Here's a little teaser....

[Viper666]

this should post with a quad fx CPU .correct?Anyone got an Idea?ASUS KFSN4-DRE I have nothing else to test it with

[Viper666]

The focus should be moved from "100% scaling" vs. "122% scaling" (surely meant ironically) to something like 60% vs. 80%. Then all people here should be happy As we all know, one type of CPU in a multi CPU setup is hampered more by FSB, inter core traffic and mem BW than another type of CPU with its direct connection links, a cache shared by 4 cores and so on..

Will a car run twice as fast by doubling the number of cylinders?

AMD cars do LMAO

[Donster]

Ive been away from building pcs for a while now .... how comes theres no 3d benchmark results for barca ?

[BeardyMan]

Stephen are you ill or so? or on to something? no updates ?

[PhilDoc]

Ive been away from building pcs for a while now .... how comes theres no 3d benchmark results for barca ?

There are some in here, you just have to look through the thread a little. Go back about a week or so. Also, if you go to the news section there's an old thread about 3dmark and the Barcalona

[KTE]

Intel Tigerton and AMD Barcelona 16-core SPECint_rate 2006 (1) and SPECfp_rate 2006 (2) results:

(1)
Tigerton 2.93GHz (peak/base) 211/184
Barcelona 2.5GHz (peak/base) 188/163

(2)
Tigerton 2.93GHz (peak/base) 113/106
Barcelona 2.5GHz (peak/base) 160/149

Both CPUs unavailable yet.

Latest Zilla Fud
K10 to cost less than Core 2 Quad
Agena FX & X4 from 2.2-2.8GHz 2007 and higher in Q1 08

[VulgarHandle]

Intel Tigerton and AMD Barcelona 16-core SPECint_rate 2006 (1) and SPECfp_rate 2006 (2) results:

(1)
Tigerton 2.93GHz (peak/base) 211/184
Barcelona 2.5GHz (peak/base) 188/163

(2)
Tigerton 2.93GHz (peak/base) 113/106
Barcelona 2.5GHz (peak/base) 160/149

Both CPUs unavailable yet.

Latest Zilla Fud
K10 to cost less than Core 2 Quad
Agena FX & X4 from 2.2-2.8GHz 2007 and higher in Q1 08

wow, on par on the integer (based on clock speed), and smashes intel on fp with lower clock (expected of course)

[JohannesRS]

Nice numbers. Am I wrong or they refer to 4-way systems?

[s7e9h3n]

doesnt work in ASUS KFSN4-DRE 1.7 BA's no post

There's an issue with the memory. If you can, get ahold of some Corsair Registered Ecc....the board should post with those sticks installed.

Stephen are you ill or so? or on to something? no updates ?

I'm fine, just a bit busy. I'll be updating in soon. Just shuffling around some cpu's.

Intel Tigerton and AMD Barcelona 16-core SPECint_rate 2006 (1) and SPECfp_rate 2006 (2) results:

(1)
Tigerton 2.93GHz (peak/base) 211/184
Barcelona 2.5GHz (peak/base) 188/163

(2)
Tigerton 2.93GHz (peak/base) 113/106
Barcelona 2.5GHz (peak/base) 160/149

Both CPUs unavailable yet.

I'm expecting b2's in the VERY near future

[BeardyMan]

any new motherboards as well? where you can test some good old D9GHM on it?

[JohannesRS]

From what all can get so, no news still from all bugs front?

@Dave: Dave, any news from your "cold and long" testing with loads of processors, motherboards, bioses and memories? I think many people is long awaiting for those...

[Viper666]

There's an issue with the memory. If you can, get ahold of some Corsair Registered Ecc....the board should post with those sticks installed.



I'm fine, just a bit busy. I'll be updating in soon. Just shuffling around some cpu's.



I'm expecting b2's in the VERY near future

Thanks S7, Asus is sending me the 1004 bios,I hope it will work with the memory I have even though I dont like what I got its Kingston, Ihave RMAed and spent a ton of money on shipping cost alone over the past lil bit here.Do you think 1004 will make it work? this memory worked fine in the Quad FX

The AMD Guy I talked to told me wait for B2

[xPliziT]

Hi there,

could anyone with a barcy eventually run this easy bench?
It seems to favor AMD CPU in general, so it would be interesting to see how it performs on the new opterons.

Download it here:
http://home.istar.ca/~lyster/pi.html

run the program on comand line window.
Enter the follwing:
qpi 10mi pi.txt


My Results
QPI-QuickPi v4.0a, (c) 2000-2006 S. Pagliarulo
Freely distributable, email: s_pagliarulo@hotmail.com

o AMD Athlon(tm) 64 X2 Dual Core Processor 6400+ detected
o Processor speed measured at 3.50 GHz
o Single processor with dual cores
o 1.3 GB of memory available
o Using default training data

Computation of Pi to 10,000,000 digits
Method used : Chudnovsky
Started : Sun Oct 21 20:50:13 2007

Series size : 705137 (10,000,004 digits)
Series processing time : 16.56
Final value time : 2.41

Total time : 18.97 seconds
Total memory used : 123,614,436 (117.89 MB)
Processor utilization : 120.68%

