Intel Core 2 Duo E6600 compared directly to an Opteron 165 @ 3GHz

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

bunnyfubbles

Lifer
Sep 3, 2001
12,248
3
0
Originally posted by: RallyMaster
and AMD takes a beatdown....major beatdown. can't even describe how much pwnage was just dealt to AMD.

? The benchmarks are incomplete. Unless you run SuperPi for a living, I don't see any major conclusions here.

And the only game that is complete shows a virtual tie...

1600x1200, 4xAA 16xAF, Everything Maximum
Opteron 165 @ 3GHz: 134.15fps

Conroe @ 3GHz: 138.41fps

The only thing else left that is complete would be:

MPG to XViD using VirtualDub
5min Video of COD2 recorded with FRAPS

Opteron 165 @ 3GHz: 13min 40sec

Conroe @ 3GHz: 10min 1sec

Conroe @ Max Clocks:

Conroe definitely wins, but the "review" is far from complete.
 

compgeek89

Golden Member
Dec 11, 2004
1,860
0
76
The review is NOT complete, Im waiting for a NEW set of intel SLI modified drivers..i was supposed to have the yesterday, im as impatient as you are, and apologize for the delay.
 

HopJokey

Platinum Member
May 6, 2005
2,110
0
0
Originally posted by: the cobbler
Originally posted by: theteamaqua
ok heres what i ahve seen, AM 2 performs better in memory bandwidth in sis sandra(much like what netburst did) , but in other banchmarking tools like Divx, cinebench,3dmark, aquamark, games, they are slower, yes thats right i said slower, and i doubt they can get any faster with the "4x4" platform which can potentially cost $2300 just in cpu/mobo,

what does the FX 62 have over FX 60? besides 200MHz boost and more memory speed b/c of DDR2,

AMD had a good run for the past 3 years, at least from july 23 to somewhere b4 K8L launches, its intel's IMO, im selling my AMD rig on ebay

as for the 65nm vs 90 nm, well Intel always used smaller silicon process compare to AMD, and thats their advantage, so comparing the best products from both company can offer, i think its fair


1) both have been @ 90nm process for years, and Intel was not competitive on that process. wait until AMD 65nm to make that kind of claim, imho.

2) AM2 enjoys whopping jump up in bandwidth (like AMD needed more) over S939. this is why you see AM2 (even S939) still taking the crown at high resolution/AA/AF from Conroe.

Conroe is good stuff for benchies, great big huge E-Penis and so forth. I am currently lmfao at all these people ready to dump their AMD gaming rigs for Conroe which have the potential to be slower still than AMD in those apps at those resolutions.
Dothan was a pretty good (and competitive) 90nm part..

Also at high resolution/AA/AF for games how much impact on fps does memory bandwidth have? Is there a reason to this? Isn't this just a video card bottleneck working here so are you saying that if we remove that bottleneck (say by putting a hypothetical super fast video card like a G80 or R600 next gen part) at high res/AA/AF an AM2 FX-62 at 3.0Ghz would soundly beat a Conroe at 3.0Ghz (or equivalent clock)?
 

TrevorRC

Senior member
Jan 8, 2006
989
0
0
Noticing that the gaming scores are near-identical. That's all I care about.

Looks to me like I'll be sticking with AMD...probably dump my current CPU pre-65nm release, then live without a computer for a week or two. Higher clocks, cheaper cost and so on.

--Trevor
 

BlingBlingArsch

Golden Member
May 10, 2005
1,249
0
0
whats takin u so long to do the game tests, nothing has been added for like 2 days it seems. iam contanstly checkin tha thread and i dont want smalltalk, i want numbers.
 

compgeek89

Golden Member
Dec 11, 2004
1,860
0
76
I am DESPERATELY trying to get ahold of the drivers necessary :(

If anyone has a connection to more recent drivers for SLI on Intel please contact me privately.
 

classy

Lifer
Oct 12, 1999
15,219
1
81
Originally posted by: compgeek89
The review is NOT complete, Im waiting for a NEW set of intel SLI modified drivers..i was supposed to have the yesterday, im as impatient as you are, and apologize for the delay.

