Intel Demos New P4 'Extreme Edition'

typobox43 writes "Louis Burns of Intel displayed a "high-definition video stream running on a 'mystery' desktop processor." This processor turned out to be the new Intel Pentium 4 Extreme Edition 3.20 GHz, with an extra 2 Megabytes of cache."

Comments

Possible Advertising Campaign?

[The_Rippa]

Saturday. Saturday! SATURDAY!

At Intel Headquarters!

Witness the unveiling of the next...

Biggest!

Meanest!

Fastest processor you can imagine.

Pen-Pent-Pentium EXXXXXTREME

It's 3.2 gigahertz of binary badness.

Come witness as it peforms calculations at mind-boggling speeds!

Special Guest The Blue Man Group

Tickets start at $20 for adults, discounts for children and seniors

If you miss this, you'd better be dead... or in jail...

And if you're in jail, break out!

Re:Possible Advertising Campaign?

[L-Train8]

Extreme? Whenever I hear that word in an advertising campaign, I think of Homer Simpson as Poochy the Pooch: "Hey kids, remember to recycle... to the extreme!"

Re:Possible Advertising Campaign?

[m00by]

dude, you so TOTALLY forgot the most important part: you pay for the whole seat, but you'll only need the EDGE!!! =D

Re:Possible Advertising Campaign?

[Mike Hawk]

$20 for the Blue Man Group? Tickets for the Vegas show are like $90. Thats too good a bargain not to scalp them.

Re:Possible Advertising Campaign?

[nrmrvrk]

Special guess celebrity: Randy Macho Man Savage:

Kid: "Pentium chips aren't an extreme sport, Macho Man..."

MM: "PENTIUM CHIPS NOT EXTREEEEEMMMMEEE!?!?! OOHHHH YEEAAAAH!!!!"

Re:Possible Advertising Campaign?

[inkedmn]

you forget free snow cones for the kids

Re:Possible Advertising Campaign?

[MrLint]

Pentium Extreme! its just as good as having a 64 bit CPU!

Re:Possible Advertising Campaign?

[Anonymous Coward]

Pentium EXXXXXTREME*

*extras Xs may result from occasional floating-point errors.

Re:Possible Advertising Campaign?

[GMFTatsujin]

Change "pay for" to "license" and you've got a deal!

Re:Possible Advertising Campaign?

[Anonymous Coward]

Um, more cache helps every process on the system. 64 bits doesn't help anything on the desktop because right now no desktop application needs it. Even if it did, 64 bits would not help performance at all. So I would say that Pentium Extreme is much better than a 64 bit CPU, especially for the Pentium Extreme target market.

Re:Possible Advertising Campaign?

[Ibag]

When I hear the word extreme in an advertising campaign, I think of this. God bless Maddox.

Re:Possible Advertising Campaign?

[jdehnert]

You forgot...

BE THEEEERE!!!

Re:Possible Advertising Campaign?

[TiMac]

Maddox forgot all about Apple's use of "Extreme" already...

Quartz Extreme, Airport Extreme, and probably more that marketing hasn't told us about yet.

I was just PRAYING earlier in the year that they wouldn't call the 970-based Power Macs "Power Mac Extreme" to follow their recent nomenclature...

Re:Possible Advertising Campaign?

[otis wildflower]

It's BOOOOLLLLLDDDD!!!!!

Re:Possible Advertising Campaign?

[Doktor Memory]

...powered by SURGE!

Re:Possible Advertising Campaign?

[foog]

oh, it was kind of cool once. In 1992 or so...

Re:Possible Advertising Campaign?

[greenstork]

Dude, I'll bring my street luge and Mountain Dew, I'm there man

Re:Possible Advertising Campaign?

[innosent]

...Pentium Extreme is much better than a 64 bit CPU, especially for the Pentium Extreme target market.

What, you mean for people who run 32-bit code?
The cache is something that's been sorely needed though, maybe Intel's finally going to make decent processors for people who need to run more than Word and Madden 2004. The problem is, they're stuck with the x86 instruction set architecture, which is crufty at best. Maybe once people start buying Itaniums (if ever), they'll get better.

Tips for Intel designers:
1) Get rid of 20-stage pipeline, it's too long for anything serious.
2) As a follow up to 1, try to actually get some work done in a clock cycle.
3) Throw out the x86 ISA.
4) Look at the MIPS ISA.
5) Realize that it's actually possible to understand the MIPS arch, and that it still works great for multimedia, math, and general use.
6) Buy the rights to the MIPS ISA, make small improvements (get rid of branch delay slot, load delay slot), speed it up, and design new Intel processors from the improved ISA.
7) Release versions of processors with 4MB Cache (2MB each I$, D$) for consumers, and 24MB Cache (8MB I$, 16MB D$) for servers/clustering/etc.
8) Release Motherboards for 1, 2, and 4 CPU configurations.
9) ...
10) Profit!

Re:Possible Advertising Campaign?

[Anonymous Coward]

Intel has managed to design and produce the fastest processor in the world for integer ops, the P4. What exactly are your qualifications in this field, that you feel so compelled to tell them how it's done?

Re:Possible Advertising Campaign?

[Anonymous Coward]

Intel has managed to design and produce the fastest processor in the world for integer ops, the P4. What exactly are your qualifications in this field, that you feel so compelled to tell them how it's done?

Dont forget the fastst processor in the world for fp ops, the Itanium 2. Simply stunning SPECfp scores.

Re:Possible Advertising Campaign?

[grolschie]

err... dude. It doesn't matter how much cache they strap to these processors, deep down you'll still know it's a P4. :-(

Re:Possible Advertising Campaign?

[jea6]

Well, for one thing, he rated 7 +5 comments on Slashdot in his past 24 posts, most being either Interesting or Informative. If that isn't qualified around here, I don't know what is.

Re:Possible Advertising Campaign?

[cheezedawg]

1) Get rid of 20-stage pipeline, it's too long for anything serious.

No its not. In fact, according to this research, the P4 pipeline is not deep enough. That paper concludes that P4 performance could be improved by up to 90% by increasing the pipeline depth to around 50 stages and increasing the cache size.

Do you actually think that Intel didn't know the consequences of increasing the pipeline depth? The Intel engineers didn't just guess on the P4 architecture- it was a very deliberate design decision. Judging by the P4's performance gains, it was a pretty good decision, too.

Re:Possible Advertising Campaign?

[Anonymous Coward]

Man, I sure hope those intel designers are reading this page so they can take these great tips from 'innosent'. I bet they never thought about any of those things before. Good thing they have a smart guy like innosent to enlighten them. He is, afterall, one of the leading experts on CPU design.

You are a moron.

Re:Possible Advertising Campaign?

[fitten]

Says a lot about folks who mod, that's for sure...

Re:Possible Advertising Campaign?

[Anonymous Coward]

His posts are full of the pseudo-technical babble that mods take at face value. He throws buzzwords around that any college freshman studying EE has learned, and he does that to pander to the wide-eyed zombies that think that /. is the tech industries bible.

Here is what I am talking about, point by point:

1) Get rid of 20-stage pipeline, it's too long for anything serious

This is hardly an original thought. He probably read some review at amdr0x.com or intelsux.com about the Pentium 4 pipeline, so now he thinks he is an expert. Nevermind that this is still largely a matter of opinion.

) As a follow up to 1, try to actually get some work done in a clock cycle.

Hm- sounds like he is making a clever observation about the Pentium 4, but it ignores the major design goal of the pentium 4 that makes it possible to scale the clock frequencies at an unbelievable rate. These clock increases have more than made up for the decreased per clock efficiency.

3) Throw out the x86 ISA.

Once again, hardly an original thought. In fact, Intel has been trying to do that since 1981 when it tried to release the iAPX 432 processor and more recently with Itanium. However, the economics of just "throwing out" x86 makes this statement meaningless.

4) Look at the MIPS ISA.
5) Realize that it's actually possible to understand the MIPS arch, and that it still works great for multimedia, math, and general use.

Ooooh- he must have taken Computer Architecture 101 to learn about MIPS! He is so l33t.

And what makes him think that Intel hasn't looked at MIPS?

6) Buy the rights to the MIPS ISA, make small improvements (get rid of branch delay slot, load delay slot), speed it up, and design new Intel processors from the improved ISA.

WOW! He actually knows some specifics about MIPS! Intel should listen to him!

7) Release versions of processors with 4MB Cache (2MB each I$, D$) for consumers, and 24MB Cache (8MB I$, 16MB D$) for servers/clustering/etc.

Yes- increasing cache is good. However, the on chip cache already takes up almost 50% of the die area, and releasing chips with such bloated cache sizes would lead to huge expensive chips with very low yields. Hold your horses until the 90 or 65 nm processes are in full use.

8) Release Motherboards for 1, 2, and 4 CPU configurations

This is trying to appeal to the socialist anti-corporation "everything should be free-as-in-beer" types that frequent slashdot, but if Intel can make more money by keeping the product lines seperate with value added features like MP, I say more power to them.

Yeah- he must be really qualified.

Re:Possible Advertising Campaign?

[Anonymous Coward]

But for Apple, "Extreme" snowy white smoothness and candy-colored liquid-jelly UI elements doesn't bring to mind monster truck rallies.

Re:Possible Advertising Campaign?

[Anonymous Coward]

Make sure to say it like the moster truck ads: Intel EXTREEME PERFORMANCE!! SUNDAY! SUNDAY!

Re:Possible Advertising Campaign?

[shokk]

Whatever. You're only renting everything anyway. You only think you own it. Licenses only piss you off because you own it in recurring payments. Not much different from a concert seat.

Re:Possible Advertising Campaign?

[Salamander]

1) Get rid of 20-stage pipeline, it's too long for anything serious.
2) As a follow up to 1, try to actually get some work done in a clock cycle.
3) Throw out the x86 ISA.
4) Look at the MIPS ISA.
5) Realize that it's actually possible to understand the MIPS arch, and that it still works great for multimedia, math, and general use.
6) Buy the rights to the MIPS ISA, make small improvements (get rid of branch delay slot, load delay slot), speed it up, and design new Intel processors from the improved ISA.

Looks like someone needs to read Hennessy and Patterson. The deep pipelines are there for a reason. The branch and load delay slots are there for a reason. Those are things compilers can deal with, and it wouldn't be possible to reach these clock rates without them. Let's review the old formula, shall we?

WPS = WPI * IPC * CPS, where WPS = work per second WPI = work per instruction IPC = instructions per cycle CPS = cycles per second

Deep pipelines are critical for improving both IPC and CPS, and delay slots help too. It might seem like you're sacrificing WPS, but studies have shown that such an effect is minimal; processors spend practically all of their time implementing the simple instructions anyway for most code. If you increase two factors significantly, and reduce the third only slightly, the result is - obviously - a net win. History has proven that H&P knew what they were talking about, and that their analysis played a pivotal role in achieving the sorts of performance we see today.

Yeah, the x86 instruction set sucks. Oh well. Unless you write compilers there's no reason to care (no, not even for embedded or high-performance work, which I've done). If you do write compilers, think of it as job security. Compatibility with existing applications really does matter more than aesthetics.

Re:Possible Advertising Campaign?

[mrm677]

1) Get rid of 20-stage pipeline, it's too long for anything serious.


No its not. It enables a high clock rate and with good branch prediction and selective replays, it is just fine.


2) As a follow up to 1, try to actually get some work done in a clock cycle.


Read some studies about the available ILP (instruction-level parallelism) in common applications. There really isn't much of it unless instruction windows are made huge which isn't feasible. This is why simulaneous-multithreading (hyperthreading) made it into an actual chip because it takes advantage of low ILP.


3) Throw out the x86 ISA.


Pentium4 is the leader of SpecINT. Not AMD, not Sun (RISC), not IBM (RISC), not MIPS (RISC). Some players, such as MIPS, don't have the resources to compete however IBM does.

And look whats happening to Itanium? Disaster even with its oh so elegant ISA.


4) Look at the MIPS ISA.


What about it? Yes, very clean and orthogonal. Intel and AMD have proven that an ISA is irrelevant to achieving high performance.


5) Realize that it's actually possible to understand the MIPS arch, and that it still works great for multimedia, math, and general use.


Realize that undergraduate computer architecture is simply an introduction.


6) Buy the rights to the MIPS ISA, make small improvements (get rid of branch delay slot, load delay slot), speed it up, and design new Intel processors from the improved ISA.


Unnecessary. Besides, I like software compatibility and binary translation just doesn't work at this level.


7) Release versions of processors with 4MB Cache (2MB each I$, D$) for consumers, and 24MB Cache (8MB I$, 16MB D$) for servers/clustering/etc.


2MB each for I$ and D$? Then you must be referring to L1 caches which are the only ones typically separated into data and instruction. And how many cycles would it take to access those caches?


8) Release Motherboards for 1, 2, and 4 CPU configurations.
9) ...
10) Profit!


Read Intel's annual report. They are quite profitable already and don't need the advice from someone with a B.S. in EE or CS and marvels at how great the MIPS ISA is.

Re:Possible Advertising Campaign?

[Anonymous Coward]

Mod parent up!!!! Very true!!

Re:Possible Advertising Campaign?

[Jugalator]

Hmm, but it would take a heck of a long time to fill up that pipeline so it was used to 100%, and it seems hard to make optimal use of it. Imagine a simple failed branch prediction. Eeeeempty pipeline and fiiiill it up again. :-S

It places high demands on the rest of the architecture to make an optimal use of such a super long pipeline.

But maybe it's worth it then, even when considering the problems with it.

Re:Possible Advertising Campaign?

[innosent]

Yes, the deep pipelines ARE there for a reason, to do a ton of integer and floating point operations as fast as possible (so that SPECint and SPECfp scores are high). Now, back in the real world, we have real programs, and real programs branch. If the branch prediction fails, or if there is an unconditional jump or function call, the rest of the pipeline has to be thrown out. Wasting 19 clock cycles is bad, especially if you have a situation (like simulation) where branch prediction is unusable or usually fails, and you have 5 or 6 instructions, then a jump. In 26/27 clock cycles, you executed 6/7 instructions (assuming no fp, which could delay the pipeline further). This may seem like an extreme example, but it really isn't when you consider that most real programs deal with input of some sort.

As for the branch and load delay slots, yeah, they needed them when the MIPS ISA was set, but since the R10K, branch delays have been unnecessary. HP's PA-RISC doesn't use a branch delay, either (it just starts executing the next op, and throws it out if the branch is taken). Load delays will happen, but if the Memory stage is split for cache hits/misses, and cache is increased, the delay can be significantly reduced, and the result could be forwarded back to the execute stage, with maybe a half-cycle delay, rather than the current complete cycle.

And finally, no, compatibility doesn't matter more than performance/efficiency (depending on use), and the x86 ISA limits performance/efficiency. The Pentium 4s may be fast, but they are far from efficient. Market factors being what they are, it's all academic, but there really are serious performance problems with Intel's chips, it's just not economically practical to use the alternatives. And yes, I've read H&P, and they're mostly right, except that once the economics of speed reach a certain point (which happened probably 2 years ago), the processing power available allows large scale data processing (data mining, graphics, simulation, intelligent network filters, [anything that has a lot of data to make decisions on]) to be feasable, and there is an increase in the density of jumps in code. With more jumps, there are more pipeline stalls, and the longer pipelines multiply this problem. For a lot of these applications, a 400MHz MIPS processor will run circles around the latest 3.2+GHz Pentium.

Re:Possible Advertising Campaign?

[Anonymous Coward]

Aha. But if branching bothers you, then you should probably have a look at the Itanium architecture, which tackles branching in a completely different manner.

Re:Possible Advertising Campaign?

[Carewolf]

4) Look at the MIPS ISA.

Why?? When they already owns the rights to the Alpha architecture, which is like:
MIPS eXtreme!

Re:Possible Advertising Campaign?

[Anonymous Coward]

If there wasn't the problem of "Eeeeempty pipeline and fiiiill it up again" there would have been no optimum pipeline length. Just make it as long as you can (which has limits too, since you can't really do a 32 bit register to register XOR in -say- 200 steps.)

Re:Possible Advertising Campaign?

[Anonymous Coward]

I'm with you 99%.

Re:Possible Advertising Campaign?

[Salamander]

back in the real world, we have real programs, and real programs branch. If the branch prediction fails, or if there is an unconditional jump or function call, the rest of the pipeline has to be thrown out.

In that real world, how often do those branches occur? "Every eight instructions" was never really true except for the most CISCy architectures, and the number has only increased since then. Would you rather have a bigger bubble at 3GHz, or a smaller one at 1GHz while also reducing parallelism for code that doesn't branch as often? Do the math. That's the real lesson of H&P: do the math instead of making wild-ass guesses about what works better.

As for the branch and load delay slots, yeah, they needed them when the MIPS ISA was set, but since the R10K, branch delays have been unnecessary. HP's PA-RISC doesn't use a branch delay, either (it just starts executing the next op, and throws it out if the branch is taken).

Speculative execution is a waste of functional units that could be used for real work instead of solving what should be the compiler's problem.

there is an increase in the density of jumps in code

Got any sources for that, Sparky? When you factor out the effects of things like abusing dispatch tables because it's "too hard" to deal with C++ semantics or JVM interpretation the right way, the frequency of jumps in real code written for real performance is going down.

Re:Possible Advertising Campaign?

[pmz]

When I hear the word extreme in an advertising campaign, I think of this.

Damn that was funny. He should try some of that tie-died ketchup, next.

Re:Possible Advertising Campaign?

[jdray]

There are two words in techno-marketing that really turn my off: Extreme and Ultra.

Okay, I guess the latter is a prefix, not a word, unless you're Sun, who, laughably, has a whole series of "Ultra" things. I guess things over there just keep getting ultra-er.

Re:Possible Advertising Campaign?

[Anonymous Coward]

3) Throw out the x86 ISA.

Pentium4 is the leader of SpecINT. Not AMD, not Sun (RISC), not IBM (RISC), not MIPS (RISC). Some players, such as MIPS, don't have the resources to compete however IBM does. And look whats happening to Itanium? Disaster even with its oh so elegant ISA.

That's odd. The spec.org website shows that both the Opteron and Itanium2 beat the P4 in SpecINT. Only these two have a higher SpecINT peak over 1300. Or maybe I missed one. Please double check for me under the CPU2000 benchmarks for specINT (not specINT_rate). As for Itanium being a disaster, well that just shows you only seem to be interested in the desktop marketplace, not the real world. Funny how a P4 needs more than 2x the MHz to perform less work than a Itanium2. Go ahead, pick a benchmark. ANY benchmark. Just be sure to compare a 1.5GHz pentium against a 1.5GHz Itanium2.

And please don't confuse clock speed with performance. Intel and AMD have proven that an ISA is irrelevant to achieving high clock speed.

Re:Possible Advertising Campaign?

[Anonymous Coward]

get a Mac, it's just as good as having a 64-bit CPU. oh wait, it DOES have a 64-bit CPU, or two if you like:-)

Re:Possible Advertising Campaign?

[Anonymous Coward]

Man, you should scalp them. Then, with a few sales, you could buy yourself an N-Gage or two. Have you thought about THAT??

I mean, seriously - Tony Hawk Pro Skater on that excellent device looks about a million times better than THPS4 on the PS2. The N-Gage fucking owns!

Re:Possible Advertising Campaign?

[jea6]

I was introduced to a word that describes the parent post well. sarchasm: The gulf between the author of sarcastic wit and the person who doesn't get it.

Aah! My P4!

[Anonymous Coward]

I'm not supposed to get jigs in it!

Re:Aah! My P4!

[Anonymous Coward]

that's "jizz"

maddox influence?

[edrugtrader]

they must be reading maddox's site

Re:maddox influence?

[cant_get_a_good_nick]

Of course, this is my favorite page on his site

Re:maddox influence?

[Anonymous Coward]

I don't like the way Mr. Sun is looking at that testicle at all.

Re:maddox influence?

[Anthracks]

Damn, I was hoping you were talking about Tommy Maddox, one and only MVP of Vince McMahon's eXtreme Football League and quarterback for the Los Angeles eXtreme! Let's see how much trouble I get in for trying to Slashdot my school...

I have a 3.2Ghz PC that I bought for home...

[Osrin]

... I struggle to tax it with anything I do, including some of the more intensive games.

This "extreme" version of the chip has to be aimed at a very niche market, at least for the next couple of years until more processor intensive software catches up.

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

I struggle to tax it with anything I do

Oh, you must work for the IRS.

Re:I have a 3.2Ghz PC that I bought for home...

[Fulcrum of Evil]

This "extreme" version of the chip has to be aimed at a very niche market

Yes, it's the 'mine's bigger' market, though I wouldn't call it niche, exactly.

Re:I have a 3.2Ghz PC that I bought for home...

[MerlynEmrys67]

The scary thing is that I was listening to "Stuart" right now as I read your .sig. Talking about being scared

Re:I have a 3.2Ghz PC that I bought for home...

[twoslice]

... I struggle to tax it with anything I do

Try taxing it with something a computer would do...

Re:I have a 3.2Ghz PC that I bought for home...

[Rosco P. Coltrane]

I struggle to tax it with anything I do, including some of the more intensive games.

Try "nice -n 20 dd bs=1 if=/dev/zero of=/dev/null" ...

Re:I have a 3.2Ghz PC that I bought for home...

[Kenja]

I can tax any computer out there by loading up the source code for my last large project (1+ mill lines o code) and hitting compile. Sure it only takes a min and a half on my current compiling system (Quad 550 Xeon with 2 megs cache per CPU) but on a faster system it would take less time.

Re:I have a 3.2Ghz PC that I bought for home...

[ergo98]

"until more processor intensive software catches up"

Let's see...DVD-RWs are now around $130 US...DVD-RW discs are about $2 each...converting two hours of miniDV digital video (yeah the DV is redundant...suck it) to MPEG2, or divx...adding some effects...

Yeah, I'd say that market is here, thanks.

Re:I have a 3.2Ghz PC that I bought for home...

[iosmart]

like the people who wear the "my computer is bigger, better, and faster than yours." shirt? yeah, i have one of those...except it was back when my 1.6 athlon-mp was crazy fast. i woulda linked to the shirt but it looks like thinkgeek has stopped selling it :-(

Re:I have a 3.2Ghz PC that I bought for home...

[timeOday]

Try video editing. You will be annoyed by the slowness of your shiny new chip.

Re:I have a 3.2Ghz PC that I bought for home...

[jawtheshark]

Uhm, I dunno about you.... But if my compiles take more than a minute, I just go out and take coffee... chat with the cute gal from marketing and eventually pull some practical jokes on my coworkers.

Of course, other people actually work at work.

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

The market that requires that whole extra 2, COUNT 'EM, TWO MEGABYTES of cache! So that gives it, what, 4 total? Wow. That's only 512k less than my 2 year old Mac.

I can't WAIT.

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

According to research at an English university, people will always buy whatever the guy in the shop says is "best".

Re:I have a 3.2Ghz PC that I bought for home...

[Rainier Wolfecastle]

OT, I know, but damn, the Dead Milkmen rock the veritable casbah. Nice quote.

Re:I have a 3.2Ghz PC that I bought for home...

[be-fan]

Try compiling KDE. My 2GHz P4 struggles to do it in under a working day. Heck, it takes nearly a minute just to recompile a KDE theme after making a change to it!

Re:I have a 3.2Ghz PC that I bought for home...

[Osrin]

The OS that came on the PC was already compiled... so no need, or maybe I'm just behind the times.

Re:I have a 3.2Ghz PC that I bought for home...

[Mitchell Mebane]

Try MPEG-4 encoding. With all the XviD settings at max, I get ~4 FPS on my Athlon XP 2400+ when transcoding DVDs at full 16:9 resolution.

Re:I have a 3.2Ghz PC that I bought for home...

[jerkychew]

Dude... most excellent Dead Milkmen reference.

"You know what Stuart? I like you. You're not like the others here, in the trailer park..."

Re:I have a 3.2Ghz PC that I bought for home...

[be-fan]

KDE is just an example. Any large C++ program would do. My point was that developers are one of the catagories that need large machines.

Re:I have a 3.2Ghz PC that I bought for home...

[mOdQuArK!]

Obviously, you haven't been running many nuclear reaction simulations - oops, I wasn't supposed to mention that online :-S

On a (slightly) more serious note, I bet a couple hundred simultaneous video-rendering jobs will bring your system to its knees. If you aren't taxing your system, then you haven't really been trying.

Re:I have a 3.2Ghz PC that I bought for home...

[homer_ca]

OH NO! He found us out! Gentoo Linux is a conspiracy by the CPU makers and electric companies to make us feel inadequate about our processors. Taking all day and night to compile Gnome, KDE and Mozilla just for a point upgrade from x.xxx.xxx-r1 to x.xxx.xxx-r2 is sure a pain in the ass. How many kW-hours you think we burned up with our 80W Athlons and P4's just for that?

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

try running Crusin' USA in MAME. You still wont have enough power.

Re:I have a 3.2Ghz PC that I bought for home...

[Khaki_Dockers]

I hate to rag on someone like this, and actually it's an enormous group of people from my estimation, but the numbnuts that'll feed intel with this latest, greatest processor, are the people who did it the last time, and the time before it. a HUGE segment of gamers think that cold cathode lights will improve performance, because IT COSTS MONEY!! they think a 200 dollar P4 1.7 will beat a athlon 2.4 because the P4 costs more, etc. etc. people buy expensive things, they get ripped off. it's never been a niche market, it never will.

Re:I have a 3.2Ghz PC that I bought for home...

[Glonoinha]

All a matter of time.

I remember reading once in a Usenet thread - some guy was trolling and asked 'will my P90, overclocked to 100MHz, be enough to handle the flight combat simulators you guys are discussing?'

The first time I read it it was hilarious because he was either bragging or dreaming, the P90 chip was out in limited supply at the time and was easily 50% faster than the common P60 machine used by the sim-gamers, not to mention the overclocking it. Of course it was going to be fast enough.

The second time I saw it (a few years later) it was hilarious because the bare minimum system for any sim/game was a PII/300 with a 3D graphics card and his P90 was so pitifully underpowered it didn't have a chance.

So we get to enjoy the 'is this CPU enough' question twice, generally, for any given CPU. Just a matter of timing.

Re:I have a 3.2Ghz PC that I bought for home...

[Glonoinha]

I am not even going to pretend to know what kind of read/write caching is available on the Linux developers platform, but on Win2000 machines I have found (particularly when doing massive C++ compiles) that adding a ton of memory to the system (ok not exactly a ton, but upgrading from 128M to 1152M) makes a WORLD of difference - Win2k uses the extra memory to cache (read/write, configurable in the settings under the device manager) writes back to the drive and then flush them all out at once. If 1G doesn't seem to be doing it, crank it up to 2G. If you have more than 2G of source code for a single compile ... I will buy you a beer or something.

Granted a faster CPU doesn't hurt, but a 1.2GHz box with a Gig of RAM is going to outrun a 2.4GHz machine with 128M of RAM.

Re:I have a 3.2Ghz PC that I bought for home...

[bestguruever]

Judging by the spam I get, I'd have to agree its not a niche market

Re:I have a 3.2Ghz PC that I bought for home...

[be-fan]

Linux has the same mechanism --- its called the page cache. Anyway, this machine has a good amount of memory (640MB). However, memory isn't the bottleneck for the KDE build. kdelibs, for example, is only about 90MB. g++ actually pegs the CPU most of the time. Compiling C++ is just plain slow. Compiling GNOME (about as large, but in C) takes 1/4 the time.

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

I was thinking about trying out "Tux Racer", however I'm not sure if my 4-way AMD Opteron 2GHz and GeForce FX 5900 combination is quite powerful enough to handle it. I don't want to waste all of that time downloading it if I can't even play it, after all I am still using a 2400 baud modem.

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

I've done lots of video editing older hardware (P3 600Mhz and Athlon 800MHz) even, and there is no problem with speed.

Now raytracing and rendering are a different story. Pull up Terragen with all detail options turned on rendering a complex scene at 1600x1200. On my P3 laptop it took 1 hour 45 minutes, on my friends G4 Mac the same scene took over 16 hours (and yes, it is running OS X).

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

Wow, Apple has a processor that runs at a WHOPPING 2GHz, that's only like 500MHz less than my 2 year old PC.

Re:I have a 3.2Ghz PC that I bought for home...

[Anonymous Coward]

Just play it in UltraHLE. My old K6-400 w/Voodoo 3 ran Cruisin' USA and Zelda 64 flawlessly using that emulator.

Re:I have a 3.2Ghz PC that I bought for home...

[nusuth]

I want my beer. Please verify by compiling openoffice.org Although the source itself is not 2GB, make process creates more than 2GB worth of temporary files (with gentoo ebuilds but I don't think it is any different on any other linux.) Add that the memory OS and compiler, linker etc. uses, source files and target files, the process may benefit from as much as 3GB memory.

Obligatory Waynes World

[Znonymous Coward]

Extreme close up! Whhoooooooooooo... Whhoooooooooooo.

Re:Obligatory Waynes World

[Anonymous Coward]

I assume you have windows on your E-machine. Look up the properties of your sound in "my computer" - Device manager. Post the sound chip ID, as well as the IRQ and DMA channels used.

Re:Obligatory Waynes World

[Znonymous Coward]

I use Linux and Mac OS X but thanks for asking.

Level Three Cache

[Master Bait]

Ho hum. I suppose if it was level two cache, Intel would have said so very loudly, so they just call it 'cache'.

Re:Level Three Cache

[Anonymous Coward]

Don't knock it, even if its operating at 1/3 the cpu bus, an extra 2MB of level 3 cache will give a significant boost to things like video games and many other interactive cpu intensive applications.

Re:Level Three Cache

[Gherald]

If it performs anything like this, then I'd say its useless and hardly worth the premium price its sure to cost.

With September 22nd drawing nigher, what the hell is Intel thinking?

Re:Level Three Cache

[Anonymous Coward]

Actually that review is not really going to apply to the P4"Extreme" since the bus speed will not be hampered like the xeon MB they tested. Also the P4 will be hyperthreaded which the Xeon was not.

The xeon IS actually faster clock for clock if those variables are moved. It's just that with the P4 being 50% faster on clock speed, HT enabled and a faster memory buss, the Xeon was looking pretty poor extra cache or no in comparison.

So instead, picture the current P4 with the added cache. It wont double the speed or anything, but it'll be a welcome improvement for hardly much more than the current top end p4.

But let's just wait for the benchmarks before prematurely praising or dismissing it then shall we?

Re:Level Three Cache

[Gherald]

I realize there are some significant differences between the P4 and Xeon test setups, but I still think if you take the ratio of L3 Xeon performance to the non-L3 and use it to calculate the performance of an L3 P4 based on the current non-L3 ones... you can have a very good idea of how well the L3 P4 will perform.

Re:Level Three Cache

[lederhosen]

I think it is L2 cache.

Re:Level Three Cache

[lederhosen]

No, it is L3, you are right. I am a moron!

Re:Level Three Cache

[Anonymous Coward]

Well, since at Intel caches, start at zero (L0, L1, L2), that Level 3 cache is technically an L2 cache. :)

Re:Level Three Cache

[philthedrill]

The tradeoff with cache hierarchy is access time on hits vs. size.

You could increase the size of your L2 (or L1 for that matter), but you do this at the risk of sacrificing cycle time. This is part of the reason the P4 has such a tiny L1 D-cache.

With clock speeds climbing higher, the amount of time for a signal to traverse across a chip is no longer trivial, so retrieving data within N clock cycles is unrealistic with a large cache.

To add to that, the benefit over an L3 hit (even though it's much slower than L1 or L2) is that it's still much faster than main memory. DRAM is built for capacity.

Adding cache is somewhat of an easy way out in terms of adding performance with your transistor budget. You keep your power density reasonable and you also don't change the microarchitecture.

In conclusion, I think drugs are bad. The end.

Re:Level Three Cache

[Jeff DeMaagd]

Cache on that order of size only helps for memory or bandwidth intensive applications. If the data and instructions for any particular thread fit in 512k, then you get no benefit for that thread.

Re:Level Three Cache

[Anonymous Coward]

Xeon does have Hyperthreading.

Re:Level Three Cache

[timeOday]

Don't knock it, even if its operating at 1/3 the cpu bus, an extra 2MB of level 3 cache will give a significant boost to things like video games and many other interactive cpu intensive applications.

...or not. Check Dual Xeon Duo: What Good Is the L3 Cache? at Tom's Hardware.

"...the top-model Xeon MP with a 2-MB L3 cache and 2.8 GHz costs $3692. "
...
"Our benchmark tests, which exactly reproduce workstation usage with appropriate applications, show that the additional L3 cache hardly adds any speed. Popular 3D software, such as 3D Studio Max, Cinema 4D and Lightwave, show no improvement in performance."
...
"To anyone faced with the choice of the dual Xeon 3.06 with or without the L3 cache, we would recommend the version without the additional cache. The additional cost is not justified."

Multiprocessor?

[tinrobot]

The rumors are that this chips are the same or very similar to the $4000 Xeon MPs with 2MB cache. I wonder if these will work on the workstation class MP motherboards. Would be sweeeeet.

Re:Multiprocessor?

[Anonymous Coward]

No it won't. "Extreme edition" will use Socket 478 while Xeon MPs use Socket 603/604.

BTW, the chip isn't similar, it's _exactly_ the same as XeonMP.

Re:Multiprocessor?

[loopWork]

Different pinout, so no.

Re:Multiprocessor?

[Anonymous Coward]

Exactly the same, minus some fuses blowin in manufacturing to disable MP that is...

Re:Multiprocessor?

[homer_ca]

Even if the packaging is different the chip die might still be the same. From the Intel engineer point of view, there's no R&D involved at all. Slap a Xeon 2MB core into a Socket 478 package. From the hobbyist end user point of view, there's really no way to repackage this core to fit in a Xeon server board.

Re:Multiprocessor?

[Wesley Felter]

Exactly. What a beautiful example of artificial market segmentation.

Re:Multiprocessor?

[Anonymous Coward]

Actually, Intel is planning to release a 2MB Xeon in the cheaper segement too. My guess is that the prices will be within a few bucks.

If anything is "segmented", it's the chipsets. Xeon runs on conservative boards with ECC, P4 Gamer systems are more 'xtreme'.

Re:Multiprocessor?

[DoctorRad]

Intel made this mistake once with the first-generation Celeron chips. These were, IIRC, based on the Pentium Pro core, and it was found that they'd double-up just fine. The Abit BP6 motherboard lets you run two of these very nicely.

I don't think they'll make the same mistake again.

Matt...

Interesting, but...

[JLSigman]

...of all the people who are going to run out and buy it, who REALLY needs that much juice? This'll just turn into another case of "I gotta have the latest greatest biggest baddest" whatever.

Re:Interesting, but...

[Anonymous Coward]

Did you just copy and paste this post from the last /. article about a new processor? Or maybe the one before that? Or the one before that?

Re:Interesting, but...

[agent dero]

You'd be suprised, in CompSci at school today, one kid was talking about an acrylic PSU!!?!

Then another kid asked if they had acryllic hard drives, they then proceded to talk for 15 minutes about how cool that would be, and how bitchin acrylic cases are.

Whatever happened to buying a computer because it...it.. _works_

Re:Interesting, but...

[Anonymous Coward]

I think its a stretch to call your evening classes at ITT Tech "Computer Science". Try going to a real school.

Re:Interesting, but...

[NanoGator]

"...of all the people who are going to run out and buy it, who REALLY needs that much juice?"

I do. I'm a 3D artist, I do rendering. However, as an artist, the processor must be economically feasible as well.

Am I going to go buy one now? No. But if Intel keeps pushing chips with huge caches, and assuming these caches make a difference in LW, it will affect whether or not my next machine is an AMD or an Intel.

For the record, AMD won the last round. I have a dual Athlon at home that runs circles around a 3.2 gig P4.

Re:Interesting, but...

[militantbob]

I need that much juice. My Windows machine typically handles mIRC, Yahoo Messenger, 3D Studio Max, 3 or 4 IE instances, etc.

Of course, most of those programs are essentially idle at any given moment. But when I'm trying to render a massive 3dmax scene while switching over IRC to ramble libertarianesquely about the failings and dangers of big government, while at the same time opening/reading 3 or 4 web documents... my machine bogs down on me. Now, this machine is a P4-2.9GHz with a gig of RAM on SCSI disks... perhaps the extended speed and cache on the new CPU would make a difference.

At the same time, I could use one of these on my colo box, which is hosting 17 domains with about 3,000 pageloads per hour. Then again, I could always get something other than x86, if I weren't a cheap and ignorant bastard.

Re:Interesting, but...

[Anonymous Coward]

So here we go with "AMD64" == "Kick Ass!" "G5" == "Kick Ass!" "P4Extreme(Ugh)" == "Why would anyone need that much power?"

But I'll bite. Real 3d work would be bogging if processors were 50 times faster than they are today.

Various encoding and other real-time DSP work that still require very expensive hardware (More expensive than the PC typically by several times).

Etc.

Re:Interesting, but...

[Graspee_Leemoor]

" I need that much juice. My Windows machine typically handles mIRC, Yahoo Messenger, 3D Studio Max, 3 or 4 IE instances, etc."

3 or (gasp!) 4!!!! instances of IE?!?!?!?

Dude, you are XTREEEEEEEEEEME!

graspee

Re:Interesting, but...

[geekoid]

of course not, but that statement was true about every processor ever released to the civlian market.

Re:Interesting, but...

[benzapp]

I am pretty sure 3D rendering is not going to benefit from additional cache too much though.

Re:Interesting, but...

[commodoresloat]

...of all the people who are going to run out and buy it, who REALLY needs that much juice?

Some of these people even believe they need more than 640K of RAM.

Re:Interesting, but...

[Anonymous Coward]

But when I'm trying to render a massive 3dmax scene while switching over

Could you address this by lowering the priority of the render process? It would get a little less CPU so the render would take longer -- but OTOH your lightweight mIRC and browser processes would get more CPU faster and therefore be more responsive.

I could use one of these on my colo box, which is hosting 17 domains with about 3,000 pageloads per hour.

Would bigger CPU cache really help this? Web serving is mostly just read file, send file over network, so I would think that bigger disk cache (on the drive controller itself, plus more system RAM) would be a bigger help here.

Re:Interesting, but...

[wondafucka]

I can think of an example:

If you are using a PC to emulate a real time system, that cache would help out in keeping your processes from having to access the slower memory. That means a few extra operations per unit time.

More specifically, if you are doing real time wave modeling that would mean a slightly more complex waveform. Shure you could use a DSP, but if you are in the development stage, it's usually easier to use a desktop system.

Another example would be if you are approximating physical hardware at high speeds. That's a few extra logic gates.

I'm not going to buy one, but if I had the money I would consider it.

Re:Interesting, but...

[Anonymous Coward]

Why not? I don't do 3D rendering (used to) but I am curious.

Re:Interesting, but...

[lannygodsey]

3 or 4 IE windows? Have you tried Avant Browser? It's an IE wrapper than gives it tabs, blocks popups and a few other things IE should be.

I have to use IE for some things @work and this makes it O.K. http://www.avantbrowser.com/

Re:Interesting, but...

[DigiShaman]

Ever see the inside of a computer loaded with dust and stuff? Now imagine looking at that nappy box a few months later after you mod up your rig with an acrylic case. YUK!!!

Re:Interesting, but...

[Glonoinha]

A faster CPU would not be nearly as beneficial in your case as a dual CPU setup (or potentially a hyperthreading setup.)

None of the programs individually are going to go any faster, but the system as a whole is going to go a lot faster, and feel more responsive also.

Re:Interesting, but...

[Anonymous Coward]

Holy Cow, I didn't realize Windows was THAT boated!

I run 7 or 8 Mozilla, 4 or 5 ssh/telnets to other hosts, Xmms and BitchX for IRC. While doing this I can also switch between either gimp or povray while writing docs in OOoRC4 and vmware running a WinXP session into work (requires Windows for the VPN connect).

All on a measly P3-800 with all SCSI drives, granted I was purposely torturing it when I had all of this open, but slow-downs were insignificant.

Man I'm never going back to WinSLOW.

Re:Interesting, but...

[0123456]

"I have a dual Athlon at home that runs circles around a 3.2 gig P4."

Why would you buy an Athlon machine for Lightwave, when it's so SSE2-intensive? Even my P4-2.4 beats an Athlon 3200+ in most Lightwave rendering benchmarks I've seen. Or were you using another machine for Lightwave?

Re:Interesting, but...

[militantbob]

Yeah, really.

A couple years ago I had a Linux machine... it ran qmail, wu-ftpd, apache, ircd, sshd, 2 or 3 console sessions, X, enlightenment, xchat, netscape, gaim, and a few xterms...

And it was only a p233mmx with 64mb.

There's something really wrong with the contrast between that old machine and my 'newer, better' one.

Re:Interesting, but...

[relativePositioning]

Here is a technique that every 3D person should know. Open up the task manager, click the processes tab, find the 3D app's process, right-click on it and set priority to below normal or low. This will give priority to your other wimpy apps but will cost the 3D application almost no processor time.

Re:Interesting, but...

[militantbob]

btw. i'm not sure what's so funny about what i said. someone wanna explain?

XTREME

[Anonymous Coward]

Great, another company that doesn't just sell a product, it sells an EXTREME product.

Re:XTREME

[Recovery1]

Oh, but it sounds so much cooler then saying "Brand new product we are hoping you'll be suckers enough to buy."

Ohpleaseohpleaseohpleaseohpleas

[JoeLinux]

I wanna see major competition between Intel and AMD. That way I can get my 875P motherboard "tossed in free with the purchase of any Intel Pentium 4 Extreme(tm) Processor." It's about time I upgraded from a Celeron 433 anyway. Ghost Recon plays more like Ghost Recon: The Slideshow.

Joe

Re:Ohpleaseohpleaseohpleaseohpleas

[bhtooefr]

You know, Intel MIGHT just do that - the CPU's cost $700 for 1000 units, and it shouldn't cost too much to throw in a cheap board - whereas AMD x86-64 ain't cheap (keep in mind, the P4X is a P4 with more cache - the P4 is rather common). Put it on a cheapo board, and it doesn't cost that much more.

BTW, have you tried overclocking? My 466 Celeron runs fine at 507MHz. Also, your mobo SHOULD take at least an 800MHz CPU (if the maxfsb is 66).

Imagine a beoGKRRRZTTFFFT

[Rosco P. Coltrane]

[NO CARRIER]

Re:Imagine a beoGKRRRZTTFFFT

[CrackHappy]

Oh man, you're showing your age there.

I bet you can whistle a 1200 baud modulated signal in your sleep too.

Re:Imagine a beoGKRRRZTTFFFT

[geekoid]

I can, but only through my nose, if I have a cold, and I am asleep.

Re:Imagine a beoGKRRRZTTFFFT

[Anonymous Coward]

This is both amuses and saddens me at the same time. Thanks for the brief trip down memory lane.

Tom's Hardware reviewed a similar Xeon...

[tugrul]

... with 1MB of L3. The results weren't that exciting.

Re:Tom's Hardware reviewed a similar Xeon...

[ciroknight]

But at the same time, the Level Three Cache is MUCH "further away" from the core in the sense that it takes much longer for data to travel accross the lines of the processor to get to it. Level Two isn't much closer, but that little edge does make a huge difference in this case. Game developers now have room to seriously push their applications because the processor will be able to cache more (data||instructions). It should vastly improve scores on very memory intensive apps.

On the other hand, I would much rather see them quadruple the size of the Level One Cache. This would improve performance on these processors, but at the same time, without the extra registers that a 64-bit chip would have, these improvements are limited by their usefulness, not to mention they would take up loads more valuable core real estate. I can't wait to see Intel move to a 64-bit chip with a 2 meg level 2 and maybe a 128k level one... we'd start to see chips FLY....

Re:Tom's Hardware reviewed a similar Xeon...

[Naito]

Increasing L1 cache on the P4 would render the other caches almost useless, because the P4 uses an inclusive cache architecture. All the data in L1 is stored in L2 also, and everything in L2 is in L3 also. So if L1 was big, L2 would end up being just a copy of L1 with nothing new, unless they made it bigger. And if they made ANY of it bigger, they'd start incurring some major latency, dropping the P4's IPC even more.

Re:Tom's Hardware reviewed a similar Xeon...

[Jeff DeMaagd]

I am curious to learn why. Last time I've seen comparisons (it might be at AnandTech), a 2.0GHz Xeon with 2MB cache compared well with the 2.8Xeon with only 512k, and I think both had the same FSB.

Re:Tom's Hardware reviewed a similar Xeon...

[Anonymous Coward]

Thing is that the bigger the cache the longer it takes to look stuff up in it. So if Intel were to quadruple the L1 cache, it would probably at least double if not quadruple or even octuple the access time to the L1 cache. Say right now L1 is 1 cycle away (I'm just guessing, it could easily be 2 or 3 cycles) then you'd be looking at a min of twice that or 2 cycles away up to 8 cycles. Compared to main memory access, that is still orders of magnitude faster, but now ALL accesses have to go through this slower inferface. When you consider that the average hit rate for L1 cache is probably in the neighborhood of 96% that means that you've now at least doubled the time required to access memory for 96% of accesses. Sure, one miss is still orders of magnitude slower, but you've now at least added an average of one extra miss for every 100 instructions if not worse.

So, all in all, the moral of the story is that cpu cache sizing is very, very tricky and seemingly useful changes can easily hurt your average case performance by a significant amount.

Re:Tom's Hardware reviewed a similar Xeon...

[Malfourmed]

Game developers now have room to seriously push their applications because the processor will be able to cache more (data||instructions). It should vastly improve scores on very memory intensive apps.

Yeah, but will it vastly improve my scores on very memory intensive games?? :D

----
Writing contest - $100 in prizes - entries close 28 Sept 2003

Re:Tom's Hardware reviewed a similar Xeon...

[akuma(x86)]

Computer architecture 101.

Average memory access latency per memory access =
(L1_hit_rate * L1_hit_cycle_time) +
(L1_miss_L2_hit_rate * L2_hit_cycle_time) +
(L2_miss_L3_hit_rate * L3_hit_cycle_time) +
(L3_miss_rate * DRAM_latency)

80-95% of your accesses will hit the 8k L1 in typical applications. This is the vast majority of the accesses. The latency of this cache is TINY on a P4. Do the math for a 3.2GHz 3 cycle cache.

Given a curve of cache-size vs. latency and hit rates for all the cache sizes, the optimal hierarchy is a simple optimization problem. I can assure you that this equation has been solved and the optimal heirarchy has been chosen (given the other constraints of obviously die-size and power).

Quadrupling the L1 will double the latency and kill your average access time, making your chip almost certainly slower.

Bigger caches mean longer latencies. It's limited by the basic laws of physics. There's only so much distance you can traverse in a ceratin amount of time and larger caches have longer distances (meaning higher RC delays).

The reason we want larger outer level caches is because the DRAM_latency is enourmous and has an impact on average access time. Hardware prefetching can also help to alleviate this problem - This solution is available on both Athlon and P4 chips and will only get better in the future because it is absolutely critical to hide this DRAM latency.

Ok, now to address the notion that more registers will improve performance...
You won't get as much performance out of more registers as you might think. First of all, when the compiler runs out of registers it spills the excess to the stack -- pushing it out with a store (spill) and reading it back in with a load (fill).

In modern processors (just about every chip out on the market), there is the concept of store buffers. Each store writes it's data to a store buffer. Subsequent loads that require data from stores, get their data by forwarding out of the store buffer. So -- the spilled store writes the buffer and the fill load reads the buffer -- all of this happening much faster than a memory access because it's just reading out a local on-chip buffer, so the load looks more like a fast register read. This architectural trick emulates the effect of having more registers, subject to the size of your store buffer. There are even more advanced architectural tricks you can play to completely eliminate the spill-fill pair from the critical path (look up memory-renaming in the literature).

If you're worried about chip-real estate, you should be very concerned that a 64-bit application's pointers will take up twice as much space effectively making your caches and memory bandwidth appear smaller.

Re:Tom's Hardware reviewed a similar Xeon...

[Anonymous Coward]

While somewhat true this is also quite misleading. Think about this. How many applications at any one time are running on your machine, 3-4? On mine there are 25. Sure many are just sitting there waiting. But guess what happens when they 'wake up'. Forget your whole cache. It is now garbage. It will basicly flush the whole thing while it reloads your program. Now what if you had a bigger cache? That is where you will see a bigger benefit. Not on the 1 program 1 test. You will see it on 2 programms running at the same time doing different things. Most of those 256k memory graphs all go to hell when you put in 2 or 3 apps all trying to get at memory all at once.

I belive we will not see anything really cool till the whole memory subsystem is on chip. Removing connectors and chips is what saves us money. But it also has a side effect of shortning the electrical path and boosting performance (moores 'law').

You will see instead things like 'write back buffers' and the like to handle what cache does now once memory is on chip. The pin count on the cpu should drop dramaticly thus lowering cost even more.

I belive currently the reason the opteron is doing better because it has that memory controler built in not because it has a bigger cache. 1 less link in the memory chain. 1 less thing to have a HUGE latency. Once all these things are all in 1 chip (it will happen) it will be very interesting then...

Re:Tom's Hardware reviewed a similar Xeon...

[Anonymous Coward]

hah, quadruple the L1 cache size?? you kidding?that would make the cache slower, and intel might have to decrease clockspeed and do more work per cycle!

Re:Tom's Hardware reviewed a similar Xeon...

[akuma(x86)]

Lets look at some numbers...

What is the standard time slice quantum for windows and linux typically? That is to say, what is the typical rate of context switches? If I recall correctly, it's on the order of 100 per second.

That's 1 context switch every 0.01 seconds. Lets suppose now that I have a typical P4 system with 6.4GB/sec of memory bandwidth. I can fill the entire 2M cache in roughly

0.002/6.4 seconds = 0.0003125 seconds

That's only 3% of the entire time slice quantum! That's assuming the thread will want all of the cache (which is unlikely for many apps).

So, yes you may be running 25 programs, but they only switch between each other 100 times a second. The cache re-fill only takes 3% of a minimum quantum, leaving 97% of the time left to read the filled data out of your caches (which is highly likely due to the principle of locality and LRU replacement policies).

Now, if you're running something server-like TPCC with a bazillion threads, then you'll probably context switch a lot more, but the P4 isn't designed for transactional servers. This is why servers systems have larger caches and multiple processors and multiple SMT threads within each processor -- among many other reasons :)

Putting the DRAM on die is a nice idea, but it's not cost effective at this time. You only have a finite die size. The cost of the die increases exponentially with area, so there's a hard cap on how big you can make the chip. The area that you would have spent on the DRAM could be used more profitably on other chip features. Perhaps in the future when we get many more transistors, it may be feasible.

Re:Tom's Hardware reviewed a similar Xeon...

[Jah-Wren Ryel]

Why do you even accept his premise that a context switch causes a cache flush? That's just silly. So his process wakes up, and runs a short loop or two that hits maybe 5-10 pages and then goes back to sleep. That's easily less than 64K that needs to moved back and forth for most wakeups. Trivial.

Re:Tom's Hardware reviewed a similar Xeon...

[akuma(x86)]

I was trying to show a worst case analysis to show that even cache flushes don't matter as much as he/she would claim.

Re:Tom's Hardware reviewed a similar Xeon...

[Anonymous Coward]

It's different level cache. Xeons will come with up to 2MB L2 cache. The cache talked about in the article (and Tom's review) is L3. L2 cache is FAST; L3 cache is much farther away.

text incase of /.ing

[Anonymous Coward]

Intel Developer Forum Cache for questions

By Nebojsa Novakovic: Tuesday 16 September 2003, 18:14
WHEN, AT today's IDF opening, Louis Burns demonstrated a high-definition video stream running on a "mystery" desktop processor, everyone must hve thought it was the upcoming Prescott part. Wrong! It was the (also upcoming), previously unheard of, even at The Inq, Intel(R) Pentium(R) 4 processor Extreme Edition 3.20 GHz , with an extra 2 Megabytes of pron. In Intel's own words, "this new processor will be targeted at high-end gamers and computing power users."

As a matter of fact, 2MB cache will help a lot those users whose apps (including games and such) have a lot of big cache-friendly *wink* pieces of code and data, but probably not the data-streaming intensive stuff. I do expect to see speedups anywhere from 2% to 20% depending on the application, maybe some more if using multithreading/multitasking (large cache can keep in code / date pieces from more threads).

However, this doesn't seem to be a new CPU in reality - after all, Intel is doing very well with its XeonMP 2.8 GHz 2 MB cache CPU, and how much effort does it really take to repackage it for the 3.2 GHz / 800 FSB desktop with less stringent thermal and reliability requirements than the big iron, anyway?

Intel would gain a lot with this move. If, touch wood, there are problems with Prescott, a large-cache Pentium4 part will provide some buffer against large-cache Athlon64 (i.e. rebadged Opteron) parts. At the same time, enormous extra benefits from the economies of scale would further reduce the identical die XeonMP manufacturing cost, helping Intel compete better on the quad-CPU server front as well. Interesting move? I think so. Let's see how the beast performs in real!

Re:text incase of /.ing

[Covener]

with an extra 2 Megabytes of pron.

2 Megabytes of pron?

Re:text incase of /.ing

[johny_qst]

yeah, shouldn't that be with an extra 2 Megabytes of pr0n

Re:text incase of /.ing

[Davorama]

Yes, makes that 'touch wood' reference take on a whole new meaning....

Lone voice...

[nacturation]

A lone voice pipes up from the back of the room, "Get a Macintosh!"

More impressed with AMD.

[Thinkit3]

What I'm really not impressed with is Intel saying desktop users don't need sixty-four bit. Well, we don't need gobs of cache. We need sixty-four bits.

Re:More impressed with AMD.

[C. Mattix]

The difference is that all the existing apps would need to be recompiled to fully use the 64bit. Even lowly DOS can use performance improvements with a larger cache. And with Hyperthreading the number of clocks per instruction is very small, this lends itself to using a larger cache more often.

See also:

Ars Technia on Caching

Re:More impressed with AMD.

[Anonymous Coward]

> We need sixty-four bits.

Why does a 512MB desktop computer need 64 bits?

Re:More impressed with AMD.

[Aadain2001]

Why do we need 64 bits? Are you running into the 4GB memmory limit the current 32 bit line of CPUs has? Do you personally need terabytes of RAM? Because that's about all you get when you move up to a 64 bit processor. Sure, they also throw in a few optimizations so it appears to go a bit faster doing common applications, but there is no real benifit to desktop users. There are real benifits to servers and data centers, which is why Intel aims their 64 bit Itanium products towards that market instead of the desktop. A large amount of cache though will produce real benifits that everyone and everyone program can use. AMD is gambling on the 64 bit on the desktop, its not a for sure "we need this" thing. That's why Intel isn't doing it. If it was a definate need, Intel would have done it already.

Re:More impressed with AMD.

[Anonymous Coward]

I want to mmap /dev/hda for my 40Go HD !

Re:More impressed with AMD.

[benzapp]

Thats not true at all. X86-64 allows for virtual 386 processors just as the 386 allows for virtual 8086 processors. You will be able to run non-emulated virtual machines very easily on the Athlon64. You could theoretically run multiple versions of Linux simultaneously on the same machine.

At the very least, more memory will be available for 32-bit applications, as the OS will be able to give a single application the whole available block.

This is all no different than running DOS apps on OS/2 back in the day. THe average DOS user had to work at keeping the minimum amount of memory available, but with OS/2 each DOS session had the full 640K available, I was able to run 3 DOS sessions with BBS software running simultaneously without any trouble whatsover.

The point is this if you want to run different operating systems concurrently, this will be great. It will perform much faster than VMWare or VirtualPC. If you run apps which need a lot of memory, and they are 32-bit, you will instantly benefit by using a 64-bit processor.

The real benefit of a 64-bit processor is not just the processing power, but the available memory. This was of course, the main benefit of the 386 for at minimum 10 years. And thats only if you consider Windows 95 to be a 32-bit OS.

Re:More impressed with AMD.

[Atmchicago]

Lots of people tote 64-bit as the giant leap into the future, that will make eveything a million times faster/better etc...

Yeah, 64-bit is where we are headed, and a good thing, but it's not like computers don't work fast running 32-bit.

Re:More impressed with AMD.

[Anonymous Coward]

You get the dumbass of the day award.

Please note- he didn't say that you need to recompile an application to run it on an Athlon64. He said you need to recompile an application to use 64 bits. Your entire rant was offtopic and unnecessary.

Re:More impressed with AMD.

[Anonymous Coward]

The P4 FSB has 36 address signals on the FSB- it is trivial to enable up to 64 GB of system memory on a P4 while using 32 bit virtual addresses for every process (so no one process can use more than 4GB).

Re:More impressed with AMD.

[Anonymous Coward]

For what?

Seriously, which desktop applications are likely to be improved by 64 bits? The main applications that I see being improved are database applications, and large engineering apps. Some video editing applications may be slightly improved, but I can't picture it being a large improvement. Graphics work won't gain hugely; they can get gains from going to large SIMD values, but not so much from 64 bit integers.

Re:More impressed with AMD.

[drinkypoo]

You know, there WERE other operating systems which ran on the 386. SCO Unix for example :D

Re:More impressed with AMD.

[be-fan]

The difference is that all the existing apps would need to be recompiled to fully use the 64bit.
>>>>>>>>
USE="x86-64" emerge -D world :)

The smiley isn't a part of the command...

Re:More impressed with AMD.

[Anonymous Coward]

64 bits won't make much faster. Sure, some apps which can use 64 bits effectively will be improved. Apps which can't will actually slow down. As integers and pointers get larger, they take more space in the cache, making the cache less effective, and they take more memory bandwidth. Blind adherence to the 64-bit mantra is pointless.

AMD is showing big improvements on some apps, but much of that improvement is actually coming from the inclusion of a memory controller on the processor die, making memory significantly faster. These improvements have NOTHING to do with 64-bitness.

Re:More impressed with AMD.

[wren337]

Reminds me of when the 386 came out and everyone said protected mode would be useless "until all of the apps were recompiled".

Re:More impressed with AMD.

[Anonymous Coward]

Rubbish. Back in the good old Apple ][ days you were (or maybe your dad was?) probably saying "do you personally need megabytes of RAM?"

The answer is, and ever will be, "hell yes." And when everyone is on 64 bit machines we can all rest for 32 years until the next guy comes along to ask "do you personally need zetabytes of RAM?"

Re:More impressed with AMD.

[Anonymous Coward]

Your .sig makes as much sense as bundling toothpicks with dental floss.

Re:More impressed with AMD.

[dustinmarc]

What I'm really not impressed with is Intel saying desktop users don't need sixty-four bit. Well, we don't need gobs of cache. We need sixty-four bits.

You may be surprised. If Intel keeps pushing how much L1 or L2 cache it can increase performance dramatically. Probably the best measure of processor performance is instructions per second in a normal computing environment. Probably the biggest cause of lowering the number of ips is due to a cache miss. Since cache is several times faster than main memory it's easy to figure the bigger the cache, the less misses, which means less time waiting for RAM and thus a higher number of ips.

For video, especially compression, this is a huge benefit. For instance an MPEG encoder using a typical full-search algorithm needs information from 2 different frames to calculate motion vectors. With a high quality video, or a video with large frames, this much data can't be stored in cache alone. The number of cache misses is pretty big. That's why it takes so long to compress video. A large cache will speed up this process greatly.

I'm glossing over a lot of details but take my word for it, a larger cache is very beneficial and not only for video compression. It is also useful for gamers and high-end users which is who Intel is marketing it for. Even the casual user may notice a difference in context-switching, etc.

Re:More impressed with AMD.

[Weirsbaski]

The difference is that all the existing apps would need to be recompiled to fully use the 64bit. Even lowly DOS can use performance improvements with a larger cache.

When's the last time you thought to yourself "you know, DOS is pretty good, but if it ran just a little bit faster on my multi-gigahertz processor ..." ?

Re:More impressed with AMD.

[Anonymous Coward]

When the 286 came out the protected mode WAS useless "until all of the apps were recompiled".

That happened some 10 years later or so.

Re:More impressed with AMD.

[Anonymous Coward]

With DOS, most of your low memory already fits in cache. You DON'T nead more than 1MB.

Honest. :)

Ack!! The puns!

[WTFmonkey]

Oh, gawd, that's dreadful. I load the page the page, and the first thing I see is an ad for RAM-- something about a "DIMM sum." Then, the next line is something about the Intel Developer's Forum: "Cache for Questions."

Sorry if I've been bitching at advertisers a lot lately, but leave the punning to those who are good at it, please, namely Shakespeare, Groucho Marx, and John Cleese.

Re:Ack!! The puns!

[wirde]

Sorry if I've been bitching at advertisers a lot lately, but leave the punning to those who are good at it, please, namely Shakespeare, Groucho Marx, and John Cleese.

Huh? What about The Sun?

Re:Ack!! The puns!

[Anonymous Coward]

There were only three who could actually say puns. And they were all bad at it.

Re:Ack!! The puns!

[SomeGuyFromCA]

"We wrote 10 wordplay slogans to lure customers."
"Did any of them work?"
"Nope. No pun in ten did."

Re:Ack!! The puns!

[Anonymous Coward]

Overrated on a +1 comment? Overrated?!

MODS ON CRACK! MODS ON CRACK! MODS ON CRACK!

Lameness filter encountered. Post aborted!
Reason: Don't use so many caps. It's like YELLING.Lameness filter encountered. Post aborted!
Reason: Don't use so many caps. It's like YELLING.

Re:Ack!! The puns!

[crizh]

He said good at it...

Extreme!

[TWX]

NEW Extreme Computing! Computers Gone Wild!!! College Spring Break!

A preview...

man shouting to sexy computer chassis:Take it off! (shakes a strand of discrete components) Take that top off!

Computer chassis lid is removed, computer shakes it's microprocessor seductively

Yours for only $349.95! Check us out at www.intel.com!!

NEG AS IN SMEG

[Anonymous Coward]

nope, sorry, nothing remotely like that, as YOU FAIL IT

probably not

[CowBovNeal]

Allowing these processors to work on the Xeon motherboards will only cannibalize sales of the Xeon.

Re:probably not

[Kenja]

That would be like allowing low cost Celleron CPUs to be used in MP systems rather then the high cost PII/PIII. Oh wait.....

Re:probably not

[Jeff DeMaagd]

The pinouts are different anyway. I _think_ the P4 is capable of two-way operation but I haven't seen any dual P4 mainboards, I think I have seen it on some chipset spec sheets.

Re:probably not

[Anonymous Coward]

I _think_ the P4 is capable of two-way operation but I haven't seen any dual P4 mainboards,

That's because: it isn't, and there aren't.

I think I have seen it on some chipset spec sheets.

You've seen workstation and server Xeon motherboards. #rd party MB makers like ASUS usually have 1 model that takes P4 Xeons.

Extreme price...

[Anonymous Coward]

$740 in 1,000 unit quantities. I think I'll pass.

Re:Extreme price...

[Anonymous Coward]

Yeah, but in 5 years you'll be posting about how you have one of those old dusty P4 extremes in your closet that runs your mail and dns servers.

Re:Extreme price...

[TheAwfulTruth]

Yeah but $750 in 2 months for a chip that today costs almost 4 grand? Might be more than a lot of people can pay, but is actually a pretty decent price! (I've paid a lot more for a lot less in the past)

Re:Extreme price...

[brucmack]

The article also states that it won't be available retail for some time... since it's going to be bundled with OEMs anyway, it'll likely be cheaper overall.

I believe that price is just meant to say that it's the same as Athlon 64.

Re:Extreme price...

[Anonymous Coward]

If I just had the $740, I'd go right away and buy a 1000 of them. Then I'd make $260 by selling then $1 each.

Cinema-like video

[Rosco P. Coltrane]

Louis Burns of Intel displayed a "high-definition video stream running on a 'mystery' desktop processor.

Gosh, one of these days I'll have to take a sneak peak at the hardware they run in that mystery little room in my local theater. The monitor is so big, the soundcard is great, and I can see it all for a buck!

I will not buy anything that has X, eXtreme,

[teamhasnoi]

X-treme, XXXtreme, X-tream, XT-ream, AXEtreme, Xtreme, or is generally Xed-up in anyway.

Please send a message to the X-tra stupid Advertising XX-cutives that X in the name is X-tremely dated and not an X-ellent idea.

The new marketing buzzword is 'Shit-Hot', as in "The new Intel Shit-Hot P4!"

Thanks.

Re:I will not buy anything that has X, eXtreme,

[The Tyro]

That's Sierra-Hotel to you, rookie.

Always use the phonetic alphabet... it makes you more 1337.

Re:I will not buy anything that has X, eXtreme,

[Anonymous Coward]

teamhasnoi: (furrows brow) "I am constipated, yet no product meets my stringent purchasing guidelines."

Later that day...

teamhasnoi: (furrows brow again) "I am horny, but alas, I cannot purchase any sex."

Re:I will not buy anything that has X, eXtreme,

[Tumbleweed]

This 'Extreme' naming just reminds me of the Stargate SG-1 episode with the 'Wormhole Xtreme' TV show. :)

I sure hope there's an 'X' on the packaging...or maybe racing stripes and a VTEC sticker.

Re:I will not buy anything that has X, eXtreme,

[svallarian]

Wouldn't that be better suited to an Athlon advertising campaign?

Steven V.

Re:I will not buy anything that has X, eXtreme,

[decipher_saint]

X-treme, XXXtreme, X-tream, XT-ream, AXEtreme, Xtreme, or is generally Xed-up in anyway.
Please send a message to the X-tra stupid Advertising XX-cutives that X in the name is X-tremely dated and not an X-ellent idea.

The new marketing buzzword is 'Shit-Hot', as in "The new Intel Shit-Hot P4!"

Thanks.

Fine you can use SGML and everyone else can use XML... ...to the extreme!

Re:I will not buy anything that has X, eXtreme,

[eatdave13]

Eat sierra you alpha-hole! I watched your mom do it in an x-ray-rated movie directly from the charlie-oscar-romeo-november-hotel-oscar-lima-echo . Hey, you're right!

Re:I will not buy anything that has X, eXtreme,

[Anonymous Coward]

What does this have to do with anything?

Re:I will not buy anything that has X, eXtreme,

[Anonymous Coward]

This 'Extreme' naming just reminds me of the Stargate SG-1 episode with the 'Wormhole Xtreme' TV show.

You sure that wasn't the goatse guy?

64bit vs 32bit

[tonywestonuk]

So, correct me if I am wrong, but isn't what the inquirer saying is that intel could be repackaging the xenon processor as an pentium.....

I thought the xenon was 64 bit, and ran 32 bits under emulation much slower than current pentiums.... So, I think they may be barking up the wrong tree..

Intel just found a way of squeezing more cache on the chip!

Tony.


Buy 3 Led light Keychains from me for a fiver . see here Thanks.

Re:64bit vs 32bit

[Anonymous Coward]

You are thinking of Itanium. Xeon is 32 bit.

Now get back to work.

Re:64bit vs 32bit

[exhilaration]

thought the xenon was 64 bit, and ran 32 bits under emulation much slower than current pentiums...

No, you're thinking of the Itanuim.

Re:64bit vs 32bit

[2nd Post!]

Nope, the only 'desktop' 64 bit processors come from IBM and AMD;
AMD Opteron
AMD Athlon64
IBM PPC970

Intel's 64 bit solutions is the Itanium! Anything with the Pentium moniker is 32 bit. The Itanium is the one which suffers 32 bit emulation lag.

So if you want 64 bit, you're stuck with, realistically, a Mac or some brand of Athlon CPU.

Re:64bit vs 32bit

[Naito]

You're thinking of the Itanium. It used to run 32bit X86 under a hardware emulator, but that was about as fast as the Pentium MMX. Intel has since switched to using a software emulator, something like Transmeta does with the Cruesoe, and it's actually faster than the hardware emulator, about the same speed as a Pentium III now.

The Xeon is a Pentium4 in different packaging and with SMP enabled. Actually, SMP is probably enabled with the Pentium4 too, but since there are no such motherboards and you can't plug them into Xeon DP mobos, nobody can test that. Xeons already come in versions that have up to 8MB of L3 cache, the new Pentium 4 is probably just a rebadged Xeon certified to run on an 800Mhz bus.

Re:64bit vs 32bit

[chill]

You're confused.

The Xeon series has always been Intel's "server" chips. Mostly a different pin out and lots more cache. They're souped up versions of the normal chips.

The Itanium is the 64-bit unit.

Re:64bit vs 32bit

[Anonymous Coward]

Sun has sold sub-$1000 64-bit Sparc machines for some time. And don't call them "workstations", because they're crap.

Re:64bit vs 32bit

[Slack3r78]

The Xenon is 32 bit. You're thinking of the Itanium.

Re:64bit vs 32bit

[Anonymous Coward]

Check your facts- Xenon is 64 bit. I think I read about a deluxe 68 bit version too, but they only use that on big servers because it is expensive.

Re:64bit vs 32bit

[Anonymous Coward]

Nuh-uh!

Re:64bit vs 32bit

[bhtooefr]

What about the Alphas that hit 800MHz until I stopped following them? All I remember was the 700MHz models were about as fast as a 233MHz Pentium II when running 32-bit apps, and they could only run Windows NT 4 (this was back when it was Windows or bust).

Re:64bit vs 32bit

[funwithBSD]

Alphas are being phased out, they maxed at just over 1GHz.
I have heard that a third party may continue to make them at 1.2GHz.
I know you can run *BSD's on them, OpenVMS and I am sure a Linux. We also saw a NT2000 port, but it was never sold.

Future OpenVMS versions will run on server class Itaniums

Re:64bit vs 32bit

[drinkypoo]

The basic difference between a pentium and a xeon has always been that the pentium has had half-speed l2 cache, and the xeon has had full-speed l2 cache. Sometimes they have come in a different form factor, especially recently. Xeons have been known to come with no more L2 cache than the Pentium with the same core (512kB) and as siblings to this comment indicate, up to eight megs. So Xeon was basically just the chip for people who needed x86 processors with more cache.

Re:64bit vs 32bit

[Brandybuck]

You're both wrong. Xenon is a noble gas. Xeon is a 32 bit processor.

Re:64bit vs 32bit

[bhtooefr]

Actually, I can give a damn good example of Alphas used well. It's Super Dimensional Fortress (telnet link). They seem to be under some heavy load - I'm only getting 3 KB/sec off of them... but if you're bored (or think P4 heatsinks are small), go here.

BTW, isn't it the dual-cpu Alpha support in Linux that SCO isn't happy about?

Re:64bit vs 32bit

[bhtooefr]

Also, I found an interesting tidbit out. Compaq bought out DEC, and HP bought out Compaq, right? Well, the "third party" that you mention making 1.2 GHz boxes actually has topped your speed by 50 MHz - and it's HP.

Here it is...

special

[potpie]

why is this special? oooh- they made a better processor! It's not as if it wasn't going to happen sometime this week- what with the fast pace of technological advancement these days. Nice reference to Maddox's "x-treme" ad campaign satire.

Paper Launch?

[Dumass]

Not to be labeled a fanboy (although not necessarily denying that status)... but this sounds like a paper launch just to take some press away from AMD.

"He [Burns] said the chip will be available to buy in the 30-60-day timeframe." from this article.

Prescott is going to be late and has been getting bad press for not being backward compatible with current motherboards. Why not make some noise with a product that wont be around for another month?

Re:Paper Launch?

[Anonymous Coward]

Where is the bad press about Prescott not being backward compatible with current motherboards? I haven't seen it. Better yet, why would there be any bad press? We always have to buy new motherboards when Intel releases newer chips with slightly different architectures. AMD is no different.

Re:Paper Launch?

[Dumass]

P4 has had 3 sockets (including the new Prescott) with the same types of changes that Athlon has gone through. Don't forget that the server chip (Xeon) has had a few sockets of its own. Athlon has had 2 (including Slot A). Since Thunderbird, there have been no changes to Athlon's socket. Compare the amount of time that each chip has spent on its current socket. Athlon has had around 2 years, P4 around 1.

Here's some press about Prescott:
http://www.theinquirer.net/?article=10438
http://www.theinquirer.net/?article=8011

And to prove that I don't only read The Inq:
http://www.anandtech.com/news/shownews.html?i=2001 4

I could probably dig up some more if you are really interested.

Re:Paper Launch?

[Jeff DeMaagd]

My impression is that the P4 Xeon has only had two sockets so far, 603 and 604. The P3 Xeon at least had slot 2, I don't know if they had a socket or not. If you bundled all of Xeon together, then yes, they've had at least three sockets.

I never really expected it to be compatible, and there's so little information on what is in Prescott, I simply assumed it to be a P5 anyway.

Re:Paper Launch?

[Dumass]

My comment was intended to point out that the entire Athlon line (server and desktop/workstation parts included) has had 2 sockets/slots. Pentium has had 5 (server/desktop combined).

From what I've read (and I could very well be wrong, so don't take this as fact), Prescott is P4 with 13 new instructions, a la MMX or SSE, in a new package on a new process. Somewhat like Athlon vs. Athlon XP (although that didn't include a die shrink until later).

I'm so sick of "extreme" this and "Xtreme" that

[ChicagoFan]

Never mind *technical* buzzwords. I think the biggest buzzword of the last decade has been the word "extreme". You'd think with the web right there at people's fingers and mice, they could make use of an online thesaurus to search for a word that's not quite so tired and overused.

When your processor press release sounds like a Mountain Dew commercial, it's hard to take that information seriously.

ChicagoFan

Re:I'm so sick of "extreme" this and "Xtreme" that

[kfg]

Well something had to replace "turbo."

I was so disappointed when I cracked the case of my 286 and didn't find an HKK in there somewhere. I had thought that maybe I could replace the one big one with two little ones to reduce net lag.

From the marketing point of view the advantage of "Xtreme" is that you can't prove it isn't in there somewhere. Maybe keeping the Magic Smoke in or something.

The disadvantage is that they can't play games by making you think the go slower button is really a go faster button. I hated having to explain that to people, they always got mad at me.

Killing the messenger always works, if you're a nitwit.

KFG

Re:I'm so sick of "extreme" this and "Xtreme" that

[mr.henry]

IANAAM (I am not an advertising major..), but apparently the "X" is supposed to make us think of sex, and therefore make whatever product a company is pitching more appealing. "SX" is even more blatant. In product model lineups, it's everywhere.

With that in mind, and seeing past the fnords, LX or LS (think Lexus LS 400, or whatever the latest is), is the most appealing of all: lesbian sex.

I hope I don't come across as crazy or perverted, but advertising will do ANYTHING to sell crap to people.

Re:I'm so sick of "extreme" this and "Xtreme" that

[Anonymous Coward]

Whatever you say, mr.horny

Re:I'm so sick of "extreme" this and "Xtreme" that

[questamor]

from thesaurus.com:

Entry:
extreme

Synonyms:
acute, consummate, great, greatest, high, highest, intense, maximal, maximum, severe, sovereign, supreme, top, ultimate, utmost, uttermost

Antonyms:
limited, mild, moderate

"P4 Acute" - sounds like a Honda
"P4 Consummate" - sounds like something you'd cook
"P4 Great" - sounds retarded
"P4 High" - sounds like a urinal competition
"P4 Intense" - sounds like a pun
"P4 Maximal" - sounds like a condom
"P4 Sovereign" - sounds like an archaeological discovery
"P4 Severe" - sounds like a poisons classification
"P4 Supreme" - sounds like a pizza
"P4 top" - sounds like a urinal competition to avoid
"P4 ultimate" - sounds like more condoms
"P4 utmost" - sounds dumb

I'll settle for P4 Extreme

Re:I'm so sick of "extreme" this and "Xtreme" that

[Hoser McMoose]

Forget all of those, I'm still waiting for my VTEC P4 Type-R!

Re:I'm so sick of "extreme" this and "Xtreme" that

[Anonymous Coward]

... apparently the "X" is supposed to make us think of sex, and therefore make whatever product a company is pitching more appealing. "SX" is even more blatant. In product model lineups, it's everywhere.

Intel previous tried that with "SX" models, as in 486SX. Most people would now see that as 486SUX and avoid it like the plague.

Maybe if they reversed their introduction it would work. That is, come out first with the SX models, and follow up with the regular model. I can hear the advertising now....

Now, for the first time ever! An Extreme Pentium that SUXs less.

It might sell.

Re:I'm so sick of "extreme" this and "Xtreme" that

[Valar]

No wonder I'm always using that register...

Re:I'm so sick of "extreme" this and "Xtreme" that

[Anonymous Coward]

Dammit, all it made me think of was windows XP and how crappy that is.
Oh wait, its ok to not think of sex, 'cause I'm married...

Re:I'm so sick of "extreme" this and "Xtreme" that

[joebolte]

I agree that putting X and extreme in everything gets really boring, but I think /.ers may be being a bit hypocritical if the criticize it. After all - UniX, LinuX, wX, xcdroast, just calling the window system "X," etc. is playing in to the same thing.

Also, this putting X's into product names is nothing new. Tom Wolfe commented on the same thing in relation to the California Custom Car scene back in the sixties.

Re:I'm so sick of "extreme" this and "Xtreme" that

[estes_grover]

...apparently the "X" is supposed to make us think of sex..."SX" is even more blatant.

Cooool - i guess that old SX40 i got propping open the door was waaay ahead of its time.

extra 2 mb's?

[mOoZik]

So it's not just a 2 meg cache but is in ADDITION to an existing amount? 256? 512? I'm confused.

Re:extra 2 mb's?

[Anonymous Coward]

Yes but the L3 cache in inclusive, so you get only 2MB of (L1+L2+L3) cache.

Re:extra 2 mb's?

[FrozenDownload]

sigh...i read the same thing, and I was very confused.

Pentium(R) 4 processor Extreme Edition 3.20 GHz , with an extra 2 Megabytes. In Intel's own words, "this new processor will be targeted at high-end gamers and computing power users."

then later on ->

As a matter of fact, 2MB cache will help a lot those users whose apps (including games and such) have a lot of big cache-friendly pieces...

I dont quite see where the extra 2 megs comes in. It isn't quite extra if it is inclusive to the other space? lol
or even IF they plan it to be "extra" what will the base be?

Re:extra 2 mb's?

[Bored Huge Krill]

yes, it is in addition. There's a report at Anandtech that writes up some of the details

What is new is a 2MB L3 cache. It's made clear that this is in addition to the existing 512kB L2 cache on current P4s, making a total cache size of 2.5MB

What does this say about Prescott?

[freidog]

Intel pushing out an (unitl recently) unkown "Extreme performance" CPU targeted for the same window (October-November) as Prescott. Considering this seems like nothing more than a Xeon MP with a bit higher clock speed, and certified for the 800mhz bus, i find it odd they would release it 30-60 days from now if they just wanted to steal the thunder from AMD's Athlon64 launch latter this month. I would have expected it out now to accomplish that. just wondering if they're buying time to settle some issues with .09 or the power dissipation of Prescott....

database searches

[chipace]

This chip would be great for database searches... it has more cache than uni-processor xeons and it probably will be cheaper. Thanks gamers! I guess the wait for Prescott is real... seeing that Intel had this chip on tap.

CNET article with more details

[HeroicAutobot]

CNET has an article with more details (or speculation more likely).

Some interesting quotes:

"The performance boost is awesome," Burns said Tuesday during a speech at the Intel Developer Forum here.

"It is a Xeon with a different pin-out, or least that's what it looks like to me," said Nathan Brookwood, an analyst at Insight 64.

Intel did not disclose the price of the Pentium 4 Extreme Edition. It likely will be as expensive as its counterpart, the 2.8GHz Xeon with 2MB cache. That chip sells for $3,692 in quantities of 1,000.

"It absolutely will be kind of pricey," Brookwood said.

Re:CNET article with more details

[WTFmonkey]

"It is a Xeon with a different pin-out, or least that's what it looks like to me," said Nathan Brookwood, an analyst at Insight 64.

Well, I guess we can see why he works at a company called "Insight" now, can't we?

Wow!

[ENOENT]

3.2 GHz! That's 6.7% faster than 3.0 GHz! You feel the need to send money to Intel! Fnord! Imagine how fast the Internet will be if you have one of these on your desktop! You will need a neon-colored bunny suit just to look at your computer! You will be assimilated by the Blue Man Group!

Re:Wow!

[Aadain2001]

I remember the days when an increase in only 50 MHz as a big deal! Now they role at a 200 MHz increase and people say "it's only a 6.7% increase, big deal"?!?. What, do you expect them to role out a 1 GHz increase with each new chip they put out? Time for a little visit back to reality.

Re:Wow!

[MarcoAtWork]

maybe you should get off the hype machine...

Yeah, at one point a 4MHz increase was a big deal, but over an 8MHz clock you can see that the jump was quite significant. I remember when my friend had just bought a 486/33 and others a 486/40 and the DX2/66 came out: nearly a 100% clock speed increase.

More recently from a p3-550 in a very short time we went to p3-733 and p3-1GHz (quite another percentage jump).

Excuse me if the past year or two of incremental 100-150MHz clock increases doesn't excite me that much: I honestly would have thought that by this point we'd already be past 4GHz, if not closer to 5 (given that IIRC Intel demonstrated a p4 engineering sample last year that was clocked at 4GHz, didn't they?)

Re:Wow!

[saddino]

I think you're missing the point. Back in the days when a 50MHz increase was a big deal, the percentage increase was much greater, hence the perceived worth of upgrading was much better -- who wouldn't want a machine that was 50% or 100% faster? Now that processors speeds are so high (comparatively), a 200MHz increase just doesn't seem worth an upgrade.

Re:Wow!

[drinkypoo]

People are still paying for incremental upgrades. If you can convince them to stop, then intel will start producing faster processors.

Re:Wow!

[Brandybuck]

Back when I had a 400MHz Pentium, a 200MHz made a big deal. It was worth upgrading if the price was right. But now I have a 2.8GHz Pentium. Adding 200MHz to that is nothing. I wouldn't be able to notice the speed difference without a stopwatch.

I think some of the speed fan need to get a life. I bought my 2.8GHz P4 one month ago for $298. A 3.2GHz (non-extreme) P4 from the same retailer was $698. The decision was trivial.

Re:Wow!

[dustinmarc]

Sure, Intel is releasing smaller percentage increments, they make more money from it. The nice thing is that they still follow Moore's law. So ultimately chip speeds still double about every two years.

Re:Wow!

[Aadain2001]

There's some truth to that. Intel engineers would love to be told "Ok, the 3GHz is out, get working on the 4GHz" and not have to worry about getting the 3.1, 3.2, 3.3, ..., 3.9GHz versions working first. But as long as there are people willing to buy those speeds, or a competitor is going to put out a 3.1GHz and say they are now the best because they have a faster chip, Intel will keep making the incremental upgrades.

Re:Wow!

[jrockway]

Not only does it not seem worth upgrading, it doesn't affect performance much. I have an Athlon 2500+ (for $85) that overclocks (with no change in Vcore) to a 3000+. The CPU now thinks it's a 3000+ (well Linux thinks so and so does the BIOS). Final results? Kernel compile is down 15 seconds. Now if you were going to buy a 3000+, it would cost $200 more than the 2500+. For 15 seconds off the kernel compile. That's what we like to call a ripoff. OTOH, just overclock the 2500+ (multiplier goes from 11x to 13x) and get the extra "speed" for free.

Re:Wow!

[silverHat]

Yeah....It's funny how the current 3.2ghz P4 costs $605[pricewatch.com], and no doubtedly the eXXXX...treme version will cost only that much more for a 6.7% increase with more cache (should i be happy?) Well, I got your 6.7% right here. For about 300 dollars, you can buy a 2.4C P4 and a decent mobo, and OC the pentium to a stable 3.4ghz+ on AIR [overclockers.com] The 512k L2 cache I believe is more than sufficient for todays desktop applications Just make sure the RAM is fast enough.

Re:Wow!

[Lifewolf]

You will be assimilated by the Blue Man Group!

Yay! Twinkies!

MODS: quit modding this crap up

[zapp]

a beowolf joke plus a no carrier joke... come on mods, this is not worth spending your points!

Re:MODS: quit modding this crap up

[CracktownHts]

a beowolf joke plus a no carrier joke... come on mods, this is not worth spending your points!

I, for one, welcome our new lame joke recycling overlords.

Re:MODS: quit modding this crap up

[Anonymous Coward]

In soviet russia they skip right to step 3 ... profit.

Re:MODS: quit modding this crap up

[akedia]

I, for one, welcome our new lame joke recycling overlords.

In Soviet Russia, lame joke recycles you!

I wonder if you can use it with BTX motherboard

[Ratface]

Would that give you a PC with BuTtoX Extreme inside?? :-D

High Def video is not that CPU intensive!

[mr.henry]

Wow, my AMD Duron 1000 can show a 'high definition video stream.' I bet it looked gorgeous and impressed the hell out of everybody (as good HD usually does), but this isn't really a difficult thing to do.

Show something useful, like, say, transcoding a dual layer DVD so it will fit on a single DVD-R.

Re:High Def video is not that CPU intensive!

[drinkypoo]

That already only takes 26 minutes on my athlon xp 2000+ with ddr333. That's twice the machine this is, in most ways, so I assume it would just whip my ass, but almost no one needs to do that any faster than in just shy of half an hour.

Re:High Def video is not that CPU intensive!

[Anonymous Coward]

Dude, you can never pirate too fast.

Re:High Def video is not that CPU intensive!

[Anonymous Coward]

*cough* Bullshit.. just try decoding any of these HD MPEG4 videos with a Duron 1GHz. Can you say less than 3fps?

breaking news

[colinleroy]

as usual from Intel... Just add some megahertz and some cache, it'll be good enough. On paper. What about improving technologies, like IBM or AMD do ?

Re:breaking news

[dustinmarc]

It's funny you should say this considering that Intel invented the technology....

Re:breaking news

[Anonymous Coward]

What about improving technologies, like IBM or AMD do ?

Hmm. Intel never improve or create technologies. Riiiiight..
PCI
USB / USB2.0
AGP
SSE / SSE2
Hyper-Threading Technology
AC'97
PCI-Express
ATX/uATX form-factors (including soft-on PSU)
Serial ATA

All of the above are technologies that Intel either developed alone, or was instrumental in steering the planning committee for. And that's just in the last few years.
Consider that, next time you install your oh-so-innovative AMD processor in your ATX case with your PCI soundcard and AGP graphics card and USB modem - it means that most of the components of your PC owe their existence to the same Intel that you claim never innovates.
You were trying to make a point, fanboy?

Re:breaking news

[colinleroy]

Yes, they did so 20 years ago. I can't understand why today's Intel processors are so backward-compatible. They would have better results designing something from scratch (or almost). x86 is a very inefficient architecture (compare its performance to Alpha, G4, G5 and Sparc ; compare its power consumption, too).

ATI Inside?

[Anonymous Coward]

I wonder if it was running this ATI card. That puppy is really going to smoke half-life 2.

GI Schmoe

[Agent R]

I hope this Pentium 4 Extreme isn't anything like GI Joe Extreme. What a bad piece of animation that was.

Just give me a stable processor that won't explode if it overheats.

Processor-Intensive SW: Engineering Applications

[reporter]

This "extreme" version of the chip has to be aimed at a very niche market, at least for the next couple of years until more processor intensive software catches up.

The processor-intensive software is already here. It is called HSpice, Verilog, fluid-dynamics simulation, etc. The Pentium 4 has done nicely in the engineering workstation market, and the "Extreme Edition" should do even better.

Please check the SPEC web site for a performance evaluation of the Pentium 4's floating-point (FP) performance. In particular, it outperforms the UltraSPARC III even though the latter has a 2-to-1 advantage in the width of its databus -- 64 bits versus 32 bits.

What changed the x86 chips from also-ran losers in FP performance to the kings of the hill? SSE.

The SSE extension to the x86 instruction set architecture (ISA) opened up a whole new world of applications for the Pentium III and successors. Older Pentiums were saddled with a FP stack that hurt their performance. The SSE extension established a directly addressable bank of 8 128-bit registers or 32 32-bit registers for FP operations. As a result, the Pentium 4 outperforms the UltraSPARC III on video applications.

At 3.2 GHz, the "Extreme Edition" of the Pentium 4 should help the Pentium 4 to capture even more of the engineering workstation market. Nowadays, the first-choice workstation among engineers in Silicon Valley and Boston's Route 128 is Linux running on a fast Pentium/Athlon, not Solaris lumbering on a slow UltraSPARC III.

... from the desk of the reporter

You forgot the xception...

[Anonymous Coward]

X-rated.

1GHz is plenty!

[Daniel Wood]

Yeah, you probably were one of those people too.

This CPU is aimed at the gaming/multimedia community. All that extra cache should make Doom3/HL2 speed along a little better. It should also help us that encode DVDs/DivX on the fly. What supprises me is that they didn't finally go to 1GHz FSB. Yeah, I know, that would mean you need DDR500(PC4000). While I'm sure you make have problems taxing your 3.2GHz CPU with MS Word or Counter-Strike, I am left longing for more CPU power with my Dual Athlon MP2100+ when De-interlacing video from my DV cam or running Urban Terror with SMP @ 1152x864x32+2xAA+AF.

You say software needs to catch up? I say hardware has and will be playing catchup for a long time. I'm sure that hardware will never exceed the demands of software. Multimedia has been the driving force behind computers since they became "good enough for Twin/QuatroPro/WP/etc." and will continue to be that driving force.

Re:1GHz is plenty!

[Anonymous Coward]

What supprises me is that they didn't finally go to 1GHz FSB

This isn't a new chipset, its just a different packaging of an existing CPU. No current chipsets support a > 800 MHz FSB, and DDR500 RAM doesnt exist yet. Getting a chipset to support a 1GHz FSB would require a heafty redesign and validation. Not a very smart thing to do for a niche product that won't generate very much revenue.

Re:1GHz is plenty!

[Pharmboy]

What supprises me is that they didn't finally go to 1GHz FSB. , I know, that would mean you need DDR500(PC4000).

Actually, no you don't. Apple sells their dual 2ghz box, that has a 1ghz fsb (dual pipe), and 400mhz ram. goto apple.com/powermac for info. It obviously doesn't talk to the ram that fast, but 1g pipe to the chipset doesn't suck either. Oh yea, and up to 8gb of ram so far. its a bit different in other aspects as well.

I am just waiting to score one of the dual 2.0 boxes used (cant afford $3500) but that will take a while. They also bench out better cycle to cycle that intel (similar to amd or better) Its actually the IBM 970 cpus (reduced power 4 cpu) that IBM is said to be releasing soon in entry level servers, with 4 cpus, for $3500, for Linux.

Re:1GHz is plenty!

[Hoser McMoose]

I actually kind of wonder about this chip. I wouldn't really expect the extra cache to help DVD->Div-X encoding all that much, and even gaming isn't likely to see that huge of an improvement. Large cache like this tends to really benefit server applications, while desktop applications benefit most from low-latency and high bandwidth memory connections, as well as raw processing power.

It will be interesting to see if this chip really does significantly improve performance for desktop systems or not. I'm not holding my breath for any earth-shattering performance improvements on the desktop end of things, though it would probably make a hell of a chip for a single processor server.

Re:1GHz is plenty!

[Anonymous Coward]

DDR500 RAM doesn't exist yet?! Both Corsair and Kingston produce it, and have done for over a month.

MOD PARENT UP!!

[Anonymous Coward]

WTG! 2megabytes of pron LOL *wink* !!
MOD PARENT UP!! Excellent troll LOL!!
LOOKS Like the mods need glasses

XBOX

[Anonymous Coward]

just spend $200 and get an xbox with ghost recon

*sigh*

[tekiegreg]

Here I am on my old Pentium III 800, Visual Studio.net (gasp!) compiles in no time flat for the projects that I work on, and I play nothing more than Starcraft and Civ III. I do wonder how Intel manages to stay in business (oh wait the chip industry is dying, never mind...).

Perhaps Intel should focus more on....(fill in the blank please, Intel seems to only be really focusing on chips...)

Re:*sigh*

[smart.id]

I'll tell you how Intel stays in business. I know of people who buy a new, state of the art computer every year. People think that you need to in order to "keep up" with the latest technology. I can understand why you're using a 800mhz computer if it works fine for you, I have a 1ghz T-bird that has lasted me 3 years, and I don't need anything better. However there are people who feel they do, and also need to buy every single type of PC possible (tablet PC, pocket PC, laptop PC, desktop PC etc.)

Re:*sigh*

[Hoser McMoose]

Err, Intel has got to be one of the most consistantly profitable companies of all time! There is hardly ever a QUARTER (that's three months) were they don't post $2 billion in profit!

Ohh, and almost all that profit comes from their x86 chip sales. All other divisions of the company tend to mostly just break even. Chip industry dying indeed! I think ANY company would like to die just such a death, sounds to me like they're already in corporate heaven.

ATTENTION!

[Anonymous Coward]

Parent post is the polar opposite of funny. Suggest poster should never post again.

Is this a ploy for extra money?

[greymond]

I guess you could say I'm a big Intel fan-boy (only because i've had bad experiences with AMD and Cyrx that has left me jaded) but i'm not about to get all sprung about this. Right now i'm using a p4 1.6 with 1g mem and 5200fx, it runs every game ive played fine - even in multiplayer (games = BF1942, WC3, NwN+Exp, etc...)

The way my system is setup I can support HT so eventually i'll max my system out using a 3.06 with HT. And that will last me another couple years at least.

But right now I feel robbed. Robbed because there are no games or apps (I use Photoshop, Illustrator, and Quark mostly) that really push the envelope. Sure Photoshop can always be improved but thats the only exception I have - none of the games are laggy. I used to somewhat enjoy wasting money on buying more memory and a new cpu every 6 months because "something" new was coming out that would utilize some new technology or something.

It's my opinion that there is very little innovation going on in the land of computers. Sure we're making things faster (somewhat) but it still costs a SHITLOAD and doesn't give us a shitload in return....I want to see Intel create a NEW chip that actually R0x0rz my b0x3n...

OH and before somene shoots off with an Apple remark - I like Apple when working/talking to my printers and doing 1 thing at a time - I hate multitasking on my mac - even tho OS X is 100x's better than 9 handled things, it just seems to lag more than my pc does when going back and forth between apps and opening/saving large files. (i'm sure you can disagree but it's a preference thing - much like my views on AMD/VIA/Cyrix, etc...)

Re:Is this a ploy for extra money?

[Anonymous Coward]

Put some more RAM in that OS X machine you stingy bastard. OS X's multitasking is much better than Window's if you give it enough RAM.

Re:Is this a ploy for extra money?

[Anonymous Coward]

Yeah I guess thats why the G5's can go up to 8gigs of memory - you need 1gig for each application you have open.

Wait for the next version after this

[PanchoVilla]

In a couple of months Intel will be releasing the new Pentium IV TypeR !!!! The heatsink will even have one of those ugly shopping cart handle type spoilers and NEON too...........

EXTREME? How about stable?

[narftrek]

What is it with Extreme as the buzzword these days? When you hear extreme you think of people jumping off cliffs or launching motorcycles off tall things. Things that some may consider DANGEROUS or STUPID. It can also mean "on the edge" as in pushing the limits or ground breaking technology. I don't know about the rest of you but I don't want a computer that pushes the edge, is dangerous, or stupid. I want a nice stable (as in doesn't crash 10 times a day) computer that I can watch my pr0n on. Is that too much to ask? Extreme is worn out in my book-pick a new buzzword.

Athlon Unleashed

[Anonymous Coward]

I'm still waiting for the new "Athlon Unleashed".

Re:Athlon Unleashed

[Anonymous Coward]

Wasn't there a video of that on Tom's Hardware where the Athlon was "Unleashed" when the heatsink was removed while running?

I won't buy anything that has XP in it

[jbischof]

Windows XP Athlon XPripoff

Re:I won't buy anything that has XP in it

[Anonymous Coward]

To me, "XP" seems about like "Win" used to be. Which is why I avoid any hardware made to work with Windows XP, as it's probably some software-driven crap that only works with terrible buggy drivers.

Part 2 of Article up now

[dubiousdave]

The second part of their article is here.

Who said the 2MB cache was L2?

[tugrul]

According to this ExtremeTech article about this cpu, its L3.

The gaming-optimized Pentium 4 contains 2 Mbytes of level-3 cache, and will work with existing "Springdale" and "Canterwood" chipsets, Burns said.

2Mb extra cache - nice server CPU maybe

[Moderation abuser]

We have some highly CPU intensive applications which could make use of it.

Awwww

[sys$manager]

I was hoping for Pentuim 4 Turbo Alpha.

HAAAAADOUKEN

Oh no, not again

[jazman]

Here we go again. Fastest chip in the world; nobody, not even weather computer people could ever want faster, blah blah, bollocks bollocks; everyone knows in 12 months granny will want a PC with one in it and we'll all "need" something with a gazyllion terabytes of RAM and that runs at a googolplex hertz just to do some silly emails and stuff.

Was looking through a 1984 copy of Personal Computer World and it was saying exactly the same about the new 2MHz 8086 or whatever. Would've thought those crazy marketers'd have learnt by now that in IT there's a new "fastest in the world" every few months.

Still, I suppose some people will be new around here and be impressed by this sort of crap. I know I was, first time (probably in 1981 or thereabouts) I saw a front page "Fastest in the World!" story; second time round I thought hang on, haven't we been here before?

And yes, I know that's before some of you here were born, before any of you point it out and depress me even further.

Re:Oh no, not again

[Brandybuck]

I remember buying a 8086 with a turbo button. I could run at 11MHz instead of 4.77! Whoo hoo!

Re:Oh no, not again

[Anonymous Coward]

That's the year I was born!

Re:Oh no, not again

[Jugalator]

Hmm.. 130% improvement then...

A turbo button would be nice today. 2.8 GHz without, but... 6.5 GHz with turbo! :-D

Re:Oh no, not again

[CavyDriver]

nobody, not even weather computer people could ever want faster

Uhh, nope. In the high end world of computer simulations there is still a need for faster processors. Why?

1) There are jobs that just don't parallelize well or even at all. In this case you still need a single fast CPU.

2) There are still things that cannot be simulated because there are not fast enough computers available. If you would like a good simple real life example, a flag or piece of string flapping in the wind is nearly impossible to simulate.

Am I missing something?

[WndrBr3d]

Perhaps I'm confused here, but I remember TomsHardware doing an article on the new Barton processors with double the cache (512k) didnt produce really noticable performance increases in most 'high end user' applications (gaming/video encoding.

Could Intel be planning a compiler that would utilize this cache??

Re:Am I missing something?

[Indy1]

the P4's archtecture (sp) is such that it is incredibly sensitive to cache misses due to its long
pipeline (20 i believe). Thats why higher memory bandwidth and larger caches make such a huge difference on the P4. Where as the Athlons have a much shorter pipeline (12 i believe), the extra memory and cache dont help out as much.

Re:Processor-Intensive SW: Engineering Application

[Hoser McMoose]

In particular, it outperforms the UltraSPARC III even though the latter has a 2-to-1 advantage in the width of its databus -- 64 bits versus 32 bits.

Err.. The P4 has a 64-bit data bus. The UltraSparcIII has quite a different databus (due to it's integrated memory controller), but when you look at memory bandwidth, the USIII has 2.4GB/s of memory bandwidth while the P4 has 6.4GB/s.

What changed the x86 chips from also-ran losers in FP performance to the kings of the hill? SSE.

Less than 5% of SpecFP scores make use of SSE. The performance comes mainly from the P4 having a lot of memory bandwidth. The only chips with more memory bandwidth are the Alpha 21364, the Power4, the Itanium2 and the Opteron. Ohh, take a guess as to which chips get higher SpecFP scores than the Pentium4 does.

Where's the dealcloser?

[contrasutra]

Well, im EXTREMEly disappointed.

If i'm an uber-kewl l33t gam3r k1d, and i'm buying this just to show my equally pathetic friends that im the 0wn3r, how am I suppose to do this without a TEESHIRT!! I NEED TO SHOW OFF!

Intel may never get their market.

What a waste of work time The AMD idea is better

[Anonymous Coward]

The FX chip in the next range will have a double ram access bus. Double ram reading speed to compete with a 2 meg cache. Yep linux running 3000 processes it not even a byte per process. Board with 2 G of ram it is almost nothing. Now AMD double access bus it does not matter how much ram or how may processes are running it will still double the speed of the ram access. Also the controler to the ram bus is in the processor chip in the AMD. Basicly the Ram is a cache just a slow one in some cases.

Basicly where is the new chip Intel.

Mod up

[Anonymous Coward]

Some ass gobbler modded this down but it's the most insightful post I've seen in this discussion. I saw this story while browsing CPUs online for purchase as I have exactly this issue. I like trying out video filters and such, but it isn't really something to do frequently when it takes an hour to render out.

Re:Pentium is dying!

[Anonymous Coward]

Gaming involves millions of complex instructions, which have to be split in to tiny bits to spread over the small bitwidth and reigster, using up clock cycles, 64 bit will allow you to more stuff without register switching all the time

#1- Instructions are not "split up" and put into registers, they are fetched from the instruction cache and decoded.

#2- You do realize that none of that has to do with 64 bit computing, don't you? Sure, AMD added some extra GP registers in its Athlon64, but that has nothing to do with 64 bits. The ONLY thing that 64 bit computing adds is higher memory addressing. It does not change your performance or speed or IPC of the processor in any way.

Re:Processor-Intensive SW: Engineering Application

[mczak]

SSE is single-precision (32bit) floats only, so pretty useless for scientific calculations (usually require doubles).
However, I believe the intel compiler uses SSE2 (which can handle 64bit floats) exclusively for float code, since the P4 legacy fpu is just slow. Of course there are compiler switches for the compiler so the code also runs on good old Athlon, Athlon XP, PIII (which lack SSE2, the Athlon also lacks SSE) - and those aren't exactly slow doing float calculations neither.

Veritable?

[G-funk]

-- I do not think it means what you think it means.

Re:Veritable?

[LoveShack]

OT, I know, but damn, The Princess Bride rocks the veritable casbah. Nice quote.

Re:Processor-Intensive SW: Engineering Application

[Anonymous Coward]

The P4 has a 64-bit data bus.

Moreover, Intel chips have had 64-bit data busses since the original Pentium.

Databus of Pentium 4 is 32 bits, not 64 bits.

[Anonymous Coward]

The parent article is referring to the internal data bus. The internal data bus of the Pentium 4 is 32 bits because it is a 32-bit machine. The internal data bus of the UltraSPARC III is 64 bits.

The Pentium 4 really obliterates the UltraSPARC III in performance even though the internal data chunks of the latter are 64 bits versus 32 bits for the former. How does Intel do this "magic" in performance? Does anyone have an explanation?

Re:Databus of Pentium 4 is 32 bits, not 64 bits.

[Arker]

He already explained it, it's not magic, it's mostly memory bandwidth. The size of that internal bus doesn't mean squat when it's sitting waiting for data from main memory.

Re:Databus of Pentium 4 is 32 bits, not 64 bits.

[Hoser McMoose]

Err, no. The internal data bus of the P4 is 256-bits wide, at least if you're talking about it's L2 cache bus. L1 cache doesn't really have a "bus", especially not the P4's trace cache (it's replacement for an L1 i-cache), but if my memory serves me correctly, the L1 d-cache of the P4 can read or write a pair of 64-byte values in 2 clock cycles. I guess that makes it's "bus" 128 bytes (not bits) wide. I don't know the bus width of this new L3 cache on this P4 "Extreme", aka a XeonMP, but I would guess it's 64-bits wide.

I haven't got a clue as to the internal data bus of the USIII, but I would guess that it's either 128-bit or 256-bit wide. Side note: the Power4 uses a MASSIVE 1024-bit wide internal bus, one of the reasons for it's impressive performance.

The only situation where the USIII has 64-bits and the P4 has 32-bits is if you are talking about integer registers or memory pointer width, neither of which are going to play a role in Spec CFP scores.

Re:Databus of Pentium 4 is 32 bits, not 64 bits.

[Hoser McMoose]

I should probably point out that while throwing all of these bit-widths around, that's only half of the picture. The other half is, of course, the number of data transfers per second (ie the clock speed x the number of transfers per clock cycle).

A good example of this is to compare the memory bandwidth of the USIII and the P4. The USIII uses a 256-bit bus to memory, while the P4 uses only a 64-bit wide data bus (shared for memory and I/O). However, the USIII's bus runs at only 75MT/s (75MHz SDR). The P4's bus runs at 800MT/s (200MHz QDR). End result is that even though the P4's data bus is only 64-bits wide to the chipset, it offers considerably more bandwidth than the UltraSparc bus.

Similarly AMD is able to get lots of bandwidth out of their Hypertransport links, even though they max out at only 16-bits in each direction. A 32-bit HT link can manage 6.4GB/s of aggregate bandwidth (3.2GB/s in each direction) due to it's 1600MT/s (800MHz DDR) speed.

For data buses, narrow but fast point-to-point connections are taking over from wide bus slow buses. Even hard drives are moving in this direction with Serial ATA and Serial Attached SCSI.

Re:Databus of Pentium 4 is 32 bits, not 64 bits.

[turgid]

A good example of this is to compare the memory bandwidth of the USIII and the P4. The USIII uses a 256-bit bus to memory, while the P4 uses only a 64-bit wide data bus (shared for memory and I/O). However, the USIII's bus runs at only 75MT/s (75MHz SDR). The P4's bus runs at 800MT/s (200MHz QDR). End result is that even though the P4's data bus is only 64-bits wide to the chipset, it offers considerably more bandwidth than the UltraSparc bus.

What about latency? This is important in multitasking and multithreading, so I've heard.

Anything to do with Doom III, Half-life 2, etc?

[Anonymous Coward]

2mb cache? Half-life 2, Doom III, and a couple other titles are coming out soon.

It looks like manufacturers are clinging to the hope that these titles will push a few for a hardcore upgrade.

Games seem to be the only thing left to push the high-end hardware sector along.

Faster CPU lengthen useful life of machine

[AHumbleOpinion]

This "extreme" version of the chip has to be aimed at a very niche market, at least for the next couple of years until more processor intensive software catches up.

While I agree with one of the other posters that many high end CPUs are sold to the "mine's bigger" crowd, Intel naming surely supports this idea, there are some legitimate advantages to getting a faster CPU even when you don't have a need for the additional computational power. I'm getting along well with a P3 1.2G but towards the end of the year I will be building myself a new machine. A P4 2.26G 533FSB would be fine but I'll put together a 3.0G 800FSB dual-channel DDR because it will only have a relatively modest price increase (then, not now) but it will add a year or so to the useful life of the machine. For years I've had the same strategy. A high quality motherboard for US$150 or less, the fastest CPU for US$250 or less, the largest HD for US$150 or less, ... My systems are not the fastest out there but they are close to the price/performance sweet spot and have good longevity since they are far more than what I need at the time.

Re:Faster CPU lengthen useful life of machine

[Glonoinha]

The sweet spot I have found :

Wait until you just can't bear to run whatever new program has come out that you need (want) to run. For a bunch of us that is going to be TFC2/HL2/DoomIII/DNF - whichever comes out first.

Suffer for a month longer on your system, try running it with all the settings turned down.

Whatever is the fastest machine out there, come down two steps (in this case the 3.2GHz-Xtreme is fastest, the 3.2GHz is next, dropping one more to the 3.0GHz with HT) and that is pretty much the sweetest spot on the curve. They charge a mega premium for the best. They charge a medium premium for the second best, but the third one is where that flattens out. Figure out how much RAM you think you need and buy twice as much, ditto drive space. If the SATA drives have gotten reasonable in pricing I recommend two of those in a RAID 0 array if performance is your thing, else just go with a bunch of IDE space. As for video card, however much you have left in your budget.

Re:Faster CPU lengthen useful life of machine

[leifm]

I prefer to just wait 4-5 years on software, that way it runs fan-fucking-tastic on whatever hardware I own at the time. Quake II/III run great on my machine, settings turned up and everything.

It's a WORKstation because you find it at WORK

[yerricde]

A "workstation" is a desktop computer that is primarily found at work. Computers typically don't end up in homes until commercial games are available for their platform. Therefore, a "workstation" is a desktop computer whose platform has few available commercial games.

The Solaris operating environment on the SPARC architecture has even fewer commercial games than the Mac platform.

Conclusion: Sun desktop machines with SPARC CPUs are "workstations."

Re:It's a WORKstation because you find it at WORK

[Anonymous Coward]

Sorry, that's a load of shit. There's no games for a Windows-based duel Xeon workstation? This Sun machine stopped being a workstation when I put it in my house?

The industry has traditionally defined workstations as desktop machines with some extra oomph for commercial applications - SMP, ECC memory, SCSI, professional video boards, etc. Traditionally a multiuser OS was also required, but all the non-multiuser OSes are pretty much dead at this point.

That shitty sparc does not qualify. Neither does a shitty DEC Multia. They're both PCs.

Re:It's a WORKstation because you find it at WORK

[dago]

Well, on my desk (or under), I have a sparc with ECC, SCSI and it can host an optionnal professional video board.
Some other are also SMP.

Ok, feeding a troll, but anyway

Re:It's a WORKstation because you find it at WORK

[Anonymous Coward]

Talking about the Blade 100/150, which is very substandard when compared to any of the "workstations" over at Dell.

Extreme ironing..

[adeyadey]

Difficult creases in your clothes? I prefer a bit of Extreme Ironing meself.. (http://www.extremeironing.com/)

Re:EXTREME? How about stable?

[shadowcabbit]

Being an excessive literalist myself, I always think that "extreme" indicates that something is at the far end of a spectrum.

For example, this new chip is at the far positive end of the price spectrum, and at the far negative end of the "will I really need this in the next three years" spectrum. It, being on the far ends of two spectrums, qualifies as EXXTREME.

(Nevermind that my first online nick had xtreme in it. I was 15, sue me.)

Old IBM XT?

[WatertonMan]

Wasn't the 80286 in the old IBM XTs the "extreme" chip. At least I thought that was what the XT stood for. Maybe it stood for extra? Anyone know?

Re:Old IBM XT?

[shadowcabbit]

On my friend's computer it seemed to stand for "Xcruciating Torment". He hated that thing.

Re:Old IBM XT?

[steve_ellis]

The XT was still an 8088-based (or was it 8086??) machine, you are thinking of the AT (presumably Advanced Technology??). I believe the XT's claim to fame was the addition of harddrive storage (as well as more memory supported on the motherboard)

Re:Old IBM XT?

[Bored Huge Krill]

no, the XT stood for "extended technology" and it used an 8086 (as opposed to the 8088 in the base PC). The AT ("advanced technology") was the first IBM machine to use an 80286

Re:Old IBM XT?

[makapuf]

except 286 were at, xt were 8086 & 8088.

LMAO

[Mitchell Mebane]

2 megs of pron, eh? I want it! ;D

Anyways, here's the _real_ text:

Pentium 4 Extreme Edition - XeonMP yields are damn good, it seems!

Intel Developer Forum Cache for questions

By Nebojsa Novakovic: Tuesday 16 September 2003, 18:14
WHEN, AT today's IDF opening, Louis Burns demonstrated a high-definition video stream running on a "mystery" desktop processor, everyone must hve thought it was the upcoming Prescott part. Wrong! It was the (also upcoming), previously unheard of, even at The Inq, Intel(R) Pentium(R) 4 processor Extreme Edition 3.20 GHz , with an extra 2 Megabytes. In Intel's own words, "this new processor will be targeted at high-end gamers and computing power users."

As a matter of fact, 2MB cache will help a lot those users whose apps (including games and such) have a lot of big cache-friendly pieces of code and data, but probably not the data-streaming intensive stuff. I do expect to see speedups anywhere from 2% to 20% depending on the application, maybe some more if using multithreading/multitasking (large cache can keep in code / date pieces from more threads).

However, this doesn't seem to be a new CPU in reality - after all, Intel is doing very well with its XeonMP 2.8 GHz 2 MB cache CPU, and how much effort does it really take to repackage it for the 3.2 GHz / 800 FSB desktop with less stringent thermal and reliability requirements than the big iron, anyway?

Intel would gain a lot with this move. If, touch wood, there are problems with Prescott, a large-cache Pentium4 part will provide some buffer against large-cache Athlon64 (i.e. rebadged Opteron) parts. At the same time, enormous extra benefits from the economies of scale would further reduce the identical die XeonMP manufacturing cost, helping Intel compete better on the quad-CPU server front as well. Interesting move? I think so. Let's see how the beast performs in real!

Re:LMAO

[crizh]

What's truely funny is what Burns said to the Inq just before the speach he unveiled this processor in...

"LOUIS 'THE THIRTEENTH' BURNS suffered a slight jaw-drop today as we asked him about the Pentium 4 Extreme Edition Nova told the world about over here(link to original article-crizh)

The regal one denied all knowledge of any such chip, though there was something in his startled manner that suggested otherwise. "Never heard of it," he muttered.

"Where you from?" he demanded to know.

"The INQUIRER."

"Ah, you're not Mike," he said, which of course was true.

"Anything else?" He growled.

"You still read the INQUIRER first thing every morning?"

"Yeah," he fessed. Though his kids he said were always puzzled by the amount of times his royal features gaze out from our front page, he said.

"Say Hi to Mike," he said. Hi Mike.

Burns is due on stage in a couple of hours. Will he talk about an Intel Extreme Edition Pentium 4? That remains to be seen. But if he does, I guess he's fuming right now. Best of luck Lou. "

(quoted from the Inq)

OfficeSpace

[Anonymous Coward]

HIYA FOLKS! How bout some Jalapeno poppers, and EXTREEEME fajitas?
(obligatory OfficeSpace reference)

128 & 32 bit Pentium versus 64 bit Athlon

[zymano]

If your thinking there is going to be a huge difference , you would be wrong. It's how efficiently you use the silicon and not the bits. 64 bits only means you can access more memory. This whole bigger than 32 bits is better is misleading .

Apples compared with Oranges

[zealotasd]

You are making a incorrect comparison in computing technologies. UltraSPARC III is for higher precision, but it is way out of its competitive market by two years ago. Pentium 4 is built for highest performance at the expense of power consumption. In a more objective comparison with the UltraSPARC III, we would compare performance/initial cost/power consumption (and forecasted power consumption cost to price barrier). UltraSPARC III is built for good performance on its implemented hardware, thus it utilizes its bus and memory architecture to optimum. The Pentium 4 does not perform with the mathematical precision and architecture efficiency as does a UltraSPARC III. The Pentium 4's memory architecture isn't even being used to full efficiency because of the nature of x86 being a pro-legacy architecture.

The biggest black sheep of the industry is the legendary Alpha architecture. It's a 100% 64bit precision platform with highest efficiency per watt and it was purposely bought by Intel to be silenced and migrate all its users to the Itanium architecture. Not even an Itanium2 can perform as well as an Alpha of two years ago (21264/ev6). The only downfall of Alpha is the legitimate and objective comparison of performance/initial cost as being the notion it is highly non-competitive with other offers. The reason it is not as competitive with other architectures is not based on fabrication costs: it is based on it being the better architecure that was purchased before its parents' bankruptcy (DEC...Compaq?), and to try to recover the R&D costs of the overly-invested lesser architecture known as Itanium.

People who still use Alpha already know that if it is buried then the only logical successor would be a Power4 hands down. All the while, HP's PA-RISC is being incorporated into the same Itanium architecture to migrate its dwindling userbase to Itanium. So much is going wrong in the idustry it makes me sick to the stomach.

Re:Apples compared with Oranges

[Hoser McMoose]

Please allow me to be the first to say:

What the hell are you talking about?!?!

UltraSparc is "higher percision"?! I think you've been hitting the crack-pipe a bit too much! Both chips offer 64-bit floating point registers, and with the P4 you can even use 80-bit fp values (though to the best of my knowledge, no one does). They have the exact same percision, as does every other major processor design on the plannet, just like they have for quite some time!

Now you think Intel bought the Alpha to "silence" it? Intel bought the corpse of the Alpha post-mortem. It died through YEARS of mismanagement and poor marketing at Digital. Intel gave them a helping hand by first buying the fab for the Alpha, since Digital was doing a piss-poor job of actually manufacturing the chips. They didn't get the rest of the chip design until after Digital was bought out by Compaq and than being prepped to be merged with HP. This was years after all real talent behind the Alpha had left.

Yeah, Alpha was a great processor in it's day, but it's day is LONG since gone. Ohh, and the Itanium2 beats the snot out of EV6 in every benchmark that compares the two! Look at Spec, look at TPC, look at any numbers you want, the EV6 loses them all. Even the "new" EV7 (a chip that should have arrived long ago, but just came out at the start of this year) is struggling pretty badly, mainly because after all the mismangement of Alpha lead no one to care about it anymore.

As for most old Alpha users, my guess is that most are switching to x86, and possibly thinking about x86-64. Heck, even Cray is switching to Opterons for their next big machine. When the Alpha first came out it might have been able to beat x86 solidly, but now, even with boatloads of bandwidth, the EV7 has a heck of a time beating x86.

Extremely irritating

[0divide]

this whole "extreme" moniker is getting so damn tired. When I first heard about "Airport Extreme" I kept thinking of some kind of ice cream dessert, with nuts and marshmallow cream.

Pentium Extreme Edition makes me think of some kind of fiber supplement--when your Pentium needs "extreme" cleaning or something.

are there "exteme" computer users out there? what, do they take the cover off the machine and play speed metal while surfing in bike pants and sunglasses?

who does the marketing for these losers?

Re:Extremely irritating

[normal_guy]

Yes, we do play speed metal while surfing in bike pants. No sunglasses.

And here I was..

[bob670]

being all pleased with my 2gHz Celeron laptop I just got. I know there's a market for this, but it really strikes as just marketing to piss on AMDs 64bit parade (kind of like how nVidia always conveniently had a new Deotonator release that would boost performance 29% every time ATI released a new card, back when nVidia was on top performance).

Speed / Cache is irrelevant *soon*

[Bruha]

In the next year we'll see the first solid state hard drives (Some that will run fast or faster than the processor) and faster RAM that would run the same speed as the processor.

Cache on a processor would be redundant if you can access the RAM at the same speeds. AMD is aware of this and are working to make compatible products.

Solid state drive/memory that runs at compatible speeds as the processor will probably reduce the need for what we call ram these days and operating systems could just use the drive for it's RAM.

If you're thinking of buying the latest and greatest I'd wait. Many things are about to happen and it'll be worth just keeping the money in the bank for now. Most people in the 2ghz range dont need any upgrades right now unless it's in the graphics department. I bought a ATI 9700 Pro last Janurary and it made more of a difference in my games than having any faster processor could.

Re:Speed / Cache is irrelevant *soon*

[darkwiz]

In the next year we'll see the first solid state hard drives (Some that will run fast or faster than the processor) and faster RAM that would run the same speed as the processor.

Cache on a processor would be redundant if you can access the RAM at the same speeds. AMD is aware of this and are working to make compatible products.

No.... we won't. What you are describing is insane. Come on: 3.2GHz x 32 bits? Access/transfer times over a full scale bus with a latency in picoseconds? Um... no.

There is a reason no one has done that yet - made system RAM the same speed as the CPU - and it ain't economics: it is physics. Nature does not take bribes.

Look, it isn't that it is too expensive to make fast RAM. And it isn't the distance - it is the capacitance. The problem with fast RAM is getting that signal off chip to the CPU. And the wires that connect the RAM and CPU are orders of magnitude higher capacitance than the wires on chip. That is a fundamental problem which you won't overcome without a fundamental change in how you move the data around.

Solid state drive/memory that runs at compatible speeds as the processor will probably reduce the need for what we call ram these days and operating systems could just use the drive for it's RAM.

Um.. no. Never will that be the case except in situations where using an archaicly small amount of processing power is adequate. Storage technology, as it is formulated now, cannot approach the speed of access, communication, and storage that even a low grade CPU would use for cache.

Maybe - MAYBE when we are using diamond wafers, high-temperature-superconductor-nanotube-quantum-d ot wires, and other buzzwords.

Re:Speed / Cache is irrelevant *soon*

[gerardrj]

On-die cache will never be "irrelevant". Even if the main RAM system runs at clock speed (or faster), there is still latency in getting the signals from the CPU, to RAM and getting a response.
The difference could be as much as 10x.
A solid state "disk" (SSD) would suffer even higher latencies with all the command overhead and the several bus systems that must be traversed/translated.

We've been promised SSDs for years. The last time I saw SSD units that were of a usable size, and reasonably priced compared to rotational media was in the mid 80s when 128KB was a lot of storage. Of course, the mid 80s is also the last time I saw a desktop computer that ran the CPU, RAM and system bus at the same clock speed.

Re:Speed / Cache is irrelevant *soon*

[akuma(x86)]

It's amazing how people spew out assertions without first stopping to think.

The fastest velocity that any human has observed is the speed of light. Some theoretical physicists believe that this is an absolute upper bound on speed :) Last time I checked, that's 3e8 meters/second...

Or in chip dimensions, that's 30 cm/nanosecond. At 10 GHz (0.1 nanosecond cycle time), light can only travel 3 cm in a clock cycle!!! That's approaching the typical dimensions of a chip! That's an absolute upper bound on speed -- this is light traveling through a vaccuum. Electrons traveling through semiconductor material won't go near as fast. You need "on the order" of a clock cycle just to traverse a typical chip. The reality is that as we scale technology forward, it may take several clock cycles just to traverse a chip.

I think you need to stop and reconsider your assertions. Latency is here to stay...it ain't going away unless someone finds a way around the current laws of physics.

What we need to do is realize this and architect around it by trying to hide latency in various ways (caching/buffering, prefetching, etc...)

Re:Speed / Cache is irrelevant *soon*

[Firehawke]

Now, if I understand the theory of SSDs correctly.. SSDs were moderately common back in the mid-80s in the form of Hard Cards, expansion cards that acted as disk drives but used memory chips. The largest I ever saw of these was 20MB, and I've always wondered from time to time if the technology would still hold up today with modern memory architecture. It would probably be more expensive, but I do have to wonder at what kind of performance such a drive system would use.

Re:Speed / Cache is irrelevant *soon*

[gerardrj]

The reason SSDs were common was they plugged in to the computer's internal bus system. The floppy and hard drive systems were connected via a very slow bus. Combine the slow bus with low rotational speed, and high seek times and even an expensive SSD could save you hours of waiting every day, thus lowering the TCO.

HD manufacturers have made super efficient motors, added lots of cache, and made the actuators move very quickly and reliably. A 120GB HD today daws only about 6W during read/write (4 at idle). When it transfers data, it does so over a bus that's about 2x faster than the sustainable write speed of the drive. You can get all of that for less than $1/GB

PC100 would make a nice SSD, it's still in productions and reatively inexpensive. Add a PCI controller, memory controller, and an intermediate microcontroller to translate sectors to RAM space and you've got a RAM disk. Problem is that even at aggressive wholesale such a device would cost about $25/GB.

Sure it would run circles around a HDD performance wise, but doubt the time savings in daily use would amount to much. For SSDs to become viable again we need a RAM spec that isn't based on speed, but pure density

Re:Speed / Cache is irrelevant *soon*

[GISGEOLOGYGEEK]

Solid state hard drives have existed for at least 5 years. Have you never seen those funky PCI cards with half a dozen sticks of ram on them?

Coming Soon!

[Luigi30]

The Pentium Alpha Zero 4 Plus Championship Rainbow Remix Euro Fighter vs. Capcom 2 Revision 1.7!

If you're wondering where the L3 cache came from..

[SuperKendall]

They stole it from Apple's new Powerbooks. So that's where it went!

If you like Extreme Teens..

[Anonymous Coward]

Hmm.. forgot what I was going to say. I will jerk off now.

IEEE 754 floating point: pros and cons

[Latent Heat]

This discussion is taking a hardware-geek turn in talking flops and mips and bits and watts. People are forgetting that Intel (at least the traditional 8087-style FPU, don't know about SSE) is considered the Gold Standard for IEEE 754 while the Alpha floating point is alleged to cut corners, and you need to factor this all in for comparing power and performance and flops per kwHr.

"Professor Floating Point" argues (trolls?) that Java sucks for floating point, but you need to wade through his entertaining read to see what he says about precision and processors and flags and traps. This dude (you have to scroll down to the section on IEEE 754) takes the opposite view: the two dudes are Dole and Clinton on that stupid "60 Minutes" thing, but between the two of them you kind of get the picture.

The picture is that a lot of why Intel is slow is they are doing an arguably higher-quality floating point, and dude number 2 reinforces the view that Alphas and SPARCs and other RISCs do what dude number 1 says is cheap floating point because they don't want to be slowed down.

Check the two references out and you decide whether everything Intel does is really needed, but it is important to know that an Alpha is not just a better, cheaper, faster chip -- it does its floating point differently, and some people (or perhaps just Kahan, dude number 1) care a great deal about this.

Think of the 8087 as a vehicle with "off-road" capability -- yes, it is slower and guzzles more gas, but if you gotta go off road, you gotta have it.

For the Love of God, Do not let my GF see this

[simontek2]

Shes a Computer art student at Scad, who thinks she has to upgrade all the time!

Re:For the Love of God, Do not let my GF see this

[Anonymous Coward]

Lol.. ok come clean. The whole purpose of your post was to gloat about having a girlfriend and nothing else. Fucker.

Re:EXTREME? How about stable?

[Bored Huge Krill]

ok, so you probably don't want one of these. Not everybody will, and aren't expected to. Intel have stated that this is not supposed to be "mainstream" ever. I have my doubts about that last bit, though...

Re:Pentium is dying!

[ChaoticLimbs]

so, I could do just as much work then with a sixteen bit computer as a 32 bit computer so long as both had the same amount of RAM?

I saw this once before...

[lhpineapple]

It started off with Street Fighter 2. Then Street Fighter 2 Champion Edition. Then Street Fighter 2 Turbo. Then Street Fighter 2: The New Challengers...

I think it took them 3 or 4 years to actually get to Street Fighter 3.

Re:I saw this once before...

[Anonymous Coward]

Closer to five or six, I believe. Alpha was out before SF3, and that was 1995. Using the US release date for SF2 (1991, IIRC) puts it about five to six years.

One word:

[Anonymous Coward]

Opteron

Woo and Yay

[NeoGeo64]

Anything to compile GNOME faster...

Still doesn't beat a dual 2GHz G5 machine.

Re:Was it named after Apple's AirPort Extreme? *NM

[Anonymous Coward]

Dear /. Editors: if you took any time to think, u'd recognize a subtle commentary on Intel's potential attempt to lower the value of Apple's marketing of AirPort Extreme. Ever heard of Centrino or WiFi Everywhere? Intel has much invested and much at stake there. In any case, it would seem Apple's use of "Extreme" in product names is gaining in popularity.

Intel's chips are in one sense, MIPS chips

[Anonymous Coward]

Part of Intel's insanely complex P4 architecture is that it does benefit greatly from the knowledge of traditional MIPS or rather RISC designs. Essentially, all supported instruction sets are broken down into "micro-ops", which is a small set of very efficient, simple, and fast operations. These are then sent through the system yielding MIPS performance, but with compatibility with past instruction sets.
Whoa, so the P4 is "emulating" IA-32, x86? Yup. Tack that in with the complexity of executing instructions out of order, branch logic and other optimizing parts. That's why it has a 20-stage pipeline for code execution with an undocumented amount for decoding/translation.
It's interesting how much different the P4 design is from the Athlon XP. In one sense, Intel's direction was to do everything as quickly as possible where AMD's was to keep things simple and efficient per clock. Who's winning? Hard to tell... For the majority of consumers, it's apples and oranges. Err, Pears and Oranges.

*soon* is a decade or more away

[MarcQuadra]

I'm sorry, but I get a strong feeling that you're way out in dreamland.

Let's get some things straight first, no hard drive or ram can be 'as fast' as a processor, because that's like saying my coffee cup is as fast as my bicycle, it's meaningless.

A solid-state hard drive has to get all that data addressed, and it has to pump it over some sort of pipe at least several inches long, the addressing will put a buttload of latency in there, and that pipe would bring the bandwidth WAY down.

Now I wouldn't fuck with a solid-state drive at the end of a Ultra320-SCSI pipe, but that's STILL 1/10th the memory bandwidth of a modern DDR400 system. BTW, those have been around for AGES, I used to have a solid-state SCSI drive on my old 25MHz Mac, it was pretty fast, but nowhere near the internal ramdisk's speed.

RAM now works 'at the speed of the processor' if you think about it. My Athlon can chew about 2100MB/sec which is EXACTLY output of the memory I'm running (that's the 'sync' in SDRAM). The only way to change that would be to 'widen up' the CPU FSB. You could put single, dual, or quad-channel memory on your Athlon and it wouldn't make a dime's worth of difference in any benchmarks, the back of the chip is the limit, and current RAM meeets that need.

It's general knowledge that the more storage you can arrange for, the more complex your addressing system has to be to keep it tamed. Here's an example:

When your CPU asks for something it needs from RAM it asks for the contents of a block of memory, whose address is held in a pointer that is dynamic, but readily available. The CPU just 'gets' it. It's even better if that block is already in the cache, as the cache buffers will satisfy the request before the memory controller even bothers to retrieve the block from RAM.

When your app needs something from DISK it has to send a request through the OS (in RAM) to do a lookup in the filesystem and give an address, which is shuttled over to the disk driver to fetch from the drive and back to a generic filesystem driver to present to the app. Should the filesystem not have that data cached it has to perform a complex lookup of where the hell that file actually is on the disk, often traversing several directory files. It's very complex.

What you're saying will eventually happen, but not for at least a decade. Someday we WILL drop the two-tiered approach to personal computing, and ther'll be 'unified storage' for running apps and storing files (like the palm pilot, but better) and it will be good. Until then we've not yet miniaturized the electronics enough to move over to that paradigm. I think nanotech/biotech will play a HUGE role in making the memory, cpu, and IO processor components small enough to run cool and unplugged.

holy....

[MoFoQ]

MoFoQ drops the spammer he was torturing with his pitchfork and flaming torch and starts drooling....

man...wish I had a 800FSB board, and here I am with a 533FSB board like a sucker....

IEEE754 is only a suggestion, like pants.

[zealotasd]

From the URL you provided, I received the opposite idea. I don't like diving into metaphores, as I think if slashdot.org participants were awarded on the accuracy of their metaphores then the Trolls would be posting minimum of "+5: Shakespeare." The following quotes, taken in context of the sections they were given, speak for themselves that IEEE754 is a standard that allows poorly written FPU code to be written on an architecture that supports a poorly-written FPU compliancy. At-least with non-IEEE754 architectures, code is written for the architecture or abstracted within the application for somewhat cross-architecture porting, and bugs are not hidden. I'm not saying IEEE754 is a bad standard, it just allows bad things to happen while allowing the application to progress without crashing. Yet, my idea of cutting corners is obviously different then yours. I understand "cutting corners" in both a Biblical perspective as well as the flawed perspective as denoting quality. The Alpha is not cutting corners, neither are the other architectures. The architecture designers simply recognize IEEE754 as what it realy is and let the software-side of things implement IEEE754 while making the FPU at the circuit level have performance and optimizations. Perhaps, I still don't understand your idea of non-IEEE754 architectures "cutting corners", whereas they lose precision in some way? I don't see any bug-reports of Alpha losing precision in its non-IEEE754 FPU, do you? I'm getting ahead of myself, here are the following quotes from the ,a href="http://www.netbsd.org/People/Pages/ross-essa ys.html">webpage URL you provided in your post that I consider to be self-explanitory...

Con's of IEEE754

4.It is expensive to implement. When the standard was ratified, exactly two companies produced floating point arithmetic chips that attached to microprocessor CPU chips. Intel sold the 8087 and National Semiconductor sold a chip for the NS32 processors.

Here is how they compared on multiply:
.....70 uS Intel, full IEEE 754 support in HW
.....5 uS NSC, no IEEE 754 support

Example two, about 17 years later...

Here is how they compared on SpecFP95:
.....9 Intel, Pentium II, IEEE 754 compatible
.....50 DEC Alpha 21264, IEEE 754 hostile

5.The targeted users have ignored it for 20 years.

6.No RISC microprocessor has ever implemented IEEE 754 floating point.

7. Buyers and Manufacturers of scientific (number crunching) computers are often IEEE-754 hostile.

10. The exception mechanism is incompatible with modern kernels and computing hardware.

Pro's of IEEE754

Continuing computations after errors does have a certain value...
Unfortunately, the net result of this new behavior in practice is something much darker... inevitable -- is that people aren't finding the stupid divide-by-zero bugs any more. Is this really a big problem?...2,000 programs in The NetBSD Packages Collection. Many of these programs use at least a little bit of floating point, often for noncritical functions such as statistics or image conversion. Almost every last one of these is developed or is maintained on the PeeCee. Consequently, it isn't very important for those developers to fix the stupid math bugs, especially if it just affects the statistics in cycle zero, or the color chosen when the intensity bits are zero, or whatever. The problem here is that the breakage just isn't serious enough, when you do have error-ignoring IEEE 754 floating point. As a result at any given time many of these packages are broken on a system that doesn't implement the IEEE 754 floating point standard. When run on such a system, the program traps and stops. That the trapping routine lacked any importance in the active case is now of no help. While one can argue that it's just great that 754 enabled

Re:Revelation

[Anonymous Coward]

Actually, you are not her type.

She's into guys who are smart but also well-rounded with other interests such as cycling and lindy hop.

Sorry to burst your bubble.

3.20 GHz ???

[mmu_man]

Just to display video in a window ?
They must be using winXP...

Nice to see that many years after they can do what BeOS does on a P11 300 :)

The level 3 cache didn't help Xeon much

[Anonymous Coward]

Don't hold your breath. Read the artilcle here:

http://www.tomshardware.com/cpu/20030811/index.h tm l

What difference do 5% make? The new CPU won't be only 5% more expensive...

Re:Processor-Intensive SW: Engineering Application

[Moraelin]

Well, honestly, nowadays "it beats a Sun box" doesn't even say much. It's just marginally more meaningful than "it beats my old ZX Spectrum."

The Suns still struggle barely above 1 GHz, have a slow cache, and so on. It also doesn't help that they're still saddled with SDRAM memory, too. (At least in the case of the cheaper workstations, on a 32 bit memory bus too.) If we're talking programs that draw something, it also doesn't help that they're saddled with outdated _and_ overpriced video cards. And so on.

Even without SSE, there's no way in heck for that UltraSparc III to keep up with a P4. E.g., Sun's Java doesn't even generate SSE code, and it still runs faster on Windows than on Solaris. Go figure.

For all the BS about the advantages of 64 bits, the reality is that in 64 bit mode an UltraSparc actually runs _slower_. So be thankful that most of the apps for it (and certainly all benchmarks) really are compiled in 32 bit mode.

Frankly, other than a few PHBs, and a couple of people who think they're some form of resitance against Wintel if they buy Suns, the rest of us don't even consider Sun to still be in the race any more.

So yeah, your words about running Linux on an Athlon or Pentium reflect exactly what I'd say to anyone considering a Sun box: Get the cheapest PC that Dell sells, or build your own Duron system, install Linux on it, and there you go. You now have a Unix workstation, and it runs circles around any of Sun's workstations. Or, much as I'm no Mac fan, get a Mac. It'll be 64 bit, and based on BSD too.

Re:Processor-Intensive SW: Engineering Application

[Junks Jerzey]

Please check the SPEC web site [spec.org] for a performance evaluation of the Pentium 4's floating-point (FP) performance. In particular, it outperforms the UltraSPARC III even though the latter has a 2-to-1 advantage in the width of its databus -- 64 bits versus 32 bits.

Note that:
1. The Pentium ONE had a 64-bit data bus. This is nothing new. The FPU was able to load and store 64-bit values directly.
2. The x86 FPU is EIGHTY bits internally. Not 32, not 64, but 80. Again, this is not new.

Interesting...

[pmz]

Will Extreme Programming work on this CPU?

Well, perhaps not until Microsoft releases "Windows XP Service Pack to the Max!"

i think this will be good for high end users

[john_uy]

the 2mb additional cache will virtually make the p4 processor look like a dual p4 1.6ghz with 1mb of cache each (when hyperthreading is enabled.) their hyperthreading feature will be much more effective. the processor will be able to cache more threads and avoid too much misses thereby greatly improving performance for cache intensive applications.

this will give a big performance boost to memory and cpu intensive applications (not much on games) such as graphic programs like photoshop, aftereffects, lightwave, maya, etc. in addition, it will also perform better in low end server environments where database and application serving programs are used. this is a good option to use in blade servers using p4 (instead of getting xeon.) aside from server applications, possible scientific applications will be boosted because of the big cache (where they can use simd for faster and higher precision calculations.)

i am surprised that a lot of people look at faster cpus for just games. there are a lot more applications that will benefit than games.

I love the Dead Milkmen...wait... oh hell

[FreedomOfSpea-MMNnnf]

I tripped over the ottoman one two many times

I tripped over the oto-man

and al-most lost my mi-iiiiiiiiiiiiinnnnnnnnnd

off topic but still X-Treme to the max

Prescott and XTREME thoughts

[gbulmash]

The developer's forum still runs today. They could be saving a Prescott-related announcement until today to go out with a bang.

Of course, if they make no announcement, their silence may be even more of a bang, as it will prompt all sorts of speculation.

As for the Xtreme... I just don't see it selling for too much more than the regular P4 3.2. If its max speed boost (due to cache) is 20%, who would pay a 150-300% premium over the other P4, regardless of how hardcore they are?

Intel's probably modeled the price sensitivity of its target market three ways from Sunday and set the price in the upper ranges of what they think will work, but still... It's probably going to have to street for $1000 or less for it to move in any sort of quantity

BTW, xicomputer.com is taking orders for Athlon 64 systems already. They're not providing any specs about the chip, but they're selling it.

All I can say is that I'm ready for a hardware upgrade, but have been waiting for both Prescott and Athlon 64 to release, so all the hardcores can do their benchmarks and argue which one is superior. I could lurk, read, and figure out which chip I want.

But if Intel continues to play coy with the Prescott release date, reference systems for reviewers, and competitive info... F'em. I'm not buying the P4 XTreme.

I can afford to upgrade once every couple of years and I'm jonesing for a big dose of new. If AMD's ofering a pure fix while Intel's just cutting the product with cache... I may just shift my buying over to AMD's street corner.

- Greg

80286 was the AT

[Anonymous Coward]

8086/88 was in the XT
80286 was the AT

Stoopid /. kiddie

Hey flamebait moderator, I own a Powerbook!!

[SuperKendall]

Can't anybody take a joke? Sheesh!