This is refreshing for a change. A new architecture where the midrange (E6600) is about the same performance or faster than the previous top of the line (FX-62). Just look at how many different generations of processors were rated at the "3200" mark or only marginally higher. On top of that, it's about $800 less than what the FX-62 was selling for last month!

My upgrade strategy from my 486SX-25 up until my Athlon XP 1600+ was to upgrade to a CPU which was 2X fast as my current one, as long as it cost between $150 - $200. This meant an upgrade about every 2 years, and was always a new architecture each time until it stalled out because of the K8 (both on timing and price), so I ended up with 3 different generations of K7 (Thunderbird, Palomino, and Barton). I'd have to say that my most exciting and longest-lived upgrade was the Palomino, which I might've had longer had it not died a mysterious death after over 3 years of dependable service.

Now that Intel is competitive both performance-wise and price-wise, hopefully AMD will become competitive price-wise. When you look at the big picture, the performance leap is noticable, but not huge over what is currently available, especially considering the disturbing lack of 2MB L2 cache benchmarks. The real story is the down-to-earth price they're charging for a well-designed chip, which puts AMD firmly back into the role of the underdog. Well, that and what I like to refer to as the end of the "Netburst ordeal".

I think I'll have to revise my pricing guideline and go for the E6600. The thought of having a chip with the performance of an FX-62 is just too tempting.

From Kyle at [H]:
We have proven here that the flurry of canned benchmarks based on timedemos showing huge gains with Core 2 processors are virtually worthless in rating the true gaming performance of these processors today.

What a slap in the face to everyone else, including Damage. Are Kyle's claims true? Or does it only prove that people using that specific video card will acheive that performance?

The CPU's are competitively priced, however people should take note that the motherboards for the Conroe costs nearly $300.

This sucks i just got an athlon 64 oh well guess its already time to go back to intel :( ..........I won't let them see me cry

time for every one to come up for air, this is my favorite site, but i do go to {h} every day, i save the best for last and that tech report. but every one is entitled to their opinion, so let kyle have his.

I just read the [H] review. More like [L] for limp. I'm just glad TR does such a good job with this kind of stuff.

70 degrees C??

Aren't these chips supposed to be more power efficient, and the TDP for the X6800 is 75W, is it not? How does this correspond to 70 degrees on a monster HSF running its fan at full blast?? I've only heard of Prescotts reaching those temperatures, and this chip is supposed to have half the heat dissipation.

From the CNPS9500 cooler review:

Zalman has validated the cooler for all of AMD's and Intel's dual-core processors, and we couldn't even get a 3.6GHz Prescott to break 65C.

What kind of prices can we expect on the E6600s when they hit the street? Certainly not $316 :P

So, that's it??? Athlon beats it in few test including some of the more important ones like Cinema4D rendering and lame encoding, in others it's almost right up there. ( I'm ignoring the overcloked fake Conroe part )

Intel clearly stated at least 30% across the board on Core duo 6700, and that chip actually either loses or is barely faster than the athlon.

I told you so.

All AMD has to do is cut prices of those overpiced top parts in half and I'm getting one ( and they will ). Core Duo = even more disappointing than expected.

Does anyone else find it humorous that Dell -- after many years of being a fanatical Intel-only shop -- finally decides to start using AMD CPUs just as AMD cedes the performance crown back to Intel?

Why did you choose to only O/C the least bang/buck processor (6800) over the 6600? If the 6600 ramps up as much as the 6800, you have an absolutely incredible bargain.

I've been trying to think of any scenario by which AMD could possibly respond to this in the next year. Here's the best idea I can come up with--

If you look at the transistor counts of the 4MB Core 2 and the 1 MB (512k per core) X2, you'll notice the Core2 is approximately twice as "big". This suggests that when AMD moves to 65 nm, the X2 will be half the die size of the Core 2.

We also see that with socket AM2, AMD has quite a bit of surplus memory bandwidth.

So, doesn't this mean that AMD could build a quad-core A64 as soon as they make the shrink to 65nm, that would be approximately the same cost to manufacture as Intel's Core 2 Duo? Granted, it's easier to make cache than logic circuits, so maybe a quad core K8 would be somewhat more expensive to make than the dual core Core2, but probably not too much more.

So, for just a little bit more than the price of a Core 2 Duo, could AMD sell an A64 X4? I would think that spending those transistors on extra cores rather than cache would improve performance quite a bit in cases where all four cores can be used.

A review using real cpu heavy games:)

http://www.simhq.com/_technology2/technology_090b.html

OK, I was kidding before. Intel did win. It's obvious. Kitty is indeed just rolling with the punches.

My only gripes:
1. They led reviewers along for a ride. You are their PR machine, none of us can really get these things yet. You may shrug it off, but in a few weeks I bet we're going to see...
2. The point should be made that Intel is NDA'ing this now to get full impact from their processors. If AMD had lowered their prices a week ago, then wouldn't the eval of these things have less impact? Or does Intel's audience solely care about performance at any price?

Intel is the champ, now I think TR should look back every few months and verify availability of these things in retail and OEM packages. (The funfun irony is biting that for me at the end of this article is a pricegrabber link to buy AMD only processors!) If we can't get them, then we've all just played a game and perhaps we've all lost. I hope not, because for all the AMD love I have, I only can purchase intel at work, and these are obviously next up on the plate.

In fact, after seeing the Core 2 in action, many folks may be wondering how AMD is going to keep up. The Athlon 64 X2 4200+ currently lists for more than the Core 2 Duo E6600, and that's just not gonna cut it. Fortunately, AMD has confirmed to us that a major price move is coming in July. We don't have the specifics just yet, but they say they intend to maintain a competitive price-performance ratio. That may mean we'll see the dramatic price cuts rumored to be coming, which would be a good start.

Well, if AMD's 3-year old architectures are the best that AMD can do for the foreseeable future, then I think these kinds of comments might have merit. But honestly, I cannot see that AMD is likely to have too many problems "regaining" the performance crown in the near term, for several reasons that seem pretty good to me:

(1) The 4-mb L2 cache versions of Core2 you tested each have a significant multiple of the L2 cache found in the Athlons you tested. I am sure that in no small measure Core2's performance is intimately tied with this 4mb L2 cache--which is either 2x, 4x or 8x the size of the 90nm Athlon's L2 cache, depending on which Athlon you are comparing. If you had been given a 2mb L2 Core2 sample to test, I have a feeling that it would have been very revealing, and that that's exactly why Intel wants to start shipping its obviously lower-yielding 4mb Core2's first...;)

(2) As you were forced to restrict your Core2 testing to an engineering sample as opposed to a production-grade sample (all the Athlons you tested were production products), this indicates to me that yields may well be an issue moving ahead for the Core2 4-mb L2 variants you tested. We'll see.

(3) As you note, you are comparing Intel's yet-to-ship 65nm products with AMD's older 90nm architectures, so we have yet to see what AMD will do with 65nm (but maybe Dell has already caught a glimpse or two...;))

(4) As you discovered, even though its technology is much older than Core2's, AMD's Athlon still holds a significant advantage in latencies and in available bandwidth, and so there is a whole lot, imo, that AMD can do inside its upcoming 65nm Athlon architectures to better leverage those advantages--whereas, up to now, there's been little reason for AMD to have to make such architectural improvements in response to Intel's Netburst architectures.

Core2 looks very nice in comparison to Netburst, but I will be very surprised if even the 4mb L2 Core2 variants tested here will trouble AMD for long.

What about HT ( Hyper-Threading ) ???? Just curious. I heard that they'll include it later on or something like that? Why not not now? What is the reason for this?

Anyone else feel the pillars of the Earth shaking? I'm pretty new to the computer world, with under 4 years' experience, so this is the first time I've seen the performance crown passed in such a dramatic way. Ever since I started building machines, it was AMD, AMD, AMD, but Intel's just smashed their way into the very top.

