the user about hwbot (they were unaware of the site). It would potentially be a great hwbot integrated (like wprime) program or you can use the online validation they will provide!

The programmers nickname on i4memory.com is "alice" if you want to PM. Or use the contact us form on the website.

Quote:

http://www.maxxpi.net/




features:

MMX / SSE Support
hardware based time measurement to, approx. 1-2ms exactly
arithmetic-deep: 128 m max.
memory usage approx. 1.2gb ram (with 128M)
multithread (not multicore, MaxxPi² will do multicore, not the preview)

for example an 4000mhz intel (+sse2), 6mb Cache will give:
128 m = approx. 14min
32 m = approx. 3min
1 m = approx. 2.4sec.

description of score: K / sec.:

is given to compare easier achieved results,
so instead of having to say 2min 32sec 343ms...
now you need only to tell on single number, k/sec.
it is nothing else than: the number of calculated decimal places per sec. in K (1024)

for example:

1m became 356.2 (K/sec.) means that the cpu was able to
calculate 364,748.8 (1024 x 356.2) decimal places per sec.

AI (Shortest path, Genetic Algorithm / Traveling Salesman Problem TSP):



Memory Reliability:



Image Processing (Blurring):



some statistics:

http://www.maxxpi.net/pages/result-b...statistics.php

and here some comparisons for example:

memory: http://www.maxxpi.net/pages/result-b...0---memory.php

flops: http://www.maxxpi.net/pages/result-b...10---flops.php
__________________

Last edited by TheKarmakazi; 05-04-2009 at 14:28.
Reply With Quote
  #2  
Old 05-04-2009, 15:09
Dualist Dualist is offline
robo cop
 
Join Date: Mar 2007
Posts: 276
Default

Looks very interesting. Should be a good addition
__________________
Quote:
Originally Posted by Massman View Post
... seems like I seriously bunnyed up
Quote:
Originally Posted by Trouffman View Post
Is christian_ney considered as a bug?
Reply With Quote
  #3  
Old 05-04-2009, 17:55
allegratorial allegratorial is offline
maintenance bot
 
Join Date: Jan 2009
Location: UNITED STATES
Posts: 15
Default

Interesting.

Speed-wise it's slower than PiFast, QuickPI, and y-cruncher.

But it has a GUI
__________________

Last edited by allegratorial; 05-08-2009 at 16:00. Reason: at the request of somebody...
Reply With Quote
  #4  
Old 05-04-2009, 18:16
allegratorial allegratorial is offline
maintenance bot
 
Join Date: Jan 2009
Location: UNITED STATES
Posts: 15
Default

huh? how is this multi-threaded?
__________________
Reply With Quote
  #5  
Old 05-04-2009, 18:23
TheKarmakazi TheKarmakazi is offline
robo cop
 
Join Date: Feb 2008
Location: Buffalo, NY USA
Posts: 209
Default

it may not be as fast as some, thats true. But very nice clean interface, online validation and error checking, integration with cpu-z. Multi core / multithread compatible. Vareity of benchmarks and tests run with only one program. Compatible with XP 32/64, Vista 32/64, Win 7 32/64 etc.

Plus its still in development phase so maybe easier to get features implemented!

jmke: I wasnt sure what the requirements were for hwbot integration but I guess this one wont cut it Still looks like a nice app though
__________________
Reply With Quote
  #6  
Old 05-05-2009, 01:10
borandi's Avatar
borandi borandi is offline
robo cop
 
Join Date: Feb 2009
Location: UNITED KINGDOM
Posts: 476
Default

Quote:
Originally Posted by TheKarmakazi View Post
jmke: I wasnt sure what the requirements were for hwbot integration but I guess this one wont cut it Still looks like a nice app though
Get it to have a 1-button test (like wPrime) that just runs a test. But the test should do something other than wPrime (the TSP looks interesting).

Then some form of checksum checking system for online submission, or direct HWBot submission like wPrime. Sorted.

The program looks interesting for sure
__________________
Bro, do you even bench?



Quote:
"who cares about Christian Ney, this is overclocking" - John Lam
ReviewBros
Reply With Quote
  #7  
Old 05-05-2009, 12:18
K404's Avatar
K404 K404 is offline
nVidia FTW!
 
Join Date: Jan 1970
Location: Glasgowvic, Scotistan. Great Kingdom of United Britain..
Posts: 3,230
Default

Quote:
Originally Posted by jmke View Post
requirements for HWBotification:
- one/two settings accessed by easy launch button from the main menu
- a single score as result
- cheat proof or at least have some implementations to prevent it
- checksum available and correct, ability to verify checksum and score
- for best integration: online submission to HWBot's API
Technically that rules out SPi though?

