Intel Releases "Fastest Chip Ever"

mao che minh writes "From News Factor Network: Intel has released the world's fastest chip ever. The new P4 runs at 3.06GHz, at 3 billion cycles per second. Man, and I'm still squeezing the last bit of life out of my Pentium 233!" Tom's Hardware already has a review up about it, and it looks to live up to most of the hype.

Comments

Overclock it

[anonymous coword]

Then you will have THE CHIP FASTER THAN THE FASTEST chip

Is it fast enough to get fp?

Re:Overclock it

[MrScience]

Don't you get it? They've already overclocked it. That's the only way they could get these out... and the only reason why they are so hard to find (it's so overclocked only a very tiny percent of the chips can even handle it).

Re:Overclock it

[dildatron]

while you are of course correct, one may be ble to overclock the overclocked by using super duper cooling. the limit of overclocking is limited often by heat, so if you can get rid of more heat, you might be able to squeeze a bit more out of a given chip.

for practical purpouses you are right, though. there is absolutely no reason you would buy this chip if you wanted to overclock it.

Re:Overclock it

[spruce]

So in summary, if you overclock this chip, it will either be slower, or faster, or right in line with what you would expect.

Thanks!:)

Re:Overclock it

[AaronPSU79]

Actually this chip overclocks pretty well, [H]ardOCP got it up to 3.68 GHz air cooled and 3.82 GHz water cooled. Not bad at all. http://www.hardocp.com/article.html?art=Mzg4

Re:Overclock it

[ncc74656]

Don't you get it? They've already overclocked it. That's the only way they could get these out... and the only reason why they are so hard to find (it's so overclocked only a very tiny percent of the chips can even handle it).

...and it's probably still slower than a dual Athlon MP rig. I compared a 2.8-GHz P4 rig we recently built at work to the dual 1900+ in my office (both with 512 MB of DDR and 15krpm SCSI hard drives). On a TMPGEnc MPEG-2 encoding job from a Huffyuv-compressed AVI, the dual Athlon ran 41% faster (7:56 for the Athlon vs. 13:21 for the P4). Cost for the two was about the same. Dual Xeons would be faster still, but one 2.6-GHz Xeon costs more than double what you'd pay for a pair of Athlon MP 2200s (so sez Pricewatch). Dual P4s? Forget it...Intel doesn't support it.

Re:Overclock it

[coryboehne]

Yes, but which one would it be? I was asking for speculation, not presenting speculation :)

Re:Overclock it

[iomud]

Round here we like to call it ten pounds of crap in a five pound bag.

Re:Overclock it

[uberdave]

Well, of course it would be faster. When you overclock, you speed up the entire chip, not just certain portions of it.

Of course then you'd have to deal with overhyped threads...

Re:Overclock it

[coryboehne]

Ok, we've missed each other, I mean would it be faster then what you would normally expect from overclocking, slower than what you would normally expect from overclocking, or right in line with what you would normally expect from overclocking the chip... phew!

I've never heard of overhyped threads, would you care to expound upon that a bit further?

Re:Overclock it

[deaddeng]

Sorry, but the 3.06GHz HT crushes any dual AthlonMP. So sad...

"The Hyperthreaded CPU has a big advantage over DUAL Athlons: it performs very well in single threaded applications too. In those applications the Dual Athlon configurations is hampered by the higher latency and lower bandwidth of the AMD760MP chipset and its memory subsystem."

http://www.aceshardware.com/read.jsp?id=50000328

Re:Overclock it

[uberdave]

I've never heard of overhyped threads, would you care to expound upon that a bit further?

Threads like "MPAA/RIAA, Content Protection and linking to DeCSS", "Microsoft is taking over the World", and this one.

Re:Overclock it

[jpmorgan]

Are you talking about the P4, or the AthlonXP? It's not clear.

If you're talking about the Athlon, the problem is much more fundamental than heat, it's a signal distribution problem. Basically, the chip is running so fast that the time it takes for a signal to get from one component to another is more than a clock cycle. This is why with the latest release of the AthlonXP, AMD had to add more layers and do more wiring optimisation to shrink the effective distances between components (closer = faster signal propagation, obviously).

The P4 is capable of handling much higher clock rates than the AthlonXP, since the NetBurst architecture isn't designed with the assumption that all signals will propagate within a single clock cycle. My rough calculations show that the P4 could probably be clocked up to about 30ghz before you hit the same signal propagation issues the Athlon is having now. Of course, there are more traditional overclocking concerns between 3ghz and 30ghz. :P

Re:Overclock it

[Anonymous DWord]

Man, recursive overhyped overclocked threads (hyper-ROOT-threading?). The mind b0ggles!

Re:Overclock it

[JebusIsLord]

However apparently they still overclock like mad, so my guess is that they changed the process some, since previous models hit 3ghz pretty easy but not much more.

The real question is...

[Quaoar]

How high can it be overclocked before melting and turning your machine into a firey inferno?

Re:The real question is...

[jonnythan]

With 82 watts of heat produced, I don't think you need to overclock it at all for that to happen.

Re:The real question is...

[the way, what're you]

With 82 watts of heat produced, ...

Pop quiz time! Fill in the blank.

Water boils at _____ watts.

Re:The real question is...

[rodgerd]

Actually, at peak HT, it produces over 100.

Re:The real question is...

[evilviper]

Maybe they were jelous of being behind AMD in the "Space Heater" performance ratings...

Re:The real question is...

[Jonny 290]

I'd mod you -1 Redund if i could. It's just that it took him about ten words to say what you just did in 100. Hehehe.

fast chip?

[AmigaAvenger]

Maybe I'm missing the point, but isn't every new chip the manufcatures release the 'fastest chip ever'

I remember when the Pentium 200 was the fastest chip ever!

Re:fast chip?

[moertle]

Ah, but when the Pentium 200 was out DEC had a 500 MHz Alpha.

Re:fast chip?

[n1ywb]

No, Intel is also releasing new ARMs and they're not the fastest chips ever. Neither is the Transmeta Crusoe.

Re:fast chip?

[aridhol]

Yeah, maybe they should rename it to "The Fastest Chip Ever or Until Someone Else Makes Something Faster", but it just doesn't roll of the tongue the same way.

Re:fast chip?

[Mr. Sketch]

You must be new... I can recall the PCWorld issue that reviewed the brand new 286/12 processor (yes that's a screaming 12Mhz, with turbo turned on of course) and the reviewer sayed "It ran so fast it left skid marks on my desktop!". How's that for fast?

Re:fast chip?

[BTWR]

Not necessarily...

Intel Celerons, for example, are contually made and are always behind the Pentium line of processors.

Although technically you could always say "Intel has just announced the fastest Celeron ever!"

Re:fast chip?

[Sloppy]

Maybe I'm missing the point, but isn't every new chip the manufcatures release the 'fastest chip ever'

Yeah, but this one is even faster!

Re:fast chip?

[JonWan]

"It ran so fast it left skid marks on my desktop!". How's that for fast?

Well I guess this one leaves skid marks in your pants.

Re:fast chip?

[bjtuna]

Up till a few years ago, most cases that came with turbo buttons did just that. I believe the turbo switch was originally shorting a pin on the mobo that sped something up, but a side effect was that it would change your LED so that it would report the new clock speed.

I'm not sure when chipsets stopped supporting whatever it was that "turbo" did, but case manufacturers continued to include the turbo switch. I remember when I built my P-166 tower in a case I bought from Dalco a few years back... I set the LED so it said '166', but when you toggled the Turbo switch, the first digit disappeared and it just said 66. Heh.

Re:fast chip?

[uberdave]

I keep these straight by referring to them as Delux and Sux.

Re:fast chip?

[Jucius Maximus]

"don't forget. the SX came out AFTER the DX."

But did you want a 486 SX/25 or a 386 DX/33 ?

I can't remember what my decision was.

Re:fast chip?

[Des Herriott]

I'm looking forward to the release of 10GHz chips. 'Cos then we can overclock them and say, "But this one goes up to 11".

Smokin!

[Blimey85]

I really wish that AMD could keep up with Intel. I have two dual processor machines and they run at 3.2G's but I don't think they would fare all that well against a single processor machine running at 3.06G since not everything can use both and it doesn't seem like all that much uses both effectively.

It's too bad that Intel charges so much for their chips.. and this thing being the hottest thing at the moment... or when it's actually in stores.. going to be a while before I can get one. Damn.

Re:Smokin!

[sketerpot]

Your apache server ought to run pretty well on an SMP system. If you want to be sure to get plenty of performance, I would recommend using apache 2.x, which uses threads and processes, AFAIK. These can be split among your two processors by the kernel. Your email software is probably also multiprocess (or multithreaded, which under current linux versions is mostly the same thing), so it can also take advantage of SMP.

Your text editor probably won't get any boost from another processor, but if you're setting up a server then there's nothing to worry about.

Re:Smokin!

[spike hay]

I have two dual processor machines and they run at 3.2G's but I don't think they would fare all that well against a single processor machine running at 3.06G since not everything can use both and it doesn't seem like all that much uses both effectively.

Actually, the hyperthreading only helps in apps that support hyperthreading. Your dual processors are hyperthreaded. So any hyperthreading app that takes advantage of the P4 will also take advantage of your dual processor setup.

I imagine two different processors would be much better than 1 hyperthreaded processor.

Also, they only mention a 25% performance increase. Dual processors running hyperthreaded apps have at least a 60% performance increase. However, I bet this P4 would beat your machine in non-hyperthreaded apps.

Re:Smokin!

[spike hay]

This P4 is has hyperthreading. It is like 2 virtual processors, at least for apps that support hyperthreading.

Dual processors also take advantage of apps that have hyperthreading. The P4 has a 25% increase in performance in hyperthreaded apps, while a dual athlon gets at least a 60% increase, usually much more than that.

So, if Apache supports huyperthreading (I'm sure it does) you would be much better off getting dual athlon XPs rather than this P4. It be much faster and cost the same or less. (BTW, get XPs instead of MPs. They are really the same thing)

Re:Smokin!

[dnoyeb]

You contradict yourself. You say you wish AMD could keep up with intel, then you mention that not many apps use both CPUs (and thus hyperthreading) effectively.

I think AMD realizes that multiprocessing is not something the average user will ever benefit from. But they are falling behind in the marketing department on this one.

Re:Smokin!

[Blimey85]

How did I contradict myself? I said that I wished AMD could keep up with Intel.. translation: I wish AMD was shipping a 3Ghz chip. Then I say that not many apps use dual cpu's effectively. The two comments had nothing to do with each other. Dual processors are great for some things. Having a pair of 3Ghz would be better than what I have now for sure. If AMD shipped a 3Ghz processor, my next machine would most likely be a single processor machine rather than a dually.

Re:Smokin!

[timeOday]

Hyperthreading isn't AMD's problem. The problem is AMD has fallen behind in performance and are losing tons of money which they don't have.

Re:Smokin!

[Henry V .009]

Applications do not have hyperthreading. That is a marketing term.

Applications can be multi-threaded however. The operating system will take care sending the threads where they need to go. Hyperthreading merely makes a single CPU look like two CPUs to the OS. And since two different programs run in two different threads (your OS is multi-threaded!) you can even see speed advantages when you are using more than one single-threaded program at once.

And your percentages are bull. It all varies widely from application to application. You will therefore have to see application specific benchmarks for whatever you plan to run.

Re:Smokin!

[jpmorgan]

I think AMD realizes that multiprocessing is not something the average user will ever benefit from.

If you think that, you've obviously never used a dual processor box. A multiprocessor against a uniprocessor (but with the CPU twice as fast) may not outperform the uniprocessor at Quake3, but it'll feel far more responsive and quicker in most things you do. It's just the way modern operating systems work.

Re:Smokin!

[evilviper]

AMD is doing well on the home users' desktops. Where Intel has a hold is on high-end workstation, and servers.

That's where MP is important, and where AMD falls behind.

Re:Smokin!

[Grishnakh]

It's too bad that Intel charges so much for their chips..

No, it's too bad people are stupid enough to pay what Intel charges. Intel's actually pretty smart here; they charge what the market will bear, and seem to be doing well in spite of charging way more than AMD for processors.

My last processor was an AMD though because I wanted something very inexpensive that performed well, and AMD could deliver that, while Intel's comparable offering would have cost more than twice as much.

What's too bad is that more people don't shop this way, so that 2000 AMD employees wouldn't have to be laid off this week.

Re:Smokin!

[dnoyeb]

That is simply incorrect. Not only do I have a multiprocessor system but I have a masters in that area. so what you take for obvious is false.

"far more responsive"? Why if I may ask? There is no reason why a multiprocessor system should feel "far more responsive" than a significantly faster uniprocessor system.

Should I buy it?

[Anonymous Coward]

Will my Internet work faster if I get it?

