Please help!!! Benchmarks... (WinXP)!

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
A few colleagues of mine are a little skeptical about the Core 2 Duo.
Click to expand...

Your colleagues obviously don't know much about computers...sry

they've come to like A64s
Click to expand...

Your colleagues must be cavemen!
I've told them that C2Ds are the way to go right now in almost all cases.
Click to expand...
Um, don't you mean ALL CASES...the $320 e6600 kills the $740 fx-62 (price to performance and in over 90% of benchies)....and has the ability to be overclocked like mad. You can get an e6600 at 2.4 up to 3.4 with ease...and beyond that with simple voltage tuning.

Tell your friends to do some research about the c2d...or simply check out this link...
 
Thanks for all the replies, everyone!!!

ElMoIsEviL: My god, man! 9 seconds??? 4.7Ghz? Congratuf*ckinglations 8O 8O 8O

13thmonkey: That's an unusual result... 30.5s is around the E6300s performance... I don't really get it... Well, I'll keep my eyes open for other results and I'll take the average...

labbbby: Thanks. Posting here is always interesting! Also, there must be a lot of stuff running in the background of your IBM T43... That's one of the main advantages of dual-core: the crap can go one way (read: core) while the interesting stuff goes another. 8)

Avalanche: Your results are more like the ones from three0duster now... That background stuff was indeed slowing you down. Oh! And I'm curious as to the results for that A64 X2 @ 2.9Ghz!!! Thanks!!

Grimmy: Hmm, relatively bad result for my program (if compared to 3.2Ghz prescott) - your 3.0C is probably better than that on average code. Must be from compiler optimizations for Prescott... My colleagues insisted that I use the Intel Fortran Compiler (9.1)...

DaSickNinja: A plane? A Bird? No, a 4Ghz C2X... Bad joke, I know... but what else can I say?... Ah, I know, I could say thanks again for that Xeon 3060 result. It might be of interest to my friends!! 😀

Psychoholic: Hmmm, great result - considering how easy it is to get a C2D to 3.0Ghz, this is beginning to look more and more like the ideal cost-efficent solution! Thanks!

jaydeejohnHmmm, JD, your Sandy doesn't quite measure up... It's obviously Intel's fault. (using Intel Compiler...) The thing is that Intel's compiler is readily and freely available to all interested (for linux), which has tilted this benchmark towards using Intel's compilers...

Grizzman Wow, for this particular program, your computer isn't much slower than a X6800! Quite good...

And finally, three0duster, this code was compiled using Intel's compilers... And they use a lot of vectorization, as far as I know, which might be giving better results for SSE2-optimized CPUs...

A request: is there any way any of you could run this on a stock system? I mean, a E6400/6600/6700? I'd like to get a few more results on the C2D lineup... Thanks!
 
Your colleagues obviously don't know much about computers...sry
Click to expand...

No they don't! In the words of Dr McCoy, "Damnit, (we're) physicists, not programmers!" 8O

I think that would probably be their answer to your statement! :lol:

Tell your friends to do some research about the c2d...
Click to expand...

Actually, I would, but they probably don't have the time to properly deal with this issue. It's rather sad, because that causes some bizarre situations. For instance, I've seen a 3.0Ghz Prescott inside a case without any fans (only the CPU heatsink+fan). Also, I've seen cases with holes for 80, 92 and 120mm fans with 3.2Ghz Prescotts used for 100% CPU usage processing with only an 80mm fan, and so on. Alas, this last 3.2Ghz Prescott had 4GB system memory, a rather good Intel board, but a lousy sub-US$30 or so PSU!!

That's exactly why I'm giving a helping hand... To avoid those very sad situations... :wink:
 
Thanks for all the replies, everyone!!!

13thmonkey: That's an unusual result... 30.5s is around the E6300s performance... I don't really get it... Well, I'll keep my eyes open for other results and I'll take the average...
Click to expand...

Mine is stock, but its using DDR, and this might actually be a case where it makes more than 5% difference. So discount mine for averaging purposes.

Its all an argument for me getting a proper board and some decent RAM but waiting for the price to come down and see what the nforce 6xx boards are like. Will have to wait till the gf is away so I can do it nice and quick so she doesn't notice...
 
Ok....137 seconds on a intel P4 1.7GH(Williamete) with 512 DDR

For a while there i started to worry it will ever finish the test :) . What did you do with that code? Programed it to find the answer for the ultimate question about life and every thing???
 
RE:Bmark2..
30.09375 AMD Sandy 3700 mildly oc'd/OCZ 3500..
???? Did we win something???:)
no optimization lots a' junk in background..
 
13thmonkey: Yeah, well, the performance seems to depend quite a bit on the memory quality (latencies, and so on) so you're probably right.

PSYCHoHoLiC: That's a good stock result. I'm thinking your memory probably helps you... Thanks.

sareiodata: Nothing quite so complicated... In any case, the answer to that one we already know: it should be 42. But to what question?... 😛 Oh, and interestingly enough, you have nearly exaclty (1% deviation) the same performance as my colleague's current machine, an AXP 1700. Quite appropriate!! 😀

Nitro350Z: Hey, that's quite adequate performance!! It's probably because of SSE2 and Intel optimizations. Congratulations!

dokk: Hmm your result is indeed quite good! Mildly OCed to what speed?...
 
Is is my result

P4 Intel Celeron D320 2.4GHz FSB 533 Prescott core
2x512MB Geil Value Ram
ASUS P4V8X-X

Total time - 43.79688s
Process - 43.76562s

Not bad 😳 [/img]
 
Thanks for all the results!

