Slashdot Mirror
Tom's Hardware Retracts P4 Endorsement
Dice writes: "More benchmarks have come in and Tom (of Tom's Hardware) is expressing doubt in regards to the P4 in this article, "I have to admit that I started off being a believer in Pentium 4 and I still respect Pentium 4's future potential. However, right now I am genuinely disappointed. For the time being, I wouldn't let any of my friends or family members buy a Pentium 4 system. It's simply not justifiable."" Intel is definitely not impressing the hardware reviewers with their new chip.
Where have you been for the last 10 years ? To make money on stocks you don't have to guess how the actual company that has the stocks are performing but rather how everyone else will do.
> Yes, but won't it take YEARS, if ever, for this "advantage" to actually benefit the majority of apps that will be run on a P4?
Not really, closed source applications that can always use more performance will have P4 optimizations quickly. The benefit for other apps will be imperceptable so replacing them would be a waste of effort.
Open source people can just recompile when gcc catches up 8-)
> Who is going to rush out to support SSE2 instructions for a chip that isn't likely to sell very well?
Don't underestimate the power of inertia, marketing, and a good "shrink and tweak" to sell a lot of P4s.
-- Eric
The g4 have been stuck at 450 and 500Mhz for months, while x86 has hit 1.7ghz. Beacuse of the red-hot compition between AMD and Intel, the PowerPC arch has just been left in the dust. I havn't seen any benchmarks comparing a 1.2ghz athlon to a g4, doing video, content creation. But I *seriously* doubt the g4 would be able to beat the athlon...
ReadThe ReflectionEngine, a cyberpunk style n
What operating systems support SMP on the AMD motherboard (I'd heard that it isn't the same MPS spec)?
Adding an apostrophe is an acceptable, albeit becoming outdated, method of expressing plural acronyms. It was just the way I was taught in school, so that's the way I do it. Besides, where do you get the notion I was grammar flaming? I corrected a number, a fact. That's quite different than picking on spelling. You're the only one grammar flaming here.
Do Dell servers support SMP? My SMP system has more than 16 IRQs.
Intel's offers neigh hold a candle to Sun's offerings. An E10000 can have 64 processors and 64 gigs of RAM (which is SDRAM by the way). P3s max out at two processors and ~6.8 gigs of RAM (IIRC) but due to practical hardware limitations usually max out at 4 gigs. Xeons on the other hand can page as much memory as SPARC chips but the GX chipset limits the number of processors to 8 meaning you have to build a cluster in order to get 64 chips. The E10000 scales to 64 processors with no clustering which is an inherent speed advantage. There are workarounds for everything of course, do theoretically you CAN hack together some P3 or P4 boxes to handle huge databases and do a bunch of fancy shit but it will still cost you alot of money.
I'm a loner Dottie, a Rebel.
--
Why do you fear the moderation? Liberate yourself, speak freely !!!
degrees is a mesure of temprature, not heat. Very diffrent things, actualy
ReadThe ReflectionEngine, a cyberpunk style n
Doesn't ACPI get around this problem of IRQs? I've heard that Win2K on an ACPI systems puts all of the devices on to the same IRQ.
Ever heard of the saying "Can't please everyone"?
Well I am sure that no matter how hard the try they are going to get people who will blame them for being a bad company after being fired.
I mean, who wants to face becoming older, or slower, or even incompetence.
NOBODY!
At the same time though form those people some may be legitimate claims and of course they shouldn't go without being noticed, but what's to say a whole company is bad because of a few incidents?
Thank you for reminding me of that. It's been so long since I used DOS/Win32 that I forgot that certain cards like the Live! do use multiple IRQs. However, I would hardly qualify this as a "usual" case, but it is a case, nonetheless.
Poor device support was one of the reasons I couldn't stand Windows (9x) anymore. I'm sick of hearing of "hardware incompatibilities" with alternate operating systems. Sure, it may not be as easy as "Hardware Wizard found new hardware and is installing," but at least you don't have to wrestle with driver modules and such with other operating systems (Well, maybe OS/2). Maybe it's just me.
"Give the P4 time, it's not worth it to buy it right now. But as code becomes SSE2-optimized and the like, the Pentium 4 will strut it's stuff, EXACTLY as the Pentium Pro behaved when it was first released."
True, the Pentium Pro didn't impress many when it came out, and reincarnated as the Pentium II, it did beat the AMD K6.
"This is like benchmarking an Athlon on 16-bit code on a performance-per-clock basis as a Pentium. The Pentium would waste the Athlon in 16-bit code, because the Athlon is simply not meant to run with 16-bit code very well (ala using x87 FPU on the P4. The P4 wants to use SSE2, which is superior.)"
This is open to opinion. AMD's Athlon/Duron is much more important in the marketplace today than the K6 was when the Pentium Pro was released. There will be no rush to SSE2 until there are a lot of P4's out there on desktops. It isn't worth it to the software companies (even Microsoft) do waste the effort. Simple point of fact is that most all CURRENT apps run better on a 1.2 GHz Athlon than they will on anything less than a 2 GHz P4.
BTW, AMD is going to also support SSE2, I believe they have licensed it. I know it will be in the `Hammer 64-bit chip, but I'd bet it will end up in future Athlons.
Actually, it's to Intel's advantage that AMD use SSE2, as it will make it an industry standard. Intel right now isn't in a position to dictate industry standards, it lost this ability when they went to proprietary motherboards (Slot 1), then really lost it with RAMBUS. Slot 1 and RAMBUS did to Intel what Micro Channel did to IBM.
=== The price of freedom is eternal vigilance
Maybe I live in a cave, but I personally have ever seen an AMD advertisment on TV or in the trade mags
This is a true statement. Remember MS vs IBM in the OS/2 wars? IF IBM would have had a marketing department like MS has, we'd all be using OS/2 now vs Windows. (and yes, I run Linux also) IBM's marketing deptartment sucks, period. AMD better learn that lesson quickly!
Life was hell, then I discovered Linux...
The whole 15 IRQ thing is annoying! Yes, yes, I know that a lot of you will flame me and say "get a SCSI system," but some people can't always afford that kind of system (but I would love one!)
So then it seems your whole basis for complaint is that you don't make enough money to buy the things you want. Am I missing something here?
-----------------------
Nicotine free Amish .sig.
Intel is doing the only thing they can do - make faster processors than their previous ones. But I think that we've finally reached a point where a fast processor doesn't mean a whole lot. You won't be able to tell the difference between an 800MHz processor and a 1.5GHz processor unless you are benchmarking. Even so, 1.5GHz is almost twice the clock speed of an 800MHz processor. You can tell the difference between a 166MHz computer and a 350MHz computer.
The PIV is powerful alright, but I don't see the need to be at 1.5GHz right now when most people are fine under 500MHz.
What's really sad is that a lot of people think MHz is equivilent to power, when this is far from true. a 1.2GHz Athlon could toast a 1.5GHzP4 in most things, which proves that more MHz doesn't equal more power. And a G4 at 500MHz is often comparable to a PIII at 800MHz. Intel designed the proc for higher GHz with less power. People who look for high GHz will be swayed to Intel thinking they are getting a more powerful chip.
I'm sick and tired of seeing 1GHz machines come with 64MB of RAM. When will people learn that processor speed doesn't equal performance?
No, lotteries are for people who'd rather throw away $5 on a chance to win millions, rather than $5 on cigarettes and a chance to die.
Are you claiming that millions of $$ will keep you from dying?
Of course, if you DID win, you could almost afford a P4 system..
Leading the partnership for a Slashdot-Free Slashdot, Son of Dog
Lets face it; many of Intels 'new' chips don't make immediate sense, but who was buying the predecesor to any of the above chips once the new style had been in the market for a while.
Well, I don't think any of the reviewers giving the P4 bad reviews are hesitating to point out that this is a design for the future, that the primary goal of the P4 is to be the start of a whole new line of performance improvements, that the P3's are pretty much at the end of the line, etc. etc. None of them are denying that the P4 has a lot of potential, just as all the chips you listed had far more potential than the chips that preceded them.
But the purpose of a review is simply to give people an idea about whether they should splash out their cash or not. And the opinion seems to be fairly unanimous at present that your cash should be kept in your pocket, or spent on an Athlon. That's all.
These reviewers absolutely love their benchmarks. If a 2GHz P4 is romping it in with the best benchmarks in a year's time, you can bet they'll be singing its praises, alleged pro-AMD bias or not..
Fighting over the newest and most bestest Pentium based system is getting a bit old. The P4 is DIFFERENT from the P3 it isn't merely a core modification like the P3 was to the P2 and the P2 was to the PPro. The P4 is something Intel wants to promote as a real solution until they are able to roll out IA64 in the mainstream. If it costs too much or doesn't do what you want, don't buy the fucking thing. I'm not going to buy one. Let the OEMs buy a shitload of them and package them in consumer PCs. In terms of actual core quality, I'd vie for a Xeon over a regular P3 or P4. I get the option of lots of memory, better pipelining, and better branch prediction. Along with that I get the option of an enormous L2 cache which is very important when you're using the same instructions over and over, say when I'm doing 3D rendering or have a web server running; it's also going to increase performance on large numbers of non-repeating instructions (i.e. a large number of individual apps running all doing their own thing, say on a large multi-user system). The P4 is in the same situation the K6-3 was, AMD haters hoped it would kill AMD while supports defended it as more of a testbed than anything else. The P4 is a production test of some new approaches to things. It won't sell well in retail sales but I bet OEMs will eat it up because their customers will want it. The Duron many of you bought is a combination of technologies used in the K6-3 and Athlon, if the K6-3 had never been released the Duron might cost you a lot more due to the fact that some techniques were untried which means unrefined. In production cases, anything you try to do without first perfecting will cost you alot of money.
I'm a loner Dottie, a Rebel.
The limit on the number of interrupts was solved by PCI. The PCI bus has four shared interrupt lines. We just need to get rid of all of the legacy ISA crap and build systems with USB, IEEE-1394 and PCI.
Mea navis aericumbens anguillis abundat
Pentium's 4 architecture leaves much to be desired. It's price compared to equivalent AMD chips is way to high... the price difference between an ATHLON and a P4 would be better put to use by buying more RAM, a better motherboard or video card -- you'd end up getting a lot more bang for your buck with that.