As for speed...the slower the better. We need marathons as well as sprints for benching. SPi 1M is the first bench tried under LN2, lets be honest. We need bigger challenges not smaller ones.
Reply With Quote
  #8  
Old 05-05-2009, 12:39
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

Quote:
Originally Posted by allegratorial View Post
Interesting.

Speed-wise it's absolutely pathetic compared to:

http://www.numberworld.org/y-cruncher/
and it's also a lot slower than PiFast and QuickPi.
But it has a GUI Apparently, the author the y-cruncher program has plans for a GUI but nothing fancy like this.
http://www.numberworld.org/y-crunche...ry.html#Future
it's pretty fast, fast enough for benching *and! comparing.
it doesnt have to be the *fastest.
it's precise, uses CRC and the HW based clock (not winclock) and provides one result (k/sec.)
witch is very easy to compare. it does not need *any additional librarys / installation.

I know him, he specially choosed the gauss algo. because of
its high *and continues (no fluctuation) cpu load.

there are no optimizations for any cpu-manufacturer on board.
all of them will benched with the same non-optimized code.
As he said, this will show the real world, better.

he also has an *incredible fast chudnovsky
algorithm (incl. binary splitting), but this one will not produce
that clean load on the cpu/memory as the gauss do.
You can see that via performance monitoring Unit (PMU-CPU).

so i don't think he will include this into maxxpi.

His MaxxPI² is very professional,
i was one of the first beta-testers on board.

example:
is there an difference between dual/tripple channel on x58?
search the web, you will find nothing. try maxxpi2 and you will see it
(memory <> overall-memory)

this is also very interesting:



and viewing/exporting own results (excel):



cu
__________________

Last edited by Tharamis; 05-05-2009 at 13:27. Reason: forgotten something...
Reply With Quote
  #9  
Old 05-05-2009, 14:46
allegratorial allegratorial is offline
maintenance bot
 
Join Date: Jan 2009
Location: UNITED STATES
Posts: 15
Default

Quote:
Originally Posted by Tharamis View Post

he also has an *incredible fast chudnovsky
algorithm (incl. binary splitting), but this one will not produce
that clean load on the cpu/memory as the gauss do.
You can see that via performance monitoring Unit (PMU-CPU).

so i don't think he will include this into maxxpi.
True, it pretty fast. Here's what the numbers look like at 32M on a friend's 2.66 GHz Harpertowns BSEL to 3.2.

MaxxPi 1.35 - 213.36
PiFast 4.3 - 101.81
QuickPi 4.5 (x64) - 44.51
y-cruncher 0.3.2 (x64 SSE3) - 14.68

Any idea where his Chudnovsky implementation stands? I'm sure the pi-community would like to see it. (since all they seem to care about is speed)


Also, if it's single-threaded (since it clearly is), why would it matter which formula (gauss vs. chudnovsky) is used? Regardless of resource distribution, it would still be 100% cpu over 1 core anyway.


Doesn't look like fast and pretty will ever go together...
__________________
Reply With Quote
  #10  
Old 05-05-2009, 16:17
Tharamis Tharamis is offline
maintenance bot
 
Join Date: May 2009
Posts: 24
Default

hi,

Quote:
Originally Posted by allegratorial View Post
True, it pretty fast. Here's what the numbers look like at 32M on a friend's 2.66 GHz Harpertowns BSEL to 3.2.

MaxxPi 1.35 - 213.36
PiFast 4.3 - 101.81
QuickPi 4.5 (x64) - 44.51
y-cruncher 0.3.2 (x64 SSE3) - 14.68
great results!, here with i7 at 4ghz:

pifast4.3: 65.6 sec
maxxpi1.35: 127.1 sec
superpi: 581.5 sec.

pretty fast... :-)

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.

Quote:
Originally Posted by allegratorial View Post
Any idea where his Chudnovsky implementation stands? I'm sure the pi-community would like to see it.
hmm good question...!?!

Quote:
Originally Posted by allegratorial View Post
(since all they seem to care about is speed)
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.

Quote:
Originally Posted by allegratorial View Post
Also, if it's single-threaded (since it clearly is), why would it matter which formula (gauss vs. chudnovsky) is used? Regardless of resource distribution, it would still be 100% cpu over 1 core anyway.
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.

Quote:
Originally Posted by allegratorial View Post
Doesn't look like fast and pretty will ever go together...
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.

Quote:
Originally Posted by K404 View Post
As for speed...the slower the better. We need marathons as well as sprints for benching. SPi 1M is the first bench tried under LN2, lets be honest. We need bigger challenges not smaller ones.
that's the point!

cu
__________________

Last edited by Tharamis; 05-06-2009 at 10:06. Reason: sry, i must edit.
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 19:49.


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