Re:Should I buy it?

[BTWR]

As Homer Simpsons once said...

"Ooh... they have the internet on computers now!"

Also on AnandTech

[johnkp]

Intel's Pentium 4 3.06GHz: Hyper-Threading on Desktops

Personal PC's

[Uhh_Duh]

I don't understand the need to always be on the bleeding edge of technology. Intel loves to push these newer faster chips down the throats of consumers, but I've got 600MHz Intel chip and a 2ghz intel chip, both running Windows 2000, and I swear I can't tell any difference between 600MHz and 2ghz for normal usage -- and I consider myself a power user.(Granted, I don't do 3D rendering or massive number crunching on a daily basis, but how many of your average consumers do?)

I won't be running out to buy this any time soon -- especially when I can the $200 Walmart computer is less than the cost of this CPU.

Call me old fashioned, but geeze.. Intel already gets plenty of money from my pocketbook for little performance gain. Something needs to be done about the rest of PC hardware before the speed of the CPU is going to make a massive difference.

Re:Personal PC's

[kin_korn_karn]

The need is driven by games. I'm a gamer, so I have to have all of this bleeding edge hardware.

However, I regularly tell non-gamers that they shouldn't upgrade unless their PC doesn't do what they want it to do. The push for faster-better-stronger hardware is out of hand, the average consumer doesn't need any more than a 600mhz.. but they do need lots of RAM and a big hard drive.

Re:Personal PC's

[Wild Bill Hickock]

I agree to your points but i would like to say something. A faster CPU it would be good for multimedia usage (something that the average person would care about). For example ripping DVDs or CDs, or as you mentioned rendering. You can't tell the difference between your 600 MHz and 2 GHz PCs because of several reasons. One that comes to mind is the hard drive bottleneck The IDE drives today are just too slow for the current systems. I believe that in order to build a PC that takes full advantage of the processor power and memory bandwidth it would be expensive. but as you know The average joe hears 3.06 GHz and thinks that is faster than his current PC. He doesn't think about other limitations of the system as a whole. Which if you think about it, boils down to marketing.

Re:Personal PC's

[StarHeart]

While I agree with your general point I can definitely tell the difference between my father's dual P3 650 with 1gb of memory and my Athlon XP 1700+ with 768mb. The faster being my Athlon XP. It isn't a huge difference, but a general snappiness feel.

Re:Personal PC's

[cardshark2001]

I don't understand the need to always be on the bleeding edge of technology

You obviously have not played the leaked doom demo.

Re:Personal PC's

[Sloppy]

..I consider myself a power user.(Granted, I don't do 3D rendering or massive number crunching on a daily basis, but how many of your average consumers do?)

Just what the heck is a "power user" if it's not someone who overworks their computer?

And a different take: how can a "power user" and a "average consumer" be the same thing?

Is "power user" just a completely content-free bland empty label of generic gray nothingness?

Re:Personal PC's

[_ph1ux_]

Exactly. The thing is that the testing that is done on Intels latest chips is done in a subjective manner for marketing purposes. They test the chips on machines that are not Identicle - with software that is optimized for that chips instruction set - then push it out to people saying "Look at Intel's latest chip! The new $Pentium-X will runs your applications so much faster!"

but there is a threshold we will hit - on the consumer level - and that day (although still a bit away) is coming faster and faster with every release. It is the subjective speed threshold, where the Human is the bottleneck. Where the computer can do anything the user can so fast - that the computer is then waiting on input from the user.

All input from a human comes in little spurts - and therefore will be processed by the CPU before the next batch comes in.

The point is that there is a somewhat finite desktop market incentive for faster processors, in that, for the average user - there will be a time, sooner than later, where they find that the machine they have is fast enough, featureful enough and big enough (storgage) to meet their (rather long term) needs.

Re:Personal PC's

[miffo.swe]

"Something needs to be done about the rest of PC hardware before the speed of the CPU is going to make a massive difference."

Your ohh so right on that one. If we do get some new killer app that eats cycles like crazy it won help that much. The CPU will regardless run circles around memory, HD and the bus. Thats why you cant tell the difference of a 1,6 Ghz and a 2,Ghz without a benchmarking program. All that cpu power is sitting and waiting to be fed instructions that doesnt come.

A faster replacement for the HD is long overdue but then again my last PC had a floppy!

Re:Personal PC's

[theLOUDroom]

You obviously don't play games, compile things, encode media, etc.
Yeah, obviously, if you only do wordprocessing and web browsing you're not going to notice the difference. Those are not the only uses for a computer.
I personally, max out my athlon's cpu load all the time, playing games, compiling things, using the autospellchecker in gaim :), etc.

Re:Personal PC's

[bogie]

"Something needs to be done about the rest of PC hardware before the speed of the CPU is going to make a massive difference."

Are you being serious? I never understand why this comment gets modded up every few months when the newest cpu come out.

Did you even read the article? Cpu speed makes a massive difference on many applications. Compared to say a XP1800 the baseline cpu for the article, the newest 3.0GHz cpu can cut your worktime in half on some apps.

Your 600 MHz cpu may be fine for word processing and email, but its bottom line of useful for say Windows XP as it ships by default or the newest version of Redhat. There is a difference whether or not an app or OS will run on a cpu and whether it runs quickly. It certainly isn't fast enough for the latest games.

".(Granted, I don't do 3D rendering or massive number crunching on a daily basis, but how many of your average consumers do?)"

I'd love to see you try to play UT2003, Mafia, Morrowind, or NOLF above 640x480 16bit on that 600MHz cpu. Beyond playing games which is huge industry, "average consumers" are doing things like editing their digital photos and movies.

Until apps and games launch and run instantly and perfectly there will never be a fast enough CPU or Video card Period.

Re:Personal PC's

[jpt.d]

with microsoft trying to push people to upgrade all the time with ever more advanced methods of crashing your computer (known as windows, forgive me, I just installed win98 on a computer for a friend (i am a mac user now) and it took 30 to 50 reboots before I was done with it).

Microsoft will not be able to sell these new things to consumers. However at the same time you will not be able to sell certain other things to consumers either. For example, SerialATA is coming out now. If we take it a few years from now when consumers have met this point, SerialATA will never come very popular because nobody (not very many - new being a niche market) is upgrading to the very new anymore. Or same with FireWire 2 (my laptop for instance, how would you put that on (its an ibook... no pcmcia)). Interfaces don't exactly make the things faster. So harddrives haven't really increased in speed in the last 5 years (usually they are running at 5 to 10 mb/s for most things or even lower). RAM has increased, but bus speeds haven't really increased that much.

So yes, the underlying things that the cpus rely on must increase a lot before people will use them. Maybe it will be too late for them, if people go portable then they probably won't care too much. Look at apple's ibooks, they are cheaper than their desktops and they are very competitive to other offerings. Apple is the way to go now guys...

Re:Personal PC's

[WeaponOfChoice]

completely agree, have an old 4 way xeon (450's with 1Mb) and it smokes my 2Ghz on anything remotely complicated...

Re:Personal PC's

[ActiveSX]

I just installed win98 on a computer for a friend (i am a mac user now)

I would be too after an experience like that.

Re:Personal PC's

[scot4875]

I'm a gamer, too, but I'm not a mark. I play plenty of games released this year (UT2003, WC3, for example) on a 1 Ghz P3 w/ Radeon 8500 with absolutely *no* problems.

By not staying on the bleeding edge of hardware, I have extra money to buy more games. I don't buy hardware that will be able to play a game that may/may not come out sometime in the next year, I buy based on what's available *now*. There's no f'in reason to have a 3 Ghz CPU for any game currently on the market.

I'd say that my current PC (minus the monitor, which cost $300, because I wanted a nice monitor) cost a total of maybe $500 to build. That's LESS than the price of this CPU.

Go ahead and buy it if you want, but it really won't make your dick any bigger.

--Jeremy

Re:Personal PC's

[be-fan]

..sigh... Every time one of these articles come out. First, if you can't tell the difference between 2GHz and 600MHz, you're dead. My 2GHz machine is nowhere near fast enough, even just running Konqueror and KMail. Second, more people need the power than you'd think. I write C++ code with some very heavily templated libraries. G++ eats my processor for lunch (I've got enough RAM that it's not HD-bound). Add to that 3D rendering (messing around with Blender for some 3D work) and numerical computation (simulations, Octave, Mathematica) and I probably won't ever have enough CPU. And I don't even do gaming!

Re:Personal PC's

[be-fan]

Oh, and people who actually do, say work on their computers, need the horsepower. When you add up all the people that do architecture, engineering, computational science, video/audio editing, desktop publishing, etc, that's a very significant percentage of the computer using population. Certainly enough to sustain some profit-margins for high-end chips like this. And that's without the gamers...

Re:Personal PC's

[grammar fascist]

Woo hoo! I might just get some "Informative" karma over this one... :)

Here's the deal: the DOOM III demo was a debug build. If you've got it, do a "strings" on it. You'll see a bunch of debug symbols.

That means no optimizations, and tons (I mean tons) of code to make tracking down problems like memory leaks easier. That kind of build will naturally munch processor cycles like crazy.

Corroborating evidence: the alpha is very CPU-bound, which should be surprising given how the algorithms it uses for rendering eat GPUs for lunch.

All the same, with features like per-poly collision detection, I expect the final version to do much better on a 3.06GHz chip than a 1.2GHz chip.

</off-topic>

Re:Personal PC's

[freeweed]

I'd agree 100% with you, except I've been around perhaps a bit longer:

I can remember well when the first gen Pentiums came out, and the exact same argument was used: 'the average consumer has no need for anything more than a mid-grade 486'. Except for high-end rendering packages and games. Putting that spiffy new P75 in your machine added hundreds to the cost over a 486-66, with no noticable speed increase, unless you were a power user.

And here we are in 2002, arguing that no one needs more than 600Mhz. Know what? You're entirely correct - for now. 7 years from now, your statement will look pretty silly, just as my claiming a 486 can handle your every need would look silly today.

Intel doesn't introduce new chips for mass market consumption the very next day; they know full well that only the diehards buy that early, and the chips are priced accordingly - ie: their business model runs around this very assumption. The idea is to hit the early-adopters first, and in time the demand will rise.

Of course, maybe I'm just a bit too cynical. I've seen this issue on an almost monthly basis ever since I was told '64k? who needs that much RAM? my games run just fine on my Vic20'.

Re:Personal PC's

[PunchMonkey]

The need is driven by games. I'm a gamer, so I have to have all of this bleeding edge hardware.

hehehehehe....

"have to have"

hehe....

Re:Personal PC's

[drinkypoo]

It's also pushed by content creation, which is to say, media encoding, and shit like photoshop and after effects filters. These are operations which usually parallelize well (if by no other method than handing each portion of an image or frame to a different thread) but which nonetheless consume an inane amount of CPU time. It takes about six to eight hours to encode (from an AVI source) a SVCD (MPEG2 at about half DVD resolution) on my 1.4GHz athlon. That's really damned annoying. Those people who are encoding or transcoding DVDs are going to need at least four times the processing power to get that kind of speed doing a DVD. Of course I also only have PC133 SDRAM so that doesn't help me any, but MPEG encoding is widely held to be mostly CPU bound in almost every situation.

If all you do is email and word processing, maybe run a spreadsheet or quicken or something, you can quite reasonably use a Pentium 2 or something similar. You can get machines like that for about $140 (surplus) plus add 512mb or so of ram to them for about $50. Tons of ram will be more helpful to the average user (who has lots of programs open but they aren't doing much in the background) than anything else, including the latest greatest CPU; as long as they have a 686-class processor they should be fine.

Meanwhile, those people doing video editing actually have a greater need for processing power than the average gamer, which is why they have dual xeons and shit like that. We (Gamers) don't, which is why a single P4 or Athlon XP is more than enough to run any game on the market today. My 1.4 gig tbird, 512MB PC133, and GF3Ti200 is enough to play any game available today at a reasonable res and frame rate. Sure, I'd like more, but I'm not getting fragged because my computer isn't fast enough. I'm getting fragged because I suck.

Re:Personal PC's

[xenocide2]

I guess what they say is true, "software expands to the capacity of the hardware."

Re:Personal PC's

[drinkypoo]

My 2GHz machine is nowhere near fast enough, even just running Konqueror and KMail.

You must have done something very, very wrong. I have a 1.4ghz athlon tbird system with 512mb PC133 and I run Windows XP, and I use Mozilla for mail. It is quite peppy. Oh, windows don't draw instantly, but mail (including that with HTML) is displayed very promptly. Plus I am at all times running Kazaa, an X server with a couple xterms in it, and assorted other doodads, sometimes a MPEG transcode or similar, running at idle.

Your other points (rendering and similar) are valid but if you can't read and respond to very large, complex email on a machine of, say, 1GHz, you are doing something wrong.

Re:Personal PC's

[be-fan]

It's not just code bloat. Perhaps it's more along the lines of "what user's want to do with their hardware is more than the hardware allows."

Re:Personal PC's

[be-fan]

I never said I can't read and respond to them. I said that it's not fast enough. Fast enough, for me, is when everything is instant. When I can resize windows like crazy and never, ever, detect a hint of redraw. In real life, if I shove my calculator across my desk, it doesn't visibly redraw. Why should my windows be any different? I mean, Konqueror and KMail are usable for me, don't get me wrong. I'd even say it's pretty peppy. But is it fast enough? No.

Re:Personal PC's

[SomeOtherGuy]

My 2GHz machine is nowhere near fast enough, even just running Konqueror and KMail.

Me and my trusty 300Mhz Celeron don't feel sorry for you.

P.S. -- If I won't be able to browse and read email with a 2Ghz -- then that is the last straw, I am going back to my C64.

Re:Personal PC's

[QuietRiot]

SerialATA will never come very popular ...

Can I quote you on that?

Re:Personal PC's

[evilviper]

Unfortunately, that $200 PC has a 5400RPM hard drive, and the processor performs similar to a ~400MHz chip, despite what they claim.

That's fine for those that are looking for a low-end machine, but for $100 more you can get a mochine with higher specs, and a processor that is realistically 3Xs faster.

Re:Personal PC's

[jpt.d]

Go ahead :-)

I also predicted that Microsoft would name a version of windows after the Millennium (at the time everything and their dog was the millennium something) and guess what they did... rebranded Windows 98 as Windows Me[llennium]. Of course its easy to say that after the fact, but you don't have to trust me until you have installed Palladium (it will be a cold day in hell...)

Re:Personal PC's

[cant_get_a_good_nick]

I think the problem right now is software not hardware. When the PowerPC was replacing the Motorola 68K, people were saying that it had the potential for making a quantum leap in the way people worked (and played) with their computer. Stuff that was impossible because it would be slow. Instead of small quantitative changes (wow! this menu takes 1.3 sconds to show instead of 1.6 seconds) there now can be qualitative changes. New interface paradigms. Theming, ha, thats wasted cycles, we're going to change the WIMP metaphor. We'd have audio interaction with the computer, whats a menu? As Much as I like bashing Microsoft, Microsoft Bob and the Bastard Child From Hell Clippy are some of the few variations from the WIMP metaphor that we've seen. The main reason they failed was that Microsoft didn't have them get out of the user's way. But another reason why is not enough processing power. Not only did you get interrupted, but for stupid advice. Wasn't worth the trade off.

When System 7 came out [my opinion is Apple did a great job of getting System 7 out and working on the PowerPCs at the same time] they announced their roadmap for the future. There were three OS teams, blue, pink, and red. Blue actually was the System 7 team, if you open the System & kernel, you can find strings about the Blue Meanies, the System 7 team. Pink was the near term really cool stuff, a new OS that they were supposed to do with IBM, and formed a company called Taligent. Was supposed to be a brand new OS, Object Oriented design. Based on the Document, not the Application. A large part of this was called OpenDoc. They shipped parts of it, but never a real whole. The best part they actually shipped was CyberDog, but gradually all of this went away. Some say the reason it failed is lack of computing power. There were other reasons - new programming paradigm, hard to get developers on board. Instability in one part can lead to bad interactions in another part as well. But it took up a lot of RAM and a lot of horsepower, too much for a 60Mhz PPC601. I'd love to see what they had thought for for Red, that was their super long term whiz-bang stuff.

So where are the new interfaces? Instead of wondering why folks are pushing faster boxes, wonder why we're still using the WIMP metaphor over 23 years after the introduction of the Xerox Star? The base comp had 384KB RAM, smaller than L1 cache on some chips. But anyone who was familiar with a Xerox Star and plop them in front of a Windows box, and they'd probably recgnize it all.

Then again, you don't have to buy a 3GHz Pentim 4 if you want. If it doesn't fit your fancy, then don't Luckily enough we still have options, Athlon, and to a lesser extent, VIA C3s are alternatives.

Re:Personal PC's

[evilviper]

It might slow down, but never stop. When the processors are so fast that regular apps run fast, they will come out with holograph displays, and a 3D printer to go with it. Of course there is no limit to the number of things that could go on in the background. Soon Office will be checking your historical actions to decide if autocorrect should be used in this instance.

Above all... to compensate for lack of bandwidth, compression schemes are requring more and more power all thi time.

I don't think it's ever going to get to be an appliance, sales-wise.

Re:Personal PC's

[dasunt]

be-fan writes:
First, if you can't tell the difference between 2GHz and 600MHz, you're dead.

Is it the CPU that makes the newer machine seem faster, or the additional memory and faster hard drive?

For loading applications/OS's, first the program has to be read from hard disk, and then (to keep it from swapping), it should remain in memory. CPU might make a tiny difference, but outside of rendering and a few other things (like some games), you shouldn't see a noticeable difference. (And for gamers, video cards make a big difference as well.)

Then again, it doesn't matter how fast of a CPU I buy, I'll find something that will push it to 100% usage. Currently, that's divx5 encoding. But I'm not the typical user.

Re:Personal PC's

[be-fan]

Well in my case, I always make sure to get huge amounts of RAM, so hard-drive is usually not an issue. For example, moving from my 1.5 GHz Athlon to my 2GHz P4 actually had a noticible benifet for me. Many KDE apps seem to be CPU bound. Particularly, Konqueror pegs the CPU while reformatting HTML pages, and all KDE apps start out by processing 60,000 relocations. The faster CPU reduced the redraw problems, and apps started up faster. A huge difference? Of course not, but enough to notice it.

PS> Now, upgrading for 500 more megahertz (essentially similar end-level performance, though) was not pure wantonness. I moved from a desktop to a laptop.

Re:Personal PC's

[Junks Jerzey]

The need is driven by games. I'm a gamer, so I have to have all of this bleeding edge hardware.

That was true back in 1999, but no longer. The sweet spot was around an 866MHz Pentium III and a GeForce 2. There's just about nothing that needs more power than that, even recently released games. Sure, you could come up with a counter example, but the law of diminishing returns has kicked-in in a severe way.

Re:Personal PC's

[Junks Jerzey]

Doom 3

Wow, so you've actually run an official, finished, optimized version of Doom 3? Impressive, considering that id hasn't even gotten that far yet.

Re:Personal PC's

[Junks Jerzey]

If you think the final thing will run acceptably on a sub ghz P3.. then wtf are you smoking..

Doom 3 is a leap in graphics, not, by all accounts, gameplay. To do all the fancy vertex shader stuff, you'll need a GF3 or better. And textures are not an issue, because you can cram a lot of 8:1 compressed textures into 128MB (that's effectively 1GB of video memory memory--256 times what that PlayStation 2 has). How much work does the CPU need to do? Remember, all of the skinning and multiple passes and all that is happening on the video card. So, sure, there's a lot to do on the CPU side, but maybe much less than you are expecting. The AI in Quake 3 was barely a leap above the AI in DOOM, for example.

Bottom line: It is possible that DOOM 3 will be just fine on an 866MHz P3.

Re:Personal PC's

[Toraz Chryx]

You appear to be forgetting that Doom 3 also has rather intensive physics to handle, I'm not expecting the AI to eat processor power, I _AM_ expecting the physics to do so however.

Wow!

[darkov]

And if they had a 40-stage pipeline they could go to 6GHz! Then I'd be really impressed.

Not just the Mhz ramp: hyperthreading/SMT

[mccalli]

This chip is more interesting than just the normal megahertz hike. It's the first of the desktop hyperhreaded chips - previously only available in the Xeon range (well, from Intel anyway. Other manufacturers had them).

This is something I'm interested in. I currently run a dual-CPU box of two 533Mhz Celerons on a BP6 board. I've wanted my next machine to be a dual-CPU has well, but now I'm not certain. Perhaps the hyperthreading will take care of that for me? Who knows, it's too early to say as yet. But I'll be keeping an eye out on the benchmarks for this chip, whereas I've more or less ignored the Mhz races for the last couple of years.

Cheers,
Ian

Re:Not just the Mhz ramp: hyperthreading/SMT

[_ph1ux_]

Does slashdot really run faster on dual CPUs?

Re:Not just the Mhz ramp: hyperthreading/SMT

[mike3411]

I'd say that 2 fast Athlon MP's (or XP's that take about 3 seconds to mod for SMP) are still going to be significantly faster for anything that can take advantage of multithreading. I don't know what tasks you have in mind, but with the price/performance ratio of dual Athlons, I can't imagine that Intel uniprocessor systems will be able to compete for some time. Of course, benchmarks wil really tell, I'd suggest 2cpu (http://www.2cpu.com) for more info, the forums are also filled with people who possess a great deal more knowledge than I do about system performance, esp. concerning multiple CPUs.

Re:Not just the Mhz ramp: hyperthreading/SMT

[dnoyeb]

I have duals just because...But I recognize their is no benefit whatsoever for the average user.

Better to get a faster single when one is available. Dual is for when the fastest single is not fast enough...The bp6 was an anamoly. I don't think that will happen again.

personally I am not even seeking to get dual XPs...

Re:Not just the Mhz ramp: hyperthreading/SMT

[cDarwin]

That depends on whether you view articles nested or threaded.

Re:Not just the Mhz ramp: hyperthreading/SMT

[mccalli]

Does slashdot really run faster on dual CPUs?

No. But it runs faster on multiple CPUs when you're also running a dozen other processes, including Virtual PC, which itself is running another six or seven processes.

Cheers,
Ian

82 watts!

[paradesign]

thats insane. Thats equal to what, two or three G4s?

Re:82 watts!

[fobbman]

Hey! New Intel marketing hype!

Re:82 watts!

[Anonymous Coward]

Intel should run an ad where a family sits around the (Apple) computer in coats and earmuffs, shaking from the cold.. everything blue, with icicles (sp).

Then switch to a shot of a family in a cozy room, all basking around the glow of a warm intel machine..

Intel .. Keeps You Warm! (tm)

Hey, those marketers can sell anything, right?

Re:82 watts!

[Rolo Tomasi]

Yeah, but you need four G4s to get the same performance, so on a performance/W basis, the P4 wins. And of course, there are only so many applications that will benefit from multiple processors.

Besides, 80W isn't that bad. Both the Itanium and the Power4 burn away around 150W IIRC. And the latest Alphas have passed the 100W mark as well.

Re:82 watts!

[WatertonMan]

But it also faster than two or three G4's. This is great for the Windows/Linux world but kind of sad for competition which may led to problems down the road. After all I think both AMDs offering and the IBM 970 will have trouble competing it Intel continues in their path.

I love OSX. But the hardware really is falling behind. . .

Re:82 watts!

[WatertonMan]

That's not entirely true. The G4 outperforms the speed difference in some functions. However not as much as many Mac zealots claim. Programs that utilize Altivect really have a dramatic speed increase. Unfortunately one of the programs that benefit best from Altivect has a bug that makes it unstable with that code. Adobe's fix? Disable it, making Photoshop considerably slower.

Re:82 watts!

[Rolo Tomasi]

Exactly. If you need low-power CPUs, there is VIAs C3 (can be passively cooled up to 800 MHz IIRC) or the Transmeta Crusoe (no idea if there are standard ATX boards for the Crusoe though). I don't know if Intel still makes Tualatin Celerons and PIIIs, but those are quite fast also and burn around 30 Watts, and they are about 20 to 30 percent faster on a clock-per-clock basis than the PIV. It's not like anyone is forcing you to buy anything.

They have even faster ones already.

[IdleTime]

According to the Computer Power User magazine, Intel demonstrated a P4 4.1 GHz at the Intel Developer Forum. They even showed it overclocked to 4.65GHz with extensive cooling.

I expect it will still take a year or two before they become generally available.

don't you mean...

[bovril]

...a midi-towering inferno?

100 watts....

[Steveftoth]

if you read the article they say that it can use up to 100 watts of power when you are using the chip to the utmost. That's a lot of power. Much more then the AMDs. Anyone else think that's a little extreme? I mean I'm all for more speed, but cost aside, this seems to be a huge factor in actually getting one of these systems. You also have to get a new motherboard.

For server applications it's not as useful because you can't build dense systems. Since server applications are by their very nature more multithreaded then workstation, I would imageine that they would get much hotter. You'd need a lot more cooling. Also, don't the chips SLOW DOWN automatically when they get too hot, thus negating any increase in speed you might get from them.

Notice that the new heat sink is larger as well.

Not trying to bash it, but it seems like the older chips are still going to be better until they get this whole heat issue under control. I run my system almost 24x7 like I'm sure many people on /. do so I think that running a system all the time (with SETI or whatnot) would be expensive.

Re:100 watts....

[ektor]

Every one of the recent processors from both Intel and AMD are very much power hungry. While the P4 3.06 pushes 80 watts the top-of-the-line AMD is not far behind.

See this article from Tom's Hardware.

Sadly this trend won't go away anytime soon. When you pack that many transistors running at ultrahigh frenquencies in a tiny package you have to pay somehow.

Re:100 watts....

[EvilTwinSkippy]

I really miss the days when you could set up a box with no moving parts. I have 2 applications for computer where a cooling fan is inneffective, or where a cooling fan in completely undesired.

I run the network at a science museum. We have kiosks (Linux of course) that sit and run all day, every day, unattended. Every so often a power fan starts making noise and vistors complain, or it quietly quits and the processor french fries either itself or surrounding electronics.

My second appliction is running a server for volunteer checkin at a folk festival. We set it up at the begining of the week in a dusty, damp, humid shed that serves as the office. It has to run, hot or cold, dry or damp, all weekend. A fan sucks (literally and figuratively) because it draw in dust when the weather is dry, and spins to almost no avail if the weather is too humid. I presently use a clocked down K6 that doesn't need a fan.

I realize I am starting to wander into the realms of embedded devices, is it so much to ask that my next computer be quieter than the diswasher?

Re:100 watts....

[fermion]

In further news, Intel states the chip also targets the excess energy stores of the earth. A high official states "No one really needs this much speed, and we could have designed a more efficient chip, but we want to be in sync with the current world economy. We hope that as we approach 100 watts of generated heat, people will appreciate the effort we put into making a chip that can truly utilize the excess energy stores we have in the U.S. We absolutely feel that the synergies between the various excesses will combine to spark phenomenal sales."

This official describes the marketing for this chip as similar to the automaker's model for an SUV. The chip is overly complicated for the job, bulky, has few application, yet consumes large amounts of energy and space. In short, Intel will market this as a status symbol among geeks. Sublimate your fear of your inadequacies into the fact that you can afford the TCO for this chip. Sublimate your lack of dates into the knowledge that no one will be as fast as you are. To compensate for the fact that the chip hides in computer, a hologram poster will be supplied, complete with a sufficiently clad female, that the child may hang on the bedroom door or locker without parental or school complaint.

Re:100 watts....

[ctr2sprt]

Also, don't the chips SLOW DOWN automatically when they get too hot, thus negating any increase in speed you might get from them.

Yes, they do, but "too hot" is about twice the normal operating temperature of the CPU, and "slow down" is "to nearly a halt." I seem to remember someone donating a CPU to test this, and I believe the hysteresis was 80-85C. He got to those temperatures, by the way, by unplugging his CPU fan and running SETI@home.

I don't think this part, at least, is something people need to worry about. I don't know if it ever was, but I heard the stories about early P4s too.

Re:100 watts....

[be-fan]

Wow. 100 watts. One whole lightbulb. Big whoop. The Power4 sucks on the better part of a kilowatt. Now that's rather significant.

Re:100 watts....

[Steveftoth]

http://zdnet.com.com/2100-1103-892836.html
This article says it's only 125Watts, so just a little bit more. What can I say, IBM is always one step ahead of Intel.

But seriously, Intel wants normal people to buy these things. And this is just the processor, not to mention the HD, CDROM, motherboard, video card etc... Adds up to a lot of power. Not many people are going to be using the Power4 that uses 125 watts, Apple is using the lower power version AFAIK.

Re:100 watts....

[be-fan]

125 watts per die, 4 die per module, so north of 500 watts (thanks to the 128MB of L3 cache) per module.

Re:100 watts....

[SailorBob]

If you're really interested in exact numbers, the cpu spec page at chipgeek shows the Thoroughbred-B core 2800+ putting out 74.3 watts and the Updated Northwood core 2800 putting out 68.4. It's hard to imagine that going to 3000 would bring the number up to 100 watts. I wonder how they got the 100 watt number? I mean, what did they use to measure it?

Too bad

[tcd004]

It runs so hot that it has to be cooled by the blood of a virgin every hour.

tcd004

Re:Too bad

[fobbman]

They're catering to the geek market. That's a large supply of virgins right there.

Congratulations, it's a CPU, and an oven

[ka9dgx]

So, we have a CPU with an internal clock faster than the 2.5 Ghz in my Microwave oven. Does it come with a carousel to keep the heating even?

--Mike--

Re:Congratulations, it's a CPU, and an oven

[neafevoc]

Does it come with a carousel to keep the heating even?

I normally use my AOL coasters for that.

Re:Congratulations, it's a CPU, and an oven

[dildatron]

You are confusing the length of a radio frequency wave with the frequency with which somthing processes.

however, since it's a joke, I'll allow it.

Re:Congratulations, it's a CPU, and an oven

[SpinyNorman]

Cool! We'll be able to heat up pop-tarts in the CD drive!

Re:Congratulations, it's a CPU, and an oven

[Cyno01]

Me: My cpu is a P4 1.8GHz. Friend: Well, my cordless fone is 2.4GHz, so nuh. :P

Re:Congratulations, it's a CPU, and an oven

[caveat]

well yeah, but wouldn't an RF field oscillating at 3.06GHz emit microwave radiaton? i know at the lab, a lot of our pcs run without cases, and we have to keep them away from instruments that runa t approximately the same frequencies...

Re:Congratulations, it's a CPU, and an oven

[ottffssent]

The length of an RF wave is its period, not its frequency. Period is 1/frequency. A 3GHz processor has a clock signal oscillating at 3GHz, just like a 3GHz RF wave, or anything else vibrating at that frequency.

It is you who are confused about a great many things.

Other Important Questions

[ewhac]

Does this new chip have support for Digital Restrictions Mechanisms? Does it still have the universally reviled serial number feature? Can it still be shut off?

Schwab

What happened??

[Ec|ipse]

So with a 3gig cpu running with 1gig memory and a 100gig of harddrive space. Is this something we can expect?

User 1 "Did my computer just crash?"
User 2 "Couldn't tell, happened to fast."

Re:What happened??

[Subcarrier]

User 1 "Did my computer just crash?"
User 2 "Couldn't tell, happened to fast."

Light travels less than 10 centimeters in one clock cycle. Of course you didn't see it crash.

Re:What happened??

[_ph1ux_]

The term Mega- is so last month!

Re:What happened??

[3-State Bit]

that's hadn't happened, you uncouth philistine. Learn some grammar.

Your post also reminds me of woody allen's "match wits with inspector ford", a series of short scenarios set up to perplex the Inspector, followed by an explanation of ridiculous absurdity.

Funny stuff.

Obnoxius complaints

[aengblom]

Err, any chip that Intel is releasing has faster brothers and sisters in the lab ;-)

Oh wait I'm grumpy without the tags anyway...

Err, hasn't their been some other chip that's faster than this? (Ok, maybe not at a competitive price) but... wouldn't calling Intel's fastest desktop processor the "fastest every" be like calling a corvette or something the fastest land car ever?

(Second part an actualy question!)

microwaves

[kippy]

I remember when processors started breaking the GhZ benchmark, people were making jokes about how we're starting to get to the point where the things will be emiting microwaves since they are in the GhZ's.

anyone know how close we are now? will this new chip boil water from a distance?

even if we're a couple years off from that, are we going to need sheilding in our cases soon so that we don't cook our lower legs? if so, does anyone else thing that this would cause a lot of problems since compUSA won't take that into account when they do an upgrade?

Just some thoughts...

Re:microwaves

[Phosphor3k]

Just because the clock ticks 3 billion times a second dosnt mean it throws of radition that will boil things. Why does this always come up?

Re:microwaves

[karlm]

Yes, at somewhere around 2.4 to 2.5 GHz your system will be putting out microwaves identical to those produced in your microwave oven. However, the power levels are really low.

BTW, your case is already shielded. Look at older Macintosh computers: some of them have plastic cases with aluminum paint on the inside for shielding. Sure, I think the iMac and other clear cases are probably poorly shielded, but for the most part computer cases are somewhat shielded. In any case, your wireless Ethernet card puts out much higher levels of microwave radation.

Re:Oh I'm so not excited.

[(startx)]

yearly? your rich! I can afford to upgrade once every 4 years or so. I went from my 486/33 to 300Mhz k6/2 to my 1.4Ghz T-bird. Why yes, is is toasty in here :-)

Why? Are you afraid of a fiery inferno?

[burgburgburg]

Crybaby.

No Pain, No Gain!

Feel the burn!

Be the burn!

GHz vs. Billion Cycles Per Second

[Zordak]

The new P4 runs at 3.06GHz, at 3 billion cycles per second.

That's nothing. I hear AMD is going to come out with a 3.06GHz chip that runs at 4 billion cycles per second!

Re:GHz vs. Billion Cycles Per Second

[distributed.karma]

Um, remember that in Europe, billion means 10^12. That's THz for the prefix freaks.

(Considering that bi means two, billion = million^2 makes a lot of sense. Or can someone explain where the two is in the American billion?)

Re:GHz vs. Billion Cycles Per Second

[Rolo Tomasi]

Dunno 'bout them island apes, but in Germany it's:

Million(1E6), Milliarde(1E9), Billion(1E12), Billiarde(1E15), etc.

Hmm, I just realize that -illion is for even exponents, -illiarde for odd ones. Cool.

Common misconception: GHz != performance

[vlad_petric]

While I don't disagree with the fact that this is probably the chip with the highest clockrate ever built, performance has another ingredient - instructions per cycle (IPC). Now, clockrate remains the same, while IPC is strictly tied to a benchmark, and that's why people buy GHz, not performance.

Such claims have to be backed by benchmark runs. The PIV, when released, had a perf improvemnt of only 15->20% when running at 1.5GHz compared to a PIII running at 1 GHz

Re:Common misconception: GHz != performance

[vlad_petric]

Clockrate is a fixed constant of the processor. IPC (instructions per cycle) also depends on the instruction level parallelism of the program. Taking it to the extreme, if your program is a chain of fully-dependent instructions (i.e. instruction n+1 depends on the outcome of instruction n, for all n), the processor won't ever be able to execute 2 instructions in a single cycle (there's a lot more to this, like cache misses & branch mispredictions)

K7 processor manage to beat Pentium processors running at the same frequency precisely because of IPC (they get more work done per cycle)

Re:Intel

[Twirlip of the Mists]

X86 is a joke and anyone who is buying a processor these days should just wait and watch, that's what I've been doing since 1998.

How's that working out for you? ;-)

I'll upgrade...

[Gerald]

...when they come out with a 4.77 GHz version.

Re:I'll upgrade...

[Atomizer]

...when they come out with a 4.77 GHz version.

and also if it comes with 640MB of memory. That should be enough memory for anybody.

Re:I'll upgrade...

[bmwm3nut]

if i had mod points, i'd vote this funny. i was just about to boot up my old 4.77 MHz IBM PC from 1981 this weekend just to see if it still works. lets see, it'll probably be about 22 years from 1981 when they get 4.77 GHz...that's (22years/18month) about 15 moore-time-units...2^15 is much larger than 1000 (about 32000). that's interesting.

Re:I'll upgrade...

[RzUpAnmsCwrds]

Yeah, but a 4.77 ghz P4 would be a lot faster than a 4.77 ghz IBM PC.

What's the point?

[Anonymous Coward]

I mean really..what is the point? You have a superfast chip and you're STILL doing everything else wrong. Why are we just speeding up the CPU? Why are we not designing a better computer that doesn't NEED to ram everything through the CPU?

We're only getting a shadow of an idea with our GPU's...I believe Apple is the "first" to start making use of the video card's GPU for day-to-day stuff. And this is a GOOD thing.

Former Amiga users know what I'm talking about. There's a damn good reason why a computer with a "mere" 68000 was able to run circles around the PC's of it's day, and easily keep pace with more advanced intel chips.

Re:What's the point?

[Sloppy]

Why are we not designing a better computer that doesn't NEED to ram everything through the CPU?

Why would a computer designer do that, when Intel and AMD are already spending a lot of money making CPUs faster? Doing things wrong gets you more bang for the buck, because even though wrongness causes a performance hit, someone else is putting in lots of bucks for you.

I think innovation in overall computer design (like the Amiga exhibited) may happen again some day, but not until the CPU makers hit certain limits where Moore's Law stops working. At that point, it will start to make sense to work on other parts of the computer, because Intel's and AMD's R&D dollars won't be helping you as much anymore. But it'll be a while, if it happens at all.

Right now, if someone comes out with a truly new computer, people will just complain about how expensive it is. It will also be obsolete by the time it hits the market.

Re:What's the point?

