Intel's Conroe Previewed and Benchmarked

DrFishstik writes "Anandtech has a few preliminary benchmarks on Intel's new Conroe architecture. From the article: 'As far as we could tell, there was nothing fishy going on with the benchmarks or the install. Both systems [AMD 2.8Ghz OC and Conroe] were clean and used the latest versions of all of the drivers.'"

Comments

Shock news.

[supersnail]

Next years Intel chip will run faster than last years AMD chip!

Re:Shock news.

[Anonymous Coward]

Not really. You should have said: "this year's second fastest Intel chip will be way faster than AMD's chip which will be released in June 2006".

Let's look at the facts:
- They benchmarked 2.667GHz Conroe against 2.8GHz Athlon64 FX (FX-60 with 200MHz overclock)
- 2.8GHz Athlon64 FX will be released in June
- 2.667GHz Conroe will be released somewhere in Q3 2006
- Conroe Extreme editition clocked to at least 3.0GHz will be released somewhere in Q3 2006 (there have been rumours about 3.33GHz version)

Based on those benchmarks, fastest Athlon64 FX won't have a chance against 3.0GHz Conroe XE (which will have also faster FSB compared to Conroe benchmarked here), even if you into account that Athlon64 FX will soon support DDR2.

Re:Shock news.

[Saven Marek]

THe simple fact remains that intel needed to do these tests at all, side by side. That's an admission on their part that AMD is beating them and beating them hard. I've heard AMD has some new stuff in the pipeline that'll put conroe out of its misery once and for all.

Given Intel's release date fiasco's it'll probably come out before conroe too.

Re:Shock news.

[supersnail]

Mmmm your right -- but how often do you get a first post oppertunity!
Check the facts and you lose it.

Re:Shock news.

[Ravenscall]

Let's look at the facts:
- They benchmarked 2.667GHz Conroe against 2.8GHz Athlon64 FX (FX-60 with 200MHz overclock)

So they are taking the AMD processor out of spec which can affect performance. Also, the forthcoming AMD processors are a new core architecture and will support faster RAM with an onboard memory controller. I think benchmarks of the final products will be much different. This is the same type of dog and pony show Intel has been doing since they released the Celeron (and possibly before, but that is when I started paying attention to hardware marketing).

- 2.8GHz Athlon64 FX will be released in June
- 2.667GHz Conroe will be released somewhere in Q3 2006
- Conroe Extreme editition clocked to at least 3.0GHz will be released somewhere in Q3 2006 (there have been rumours about 3.33GHz version)

If you think those numbers mean anything, I would like to know what cave you have been living in for the past 3 years.

Re:Shock news.

[timeOday]

Also, the forthcoming AMD processors are a new core architecture and will support faster RAM with an onboard memory controller.

The review did address that, as best they could:

While we're still comparing to Socket-939 and only using RD480, it does seem very unlikely that AMD would be able to make up this much of a deficit with Socket-AM2 and RD580. Especially looking at titles like F.E.A.R. where Conroe's performance advantage averages over 40%, it looks like Intel's confidence has been well placed.

As for your assertion that MHz don't mean anything, that's just wrong. Within a single architecture, speed is nearly proportional to MHz. For a 2.66 GHz Intel to crush a 2.8 GHz AMD so convincingly, does not mean good things for AMD if the Intel can easily reach 3 GHz. It means AMD would have to be at about 3.8 GHz to keep pace: 2.8*(3/2.66)*1.2 = 3.7895 assuming these benchmarks show a 20% lead for Intel.

The real hope for AMD here is that these results won't hold to other benchmarks in general. Apparently this set of benchmarks was handpicked by Intel, so that's almost certainly the case to some degree.

Re:Shock news.

[MSisNOT4Sale]

First off, the benches that intel provided are fishy. Using a motherboard that doesn't even recognize the FX-60? This just stinks of marketing poop and anand just stepped into it. Is there a benchmark of a crossfire/nvidia sli setup using a FX-60 and F.E.A.R.?

Re:Shock news.

[fitten]

The things that AMD has said that they have are F, G, and H revisions of the K8 core (the core that the Athlon64, Turion, Sempron64s, and Opterons are based on) which, other than DDR2 support, not much more information is available. There is another revision called the K8L which will supposedly have 2x the FPU units for about a 50% gain in FPU performance. These will most likely be HPC blade Opterons or some such.

DDR2-800 support, which is the known upgrade, basically adds bandwidth to a chip that isn't bandwidth starved as it is. Current speculation is that the new DDR2-800 Athlon64s will show up to a 10% performance increase on extreme bandwidth benchmarks (synthetics and HPC crunchers, for example).

THe simple fact remains that intel needed to do these tests at all, side by side. That's an admission on their part that AMD is beating them and beating them hard.

Intel has publicly stated (admitted) this already. This demo is to show that the chips they have planned for Q3'06 release (speculation is that they will be delivering machines based on it in July which is the very beginning of Q3, which is only 4 months away) perform well.

By the way, if speculation is that machines will be selling in July, this would imply that the chips are in manufacturing even as we speak. This means that Apple is most likely to announce availability of the new Intel based Power Macs around this time, as well and the various benchmark sites to have their hands on 'pre-production' machines in two to three months tops. We'll be able to see the real story then.

The only announced things from AMD even remotely in this time frame (specifically July and Q3'06) are the AM2 socket for DDR2-800 and a speed bump of the FX-62 to 2.8GHz (which is the equivalent of the overclocked part in the demo). Given that DDR2-800 is expected to be a 10% speed bump at most in most cases and that Conroe will be available at 3GHz (if not higher as rumored - 3.33GHz), I predict (a rather easy prediction to make) that AMD will be playing catch-up for once in the past few years.

Re:Shock news.

[fitten]

Oh... another thing that I forgot was that somewhere in the F, G, and H revisions (and probably the L), HT was supposed to be bumped up to 333MHz (1.333GHz effective) from the current 200MHz (1GHz effective). Given that tests have already shown that 800MHz effective HT performance is statistically equal to 1000MHz effective HT performance, boosting HT speed will probably give a small (1% to 3%) performance increase at best. In actuality, the HT speed increase is required for DDR2-800 to run at its best so the performance gain for it is probably inclusive to any gains shown by DDR2-800 adoption.

Wait and see

[xming]

As pointed out by Ars http://arstechnica.com/news.ars/post/20060307-6334 .html I think we should wait and see for the more objective benchmarks. Anyway 2006 will be a good year for CPUs

Re:Wait and see

[imsabbel]

That ars technica article is denial at its best.
"intel faster? CANNOT BE!!!111"
Sorry, i am as much an AMD fanboy as anybody (hey, their stock financed the car i am driving right now), but besides dual core and adapting sse2/3, VERY little has been done to beef up the aging k8 core (which is byitself also little more than a k7 with on die memory controller).
In a race, standing still will only lead to a loss.

Amd just now is in a position where their flagship is in fact a 7 year old core design, they are one die-shrink behind, and their cache technology is about 4 years behind intel (they need twice as much space per Mbyte cache on the same process size, plus are a factor of 4 slower).

Its time for a _real_ K9 just in the same way intel needed something new after netburst.

Re:Wait and see

[Bert64]

And Intel's new chips are based on the Pentium-M, which is still heavily based on the Pentium-PRO that dates from the early 90s...
Intel's attempt to produce a new architecture (netburst/p4) resulted in an underperforming overheating mess, so they're going back to one that works.

Re:Wait and see

[hattig]

The K8 core is as similar to the K7 core as the conroe core is to a PIII core.

I.e., at first glance there are similarities which can lead to the obvious thought that the K8 core is just a K7 core with memory controller, but actually they're completely revamped, overhauled, enhanced and redone.

I agree that it is time for AMD to get a "K9" out of the door as the K8 as it is won't compete against Intel's offerings unless AMD somehow get 3.6GHz out of 65nm at launch (which is extremely unlikely). Of course, K8L will probably put AMD back into the lead in terms of floating point anyway, but integer is going to be very weak.

Unless AMD is sandbagging - but that's a faint hope for even the most ardent AMD fanboy. I think they miscalculated Intel this time around.

Which of AMD or Intel has the most fangirls?

Re:Wait and see

[mooingyak]

ars technica != Anandtech

Good summary of the Anandtech article though.

Re:Wait and see

[Sketch]

Never mind that the AMD K7 was a carbon copys of the P6 microarchitecture, with incremental tweaks most probably applied to account for P6 shortcomings found in the field. That's an euphemism for `AMD stole Intel's field experience.' The K8 core is only an incremental tweak of K7, the major feature being the on-die memory controller.
So really, AMD can't blame Intel for using P6-derived cores since they're doing the same (not to mention the ethics of stealing a competitor's design). Also, their incremental tweaks aren't really that significant -- process technology changes account for the larger share of performance increase.

I don't think anyone who knows much about CPU's is "blaming" Intel for going back to the P6 core. The P6 was basically the beginning of the modern x86 CPU...more RISC than CISC. As for the K7, this article has a good summary of how it's similar to, but different from, the P6. I think a better euphamism would be 'AMD decided to build a better P6'. (Which is actually what Intel has done as well. This isn't just a faster P6, it's redesign of the P6.) AMD 'stole' from Intel about as much as Intel stole from DEC and Motorola and other RISC CPUs in building the P6, and as much as they had 'stolen' from the CPUs before them. Welcome to the evolution of the CPU, where every CPU is not designed in a vacuum with no relation to anything else, but is built on previous technology and ideas.

Intel tried to raise the bar with the P4 designs, applying some risky design features like hyperpipelined design, and unfortunately the strategy didn't work out all that well, in no small part due to power issues. Moreover they had to endure fanboy cries of `designed by marketing!', but that's the price one pays for exploring new ground in computer architecture. Meanwhile AMD will be content to follow on Intel's successful footsteps as they've always done.

Too bad the superior engineers at Intel weren't smart enough to copy AMD's supposedly "minor tweaks" and bring out a competitive CPU in the last couple of years, and instead chose to stick with their risky design which essentially hit a clock speed wall that they were apparently unable to predict. Surely the geniuses at Intel could have designed a better P6 than that "copy" which was beating their P4 in less than 6 years. Or perhaps they didn't care, and thought that marketing would keep them on top?

Re:Wait and see

[imsabbel]

Well, thats certainly not correct.
The internal differences between p6 and k7 are enormous.
From microops sheduling (k7 using packed microops, in some kind of on the fly VLIW ) to the execution units (fully piplelined and superscalar FPU, for example, compared the non-fully piplelined scalar one), virtually the only thing thats the same is the fact it eats x86 opcode.

But the fact is that the changes between p6 and the new p-m derivates are VERY much larger than the change from 99s k7 to the latest k8.
Just look at a current die-photo of a k8... back in 99, the core transistor count was at the edge of what was possible economically, with l2 cache externally implemented. Nowadays, the nearly unchanged core is just a small lump on the side of the large and not very dense l2 cache-array...

A better competetion

[poeidon1]

With AMD taking the performance lead now and Intel gearing up for getting the top performer position again, I think we are going to see nicer battles now, much nicer than the GHz ones with AMD now much better in its market position and its new fabs.

AMD Processor Model Unknown

[Bert64]

Did anyone else notice that the AMD motherboard didn't detect the processor correctly?

Re:AMD Processor Model Unknown

[Malenfrant]

Yes, exactly. They tested it against a chip which is not listed as being as fast, overclocked so they can pretend it is as fast. Big surprise it didn't perform as well

Re:AMD Processor Model Unknown

[iainl]

The AMD was overclocked to the timings of the one that won't officially be released until June - unsurprisingly, AMD won't let them have a pre-production chip to demonstrate how their one is even faster.

Re:The Conclusion

[mtenhagen]

Well the "new" intels are faster then the "old" amd's. This not suprising Intel has plenty of cash and had to come out with a cpu faster then amd.

The big question will be how will this compare to the next generation of AMD cpu's. And what will the price be. If amd will be faster per dollar the rise of amd will continue.

What about RAM?

[Anonymous Coward]

The AMD system used 1GB of DDR400 running at 2-2-2/1T timings, while the Intel system used 1GB of DDR2-667 running at 4-4-4.

and:

Intel told us to expect an average performance advantage of around 20% across all benchmarks.

Did they really expect around 20% better performance, while using 66% faster RAM? That seems at least unfair to me... Especially the encoding tests, whose results depend heavily on RAM access.

Re:The Conclusion

[Kjella]

by the time this new intel is out, AMD should already be well and truly released. probably also embedding themselves more in Dell's good books and taking more than 80% of the market. Intel are fighting the loosing battle.

1) AMD has something like 20% of the processor market, including OEMs. They couldn't deliver 80% of the market in many years even if the market wanted it.
2) AMD has no major process/architecture shifts between now and Conroe's release.
3) The AMD chip was already overclocked (but then again, they may have gotten a golden sample from Intel).
4) It's losing, not loosing.

TrustedComputing Inside (TM)

[Alsee]

Code-named Conroe... blah blah blah ...feature "security" is expected to be discussed in the framework of a technology that is based on standards set by the Trusted Computing Group and carries the code-name "LaGrande."

Intel's new chips have a Trust Enforcer chip embedded inside the CPU itself. Each chip features a unique serial number, DRM enforcement, Sealed Storage to prohibit you from reading your own files on your hard drive, and Remote Attestation to act as a spy on your computer to log your hardware and what software you run and to securely transmit that spy report to other people over the internet. The chip has your computer's master key locked inside, and you are forbidden to know your master key to control your own computer. Other models of the Trust chip are boobytrapped to self destruct if you attempt to get you key out, and I'd wager these CPUs are boobytrapped to self destruct as well.

Evil as hell.

-

Re:TrustedComputing Inside (TM)

[Jeff DeMaagd]

Maybe it is time to ask that AMD gets out of the Trusted alliance before their chips are like that?

Re:TrustedComputing Inside (TM)

[Kjella]

So... let's take the headcount:

1) Non-geeks who'll buy a new PC not caring and/or approving of whatever "security features" the salesmen told them about.
2) Geeks who'll just bite the bullet and run TCPA/Windows anyway.
3) Geeks who'll be on the TCPA/OS X-x86 platform.
4) Geeks that'll use Linux or turn off TCPA, but will still want new and faster processors.
5) Geeks who won't buy the 'evil as hell' processor.

Oh yeah, Intel is doomed now.

4) is the final nail in the coffin. It's like trying to stop people from buying an iPod which they plan to fill up with their CD collection, because it could also play DRM-protected AACs. That battle is already lost. It only remains to see what content will succeed at DRM, and which will be rejected by the consumers. I'm not too hopeful...

Re:TrustedComputing Inside (TM)

[Anonymous Coward]

There's surprisingly little discussion of this... I remember, about 8 years ago, hearing an Intel engineer talking about how the next step in security was going to be ensuring that a PC was secure against its owner -- along with his updates on such things as encryption from end-to-end with media. I said at the time that what Intel was planning was nothing less than a total lockdown of the previously open PC platform.

And here we are... the final step. With this hardware in a PC, it does not belong to you... you have paid for a car with the bonnet welded shut and no keys.

It's important for everyone to realise just what an enormous amount of control this hardware gives to technology companies... in simple terms: your PC will be nothing more than a set-top box. Technology companies are furiously spinning this as improved security... which is not entirely wrong. This hardware does have security benefits... but as things stand, *YOU*, the person who paid money for the machine, are not in control of it. As others have noted, trusted computing is about them not trusting you -- and not about you trusting your machine. On the contrary, the only thing you can trust is that machines with this hardware are not working for you. Hence the strong link with DRM -- this hardware will enforce DRM on a PC, not to mention allow companies to make any FOSS proprietary (see the discussions about the GPL v3 for examples). Indeed, the TCPA system was designed in conjuction with the RIAA and the MPAA. It's supported by all the technology companies. And don't think that Linux distributors are against it either -- Red Hat is busy working with IBM to produce a TCPA version of Linux... software that cannot be modified by you and continue to work as it did. How about Gstreamer - the media framework used in GNOME? the company behind that has developers who are actively welcoming the introduction of signed Linux kernels (yes, Christian Schaller, I'm talking about you) that will ensure that media is never intercepted and stored... but which will also no longer function if you modify them, or even recompile them yourself. Source code means little in a Trusted Computing world, all that matters is who digitally signed the binary... and this hardware will enforce that. Companies like Red Hat, IBM, Novell, Fluendo etc can all effectively take ownership of FOSS code. Remember: DRM is all about applications. To control data, you must control what applications can access it. DRM is about apps, not data.

