Multi Core Pi v2.000

Toggle sidebar Toggle sidebar
Edgemeal

Edgemeal

Senior member
Dec 8, 2007
211
57
101
Multi Core Pi v2.000 Download Links
http://www.pcgamingxtreme.ro/
http://www.pcgamingxtreme.ro/forum/viewtopic.php?f=6&t=1910

I'm not sure how well this plays out as a multicore benchmark but for anyone interested please post a screen shot including,...
1) Multi Core Pi v2.000 @ default (80K Pi decimals).
2) CPU-Z CPU tab.
3) CPU-Z Memory tab.
----------------------------------------------------------------------

Multi Core Pi v2.000 Scores
Code: 
[B]
   Name             Time       CPU        CPU MHz  Ram MHz  Timings             ScreenShot[/B]
01 [URL="http://forums.anandtech.com/showpost.php?p=34625878&postcount=32"]YBS1[/URL]             00:15.445  3930K      4800.28       -   -                   [URL="http://pics.bbzzdd.com/users/YBS1/Capture5.JPG"]01[/URL]
02 [URL="http://forums.anandtech.com/showpost.php?p=34621085&postcount=16"]skipsneeky2[/URL]      00:18.580  3930K      4000.08       -   -                   [URL="http://i45.tinypic.com/2gxigrl.png"]02[/URL]
03 [URL="http://forums.anandtech.com/showpost.php?p=34623946&postcount=27"]John Tauwhare[/URL]    00:20.493  2700K      5399.90   933.4   11-11-11-28-2T      [URL="http://i1289.photobucket.com/albums/b511/John_Tauwhare/MulticorePIJohnTi7-2700K54_zps28034e61.png"]03[/URL]
04 [URL="http://forums.anandtech.com/showpost.php?p=34620635&postcount=9"]Maximilian[/URL]       00:21.054  3930K            *       -   -                   [URL="http://i.imgur.com/TWb6iFi.png"]04[/URL]
05 [URL="http://forums.anandtech.com/showpost.php?p=34623087&postcount=24"]Klunt Bumskrint[/URL]  00:22.784  3770K      4501.50   800.3   9-9-9-24-2T         [URL="http://i15.photobucket.com/albums/a399/Cokey000/Capture-13.png"]05[/URL]
06 [URL="http://forums.anandtech.com/showpost.php?p=34621611&postcount=20"]SocketF[/URL]          00:26.522  W3520      4309.10   820.8   10-10-10-28-111-1T  [URL="http://img715.imageshack.us/img715/4646/multipi1.png"]06[/URL]
07 [URL="http://forums.anandtech.com/showpost.php?p=34621050&postcount=15"]Idontcare[/URL]        00:27.492  2600K      4000.00   936.6   10-10-10-28-2T      [URL="http://i272.photobucket.com/albums/jj163/idontcare_photo_bucket/Intel%20Core%20i7-2600K/Multi-CorePiver2002600k40GHz.png"]07[/URL]
08 [URL="http://forums.anandtech.com/showpost.php?p=34637407&postcount=42"]ThatsABigOne[/URL]     00:32.332  i7 960     3679.80   800.0   7-8-8-24-88-1T      [URL="http://i.imgur.com/5c5wV62.png"]08[/URL]
09 [URL="http://forums.anandtech.com/showpost.php?p=34620295&postcount=2"]Vectronic[/URL]        00:36.112  3570K      4500.05  1000.0   10-10-10-24-2T      [URL="http://i.imgur.com/fIG1dVc.png"]09[/URL]