_______________
SitePoint.com - Resources to Build and Grow Your Site
_______________
I don't care who wins the fight. The bottom line is that the more competition in terms of speed, advancements and price, the better for the public. As for AMD not having enough advertisements... Just remember for every advertisement that Intel buys air-time for; you pay for it (at least those who pay the extra dough to buy an Intel processor). Five months ago a top of the line Ghz processor cost at least a thousand dollars for both Intel and AMD. Now AMD's top of the line processor barely costs half that. Not because of efforts to bring joy to the consummer, but to compete with Intel dropping their prices BIG TIME. Bottom line, I hope that they switch places in terms of all of those factors (price, speed, core advancements), because it means more of the good things (speed/adv's) and less of the bad things (price).
Quake 3 does include SSE enhancements. we are probably seeing the bitchin p4 SSE2/MMX engine accelerating q3. ~q~
Lawrence Person (lawrencepersonh@gmailh.com (remove all "h"s to mail)
http://www.lawrenceperson.com/
Many people accuse Tom of being biased against this or that etc... the reason is probably that in Europe, newspaper and journalistw have the tradition of voicing opinions. There are very few newspapers or news programs that keep the "no opinion" line that you can find in many US medias (like CNN). So, Tom is not affraid of saying "Intel sucks" when truly, Intel products sucks. That might not be neutral, but that is not the way journalism work in Europe (and remember, Tom is German).
Well, that's my idea on it anyway...
I wonder if this is entirely true. I have no
knowledge of chip design and layout issues
but it seems to me that sheer interest in
advanced features wouldn't push Intel to
P4-like processors. It is twice the size of
P3, which means less chips per wafer and less
yeild.
I would guess that stuff like SIMD can be kept
proprietary, so they push on that at the
expense of generic x86.
It also seems as though Intel squeezes about a
factor of 5 increase between initial
introduction speed and how far the core is
eventually pushed. Is there a reason? And if so,
should we expect P4 core to be pushed to 5Ghz
before topping out.
Maybe the downfall of the P4 will convince Intel to rethink their plans, maybe not... either way both Intel and AMD are still rehashing the x86.
but, I have to take exception to the following statement:
The mere fact that Corps. act as though they own the laws and can do whatever they wish does not mean I have to accept that as a 'fact of life'. Replace 'corporation' with 'mafioso', and that line suddenly looks less appealing, even though the Mafia has been (and probably still is) supporting some causes which might, by some, be seen as beneficial for society. Like ridding neighbourhoods of crime (by criminal means, but still). That does not negate the poisoning role of the Mafia (or other crime syndicates) in several public institutions.
So, cheers to Intel for their insight that free software and business can go together. But boo to them (and all other nasty corporations) for their continued disregard of 'the intent of the law', for their heavy-handedness, their lies and their greed.
--frank[at]unternet.org
It looks like Intel is going to have to sit on the back burner as AMD once did for so many years. Rethink the architechture, for now, David has beaten Goliath
I think most of the big buyers of SMP stuff are rather conservative.
;).
A handful of overclocking geeks isn't going to help your bottomline, but just a handful of datacenters with hundreds of multi-cpu "enterprise" (aka high margin) servers will. Smirk if you want, but yes AMD's hope is in those "idiots" who fork out USD500,000 just to barely run Microsoft Exchange 2000
So it's no point for AMD to start pushing out SMP stuff before they've gained recognition for producing stuff that won't get the IT head fired.
That said, I think AMD is well on its way to that level- especially with all the Intel vs AMD press, and Intel's recent screwups.
Step 1) When people start comparing, it means to them you're comparable.
Step 2) The other side screws up, so now you just might be better.
In short, everything seems to be going as planned for AMD. If AMD can come up with a real good SMP chipset next year and convince a major server vendor, then they're all set. I'd pick IBM as a possible target vendor. They seem to support almost anything, as long as they know their customers will fork out the money. Their mainframe unit even pushed Cisco stuff, despite grumbles from their network unit.
Even Dell might buckle eventually, especially if other server vendors start posting significantly better benchmarks than they can with pure Intel stuff.
Heh, Sun's going to have even more pressure. With higher and higher DDR and RDRAM data bandwidths, I think IA32 servers will start giving really impressive DB/Web performance despite smaller cpu caches than Sun's SPARC processors.
Cheerio,
Link.
Using this same logic, why not compare four 500MHz PowerPC G4's to one Athlon? One Athlon actually costs more!! Load Linux onto these machines and which do you think will be faster - 2GHz of PowerPC or 1.2GHz of Thunderbird? On well threaded apps, the PowerPC would kick ass!!! (This is in theory - is it like this in real life?)
d ex.shtml
While this logic isn't actually flawed it just never turns out to be this way in the real world. First of all, someone actually has to make the machines (forget about that G4 idea.) I'm sure the AMD760 MBs will be out soon but, until they are, it really isn't a fair comparison. Also, don't forget about the additional hardware required to support dual CPUs - larger power supply, extra cooling, more expensive motherboard. The best "real life" test would be to take a computer manufacturer (like Gateway) and compare their offerings. Not a perfect comparison but at least it takes into account _all_ of the costs associated with building a computer.
PS: I got the Athlon price from this:
http://www.hiphardware.com/economics/cpuprices/in
I'm sure it is slightly outdated and that it is now possible to get a better price. The G4 sells for $150.
Willy
Clock that Athlon to 1.5 GHz and it will a heck of a lot more than 54 Watts. I think they projected something like 80-90 Watts. But the P4 will still be at 52 Watts. Thus, the P4 has WAY more headroom for frequency increases.
Well, let's look at it face on: The P4 with about 1.5 GHz easily beats the PIII with 1.0GHz (and probably even 1.13). A big part of it is probably due to better memory performance. But as the PIII seems to be at the end of the lane MHz wise (as the 1.13 GHz "launch" showed) the P4 is just at the start of it. The P4 was desinged to run at a higher speed, it makes no sense to compare it to a PIII clock by clock because the PIII can never reach those clockrates. We'll probably see 2GHz P4s in the stores within the next year (intel announced them for Q3) competing with Athlons (Palomino) clocked at about 1.6 GHz.
If you look at it this way the 1.2 GHz Athlon and the 1.5 P4 are just the top of the line processors so it's just right to compare their performance. But to be fair one should allow for some increased performance for the P4 with some firmware updates (a few percent), and, more importantly, consider that the P4 design will probably go a longer way. So the P4 will set high standards for the next major overhaul of the Athlon core (Thoroughbred) and AMD moving to 64-bit with the ClawHammer.
This means we'll see some serious competition between intel and AMD next year, and that's just what benefits the consumers most.
"By the way if anyone here is in advertising or marketing... kill yourself." -- Bill Hicks
I don't think you understand you chips achieve low power. It has very little to do with the linewidth, and everything to do with power management.
The 1.5 GHz P4 and 1.2 GHz Athlon use very similar processes (at least, both 0.18 um). The P4 has about TRIPLE the number of transistors (which says that the P4 SHOULD consume way more power), and runs at 25% faster clock speed (which says the same thing). YET P4 USES LESS POWER THAN ATHLON (54 watts vs. 52 watts).
The explanation is the P4 has all sorts of fancy power management built in to make it consume so little power. At least, it has this in much better quantity that the Athlon does. This is an extremely important competitive advantage of P4.
Well he was pretty damn right about nVidia vs 3dfx. Voodoo 3 was crippled hardware and, despite honorable framerate, was really lagging behind the TNT/TNT2 features. The Voodoo 3 was basically an overclocked 3D core from the Voodoo 2 with a 2D engine. And since the Voodoo 2 was basically the same core as the original Voodoo, but with a second pipeline. So Voodoo 3 was a very fast Voodoo 1, but with no real feature improvement (same max 256x256 textures, same lack of AGP support, same 16 bit rendering, etc.)
Looking back to what nVidia and 3dfx are now, I don't think Tom made a mistake in saying 3dfx products sucked...
Don't forget that p4 has a whole set of new instructions and I'm wondering how much optomosed soft will affect the performance.... Still even if it is a significant increase I can't see the majority of users getting benefit from it until more p4 optomoised soft comes on the market - and why would you buy a chip that runs your current apps slowly.....
I think Intel is definitely aware that this cpu, in its current incarnation, is not satisfactory. But it does do one thing for them. It let's them say they have a faster clock speed than AMD. That's why they released it. They don't really care how well it works right now, they just want bragging rights. Eventually, they'll have to remedy it, but for now, I think they have accomplished their goal. To people who know better, it works against them, but to people like their stockholders and the average consumer, it lets them say, "Hey, look what we did. Look how fast our new processors are."
-N
Would you compare a Porshe against 5 Escorts?
Everything in this post is false.
It turns out that a few points make it more attractive to use a more complex core:
This means that making the processor core itself larger doesn't have as big an impact on the size of the chip as a whole as one might think.
When yields approach reasonable ranges, as they always do eventually, the cost of an individual chip drops dramatically. Most of the cost of a module is support cost for the company that produced it, as opposed to raw silicon cost. While the die cost is still significant - especially when you're still fine-tuning and have low yields - chip size isn't as big a problem as it might first appear to be.
We've reached the point where adding more cache to a chip doesn't help very much (for many applications, at least). Thus, the only way to get a performance gain over one's competitors is to produce a chip that has a higher clock rate for a given linewidth, or that can find more instructions to issue per clock, or both. To do this requires a redesign. We're nowhere near having a perfect design yet; there are always new things that can be added, especially as linewidth shrinks and transistor counts rise. So, redesign remains a useful way of improving performance.
In summary, the cost of using more transistors actually isn't that high, and the benefits of a redesign using these transistors are potentially great. So, the new cores continue.
Bzzzzt! Thanks for playing. The P60/65 was an evolutionary cul-de-sac. Different socket than Socket 5/7, often with 486-grade parts, fdiv bug, in short, a total disaster.
The AMD K5-133 handily kicked the P60's ass.
It took the Socket 5 P75 to establish the Pentium platform...you can endlessly upgrade a board from that generation on.
---- Hey Grrl Geeks! Your very own geek news site has arrived!
Knowledge is power. Knowledge shared is power multiplied.
I'm sorry. But I've grown QUITE tired of Tom "Got my agenda" Pabst in the last couple years. Hope his site does well and that it draws in a lot of advertising revenue. But he's NOT got his finger on the pulse of the business user.