[rodgerd]

You'd be wrong. Windows has been using the 2D acceleration functions of graphics chips for years. Apple have only needed to use the 3D features because they've used more eye-candy.

Re:What's the point?

[Benley]

I believe Apple is the "first" to start making use of the video card's GPU for day-to-day stuff.

*ahem* SGI's IRIX has been doing this for more than a decade. Their systems have always amazed me - just today, in fact, I managed to get an old Onyx system working. It's got a pair of 75mhz r8k cpus and a RealityEngine2. That's not a typo - 2x75mhz. Even with such slow CPUs, the user interface is lightning quick because of how well the OS makes use of the video hardware. Granted, the r8000 was a very unusual CPU in how effecient it was per clock, but still...

Re:What's the point?

[jpmorgan]

I believe Apple is the "first" to start making use of the video card's GPU for day-to-day stuff.

You are completely wrong. Did you actually bother to check this 'fact'? That's like claiming Apple was the 'first' to ship protected memory and preemptive multitasking in an operating system.

Hardcore systems produced by the likes of SGI have done this for years. WindowsXP supported this from its release, introducing it to consumer operating systems a year before Apple shipped their hardware accelerated Quartz in Jaguar.

For Christ's sake, even Berlin had this before Apple. And you know, that's just sad.

'Fastest' chip ever?

[penguin_punk]

How come I see myself returning to this article some day in the near future and scoffing at the "3.06GHz" label?

Does this remind anyone of the Popular Science articles where Planes may someday make transatlantic flights and In the 70's, automobiles will be obsolete, as personal gyrocopters will likely be the main method of transportaion.

Hell, I propose that in 2008, my shoelace-tying machine will be run off of a 3Ghz processor.

I'm not trying to bring down this article, as much as I'm bringing to light the humour behind the title.

Geez. I hope my dog doesn't piss on my shoe-tying machine.

So THAT'S what GHz means ...

[Hektor_Troy]

The new P4 runs at 3.06GHz, at 3 billion cycles per second.

Thank you for pointing out to me, that G == Giga == Billion and that Hz == Hertz == cycles per second.

This message was brought to you by the Department of Redundancy Department, who was happy to bring you this message.

Re:doesn't this happen like every month?

[Twirlip of the Mists]

We break a new GHz barrier every month?

What GHz "barrier?" It's not like 3 GHz was theoretically impossible or anything. This is just a matter of making something go slightly faster than it did yesterday.

Or is it the big round number that impresses you?

Re:Processor is not the bottle neck

[error0x100]

I have to wonder though, WHY? Todays software seems to need insane amounts of RAM compared to five or ten or fifteen years ago, and yet we don't seem to be all that much better off. Programmers just seem to squander the RAM faster than the RAM manufacturers can make it. Software expands to fill all available RAM. Its not even a joke. Why should "calc.exe" need 1-3MB RAM? The process running the task bar on my Win2K machine needs about 3MB of RAM, which is ridiculously high since all it has is a few buttons and icons and shows the time and has a menu, and yet the same thing in Windows XP typically needs close to 10 MB RAM. Windows Explorer in XP is MUCH slower than in earlier versions of Windows. Something is wrong with this picture.

I wish programmers would make some effort to optimize the stuff. Perhaps better tools would be useful. As a C++ developer, I would like a tool that shows me a breakdown of how much RAM is being used by which parts of my program. If such tools were commonplace, programmers would be able to quickly isolate the parts of the their programs that are hogging the most memory.

Re:Tech Specs

[airrage]

Oh my goodness...I fell for this AT WORK!!!

HT on Linux or FreeBSD...

[cowmix]

Does anyone know if Linux of FreeBSD sees any benefit from the 'hyperthreading' technology? All the things I am reading say that you OS needs to support threads, but how does the processor know what is a thread, and what is a process?

Re:HT on Linux or FreeBSD...

[BlueLines]

i've had a developer maryville board on my desk at work for the past 2 months (p4 2.8ghz). my experience with it so far hasn't been particularly impressive. i mean, it presents itself as 2 cpus to the underlying os (works w/ xp, .net rc1, and linux), but when you do something that actually taxes both cpus (make -j8 bzImage or what have you) there's a lot of thrashing and no true performance gain. i like the idea that no one program can totally lock up your cpu (netscape / q3 / X / etc), but i haven't seen any gains in day to day use.

i'm curious how oracle / msft will deal with the licensing issues that will come about from presenting virtual cpus.

-BlueLines

Re:HT on Linux or FreeBSD...

[Elladan]

Linux supports HT. No OS as far as I know sees much benefit from it.

The difference between a "process" and a "thread" is pretty small. A thread is just a process with shared page tables, for the most part. This means that there's less overhead switching between two threads, since you don't have to flush the TLB and caches. The processor per se knows absolutely nothing about any of this - it just knows when the OS commands it to flush the TLB and the caches, and change the page table addresses.

The basic point of HT is that it's sort-of another CPU, but it's just leeching unused resources from the main CPU. So, the scheduling logic in the OS needs to understand that it's not a real CPU, and thus should be grouped with the real CPU it's associated with. Linux 2.5/2.6 will support these tweaks, with 2.4 you'll need some sort of patch currently. Without the tweaks, you still get HT, it just doesn't help much.

But really, it never helps that much. Don't expect a 2x speedup or anything, even if your system is running heavily threaded applications.

Re:HT on Linux or FreeBSD...

[paradesign]

BeOS supports it.

Re:HT on Linux or FreeBSD...

[Sloppy]

how does the processor know what is a thread, and what is a process?

It doesn't, which is why the OS needs to support it. The chip still just has one cache, so HT might make it run slower if the threads are accessing different memory a lot.

To get best performance, it seems to me, you might want the OS to only schedule the second thread when there's already another thread from the same process, running on that processor. Then they might share the same instruction cache (if the two threads are doing the same thing) or the same data cache (if one thread is "consuming" the output of the other) and actually result in better performance.

In other words, it's the OS' job to really make sure it gets used for multithreading, and not multiprocessing, becuase using it for multiprocessing would be a Bad Thing.

Re:HT on Linux or FreeBSD...

[velkro]

I've got 85 Dual 2.4ghz Xeons running for 2 months now with HT enabled (both Linux + Win2k), and I concur. While each box appears to have 4 CPUS if you query the OS, running even make -j5 bzImage thrashes the heck out of the systems, negating any possible performance gain.

Re:HT on Linux or FreeBSD...

[Tim C]

Don't expect a 2x speedup or anything, even if your system is running heavily threaded applications.

You can't expect a 2x speed up even with 2 real physical processors - then, iirc, you tend to get a factor of about 1.8 increase.

HT is supposed to give you about a 25% increase in performance on desktop machines, or 30% on servers. I've not used an HT-enabled machine, though, so I can't vouch for those figures.

Re:HT on Linux or FreeBSD...

[Wesley Felter]

HyperThreading reduces my kernel compile time from about 1:10 to about 1:00; it's not a big deal, but it's free so I'll take it.

Re:HT on Linux or FreeBSD...

[wsloand]

it presents itself as 2 cpus to the underlying os... , but when you do something that actually taxes both cpus (make -j8 bzImage or what have you) there's a lot of thrashing and no true performance gain.

It sounds like you're not using the system correctly. With a make -j8, you're going to be accessing the hard drive a lot. If you have enough ram to make everything from a ram disk it will provide a noticable gain. Try something like running a massively parallel computational task and see what the benefits turn out to be.

Re:HT on Linux or FreeBSD...

[jpmorgan]

Why are you running make -j5 on a system with 2 CPUs? That seems a bit ridiculous to me.

Hyperthreading presents two CPUs to the operating system to provide certain performance benefits. That doesn't mean it's magically two CPUs - you've still only got the computational resources of one.

make -j2 will tax out the CPUs resouces as much as possible, and the two 'unused' virtual CPUs can be used by the OS kernel to do its housekeeping without interrupting the processes running in the foreground as much. Hyperthreading is a subtle technology. Beating it with a sledgehammer isn't the way to achieve the best results.

Re:HT on Linux or FreeBSD...

[stephend]

Microsoft have already stated that they will charge for the number of virtual processors (don't have the URL to hand).

AFAIK that means you have to have XP Pro or higher to run with HT enabled (as XP Home only allows a single processor).

Oracle will loudly complain about the liberties that MS are taking and then quietly do exactly the same thing in six months time.

faster faster faster...

[u19925]

so now you can waste your time on computer even faster.

Does anyone want 40" knife in kitchen? yes, if you just want to play games with it.

I call bullshit...

[Cervantes]

...with 50 percent saying they play video games while also burning CDs

I'm sorry, but there are only two explanations. One is that half of users out there are running maxxxed out machines that can handle that load (yes, with winblows). In which case, why the push for new chips?

The other explanation is that users really are burning cd's while playing games, in which case, the RIAA can pack up and go home, because those hundreds of thousands of CD's are obviously ending up as coasters, not as pirate booty.

I know, I know... I show my age when I remember the days where you clicked "burn" and ran like hell. I still remember the setup I had that would coaster the disk if I moved the mouse during the TOC writing. Admittedly, it was a brand new 1x burner, but still....

And considering my ole Celeron 300a runs Win2k just fine, why in the blue blazes would I need a 3G? Seems computers have hit the plateau... the average user gets along just fine with what they have, it's only professionals and gamers who really snap up the new hardware.

I'm gonna start a bet... how long can my 300 run before it's finally too slow?

(and to stop your flames, RedHat goes on my 1Ghz. So there)

Re:I call bullshit...

[istartedi]

how long can my 300 run before it's finally too slow

At least as long as mine. When I first got it, I OC'd it to 450. Although it was cool to see stuff compile so much faster, I decided that having IE occasionally display liquid web pages (text rendering and then not being erased--if you've seen it then you know what I'm talking about; it looks a lot like the page is melting). So I never OC'd again. /me often wonders how many "Windows problems" are really "crappy hardware from Best Buy problems". I know that some mfcts have actually stuck OC'd CPUs in a box, and that's the kind of wierd crap that happens; but I digress.

The only reason I am thinking of getting a new computer is so I can have something mobile/wireless. The new Tablet PC might be the ticket, but I want to let the early adopters deal with the bugs first. I also will want to use my new mobile as a desktop with a full keyboard that doesn't have the superfluous Windows keys. I have a feeling that will be the real challenge. Does *anybody* make a USB keyboard with the "fat" enter key, no Windows keys, etc? I'm typing this on an IBM keyboard that comes close to that ideal, except it has a small enter key, an oversized backslash key (?) and a large backspace key. My favorite keyboard, an Acer circa 1995, sits propped up in the corner, the victim of "stickiness" from overuse and an unfortunate incident involving an AT to PS2 adapter that broke apart.

Quite true, actually

[phorm]

So you've got a smoking video card, a super fast processor, and some other fancy peripherals. You've stocked your machine up... except you haven't taken the time to upgrade your RAM about the 256MB of PC2100 DDR...
ooops, mistake!
RAM does definately make a difference. It used to be that after a certain amount of RAM, the speed difference was negligable, but since then OS's and apps have been chewing up more and more memory.

Once your monster-fragging memory-chewing game starts getting near memory limits, you are going to see performance loss, even on a high-end processor. You'll start hearing that annoying clickety-clickety-clack sound, which often indicates your hard-drive is whirring away storing up swap space.
Even if you've got a nice new 7200RPM (or higher in SCSI) hard drive, it's not going to get near the transfer speed as your RAM, as you're limited by the mechanical medium. Suddenly, your game will start stuttering, and some bigass monster or perhaps a dude with a show gun is going to tag advantage of this to remove your head.

I have 2 machines, an Athlon XP and an old Duron. The Athlon is by far superior, faster processor, faster bus, faster RAM, etc, etc. The Duron, however, has half a gig of RAM (and probably more soon, PC133 is cheap and abundant). While the Athlon takes the lead easily at first, it can decrease noticably in performance as I start running into heavy swap usage.
Windows XP is a big fat whale of an OS, and it sucks a lot of my RAM to begin with. Throwing a big game on top of that (and whatever helper apps multitask in the background) can put it in the red zone fairly quickly. In contrast, with 512MB of RAM, the OS tends to put its bloated self into memory, and still leave enough space for my gaming needs.

The moral of this is, that - as always - a PC is only as fast as its slowest component. In many cases, you can bottleneck at the RAM, or - when you run low on memory - a the hard disk in swap.
It's like having a car with a huge engine, and only 6" tires or a really narrow gasline. You have to have balance... and a superfast processor really isn't going to cut a big difference nowadays until everything else catches up.

Actually...

[JanneM]

I am looking to upgrade my machine. WIll I be getting the latest and greatest in insanely fast CPU:s? No. My current, aging machine (a 600Mhz Athlon) is actually well able to support just about everything I do today. While a speedup is nice, it has long since ceased to be on the 'must have'-list for me.