10 [URL="http://forums.anandtech.com/showpost.php?p=34625852&postcount=30"]crashtech[/URL]        00:37.392  2 x E5472  3003.63   320.0   5-5-5-15-20         [URL="http://i49.tinypic.com/34s2xdj.jpg"]10[/URL]
11 [URL="http://forums.anandtech.com/showpost.php?p=34623772&postcount=26"]Sable[/URL]            00:38.720  2500K      4700.00   800.1   9-9-9-24-2T         [URL="http://i47.tinypic.com/66gjdz.jpg"]11[/URL]
12 [URL="http://forums.anandtech.com/showpost.php?p=34620175&postcount=1"]Edgemeal[/URL]         00:39.959  3570K      4020.94   670.2   9-9-9-24-1T         [URL="http://img543.imageshack.us/img543/8805/multicorepiv2000.png"]12[/URL]
13 [URL="http://forums.anandtech.com/showpost.php?p=34622730&postcount=23"]YEPP[/URL]             00:49.502  i5 760     4011.00   764.0   9-9-9-24-124-2T     [URL="http://img266.imageshack.us/img266/345/pi2000.jpg"]13[/URL]
14 [URL="http://forums.anandtech.com/showpost.php?p=34637639&postcount=44"]Puppies04[/URL]        01:02.437  740QM            *   664.7   9-9-9-24-107-1T     [URL="http://i226.photobucket.com/albums/dd7/puppies04/Pi_zps4a7d1b3e.png"]14[/URL]
15 [URL="http://forums.anandtech.com/showpost.php?p=34622730&postcount=23"]YEPP[/URL]             01:09.009  Q8400      3200.00   399.6   5-5-5-18-52-2T      [URL="http://img40.imageshack.us/img40/345/pi2000.jpg"]15[/URL]
16 [URL="http://forums.anandtech.com/showpost.php?p=34620635&postcount=9"]Maximilian[/URL]       01:09.809  i5 2400s         *       -   -                   [URL="http://i.imgur.com/cXBAs7Q.png"]16[/URL]
17 [URL="http://forums.anandtech.com/showpost.php?p=34620998&postcount=13"]SPBHM[/URL]            01:10.155  i3 2100          *   665.3   9-9-9-24-1T         [URL="http://tinypic.com/view.php?pic=309hpck&s=6"]17[/URL]
18 [URL="http://forums.anandtech.com/showpost.php?p=34620681&postcount=10"]Scholzpdx[/URL]        01:19.327  Q8200      2912.00   416.0   6-6-6-18-54-2T      [URL="http://i.imgur.com/g3W6j0Q.png"]18[/URL]
19 [URL="http://forums.anandtech.com/showpost.php?p=34637407&postcount=42"]ThatsABigOne[/URL]     04:24.500  Celeron    1397.00   665.3   9-9-9-9-24-1T       [URL="http://i.imgur.com/4mB0BD6.jpg"]19[/URL]
20 [URL="http://forums.anandtech.com/showpost.php?p=34624850&postcount=28"]Prey2big[/URL]         05:17.462  P4         3257.70   200.5   2.5-3-3-7           [URL="http://oi48.tinypic.com/2j5dpa0.jpg"]20[/URL]
21 [URL="http://forums.anandtech.com/showpost.php?p=34620635&postcount=9"]Maximilian[/URL]       06:20.097  SU2300           *       -   -                   [URL="http://i.imgur.com/cOwbpuy.png"]21[/URL]
22 [URL="http://forums.anandtech.com/showpost.php?p=34620593&postcount=5"]Scholzpdx[/URL]        12:46.605  Z530             *       -   5-5-5               [URL="http://i.imgur.com/tVP4Zm4.jpg"]22[/URL]
					       * = Stock speed.
-----------------------------------------------------------------------
multicorepiv2000.png
 