Pi = 3.
1415926535 8979323846 2643383279 5028841971 6939937510 : 50
5820974944 5923078164 0628620899 8628034825 3421170679 : 100
8214808651 3282306647 0938446095 5058223172 5359408128 : 150
4811174502 8410270193 8521105559 6446229489 5493038196 : 200
4428810975 6659334461 2847564823 3786783165 2712019091 : 250
4564856692 3460348610 4543266482 1339360726 0249141273 : 300
7245870066 0631558817 4881520920 9628292540 9171536436 : 350
7892590360 0113305305 4882046652 1384146951 9415116094 : 400
3305727036 5759591953 0921861173 8193261179 3105118548 : 450
0744623799 6274956735 1885752724 8912279381 8301194912 : 500


and please try this proggy too:
PI Fast
http://numbers.computation.free.fr/C.../download.html

My results:

Program : PiFast version 4.3 (fix 1), by Xavier Gourdon
Computation of 10000000 digits of Pi
Method used : Chudnovsky
Size of FFT : 1024 K
Physical memory used : ~ 61355 K
Disk memory used : ~ 0.00 Meg
------------------------------------------------------------
Computation run information :

Start : Sun Oct 21 21:28:48 2007
End : Sun Oct 21 21:29:14 2007
Duration : 25.98 seconds
================================================== ==========
Total computation time : 25.98 seconds (~ 0.01 hours)
================================================== ==========
Pi with 10000000 digits :
Pi = 3.
1415926535 8979323846 2643383279 5028841971 6939937510 : 50
5820974944 5923078164 0628620899 8628034825 3421170679 : 100
8214808651 3282306647 0938446095 5058223172 5359408128 : 150
4811174502 8410270193 8521105559 6446229489 5493038196 : 200
4428810975 6659334461 2847564823 3786783165 2712019091 : 250
4564856692 3460348610 4543266482 1339360726 0249141273 : 300
7245870066 0631558817 4881520920 9628292540 9171536436 : 350
7892590360 0113305305 4882046652 1384146951 9415116094 : 400
3305727036 5759591953 0921861173 8193261179 3105118548 : 450
0744623799 6274956735 1885752724 8912279381 8301194912 : 500

use the standard/fastest options in the dialogs when PIFast was started and enter 10000000 (10mi), for the output question choose 0 instead of 1.

all results were taken under "normal" desktop enviroment, e.g. lots of firefox tabs and other stuff running in the background

[Opteron146]

... this memory worked fine in the Quad FX

Do you mean the L1N64 from Asus ? Then no wonder that nothing is working. You have unbuffered RAM, which wont work in KFSN4-DRE. Get yourself some nice registered RAMs. There are new DDR2-800 ones from Kingston around. Type No. is: KVR800D2D8P5/1G.

@xPliziT:
If your programs runs very well on recent Athlon64s and not that well on Intel chips, then I assume that it is due to heavy use of FPU code instead of SSE1/2/3. As there is no FPU improvement in Barcelona your code might run slightly better because of the bigger caches, and better prefetchers, but performance will be hurt badly because of Barcelona's current clock disadvatage.

cheers & good luck

Alex

[Viper666]

Do you mean the L1N64 from Asus ? Then no wonder that nothing is working. You have unbuffered RAM, which wont work in KFSN4-DRE. Get yourself some nice registered RAMs. There are new DDR2-800 ones from Kingston around. Type No. is: KVR800D2D8P5/1G.

@xPliziT:
If your programs runs very well on recent Athlon64s and not that well on Intel chips, then I assume that it is due to heavy use of FPU code instead of SSE1/2/3. As there is no FPU improvement in Barcelona your code might run slightly better because of the bigger caches, and better prefetchers, but performance will be hurt badly because of Barcelona's current clock disadvatage.

cheers & good luck

Alex

I have REG RAM dude and yes it works in the Quad FX ,They Are Opterons ! DDR2 667 REG ECC ,Im pretty sure all ASUS boards can use REG ECC

[LowRun]

Do you mean the L1N64 from Asus ? Then no wonder that nothing is working. You have unbuffered RAM, which wont work in KFSN4-DRE. Get yourself some nice registered RAMs. There are new DDR2-800 ones from Kingston around. Type No. is: KVR800D2D8P5/1G.

@xPliziT:
If your programs runs very well on recent Athlon64s and not that well on Intel chips, then I assume that it is due to heavy use of FPU code instead of SSE1/2/3. As there is no FPU improvement in Barcelona your code might run slightly better because of the bigger caches, and better prefetchers, but performance will be hurt badly because of Barcelona's current clock disadvatage.

cheers & good luck

Alex

Whoaw, three years and some before your first post

See you in 2010

[MAS]

he's very timid

[SOLDNER-MOFO64]

Whoaw, three years and some before your first post

See you in 2010

[alayashu]

guys don't push him to break his tradition:p

any news on the B2's and a good BIOS, Stephen?
seeing the activity in this forum i guess i was right,
we'll see Phenom + RD790 mobos by the time the
BIOS et all problems are fixed:P

[KTE]

Yeah Stephen, I have one question for you:

Whats the correct spelling of your name?

OK, the real qs is, any news of B2s or Phenoms?

[JohannesRS]

Hey, the best barcellona topic of the world has just suddenly DIED??????

[s7e9h3n]

Hey, the best barcellona topic of the world has just suddenly DIED??????

Well, the same can be said for my Tyan board Time for a replacement..

[THE JEW (RaVeN)]

Well, the same can be said for my Tyan board Time for a replacement..

anandtech's mobo and now yours? I realize it's a small sample size, but hopefully this isn't a common occurrence.

[Lightman]

anandtech's mobo and now yours? I realize it's a small sample size, but hopefully this isn't a common occurrence.

I think he helped his Tyan die .....

[Ace123]

|
v