As amazing as Core 2, is, let's not forget the lower end of the market, and AMD's (rumored, but likely) upcoming price drops. Depending on how low things falls, AMD X2s, especally the 3800+, 4200+, and 4400+, may take up the post of the average-user CPU. But, that will take a very aggressive price drop on AMD's part. Core 2 does seem to make single-core Athlon 64s, especially low and middle range versions, almost defunct. I don't see this as a complete and total domination of the market by Intel, though it is a very bold move. I see this as forcing AMD to rely on the average end user, not the enthusiast, to make their revenue for a while. That said, they have their work cut out for them; both because AMD is not nearly as widespread as Intel in the consumer's mind, and because AMD will need one bloody fine architecture in K8L to trump Core 2.

Honestly, when the smoke settles, I hope AMD takes the crown back. If my perceptions are correct, they brought precision and well-designed architectures to the CPU world, an alternative to Intel's hack-and-slash, high clock speed approach.

Please correct any errata in what I've said. I'm sure there are a few. ;)

CPU Quake 4 HL2 Ep 1 F.E.A.R. BF2

AMD Athlon 64 FX-62 156.7 170.0 164.0 108.7

Intel Core 2 E6800 192.5 263.5 236.0 142.3

Advantage (Intel) 22.8% 55.0% 43.9% 30.9%

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2795&p=17

In case you missed this Kitty...... :)

Friday night topic: Core 2 Duo. Discuss.

That was a great read, TR.

Finally, it's time to start a pc piggy bank.

Damage, Anand has the pricing for the E6400 and E6300 at $224 and $183 respectively instead of $183 and $163:

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2795&p=2

His numbers are what I have been seeing everywhere else for months. You might want to change the prices you listed.

but right now it looks like there may be no advantge at all.

Advantage (Intel) 22.8% 55.0% 43.9% 30.9%

Nice try Kitty.....best not to deny the obvious tho or no ones gonna reply to you anymore....

I'd hate to say this but the numbers don't match up.

The part that was previewed back in March was supposedly the 6700.

And it was crushing the FX-62.

Yet with TR review. The 6700 isn't putting up those number and the 6800 is just barely quite reaching them.

Granted Intel has a very good CPU here. No one doubts that, but honestly based on those preview number the 6600 should have been where the 6700 is at, the 6700 where the 6800 is at and the 6800 should have had us slack jawed mouths agap in awe.

That didn't happen in my opinion. And that those early preview numbers are fishy.

But if the prices come in where they are supposed to be the 6600 might be the sweet spot. I certainly am not paying $1000 for the 6800. I should have been getting that performance with a lesser chip.

Silly. They are unavailable. Oh well, I'll check this review out again in Jan. ;)

Wow. Nice to see that Intel had a second coup d'etat, with the engineers retaking the company from the market weenies.

Sign me up for some C2D loving. I'll keep my dual Opterons, cause they lift heavy loads like nobody else, but the C2D is overwhelming, unstoppable. I'mll be watching 975X/NF4Intel/NF 590 Intel mobos closely now, and I think there's a C2D on my desk in 2 or 3 months, tops.

Now if Nvidia would just enable SLI on non-NV chipsets, I'd have a 975X and be a very happy d00d.

what happened to those 30% performance gains from the intel black boxes?

all in all, though, very impressive. i don't need a new comp any time soon though, so i might even be able to hold out until K8L and see how things go from there.

Definitely a fast processor but I hope to see some more in-depth reviews come out. Most reviews have games set at 800x640 or 1600x1200 AA/Anis. Easy to see the power of conroe at the lower resolutions like 800x640 and 1024x768 but i'd like to see some 1280x1024 some AA and Anis in there too. Either way its pretty impressive.

Has anyone seen any power consumption figures for the low end Conroe's? I'm interested in upgrading the HTPC under the telly for a unit that consumes less power than my 3500 winnie... at the moment I'm looking at a 479 mobo with a T2300E which seems to be the best bang for power buck at the moment.

Grab an e6300 and OC it and you'll get my attention.