Look at that rather long table!!! 8)

table01.gif


I'm still missing the stock C2D results though... :cry:
 
Yep, mine is well off, if memory bandwidth is the issue than I'd expect the OC's to scale really well, the triplet of x6800 results do just that, i.e. just under a doubling of clock speed (FSB) is halving the result.
 
P4 Intel Celeron D320 2.4GHz FSB 533 Prescott core
2x512MB Geil Value Ram
ASUS P4V8X-X

Total time - 43.79688s
Process - 43.76562s
Click to expand...

Hmmm? 8O

Something's not right.

How come that 3.0Ghz Northwood does 47s, but your 2.4Ghz Celeron does... 43.8s? And what about those two 3.2Ghz prescott results, which give ~25-26s? I don't quite get this. Those prescotts were like 80% faster than the Northwood! Kinda makes me think there were a lot of background processes on that 3.0Ghz Northwood, since we have two distinct 3.2Ghz P4 Prescott results... That would make much more sense... and even more so if you think that a 2.53Ghz P4 bests that 3.0C's result...

Very strange, but I'm keeping track of all the averages.

Still missing good C2D stock results for E6400/E6600/E6700... :cry:
 
forgot to mention, and noticed that I didnt include this in my sig, my northwood is a 2.2ghz oc'd to 2.53(rdram wont let me push it farther without loosing stability).
 
My stats for my hardware were
X6600 Core2Extreme @ 2.4GHz (stock) = 23.62s
X6600 OC'ed to 3.02GHz = 18.03s


After killing Mcafee 2006
X6600 OC'ed to 3.02GHz = 16.11s
 
evilr00t: Don't really know. I do have access to another 3.0Ghz P4 Prescott which I'll benchmark tomorrow, along with an A64 3000 and whatever else I can put my hands on... The more the better.

Nitro350Z: I see! In any case, this rig of yours performs quite well considering its age, which is probably a testament to how well-thought-out that purchase was.

suzukii: Thanks! You gave me one of the numbers I was looking for!! Now I'm only missing E6400 and E6700 scores...

Anyone that has a system similar to the ones already benched here, please run this anyway if at all possible! I'm considering averages if possible!! [greed] More! I need more!!! Much more!!! [/greed]

Interesting: If compared to the reference AXP 1700 rig, performance per megahertz is always around 2.4x reference for A64s, and always around a nasty 3.5x-3.6x reference for the C2Ds. This basically means that a typical C2D performs +50% per clock on this particular benchmark if compared to a typical A64s. Quite impressive indeed!!! :twisted:
 
New score for X6800 @ 3.9GHz - 12.5 sec.
New score for Xeon 3060 @ stock - 18.32 sec
Score for E6400 @ stock - 19.75 sec
Score for Turion X2 @ 2.0GHz - 23.5 sec
Score for Core 2 T2700 @ 2.2 GHz - 21.2 sec
Score for E6300 @ stock - 22.65 sec
Score for Celeron D (Prescott) @ 3.06GHz - 35 sec
Score for Pentium 4 @3.8 GHz - 33.2 sec
Score for AMD FX 55 - 29.99 sec
Score for Pentium 2 @266 Mhz - 59.6 sec
Score for Dual Woodcrest 5160 @ 3.0GHz - 17.8 sec

Hope this helps.
Ninja
 
Grimmy: Hmm, relatively bad result for my program (if compared to 3.2Ghz prescott) - your 3.0C is probably better than that on average code. Must be from compiler optimizations for Prescott... My colleagues insisted that I use the Intel Fortran Compiler (9.1)...
Click to expand...

Well it might be just my setup. Although, I don't have hardly any running in the background. Perhaps my MB chipset may have something to do with it. I'm not/can't even running dual channel memory. So I'm on bascially the old i845PE chipset that has min support for the 800mhz fsb.

Perhaps someone on a P4 3.0/800 with full support would get a better time then me.
 
Wow, the "cost effective" solution of a 6300 @ stock still spanks an X2 3800 heavily OC'd. When you then factor in the easy 3.0 ghz OC on the 6300, it absolutely destroys any AMD (and I am coming from AMD, now on C2D).

I guess your buddies have found the proof they were looking for........C2D rocks the AMD's in every situation......but most importantly where it counts...in price per performance.

Wow, looking at the charts, I'm ecstatic with my 6300. :)
 
Wow, the "cost effective" solution of a 6300 @ stock still spanks an X2 3800 heavily OC'd. When you then factor in the easy 3.0 ghz OC on the 6300, it absolutely destroys any AMD (and I am coming from AMD, now on C2D).

I guess your buddies have found the proof they were looking for........C2D rocks the AMD's in every situation......but most importantly where it counts...in price per performance.

Wow, looking at the charts, I'm ecstatic with my 6300. :)
Click to expand...

Well you cant go outright and say that, This little benchmark program was compiled using Intel optimized code, so the results are a little skewed towards Intel products. Im sure if you were to code it so that it utilized two cores and wasn't so intel biased, AMD could hold a candle, but Im not saying that C2d are not more superior, Just saying dont kill all of amd from this one piece of binary
 
RAM timings –> Cas latency – 2CL, RAS to CAS Delay - 3CL, RAS Precharge- 3CL, Clcle time – 6CL

A) Celeron @ 1700 MHz, RAM - 134 MHz -> 87.8sec
B) Celeron OCed to 1920 MHz, RAM - 113 MHz -> 102 sec
C) Celeron OCed to 1853 MHz, RAM - 142 MHz -> 48 sec
D) Celeron OCed to 1818 MHz, RAM – 140 MHz ->82 sec

Why the huge difference between C and D?
I set the system twice to C &got the same result i.e. around 47-48 sec
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom