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.

[eshefer]

yup this is what I thought too.. but..

the AMD system was overclocked, and it's confirmed that AMD will not move to a 65nm process untill 2007, so the comparison has a chance of being apt. that is the competition in six months might be between very similar system from AMD and Intel.

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.

[Anonymous Coward]

And this Intel chip comes with added hardware DRM built into it.

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.

[Anonymous Coward]

One thing all sides can be sure of is that this will be the biggest flame war in years.

Re:Shock news.

[somersault]

well at least AMD has started taking a lot of market share from Intel, and Intel has been rattled - now we're going to have more healthy competition along the lines of ATI/nVidia , rather than Intel thinking it can do whatever it wants and people will still buy its products (*cough*likeMicrosoft*cough*)

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.

[Anonymous Coward]

They benchmarked against a Athlon 64 X2, not Athlon 64 FX-60....

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.

[GuyverDH]

As far as Mhz semi-equating to speed...

Doesn't anyone else recall Intel releasing a faster processor, yet having to have the CPU idle for half of them to keep it from melting down?

Seems to me that we can speculate all you want, yet, in the end, only final numbers will be able to show what's what....

At this point it's all a big phallic comparison, and everyone who jumps on board swinging their own extensions are just blowing smoke up everyone's arses.

Re:Shock news.

[DaFrogBoy]

The article compared a AMD Athlon 64 X2 *NOT* a FX.

FX is known to be better in gaming than the X2.

Re:Shock news.

[DaFrogBoy]

Corrrection:

Oddly enough they refer to comparing it with an overclocked FX. However, in their benchmark graphs, they refer to it being compared to an X2. I think the article was trying to confuse us!

Re:Shock news.

[stunt_penguin]

OK so the Intel chip will be faster......... but how much will it cost? There's no point in it being 30% faster if it's three times as expensive. We'll just have to wait and see.

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.

[ruiner5000]

Sure, except they used the old ATI RD480 and SB450 chipset combo with bad Southbridge performance, and only 16X of PCI E lanes for the graphics. Had this been a real test the Asus A8N32-SLI Deluxe would have been used. Also get your facts straight. The next FX is Socket AM2, not 939. Just overclocking the FX60 is not going to work. Since you clearly have not been paying attention then you don't realize that DDR2 performance on AM2 is superior to what Intel has gotten.

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.

Re:Shock news.

[fitten]

Bah, my math and numbers aren't right in the above but I still hold to the conclusions. There is supposed to be a speed bump of HT but I have the numbers wrong reflecting what the change is.

Re:Shock news.

[Anonymous Coward]

The FX-60 is dual core (basically an X2) and was released a few weeks ago.

Re:Shock news.

[Atmchicago]

Good call, especially when you are a snail and already unfairly disadvantaged. I say Slashdot gives priority to all users who are snails!

Re:Shock news.

[Overly Critical Guy]

I love it. Intel even has side-by-side benchmarks showing their chips have surpassed AMD's, and someone on Slashdot will find some way to make it a victory for AMD.

Conroe's not going anywhere. AMD hasn't even gone to 65nm yet!

Re:Shock news.

[NatteringNabob]

[ even if you into account that Athlon64 FX will soon support DDR2 ] How about when you take into account that the Intel chip is 65nm and the AMD is 90nm? The next AMD die shrink will likely take care of most of the performance difference. In the meantime, in the battle between products that you can't actually buy, Intel appears to have a lead on their prefered benchmarks.

Re:Shock news.

[Guspaz]

You missed the part where the parent said "within a single architecture".

Intel's heat issues started when they introduced Prescott, which was effectively a new architecture that didn't really deserve the moniker "Pentium 4".

When you compare the current P4 to the original P4, they have very little in common. Intel just stuck the P4 name on all of them for marketing reasons. In fact, if I'm reading all the coverage of Conroe correctly, they are going to call it a P4 too even though it is a completely different architecture derived from the Pentium M (which is itself derived from the Pentium III).

So the parent's point remains valid. When you compare the various initial speeds of the Conroe, since they will all be based on the same core, comparing performance based on clockspeed (between Conroe chips) will be a valid comparison.

Re:Shock news.

[Guspaz]

In fact, AM2 isn't superior to what Intel has, because DDR2 performance on AM2 is significantly SLOWER than AMD's current DDR performance.

See this article:

http://anandtech.com/weblog/default.aspx?bid=279

Essentially, AMD is having huge performance issues with AM2 with their current prototypes. At the current point in time, overclocking a current DDR1 S939 chip is going to give you faster performance than a current DDR2 AM2 chip.

Re:Shock news.

[Guspaz]

It doesn't recognize the MODEL NUMBER because they're overclocking it. This is normal on an AMD board when you overclock a CPU. The motherboard simply can't figure out what model number to report when faced with a processor that doesn't actually exist.

Re:Shock news.

[trilliwig]

It was an FX-60 (overclocked to 2.8 GHz), which is pretty much identical to the Athlon64 X2 series with the sole exception of having unlocked multipliers to support overclocking. Hexus did a review of the same machines over at http://www.hexus.net/content/item.php?item=4843.

Re:Shock news.

[GuyverDH]

LOL - the benchmarks were probably compiled by Intel as well, using the Intel compiler, with default settings too.... =D

Re:Shock news.

[mink]

You also left out that they are supposed to be adding virtualization to all processor lines (thats the rumor I read) around the middle of this year. I am not sure if this will mean anything in the performance department as I have only worked with virtualization on IBM pSeries hardware.

Re:Shock news.

[billcopc]

That's the beauty of it. Intel is (as always) desperately throwing in new technology that has more often than not backfired. AMD is calculating its move and waiting for the right time to nail Intel square between the eyes, as it has done for the past 3 years now. AMD still has old 130nm dies at 2.6ghz coming within 80% of Intel's 3.4ghz Xeons, to speak nothing of the latest 90nm cores that are more efficient both in performance and power consumption. Intel is moving to a 65nm process to ramp up clock speeds, of course the highest clocked cores will beat out current Athlon 64's, but what will happen once AMD follows into the 65nm arena ? What will Intel do when a finely tuned 65nm 3.6ghz dual-core Athlon64 comes rapping on the door ? I'll bring the popcorn, it's going to be a fun show!

Re:Shock news.

[Psiven]

Changing chipsets isn't going to cover the 40% performance gain in Fear. Also, Tom's Hardware did a comparison with a 32 lane PC-X mobo using ATI's RD580 chipset. The chipsets compared about equally with the SLI performing slighty better. It performed no where near 40% better.

Re:Shock news.

[Overly Critical Guy]

I see, so to translate your fanboi post, you have decided to believe that Intel's total victory over AMD here doesn't matter because AMD has super-secret magic plans that will instantly kill Intel.

but what will happen once AMD follows into the 65nm arena ? What will Intel do when a finely tuned 65nm 3.6ghz dual-core Athlon64 comes rapping on the door ?

Easy answer. Based on the public roadmaps of both companies, when AMD finally gets 65nm out the door, Intel will be releasing 45nm. AMD is an entire die-size behind Intel. And yet I still believe Slashdotters will be vehemently anti-Intel, because it's the cool, hip thing to do even when Intel clearly has the better chips in 2006/2007.

Re:Shock news.

[GuyverDH]

I'm not sure I'd fully agree with that statement either.

Reason one, Cache size - if they release slightly altered Conroe models, with varying sizes of cache, that of course will skew the results.

Reason two, Memory speed utilized with the system - varying speeds / quality of RAM will impact speed results, regardless of CPU mhz

Reason three, System cooling efficiency - if the machine built with the faster proc cannot cool the cpu sufficiently, typical configurations will slow the machine down to compensate for it.

Now - if you can clearly state that

#1 - All tests are done using the same memory, (the same physical RAM sticks)
#2 - All tests are done with the same chassis, motherboard, power-supply, hard drives, video card and cooling system (except for the CPU heatsink, using the one supplied with each CPU by Intel)
#3 - All tests are done with CPUs using the same size cache.

Then you should be able to compare Mhz to Mhz and get a somewhat linear result in benchmarks.

But don't tell me, that just by increasing mhz within a single architecture, that you can get an approximately linear result curve - there are too many other factors involved.

Granted, I am being nit-picky here, yet I'm just following someone else's lead in that regard.

Re:Shock news.

[billcopc]

You didn't get the point of my post. The fact is that Intel's future chips are going to beat AMD's current chips. Well yes, DUH, but AMD will also release new, faster products. The performance race never ends, I just feel that AMD is better at squeezing the most out of current manufacturing processes and IMO better architecture designs. It took Intel how long before admitting Netburst was a lost cause ? Pentium-M is great, had it been out three years ago they might have given the Athlon T-Bird a good beating, but that's not how history was written.

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

[masklinn]

That ars technica article is denial at its best. "intel faster? CANNOT BE!!!111"

I suggest you RTFA before that kind of statements, the Anandtech article can be summarized as "Intel's chip beat the crap out of AMD's, and even though we're going to wait til we can do our own benches we didn't find anything that could lead to thinking the testing rigs were tampered with. Conclusion, even the M2 socket probably won't be enough for AMD to keep the lead".

Re:Wait and see

[mcbridematt]

But isn't Conroe based on the new Intel Core (Not the current Solo or Duo) design, which is similar but not the same to the P-M? See the dirt on Wikipedia

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

[Anonymous Coward]

I R'd TFA and that's exactly how the Ars editorial came across. What the writer seems to ignore is the F.E.A.R. benchmark that Anandtech ran. It used their own demo instead of an Intel provided one (like the other benches) and it still trounced the AMD chip.

I agree that we should wait and see until truly independent benchmarks are done, but I don't see a reason to be as dismissive as the Ars writer.

Re:Wait and see

[Anonymous Coward]

Learn to read, you halfwit.

Re:Wait and see

[Anonymous Coward]

Umm, I suggest you RTFC (comment), as the grandparent was in fact referring to a link posted in another comment to arstechnica, not the original anandtech article.

Re:Wait and see

[Anonymous Coward]

It's really pathetic when a grossly inaccurate oversimplification gets scored 5 just because it somehow reinforces AMD fanboy delusions.

Re:Wait and see

[adsl]

Here's a thought and a question: It seems top me that Intel's "new" platform CPU design can be used across most sectors. i.e. Laptops, PCs and Servers at 65 nanos. They also admit that it's as cheap to produce in dual core as the Pentium single core is at 90 nanos. So here we have excellent computing power, great power saving and MASSIVE cost savings in production. Conclusion and question: Given the above will this give Intel, what looks like, a huge price advantage in terms of production and enable them to realise much greater margins than ever? This of course would enable them to underprice anything AMD could offer..... If so, for the consumer it looks like second half 2006 will be a great time to build/buy a new PC.

Re:Wait and see

[Anonymous Coward]

Anyway 2006 will be a good year for CPUs

I've been using PCs for about 18 years now and I have to ask... how the hell does anyone keep up with all these processors? My last PC purchase was an AMD Athlon XP 2400+ which was roughly like an Intel P4 2.4GHz CPU, but since then the whole thing has become immensely confusing. If you can even follow all the CPU code names you might be able to figure out why one Athlon clocked lower than another Athlon is actually faster, or what the hell the difference is between an Extreme Edition CPU and a dual core or why an FX2 AMD CPU is faster than a non-FX2 CPU... how the hell do you guys follow all this anymore? To upgrade my computer I'd need to spend several weeks relearning CPU architectures just to figure out what the hell is current. Apparently a 1.67GHz "Core" CPU is faster than a 3.6GHz P4 for example. Why even bother putting down the hertz rating at all anymore? Just assign it a production number with newer/faster CPUs getting a higher number.

/old fogey

Re:Wait and see

[acidblood]

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...

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.

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.

I'm sorry if that's not a fashionable opinion in Slashdot groupthink, but there you go.

Re:Wait and see

[Quirk]

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

K9 is bound to be a real dog.

Re:Wait and see

[the_real_bto]

It sounds like AMD and Intel are both choosing the designs that work. How can anyone blame them for that? Can you follow up with more evidence on your "carbon copy" claim? If AMD's designs are so uninspired, then why is AMD giving Intel so much trouble right now? This in spite of Intel's advantage in manufacturing technology and might.

I don't believe for a second that Intel's marketing department designed the p4. But I'm also not so naive as to think that Intel's marketing department didn't try to take advantage of high clock speeds in their advertising. They made their bed, now they get to lie in it.

AMD has delivered better speed performance at better power consumption than Intel. Kudos to Intel for trying a new design, too bad for them they stuck with it for so long. That is life in the big city. The company that delivers *results* gets rewarded. AMD has done just that.

Re:Wait and see

[Anonymous Coward]

That ars technica article is denial at its best.
"intel faster? CANNOT BE!!!111"

And pray tell, where does Hannibal write that? I did read:

For the record, I do actually expect Intel to regain the performance crown from the Athlon with Conroe.

and

So yeah, I expect Conroe to beat the current Athlon64 architecture, all things being equal.

but most importantly

But until independent benchmarks come out, there's no way to know the extent of Conroe's advantage. I really don't expect it to be as big and unambiguous as these benchmarks suggest, but of course I could very well be wrong.

But then again spouting nonsense is so much more fun than an actually RTFA.

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...

Re:Wait and see

[Sketch]

The Core is basically a redesign/relayout of the P6 on a modern process, with some things learned from the P4 thrown in.

Re:Wait and see

[acidblood]

Can you follow up with more evidence on your "carbon copy" claim?

Sure, just grab the optimization guides for both processors and compare the microarchitectural diagrams. AMD tried to disguise this a bit by changing some terminology, but after adjusting the terminologies to match, the similarities between both processors are clear.

Although I have to admit I've been fooled by this as well. I recall I came to this realization by reading H&P's classic `Computer Architecture: A Quantitative Approach' and not the optimization guides.

Re:Wait and see

[the_real_bto]

Thank you very much for the sources. I just looked over all 3 of them a bit. Very interesting. I don't have the energy or time to dig into those now, but I always appreciate a good nudge in the direction towards knowledge. They will be filed for future reference. Thanks.

Re:Wait and see

[Deliveranc3]

Launching a new socket with identical speed CPUs? Ouch ouch ouch... 5000+ my ass.

wooo 5.... no one cares about that. AMD has basically said that they aren't doing ANYTHING new in terms of performance this year, now old reports said that it would take Intel approximately 4-5 years to recover from not adopting HyperTransport (Doesn't matter how fast you run if your cpu is starved for bandwidth) but AMD should have been moving in that period.

Their prices aren't coming down the way they used to so they need to find some way to replace their product line while keeping high margins going to be tough to pull off without about 1.5 - 2x more performance.

Re:Wait and see

[Anonymous Coward]

You shouldn't be driving and posting on slashdot.

Re:Wait and see

[Anonymous Coward]

Hey retard, AMD hasn't had any radically new architecture since they bought Nexgen.

Innovators? Hardly.

Re:Wait and see

[dr.badass]

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...

This is true of Core Solo/Duo, but not of Conroe. Conroe, Merom, and Woodcrest are all a new architecture.

Re:Wait and see

[Overly Critical Guy]

Core Duo and its predecessors are based on the Pentium Pro, but Intel's new chips starting with Merom are a redesign to replace Netburst and vastly improve performance-per-watt.

Re:Wait and see

[Anonymous Coward]

Sorry, i am as much an AMD fanboy as anybody (hey, their stock financed the car i am driving right now),

You really shouldn't be driving your car and posting to slashdot at the same time.

Re:Wait and see

[Psiven]

"Solo" and "Duo" are Core. They're both Yonah processors but with either a single or dual core. They're all the same.

Yes, Core (Yonah) is pased on Pentium M. It evolved from the Pentium III Celeron, then into Northwood, to Pentium M, now Yonah and next Conroe/Merom.

Re:Wait and see

[SebNukem]

K9 is bound to be a real dog.

Actually wait a little for... KY bound to be the slickest core ever.

Re:Wait and see

[Bert64]

Yes, the K7 does copy a lot from the P6...
But the P6 takes a lot of ideas from the nexgen 5x86 designs (now owned by AMD) and the DEC Alpha designs (at least partially licensed by AMD, and DEC successfully sued intel over ripping off the alpha designs)

Re:Wait and see

[Anonymous Coward]

Arstechnica are a pack of uneducated dolts and no professional experience in this field idiots and morons.

In fact, one of their "article authors" in Jeremy Reimer is whom I submit here as living proof of my first statement.

For example: Jeremy Reimer has no computer science degree, and not even a certification like an MCSE (not even an A+), and Jeremy Reimer certainly does not possess years of professional experience in the trenches in this field.

Yet he is supposed to be some kind of authority on matters regarding computers?

Arstechnica is just a pack of hacks, or rookie students, trying to play computer-experts online.

The Conclusion

[mtenhagen]

The conclusion from the article:

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.

Also keep in mind that we are over six months away from the actual launch of Conroe, performance can go up from where it is today. We also only looked at the 2.66GHz part, the Extreme Edition version of Conroe will most likely be clocked around 3.0GHz which will extend the performance advantage even further.

AMD still does have some time to surprise us with AM2, but from what we've seen today, they are going to have to do a lot of work to close this gap. We saw performance today in the two areas that we were most concerned about with Conroe: gaming and media encoding, and in both Intel greatly exceeded our expectations. Also remember that Conroe should be lower power than the AMD offering we compared it to, although we weren't able to measure power consumption at the wall in our brief time with the systems.

Going into IDF we expected to see a good showing from Conroe, but leaving IDF, well, now we just can't wait to have it.

More from the show as we get it...

Re:The Conclusion

[Saven Marek]

I think the fact intel had to put two systems side by side to try "proving" they're faster shows their desperation. That kind of standoff where the intel can't perform just by itself without trying to compare to competition smacks of gimmickry over performance. AMD Still in the end wipes the floor with Intel and I'll bet real world benchmarks will show that yet again.

Looks like another big yawn for Intel's next generation.

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.

Re:The Conclusion

[Saven Marek]

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.

I've heard AMD's next generation will keep up the same improvements where AMD slaughter Intel and it looks like 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.

Just goes to show all the trickery in the world won't help a company that can't really "innovate". I'm surprised a sight like Anandtech didn't see it.

Re:The Conclusion

[Anonymous Coward]

These stories are like a broken record. Year after year it's the same old story.

And then reality hits for the clowns at Intel:

1) The chip will be delayed

2) The chip will run slower in realworld benchmarks

3) The chip will run hotter in realworld situations

4) Intel will continue to 'tweak' their marketing compiler to generate even higher SPEC scores

Same old, same old from Intel.

Not that I care...I'm running a quad-970 system at home and get to work on Cell stuff at work. Bwahahahahaha....

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.

Re:The Conclusion

[Glock27]

Going into IDF we expected to see a good showing from Conroe, but leaving IDF, well, now we just can't wait to have it.

Nice that this is a horse race now. Should be very good for competition going forward. Might also be a spot to short AMD and go long INTC coming up... ;-)

It also looks like Jobs made a genius move picking this point in time to go with Intel. He must have hated Netbust too. I can't wait to see the new MacMacs! (Intel PowerMacs, eh?)

Re:The Conclusion

[Anonymous Coward]

What short memory we all have. Wasn't that long ago that Intel was always better than AMD. But all the geek boys here feel empowered, much like a sports fan, by thumping their chests and proclaiming how smart they are via bizzare brand loyalty.
How about some useful information instead?

Re:The Conclusion

[Anonymous Coward]

Do you feel better now?

(Slashdot ignorance is prospering)

Re:The Conclusion

[mattspammail]

Just who do you think you are, and how do you know me so well?

You know, sometimes words can hurt.

Re:The Conclusion

[Anonymous Coward]

And if Intel didn't benchmark against AMD's chips, you'd just say Intel where too scared cos they knew they'd get owned, right? You're ability to twist reality to fit your deluded beliefes is actually quite scary.

Re:The Conclusion

[Anonymous Coward]

Here is an anand blog piece on am2. Doesn't seem to give great hope that just am2 will make a huge difference http://www.anandtech.com/weblog/default.aspx?bid=2 79

Re:The Conclusion

[twiddlingbits]

The big question is going to be is when Intel makes a chip that runs fast AND runs cool. The Conroe is still as big of space heater as the older chips are. The AMDs are just a bit slower in everyday apps (90% of the users are not gamers and don't care about FPS) but it's a hell of a lot cooler. HEAT matters to those who run data centers.

Re:The Conclusion

[jonnythan]

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.

Maybe in the notebook and server markets, or if you count total installed base.... but AMD certainly has a hell of a lot more than 20% of new desktop sales.

http://hardware.slashdot.org/article.pl?sid=06/01/ 19/1440200
"Sales of AMD-based desktops took the lead during the pivotal fourth-quarter holiday shopping season. AMD chips were found in 52.5 percent of desktop PCs sold in U.S. retail stores during that period."

Re:The Conclusion

[Omnifarious]

Well, I think they used their money largely to buy better benchmarks. What do you bet that those games used the Intel compiler and would perform a whole ton better with Intel's special AMD-detecting hooks removed?

Re:The Conclusion

[ImaNihilist]

"52.5 percent of desktop PCs sold in U.S. retail stores during that period"

Total desktop sales are a lot larger than desktop sales sold in retail stores. All that's really saying is that 52.5% of desktops PCs sold in retail stores feature an AMD processor.

Re:The Conclusion

[djohnsto]

This is just wrong. Conroe has a 65W TDP. I believe that the FX-60 is pushing 100W (if not over).

Re:The Conclusion

[jiushao]

Uh?

The Conroe here is the E6700, TDP under 65 watts (max for the mid-end E series).

The Athlon64 is a FX60, TDP of 110 watts at 2.6 GHz (whereas it is run at 2.8 GHz in the benchmarks).

Re:The Conclusion

[jiushao]

The benchmarked Conroe (the E6700) is expected to have a price-tag of $529, the benchmarked Athlon64 (at correct clock, it is overclocked a bit for the benchmark) costs $1032 on newegg.com.

So on this high-end (well, high mid-end for the Conroe since this is just the E series, there will be a higher clocked XE series as well) pricing looks to be in Intels favour at the moment. Things might shift a bit before the release, but it seems likely that the pricing will be very competitive.

Re:The Conclusion

[twiddlingbits]

Several spec sheets show Athlon 64 dual core at 93W and 65 degrees C MAXIMUM for case temp, both Intel and AMD use that measure. Internal chip temps are not published. In fact I've never seen a Thermal image of either chip showing hot spots. One review of an Athlon 64 X2 stated "...Athlon 64 X2 4800+ was able to run our benchmarks with an operating temperature of less than 40 degrees Celsius." (http://www.itreviews.co.uk/hardware/h753.htm). Until the Conroe chip goes into volume production I'm skeptical about Intel's claims.

Re:The Conclusion

[mink]

And if you go back before that AMD was as good as or better then Intel, depending on who you ask.

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.

Re:A better competetion

[Anonymous Coward]

Kind of makes you wish we had such competition on the OS side. Imagine a world with a roughly equal percent of Windows, Linux, OSX, BSD,... machines. Oh, the innovation we would see...

Re:A better competetion

[Anonymous Coward]

> Imagine a world with a roughly equal percent of Windows, Linux, OSX, BSD,... machines.

Imagine a world where 98% of software is written Java for portability.

Re:A better competetion

[Anonymous Coward]

Ieeeeeiiieeee! The tentacles!

Re:A better competetion

[Glock27]

Imagine a world where 98% of software is written Java for portability.

At one point, not that long ago, I agreed 100% with this thinking. I was still drinking the "Java will have C performance" Koolaid from Sun.

I'm now of the opinion that the "managed" languages are a short-term abberation, unless they adopt an ANDF type "freeze" approach. That is where the bytecodes are pre-compiled once into machine code, just like a traditional compiler. I'm also not happy with where Java is at as a language after 10 years of evolution. No operator overloading, feh.

Lately I've been looking at D and Dylan for some projects. Both are quite advanced compared to Java, just as portable, and from what I've seen so far both outperform it in many areas. Game and HPC programmers could really use a better language than FORTRAN/C/C++, and Java will never be it, IMO. D seems the more pragmatic of the two, while Dylan looks "better" from a pure language perspective.

If the new processor performance metric (as touted by Intel) is "performance per watt", someone should take a hard look at Java and .Net performance compared with the top compiled languages. Dylan or D would work fine as "server side" web development languages.

Re:A better competetion

[m50d]

I'm now of the opinion that the "managed" languages are a short-term abberation, unless they adopt an ANDF type "freeze" approach. That is where the bytecodes are pre-compiled once into machine code, just like a traditional compiler.

.Net is at least capable of doing this, and it does give a decent performance improvement. Personally I think it should overtake the JVM - more emphasis on choice of languages (of which java is one), and it seems to be generally improved.

Re:A better competetion

[John Courtland]

Java purposefully does not overload operators because of the thought that programmers suck and will screw it up. I personally disagree and wish I could be allowed to have to option of being a "bad programmer" in Sun's eyes.

Re:A better competetion

[Glock27]

Ah, I see a genius mod saw this as "Offtopic". Not sure how, given:

'If the new processor performance metric (as touted by Intel) is "performance per watt", someone should take a hard look at Java and .Net performance compared with the top compiled languages. Dylan or D would work fine as "server side" web development languages.'

This is completely topical with the discussion of new Intel processors. Nice job.

Re:A better competetion

[Glock27]

Java purposefully does not overload operators because of the thought that programmers suck and will screw it up. I personally disagree and wish I could be allowed to have to option of being a "bad programmer" in Sun's eyes.

I'm well aware of the rationale for no operator overloading in Java. I just completely disagree with it.

No language can protect you from bad programmers. Don't hire them, or work (long) with them. If an organization doesn't want operator overloading, they should mandate that and make sure with code audit tools. There's no justification for penalizing those of us who could make good (and clear) use of operator overloading for the *possible* abuses of inferior coders.

It's the language equivalent of a "nanny state", and I hate that. ;-)

All that said, these new Conroe processors sure look promising for running whichever type of code. =)

Re:A better competetion

[chez69]

if you've ever maintained a large piece of software written by 'clever' programmers you'll agree with sun. operator overloading gone amok is EVIL.

I'm sure it has it's uses, but my 7 years of experience tells me that maintainable code is easy to maintain in the long run =-)

For simple one person projects, I can see why you prefer it.

Re:A better competetion

[renoX]

>>
I'm now of the opinion that the "managed" languages are a short-term abberation, unless they adopt an ANDF type "freeze" approach. That is where the bytecodes are pre-compiled once into machine code, just like a traditional compiler.

I've always thought that ANDF approach wasn't so good as compiling the bytecodes could take a long time for say a 100MB binary.. At least with a JIT, you only compile what you need (the loops).
What's your rationale for preferring ANDF approach?
It's not used very much nowadays..

Re:A better competetion

[Glock27]

I've always thought that ANDF approach wasn't so good as compiling the bytecodes

Have you ever looked at dissassembled bytecode? It's very close to the original source code, which means almost all of the same optimization opportunities are there. Why shouldn't it do a good job? could take a long time for say a 100MB binary..

Do you have any 100 MB binaries? o.O

At any rate, it only happens at install time, and shouldn't be a big deal in terms of time.

At least with a JIT, you only compile what you need (the loops).

That's not quite right for HotSpot and other advanced VMs. First of all, they're constantly profiling the code to see if it needs to be JITed. Then, at whatever intervals, they actually do some translation into machine language. However, all this activity occurs every time the program runs! Doesn't it bother you a bit that all the profiling information from all previous runs is lost? If we're talking "performance per Watt" current VM approaches seem to have some serious overhead compared to compiled languages.

What's your rationale for preferring ANDF approach?

Expensive optimizations are only paid for once. It's prohibitive to even try some optimizations in JIT compilers, those need to be fast.

It's not used very much nowadays..

Yes, most languages use my preferred method. Source -> Object -> Executable/Library. Bytecodes don't really obfuscate much, might as well distribute obfuscated sources.

Traditional compiled languages can use all the expensive optimizations they want, and furthermore the program has deterministic timing when run. Neither of these statements is true of JITing VMs.

Re:A better competetion

[StikyPad]

I see a genius mod saw this as "Offtopic". Not sure how.

Probably because a joke about Java got launched into a full-blown debate about language portability. Obviously the GP should've started his post off: "Now here's a Jimmy joke about your Java that you might not like.."

Re:A better competetion

[Glock27]

I'll buy your point as soon as you convince me that:

BigDecimal t = new BigDecimal(0.0);
BigDecimal a = new BigDecimal(1.006);
BigDecimal b = new BigDecimal(5.0);
BigDecimal c = new BigDecimal(15000000.0);
t = c.add(a.pow(b));

is clearer than:

BigDecimal t = new BigDecimal(0.0);
BigDecimal a = new BigDecimal(1.006);
BigDecimal b = new BigDecimal(5.0);
BigDecimal c = new BigDecimal(15000000.0);
t = c + (a ^ b);

It's not, there's no getting around it, and the problem gets much worse with more complex examples and higher mathematics.

If there's no advantage to algebraic syntax, why provide it for primitives?

I return to my earlier point: any organization can mandate against using operator overloading if it wants to. Operator overloading is syntactic sugar, so you can continue to simply call the methods the old way if you prefer it. Just don't force everyone into your mold.

Re:A better competetion

[Glock27]

The example probably would have been better as:

t = c.add(c.multiply(a.pow(b)));

vs.

t = c + (c * (a ^ b));

BTW, I now see that I can't raise a BigDecimal to a BigDecimal power...strange omission there.

Re:A better competetion

[Glock27]

Probably because a joke about Java got launched into a full-blown debate about language portability.

I guess we'd have to ask the OP, but I don't think it was a joke. The idea of completely portable apps is a very appealing one.

Whether Java is a reasonable means to that end is an entirely different discussion. ;-)

Re:A better competetion

[StikyPad]

The idea of completely portable apps is a very appealing one.

Mostly just for developers and maybe OS-X users, although I'd wager most are too snobby to want our uncultered apps. The rest of the Windowscentric world doesn't care much either way.

AMD Processor Model Unknown

[Bert64]

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

Re:AMD Processor Model Unknown

[Anonymous Coward]

The AMD was being overclocked to make things a little more fair.

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:AMD Processor Model Unknown

[GooberToo]

That's exactly right. AMD pruposely does not chase the clock cycle monkey. That's the whole point behind their naming convention. While I admit I did not read the article, based on these comments, it's pretty clear that the comparison is complete BS. AMD, for a long time now, has gotten great performance by changing/tweaking their design and not by simply cranking up cycles....which sounds exactly what was done in this comparison.

Who knows, this test may be foretelling of AMD's next effort, but until the actual product is out, IMO, this test is meaningless.

Re:AMD Processor Model Unknown

[koafc]

From what I learn from some AMD-fans, when the AMD cpu is overclocked, it shows up as model unknown.

Re:AMD Processor Model Unknown

[mobby_6kl]

>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

What do you mean 'is not listed as being as fast'?? FX-60 is AMD's fastest desktop dual-core processor, what do you expect them to do, send ninjas to steal AMD's engineering samples?

Re:AMD Processor Model Unknown

[Malenfrant]

No, I expect a test comparing two chips to test two equivalent chips, not one fast one and one slow one overclocked, otherwise the exercise is nothing but a marketing exercise.

Re:AMD Processor Model Unknown

[Anonymous Coward]

Clueless Slashdot poster as usual...

Re:AMD Processor Model Unknown

[Overly Critical Guy]

Right, because it can't just be that Intel's new chips are kicking AMD's butt. You just "know" that AMD's pre-production chip will magically have been faster in these tests.

Where's AMD's 65nm?

Re:AMD Processor Model Unknown

[Anonymous Coward]

And we're all glad you know who you are! Now, if only we could keep you from wasting all our time with you boasting of just how clueless you really are.

Re:AMD Processor Model Unknown

[gordo3000]

so all you can take from the test is the fact that the Conroe, on certain benchmarks, beats the shit out of the fastest processor AMD has, even if it is overclocked(which according to other Toms Hardware articles, does give it a performance boost in these tests).

I know that isn't saying much, but it does say that the next generation of processors coming out in the next 6 months will beat the pants off of all the other processors that are out there. This is a big deal because it is very rare that there has been thsi kind of speed jump in just a few months.

Re:AMD Processor Model Unknown

[iainl]

Umm, I don't get your hostility. Your supposition that "Intel's new chips are kicking AMD's butt" is pretty much what I meant. All I was doing was describing (as asked) why the AMD chip wasn't recognised as a standard one by the BIOS.

Processor affinity?

[Anonymous Coward]

Does Windows have any processor affinity settings that could have been "rigged", or does the cache type between Intel's Dual Core CPUs and AMD's Dual Core CPUs differ(shared vs unique?)?

Re:Processor affinity?

[Anonymous Coward]

Yes, they do.

Athlon 64 X2s and Pentium Ds have independent cache for each core. Yonah has a unified (shared) cache for both cores, which would improve performance significantly for anything involving inter-core communication. I'm assuming Conroe/Merom will also have unified cache, although I could be wrong.

Who staged This?

[Anonymous Coward]

And as you'll notice Intel Staged the test so it will be interested to see what a Third party test will learn when the do a comparison, along with the new AMD processors, not ones that are already months old.

Weird Summary

[Anonymous Coward]

'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.'

This summary is really weird, until you rtfa and realize that the numbers are provided by Intel themselves supplied the hardware and possible even ran the tests (hard to tell exactly from the article. The submitter could have had a more meaningful summary, or were they afraid that stating the Intel connection from the get-go would bias /. readers (like that would happen, right?)

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:What about RAM?

[Anonymous Coward]

Mod parent up. The different memory systems skew the results a lot. That kind of artificial benchmarks are entirely meaningless and only show that some PR department is spending a lot on "journalism". With monopoly profits that can be done, but let's hope it will not last long.
 
On the other hand I'm a strong believer in competition and I know for sure that consumers will suffer a lot without AMD on the market. This is a perfectly rational and tremendously important reason to support the underdog. The superior tech of AMD makes it easy and stupid articles like this only reinforce the point. Oh, well, Turion, here I come...

Re:What about RAM?

[fitten]

One of the biggest arguments against DDR2-667 is that it performs roughly equal to DDR400 in real tests. If you take them with a grain of salt, the recent reviews of the AM2 Athlon64 actually ran slower on average with DDR2-667 than a comparable system (same clock rate and such) using DDR400 (although there were issues with that system being tested).

Re:What about RAM?

[Keeper]

The ram isn't 66% faster. The clockspeed of the ram is 66% higher; the latencies are also 2x higher. The performance of the ram in each system should be similar; not identical, but about as close as each different type can get.

Typical AMD fanboy rant

[Anonymous Coward]

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.

If you read the various benchmarks over the years, changing memory architecture or increasing it's speed directly does very little to increase most benchmarks more than a percentage point or two. Inceasing FSB also hasn't done much. Rather increases in processor performance are directly responsible for the disparity between the new Intel cpus coming.

Re:Typical AMD fanboy rant

[liliafan]

Additionally it is a fair test, they are using supported technologies, in the real world people are more likely to use this configuration for their systems if you can use fast RAM you will. If AMD want to take advantage of this they need to support the technology.

Please note I am a fan of AMD I have been using their processors since the k62 400mhz, but I do believe benchmarks should take full advantage of all supported technologies.

Latest chips, latest games & instant obsolesce

[digitaldc]

Since you need a really high-end PC to play most new games these days, most people will miss out on the new titles and technology. The new dual-core technology is outpacing most users purchasing power and the ability to even play the game on its lowest graphics settings.

A year or two after people spend an avg. of $1000+ for a new system, most are not going to run out and buy the latest dual core chip and ATI/NVidia video card just to play the latest new game (Quake 4, Far Cry, F.E.A.R., etc.) and then keep doing that year after year.

They need to make it so the games can be played (with the lowest settings) on any system with chips from the past 5 years IMHO. Then everyone can enjoy the game, some more detailed than others. At this point, it is just better to buy an XBOX or PS2 and just buy games that they know they can play without constantly upgrading your system.

Nice to see a manufacturer take their time!

[brucmack]

The thing I like the most about where Intel is going is that they really seem to be taking the time to do it right. They have been doing exactly the opposite for the past couple of years... Prescott was released with many good ideas that were just never put together in a way that gave a good final product. Then the Intel dual core chips were just two single core chips pasted together, not even sharing the cache... again, it just seemed like a "let's just get it out the door" solution.

Video cards are even worse, with the shorter dev cycles. How many times have we seen a manufacturer put out a video card that is essentially the same as their last model, but with a ridiculous overclock and cooling solution. It's not innovation, and spending the time to develop properly would put us as a technological society further ahead a year from now.

But Intel's really taken their time with this, and hopefully they will have gotten their 65 nm yield issues worked out by the time they want to ramp up production. Hopefully AMD will follow suit and give us some great innovation in 2007!

Re:Nice to see a manufacturer take their time!

[somersault]

it seems that Intel is still just ramping up Mhz, though this time going back to the Pentium M as a base, which was developed by their mobile division in Israel. They appear to be a lot brighter than the main engineers :p

Re:Nice to see a manufacturer take their time!

[Overly Critical Guy]

It's because of their new CEO, who reorganized Intel's engineering into groups by target markets instead of by chip models. It's why Intel owns portable computing right now and is set to make huge gains in the desktop space again.

Re:Latest chips, latest games & instant obsole

[LiquidCoooled]

I got a cheapy 64 bit sempron system with crappy onboard 6100 nVidia graphics card (theres an empty pci-e slot available for the summer)

I held off playing half life 2 because I didn't think it would run (I had a 5900 agp previously than ran it really well)

I am running now at 800*600 with full details enabled and 2x AA and I've only noticed one point where it even shudders (the chimney blowing up and falling whilst in the airboat), if anything its smoother on this card than before, and the shaders are tonnes better (water, and nobbly glass doorways especially).
The only thing I'm missing is the ability to go to super resolution, but considering what I have gained I'm willing to wait.

I was very pleasantly surprised :)

What about cost, ram cost, power, and heat?

[CFD339]

There is a lot more to picking a processor than just how fast it runs. Personally, I have more bottlenecks with I/O (as I've said before) than I do with video or processor performance.

Of importance to me in addition to raw speed are are the number of concurrent threads, the power consumption and with that the heat output I have to dissapate into my office or my lap, and of course the expense of both the processor and the ram it needs to get these kinds of speeds.

Frankly, I'm looking for which allows me to build the most efficient system for my needs at the least cost.

COnRoE?

[Anonymous Coward]

Any connection? :)

Re:Latest chips, latest games & instant obsole

[Ginger Unicorn]

I have an athlon xp1500+ with a geforce fx5200 and 512MBs of RAM and i can play anything on the market as long as i turn the settings down to "my computer is a retard" levels.

even games that say they require faster CPUs dont.

Re:Latest chips, latest games & instant obsole

[miscz]

The fact that new games have very high demands isn't not the thing that bugs me the most. I'm more angry with the fact that new games on lowest details look worse than Pong. :\

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)

[ingsocsoc]

Remember when we were pissed off about the Processor Serial Number on Pentium III's? Those were the days :)

Re:TrustedComputing Inside (TM)

[segedunum]

We have been waiting for details on LaGrande for several years and according to Chapman, we will have to continue to wait and should not hope for much information on this technology this time around. This is somewhat surprising, especially if one considers the Apple-Intel deal - in which LaGrande appears to play a key part as technology that prevents MacOS to run on any PC.

They don't seem all that keen to talk about it either......... As Alan Cox said, if you don't have the key to your own hardware then it's not about security. I see stormy waters ahead...

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.

Re:TrustedComputing Inside (TM)

[Anonymous Coward]

Believe it or not, this fight is not lost. It's not just geeks on slashdot. This hardware (with the master key hidden) takes control of the PC *completely away from a business*. It puts them completely at the mercy of Intel and Microsoft.

We can get them onside in the push to ensure that they hardware works for us, and not against us. We jsut have to make sure that they hear us and not just Intel whisphering seductive promises of better security and control over their computers.

Look at it this way: Microsoft and Intel will be selling this hardware as allowing Boss total control of his employees. The reality is that any company boss submitting to use this hardware without access to the master key is giving up any control of their data, and in fact, any privacy. This is a message that will sell... and Intel and Microsoft *will* have to listen when their business customers start telling them to fuck off. We just have to ensure that our message is heard, by the press, by business, and by your average consumer.

Re:TrustedComputing Inside (TM)

[justins]

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.

How are you trying to "get you key out"? I imagine if you're using a Dremel, that is pretty destructive.

What a bunch of dumb FUD.

Re:TrustedComputing Inside (TM)

[ivoras]

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

Either that or start buying "independant" products such as VIA's processors, which even have hardware AES and hardware random number generators, for that added safety when you're feeling paranoid.

Re:TrustedComputing Inside (TM)

[dr.badass]

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

AMD's 2006 roadmap already contains chips with identical Trusted Computing features. That is to say, it is already way too late.

Re:TrustedComputing Inside (TM)

[Anonymous Coward]

Intel's new chips have a Trust Enforcer chip embedded inside the CPU itself.

As do AMD's.

Each chip features a unique serial number

Unless I missed something, this has been standard since the PIII. Standard and off by default.

DRM enforcement

Don't like it? Don't download DRMed content.

Sealed Storage to prohibit you from reading your own files on your hard drive

What? Nobody would ever use it for that. The main use for Sealed Storage is to prevent other people from accessing your data (or more accurately, the data of your business or governmental agency), in the event of theft.

Re:TrustedComputing Inside (TM)

[Anonymous Coward]

What? Nobody would ever use it for that.

Microsoft would, and does with BitLocker.

The main use for Sealed Storage is to prevent other people from accessing your data (or more accurately, the data of your business or governmental agency), in the event of theft.

It's quite simple -- the test of the agenda behind Trusted Computing is quite simple. Give us, the owner of the machine, the master key to our own machine. I'm not keeping military secrets on my machine. I want to know what's stored on it, and I wanted to make my own choices. The purchasers of military hardware or intelligence agencies, or corporations concerned about espionage do not need to hand this master key out to those using their supplied laptops in situations where secrecy is required. Choice. Everyone can get the benefits of this hardware without the serfdom. The Trusted Computing Group will not do this, of course, since keeping the master key for each machine secret is their real aim... it's what will allow them almost total control of the computing infrastructure.

Re:TrustedComputing Inside (TM)

[Anonymous Coward]

It doesn't even require processor to be free of this "Trusted" stuff... just that it be done in such a way that he owner of the machine can assert control over it. The owner should be given the master key. That way, the benefits can be gained, with none of the downsides.

The fact that the trusted computing group is vehemently opposed to this is a giveaway as to their real agenda.

Re:TrustedComputing Inside (TM)

[Alsee]

The point is that the chip is in fact designed to be secure against the owner, and to secure the computer against the owner.

I do not dispute that you would need a suitably equipped college laboratory or similar equipment to extract your key. But they are so insanly set on securing your computer against you that that's not good enough for them. They include anti-tamper mechanisms on top of that to destry the key even if you *do* happen to be a smart skilled college student with access to a well equipped lab or something similar.

An IBM Thinkpad TV commercial even advertized the fact of the boobytrapped selfdestruct feature of these chips. Of course the IBM commercial only mentioned it in the context of protecting your data against theives and hackers. They did not advertize the fact that it was primarily designed to keep your computer and your files secure against you.

What a bunch of dumb FUD.

Are you claiming I said anything untrue??? Or do you simply think that Trusted Computing is a Good Thing and that people's computers should be secured against them using self destructing anti-owner chips?

-

WHEEEE!!!

[Damastus the WizLiz]

Its like watching Republicans and Democrats argue.

Re:WHEEEE!!!

[Grishnakh]

Yeah, it was better in the good ol' days (just last year in fact) when Intel's stuff was all obviously crap, and only a fool would try to argue that Netburst was any good at all and could actually outperform AMD's stuff.

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. :(

Re:The problem with buying a new processor

[Anonymous Coward]

No my poor friend. You are just reading reviews on the wrong sites. Here is the laptop you are dreaming about. Previously covered on slashdot also. Leave these AMD and Intel Fanboy fools to drool over their new 'petty innovations'... :)

Re:The problem with buying a new processor

[eris23007]

I feel your pain, but allow me to point out the following info:

The last time I bought a CPU, I worked for an Intel Product Dealer and got a 50%-off deal on a P4/3.0GHz Northwood with an 800MHz FSB. That was in the Fall of 2004. That machine obviously isn't the fastest thing in the world anymore, but it will still be solid for a good year or two more before I really need to consider upgrading. I'm not a gamer but I do record multichannel hidef audio on my PC.

The reason I point this out is to provide evidence for the following contention: In my opinion, the past couple of years have shown that getting in at the right time is more important than it used to be. Back in the late 90's, it was more a case of "just buy whatever's at your price point when you need a computer", because you knew it was going to be heinously obsolete after two to three (maybe you could stretch it to four) years.

Nowadays though, single-threaded execution speed is increasing at a relatively slow pace (at least in comparison to the pace at which it increased back in those days). The CPU manufacturers are focusing on improving multithreaded performance, adding multiple cores and such. This is a boon to code that is suitably multithreaded, as it will provide a speed increase of 50%-80% or more (100% isn't really possible because the dual core machines share memory-subsystem bandwidth, whether it's cache or RAM) in really well-written multithreaded code.

However, the software hasn't kept up, particularly in the case of games - most games are still primarily single-threaded. Unfortunately writing good multithreaded code is extremely difficult, and debugging it even more so. The situation is different in server-land: the types of code for processing databases, serving webpages, etc. is well-suited for multithreading and has been written this way for years, especially in *NIX-world.

Therefore, I would simply like to point out that the marginal improvement in performance from adding more cores will decrease over time - i.e., a four-core machine might improve performance 25%-50% over a dual-core machine instead of another 50%-80%, because there are just that many fewer situations in which you can take advantage of the additional resources.

The overall point I'm trying to make is, I believe the introduction of Conroe to be one of those "getting in now is good timing" sorts of situations, due to the confluence of factors I pointed out - recent introduction of a significant architectural improvement, reduced power consumption, dual-core, good platform, etc... Note that this is contingent upon the "real" benchmarks proving to be as impressive as the ones released in this article. If those prove to be false, all bets are off... ***Also note that the discussion of percentage-speed-ups is highly dependent upon the type of code and suitability for multithreading - code that is completely single-threaded will show 0% performance improvement when moving from a single-core cpu to a dual-core cpu, with the same architecture.

I think it's a bit early for benchmarks...

[Anonymous Coward]

Lets actually see the processor released before we crown it the new champion. After all, I still haven't seen a 4GHz Prescott that they demonstrated.

Re:I think it's a bit early for benchmarks...

[Anonymous Coward]

Companies and people all over the world have Conroe samples running since December (I have one in my machine). It is hillarius to compare that to the 4Ghz Prescott demo which Intel admitted, at the time, was not about to be released soon and was using exotic cooling.

Re:Latest chips, latest games & instant obsole

[jmke]

Wake-up call; NEW games are ALL VIDEO CARD dependant, CPU speed hardly matters; A 2GHZ Athlon 64 or 3ghz Pentium 4 will let you play the latest games at highest details if you have a powerful videocard. I found no difference between 2.4Ghz and 2.7ghz A64 in game-speed, the video card is the limiting factor, the CPU is plenty fast

Re:Latest chips, latest games & instant obsole

[digitaldc]

I guess what I was trying to say that most people can not afford to keep upgrading their PCs just to play the latest cool game (me included.)

I am all in favor of new tech, just don't forget the people who like new games, but may have old tech.

FPS are GPU dependant.

[Tei]

Often on videogames the limiter is the GPU, not the CPU.
Changes on CPU affect litte to nothing on FPS.

But maybe this will change?

If your game or engine required intense collision calculations, of phisic simulations. And this stuff its mostly poorly code with scripts. Or you need to compress/uncompress on-the-fly textures or sound. You will need that CPU horsepower.

Its also interesting how different RAM types and quantity of L1 cache affect this beckmarks. I think the information provides its too litte to really know anithing about the combo new games with new hardware.

Re:FPS are GPU dependant.

[DigiShaman]

While FPS increase scales much better with faster/new GPUs, it still does scale with faster CPUs as well...just not as dramatic. I can only assume these game engines are designed to have the CPU co-process whatever the GPU cannot handle at any given time. Sort-of like a mix of hardware and software graphic rendering I think. At least that's what past and present game benchmarks seem to point to.

Next gen Amd

[neuromancer2701]

I have a couple of question? To my knowledge the M2 AMDs are just the old 939s with the DDR controller changed to a DDR2 controller, so unless there is a massive improvement in memory management they can improve that much. I think the big change is going to happen when 1206 LGA comes out. I don't have any idea how much putting the PCIe bus in the processor will do but is got to be great for games. I don't understand why they had to do the 940 socket again they should have just jumped to 1206 but I guess they could not get that out in time. Does anyone know when the 1206 is supposed to come out? I think the Opteron is supposed to come early then the Athlon 64.


I need to upgrade my machine I am still running a socket A but at least it runs WoW. I will probably get a 939 when the 940 comes, hopefully that will be cheaper.

Re:Next gen Amd

[n00tz]

You're not the only one in that boat. My 3200+ is the weakest link in WoW now. When it's loading the instances or the world after I return from an instance... the CPU is put to full test. A recent upgrade to memory handed the baton to the CPU. I'm hoping on a good tax return this year.. New CPU/Mobo/VC and I'm seriously thinking about the 4400+ X2 Dual Core.

Re:Next gen Amd

[neuromancer2701]

I was thinking the same thing(about the 4400X2) I only have a 2200XP on Asus A7V8X-X

One question I was wonder how do the 165 and 170 Opterons(939) compare the 4400. The 170 is just 200MHz slower but over 50$ cheaper(newegg),any thoughts.

I upgraded to 1GB of memory as well that helped I a lot I have no idea how much the processor will change performance in Wow but I have not played it on someone else's computer to see the difference.

Re:Next gen Amd

[n00tz]

At work I have a server that has a pair of Opteron 265's (opterons built for 2 way communication) it does very nicely. of course, it also has 4GB of ram and a RAID array assisting it's performance.

I honestly don't think you'd notice much of a difference between the two. But I'm just gonna stick with the desktop CPUs for my desktop needs, and server CPUs for my server needs. Although I was VERY tempted to go dual 265's in my new rig after I assembled that server.

I know when I upgraded RAM to 1.5GB, my broadband connection, DSL 1.5Mbps, became the reason for latency in ironforge and now stormwind (because of the linked AHs) on frostmane server. but even then, it's not much.

Well no..

[Anonymous Coward]

It's not the same, because here the argument isn't on how to screw you.

Re:Well no..

[Damastus the WizLiz]

true but they are just as stubborn

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 :-)

Re:impressive benchmarks

[Deliveranc3]

Amd has recently gotten into some shenanigans releasing it's X2 line at $400+ while Intel kept releasing $250 chips.

Conroe is going into Dells Dell doesn't want to put 4800+s in he wants 3800s (equivalent)

AMD just managed to scrape out a few successful quarters selling these chips but the prices they are charging are absurd.

Mostly it's for marketting and new Fab plants... Fabs are understandable but AMD's increased spending on marketing is a huge turn off...

Last few systems AMD but Conroe does look promising, if you dig around you can find some benchmarks from overclocked Pentium-M chips... those offer pretty absurd performance compared to Pentium D chips and Conroe is based off of these with a higher thermal envelope.

Designing off a mobile chip 65 watts isn't too unreasonable and looking at the performance of old Pentium-M chips the performance is in line with p1-p2 levels of performance increase.

Looks like Intel may have pulled a Coppermine.

Re:impressive benchmarks

[jiushao]

Well, the tested Conroe E6700 is expected to retail at $529, whereas the tested Athlon64 (clocked at 2.6 GHz, the tested one is overclocked a bit) costs $1035. So even though the price of the AMD chip will no doubt go down a bit before then things still look very good for Intel.

It might also be interesting to note that the E6700 has a TDP of less than 65 watts, whereas the tested Athlon64 has a TDP of 110 watts. Amazes me that such a thing isn't brought up in the article, it is one of the factors we have decently hard data on and a clear advantage for Intel, whereas the benchmarks themselves leave a lot of room for arguing.

While one can argue a lot about the validity of the benchmarks it is quite clear that Intel is ahead in pretty much every possible metric with the information we have as of yet. It will be interesting to see how things sort themselves out this fall.

Re:impressive benchmarks

[Anonymous Coward]

and if the Intel is cheaper than the AMD ... ?

Re:impressive benchmarks

[Anonymous Coward]

if the AMD processor is 50% less than the Intel one, but only 20% slower, AMD will win it for me :-)

It makes more sense to compare system prices, not CPU prices. In your case, even if an AMD CPU is 50% cheaper, an AMD-based system may be only 10-15% cheaper. Then Intel will win in bang for the buck contest.

Re:impressive benchmarks

[bleughbleugh]

http://voodoopc.blogspot.com/2006/03/if-only-they- had-time-machine.html

some interesting points raised there

Faster, Cooler, Smaller, but

[layer3switch]

What good is it, if I can't play copies of my DVDs, MP3s and games on it? Will it even let me install whatever OS I want on it? Don't get me wrong, I love to have 20%+ performance boost, but I don't love it that much.

Give me freedom or Give me 486 (so I can run Slack on it)!

Re:Why the 200mhz OC?

[Anonymous Coward]

What's the point of OC'ing FX for that test? Maybe it wouldn't have made a big difference in numbers but showing higher clocked AMD proc get beaten by lower clocked Intel is big. I say rigged or at least tuned test to underline the speed difference with all they (Intel) could.

Is it 64-bit?

[MWales]

I haven't read anywhere if it implements the 64-bit extensions, and obviously haven't heard anything about the 64-bit performance. And I _do_ care since I'm already running Fedora Core 4 x86_64 on an Athlon64 3200+.

Re:Is it 64-bit?

[layer3switch]

old article indicated that it's actually x86 extension on 64bit processor, Conroe.

Retard

[Anonymous Coward]

The prices will eventually go fucking down. Suck it.

RAM settings....

[tinkerghost]

Am I the only one who noticed that the FSB speed on the AMD wasn't mentioned, and that the RAM settings were 2-2-2/1 for AMD and 4-4-4 for the Intel?
Call me cynical, but I would think that they could have found 2 boards somewhere that would have let them use the same RAM and FSB settings.

Re:RAM settings....

[Numbstruck]

I don't know enough about Intel to say, but it's my understanding that with the move to A64 they(AMD) technically don't have a FSB. So it's not really an apples to apples comparison, right?

Intro to A64 Architecture

Traditionally, a Northbridge existed between the memory bus and the CPU. The rate at which data is transferred between the memory and CPU is known as the front side bus. However, the Athlon64's memory controller is on-die, and as such, has no Northbridge, nor a front side bus. The Athlon64's have two independent buses; one between the memory and the on-die controller, and another bus that communicates with the other system devices- the HyperTransport bus. The CPU's clock speed is determined by the HTT speed multiplied by a clock multiplier, which is why it's often suggested to view the HTT as if it were the front side bus. However, this is about where the similarities between the two diverge. The HTT is, in fact, not a data path or bus, simply an internal time off of which the HyperTransport and CPU speeds are derived. The HyperTransport bus' effective speed is determined by an LDT(Lighting data transport) multiplier, multiplied by the HTT. Traditionally, the memory speed is derived off of the front side bus, and can be manipulated by FSB/memory ratios. In contrast, in the A64, memory speed is derived off of the CPU speed in CPU/memory ratios. This is why it's rather inaccurate to say that the memory is ever running "synchronously." The memory is always running asynchronously with respect to the CPU speed, off of which it's derived. How fast it's running with respect to the HTT does not matter at all. There is no latency hit in running the memory slower than the HTT. While the front side bus could've been traditionally double or quad-pumped, the HyperTransport's effective data rate can be anywhere from 1x to 5x it's speed on the CPU.

Re:RAM settings....

[sudorm]

DDR2 vs. DDR
DDR2 has much looser timings than DDR. Sorry, no conspiracy there.

Prices

[Namarrgon]

According to this, the Athlon FX60 was released in Jan '06 at a price of US$1031, in 1000-unit quantities. The next FX-series chip (the 2.8GHz version) will probably debut around June, at somewhere near this price.

According to this, the 2.66 GHz Conroe will be released in Q3'06 at a price of US$530, in 1000-unit quantities.

With these prices, combined with the apparent performance and power differences (Conroe has a predicted TDP of 65W, compared to the FX60 at 110W), it looks to me like we'll finally see some heavy competition from Intel. Of course, a lot can happen between now and then - Intel have had manufacturing issues in the past, AMD have a new memory controller on the way and a 65nm die shrink due early next year, and can probably squeeze out two or even three speed bumps before Conroe really hits. Who knows, they might even drop their prices a bit.

Come Q3, I'll be sitting in the ringside seats with popcorn, ready to watch the fun :-)

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:Latest chips, latest games & instant obsole

[het3]

Game developers are mostly still supporting GeForce MX cards on the low end, because there's only fifty squillion of them still out there. The upper end creeps ever upward, but the baseline for most games is still very, very reasonable. It's the best of all worlds.

How many cores?

[Anonymous Coward]

The Conroe comes in a quad core package as well as dual core. They never mentioned which one they were comparing to.

Re:How many cores?

[Cornelius the Great]

Intel has already stated that the Conroe will be dual-core only. The quad-core chip is called the Clovertown, and is due out next year (early 2007).

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)

So what

[octopus72]

Those chips aren't on the market (not even in 6 months) so it's very much vaporware. But if AMD won't be able to match intels performance, they will have to drop prices of their current generation CPUs and I'm looking forward to that (because cheap athlon xp's were best buy few years ago)

Re:So what

[Anonymous Coward]

With AMDs financial position, I _really_ doubt they want to lower their prices.

Re:Latest chips, latest games & instant obsole

[Anonymous Coward]

New motherboard/CPU/RAM every 4 years (keep the old monitor, mouse, etc), say $250. Upgrade the graphics card every 2 years (probably overkill), $100. $90 a year, the cost of 3 games - and if you don't buy 3+ games a year why upgrade anyway?
Buy not-quite-bottom end stuff, you get the best bang-for-buck and you'll be able to play anything on the market. I'm using an athlonXP 2600+ and it's still comfortably able to play anything that's around, I figure I'll do my next CPU swap when dual-core AMD gets below $120 - I might be able to use the same DDR400 RAM too, saving more cash.

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.

Let's all hope it is true

[Aaron Isotton]

You should not forget that Intel supplied both the hardware and the benchmarks. Obviously, they will only supply benchmarks where they win, and not the ones where the Athlon is better (if there are any). The F.E.A.R. benchmark seems to confirm that Conroe is really faster, but that's just one benchmark, which is not enough to convince me of Conroe's superiority.

That being said, I think it is in everybody's best interest if the benchmark results actually represent a real advantage; 20% more speed is indeed a big step forward.

Re:Let's all hope it is true

[Anonymous Coward]

The FEAR benchmark is actually the most peculiar, because FEAR is usually bottlenecked by GPU performance. The Conroe doesn't just do slightly better, it does significantly better. It does so in FEAR while its encoding performance for WMA and iTunes is only marginally better than that of the X2's. Those are actually CPU-bound tasks, and they're seeing =12% while FEAR is seeing a 23.8%-52% improvement over the X2. That should seem a tiny bit peculiar to anyone.

Re:Let's all hope it is true

[Anonymous Coward]

And if we look here, FEAR which is the most GPU-bound sees the most dramatic change (~47%) in framerate, while UT2k4 which is the most CPU-bound sees the least (~16%).

Re:Let's all hope it is true

[TooncesTheCat]

I'm not really jumping to conclusions as of yet on the gaming performance of the Conroe. Some games run better on Intel, some better on AMD. Who knows, maybe F.E.A.R is one of them. All I know is that most games are GPU intensive and not CPU dependent ( Half Life 2 is about the only game I can think of that is mainly CPU intensive ) :/

Re:Let's all hope it is true

[Lord Crc]

The FEAR benchmark is actually the most peculiar, because FEAR is usually bottlenecked by GPU performance.

Indeed. I have an AMD X2 3800+, and when overclocking my 6800GT, I saw a linear increase in framerate for each MHz I overclocked. 10% higher clock resulted in 10% more frames per second.

Re:Let's all hope it is true

[Anonymous Coward]

HL2 is GPU-intensive when used with cards capable of making use of its shaders. That's why ATI beats NVIDIA at HL2 performance and NVIDIA beats ATI at Doom 3 performance: the shaders in both titles have been designed with the shader execution models of the games in mind. FEAR is the most GPU-bound of the three titles, to the point where CPU speed matters significantly less to the end framerate.

CPU-bound games are pre-pixel shader games or games that barely make use of them like UT2k4, Serius Sam, Quake 3, Enemy Territory, Call of Duty, and so forth. If you look at Hexus's comparison of the games, the more GPU-bound a game is the greater the improvement we're seeing with Conroe over Athlon64. This is clearly of interest, because it suggests some manner of driver optimization at best, and at worst some manner of disparity in the hardware configuration of the two systems.

Re:Let's all hope it is true

[Anonymous Coward]

1 word: Memory.

Re:Let's all hope it is true

[Anonymous Coward]

All of the titles stress the memory bandwidth, but FEAR especially being most GPU-bound sees miniscule differences between various versions of the P4, Athlon FX, and Opteron equipped with the same GPUs despite differences in I/O and memory bandwidth. This Conroe has no more bandwidth than a P4 D EE.

Re:Let's all hope it is true

[Anonymous Coward]

Amusingly Anand has redacted the AMD FEAR scores (attributing them to human error) and replaced them with scores giving Conroe only a 20% lead. Hexus hasn't backed away from numbers suggesting the same degree of error in FEAR, which would I guess suggest replication of the same exact error.

Re:Latest chips, latest games & instant obsole

[armblessed]

I whole heartedly agree with you there. I've predicted for a long time that the console industry will be the provebial bullet in the heart to the PC gaming industry. There will be people who can afford the big gaming rigs and people who can't. Those who can't will buy consoles. Since there are more people who can buy consoles than big rigs, that moves market share attention to them instead of old timers who pour money into the bottomless money pit called our gaming machines.

How it effects the actual gaming titles are also up for grabs. Will there be a two tiered system where you have high end games for PC based platform and a lower end game for console platform? Will the economics of pandering to the diminishing PC market prove to be pointless since it's more profitable in numbers make games for the standard architechture of a console? Does the cost of creating games for the standard architechture of consoles prove more profitable than having to accommodate several hardware configurations of the PC market?

Personally, I think the console market is the "businessman's" way to go. I'm a die hard PC gamer but I see the writing on the wall.

WOW!

[ch424]

" hey, their stock financed the car i am driving right now "

You post on slashdot while driving? Whoa!

Re:Latest chips, latest games & instant obsole

[Anonymous Coward]

Yeah, because no game uses advanced physics or anything like that now.

Re:Most important question

[Anonymous Coward]

Anything can and will run linux with these ingrediënts:

1 basement
1 lack of (social) life
1 PC setup
1 inanimate object potentially runing Linux.

Re:Latest chips, latest games & instant obsole

[somersault]

I didnt need to buy a dual core chip to play Farcry. Admittedly I upgraded my machine to an Athlon 3000+/Geforce 6600 GT(OC) a year ago, but dual core is actually pointless for most games, and yes I should be able to play games over the next 5 years on that system, and not even at the lowest settings. I would expect to have to upgrade my graphics card again before upgrading my processor, though now that I am no longer a student, and have a job, then I could afford to buy a new machine if I thought it was actually going to be worth it. You're kind of right about buying consoles for games, apart from when it comes to games like Quake which benefit a lot from mods, and FPSs in general just suck with a joypad IMO

Re:Wait and see (faked tests)

[FirstOne]

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

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

Maybe %20 faster

[mkop]

But it will probably cost at least %50 more.

Re:Maybe %20 faster

[jmke]

No I don't think so; :)

How do Conroe, Athlon 64, and Opteron compare?

[olddotter]

I just don't know what the differences between Athlon 64 and Opteron. But what I am more interested in is how do new Intel dual cores compare to dual core Opterons.

That may not be a big difference if the difference between Athlon and Opteron is more marketing than substance. These days I have been considering Athlon CPU's as low end.

Re:How do Conroe, Athlon 64, and Opteron compare?

[Anonymous Coward]

"Compared with the AMD Athlon 64 FX chips, AMD Opteron chips are designed with three "HyperTransport" links for input/output and communication between processors, rather than a single HyperTransport link. Each HyperTransport interface provides up to 24GB per second peak bandwidth per processor to reduce I/O bottlenecks"

Source: http://www.extremetech.com/article2/0,1697,1915530 ,00.asp/
(2nd paragraph)

So would this be the 886?

[autopr0n]

Is this a new "level" of chip? Like the diffrence between the pentium (586) and P6-PIII (686) and the P4 (786)?

Mod parent up

[gr8_phk]

"And I _do_ care since I'm already running Fedora Core 4 x86_64 on an Athlon64 3200+."

I'm in the same boat. Nobody ever does 64bit benchmarks because you can't over in the Windows world. Taking the same code and compiling 64bit vs 32bit on the AMD parts can get a substantial performance increase. Running 64bit version on Intel parts (the few that can) generally degrades performance. It will be interesting to see how the 64bit performance of this new chip measure up - if it's even a 64bit part.

I personally don't care one bit about 32bit benchmarks.

Re:Mod parent up

[Anonymous Coward]

The difference will be much smaller. Intel seems to have done an incredible job of getting performance in 32 bit where the code does constant spills to memory because of the lack of registers. With twice as many registers (in practice more than twice since some have a dedicated function like the stack pointer and the compiler cannot decide to use it for anything else), the difference will be smaller since you don't run into spill hell on moderately complex code (also known as reload hell with GCC).

It also makes the compiler's job easier, which does not mean faster since you have to keep track of more physical registers.

Re:Mod parent up

[TWooster]

If you don't care one bit about 32-bit benchmarks, does that mean you care 32 bits about 64-bit benchmarks?

I'm confused.

Re:Latest chips, latest games & instant obsole

[jmke]

look up the facts before you add a worhtless comment.

Put brain into motion before mouth.

Tools and how you use them

[SuperKendall]

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

That's one scenario. Apple for instance is only using the TPM to make sure OS X will run on the boxes they sell, but it's not controlling my hard drive or stopping me from intalling Linux.

My car gomes with gasoline, a liquid capabile of setting a human being alight in a most gruesome manner. However I choose to use this tool to propel the car instead of immolating bystanders (despite my having played GTA).

You should never fear a tool itself, only improper uses of them. Nuclear weapons also bring us nuclear power.

Re:Tools and how you use them

[Anonymous Coward]

You're wrong... 1. iTunes DRM... and the future video downloads. 2. There's absolutely nothing technical stopping Apple from using the TPM in the Intel Mac to do exactly what was described. It's a simple software update. With the hardware, your Intel Mac is nothing but an expensive set-top box controlled by Steve "Disney" Jobs. I realise that you want to believe that Apple is a glorious ethical company, but it isn't... and it never was... get over it.

Re:Tools and how you use them

[Grishnakh]

You should never fear a tool itself, only improper uses of them. Nuclear weapons also bring us nuclear power.

No they don't. They're entirely different. Commercial nuclear plants use Uranium that isn't highly enriched, and the method they use to generate power is they have rods of uranium in a neutron-absorbing material (to control the chain-reaction). The heat from the rods boils water, which is heat-exchanged with other (non-radioactive) water to create steam, which drives steam turbines.

Thermonuclear bombs use highly enriched uranium, or plutonium, along with deuterium, along with an explosive mechanism to slam two pieces of the fission material together to cause an out-of-control chain reaction; this is just to cause a fission explosion to generate heat on the order of 10^7 K. This heat then in turn causes a fusion reaction with the deuterium, which releases even more energy than the original fission reaction.

The knowledge needed to create commercial nuclear power plants was all discovered with the initial atomic bombs of WWII. All the nuclear weapons research since then has had little, if any, usefulness for power generation.

If you want to make an argument for nuclear weapons being potentially useful, you might be able to say something about them coming in handy in case we detect an planet-killer asteroid on a collision course with us, although I'm sure a bunch of people will disagree with that too (too unpredictable; don't want to break up asteroid; rocket engines to divert course more effective; etc.)

Re:Tools and how you use them

[Alsee]

You should never fear a tool itself, only improper uses of them. Nuclear weapons also bring us nuclear power.

But there is a HUGE difference between the design specification for a nuclear weapon and a nuclear powerplant. The TPM is explicitly designed as an anti-owner nuclear weapon design, not a powerplant design.

Claiming that a TPM is just a "tool" that can be used for good or evil is an invalid argument. It is as invalid as citing the vitamins and nutrients in a poisoned apple. The fact that the apple has vitamins and nutrients is in NO WAY a valid defence of the poison pill in the apple.

If you want to support nuclear powerplants or poison-free apples, GREAT! I'm all for it. However that is NOT support for the TPM.

With a very tiny change in design the TPM could be changed from a nuclear weapon intoa nuclear powerplant, however the Trusted Computing Group prohibits the manufacturers from producing any legitimate pro-owner design.

Apple for instance is only using the TPM to make sure OS X will run on the boxes they sell, but it's not controlling my hard drive or stopping me from intalling Linux.


You do not correctly understand what the TPM is and does. The TPM does not stop you from installing Linux, not ever. The TPM does not control your harddrive, not ever (though one primary fuction is to lock selected files against you). The TPM is far more insidiously evil than that. The chip does not pervent you from running anything. No one would buy these evil Trusted Computing machines if they couldn't run any and all pre-existing normal software.

-

Thanks, that is good information.

[CFD339]

Its good on a number of fronts -- first, I've been leaning toward AMD for my near future machines based in part on heat and power use. These seem to show Intel in a good light -- which is hard to believe, but if true is a good thing.

Its also interesting that you point out the choice of ram will be up to board manufacturers. I wasn't considering that -- assuming that the processors would generally be driven by specific chipsets and thus tend to favor one type of ram over another.

Re:Thanks, that is good information.

[Anonymous Coward]

No, it's bad information from the GP poster.
Conroe has "only one hardware thread"? Puh-lease. Conroe is a dual-core processor. Hyper-Threading was just a half-way house to full multicore - a cunning way to maximise usage of the resources on a single-core processor, allowing two simultaneous threads to execute iff they required different resources (for example, integer and floating-point). Dual-core has no such restriction, so two full threads in hardware for Conroe.
Secondly, ignore the comments about FB-DIMMs. Desktop Conroe-based boards will use standard, non-buffered DDR2 DIMMs at 667 or 800MHz - readily available and inexpensive. High-end Server platforms will be the only ones using FBDIMMs for quite some time - they simply don't make sense on the desktop just yet.

Re:Thanks, that is good information.

[Namarrgon]

You're right of course, if a little abrasive. I assumed that everyone realised that Conroe was dual core. "Hardware threading" is a term I associate with Hyperthreading, which Conroe does not have. I should have said, "only one thread in hardware, per core".

I read somewhere yesterday, can't find it now, that Conroe platforms could use FB-DIMMs - but that may be just Woodcrest-based servers. I did say consumer boards would use ordinary DDR2. In any case, since the memory controller is in the motherboard chipset and not the CPU, Intel can pair it with whatever is appropriate.

Re:Thanks, that is good information.

[x2A]

"iff they required different resources (for example, integer and floating-point)" - this is a bit dated, modern processors have multiple execution units for most things. Hyperthreading was to allow free execution units to be assigned to another thread while the pipeline on the primary thread isn't at the point where it can send instructions to the execution units.

Hyperthreading was a cheap half way house *because* different threads of a process share the same address space, so don't need their own everything's, such as TLBs etc.

Multi-processors /are/ capable of running different processes, as each core will have it's own everything. But, because of this, smart enough OS's will try to keep threads of a single processes to one core (overflowing to the other when it makes sense to) to make use of the cached TLBs etc.

To cut a long story short, I'd say your "No, it's bad information from the GP poster" is unwarranted, the educated mind would class the chip as multiprocess, not multithread, and recognise that Namarrgon's post was a lot more informative and helpful than yours.

(unless anyone wants to tell me that both cores share the same TLBs???)

Additional info that might explain the results

[glop]

The memory bandwidth is one improvement in Conroe but there is also "Advanced Digital Media Boost" which Theregister.co.uk reported to bring single cycle SIMD instructions (whereas many SSE2 instructions on P4 have a throughput of 1 instruction every 2 cycles).

This means that the Conroe machine tested had :
1) >50% more memory bandwidth
2) 2x faster FPU/SSE2 at a given frequency.

This all makes the Conroe look good and promising. AM2 will bring an answer to point 1 (with DDR2 up to 800MHz).
As for point 2, AMD has yet to announce any improvement on the FPU/SSE2 part of their processors.

The facts currently available seem to indicate that Conroe might be better than AMD's offering when it is available.

Re:Additional info that might explain the results

[BobPaul]

As for point 2, AMD has yet to announce any improvement on the FPU/SSE2 part of their processors.

I wouldn't expect to see anything like that. AMD pretty much JUST added SSE2. I still don't have an AMD in any of my systems that supports SSE2...

The truth about LaGrande

[Wesley Felter]

As I explained yesterday, the TPM is not actually in the processor, but the processor has a few new features that allow it to cooperate with the TPM. If you buy a motherboard with a LaGrande-enabled processor but no TPM, LaGrande will not be able to work its evil magic on you.

Re:The truth about LaGrande

[Alsee]

Ah! Thanks for the link to the LaGrande specification. I read it over and it looks like it blows the whole Trusted Computing scheme wide open to a number of (relatively) easy hardware attacks. It looks like it might ...might... even be as easy as wiring a switch in across the right contacts to simply freeze out all TPM activity during an owner selected Trust-activation sequence. Then we toggle the switch to activate the TPM and let our software feed the TMP emulated configuration settings! Happy Happy!

If the TPM were embedded in the CPU itself it would have been a serious bitch to break.

-

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.

AMD K9

[rsborg]

I agree that it is time for AMD to get a "K9" out of the door

... AMD's K9... will you let it out the doggie door? Will it be more bark than bite? Man, the press will have a field day with that core designation :-)

Cue the AMD fanboys?

[Otis2222222]

I have been an AMD fan for a while now; my last couple of desktops used their processors, and the one I am using right now does as well. Having said that, I'd switch back to Intel in a heartbeat on my next system if they are faster and their price/performance ratio is good. Heck, I might even give price a pass if the benchmarks were really that much better (I like to build top of the line and upgrade every 2-3 years).

Good for Intel, I'm sure this is just the latest volley in an ever continuing war. So has Intel officially abandoned the so-called Megahertz Myth? I see that the system in this article ran at a slower clock than the AMD CPU. Interesting.

Re:Cue the AMD fanboys?

[Guspaz]

They officially abandonned the Megahertz Myth when they released the Pentium M years ago. Back then they were faced with the task of convincing consumers that a 1.7Ghz Pentium M was at least as fast as a 2.8GHz Pentium 4.

Conroe is simply the evolution of the Pentium M. From Banias to Dothan to Yonah to Merom/Conroe. The name of the product has changed as time goes on ("Pentium M" for Banias and Dothan, "Core" for Yonah, and possibly "Pentium 4" for Conroe), but the old Pentium 4 line is dead. Any naming similarity between Prescott P4s and Conroe P4s are just marketing.

It seems like Intel doesn't want to abandon the Pentium 4 moniker. They can't very well call Conroe "Core" like they will Merom, because they need a way to differentiate its higher performance from the mobile part. Personally I would like to think that Conroe qualifies as a Pentium 5 due to the completely different architecture. Heck, considering how the P4 has evolved over the years we should be at P6 or P7 by now. This time, though, the entire architecture is different, so calling it P4 is a bit silly.

Not really

[BobPaul]

Republicans and Democrats argue about stupid things like wars, taxes, pollution and crimes against humanity. We're arguing about stuff that matters: "My computer is faster than yours"

AMD is in big trouble...

[dtjohnson]

People have been waiting for Intel to destroy AMD with a better product ever since AMD came out with their Opterons, then their 'Venice' cores and then finally their 'X2' line of dual-core processors, all of which were much superior to the Intel chips. Intel already destroyed AMD a few years back when they released their Pentium 4 to compete with the original Athlon and everyone has expected the same thing again. That's probably why Dell has sat on the sidelines selling their aging, wimpy Celeron Ds and P4 systems at cheap prices.

Intel is a much bigger company, they have a lot more money, a lot of smart people, the nastiest, sleaziest marketers in the business, many more fabs, and great lawyers to fend off the AMD legal strikes too. The Intel 'Prescott' was supposed to do the job on AMD but it never came close. Now, though, the 'Conroe' looks like it is FINALLY the answer to AMDs stuff. Based on the benchmarks using Intel-supplied hardware and software, it looks like the 'Conroe' line of processors totally destroys the AMD FX-60 which is the fastest AMD processor sold today. Of course, you can't buy the 'Conroe' until September, 2006 but it will be worth the wait, based on the benchmarks anyway.

The only thing AMD has to offer is a little bit faster clock speed (aka FX62) and their upcoming AM2 socket systems which don't seem to do much of anything new other than allow DDR2 memory and a bigger cache. Looks like AMD is headed back to the bargain bin.

Re:AMD is in big trouble...

[Anonymous Coward]

> Looks like AMD is headed back to the bargain bin.

Good quality at bargain bin prices? Count me in!

Re:AMD is in big trouble...

[Sebastopol]

the nastiest, sleaziest marketers in the business,

Do you know any, or are you just making shit up? 'cause boy oh boy, those bunny people are so sleazy.

I think maybe you are confusing "marketing" with "volume pricing" and "price fixing for preferred customers".

Intel's marketing is painfully benign compared to just about every other industry, like the Marlboro (cigarrets are cool), or Kellogs (sugary cereal for kids is part of a balanced breakfast), or McDonalds, or British Petroleum (we care about the environment), or Ford (our new suv is "green"). Now THOSE are some sleazy marketing examples, lying to kids qualifies.

Re:Latest chips, latest games & instant obsole

[LurkerXXX]

Sorry, but I think you are entirely wrong there. Most folks these days have a PC at home. The only thing you really need to make your standard PC a 'gaming rig', is a decent graphics card. My old ATI 9600 runs everything out there right now, and it doesn't cost anywhere near the price of a new console. CPU? You dont' really need much. My aging Athlon 2400+ is more than required to play the latest games, and even 'cheap' new computers come with that much CPU power these days.

the benchmark is bs

[Truekaiser]

i would be saying the same thing if the tables were turned and it was amd supplying the hardware and the benchmarks. the review site has lost it's creditability in my eyes because of this. the bios on the amd mcahine they supplyed most likely doesn't support dual core athlon-fx's thus making it run with only one core which would acount for the lack of performance.

32-Bit faster than 64-Bit?

[dunc78]

I have read that 32-Bit applications typically run faster than 64-Bit applications (http://www.osnews.com/story.php?news_id=5768). As an owner of an Athlon64 3200 I am curious if you have verified that running the same code in 64-bit mode is faster. I was going to install 32-bit linux, but maybe this isn't the way to go.

Re:32-Bit faster than 64-Bit?

[higuita]

i can say that my system working at 64bits is ALOT faster in some cpu intensive runs than in 32bits...
i cracked some passwords, performed openssl tests, compiled things and tar+gzip large files and in all the 64bits won

cracking passwords with john gave more than triple the speed from 32bits

as for running my desktop setup, i cant tell a huge diference, but i also use slackware (slamd64 in 64bits), that is fast, fluxbox, no desktop environment, so i dont have much that isnt waiting more time for my input

that benchmark is probably bogus, its testing more the weak GCC support on nom-(x86|x86_64) systems

i reacall some years ago where openssl in amd64 and ppc was slower than in 32bits, due big optimizations on the last, and lack of then in the former... after some months, with the latest gcc, the difference was huge

64bit eats more memory and hd, but have no reason to be slower than 32bit, taking out only the cache bounded tests, where the bigger size of the 64bits may hit a little the performance

also, some cpu intensive programs take alot of advantage, but most of then is just a small increase (5-10% IIRC)

Re:32-Bit faster than 64-Bit?

[MWales]

On my homepage I have some benchmarks of my system running Fedora Core 4 x86 and x86_64. It seemed like most apps would either be just slightly slower (~5%), or quite a bit faster (~30-40%). I'm pretty sure the UT2004 benchmark is wrong on my site, ignore it.

Wow, just wow

[Anonymous Coward]

I'm seeing angry emotional responses to

Besides, who actually runs anything serious on a hpme

Re:Wait and see - OFFTOPIC

[FirewalkR]

Regarding your sig, I just wanted to say that Vernor Vinge rules :)

Don't noun that verb

[iliketrash]

'As far as we could tell, there was nothing fishy going on with the benchmarks or the install.'

"Install" is not a noun.

Re:Don't noun that verb

[Anonymous Coward]

Don't noun that verb

"Noun" is not a verb either.
http://m-w.com/dictionary/noun

Re:Don't noun that verb

[Courageous]

"Install" is not a noun.

I beg to differ! Consider:

"He Richard Nixon'ed the tapes".

Perfectly legal conversion of a proper noun into a verb. Interesting stuff, eh?

Perhaps you don't know the English language as well as you think you do? I suggest you consider a course in Linguistics...

C//

A couple of points on that

[Namarrgon]

The price you quote for Conroe is not a retail price, it's in 1000-unit quantities, as with all manufacturer pricing. The Newegg price is single-unit retail. I can't find figures on Conroe's die size; it's not small, but at 65nm it's likely smaller than the FX60. Still, Conroe is clearly going to be cheaper, barring something unforseen.

Regarding TDP, AMD usually quote maximum potential thermal dissipation, whereas Intel usually quote "typical" dissipation, around 75% of maximum, which would make the FX60's TDP only 82W by Intel's reckoning (Intel claim a TDP of 115W for their 3.6GHz P4, which is clearly a lot hotter than the FX60). Conroe is still coolest, but the difference is not that drastic.

AMD & IBM have been making good strides on their strained-silicon SiGe-based 65nm process, so it should provide performance & power improvements decent enough to keep AMD competitive when they get it into volume production early next year. We'll see how that turns out.

No iNTELL based Power Macs

[Joseph_Daniel_Zukige]

Obviously.

They'll call them Mac Pro or something, following the lead of MacBook Pro, but I'm pretty sure they will drop the "Power" from them. Why everyone seems surprised about this I have trouble understanding.

There's a certain irony in the name change, but that's what people get for using trademarks with too much semantic burden. (How heavy the irony will prove to be remains to be seen. The market momentum may save iNSMEL again.)

And I still think Apple had to silently deliver higher clocked G4s in the Mac Mini last fall simply because they were having trouble getting the lower clocked ones.

Re:Latest chips, latest games & instant obsole

[armblessed]

I don't deny that *if* I was referring only to the hardware requirements to convert a machine to a game rig (though we can split hairs by saying a RAM upgrade is needed as well). I'm referring to a larger economic picture.

Just the fact you post here means you have some idea how to upgrade your machine or know that is what you need when you decide to ask someone to do it for you. We are not representative of the general population of the US in this aspect. Most people do not have the knowledge or even care to have it but what they do recognize is the price difference between a console and a game machine. People generally do not make informed choices.

Most folks in the US do have a computer in their home (68% of them do) but to earn that other 32%, they're going to have to drop the price to pander to the poor who can't afford $1,000 or even $299 bucks for a computer. Not to mention the addional costs of operating system, software, upkeep, and repair. The standard economic number for tapping into that demographic is $199. Unless PCs can drop to that baseline price, PCs will never hit that 100% saturation. Think of the way almost every household has a Telelvion. It's an appliance. No thought involved. The corporations objective is saturation of an emerging market with their company's platform. This then sets the playing field for them to focus in on content which helps secure longevity.

Consoles because of their price, their appliance-like nature, and their standard platform all make buisness sense to invest in long term. In every market there will be those who who demand quality from every subtle aspect of their passion and there will be those who don't really care just as long as they get it. The pandering to those who are already "in the know" regarding gaming is coming to a close. The big corps are now casting an eye to those who are intrested but either can't afford it, don't know what to look for, or would rather have a system that is more like an appliance.

For the record, I'm not a proponent for this belief system. It's just the way business is done. We can protest all we want because the choices that the general populace make do not reflect what we, as afficiandos, know as the best choice. I've simply learned to profit from the uninformed choices the world makes to fund the high prices they charge to those of us who do make the informed choices.

You're a fucking moron

[Anonymous Coward]

N/T

Re:You're a fucking moron

[rlo1346]

Hhahaha................Bam! Why the anonomouse? WE know its you!??!

Re:Latest chips, latest games & instant obsole

[LurkerXXX]

I understand that no everyone has the knowledge to know to just make sure they get a computer that they can upgrade the video card in, but I'm lost on your point about only 68% having a PC. If the other 32% can't affort a $299 computer, I think buying a $299++(with all the accessories you need) gaming console is also going to be out of their budget.

TPM is just a lock

[SuperKendall]

Claiming TPM does not prevent you from running Linux, ever, shows a basic misunderstanding of the technology. It's simply a lock that can be applied. If a motherboard maker chooses to do so, TPM can be used to lock out other OS'es that do not have the proper key. Do some research; this was the basic fear of TPM when it was first unveiled.

Apple has used it differently, and Microsoft intends to use it to prevent unauthorized applications (like, perhaps, Open Office?) from running. In reality they probably would not prevent Open Office from running but that's how it can and will be used.