Instead, my interest is in getting a big laptop to use as a desktop replacement. Something with a decent-sized screen and keyboard, 3d (read: nvidia) graphics system and plenty of memory. Also needed is the ability to plug in a 'real' keyboard and mouse when I'm sitting by my desk. Whether the machine runs at 1.3, 1.7 or 2.2 Ghz really does not matter for me. What I'll have is a quieter desktop able to bring along wherever I am.

Looking around, this seems to be a bit of a trend; laptops are less expensive today than a few years ago and more capable. A number of my friends are also thinking along the same lines, and so are a lot of other people as well, judging by the increased sales of laptops.

Speed just is not the defining characteristic of computers today that it was just three or four years ago.

Intel ads?

[zerocool^]

Is anyone else noticing that 9 times of 10, when you refresh this page, you get an intel ad?

or, is anyone else with IE 6 having the problem of the browser thinks the entire page is a link to ads.doubleclick.com/jump/bunchofcrap438934?

I'm not saying conspiracy, i'm just saying conspicouos product placement.

I Smell a Poll...

[flogger]

What's yer CPU speed?
[]2.8 GHZ+
[]2.5GHZ+
[]1.5 GHZ+
[]1ghz+
[]500Mhz+
[]233 Mhz+
[]Cowboyneal runs the cage the powers my CPU.

The Core: Part II

[BiOFH]

82 watts of power dissipation...
The VRU has to be cooled (or be of a more expensive design) so it won't... oh god I don't wanna think what it will do... melt? catch fire? explode? Eat through the Earth's core and Bruce Willis will have to team up with Hilary Swank to save us?

On the bright side, people who live in the northern hemisphere can consolidate their heating and gaming bills.
"Bill, we got any space heaters left in the back?"
"No, but we got some of them new Pentium machines."

Where does this end? I know Moore's law will 'eventually' catch up and they'll have to move away from just throwing more transistors at the design (although, like some weird horror movie, they keep infusing a few more months into the x86's life), but, seriously, how much is too much? Where will people draw the line on power consumption for their PC? Once upon a time I thought that 30 watts for the G4e was high. That's peanuts compared to this!

Start ordering more Lieberts, y'all. And invest in air handler stocks.
===

Re:HT on Linux or FreeBSD... (fixed)

[cowmix]

Does anyone know if Linux or FreeBSD sees any benefit from the 'hyperthreading' technology? All the things I am reading say that your OS needs to support threads, but how does the processor know what is a thread and what is a process?

Bang for the buck

[LoudMusic]

I'm all for progress in processor speed, but the consumer is looking for the most bang for their buck. Knowing Intel's past, that's going to be one hella expensive chip. I wish they would focus more on making a quality chip for much less (or just charging more acceptable prices) rather than seeing how fast they can burn the suckers up.

The last few computers I've bought have all had middle of the line processors in them because the price breaks are enormous. I so absolutely no reason to purchase a top-notch processor when you can get one a couple hundred megahertz slower for more than a couple hundred dollars (US) cheaper. Those last few megahertz they're eeking out honestly don't make that big of a difference in the real world - especially at a premium price.

Newsflash!

[limekiller4]

...and in other news, today is the "Latest Day, Ever."

everyday math stuff

[sstory]

There are computer labs at NCSU with old Sparcs (running SunOS 5.7, for all you geeks in the audience) which seem to be practically equivalent, in ordinary usage terms, to P 200's. There's a huge difference, HUGE, between Maple or Mathematica on these systems, and on the PIII 500 in my office. But there's very little difference, it's almost unnoticeable, between those programs on my office computer, and the same thing on my home computer, an Athlon 1200 mhz. And I've used mathematica on a 1.7 ghz Dell in our office, and again, there's no practical difference. Maybe computing a bunch of Fourier coefficients takes 8 mins on the 500, and 4 mins on the 1.7.

Compared to the average person I do intensive computation, and I feel no pressure to upgrade. For the average user the need to upgrade must be entirely generated by marketing--right now performance improvements in hardware is irrelevant. I wonder what's going to change--assuming anything does--to make us all hunger for faster systems as we used to. I can't think of anything compelling, but i'm unsure because intel etc are spending piles of cash figuring out how to reestablish the need for improvement.

Re:HT on Linux or FreeBSD... (fixed)

[daeley]

Looks like you're hyperthreading already! ;)

Re:Moores Law and Fireplaces

[man_ls]

my system heats the room it is in to aprox. 4F hotter than the rest of the structure it is contained in. The house averages about 74F; my room is generally closer to 78-80F depending on the weather outside (south-facing windows)

The CPU and case temp themselves are quite reasonable; 43C/48C chassis/CPU, but it does add some heat to the ambient levels.

This Intel chip has to have double the heat output of my Athlon...that would be a nightmare to cool.

not as fast as the overclocked 1.8Ghz

[adpowers]

Somebody on the overclockers' forum got his 1.8 Ghz P4 C1 stepping (from a Dell computer) to 3.5 Ghz. While it doesn't have hyperthreading (which doesn't neccessarily give you performance benefits), it does have a much faster bus rate.

Re:HT on Linux or FreeBSD... (fixed)

[alexandre]

Apparently 2.4.18 or higher does support HT...

well, yeah...

[BiOFH]

... but the other day I copied a 3gb folder, burned a CD and played music all at the same time... but I'm on a Mac... and it's a laptop. Guess that's not the same. heh :)

Re:question

[spike hay]

What is a power user?

It means you use a lot of electricity with your computer.

About senseless performance increases

[eyefish]

In the old days, it was a big deal when we went from a 100 Mhz chip to a 150Mhz one. Today though a 50Mhz difference means nothing.

Don't you all think it is time for Intel and AMD to stop bringing out a new chip which is 0.000005% faster than the previous one, and instead start coming out with chips ONLY when they make a noticiable performance difference???

In other words, I hope that the next chip after the 3Ghz one is a 3.5Ghz one, then 4Ghz, 5Ghz, etc. And by the time we get to a 10Ghz chip they should start making them in 1Ghz increments.

Sure, I know clock cycles is not the whole story to performance, but geez, I see people upgrading their 1.9 Ghz systems to 2Ghz systems for several hundred bucks like if that's going to make a noticiable difference (on the other hand, it is probably because of *those* people that Intel/AMD do what they do)!!!

Yes

[TobyWong]

The intel commercial said so therefore it must be true.

Hyperthreading results

[friendofafriend]

The most interesting part of the reviews posted are the comparison between Hyper-threaded and normal mode. These nice graphs show that in all but one case, the speed is not harmed by having HT enabled, and indeed it improved the performance by up to 20%.
This will not make a single process speed up, but will make systems seem faster, as it is rare that you are only doing one thing at a time.

Re:What about Linux?

[Jim Norton]

Yes, kernel 2.4.something apparently supports it. The only Microsoft OS which supports hyperthreading properly is Windows XP. Windows 2000 and below doesn't utilize it.

So it's either Linux or XP, as far as I know.

Pricewatch

[unicron]

Pricewatch just through the mobo's up. I think I'll get the 1.5ghz chip inside one of the mobo's and that way then the 3.0ghz chips come down, I can afford one.

With 3Ghz I get 200 FPS!

[stefanlasiewski]

Wow! I just installed this new 3Ghz machine, and now Nethack runs at over 200FPS, even with full alpha light rendering and environmental audio turned on full!!!

Yawn - Hype for the sheep.

[dnoyeb]

Marketing hype. This is really nothing and I can't understand why THG is hyping this.

All this does it let the CPU have 2 apps it can switch between at. Normally the CPU has to wait on the OS to give it something to do. Now the OS can give it sort of a spare job to keep doing.

Still only 1 can run at at time though. Its NOT a multiprocessing system. Simply where the OS normally chooses which app gets to run, now the CPU can always hold 1 app in the hole, ready to run it when any down time comes along.

For those who ALWAYS run something in the background like Folding@home or SETI, they will certainly see an improvement. if the OS and CPU agree to keep that app on the CPU, it will improve performance. But it will NOT increase your fps because you will only have 1 app going then.

AND if you turn on dual cpu support in quake, you should see a performance hit if anything.

The results from THG bore this fact out. I wouldnt waste time on this if I were AMD. The everyday user still has no benefit from dual processing systems, and the servers will need TRUE dual processing systems.

Re:Yawn - Hype for the sheep.

[Steveftoth]

Actually it's a little more advanced that that.

Basically, you have all these different units for doing math in the CPU, and if one thread is only using say the interger units and another thread want to do float-point math, then the processor can actually let both run at the same time by scheduling the instructions properly. Normally, is one thread was all interger and one thread was all floating point, then the CPU would have to do a context switch to be able to run one type of instruction or the other.

It's a little more complex then that, but it's more then just holding one app in the hole. But yeah, for those running FPU intensive appls like seti, while doing normal stuff should see less slowdown.

Re:Yawn - Hype for the sheep.

[Twillerror]

I believe XP and the newest linux kernel are aware of this and have been somewhat optimized. this was a big deal when the Xeon's first started coming out. If your running and older OS, like NT 4 or an older kernal you could see a decrease. But the processor is executing both at the same time, each getting a piece of the massive pipeline.

You should read the articles on anandtech or arstechnica, THG has gotten kind of dumbed down as it's popularity has increased. Sometimes I read it first, then go on to other sites for more details. I've been writing multi-threaded apps for sometime, so this was interesting to me. It took me a while to fully understand it as well.

Re:Yawn - Hype for the sheep.

[dnoyeb]

I dont believe it requires any OS intervention other that the support of multiprocessors. XP and Linux and win2k already support this.

I feel strongly that its 90% hype. the 10% just allows them to get away with it.

Re:Yawn - Hype for the sheep.

[cartman]

SMT is not "hype for sheep."

1. SMT will become vastly more important as time progresses. This is the first iteration only. Being able to dispatch multiple threads to various pipelines, will grow increasingly more important as the number of pipelines in a CPU naturally increases. What happens when CPUs have 400 mil transistors, and 16 integer pipelines? There is no more instruction-level parallelism to exploit; the only way to get any more parallelism is at the THREAD level. In the future, we should expect to see CPUs with >4 thread contexts that perform at least 2x as well as single-context CPUs.

2. Hyperthreading does not just "hold 1 app in the hole, ready to run [during downtime]". Hyperthreading can dispatch instructions from multiple threads to various execution units simultaneously.

Re:Yawn - Hype for the sheep.

[dnoyeb]

Practical usage will be close to 0.

This as you describe it will be fantastically complicated. My professor in school's 'thing' was making a compiler to automatically multithread a program. It was wildly complicated. Its tough enough to efficiently use multiple threads. It will be even moreso of you have to consider the types of instructions the thread is running.

I don't think many apps will be designed for this because any app thats processor intensive will tend to get the whole processor anyway. For instance, a server typically has just 1 job.

Nice technology, but it seems a little costly and inefficient to use. Heck they don't even want to make 64-bit conversions...

Why are out TechSavvy ./ posters running TechCrap?

[Viewsonic]

I mean really.. stay with the times, folks.. You might outdate your own website and life. You're doing us a disservice by not using the latest and greatest.

Bah

[Dark Lord Seth]

This news doesn't matter. In a few months Intel will have brought out another, even faster chip that can do more, do it faster and more efficient while (hopefully being cooler then it's precursor.)

You arn't gonna get it in rack mount

[myowntrueself]

not without buckets of dry ice in front of it.

The back of a big render wall gets extremely warm.

Even if you could fit small enough heatsinks that would let it fit into a single space rack the heat thrown out the back of a tower of, say, 32 of these would be ridiculous.

Mind you, I *guess* you could hang old pizzas around there to warm them up.

When will 8GHz be out?

[tevenson]

Maybe Intel should concentrate on memory bandwidth instead of speed. Seems to me that all these MHz increases aren't nearly as effective as speeding up the FSB. We need a new memory interface architecture, go AMD?

After you hit about 60 fps in Q3 you're not gonna notice anything else higher.

Overkill anyone?

Re:When will 8GHz be out?

[SuiteSisterMary]

They did. It was called RDRAM, from Rambus. Everybody bitched and moaned, so they went back to using SDRAM. Or was it DDR RAM? Either way.

3060Mhz!?!

[ejaw5]

Shouldn't 640Mhz be enough processing power for everyone?

heat dissipation

[jonathanbearak]

when i leave my pizza out overnight by the computer, will it stay warm?

Re:Processor is not the bottle neck

[Proudrooster]

Why does Microsoft use insane amounts of RAM for simple applications you ask?

Simple, because it gets you to upgrade. It all starts innocent enough, you are running Windows 98 and your buddy gives you a copy of Windows ME. Wanting to be at the same level of modernity as your buddy, you install it, only to have your machine run slower. Eventually, your machine suffers from the dreaded Windows O/S decay and conveniently christmas rolls around. You then decide the old computer is going to the kids (or trash) and you get yourself a spiffy new Dell or eMachine.

This moves hardware, software, and yes another OEM Windows license that is locked to your genuine Intel processor. It also moves money out of your bank account.

I hope that clears it up. It's about getting consumers to buy more, so the latest and greatest bloat code will perform at an acceptable level of performance. Windows does a great job of masking the true power of the Intel architecture. In fact, the gap between Windows and Linux performane is growing and on identical hardware, doing identical work, Linux is 10-15% faster and tends to scale higher and support more clients as we have been seen in Samba vs Win2K, and tux vs. IIS benchmarks on identical hardware. Again, Win2k scalability has more to do with selling server licenses than creating better code. If your Win2k server runs out of ummpphh at 50 users and you have 75 users, then the solution is to buy another server from Dell and of course another OEM Win2k license locked to the CPU in the new server. Or, if you are just doing file and print server, you can scrap it all and put in a Linux box running Samba.

There is no economic incentive for Microsoft to write efficient code with a small memory footprint.

In contrast, the Linux kernel is constantly under the microscope running of embedded devices, strong-arm CPU's, and I still run a single floppy micro linux distro on a 486 (that even gets me a network stack). I am amazed at how much throughput I can get out of an old 486/100 with 32MB of RAM running Linux that booted off a floppy. It's just amazing how much power is there.

If I am wrong on any of these points, please correct me. If not, mod me up.

hyperthreading/SMT - don't get too excited

[Daniel Phillips]

This chip is more interesting than just the normal megahertz hike. It's the first of the desktop hyperhreaded chips - previously only available in the Xeon range (well, from Intel anyway. Other manufacturers had them).

It doesn't help a lot, at the very most a 20% speedup, typically much closer to 0%. This iteration just isn't that effective, maybe next time round with better management of cache or something, they might get it working more like separate chips. Right now it just doesn't.

It's good to see this entering mainstream though. It provides an incentive to write the kinds of multi-threaded applications that can actually squeeze out the full 20% speedup, and in turn, those applications will automatically be able to take advantage of real multi-processor boards without further changes. So this process ends with everybody having SMP laptops :-)

Is Intel going to add a performance-rating?

[puppetman]

"With the introduction of its Hyper-Threading technology, Intel has confirmed that constantly increasing the clock cycle is not the only way to skin the proverbial cat."

Yah, AMD has been saying that for years with their performance-ratings, and Intel's been saying that cycles-per-second was the measurement that the consumer truely understood, and was a good way to get a measure of the speed of the processor.

Wonder if Intel will adopt that, now that they have a CPU that, at lower speeds, can process more data.

Lotsa gigahertz

[Dirtside]

I was under the impression that transistors had a lower limit on how fast they could switch. A 3GHz chip, in theory, has a clock pulse every 1/3 nanosecond... and I thought transistors took a nanosecond or two to switch? How exactly does this work, then, or are my premises false?

Price/Performace....who's gonna buy this?

[PantyChewer]

Ok so I can buy this P4 with "hyperthreading" to emulate 2 processors or, I can go buy 2 Athalon MP 2200+ processors and a motherboard for less money....

consider the via eden platform

[honold]

mini-itx form factor, integrated video/ethernet/fanless cpu - just add memory and storage. link

if you wanted to go all-out on skipping the moving parts, you could run the os on compact flash using an ide to cf adapter from pcengines.com and use a cupid case with a dc power supply. just make sure to disable writing, or you'll wear it out! use mfs or a (non-essential) extra standard hard disk for data.

Re:consider the via eden platform

[JudasBlue]

Mod parent up, he is right. The Via Eden mini-ITX platform is the opposite of this 'We need more and we don't care if we need liquid nitrogen to cool it,' thinking that has been going on in the Mhz wars. For a load of applications, you don't need all that cruch, and smaller, cooler is the way to go.

Check out mini-itx.com for more on these eden processors.

Re:Wow! -- Optimal Pipeline Depth

[Erich]

There are actually some interesting papers out about optimal pipeline depth. At first they appear to have different conclusions, as they cover different architectures, but the conclusion is really sort of the same: optimum pipeline depth is about six fanout-of-four inverters per stage of work for integer paths and four for floating-point paths. Plus two (each) for overhead. That leads to crazy-long pipelines, I think the rough calculation for redoing the P4 pipeline came out to 50 stages or something.

If you do a google search on optimal pipeline depth you'll find some good results.

OMG

[scrytch]

Imagine a Beowu--WHAM WHAM WHAM WHAM.

This LARTing brought to you by the Narn Bat Squad

A friend just e-mailed me info on a faster one!

[dirvish]

This thing is faster. English summary at the bottom.

Re:Tomorrow...

[ncc74656]

...but will it have a decent mobo to run on?

Score: -1, Troll. As long as you avoid obvious crap (like anything from ECS), you should be OK. At this point, I'm somewhat partial to the MSI K7D Master.

Isn't price an issue?

[moosesocks]

Since this isn't practical to the high-end server market with the insane amount of heat it gives off, where IS it practical?

Buying JUST THE CPU will cost more than buying 2 athlon MP 2200s and a decent motherboard with it.

I'll stay with my athlon 750 which is treating me very nicely right now (although I may take the plunge and get an athlon xp)

Re:Tomorrow...

[Qrlx]

It will .

Re:Processor is not the bottle neck

[mandolin]

Why should "calc.exe" need 1-3MB RAM?

I see 1.6MB here. It's probably because the dynamically linked libraries 'calc' is loading. Maybe the "ShinyButton" DLL has a lot of other crap in it that 'calc' doesn't use.

Some of that DLL memory may, or may not, actually be shared with other processes. (depending on whether the "preferred address space" of the DLL conflicts with the addresses the program already has mapped? I'm unclear on it. On Unix, shared libraries are PIC so they can always be shared between processes.)

SMT.

[Christopher Thomas]

This chip is more interesting than just the normal megahertz hike. It's the first of the desktop hyperhreaded chips - previously only available in the Xeon range (well, from Intel anyway. Other manufacturers had them).

Which other manufacturers?

To the best of my knowledge, nobody else has built a SMT chip. The Power4 was a CMP chip (multiple cores on one die, not multiple instruction streams sharing the same core). Everything else that I've heard of outside of paper-land has had one and only one instruction stream.

SMT was a great idea, but with transistor count being less of a limit nowadays, CMP seems to have the advantage (as you don't have functional-unit contention between threads).

Re:SMT.

[cartman]

The IBM POWER RS64 that was used in the S/80s, was an SMT chip. All of those bigass IBM unix machines sold during that era had SMT.

Re:SMT.

[cartman]

I don't think that sparc is an SMT variant, however I think that MAJC might have had some form of SMT.

Uh Oh

[TheOnlyCoolTim]

I run my computer with the side of the case removed!

Tim

Re:Uh Oh

[karlm]

You're still probably getting less radiation from your case than a wireless ethernet card puts out.

OTOH, there's no such thing as too much caution when you're talking about the baby beans. I'd put my case back together if I were you, or else get metal boxers. :-)

Heat output. OMG.

[Sivar]

Tom's Hardware already has a review up about it, and it looks to live up to most of the hype.

Right. And for readers that want a review by people that actually know what they are talking about, you can read the review at Ace's Hardware.

In other news, the P3 @ 3.06GHz is indeed a fast CPU, but considering that it's maximum power dissipation is 105W to the Athlon 2800+'s 68W, it looks like people should stop making fun of the Athlon for running so hot. :)
This comparison isn't completely fair (the Pentium IV is faster), but even the P4 2.2GHz spews 70 W of heat.
At 105W, the P4 is approaching the (in)famous heat output of the Intel Itanium! This is not a good thing.
(note: regarding Tom's Hardware, I have no specific complaint about the article, just the website quality in general. The reviewers, except for Tom, have no clue and generally spew pure uninformed BS throughout their articles. Why the site is still respected is a complete mystery to me.)

skid marks

[drinkypoo]

"It ran so fast it left skid marks on my desktop!". How's that for fast?

Well I guess this one leaves skid marks in your pants.

No, that's a different kind of chip, mostly the kind made with olestra.

This is a chip from intel, so it leaves skid marks in your wallet.

Not to be confused with...

[Superfreaker]

The HOTTEST Chip around.

New! Doritos, Firery Salsa and Cheese.
An extreme, mouth-watering combination that will have you screaming for more!

Re:No FUD, just Facts

[Proudrooster]

Accuse me of FUD all you want, but examine the evidence for yourself.

Exhibit A
Win NT beats Windows 2000 in SQL Server 7 Benchmarks
What? The new O/S is slower? Must be FUD, doesn't have anything to do with bloated code and forcing users into hardware upgrades.

Exhibit B
Red Hat/Samba far outscales Windows 2000 on identical hardware
Yes your honor, it's true, at a load level of 16 clients Windows 2000 filesystem throughput flat lines vs. Red Hat Linux with Samba which is still scaling up nicely with 28 clients.

Does Windows 2000 mask the true power of the Intel hardware? Examine the report and look at the benchmark graphs. Decide for yourself if it's FUD or FACT. Note: the source is PC Magazine which if you will refer to this months copy contains many advertisements for Microsoft .NET .. Looks like PC Mag has some integrity.

Shall I continue?
Want to see why TUX stomps IIS and Apache for serving static content?
I challenge you to find the FUD in any of this. In fact, many of you might wish to save these links for future TCO discussions within your local IT departments.

PROVE ME WRONG!!!! Show me how Microsoft is doing it faster and better compared to either a) A Previous Microsoft Server Product, or b) Linux. Wave your hands and shout FUD all you want, but be prepared to back it up.

I wish someone would back me up! :)

As for my 486, I wrote a user mode driver which allows me to access the data pins on the parallel port to activate a relay and ultimately switch A/C power. (Web page coming soon.) This device can be used to remotely reboot Windows servers that BSOD, or turn on Christmas Lights add/or Coffee Pots via cron or telnet. Did I mention it all fits on a floppy, runs on a 486, and is network accessible? I am trying to shoe-horn a webserver onto the floppy now.

Obligatory TMBG quote

[distributed.karma]

You're older than you've ever been
and now you're even older
and now you're even older
and now you're even older
You're older than you've ever been
and now you're even older
and now you're older still.

only a few chips can reach 3.06 GHz? wrong

[Superfarstucker]

actually a lot of chips that are marked for sale under slower speeds can be overclocked to over 3.06 ghz using *AIR* (yes, thats right, air) cooling... needless to say rather "impressive". the new C0 stepping 1.8 ghz northwoods are very impressive overclockers, somebody reported on the overclockers.com forum a 1.7 ghz overclock on one.. thats 95% damn impressive, and on air cooling too. PIV's are great chips and very scalable, Intel probably has working chips all the way up to 4 gHz+ right now its just not as profitable for them to produce them (it would drive down the price of there current chips and create a market saturation effect... plus they obviously cost more to make due to yield rates)...

No market reward for the memory misers

[Ars-Fartsica]

Programs use more RAM because its there and its cheap and that makes it easier for programmers to use higher level toolkits to crank out code more quickly. We have the resources, why not use them??

Exactly

[Ars-Fartsica]

This quite says it all and it is worth repeating:

There is no economic incentive for Microsoft to write efficient code with a small memory footprint.

If programmer P1 can modestly abuse resources to get a program out the door faster than programmer P2 who takes the time to be miserly with ssystem resources, P1 will win and P2 will end up working for P1.

Re:Exactly

[error0x100]

:( You are right!

what about imacs?

[caveat]

dude, if it's just a kiosk, slap a 256meg stick and jaguar into some rev b (1st-gen slotloading) imacs. hell, if you use a ram disk, you can put the hd to sleep permanently and have a totally silent station.

Intel Link

[loconet]

And here is the link to Intel's view on this...

Re:Tomorrow...

[tchueh]

Correction, Tomorrow, AMD will ANNOUNCE a chip that runs faster and cheaper at half the clock speed...

Of course, actually seeing in a store will be a different matter.

Re:Processor is not the bottle neck

[error0x100]

The process running the task bar must* store each icon's pixmap in memory

Sure. Do the math. A 16x16 32-bit image in memory (mine are set to 16x16) needs 16*16*4 bytes of memory, which is 1 KB. Lets be generous and assume that you need four times that much memory for "overhead", and we have 4 KB per icon. If my entire task bar and start menu has 100 icons (that figure is high) then we're looking at 400 KB. And I *know* these are not all in memory, because I can SEE its reading them when neccesary from the hard disk. And I have 512 MB RAM, so they sure aren't in virtual memory. Win2K: PIII 667 512 MB RAM, GeForce2. WinXP: P4 1500 512 MB RAM, GeForce4. With 512MB RAM, it would be incredibly stupid for an OS to quickly put something as fundamental as the user interface into virtual memory unless absolutely necessary.

I know I get "start menu" delays from disk swapping if I've been running very memory hungry applications, but you should also be aware that the start menu is generally slow in XP because there is a deliberate built-in delay... there is a setting somewhere in the registry you can set the delay, in milliseconds. Can't remember where it is though.

If you want to make excuses for bad software, you will need to try again, .

Re:Processor is not the bottle neck

[error0x100]

quickly put something as fundamental as the user interface into virtual memory unless absolutely necessary

Hmm .. that sentence came out wrong. So before someone jumps down my throat, I meant, putting the *pages associate with a piece of the user interface as fundamental as the task bar* into virtual memory. (Especially if I've just booted a 512 MB RAM system)

Re:Processor is not the bottle neck

[error0x100]

This may be "conspiracy theoryish", but I've been getting a "subjective feeling" that Windows XP very readily puts Java applications into VM.

Something more concrete that I've noticed on my 512 MB systems, Win2K and WinXP both seem to eagerly start putting stuff into VM at 256MB RAM. Its pretty annoying, the strategy just does NOT work well, at work we write applications that typically need 200 - 300 MB RAM. In theory, we should NEVER need to swop, and yet we end up with easily 100 MB or more of paged memory. Is there somewhere you can configure this 256 MB limit?

Turbo Buttons (Was Re:fast chip?)

[mhesseltine]

What do you think it would take a motherboard manufacturer to make a board that utilized the "Turbo" switch and linked it to an adjustable BIOS setting for overclocking? Imagine, doing mundane desktop work, leave the thing off. Building a new kernel, while burning CD-ROMs and playing UT2003, hit the button and instant overclock; no reboot necessary.

Or, has this already been done, and I'm just out of the loop?

Re:Turbo Buttons (Was Re:fast chip?)

[Anonymous DWord]

Buttons? Feh, that's so 90's. Gigabyte has EasyTune.

Re:What about Linux?

[jpmorgan]

Linux 2.4 supports it in theory. However, the dynamics of scheduling virtual CPUs are slightly different from scheduling real CPUs. With real CPUs, putting a task on one CPU isn't going to directly effect the performance of another CPU.

Until the scheduling logic in Linux improves to better support HyperThreading you're not going to see a lot of direct benefit out of it unless you understand some of the subtleties of hyperthreading and use your system to take advantage of them. But, 2.5 is supposed to be much better.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Fastest chip ever?

[tshak]

Even if we narrow the scope to x86 desktop apps, it seems that based on preliminary benchmarks (with Hyperthreading enabled) AMD's AthlonXP 2800+ still reigns (albeit, by a very small margin) as the fastest chip available.

Re:Fastest chip ever?

[MikeBabcock]

I'd love to see a performance / power dissipation graph.

Or a performance / power / price graph (3d).

Re:GHz vs. Billion Cycles Per Second -- But Wait!

[Zordak]

Not really, because now you're talking more about operations per second (i.e., "flops") or possibly instructions per second (i.e. "mips"). The "GHz" measurement refers strictly to the frequency of the oscillator that is providing the clock signal to the logic gates. Incindentally, this is why GHz (or MHz in the old days; or KHz in the really old days) is a fairly useless performance metric -- it doesn't really tell you much. On the other hand, it's great for marketing because you know it's going to keep increasing incrementally, so you can keep telling people that their computers are out of date. If you're going to lust after this chip, lust after the HT stuff you were referring to. That really might make a difference. The 3GHz hype is just the latest "Ho-hum" increase in clock speed.

Crap, I need to upgrade..

[Lord Bitman]

I knew I was in trouble when the new Zaurus SL-5600 specs came out and were higher than my webserver..

In reguards to Hyper-Threading.

[nycbrujah]

Intel has a nice tutorial on the subject.
I know that of the Microsoft OS's, only the XP family supports the Hyper-Threading. I couldn't tell you if any other OS's support it.
Distilled down, the processor creates a virtual or logical second processor which assists it in using underutilized resources.
A lot of multimedia vendors would be interested in this, a lot of gaming vendors will jump at this.

Price Performance Ratio

[SailorBob]

Basically, it comes out that the XP 2800+ and the P4 3.06 Ghz are neck in neck for most real world applications, with less than 10% differance between them on anything most home or business users are going to run. So it really comes down to which is the better deal, especially in a depressed economy with tight IT budgets. At the moment, only the XP 2700+ and the P4 2.8 are shown up on pricewatch.com, with prices of $354 amd $389 respectively. Meaning that AMD still has the crown in the Price/performance arena. However, the gap is narrowing.

Yes.

[Erpo]

Yes, it's DRM-infected. It's called LaGrand technology and it's built into all new P4s and will be built into all AMD Hammer CPUs. It provides the "trusted" operating mode (in addition to regular x86 kernel mode and user mode) portion of tcpa support. With a fritz chip on board and an OS that uses Palladium, Microsoft will, for the first time ever, be put in the position of being able to charge you to access your documents. I'm not talking about the power they've always had to change file formats. I'm talking about the ability to literally refuse you access to the bits that make up the file if you don't pay up. After all, if it becomes illegal to reverse engineer file formats (How much will that cost in campaign contributions? Peanuts to microsoft.) and you're saving all your documents in MS Word DRM 2003 Palladium Edition, there's no possible legal reason for you to need to access your files with any application other than Word, right? And if Word is available on a subscription basis only and you stop paying....

As for the unique ID, no - P4s have no unique id (as far as I know). That's on the fritz chip, and not only will it be unique, but (I strongly suspect from reading the full General and PC-specific tcpa specs) it will be obtainable by anyone that can talk to your machine on a network.

----
Example:

Boss's computer: Hey, I want to send you an email, but I need to verify that you're subject to digital restriction mechanisms before I release the data to you.

Your computer: Ok. As of (this time) (this date), this machine is running in trusted mode with a trusted OS. (RSA signature and public key for verification)

Boss's computer: Hey central DRM authorization server at microsoft!

MS: Yeah?

Boss's computer: Is this public key (public key here) one that was implanted into a DRM-infected fritz chip, or is someone blowing smoke?

MS: Yeah.

Boss's computer: Ok, pc. Looks like you measure up. Here's the message: "Good morning employee! I'm offically ordering you to take risky business action X. I'm aware that this could kill off the company if it fails, but the possible payoffs are irresistable." Do not allow the user to copy, print, or otherwise manipulate this message. Delete all record of it being sent in one minute.

Your computer: Sure thing.

---- ...and that's just how the designers envision it being used. I'll leave the possible abuses of this internet-available unique pc id to your imagination.

Just had to...

[RAMMS EIN]

``I'm still squeezing the last bit of life out of my Pentium 233!''
My Pentium 200 is mostly running idle. :-) Except when I'm compiling, of course. Seriously though, most of my regular activities (web surfing, emailing, chatting, editing plain text, burning CDs, playing music) don't require much CPU power. It's memory that counts for me. So I'm just going to save money and energy by sticking to so-called obsolete hardware. If OpenBSD runs on it, what more can I wish for? (Err...)

---
Wombat's Laws of Computer Selection:
(1) If it doesn't run Unix, forget it.
(2) Any computer design over 10 years old is obsolete.
(3) Anything made by IBM is junk. (See number 2)
(4) The minimum acceptable CPU power for a single user is a
VAX/780 with a floating point accelerator.
(5) Any computer with a mouse is worthless.
-- Rich Kulawiec

Re:With 3Ghz I get---Perception & FPS (OT)

[labradore]

The human eye can only perceive about 12-15 frames per second as distinct images. Caution: the following is not substantiated fact, just informed opinion. You are invited to add corrections.

For the purposes of this comment, all our frames in frame-flipping motion representations contain sharp images by default. Above about 15 FPS, our perception changes. In the case where the frames contain temporaly (yes, temporal, not temporary) progrssive images of an object in motion, at framerates above 15 FPS we start to think we see an object in motion rather than a rapid succession of distinct images. However, as any good FPS (first person shooter, not frames per second) gamer will tell you, most people are acutely aware of the "choppiness" of motion represented by a stream of images presented at framerates between 15 and 50 FPS. In fact, people are sometimes capable of perceiving the presence of framerate acceleration at framerates above 50 FPS.

Note that framerate acceleration is not the same as the perceived acceration of a moving object represented by an image contained in a frame. Framerate acceleration is the increase or decrease of framerate during which time the images in the frames do not seem to have objective changes in motion. Rather, during framerate acceleration we perceive smoother or choppier quality of motion. Although, at framerates above 60FPS, most people can no longer perceive framerate acceleration, we can still register different physiological responses to different framerates. In other words, our eyes can tell the difference between actual motion and a frame-flipping representations of motion even at framerates up to about 72FPS. The physiological response to sharp, distinct images presented at framerates above 60FPS and below about 72FPS (at which point physiological response drops off sharply) is felt as eyetrain. This is why setting a monitor's vertical refresh rate above 72Hz helps prevent eyestrain. It also means that at framerates above about 60FPS, our brains no longer capable of processing incoming image data as fast as our eyes can supply it. In other words our visual perception bandwidth is probably limited first by our brains and second by our eyes.

Having a limited bandwith of perception is not a flaw. It is an adaptation to our surroundings. Things that move faster than we can perceive them either seem blurry or are invisible (if they move entirely through our field of vision). This adaptation gives us special feedback on the world we perceive: things that move too fast are dangerous to us and are flagged in our perception by uncomfortable blurriness.

Blurriness is also interesting because we can use it to better fool our eyes and thus fool our perception. Most film movies are presented at framerates of 25 FPS. How, then, do we watch movies without eyestrain and perceive smooth motion even though 25 FPS is well below the upper limit at which we can no longer detect both choppiness and framerate acceleration? In this case the images in the frames are pre-blurred for us. Consequently, our eyes do not detect the presence of the rapidly changing positions of sharp edges but instead register a blurry or soft edge that is more fluid. Our brains are good at interpolating movement and boundries based on blurry images and so we do not see choppiness but accept the images as smooth movement. The big question is: do we experience eye-strain at 25FPS with blurred, soft edged pictures? If film movies induce eye-strain then we can reasonably conclude that the eye detects and feeds much different sets of data to our brains when we watch the simulated motion of frame-flipping versus when we watch the motion of actual objects in continuous lighting. This would be further evidence that the eyes have more detection capacity than the brain has the ability to process. One mitigating factor in this situation is that movie theaters are darkened and the main source of light is the reflection of the image from the movie screen. In dark situations the capabilities of the eye are limited and it is possible that the darkness limits the eye's ability to detect and feed frame images to the brain due to retinal after-image effects (e.g. this is the same as the after-image we see after we stare into a bright light--the after-image may prevent the eye from properly discerning frame flips).

Your comments are welcome.

Faster chip is 8086!

[supergiovane]

8086 are still used in Space Shuttle which, orbiting at 17500 mph, makes them the fastest chips ever. However, I agree that the new P4 is the fastest chip on Earth.

Re:Faster chip is 8086!

[SuiteSisterMary]

And the reason nothing newer than a 386, or specially designed 486, is used in spacecraft is....the radiation!

Yes, out in space, there's so much more radiation floating about that you've a much higher chance of having a bit flipped. The more dense your transistors, the more likely you'll get something flipped.

I seem to recall one system using 3 486 chips that would each do a calculation, then the results would be compared. If at least two of them agreed on the result, that's what they used.

Re:Yeah,

[Alan Partridge]

you really should seriously think about getting a life

Yay! A 9.28% increase in clockrate!

[Junks Jerzey]

3.06GHz is a 9.28% increase in raw clockrate over the last fastest chip at 2.8GHz. That everyone is going nuts over this shows how sad PC hardware fanatics have become.

Bloody waste

[LarsWestergren]

82 watts, that's great!
* Higher energy bills!
* More energy wasted as the processor idles!
* More heat meaning:
-New fans, either expensive or noisy.
-More heat in the room. My own apartment is already hot enough on most summer days, and the server room I administrate at work is also too warm for comfort.

Re:Yay! A 9.28% increase in clockrate!

[MikeBabcock]

A 10% increase is pretty impressive, considering the increases we used to get out of Intel (2%ish) ...

For what its worth, I think it bears mentionning that a PR2600 from AMD is running nowhere near 2600MHz and is still holding its own against 3GHz parts from Intel. Thats some impressive engineering folks ...

Re:Yay! A 9.28% increase in clockrate!

[Junks Jerzey]

A 10% increase is pretty impressive, considering the increases we used to get out of Intel (2%ish) ...

It's a 9.28% increase in raw clockrate. The actual performance increase is less than 9.28%.