FACT: No truly large business is going to trust their system contracts to AMD. PERIOD. Note: I'm not talking about some little 100-station computer shop. I'm talking BIG business. Whose systems number into the high thousands, if not tens of thousands.
AMD is a great home computer and enthusiast platform. Little more. Several things count against it though.
What this boils down to is that big business KNOWS Intel's going to be around for them to yell at if something breaks. They don't necessarily have the same comfort zone with AMD.
Feel free to refute this all you want. I merely ask that you provide me with examples and hard data.
Chas - The one, the only.
THANK GOD!!!
Chas - The one, the only.
THANK GOD!!!
A lot of people here seem quite confused about why Intel has designed the P4 the way they did. I'll tell you why:
It's *all* about the memory bandwith
The gap in MHz between CPU and memory is growing, and has been growing for quite some time. Remember when you could still get 0-wait-state computers? That's a looong time ago now. Currently it's the memory subsystem (including caches) that is throtteling chip performance
Now: When you ramp up chip speed, you usually have to compromise on both cache size, and cache complexity. This is why a lot of chips (say the Alpha design) has direct-mapped level 1 cache, instead of n-way associative. Note that the P4 has smaller L1 cache than the P3.
So what can you do? If you want to design an architecture that will last for years, and possibly ramp to approximately 10GHz, you focus on getting the memory subsystem right first. You might even sacrifice performance pr. clock at "low" clocspeeds (by todays standards, that is), in order to insure that the basic design will last you a nice, long product cycle.
Thus the P4: Great memory bandwith, but less impressive performance pr. clock. Supporting evidence: Note that all large minis and supercomputers have really really REALLY exotic memory busses, focused on extreme throughput. We'll see more of this. Trust me!
16 IRQ's. Anyway.
May I ask what sound cards use two IRQ's? I've seen enough sound cards over the last seven/eight years or so to determine how they "usually" behave, and I cannot think of one that uses two IRQs. Sound cards (especially Creative Labs cards) historically tended to use two DMA's, which are even more scarce than IRQ since you only got eight.
In fact, the only devices I can think of that use multiple IRQs are dual-chipped video cards, serial controllers - ATAPI controllers (multi-device controllers in general), and those funky dual-channelled ethernet adapters. Most of what I just listed gives you twice as much functionality for two IRQ's, so it can be said that you really are getting your two IRQ's worth.
Anyway, since when has the 16 IRQ limit been a problem? Back in the ISA days there weren't enough types of devices to even fill up all the slots, let alone IRQs, and with the advent of PCI, IRQ sharing became possible. Now with USB, Firewire, et. al. all the devices are packed on one bus which communicates to the computer via one interrupt (Or is it two? I don't use USB or 1394).
Since I shut off my serial and parallel ports, I'm only using about half my IRQ's, and I don't use SCSI.
Media players (+ codecs), games, versions of directx all have very short release cycles.
Who is going to rush out to support SSE2 instructions for a chip that isn't likely to sell very well?
Microsoft, GNU and Borland will. They make the compilers.
Also, to use SSE2 and the P4 to it's potential, you have to upgrade EVER SINGLE APP on your PC.
Are you saying that "EVERY SINGLE APP" on your PC uses FPU instructions? I don't think so.
Hands in my pocket
While this is quite true, the companies that choose to pursue their goal of making money through sleazy licensing tactics and anti-competitive business tactics are wrong to do what they do.
I would quit a company that pulled those kinds of stunts. I very nearly did once when a company I worked for contracted out to have their stock promoted via spam.
There are many ways of making money, but if a company chooses to make money by some means other than providing a superior product, excellent service, or some other strong benefit to their customers, their a bad and evil company. Them being a company does not automatically free them from being held to a moral standard for what they choose to do.
The reason I gave up on Intel about 3-5 years ago is that it was becoming apparent that they were relying on marketing and sleazy business tactics over engineering excellence. Until they change, I will continue to do so.
Need a Python, C++, Unix, Linux develop
I don't know if Linux fully supports ACPI as I only run it on the servers I remotely admin so maybe someone else can answer this one?
I think that's why we mostly hate Microsoft. I also remember having a serious dislike of Lotus, Word Perfect, and Novell in their day (not realizing the Microsoft juggernaught would eventually run them over back then...)
Competition is good for all of us. Variety in the marketplace is also good.
But it's a good point. How many Linux user out there use Linux on non-IA chips? No, don't answer. This is a rhetorical question, not a poll! :)
And, uh, P4 is still 32bit...
... :)
Only half-way. A lot of silicon is devoted to SSE2, which if memory serves, is 64-bit. As soon as you see how fast programs are(when they're compiled with a SSE2-aware compiler), you'll start to wonder if maybe Intel processors arn't really x86 any more
The Pentium was a major leap forward from the 486, bringing forward major speed advancements...
The real "speed advancement" for the pentium was its pipeline - sure, not much - but it allowed the Pentium to increase its clock speed dramatically(4-5 times), which ended up making computers faster. The P4 is expected to end at around 7 times its current speed. Not too shabby.
Again, the P4 and P3/Athlon are all 32bit...
See my first argument - the P4 can no longer truely be called "32-bit". And the P4 actually runs strictly 32-bit floating-point ops slower than the PIII(significantly slower), but those nice SSE2 instructions are supposed to be absolutely blazing.
before Intel actually brings something worthwhile out...
What do you think that "worthwhile" processor will be? Yup, you guessed it! The P4, with a different package and maybe a slightly tweaked core. This core is expected to go up to 10GHz. You think going from 1.4(actually, they'll be releasing a 1.2GHz version too) to 10GHz is just a "short-term filler"? What Tom was referring to was the chipset/socket combo as short-term. The P4 is going to be around for quite a while - they build their processors with longevity these days.
Dave
'Round the firewall,
Out the modem,
Through the router,
Down the wire,
Barclay family motto:
Aut agere aut mori.
(Either action or death.)
Don't get it wrong. It's not that P4 performs badly on IEEE
double precision encoding algorithm. It's that it performs
badly on any x87 code. As soon as the same algorithm
gets optimized for double precision SSE2 equivalents P4
will get back on the playground. Of course I can't state
how well it will perform, but SSE optimized results of
Quake3 give a hint about it.
Leaving underpowered floating point unit could be also a
relatively smart decision on the Intel's side. This forces
software developers away from outdated stack-oriented
x87 architecture, so in the end we'll have all of the
software developed for much more optimal SSE.
Still, don't buy P4 just now...
BTW, does anybody remember that the original Pentium
was for some time codenamed P5 ? Talking about "P4"
now seems a bit funny.
Note that in many of these cases, one big argument which always favored the newer chip design was greater upgradeability. 486-100 vs. P60: about the same, but you can upgrade the Pentium to a faster Pentium later. PPro vs. PII: The PPro socket is a dead-end, slots are the wave of the future (ha!).
In the case of Athlon vs. P4, which is more upgradeable? The Athlon by far! The P4's successor will use a different interface. Basically, upgradeability is limited to getting a faster P4 (is the P4 even SMP capable?) In contrast, if you buy an Athlon now with PC100/133 ram, you can move up to DDR and/or SMP later.
The 386 (faster than a 286, but oh so expensive, and no one uses 32 bit apps yet anyway)
And no one did use 32 apps for years to come, however. The 386 was a revolution, anyway; almost the entire industry recognizes that. Bad example.
The 486 (who needs a math co-processor? Geez it's expensive)
You'd only believe you didn't need a math coprocessor if you listened to Intel's marketing. Ever buy a game? Nothing is more disappointing than buying a $2000 hunk of metal and seeing that "Sorry, this software requires a math coprocessor to run. Please contact your manufacturer."
The Pentium (Gosh 486's are available with the same or higher clock speed)
Same with the 386. The Pentium was hailed as the Macintosh killer when it was released, and rightly so since it was the first chip from Intel to even rival the decaying Motorola 68k line.
The Pentium Pro (16 bit apps actually run _slower_)
The PPro was viewed in the same exact manner as the P4 is viewed now. I'll give you that.
The Pentium II (oh, bummer, L2 cache is at half-speed, PPro is so much better...)
Ditto on this one, except by this time, it was 1997, and most people were using 32-bit apps by now.
Nah....I'm more like Tim-the-Toolman-Taylor - I want more power! When you start spending all day compiling source code or running massively parallel simulation code, you'll understand why.
8088, that was a chip.... *sigh*
MY first PC had one.... those where the days...
if (!signature) { throw std::runtime_error("No sig!"); }
Yeah, for applications compiled with those, but how many applications will be?
Look at all those nice Proccessor extensions and how long it took for everyday applications to benefit from them.
No, i don't say those extensions are useless, but it takes time until you can get that performance boost and if SSE2 catches on you can be sure that AMD either licenses it or offers something similar.
"By the way if anyone here is in advertising or marketing... kill yourself." -- Bill Hicks
Guess what? If they used the 1.2 GHz Athlon, it would have been faster than the 1.5 GHz P4 at all but the memory test! So, this link doesn't prove anything, it's bad news for Intel, actually, because the SSE2 instructions did NOT help the 1.5 GHz pentium to beat a 1.2 GHz Athlon!
Thanks, I'll use this info to prove my point in other discussions!
Sigged!
Well I was wondering when the first revision of this kind of the MPEG "benchmark" would come out. MPEG encoding/decoding is dominated by 2 or 3 very tight loops.
Which CPU looks good will depend far more on which architecture you optimised for than any inherent strengths weaknesses. Case in point. The P4 suddenly looks "bad" when Tom switched to stack F.P. based iDCT routine. Well this frankly is mere luck. You don't need F.P. to do an accurate iDCT. If the FlaskMPEG guys had used a good MMX iDCT (it *is* possible!) istead the P4 would have stood there like the MPEG CPU to end all others.
Instead its suddenly a lemon.
Acutally, I personally think Intel blew it with the decision to go to a super-long pipe. Quite a few codes *are* branchy and not all branches can be predicted. Period. The P4 always will be a brittle performer. Good on f.p. crunching with SSE and some kinds of "multimedia" stuff. A total lemon for other codes. Given the current trend to off-load a shed-load of the f.p. work to GPU's I think Intel made a bad call...
However they do deserve kudos for finally having the courage to side-line the horrible stack f.p. and put their effort into SSE2 instead (with far better potential). It think we'll see some really good f.p. numbers as the SSE2 compiler support cuts in.
Andrew
Since the P4 costs as much as two of the 1.2GHz Athlons wouldn't it make more sense to compare the P4 to a system with the AMD 760MP chipset and two of the DDR Athlon 1.2GHz CPUs?
Where can I buy the 760MP chipset? Oohhhh yeah -- it won't be available until 2H01. Heck, the 760 chipset which all of the current benchmarks are against isn't even released yet.
The P4 will not come with a multi-CPU chipset anytime soon. In fact, the P4 right now and in the next few months will definitely be a no-MP tool.
This is incorrect.
re: Shorting a stock you have INFINITE liability yet the ability to make at most 100% back you actually have infinite abilities either way (well, not infinite, but close). If I have my sell price at $10, and then end up buying the stock at 5 cents, then I've just realize a 10,000% return. a far cry from "just" a 100% return. And, about the ability to lose money... If it goes from $10 to $1000... Then if I shorted 1 share's worth, then I invested $1000, and lost $990. So, I've lost 99% of my investment. %'s can be tricky...
You'd think this logic would apply to Microsoft but people never think of them that way either. They always call them M$, Microsloth, their product Winblows, Windoze...
.sig: Open Source, Open Mind
Yes I would compare 5 escorts against 1 porsche.
They have a job to do of carting bits of data back and forwards. The five escorts can act together all carrying their data together.
In addition to this everyone knows that porsches have got small boots (trunks) and so they may be fast but have no carrying capacity (bus width?).
Cheers Xander
Slashdot Beta should die a painful death.
Yes, but Intel will certainly encourage developers to do this. As a lost of people gave said already, the x87 is a terribly obsolete piece of crap, and most developers will be happy to get rid of it. This will lead to faster acceptance, at least where it matters.
The Pentium 4 will sell well, there's no doubt about that. The PIII can't go any faster than it does now, so those looking for a faster MPU will have to go for the P4. Corporations and consumers are so used to "newer is better" they'll buy anything that Intel gives them. Haven't you seen the advertisments for 1 ghz PIII boxes with 64 MBs of RAM and TNT2 M64 cards in them?
What about AMD then? Well, AMDs fabs are already running at full capacity - there just aren't enough Athlons to go around. Also, the Intel brand is still incomparably bigger than the AMD in the real world.
Most of your apps don't stress the FPU. The areas that really need to be rewritten/recompiled are DirectX and vidcard drivers. People upgrade these a lot more often than their apps.
I really don't know why the P4 FPU is so very bad. I guess the PIII FPU doesn't scale to the kinds of clock speeds Intel are aiming for with the P4.
Don't take this post as a recommendation to buy the Pentium 4. You'll have to be a fool to buy one until 2002, especially with the dead end nature of the Willamette/Socket423. Of cource Intel knows there's no shortage of fools...
If what you say is true, then i should be able to take an R3000 and with a sufficiently small die size and make it go 3 ghz.
But I can't. If I could do that, why did SGI break the elegance of the R3k when they introduced the R4k ? The 4400 ? The un-scalable R10k ?
Remember, when you add more pipeline stages, the amount of effort you put into designing the stage, and the amount of effort you put into designing the interlocking, goes up. You pay a much higher penalty on _every_ instruction that deals with memory ( on the x86 this is quite a lot of them). There has to be some compelling reason to do this.
Why aren't people _always_ making caches run at the same speed as registers; after all it should just be an issue of shrinking the feature size, right ?
And while you point out correctly that changing the number of pipeline stages wont help a single instruction make it through faster, pipelining has always been about amortizing the cost of the entire pipeline over as many instructions as possible, so the single instruction case isn't really relevant.
The other advantage of increasing the number of stages is that it _does_ allow a clock speed increase. All other things equal, one instruction still takes the same amount of time, but if you double the pipeline length and then double the internal clock, a million instructions take about half the time as they used to.
Finally, it should be pointed out that a given design has a realistic feature-size range and mhz range. Right now the p4 is in a spot where its featuresize and mhz are _below_ the "break even" point for the architectual complexity it introduces over the PIII/athlon. But the P4 only has to hit 2.5ghz before its 20 stage pipeline is "as good" for a single instruction as a 10-stage PIII at 1.25ghz. There of course is no doubt that hte P4 will hit 2.5ghz. And remember, the entire time intel is shrinking feature size and cranking clock rate on the P4, the total instructions completed in unit time is rising.
So the "sweet spot" ghz and feature size for the P4 architecture haven't been reached yet. Until they are, expect the P4 to comparatively dissappoint. But dont expect a PIII or Athlon to ever reach 2ghz.
My opinions are my own, and do not necessarily represent those of my employer.
Intel NEVER has enabled SMP on the first stepping released of a CPU (and AMD and Cyrix have never released a commercial SMP capable platform). The P4 will support SMP probably sooner than any new Intel processor ever. [Here's a clue: the ZDNET article which stated that SMP wouldn't be supported until 2H01 is out and out false].
Actually I thought it was interesting how slow the P4 was EVEN WITH these optimizations. On most of the scores it only narrowly beat the 1GHz Athlon, so would probably have narrowly lost to the 1.2GHz version.
It really does appear that we're going to have to wait for Northwood and the new socket, and/or see whether Intel can ramp up the clock speed faster than AMD can before the P4 will offer much.
The reason the P4 doesn't support SMP is simple ... HEAT!
Some clues are in order.
The 1.5 GHz P4 puts out 30 degrees of heat. The 1.5 GHz Athlon is projected to put out 95 (!!!!) degrees of heat. Yep, that's only five degrees less that the boiling point for water.
You're only looking at it from an architecture perspective and not from a circuit perspective.
One thing that every P4 reviewer (even the haters) have remarked is that the P4 runs extremely cool, and is extremely overclockable.
The P4 @ 1.5 GHz runs at LESS THAN ONE THIRD of the temperature of an Athlon at the same speed (30 degrees vs. 95 degrees). This says, of course, that the P4 has dealt with the heat problem already, and thus has SIGNIFICANTLY more head room to increase speed, since it puts out so much less heat.
The Athlon is going to hit a speed bump because it puts out so much heat. You can't sell a processor which takes 95 degrees, so they won't even be able to do 1.5 GHz unless they radically modify the core (... Palomino is supposed to; we'll see).
It looks like you just doesn't get his point - that every new Intel processor family has been met with criticism. Later, it usually died off.
The P4 isn't that much faster than the PIII for many applications, but the architecture gives room to increase the frequency a lot more - which will give it enough speed to outspeed the PIII on all tasks and compete with the Athlon, something the PIII wasn't able to do anymore.
It doesn't seem to run most common apps much faster, but those really doesn't matter much - most office tasks aren't even remotely CPU bound. P4 does introduce SSE2, which has the potential to speed up many operations when utilized. And it vastly increases memory bandwith from the PIII, which has been a chokepoint
Would I buy one now? No. RAMBUS memory is way overpriced, and from a company I detest. The chip will go to 2 GHz and a shrink soon, DDR-supporting chipsets will be released - at that time, it might be a good alternative.
Yeah right, and pigs might fly too.
I think the more appropriate and convincing way is to lobby against and boycott the unethical corporations.
--
when everyone gives everything,
when everyone gives everything, then everyone everything will get
Is the Pentium II a descendant of the P-Pro? I thought Intel dumped the idea of doing CISC to RISC conversions. Obviously the PII takes some of the ideas of the P-Pro, but wasn't the defining feature of the P-Pro the fact that it was a RISC chip masquerading as a CISC? And if so, is the PII like that?
Pretty much all modern x86 chips, not just Intel's, are implemented this way. CISC instructions are very difficult to pipeline, which is why this technique was introduced (with one of the VAX chips, IIRC; neither Intel nor AMD nor NextGen invented it).
1) Because it started acting like a monopoly (see: its attempts to bully board makers from making Athlon-compatible motherboards);
2) The way Intel treats its employees;
3) The product serial number blunder.
Also look up: wintel.
========================
63,000 bugs in the code, 63,000 bugs,
ya get 1 whacked with a service pack,
--- Grow a pair, liberals... stop letting the Republicans bully you!
Tom's ego compares favourably to the heat sink on a P4, the power supply for an Athlon, the Mac LCD cinema display, the Razor Boomslang mouse, old HP laser printers, and IBM's "wing o' death" keyboard.
Which is to say that it's fucking enormous.
Tom, my man, I think you need to chill out. Start popping some Paxil. Quit taking yourself so fucking seriously. You're just a little shit in a big pond. You don't make or break the hardware world.
--
--
Don't like it? Respond with words, not karma.
From what I understand, there are four holes in the P4 motherboard through which mounting posts (part of the case) rise; the heat sink is then bolted to these posts? Simply clipping the heat sink onto the motherboard as has been done with past architectures won't work because the unit is too heavy and causes significant bowing of the board.
Mind you, this is all second-hand and several months old from a friend inside Intel, but a heat-sink that is too high to fit into an ATX case would be a monster indeed...
STOP . AMERICA . NOW
Aye, Tom, Kyle and Anand colaborated to use Toms Linux compilation benchmark on the three chips they had between them. As I remember none of those chips would complete the test.
Trust The Computer, The Computer is your friend.
You have to use them if you want to stay in the business. You've got rivals, you know. So you have to use whatever means possible to ensure the prosperity of your company. Once you have that, you can use existing funds to provide better service.
There's no such thing as an evil company (hehe.. not even M$), it is just an organization of people good or not very good at something trying to make money. That's it.. It is not the end of the world, they are not trying to make something evil.. I am not talking about drugs or nuclear weapons, just in general. People make money.
http://dtum.livejournal.com
You are neglecting the fact that yield is dependent on die size. Therefore your "eventually" will depend on die size and you can easily run into the problem that your design is outdated (meaning you can only ask a low price for you chips) by the time your yield is acceptable.
This is true; however, the limit is still pretty high, because part of the development of a new linewidth is tweaking of the process until adequate reliability for large dies is obtained. The customers want to be able to produce large, fancy chips, so the fab houses tailor their processes accordingly.
You can also get a bit of leeway by using more conservative design rules when laying out your chip, though this usually has the side effect of making the chip slower.
Note: All of this just applies to the size of the core, not the entire die. Most of the die is cache, which is very easy to build in a fault-tolerant manner (more rows are included than are needed, and during initial testing, faulty rows are permanently isolated from the circuit).
There is a considerable amount of research on building chips that can function despite design or fabrication faults, which should make the problem much less severe in the future (yet another thing that I have to study as a degree project). This is needed not because of the yields, but because designing a billion-transistor chip without making critical mistakes may not be practical.
just do a google or pricewatch search for them... not exactly ubiqitous but not 1.13Ghz vaporware either. A month after the 1.13 launch a total recall affected less than 200 users..
If voting were effective, it would be illegal by now.
You can bet your sweet ass that BeOS will be one of the first systems to support it - a majority of the BP6 crowd (which run dual Celerons under BeOS) will flock to the AMD 760MP like there's no tommorow. I for one will get this baby as soon as it hits the streets - in a dual Duron configuration. Cheers.
Revolution = Evolution
If I recall correctly, one of the bigger challenges Intel faces is in retaining skilled chip designers.
I'm under the impression that their best and brightest designers have fled the company, and they're now left with newbies who have simply never worked on anything approaching the scale of these CPUs.
It's an interesting problem, come to think of it: the only way you get that sort of expertise is to progress through the chip designs. You start off designing 8086, get involved in designing 80186, 286, 386... eventually, you're an expert at designing large CPUs, because you've been chiefly responsible for designing increasingly larger CPUs.
When you kick the bucket, how's that runny-nosed kid fresh from tech school ever going to cope with developing the next generation CPU? Poor little bugger hasn't ever designed a CPU at all: his training was all theoretical, and perhaps a few class projects designing variations on the 555 timer.
In all likelyhood, it's going to become one helluva problem within the next ten to twenty years, as the old school designers, who cut their teeth on simpler CPUs and were key in the development of more complex CPUs, die off or retire.
--
--
Don't like it? Respond with words, not karma.
Actually, inside Intel they call such intermediate chips as "enablers" because they "enable" manufacturing of new chipsets and related peripherals. Then the actual "new" product is released on top of the already existing infrastructure.
Same for P4 - it's an enabler.
Donate free food to the hungry at The Hunger site.
Try the other one: Server 2
What I find interesting is that when the first versions of various Intel CPU's came out, they didn't really make much sense.
But once Intel was able to quickly speed up the core CPU speed, then it did make sense. Remember the original Pentium 60/66 MHz CPU's? Everybody complained about the cooling requirements of those CPU's, but once Intel switched to the Socket 7 design and went from 75 MHz all the way up to 200 MHz, THEN the Pentium CPU's became very desirable. The same with the original 233/266 MHz Pentium II's; it was not an improvement over the Pentium MMX 233 MHz until Intel sped the CPU to 333 MHz and Intel introduced the second-generation PII's that supported PC-100 DIMM's. The same also applies with the Pentium III, which started at 450-500 MHz, but didn't become desirable until Intel sped it up all to 600 MHz (Katmai core) and 1,000 MHz (Coppermine core). (By the way it appears that Intel has finally licked their 1,000 MHz PIII production problem; it appears that supplies of the PIIIEB FC-PGA variants up to 1,000 MHz are fairly plentiful, if a bit expensive.)
Of course, Intel needs to quickly ramp up new and better CPU technologies soon. The current AMD "Thunderbird" CPU's are more than a match for the PIIIEB, especially with the new DDR-SDRAM technology. With new, cooler-running Socket A Athlons comimg in the early spring of 2001, AMD could crank up the speed of the CPU to as high as 1,700 MHz, which when combined with DDR-SDRAM could mean AMD can in many ways keep up with the Pentium 4, but at much lower cost. And with the Athlon likely supporting SSE2 instructions in the second half of 2001, a 1,700 MHz Athlon with DDR-SDRAM could do everything the P4 could do but possibly faster and definitely less costly, too.
Raymond in Mountain View, CA
Sure, if you could get latency for free. Only you can't. We are beyond the point where anything but perhaps general improvement in gate switching is an unballanced good. This is the same thing as trying to decide the optimum cache-line length. This is a function of many variables, and can't be decided in an isolated fashion.
Actually what I was saying implicitly in the original post was that the latency gap between CPU and memory is growing, so you'll have to optimize for either lower latency gap (= better caching, slower clock, smaller latency-gap) or higher clock (=simpler caching, higher IPC in core, larger latency-gap, longer cache-lines). If you go for higher clock, you want to optimize for bandwith, as this is more or less the only option.
RAMBUS is generally higher bandwith/higher latency than normal DRAM. There really isn't much we can do about the latency-gap, except perhaps going for avanced RAM architectures w. more intelligence in rows, or even individual cells, which costs transistors, which reduces storage capacity etc. etc.
That, or a whole new way of fabbing chips and/or memory. There a quite a few promising alternatives out there, memory-wise, but they are all experimental, so don't wait up
"Let me open these blinds so the snipers can see in." - Kevin Giffhorn
If what you say is true, then i should be able to take an R3000 and with a sufficiently small die size and make it go 3 ghz.
But I can't.
Sure you can.
Later chips just add features that the old chip didn't support, or redesign functional units to work more efficiently. Thus, as they're more useful and silicon (below a certain area threshold) is cheap, the later chips are used.
One example: The R3000 has an in-order pipeline. The R10000 has an out-of-order pipeline. This means that the R10000 can keep on crunching in cases where the R3000 would be stalled.
Other differences exist, but I'm having a surprising amount of difficulty finding documentation on the MIPS cores' features on the web.
What is more; before the reader's correction, Tom's benchmark results were for P4's advantage.
Donate free food to the hungry at The Hunger site.
increase its clock speed dramatically(4-5 times)
Well, as much as this whole 'fiasco' reminds me of 'fiascos' past that really weren't, I have to put in my 2 cents. I have seen the future of PC's and I'm, honestly, wondering what-the-fuck.
Has anyone heard about the heatsinks on these things? New power supply? Reminds me of a Voodoo5. In case you've never seen one, a Voodoo5 is a full length AGP card that needs to be plugged into a hard drive power source. These things are meant to be the best of the best, and instead they just run pretty fast and run hot as hell. AMD isn't a whole lot better.. The shop I work in recommends that all Athlons (.9GHz+) be equipped with a 300W power supply.
What's going on here? Whatever happened to smaller and cooler? I thought some of the best 'geek' wants were just those.. Webpads, wireless, laptops, etc. These things are none of those, and also not cheap. People like the idea of cheap, small, portable and cool. (Starts thinking about Snow Crash and Diamond Age.. Mmmm...) Instead we're basically buying these things and letting Intel know that we don't actually mind if our desktop systems are getting bigger again and starting to look like older style computers. Yeah, the downward trend of software doesn't help too, I know, but what's wrong with making what we have more.... accesible.
Yes good point... heat density is much more important than total heat... L2 cache has very low density, but all that means is that your FMACs are running really hot...
I must burn in hell, suffer and pay for my sins
But Gods the one who's losing, Satan always wins!
The Athlon is going to hit a speed bump because it puts out so much heat. You can't sell a processor which takes 95 degrees, so they won't even be able to do 1.5 GHz unless they radically modify the core (... Palomino is supposed to; we'll see).
Actually, linewidth shrinks can still be done. Power dissipation is proportional to the capacitance being charged and discharged, which is proportional to the area of the core (I'm ignoring the cache, which can be partitioned so as to scale without additional heat generation). For a given core layout, power dissipation at a given clock rate goes down as the square of the linewidth.
As clock rate is also determined by capacitance, it goes up by at most the same amount, resulting in a worst-case power dissipation the same as before the shrink - with a processor that runs much faster.
That having been said, reducing heat production will still allow you to increase the clock rate (you'd raise the clock speed, which in turn might require raising the core voltage, until the power dissipation was again at your maximum acceptable threshold).
As for the P4... Part of the reason it runs so cool is that it has a heat sink the size of Alaska sitting on top of it. What are its actual power consumption figures? These are a better basis for comparison.
>when the optimising compilers come
>out supporting SSE2
...but then again P4 with 2GHz will be out and the *overpriced* P4-1G4 you bought 6 month ago will be *old*!
Intel should get the compilers out *FAST*. And even if it's just a reference-design without bells and wistles it will impress the testers and you could expect your P-4 optimized mpeg-viewer much earlier.
With the current practice of having just few optimized apps (I do not play quake for example) to show only few people will spend the extra money.
Besides, someone needs to take a shot at those blue Intel whatever they ares.... everytime I see those ads I want to puke. What? You gotta be kidding. I'm no fan of Intel, but those are are a masterpiece. Very funny. just a question; does the average person who would see those ads actually know who Intel are and what they do? and the difference between a pII and a pIII? Or the difference between Intel and AMD?
------
Shhhhh....there's a dead body in my trunk.
Wanna see it? Fuck around and you'll be it!
_____________
this is a sig
I don't think you understand you chips achieve low power. It has very little to do with the linewidth, and everything to do with power management.
Um, I've spent the last 5 years learning how to build chips. While power management is important, total area - absolute area, not number of transistors - is also directly tied to power dissipation.
The power dissipated is simply the clock speed times the square of the core voltage times capacitance that is charged or discharged per clock.
If you optimize your chip so that only areas that are being used are clocked, you save power. This is what you were referring to.
If you lower your core voltage, you save power.
If you reduce the total area of the chip - by applying a linewidth shrink, for instance - you reduce the total capacitance (by a factor of 2, usually), and save power.
Thus, a linewidth shrink would most certainly allow an Athlon to run faster for the same power dissipation.
Also, saying that "the P4 should dissipate more power because it's bigger" isn't strictly true - all of this applies to the size of the CORE, not the size of the CACHE. The cache can be optimized to dissipate pretty much the same amount of power no matter what its size, as in any given clock, you're only accessing one or two rows of it. It's the core that's changing state all the time.
The relative sizes of the Athlon _core_ vs. the P4 _core_ are what would be important for your argument.
True, yes, but I'm morally against overclocking.
BS. dual channel PC800 RDRAM has 2x 1600MB/s badwidth whereas DDR has 2166MB/s. Because of this the rest of the argument is also BS.
RDRAM is at least 3x more expensive than SDRAM (depending on where you buy it). And it will not get any cheaper. On the other hand DDR costs only marginally more than SDR, and once they ramp up the production it will be same price. The marginal performance improvements RDRAM provides are not worth 3x the price. That's why even Intel is ditching Rambus and goind with DDR. That's also why you'll never see RDRAM based chipset from AMD.
___
___
If you think big enough, you'll never have to do it.
You missed it, tom was bashing slashdot in that article when it first cam out but today he deleted what he said about slashdot, wish I had it to post it here. It was on the first page, and suspicously, slashdot is linking to the second page of that article now.
Sneakemail is to spam filters what an ounce of prevention is to a pound of cure.
No it doesn't in 2.2, but does in 2.4, so it will probably be in the next round of distro releases.
Since the P4 costs as much as two of the 1.2GHz Athlons wouldn't it make more sense to compare the P4 to a system with the AMD 760MP chipset and two of the DDR Athlon 1.2GHz CPUs?
Has anyone seen such a comparison?
Blah. Who-or-whatever gave you that opinion. US is the only state where reporters actually ask questions like 'but-what-if-you-did-this?' or 'wouldn't-it-be-better-if-you-would-..?'. CNN not opinionated? Don't make me laugh, CNN=news for wealthy WASP living in USA, you just dont notice since you must fit the description. ok, i'll drop the W there.
Latest crack remark during CNN:
- anchor woman: It seems that over the world people have problems grasping the electoral system.
- anchor man: Well, I guess we gave them a lesson in democracy.
Whoah, and that for a two-party biased old system where the assumption is that people were/are to stupid (read to black) to elect their president directly.
Go check your ideas... Better, go watch BBC instead of CNN...
I have a cool sig too.
We don't have to be loyal to any company, just because of what they've done in the PAST. If they can't keep up the superior quality, then we, as consumers, will support the company that producest that product of higher quality.
Or simply, the consumers control the market, and if AMD is nicer to the consumer, the consumer is gonna be good to AMD. Same with Intel. Whoever pleases the consumer the most wins (for the time being)...
I don't see why I should remain loyal to a company that is starting to screw me over as a consumer. Forget that!
So, for that reason, I won't lay off Intel until they have produced a product superior in quality and value than that of its competitors.
Simple as that.
What could possibly hurt the security of the American people more than giving our own government the ability to hide its
A common thread of all these reviews is that the regular FPU on the P4 is pretty ordinary (ie, worse than on the P3), but the SSE2 instructions are great, and will speed things up immensely, if they're used. Hence the excellent specFP scores.
My question is: given that few developers these days have the mental fortitude to write hand-optimised assembly, and given that Intel's C++ compiler alleges to be able to vectorize loops to make use of the SSE2 instructions - why are they charging $500 for it?!?! (OK, it's only $400 during December).
Surely it is in their best interests to give the damn thing away, so that developers everywhere are releasing code optimised for the P4? (surely AMD should do the same, but maybe they don't have an army of compiler-writing gurus like Intel presumably does). Graphics card vendors, for example, are always exhorting you to "get the latest drivers", to improve the performance and stability of systems using their hardware. Intel should view this compiler like a driver for their CPU - get everyone (who compiles) to download it, and it'll make their hardware look better.
I have this feeling everyone's against Intel and is making sure this viewpoint gets through.
/.'s point of view.
People openly say "I want Intel to crash and burn" ALL THE TIME!, even though this isn't Tom Pabst's or
Most of you who curse Intel are hypocrites. You'll be buying Intel processors if they come back and saying "Intel rules".
You seem to disconsider Intel has developed good products and technologies despite its failures. Please don't send me Intel's top 10 (or 50, or 100) top mistakes list. I'm well aware of those.
If you think Intel charges way too much for their processors (and they do), fine, just don't buy them. You shouldn't run around screaming antipropaganda simply because they're on top.
A similar phenomena has happened to 3dfx and Netscape. They also have made bad, wrong decisions. But haven't they also broken huge amounts of ground?
So lay off Intel a bit. Cut them a little slack. Do you think RAMBUS's debacle and design difficulties all over are done on purpose?
Here's an analogy for you guys (you say if it's bad or good). Most of you (the ones from the US, at least) sometimes agree with Jon Katz about the geek kids who are made fun of by the jocks.
Intel started playing really bad american football for some reason and is like a geek kid. Don't kick them when they're down, at least not that hard. You have your reasons, I know, but enough is enough.
Flavio
P.S.: I don't work for Intel, never have and don't expect to. I wouldn't mind to, though.
Wrong! If you going to spell or grammar flame, at least make sure that you're not wrong in the particular issue you're flaming about... Examples of correct sentences:
The illegal we do immediately. The unconstitutional takes a little longer.
--Henry Kissinger
looks to be a big step towards a different model of processor (deep
pipelining, sophisticated branch prediction) which whilst I agree in
the long run is probably right, in the short run it might be a long
time before it becomes an improvement on current technology. One
could say the same about Rambus...
And if we are in the business of backing predictions about which
will be the best architecture in the long term despite less than
stellar short term performance, why should we believe the P4
architecture is better than that of rival VLIW architectures (eg. the
Crusoe)? Following Intel's lead has been the right thing to do whilst
Moore's law held, but now it rather looks broken...
When yields approach reasonable ranges, as they always do eventually,...
You are neglecting the fact that yield is dependent on die size. Therefore your "eventually" will depend on die size and you can easily run into the problem that your design is outdated (meaning you can only ask a low price for you chips) by the time your yield is acceptable. Yield is only approaching reasonable ranges fast because the manufacturers choose the die size accordingly.
You are right that the cost of the raw material is not the dominating factor for a chip, but that doesn't mean that the cost doesn't depend heavily on die size.
Perfect, no. But still it is reasonable to expect corporations to honor their responsibilities to their shareholders, their employees, their community, and to the public at large.
The other advantage of increasing the number of stages is that it _does_ allow a clock speed increase. All other things equal, one instruction still takes the same amount of time, but if you double the pipeline length and then double the internal clock, a million instructions take about half the time as they used to.
Increasing the number of stages only allows clock speed increase if the individual stages get simpler, i.e. do less work. If you split one stage into two that do the same work and double the clock you gain no direct speed increase. And that's exactly what we are seeing in the benchmarks right now. What you get with longer pipelines is more granular control of instruction execution and this might help to optimize the microcode.
The P4 isn't a chip for you and me. Wanna know why?
* In almost all kinds of applications, it is slower than an Athlon T-Bird 1.2 Ghz, and that's from a P4 1.4Ghz. Even overclocked to 1.7ghz, it's still slower.
* Almost all applications - meaning pretty much everything involving a floating point unit, including CAD, raw calculations, Office apps, and Unreal Tournament - are slower than on the lower-clocked and cheaper Athlong. Oh, and I forgot: It is atrociously slow compiling anything with gcc.
* The much slower P3s actually beat it in speed at many real-life applications.
* Tom's review compares it encoding a long DivX movie in high quality with a 1.2Ghz Athlon. The P4 needs twice as long at some tests.
* You can get a 1Ghz Athlon for less than $300 in some places, with Athlon prices dropping weekly. A 1.4Ghz P4 will cost around $1000. Prices won't be dropping anytime soon.
* The P4 needs a new socket, doesn't always play nice with all types of memory, its socket is of course incompatible with everything, it needs gigantic coolers which NECESSITATE new cases, where old cases are simply too narrow. That's right, many old cases (ATX format) simply won't take a P4+cooler.
* The P4 will not come with a multi-CPU chipset anytime soon. In fact, the P4 right now and in the next few months will definitely be a no-MP tool. MP Athlons are just around the corner, and so is the 266mhz FSB Athlon chipset for use with superfast DDR memory. Rambus, anyone?
And if you read the reviews, the only thing it's actually faster than the Athlons is at Quake3. Seeing how many buying decisions are made by completely irrelevant Q3 scores, this may be a very bad thing.
And yes, the incessant pro-AMD propagande isn't good, but have a look at face intel to see why intel really isn't a good company. Maybe that will explain some of the hostility.
Alex T-B
St Andrews
Ok, ok, this is the second time I have heard the "Pentium 4 runs at blah blah blah degrees..." Enough! What temp a chip runs at is dependent on cooling, case airflow, etc, etc. If you want to try a REAL number on for size, how about Power Disipation? Athlon 1.2 Ghz: 54 Watts. Pentium 4 1.5 Ghz: 52 Watts. (See Anandtech.com) There, doesn't that feel better? :-)
Laugh at stupidity: mod idiots +1 Funny.
Dump the x86 while youre at it, and you've got yourself a G4/450 MP :)
Intel paid Carmack to include SSE optimizations in Quake 3. That delayed the lauch of the game, but I guess the money was good enough.
___
___
If you think big enough, you'll never have to do it.
That was an incredibly acidic review.
... its a marketing game now.. and Intel will prolly do *OK*
Most of it is really true, people can claim toms is PRO-AMD all they want but...
I havent seen a single glowing review ANYWHERE, I mean come on not everyone can be Intel haters, unless theres a reason for that HMMM
I mean come on the whole P4 deal leaves a bitter taste in my mouth
I can go get a 900Mhz processor a thunderbird and make a whole system that outperforms a P4 for about the price of a p4, what gives?
I do know that toms is a little pro amd, but hes not a stupid person so there is prolly some justification for his bias, AMD actually DOES have good chips now...
Oh but people arent allowed to be biased, formulate your own opinions but I feel safe in trusting all the reviews that P4's basically stink.. anyways
I hope AMD crushes them just for the shit they are putting out
Jeremy
Not very many people compress video regularly; that's a niche market, not a major factor.
Actually, they usually buy BOTH cigarettes and lottery tickets.
Lighten up folks, the tag line was a joke. Its statistics. Get it?
What do you expect? That Intel is going to be perfect? It is a huge company, in any big company there's not only focus on delivering a good product, but there are paychecks, deadlines, taxes, legal issues, unsatisfied customers, lawsuits... Do you expect them to disregard all that and just try to make your life easier by doing miracles and pleasuring everyone?
It is brutal to compare a company like Intel with Mafioso clan. It just shows, that you are a geek that has this gut feeling against anything that's stronger than him and you don't accept that. What the hell is up with that? Grow up. It is a real world, someday you will find a job in an IT industry and will understand, that still the main motivation of any company is to make money, and not to deliver the best product ever. Get real. And as for mafioso, rent (if you don't own) a Godfather I and II. See the difference. Or even better, go to Russia, you'll enjoy it there.
http://dtum.livejournal.com
" And ethical salespeople (oxymoron) Wait a sec... I work at CompUSA... that makes me unethical?
"
This was meant as a joke, sorry. I used to be a salesperson, and it seemed everyone else selling PC's around me weren't ethical. One reason I got OUT of sales, and deleted my sales experience from my resume so no prospective employers would try to get me into sales instead of technical (despite my certifications and 7+ years experience).
If I were still a salesperson, I'd tell the customer the advantages of the Athlon, Duron, vs the Celeron/P4. Getting the customer the best deal is in the SALESPERSON's best interest because those people come back. All of Intel's marketing probably won't matter if salespeople all had the technical knowledge to explain the facts to customers.
This version of the P4 is a turkey, it requires a totally new case spec, and uses a motherboard that will be onsolete in 6 months. Dead end. It reminds me a lot of the origianl P60/P66, that used the larger socket and were 5 volts. People who got those were screwed because they couldn't upgrade to the new 3.3 volt socket 7 Pentiums.
That's another tidbit ethical salespeople would tell customers.
=== The price of freedom is eternal vigilance
Uh, like Compaq & Gateway?
love is just extroverted narcissism
Smaller, cooler, quieter. My CPU is plenty fast.
When is Intel going to dump the i386 architecture? The whole 15 IRQ thing is annoying! Yes, yes, I know that a lot of you will flame me and say "get a SCSI system," but some people can't always afford that kind of system (but I would love one!)
Most home users just buy their computers and expect them to work. Then they go off and buy all these extra periphials (that aren't USB) and eventually all the IRQs are taken. Sound cards alone usually take 2 IRQs! It doesn't take much to fill your system.
Apple changed in 1994 to PPC, cold turkey. Still a lot of the Classic Mac OS is in 680X0 assembly code but it's a fast fading memory.
While the PPC 7400 has been stuck at 500 Mhz for over a year it is still a mighty contender against the x86 family.
If the x86 world is unable to swallow the consequences of migrating towards a real architecture it will slowly but surely sink into irrelevance.
Well, it was bound to happen. I'm no AMD sympathizer but Intel kinda dropped the bomb with the release of P4.
I like Intel, I have a celeron II but Intel only released so they could be first. After they lost the 1Ghz race they wanted to "Aw" the public with the new P4.
In reality the P4 will be a great chip. I guarantee that. But now, it is in nooo way worth the money. Tom's optimizim was short lived...I think most people feel that way. The performance benifits are marginal right now...some tests even worse. But...
When they switch to DDR ram and make it SMP compatible with their smaller DIE, it will be "THE" chip to have.
Oh well, just one more pre-mature E*****tion (figure it out) by a major corparation to be the best.
Grubby
"CPU's Don't make mistakes....They just miss a few cycles sometimes..."
Having actually purchased a 286... I can actually remember the whole 286 -vs- 386 argument.. I remember my 286-16 kicking the pants off of a 386DX-16 running a couple of games. The 386 tended to have IN/OUT instructions that were nearly twice as slow (IN took 286 5 cycles and the 386 12 cycles) which caused anything that was doing heavy PIO to run much slower.
There were a number of other things that ran faster too.. Eventually though the 386 scaled to 40mhz (thanks AMD) while the 286 stalled out at 20mhz. This was more than enough clock rate to make the 386 faster at everything.
Since the P4 is slower than one Athlon in most benchmarks, pitting it against a pair of Athla is just mean.
--
All i have to say is...Give Intel Time. (The first new core design since the P Pro). Good luck to AMD.
"Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life." Terry Pratchet
To verify your Windows 2000 install is using ACPI and IRQ routing you can goto device manager and select the "Computer" tree. If it says MPS Multiprocessor or Uniprocessor then it is not, if it says ACPI Multiprocessor or Uniprocessor then it is.
This really suprizes me. I'm amazed that Intel would release a processor awhile after the development of the 1Ghz AMD that was a poor choice in the feild of vido compression. I'm not the only one to realize that video is comming and its comming quick in this world of streaming media. A processor that is only good for one proprietary algorithm with encoding is just plain rediculous. I really don't follow alot of the new hardware stories but this just can't be ignored, these issues will determine what processors we use in the future.
That's pretty much the situation I'm in right now, I'm stuck in sales until someone in tech services quits/gets fired/has an unfortunate "accident"...
"Let me open these blinds so the snipers can see in." - Kevin Giffhorn
to your comment:
Quite the opposite - in most cases, Intel's newest CPU architecture doesn't perform as well as what it replaces - at least for a while, until the compilers have been modified.
The 386 was cool bacause it could do some instructions in only one clock instead of several.
The 486 was cooler, because of more of the same, and a built in , if slightly slower FPU.
The pentium was supposed to have a FPU as fast as the 486, and added some pipelining to the mix. It also added the rest of the functionality to properly virtualize the 'windoze' approach to computing. Win 95 will run on a 486, but not as well as NT. NT is more deterministic, and doesnt get 'lost' as often.
Now, if you read the data sheets for amd's products, and how processes are pipelined, they're quite different than the intel processes they replace/emulate.
Unfortunately the K6-2 or 3 cache units are far superior to the athlon's, currently. probably ate up too much silicon at this time, but after the next shrink, it would make sense to return to the better algorythm.
All in all, these processors from the past superceded and brought new functionality to the software. It just takes awhile for people to write it into the compilers for the masses.
Anyone who writes assembler can use these capibilities today, and i think i saw the VC libs somewhere yesterday. knowing the net community, someones probably doing a recompile of Q3A for sse as we speak!
Truth isn't Truth - Guliani
Well according to a previous article at Toms, AMD will support SSE2 in future Hammer processors. Since both Intel and AMD will support SSE2 it looks like it is here to stay. I only worry a bit (or is it 16 bits?) about the loss of precision, since normal floating point uses 80 bits and as far as I understood SSE2 is only 64 bit.
Never express yourself more clearly than you are able to think. --Niels Bohr
After seeing all the reviews I'm starting to wonder whether Intel's real reason for bringing the P4 out isn't because it's more efficient, or because they've improved the architecture in any way or made any improvement over the PIII.
:)
Perhaps their only problem was that the PIII kept breaking when they pushed it up to ridiculously high clock speeds. So they moved things about a bit (no idea how, maybe increased the separation distance between the etched components), and possibly made it less efficient, but in the future able to be pushed up to 2, 3, 4GHz or whatever.
If that's right, then it's so not The Way Forward. There's a point when channeling most of your R&D into pushing the same old architecture faster and faster will provide fewer and fewer gains compared to actually designing a better one. I mean, how long is it since the first 64-bit processor now? It made me wonder how important it is for M$ that PCs stay on the 32-bit architecture for as long as possible, because of course it's much more painful for windows users to take full advantage of 64-bit processors (they'd need to get a whole new precompiled OS distribution) than linux users and the like who could recompie everything overnight.
I don't know. Maybe I've been reading too many conspiracy theories...
Dave
And, uh, P4 is still 32bit...
... :)
Only half-way. A lot of silicon is devoted to SSE2, which if memory serves, is 64-bit. As soon as you see how fast programs are(when they're compiled with a SSE2-aware compiler), you'll start to wonder if maybe Intel processors arn't really x86 any more
And the x87 compatible FPUs both the P4, athlon PIII etc uses got even longer wordlengths (80 bit ?).
Don't confuse the 'bitism' of a CPU with it's FPUs wordlength.
Traditionally a n-bit CPU is so labelled because it accesses memory in n-bit chunks. It was also at times used to refer to the width of the adressbus, but this have fallen into disuse in later years.
See my first argument - the P4 can no longer truely be called "32-bit". And the P4 actually runs strictly 32-bit floating-point ops slower than the PIII(significantly slower), but those nice SSE2 instructions are supposed to be absolutely blazing.
Hmmm... as far as I can recall there is nothing special about the SSE2 ALU (and there is only one). While the SIMD nature of SSE2 instructions can give a nice boost, it's not given that any application will benefit from them. Also this did leave the P4 with very weak x87 performance which is acutally widely used and will be present in legacy code at least 5 years more (speculation).
If the PPro can be used as a guideline I'd guess there will shortly be a new iteration of the netbus core with improved performance for x87 (much like the improved the 16 bit code performance the PII got over the PPro)
It seems to me that the P4 is at present a silly purchase. It's no good for SMP, it makes no sense to buy one in order to upgrade the CPU later when they get up to speed (the socket will change), and it's performance is uneven and a poor match for it's price. You must also use RDRAM which is still rather expensive (but does indeed seem to give a performance boost this time around as the CPU has improved off-chip transfer capasity).
It will take quite a while before P4 optimized applications will show up, until then the P4 makes little or no sense to buy.
As for the future performance of this core w.r.t attainable clockspeed I'd be carful with my speculations. Remember that we originally expected the merced (now itanium) core to be the next generation core after the P6 core. That we now got this Netbus inbetween does seem to indicate that this really is just a stopgap effort to remain competitive until the itanium becomes viable. This does of cource does not mean the netbus is _bad_, but it might point to a rather short life.
After all it seems we need to get away from the ridiculously complex decode/predict algorithms modern CPUs (Both x86 and risc) utilizes, however compiler tech is still a problem for VLIW (which othervise IMO has great promise)
ah yes i can see it now.. a beowulf cluster being used to play quake3 REALLY fast
maybe bill gate's ultimate goal is to dominate the world with quake and he's using intel as pawns to promote it?
I believe sex is highly over rated... unless it involves me
Even the most pro-AMD detractors of the P-4 seem to agree: The one thing keeping the P-4 in the race against the Athlon is it's incredible memory bandwidth. For once, Rambus isn't being gated by a 133MHz FSB--the P-4's frontside bus runs at 400MHz, which is PC800 RDRAM's "native" speed. Read any of the reviews on the P-4 and see if any of them are talking about RDRAM latency now. Dual-channel RDRAM delivers 3 times the memory bandwidth that PC2100 (266MHz) DDR-SDRAM delivers on the AMD760 chipset.
.13 micron should take them very close to 4GHz by 2003. I look forward to AMD reponse in the Clawhammer (and AMD first RDRAM-based chipset) by mid-2002.
This is why the P-4 dominates LINPACK, STREAM and all other memory-intensive benchmarks.
The irony is that Intel CEO Craig Barrett whined in the press a while back about relying on a technology (Rambus) that "gated" your performance. It's now pretty clear that previous Intel chipsets were "gating" RDRAM's performance.