Yea I think you need new drivers, because if that one game score is indicative of the others, conroe ain't much of gain at all. And thats not even am2 at that, which is a little faster. So yea get those new drivers ;), lol.


Th encoding and rendering Conroe scores are incredible though.
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
I don't give a flying monkey testicle about the game tests. We know that in gaming the only siginificant limiting factor is the graphics card for 99% of people.

What i want to see is the rest of the transcoding results. (Even then it's hard to make the test fair, optimum would be using the same HD as a second drive in the different machines with a reformat between tests, an utter bitch.) Otherwise you run into trouble with defrag, physical location etc.

Thanks, awaiting the next installation ;)
 

the Chase

Golden Member
Sep 22, 2005
1,403
0
0
Originally posted by: Bobthelost
I don't give a flying monkey testicle about the game tests. We know that in gaming the only siginificant limiting factor is the graphics card for 99% of people.

What i want to see is the rest of the transcoding results. (Even then it's hard to make the test fair, optimum would be using the same HD as a second drive in the different machines with a reformat between tests, an utter bitch.) Otherwise you run into trouble with defrag, physical location etc.

Thanks, awaiting the next installation ;)

Bahhahhahh.:D You should see them fly during mating season.:D
 

DarkKnight69

Golden Member
Jun 15, 2005
1,688
0
76
Originally posted by: Duvie
Originally posted by: keysplayr2003
Originally posted by: Duvie
So I guess you guys got that from synthetic benches so far...

While I dont disagree it may be a huge lead in the end the fact you (and trust me I want to add a comment hear) have commented after synthetic benches really states volumes where this forum has gone.....

I though I was going to read something really comprehensive...You haven't even waited until all the test were done...


disclaimer...I am going to buy the exact model so dont label me as a fanboy....just tired of some of you ppl...

You really are a rude piece of work kiddo.

If this forum has become so distasteful to you, I recommend you go find yourself another one. It's one thing to disagree with someone, but it's quite another to constantly berate and insult other users ESPECIALLY when they are only trying to provide desired information. Just go away and never return.

ALOHA.

Compgeek, nice work so far, and thank you.



i really dont take advice from you!!! I tend to spend more time elsewhere but I will be sure to stay around more to please you!!!


And by the way i AM DEAD RIGHT...the comments made at the beginning of this thread was for 5-6 synthetic marks....Until real test were done that type of intel fanboy "AMD takes a beatdown" only is flamebait....

Duvie, I had thought you had grown up...from what I saw, AMD is taking a BEATDOWN in every aspect so far...Hard to see the truth when the facts are so blatantly obvious, i agree, you should leave and not return!
 

golem

Senior member
Oct 6, 2000
838
3
76
Originally posted by: Xcobra
Originally posted by: Markfw900
AM2 is actually a few % higher by the benchmarks I saw, and you are comparing to that, and its available now, and Conroe still ISN'T available (to buy) so how is this fair ?

i think that was one of the points i failed to point out, its a good point

I don't see this as a good point. Unless you believe the shipping conroe will be substantially different than the one the OP is testing, or that Conroe won't ship within the expected time frame, why does it matter if ships next month? That's not really huge amount of time and the number arent' likely to change much. It's basically a preview.

With the point your trying to push, would it only be fair if the people who had conroe chips do nothing but sit but twiddle their thumbs or just site on their benchmarks until July 23rd or whatever the date is that conroe systems ship?
 

Megatomic

Lifer
Nov 9, 2000
20,127
6
81
I don't think that y'all remember when Duvie was running Intel machines. He's not a fanboy for either side, just his own. As are we all.
 

theteamaqua

Senior member
Jul 12, 2005
314
0
0
Originally posted by: DarkKnight69
Originally posted by: theteamaqua
so wait this article is false??

http://www.hexus.net/content/item.php?item=5692

Sha-zam duvie?

I dont think your fanboism is gonna save you this time, finally they released a better chip...


what??? i thought intel was better, but someone here suggest AM2 can do better b/c of high bandwidth

and why am i a fanboy??

if u actually read couple thread b4 i do own a AMD rig .... and if core 2 duo is better(which seems like the case), i might sell it on ebay and get a conroe rig ....

i dont care who makes it as long as they are the best
 

coldpower27

Golden Member
Jul 18, 2004
1,676
0
76
Originally posted by: theteamaqua
but cobbler seem to think that AM2 is better at higher bandwidth when gaming ....

I thnik DarkKnight was refering to Duvie when he was talking about fanboyism.

There is also nothing to suggest that AM2 high bandwidth will save it against Conroe at high resolutions were the limit is the graphics card and not the CPU.

When your GPU limited Conroe and Windsor will perform the same.
 

DarkKnight69

Golden Member
Jun 15, 2005
1,688
0
76
yea, it was directed at duvie...he always comes back with either

synthetic means sh1t
or
well, amd has this that is better...

He has nothing now...unless amd is gonna surprise the ****** out of us...but i dont see it happenin...

I have been saying it for a while, Intel has something in their P-M, now we see how they build on it!
 

dmens

Platinum Member
Mar 18, 2005
2,275
965
136
Originally posted by: Duvie
Originally posted by: Hard Ball
SuperPI is a cache bound benchmark; and it's not a surprise at all that Core Duos would do better on such a small dataset, where memory bandwidth or latency does not matter at all. Heck, my T60 with T2300 Duo can run in 36sec, significantly faster than a 2.0GHz A64. But that really means squat in real world applications.

The two most worthless benchmarks out are probably SuperPI and linPack FLOPS.

QFT...

Sold your dual dual opteron setup yet???

QFBS on the neverending cachebound superpi myth. HAHAHA. Maybe you can tell me how the conroe-is-only-good-with-4MB-dataset theory explains why woodcrest kicks ass in plenty of server loads.

On the side, the woodcrest is able to cut wallclock by over 60% compared to a 3.8ghz P4 on the sv sims I've been running, and those use over 12GB of memory. Too bad I can't compare with an opteron, but I'm pretty sure woodcrest will wallop the opteron easily on this integer load.
 

theteamaqua

Senior member
Jul 12, 2005
314
0
0
Originally posted by: DarkKnight69
yea, it was directed at duvie...he always comes back with either

synthetic means sh1t
or
well, amd has this that is better...

He has nothing now...unless amd is gonna surprise the ****** out of us...but i dont see it happenin...

I have been saying it for a while, Intel has something in their P-M, now we see how they build on it!

oh ok, cool

hout bout something like shaikira ... w/e, i forgot how to spell his name

the guy who use his blog and show how AMD is superior ... lol


thats what i thought, intel is making come back with core, and the OC ability is just insane lets not talk about in stock speed its faster than FX 62 (at least E6600 or above) in gaming/video encoding (the 2 things i care)

coolaler in XS was able to squeez 3.9Ghz out of E6700 on tower120 heatsink(which stcok speed is 2.66GHz) .... not that i think most people can do it, b/c u will need a good mobo like DFI and also DDR2 1000 since FSB is sky high
 

eLiu

Diamond Member
Jun 4, 2001
6,407
1
0
Originally posted by: Hard Ball
Originally posted by: Hyperlite
29 sec superpi for the 165 @ 3ghz? shoot, my 2ghz core duo can do a 28sec run...

that aside, this is very exciting. Things are really heating up for the first time in a while.

SuperPI is a cache bound benchmark; and it's not a surprise at all that Core Duos would do better on such a small dataset, where memory bandwidth or latency does not matter at all. Heck, my T60 with T2300 Duo can run in 36sec, significantly faster than a 2.0GHz A64. But that really means squat in real world applications.

The two most worthless benchmarks out are probably SuperPI and linPack FLOPS.

They matter if you're goign to be using the box to do FP operations like that... eigen-problems and the like. (Which I am/do).

But yes in general, I agree with you.