Slashdot Mirror
New Pentium 5 Details - 5-7ghz?
zymano writes "This article gives some details on Pentium 5. It will have 64 bit extensions and maybe a 4000 mhz frontside bus. Quote from the article,'The Pentium V is likely to fly along at between 5GHz to 7GHz, have 2MB plus of level two cache, be built on a 90 nanometer process, and have a stackable design. '"
Is it me or would you pronounce that "Nail 'em"? A dig at AMD perhaps?
Mid-Eastern Pennsylvania Gaming Convention
The fifth fifth processor.
64-bit extensions? In the same way AltiVec was 128-bit extensions?
The 4GHz bus does sound good, thought.
That what was all this school was for... to teach us how to solve our own problems. -- janeowit
... will it be able to do Math correctly?
Donte Alistair Anderson Roberts - hi son!
Karma: Chameleon
'The Pentium V is likely to fly along at between 5GHz to 7GHz, have 2MB plus of level two cache, be built on a 90 nanometer process, and have a stackable design. ' And raise the temperature of the room it's in by 50 Celsius.
Karma: Excellent^(-t/Tau), Tau=Wittiness/Trollishness
The chip will sample internally at Intel in January 2004 and will take between four to six months to get to market. The Pentium 6 will follow a very similar schedule.
The Pentium V is likely to fly along at between 5GHz to 7GHz, have 2MB plus of level two cache, be built on a 90 nanometer process, and have a stackable design.
The processor we believe, sits in the LGA 775 pin socket, and above it is a very thin heatsink. But, according to sources close to the firm's plans, another permeable heatsink can sit between this and another microprocessor module, giving a stackable design.
The final design of this arrangement is not set in stone.
According to this source, and the details have not been confirmed, a module sitting on top could provide 64-bit extensions.
And the source claimed, Microsoft is ready to launch a version of Windows called Elements with 64-bit extensions.
The idea seems to be that people can buy a 32-bit module, and then add in the 64-bit processor.
There are three samples of an arrangement of the Pentium V here in Taiwan this week, with a very thin processor and lots of wires and patches stuck on it, just to show proof of concept.
The Pentium V could have a front side bus speed of as much as 4000MHz, the source claimed, although this may be reserved for the next chip along, the Nehalem.
Looks good for your age..
.. That article sounds a bit too good to be true, I'd like to see their sources. Some of those figures seem to be plucked from thin air. ...
..
They would need some serious cooling going on at those speeds
Anyhow Imagine a Beowulf Cluster of these
Electronic Music Made Using Linux http://soundcloud.com/polyp
This article is all speculation...
Ok here, um the next AMD processors will be faster than before, have more cache, maybe some new instructions [doworkNow! then doworkNow! (ext)].
I must be an AMD insider now, l33t l33t !
Tom
Someday, I'll have a real sig.
Stackable designs sound really cool in the sense that you can cut latency between processors (for things like cache coherence) to rediculously small levels, but what about cooling? Cooling ability is roughly proportional to surface area, and two stacked chips will make twice as much heat but have almost the same surface area as only one (as two sides cancel out). This has to be a problem.
No this is not a troll. I honestly wonder how they expect to accomplish this.
Anyone know?
Cheers,
Justin
Intel is up to Pentium 5 now, my question is when will they drop the 'Pentium n' line and go with something new. By the same token, Apple as well is up to 5 with their G-line. After a while, it gets a little rediculous and reduntant, so companies come up with a new product line (Geforce FX, kinda hybrid cause nvidia didnt want to loose the geforce recognized name). I have to say that I prefer AMD's system more with the lettered naming system, XP, MP, etc since atleast its different. So how far do you think pentium will rise to? I have a hard time saying 'Pentium 7'...
"What can a thoughtful man hope for mankind on Earth, given the experience of the past million years? Nothing." -Bokonon
I still prefer AMD chips for some reason.
Outdoor digital photography, mostly in New Engl
but will actually perform the same as a 2.5 GHz Athlon
1p}{ 1 sp34k |33+ +|-|e|\| p30p13 \/\/il| 8e i/\/\pr3553|)
Proccessor. Add-on? OS. Add-on? This sounds like a clever attempt to creat a support nightmare for anybody developing for the pentium pentium. Oh well.
Looks good for your age..
The article doesn't say the processor will have 64-bit extensions. The article doesn't say anything.
Some quotes:
"The Pentium V is likely..."
"The processor we believe..."
"The final design of this arrangement is not set in stone."
"...details have not been confirmed,..."
"... the source claimed..."
"The Pentium V could have..."
"...although this may be reserved for the next chip along, the Nehalem"
This isn't news, this is BS speculation.
I swear that the PIV 2.4 Ghz machines I've used are no faster that some of the P III 1 Ghz boxes I've used. We upgraded all our development boxes at work this way and there was hardly any notable improvement... yes, the memory is tricked out so we're not having swapping issues. But you run apache, mysql, and X on one of them and it just doesn't seem like an improvement.
Are they doing a direct trade off where they ramp up the clockspeed and break the instructions down so that less is getting done per clock or something?
Cheers.
Running at 5-7GHz is absolutely retarded for a processor to do. If you look at the way that every single "wire" in the professor acts, they all must be treated like transmission lines. just sitting there and doing thost calculations to find out how much power is being delivered would be the most bit*h/bullsh*t job every. A processor running that fast would probably lend its self to using onboard optical systems (waveguides) and running parts that way so as not to have to deal with running copper or Al and doing all of the insane calculations associated with that.
Oh and by the way, i'm running a PIII750 and the only things i would upgrade to are Apple and a 64bit processor. I'm not going to upgrade for a long time.
-=gabe2=- macbook dual 2.0
The clock speed hike reminds me that the P4 is slower clock-for-clock than the P3, and makes me wonder if Intel are doing this entirely for marketing reasons. I can't help feeling that they should start looking more closely at the other end of the market. Saying that 100W is acceptable in a desktop CPU does not make it so. For a large number of people 1GHz is fast enough, and a silent 1GHz chip would be more welcome than a 5GHz chip with a built-in tornado.
I am TheRaven on Soylent News
The Pentium V is likely to fly along at between 5GHz to 7GHz, have 2MB plus of level two cache, be built on a 90 nanometer process, and have a stackable design.
Will it make coffee?
A "Pentium" ("penta"=5)came after a 486, which came after a 386, which came after a 286, which came after a plain old "86"... So this one is the "Five-five"... Such wit, those marketeers...
There's no point in raising the speed of the processor to 5GHz if the memory speed (esp. latency) can't keep pace.
4GHz front-side bus? Yeah, right.
I calculated a while ago that assuming that RAM was 5 cm away from the CPU, at 5 GHz a clock cycle would be lost on waiting for the signal to travel the 5 cm to the RAM and back.
If the speed of light is not far from being a limit at this point, then clock speed improvements can't continue working for long.
Besides, there's the question of whether it will "fly" or not. Clock speed doesn't measure performance. It especially says nothing of the performance of a new chip.
P-V should have 64bit extensions for both pointers and basic math.
64bit pointers and basic math on those pointers, are really what people desire so that more than 4GB can be trivially addressed in a single process's virtual memory space. Think about people who want to manipulate a video file that is larger than 4GB.
AltiVEC **128 bit** is just wide data manipulation and is of no use for those that require large memory footprints. It has the same 32 bit address lines and pointers at a 60MHz Pentium I.
That being said, P-V should also have more than the current 36 bit of physical address lines. I'm guessing they will have 40 usable bits or so of the address bus to physically address memory.
So if you want to put in more than 4GB of RAM you can. But if you don't, 64 bits will be useful to address more than 4GB of a video file sitting in virtual memory.
Windows will certainly get just as bloated and suck down all that speed and power. That's how it has always been, and always will be.
I'm waiting for the day that Microsoft Windows GUI will be fully raytrace/radiosity/photon map rendered.
I won't be happy unless I have a glass refracting mouse cursor made up of at least 64,000 triangles, updating at no less than 60fps. It had better be casting both a shadow and also focused light complete with chromatic aberation.
That'll show those OSX zealots!
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
I had a P5 back in 1996! And it could whip through Windows 95 like nobody's business.
If Windows can't do it alone, our IT department has a magic bag of software tools that'll finish the job. Maybe McAffee with double-dare mega hueristics wicked up full blast.
Intel: "Oh my god this is so AWESOME because we have super high gigahertzian-ness and you dooooooooon't!"
AMD: "Uh... We don't need GHz to keep up. That's what We have these new nifty + ratings eh?"
Intel: "Uh... HYPER-THREADING! WE'RE AWESOME!"
AMD: "And we have a better 64-bit processor than your dinky Itanium. It doesn't need to 'emulate'. What a bunch of idiots."
Intel: "OMG OMG! WE HAVE ULTRA 1337 SPEED! I MEAN 5-7 GHZ AND 4 GHZ FSB! I MEAN AREN'T WE COOL! 64-BIT EXTENSIONS!"
AMD: "... Shut up. Better yet, don't shut up. It's good for our business, because at least we're delivering."
Face it - the only way we'll see the end of x86 is if someone builds a new, non-x86 chip that can still run all that existing x86 code at least as well as the best existing x86 processors. Otherwise it's just another niche architecture, and no-one's going to "upgrade" to it.
Intel forgot that, or thought they could force it on people anyway. AMD remembered, but took the easy way out & just extended things. Similarly, IBM got it wrong with OS/2, and MS jumped straight in with Windows. Note how long it took before MS was able to phase out DOS completely, even so.
Why would anyone engrave "Elbereth"?
They did. It's called the Itanium. Look how well that's worked out.
Even running it outside of a server, you have to have a special version of Windows, which doesn't have all of the features that the 32-bit Windows does (Windows for the Opteron line is supposed to fix this). It's hideously expensive, meaning fewer people adopted it, which meant that costs stayed high, so there was less encouragement for people to adopt it, even within the server/workstation market in which it was sold.
AMD is going about it the right way. Allow a smooth and orderly transition. That they're going about it using a 64-bit adjustment to x86 makes it more difficult to move on to a new architecture, but perhaps in a few years, this will be looked back on as a successful model.
You can never go home again... but I guess you can shop there.
Hahaha, does anyone thing this sounds like a leak from Intel in an attempt to dampen the tide of people eyeing the Atlon 64 FX? "Hey! Don't buy our competition's superior product. We'll have something that might be as good or better ready in.. er.. half a year! And we'll try to have it on the market in quantity... er... maybe in a year if everything goes perfect! What, things have never gone perfect? Sssssh.
A bird in hand is worth two in bush. Intel, you will now pay for your complacency. You did not believe the consumer market needed the 64-bit processor; it was cheaper to milk your enchanced Pentium Pro core a little bit longer. AMD had other ideas. Well, well, well.
I love competition!
Maybe their slogan should be "The Inquirer: All your baseless speculation are belong to us."
that we should ignore all of those silly Opteron and Athlon 64 announcements in the past six months, because next year Intel will announce something that will blow them all away, and lead us all back to the One True Processor Roadmap.
Does this qualify as a pre-announcement, that just happens to be overlapping a competitor's introduction? I seem to remember that several decades ago, another three-letter company got in a decade-long heap of trouble for just that type of behavior. (Amoung others, but then there are more stories of things Intel has done to keep AMD 'present, but weak.')
The living have better things to do than to continue hating the dead.
...likely to fly along at between 5GHz to 7GHz...
...and Windows 2005 (Code named Canyonero) will still manage to slow it to a crawl!
.
You're using her as bait, Master!
If you look at media benchmarks, encoding requires a lot of processing power. So, while ripping your DVD may not take any more time on your P3-1GHz versus your P4-2.4GHz, converting it to DivX MPEG-4 for your media jukebox will take significantly longer on the P3 than the P4. In fact, decoding H.264 video and WMP9 High Definition supposedly requires 3GHz (or the equivalent in AMD doublespeak) processors. Add to that the fact that you may want to do more than one thing at once (i.e. encode video in the background and play back another), and you will quickly run into a hard wall. Check out this link for a very nice roundup of how older processors fare against newer processors. A simple DV-to-MPEG2 conversion takes approximately twice as long on a P3-1GHz than it does on a P4-2.4GHz. That's a lot of time when you have a couple of hours of video to encode. Audio and image manipulation applications, video editing and the like will also benefit in similar ways.
Games, it goes without saying, scale in a similar way and a similar doubling of performance.
The caveat: for many business applications, you will hardly notice a difference. A faster I/O subsystem and more RAM, as you mention, will pay much larger dividends for these users than any processor upgrade will. In fact, this post is being written up on my trusty P2-400MHz all-SCSI box and it's still going strong, though it's getting a bit long in the tooth.
Just a scheme by Intel to keep those who are riding the fence on their side.
5-7ghz and yet i still find no scenario to replace my pIII .5ghz. other than games but i dont game that much anyhoo
For The Best Jazz/Hip-hop fusion > COlD DUCK
I'm wondering what type of PSU they're hooking this thing up to as well. I mean, at that speed and probably power consumption, we'll be seeing a whole new line of PSU's just to power the thing (not to mention needing a new video card to take advantage of it).
Wonder if I'll have to unplug my stove in order to allow my PC access to the ol' 220V, or perhaps I'll just ask my landlord for access to the MAINS.
Either way... I have this picture of the lights in my apartment dimming and my power meter suddenly spinning around at 60rpm...
It complicates cache design, yes, but it's a solved problem.
In x86, you can store into instructions. Even right before they get executed. Even right before they get executed by another CPU. And it all works right. Now that causes architectural complications.
Think about what that means. The superscalar processor is happily going along, executing several instructions ahead simultaneously. Then information comes in that some instruction already executed but whose results have not yet been committed to memory has been overwritten. The processor has to discard everything dependent on that instruction, back up, and do it over.
It sounds horrible. But if you view it as another case of speculative execution (where, at a branch, the CPU starts executing on both paths until the branch is decided) it starts to become clear how to implement this in silicon.
The key to all this is the "retirement unit", which first appeared in the Pentium Pro. The Pentium Pro was the first "modern" x86 machine. Up until the Pentium Pro, what went on inside the CPU was reasonably closely related to the user-level instruction set. In the Pentium Pro, the user-level and internal architectures parted company. Inside a Pentium Pro/II/III/IV is a dataflow machine, pipelining little self-contained operations expressed in an internal instruction set that's quite different from the one the programmer sees. The dataflow machine is front-ended by an x86 instruction translator, and back ended by the "retirement unit". The "retirement unit" takes the outputs of the dataflow machine, figures out which ones to keep and which ones to dump, and determines what gets stored in the programmer-visible registers and memory.
In addition, the Pentium Pro and later machines have far more registers in the CPU than the programmer sees. The Pentium Pro and later have 40 or more registers storing temporary results. Storing data in a temporary variable on the stack just puts it in a register representing that stack slot. There's little or no penalty for this compared to having the value in an x86 register. Eventually the retirement unit pushes the value out to memory (i.e. cache), but the processor doesn't wait for that event.
Once architectures broke the problem apart like that, the programmer-visible instruction set didn't matter that much. This is why RISC isn't very important any more. The original RISC idea, as expressed in early MIPS machines and the DEC Alpha, was to have simple, fixed-sized instructions, a simple CPU, and execute one instruction per clock. This made sense when non-RISC machines were executing less than one instruction per clock.
But the Pentium Pro architecture changed all that. Now, more than one instruction was being executed per clock in a microprocessor. To keep up, RISC machines had to go to similarly complex architectures, losing the simplicity advantages of RISC, while keeping the code bloat of fixed-size instructions.
There are other ways to accomplish the same result. AMD does instruction translation when instructions move into the cache. Transmeta does it in software when the program is loaded. But none of today's fast machines are directly executing what the programmer wrote.
That's why instruction set architecture doesn't matter much any more.
All this takes huge transistor counts, and acres of chip designers. (Intel's acres of chip designers, each in their own tiny cubicle, with one acre of cubicles per room, are at Intel's Santa Clara facility. I've been there, but fortunately don't work there.) But it all works.
Intel has built up the megahertz myth so much that most people who don't understand the work/clock cycle dynamics would need to see this type of 4-7GHz speed to even think about why they should upgrade to a next-generation Intel processor, even if it was only marginally faster computationally when compared to the P4. How much would you like to bet this "next-gen" processor has a 75-stage pipeline and a one-trillion transistor branch prediction unit to try to keep it working, not to mention the most-likely needed nuclear power plant water cooling tower that would have to be attached, or the 240V power outlet that would be needed. Yeesh.
today is spelling optional day.
When AMD's K6 came out following up on the massive name recognition of the K5 I was really looking forward to eventually being able to buy a K9. Think about the possible slogan: "Introducing the AMD K9. No fancy names, no gimmicks, just pure processing power that dogs the competition." Then when the K9 got older people would say "K9? Man, that slow-ass chip's a dog!" Yup, I was looking forward to both. AMD could have even made that robot dog thing from Dr. Who the mascot. But then AMD blew their chance by listening to some marketing twits and called the K7 the Athlon, whatever the hell that means--sounds sort of like someone sneezing right before they jump off a diving board.
I still bought one, though.
Well MY inside source tells me that Intel are ready to release the Pentium X, running at 50 Ghz and having a FSB of 25,000 Mhz.
.......... yes, yes ........ THEY'RE ON SALE NOW!
It has their patented uber-cool ultra-wizzer-extra-special 128-bit extensions, and it also has an expansion port that you can slap an extra processor on in case AMD releases a 256-bit processor in the meantime.
This thing is going to scream, baby! It will plug into existing Slot-1 motherboards, and will be built on a 2 nanometer process.
Microsoft are believed to already have a version of Windows running on the beast, with their new 'WTF That's Friggin Incredible Mate' extensions that go hand in hand with Intels 'Fuck Me If This Isn't A Faster Chip Than AMD Has' architecture.
Wait a moment
Umm.. My first PC had 640kb of RAM, which wasn't that bad -86. Now one could copy it 200 times in the cache of the P5.. I wonder, are we 200 times happier now? :)
From zero to a superheated lump of useless plastic in only 3.25 seconds!!!!
Intel are apparently in negotiations with the Malaysian civil construction firm that built the Petronas Towers to develop the heat sinks for these little cookies...
~
~
~
-- INSERT --
...but if you looked at Anandtech and read how fast the current Pentium IV (not EE) would have to run compared to the Athlon 64 FX in the areas where it really excels, it's not that far from the truth. Of course, Intel is improving their architechture (FSB, HT, cache size +++) also, so it won't actually come to that. But I suspect the difference in clockspeed for same performance might increase.
The reason? Intel has sold the GHz (aka the MHz) myth so well, they need to increase clockspeed in order to make their own customers upgrade, even if that means the performance/cycle has to suffer more than what gives optimal performance. Unlike AMD, they can't make up PR (Performance Ratings) because it'd look stupid, while AMD has a valid excuse with their chips being fundamentally different from the Pentiums.
Kjella
Live today, because you never know what tomorrow brings
Reading these amazing specs, with the attendant oohs and aahs and lots of ifs thrown in --"it seems, it could be" -- this gives me the feeling that it might just be vapourware, brought out at this time in the time honoured tradition of microsoft announcing products that do not even exist beyond specification form simply for the reason of cornering the market. AMD's Opteron and IBM's PPC 970 (G5 in Apple's Macs) are getting more press than the desasterous Itanium or even the Itanium2 for that matter.
My feeling is that while Intel is probably less worried about the G5/PPC 970 as their marketshare is very small, but is more worried about the effect a successful Opteron could have on the market, on the one hand not needing special recoding for 64 bit apps (compatible to x86 32bit) and more importantly what the Opterons could do to the server market, causing companies to switch their 32 bit Xeon stuff to 64 bit Opteron with little effort and low price.
I seriously doubt that all of a sudden next year, CPUs will be on the market running at 5 to 7 GHz without having serious cooling problems or running away from memory.
So, in summary, I think it's Intel's marketing department in microsoft mode:Vapourware.
Quite frankly the code names for the chips are all much better than the actual release names. "Katmai" and "Mendocino" are pretty cool-sounding (well to me anyways; "Deschutes" is a bit hard though).
And does anyone remember, right after the transition to "Pentium", how everyone was calling the different generations of chips? The original Pentium was the "P5", to distinguish between the prior 486 and the Pentium Pro ("P6"), on which all the Pentium II etc. offshoots were based (i.e. PII was "based on the P6 architecture") etc. But now what do we call the Pentium-V?
If we call it "Pee Five" then do we mean P5 or P-V? P-III and P-4 there's not all that much confusion...
(I also hate the way Apple has named their machines. There are, what, six seperate classes of machines all called the "PowerMac G4"? It's kinda sad to have to distinguish which model you have by checking things like "uh, do your drive bay doors look reflective, i.e. "Mirror-like"?).
What's wrong with good old model numbers?