You are going to have to fight for your rights on this one. Apple users have rolled over and accepted the introduction of a TPM into the new Intel-based Macs... but then, they can never be relied on to say anything critical of Apple, even when they are being lied to and sold a lemon. They are quite happy to accept this. I would hope the PC crowd is different. Read Professor Ross Anderson's TCPA FAQ. Read Seth Schoen's updates on what Microsoft is planning to do with this hardware -- if that doesn't scare you, nothing will. Join things like the EFF's push to ensure that the hardware you pay good money for works for you, and not Intel/Microsoft and Hollywood. Do not assume that "someone will hack it"... this stuff is not your average dumbass security measure. Educate yourself before its too late and this technology, in its present form, becomes ubiquitous. Support the push to ensure that you, as the owner, have access to the master key... and some method of owner override. Otherwise, in five years, there will be a big brother in every single PC and no way of escaping it.

The problem with buying a new processor

[InsaneLampshade]

This is the problem i have with deciding on which processor i want to buy for my new computer...

I keep reading all these benchmarks, but then i hear afterwards "Oh, if you think that's good, just wait and see what so-and-so is comming out with next year!", so i think, oh, ok, i'll just wait for that then. Then when the new processor gets benchmarked i just hear the same thing over again.

And so... i don't think i'll ever buy a new processor... i'm always waiting for the next version. :(

impressive benchmarks

[bleughbleugh]

yep, impressive, BUT, more important than benchmarks for most Bang per buck by that I mean, if the AMD processor is 50% less than the Intel one, but only 20% slower, AMD will win it for me :-)

Isn't that highly dangerous??

[Namarrgon]

their stock financed the car i am driving right now

Man, there's gotta be some pretty heavy laws about posting on Slashdot while in control of a moving vehicle.

Re:Wait and see (faked tests)

[FirstOne]

Something isn't right, from the screenshot.

..Using an award bios last copyrighted in 2003 for AMD's latest FX-60 chip (2006)..
..Notice how the AMD Processor isn't correctly id'd in the Bios post.
..Even though.. DFI has distributed a new bios version to suport FX60..

.. This thread indicates that there is some video defect in RD480 chipset..

These red flags indicate that something is very fishy and Intel's results should not be trusted... (rigged test)

Your questions answered

[Namarrgon]

the number of concurrent threads

If you're referring to "Hyperthreading", Conroe has none that I'm aware of. One thread at a time, in hardware (whatever you like in software of course).

the power consumption and with that the heat output

Conroe is supposed to have a Thermal Design Power of only 65W. Compare this to the current 3.6GHz P4's TDP of 115W. AMD rate the Athlon FX60's TDP at 110W; however AMD quote the maximum possible thermal dissipation while Intel quotes "typical", usually 75% of maximum (which would make the FX60 about 82W by Intel's reckoning) .

of course the expense of both the processor and the ram it needs

The 2.4GHz and 2.6GHz Conroes are expected to sell for US$316 and US$530 respectively, in 1000-unit quantities (the FX60 was released at US$1031). RAM is harder; reportedly Conroe chipsets will use DDR2, but possibly packaged as new FB-DIMMs. I don't have pricing for those yet, but they'll probably cost more. Consumer motherboards may just use standard DDR2 DIMMs.

Re:Wait and see (faked tests)

[FirstOne]

Corrected link to DFI bios update.. (using ATI's RD480) chipset..

Notice items.. 1, 2, and 10..

Re:Wait and see (faked tests)

[LurkerXXX]

Wait. You are implying a hardware vendor went out of their way to give unrealistically good comparisons of their hardware compared to their competitors by mucking around with the benchmarks setup?

I hope no one tells ATI or Nvidia that this is possible. ;)

History repeats itself. Wait for the real hardware to come out and be benchmarked by independent 3rd parties before getting worked up about how great the new Intel harware is going to be, once it's not vaporware. Nothing new here folks.