Last edited:
I

Idontcare

Elite Member
Oct 10, 1999
21,110
59
91
I'm game, will post benches later. Just a quick question though, did the author ever address (discuss) any of the reasons why the previous one was so buggy?
 
TurtleCrusher

TurtleCrusher

Lifer
Apr 20, 2008
10,065
984
126
Idontcare said:
>12min! D:
Click to expand...

Hey, I'm just giving you guys a point of reference!

I use this netbook for taking notes and notes only in my writing/psyc/sociology classes.

grimpr said:
Cant wait for the new version 0.6.1, it will support AVX & FMA.
Click to expand...

Wouldn't adding those instruction sets to a benchmark make it awful for a real-life comparison? AVX has nothing to do with gaming and most PC usage scenarios.
 
Last edited:
Maximilian

Maximilian

Lifer
Feb 8, 2004
12,604
15
81
Stock 3930k
TWb6iFi.png


Stock i5 2400s
cXBAs7Q.png


Stock SU2300 (c2d celeron, low voltage thing for small notebooks, runs like a one legged turtle stuck in tar)
cOwbpuy.png
 
Last edited:
TurtleCrusher

TurtleCrusher

Lifer
Apr 20, 2008
10,065
984
126
g3W6j0Q.png


It's actually at 2912Mhz. It probably picks up speed from windows, because no multiplier/bus ratio of mine would equal to a clean 2800Mhz.
 
Last edited:
Vectronic

Vectronic

Senior member
Jan 9, 2013
489
0
0
Off-topic, but I did realize CPU-Z updated thanks to Scholzpdx
1.63:
Intel Pentium & Celeron "Ivy Bridge" CPUs
AMD Athlon X4 750K
FMA3 & 4 instructions sets
Support for systems with more than 64 CPUs
Report TDP in validation pages
Click to expand...

Nothing of note really, though now the instructions list is two lines... :|
 
I

Idontcare

Elite Member
Oct 10, 1999
21,110
59
91
I'm not convinced this guy fixed anything :hmm:

He removed the onscreen printout of the computed digits of Pi, so we no longer have any way of determining if it is still a chaotic number generator like it was in the first version.

And the reported benchmark times are still borked. :|

Here's my FX-8350 turning in 44s 884ms at 3GHz...or you can see it turn in 45s 136ms at 4GHz (yep, it got slower despite being 1GHz faster :confused:)...but at least the 5GHz run was faster than 4GHz by coming in at 44s 982ms at 5GHz (but still slower than the 3GHz bench).

I predict ver300 will do away with the pesky benchmark time altogether and replace it with a chaotic number generator as well...may I suggest RandBetween(Pi(), Pi()^2)?
 
Vectronic

Vectronic

Senior member
Jan 9, 2013
489
0
0
Idontcare said:
I predict ver300 will do away with the pesky benchmark time altogether and replace it with a chaotic number generator as well...may I suggest RandBetween(Pi(), Pi()^2)?
Click to expand...

My speed increased with each run at the same clock, about 100ms faster each time. (stopped after 4 runs), I wonder If I could hit 0 seconds.

Maybe Edgemeal (et al.) can team up with Soulkeeper who is working on the same algorithm... call it Multi-Core Tau.
 
Last edited:
lamedude

lamedude

Golden Member
Jan 14, 2011
1,214
19
81
Scholzpdx said:
Wouldn't adding those instruction sets to a benchmark make it awful for a real-life comparison? AVX has nothing to do with gaming and most PC usage scenarios.
Click to expand...
Except for developers to lazy to set a compiler switch (looks at Bethesda) x87 hasn't been relevant for years but people still use these pi benchmarks.
 
SocketF

SocketF

Senior member
Jun 2, 2006
236
0
71
Idontcare said:
Maybe it just borked for AMD chips?
Click to expand...
Could also be the power-limits. Do you have a HPM Mode? Then switch that on, also some mainboards have an "overcurrent" (or some other name) button in the UEFI BIOS, afaik Bulldozer chips throttle if they are above the limit.
There is a little bit of information in the AMD PDFs, register MSRC001_1028 "di/dt".

My scores:

multipi1.png


The score varies each run around ~1sec.
 
Last edited:
I

Idontcare

Elite Member
Oct 10, 1999
21,110
59
91
SocketF said:
Could also be the power-limits. Do you have a HPM Mode? Then switch that on, also some mainboards have an "overcurrent" (or some other name) button in the UEFI BIOS, afaik Bulldozer chips throttle if they are above the limit.
There is a little bit of information in the AMD PDFs, register MSRC001_1028 "di/dt".

My scores:

multipi1.png


The score varies each run around ~1sec.
Click to expand...

HPM is enabled, also I watch the power consumption for each run and it is very clearly using more power the higher the clockspeed (and other apps scale with clockspeed as expected on the same rig).

It is just this app, which in the prior version could not even calculate Pi correctly let alone return the same computed result twice (the value of Pi kept changing per the program)...the author took away the part where we could at least observe how poorly it calculated Pi so we just have to assume it works correctly now. But the borked benchmark timing is still there.
 
S

Sable

Golden Member
Jan 7, 2006
1,130
105
106
Wait, didn't we do this once and it was concluded the benchmark was pants?
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom