The Gigahertz Battle: How Do Today's CPUs Stack Up?

[pschmid]

Core 2 Duo is king, but how do other processors today compare on a clock-to-clock basis? We benchmarked comparable CPUs from AMD as well as Intel to see.

 

[dasickninja]

Does this mean that the CPU charts are getting updated?

 

[Dahak]

Well I guess this proves my point.Even thouugh c2d is a fine cpu,amd 64 is still a heavy weight contender,not to be easily dismissed as inferior.I have faith that AMD will come out with something very nice soon.(By soon I mean in the next few months).

Dahak

AMD X2-4400+@2.6 TOLEDO
EVGA NF4 SLI MB
2X EVGA 7950GT KO IN SLI
4X 512MB CRUCIAL BALLISTIX DDR500
WD300GIG HD/SAMSUNG 250GIG HD
ACER 22IN WIDESCREEN LCD 1600X1200
THERMALTAKE TOUGHPOWER 850WATT PSU
COOLERMASTER MINI R120
3DMARK05 13,471

 

[1Tanker]

Well I guess this proves my point.Even thouugh c2d is a fine cpu,amd 64 is still a heavy weight contender,not to be easily dismissed as inferior.I have faith that AMD will come out with something very nice soon.(By soon I mean in the next few months).

Dahak

AMD X2-4400+@2.6 TOLEDO
EVGA NF4 SLI MB
2X EVGA 7950GT KO IN SLI
4X 512MB CRUCIAL BALLISTIX DDR500
WD300GIG HD/SAMSUNG 250GIG HD
ACER 22IN WIDESCREEN LCD 1600X1200
THERMALTAKE TOUGHPOWER 850WATT PSU
COOLERMASTER MINI R120
3DMARK05 13,471

I didn't really get that from the article. What i saw was that the A64 kicked the Northwood P4's butt pretty good, and the C2D kicked the A64's butt equally as good. 8)

 

[1Tanker]

Does this mean that the CPU charts are getting updated?

They assembled all those CPU's, spent all that time running the tests.....yet they can't? find an E6300, or don't have the time to test one. I think won't is the more appropriate word. :x

 

[gOJDO]

AVERAGE:
C2D vs Core Duo(Yonah) = +15% at same frequency
C2D vs K8 = +20% at same frequency
C2D vs PentiumD = +80% at same frequency

 

[RichPLS]

:roll: methinks the point was different arch running at same frequency produces different performances....

 

[neocristi]

I have my venice 3200+ 939 overclocked to 2400 Mhz for allmost 2 years now.... What a fine CPU it is.

 

[Grimmy]

I still have my P4 northy 3.0 HT CPU.

It still sits, begs, and rolls over on command...

And they say you can't teach an old dog new tricks

 

[Reynod]

I note you compared crippled AMD64 dual cores (half the cache disabled).
How can we trust you when you publish this kind of review?


The 939 choice of CPU should have been the 4800+ (e4 Toledo) not the 4600+ crippled MANCHESTER E6 core you used instead.

You also should have used a 4000+ sandiego core - SSE3 ??? not the earlier 130nm 4000+ core.

The AM2 choice should have been a declocked FX62 down to FX53 speed ... just lower the multilier.

Once again Toms has put AMD at every disadvantage possible in the benchmark comparisons.

But the free rent from the Intel building your located in must be nice.

You always seem to put just enough spin on the articles ...

I remember when the world was bagging Netburst after the A64 was introduced and you sere still showing bent benchmarks here showin the Extreme in front ... man that made me laugh.

Toms will always be known as a site that Intel bought and uses as part of it's longterm marketing strategy ... its just sad that the overage Joe believes what you tell them.

For the rest of us .... well.

 

[dasickninja]

Here we go again. Yet another person comes out of the wood work to claim bias.

 

[Grimmy]

Someone throw that guy a t-shirt to make him feel better.

Here's one:

GENERIC "Stomped" AMD Shirt - Green - OEM

 

[Phrozt]

How much do you think FSB, different memory and different vid cards played into the results??

Not to mention different generations of motherboards offering different efficiencies...

It's an interesting article, but leaves a bit to be desired.

I'm not sure what he means by:

When compared to the Athlon 64 4000+ single core, the AMD processor offers up to 50% more performance, showing that the AMD64 architecture is far superior to Intel's NetBurst.

Could someone explain that to me?

 

[TeraMedia]

I applaud the article for it's goal: demonstrating architectural efficiencies and microcore design

I really would have liked to see a comparison to Pentium Ds. They say they can't run them at 2.4 GHz so they can't do an Apples-to-Apples comparison... but what about Oranges-to-Oranges? Can they compare a Core 2 Duo at 2.66 to a Pentium D 805 at 2.66? I guess I have to check the charts for that. I know the D's will get smoked (if they haven't already burned up), but I wanted to see by how much for future design purposes.

Anyway, thanks for putting this together. Interesting read.

 

[dasickninja]

Is there a "I

AMD" shirt? I think he needs one of those too.

He's complaining about the difference between disabled and native caches. If that isn't nit picking, I don't know what the hell could be. The difference is nominal at best.

Raise a E4300 FSB strap to 266, and drop the multi to 7. Now compare the scores in cache dependent tasks to an E6300.

 

[k61824]

Argh! when will these people do things correctly?

If the Pentium D and P4 has a minimum clock speed of 2.8 GHz, you don't exclude them from the discussion, you raise the clock speeds (Remind you it is called Gigahertz race) for most of the other procs when you can overclock them (which most of them do have the ability to clock at 2.8 GHz), especially when all sorts of liquid coolings and others are available (should normalize the temperatures as well, but that's minor)

also if you have socket 478 P4, where is the socket A /socket 754 AMD stuff?

Yet another point of completion suggestion.

 

[Phrozt]

I've never been a fan of pschmid's reviews... I've said over and over and over and over and over and OVER again that he needs to go back to 6th grade and learn what the terms "Control" and "Variables" mean.

 

[Therlian]

How much do you think FSB, different memory and different vid cards played into the results??

This is something that kind of always bugged me about benchmarks. If the article is about testing CPU's, then I think it would be better to use as similar components as possible with all of the CPU's tested. For example, if they are using ASUS mobos, then have all of them use ASUS, or if they use BFG video cards, then have all of the setups use BFG. Also, try to keep the chipsets the same manufacturer. I'm just saying I think all of other non-reviewed components should be as close to identical as possible.

 

[BaronMatrix]

Core 2 Duo is king, but how do other processors today compare on a clock-to-clock basis? We benchmarked comparable CPUs from AMD as well as Intel to see.

Pretty good review. It shows that the only way to tell C2D and X2 apart is to either crank up your SLI 8000GTX res up to 2560 on Oblivion or actually use a measuring tool.

Cloack for clock it seems much closer than PD to X2, but as you said X2 is still worth it especially with the lower prices.

Flame On!

 

[srgess]

Something is weird in the benchmark with winrar, i took the same amount of file and same total mb. I did the compression to Best etc and i was at 3 min compression and my proc is at 2.4ghz e6300 and in their benchmark its like 1 min. How can this be possible ?

 

[ajfink]

Here we go again. Yet another person comes out of the wood work to claim bias.

He did make a few small, good points. They probably just used whatever CPUs were on hand, and I can't blame them for that, because this whole article reeks of "Hey, guys, we need to do review SOMETHING. What have we got lying around?"

 

[bkiserx7]

I wish they would have included my 2.4 Ghz Athlon XP-M

 

[bkiserx7]

I note you compared crippled AMD64 dual cores (half the cache disabled).
How can we trust you when you publish this kind of review?


The 939 choice of CPU should have been the 4800+ (e4 Toledo) not the 4600+ crippled MANCHESTER E6 core you used instead.

You also should have used a 4000+ sandiego core - SSE3 ??? not the earlier 130nm 4000+ core.

The AM2 choice should have been a declocked FX62 down to FX53 speed ... just lower the multilier.

Once again Toms has put AMD at every disadvantage possible in the benchmark comparisons.

But the free rent from the Intel building your located in must be nice.

You always seem to put just enough spin on the articles ...

I remember when the world was bagging Netburst after the A64 was introduced and you sere still showing bent benchmarks here showin the Extreme in front ... man that made me laugh.

Toms will always be known as a site that Intel bought and uses as part of it's longterm marketing strategy ... its just sad that the overage Joe believes what you tell them.

For the rest of us .... well.

I agree

 

[lca806]

Why were there no math / computational benchmarks done? Specifically the sandra 2007 floating points and integer benchmarks. I know I am a minority in that I care about floating point math computations and this site mainly caters to the designers and those who focus more on rendering. So... oh well I can wait for the charts to get updated!

Until then, I play the wait and see game.

 

[pmr]

Something is weird in the benchmark with winrar, i took the same amount of file and same total mb. I did the compression to Best etc and i was at 3 min compression and my proc is at 2.4ghz e6300 and in their benchmark its like 1 min. How can this be possible ?

(303 MB, 47 Files, 2 Folders)

Did you split 303Mb in 47 files and two folders?

Also as an example, video or jpeg pics take more time to compress than mp3 music