Another interesting part of Tom's original benchmark was where he mentions in passing that he had successfully overclocked the P-4's FSB to 125MHz (on an Asus P4T board). With the FSB @ 125Mhz, the RDRAM is running at 1GHz, a pretty nice 25% overclock, the equivalent of running DDR-2100 SDRAM (266Mhz) at PC2625 speeds (333MHz). Dual RDRAM channels @ 1GHz deliver 4GB/sec of memory bandwidth to a 2GHz P-4. Even ovclocked, PC2100 can't even do 1/2 that--the data lines are double-clocked, but the address lines are not. Most of the performance improvement claimed by DDR is due to raising the Athlon's FSB to 133Mhz (double-clocked to 266Mhz) from 100Mhz. Wait until you see some benchmarks on the VIA KT133A chipset and compare them to the AMD750; you will be wondering what, exactly, DDR is doing for you over PC133 SDRAM.
That is a serious advantage for the P-4, despite it's other shortcomings on legacy apps. You can't overclock any AMD-based chipset for Athlon by more than 10%--the timing tollerances on the EV6 bus are just too close. Thank god you can still unlock the multipliers or overclocking an Athlon would be almost impossible.
At the end of the day, Moore's law says that we are going to see 12GHz CPUs by 2005 (or a 1GHz CPU will cost less than $40). The Pentium-4 is a step in that direction--a die shrink to
---
If you have a better explanation for why quake3 is fastest on a p4, then present it. Personaly i cant see how q3 is running faster on a p4 when the fpu is so much weaker and apps run slower. ~q~
And you'd put 4 G4s in what motherboard? Apple on makes a dualie AFAIK. That leaves you with what, an RS6000 (which I'm not even sure runs the 750s, only 604es or the Power line) to compare it to?
Whow, what planet are you on? I have an Athlon 750MHz, 256M Ram, nice 32M vid card... I'll trade you for 4 500MHz G4 Boxes, what's your address ;-)
Go look at John Carmack's plan. He's beeg asked this about a million times. There are NO SSE optimizations in Quake 3, let alone SSE2 optimizations.
Where you you AMDroids get this crap?
---
This reminds me of...
The 386 (faster than a 286, but oh so expensive, and no one uses 32 bit apps yet anyway)
The 486 (who needs a math co-processor? Geez it's expensive)
The Pentium (Gosh 486's are available with the same or higher clock speed)
The Pentium Pro (16 bit apps actually run _slower_)
The Pentium II (oh, bummer, L2 cache is at half-speed, PPro is so much better...)
Lets face it; many of Intels 'new' chips don't make immediate sense, but who was buying the predecesor to any of the above chips once the new style had been in the market for a while.
Personally, I'm looking forward to the new AMD chips. As always, more bang for the buck than Intel.
"But actually trying to use m4 as a general-purpose langage would be deeply perverse" --ESR
Well, this Pentium 4 processor isn't for everyone.
I read (red) on Intel's web page that the Pentium 4 has expanded the 80-bit registers in the FPU to 128-bit. That's not something they would do for "compatibility." I think they want SSE2 to be used for ordinary precision levels, and the old FPU is now a "high-precision" calculation unit. I expect it takes a great deal more time to calculate things to that high degree of precision. I doubt they put a lot of extra hardware to it; they're probably relying on microcode. Most of that time is wasted in legacy apps because (a) legacy apps cannot turn the high accuracy off, even if an instruction to do so exists, and (b) even though I assume new instructions have been added to load and store 128-bit floats, no existing applications use those instructions. So the high accuracy is there waiting to be used, and in the meantime, it's burning up clock cycles on existing floating-point code.
Intel's 80-bit FPU was always something of an oddity. It allowed intermediate results to have more precision than the inputs and outputs. If you are aware of this, you can actually use it. Even though the Alpha can calculate much faster clock-for-clock, the old x86 is actually more accurate. Except for the old Pentiums with the FDIV flaw. And the Pentium 4 is (assuming no errata) more accurate still.
If your emphasis is on speed instead of precision, you'll be "forced" either to use SSE2 instructions -- or buy an Athlon. But if you're one of those rare people who actually need 128-bit precision floating-point, the Pentium 4 is the only chip that currently supports it in hardware.
-- Sunlighter
Sunlit World Scheme. Weird and different.
They obviously optimised _some_ instructions. Do you think that they wanted the "PhotoShop" benchmark crown back off Apple/PPC?
FP.
Also FatPhil on SoylentNews, id 863
Intel, like Microsoft, is a derivator. Much of what they've done is simply an adaptation of previous work. No problem; that is what everyone does. Problem starts when they release two chips in a row which are demonstrably slower than chips in the stable already, but more expensive. When the P-III came out, I remember benchmarks showing it slower than the P-II for many things. The P-IV shouldn't surprise anyone. It took forever to create a P-II that was faster than a PPro in multiprocessor configurations. None of this is really bad, but Intel proceeds to advertise as if these are a lot faster than previous models, and the fastest on the market, which is laughable.
A society that will trade a little liberty for a little order will lose both and deserve neither. - Thomas Jefferson
Anyone who believes that AMD should start spending the kind of money on marketing that Intel does needs to have a note pinned to their jacket and take a ride on the short bus.
One of the big reasons that AMD can charge less for their processors is lower overhead. One of the biggest pits that money can fall into is advertising. You get the talk every year about how much it costs to buy commercial time during the Super Bowl. What is never discussed is how much it costs to put three or four commercials per NIGHT into prime-time network television every year. Start adding in cable/satellite stations (you don't want to miss any market segment in the viewing population) and the costs go through the roof.
I've found it hard to watch television on just about any station without seeing Intel advertisements, and this tells me that Intel is spending a LOT of money on advertising. I, for one, don't want AMD to go down that road if they can keep making inroads on price. After all, with companies like Compaq and Gateway selling Athlon-based PCs, it's not just the computer "geek" community buying the things. More and more, companies are going to be using AMD processors because of the price/MHz advantage, which is what most customers are looking for.
In closing: "Keep the advertising to the lowest roar you can manage, AMD. That way I don't have to win a lottery to buy a 'top-of-the-line' computer."
Yup...
I don't know about the sucking up to AMD thing though, they're pro-competition. Intel has a lousy track record.
In the day of the Celeron 300a, Tom's Hardware was all over the fact that it was the best bang for the buck. If they're pro-AMD it is because AMD is in the lead right now.
Almost every Intel CPU, unlike AMD and Cyrix, supported SMP. I still can figure what was the point of not supporting SMP in P4. AMD is cheaper/faster for the home systems, but P4 could still sell well for multiprocessor servers. No SMP support means no one in their mind would buy it when you can get 2x P3 1Ghz for less that you would pay for single P4 1.5Ghz with no ability to add 2nd CPU to the server.
Intel's size and larger R&D budget would normally mean they could actually release *faster* chips rather than slower ones. Perhaps this new floating point system will take off, but if not, Intel just bet the farm figuratively on something that is completely untested.
A society that will trade a little liberty for a little order will lose both and deserve neither. - Thomas Jefferson
...to count the dimpled chads the first time, and durring the recount he found some extra cpu cycles for designated for the Athlon.
I am become Troll, destroyer of threads
True enough, but most software manufacturers don't set their release schedules based on when Intel puts out a new instruction set extension. Apart for the ones you mentioned (media players and directx), a lot of commercial software goes through version revs at best at 1 to 2 year separations because of the expense of the beta/gamma testing and release cycle. Games are released according to a schedule (i.e. Christmas season) and I would think if adding support for a new feature can be done within the release schedule it will be done; otherwise expect to wait a few months for a patch. Do you remember how long it took for a Mechwarrior II patch using DirectX to become available? SSE2 support may take a less time for games because those functions should already be modularized to work with SSE or 3DNow.
Microsoft Visual Studio will probably rev soon with VB 7.0. Office 10 is going into beta testing but, since Office 2000 was only released in late '99, even if MS releases Office 10 before 4Q2001, I doubt that there will be many people switching to it before 1Q2002. The latter is about when ClawHammer is supposed to come out with SSE2 FP support. AMD may have a bit of a performance gap in 4Q but they can plan for it and come back swinging in 2002. That gap will be a lot easier to cross if they can start making inroads into corporate sales with AMD760MP dual processor servers by then. Otherwise they can hold their breath and drop ASPs for the quarter.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
As others have pointed out, these new chips are *usually* slower on 'legacy' code (cf original 5volt pentiums etc).
All this is a bit preliminary, ie P4 is certainly expensive and not that great Right Now, but almost certainly will rock when there is the code to support it (as there certainly will be).
I am a big AMD supporter, I think that they have done extremely well. But now, they are under the pump, just as they had Intel under the pump for the last 18 months.
All is well, just don't buy a P4 right now. Maybe the vapourware dual Athlon chipsets and motherboard will come out in time to toast the P4, but otherwise Athlon has some serious competition.
Lovely reply. I fully agree with everything you said. However, I've got a real feeling that chas' post was a troll. If he isn't trolling, he just made himself look like a complete idiot to anybody who has been gathering their AMD information from sources better informed (or less biased) than investment analysts.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
Tom tends to go for somewhat sensationalist reporting. He's no worse than any of the major media in this and my guess is he does it for the same reason: to get free publicity and more ad revenue. Personally I find it annoying and that it detracts from his ability to present an objective view. I actually appreciate a lot of the research and digging that he's done over the years but his approach detracts from his message because many readers doubt his objectivity; this thread is a case in point.
Personally I think Tom dislikes Slashdot because it's unappreciated that he's trolling on a scale no slashdot karma whore can ever hope to match. :-) Many of his T'sHG articles have run a fine line between -1 Troll and +1 Informative or Insightful.
Laissez lire, et laissez danser; ces deux amusements ne feront jamais de mal au monde. - Voltaire
When I threw this same story in the hopper this morning, I did it with something along the lines of "Happy Thanksgiving and the P4 is a turkey". With Grove gone, these guys don't have a clue anymore...
Now maybe the P4 when later revised will be better, but what reason is there to buy it?
-It's slower clock for clock than a P3 or an Athlon... In fact, a 1.2 GHz Athlon is probably a bit faster than the 1.4 GHz P4.
-It's many times more expensive than a comparable Athlon system.
-It's FPU performance is not at all up to the Athlon.
-You are stuck with RAMBUS and the buggy Intel RAMBUS chipsets.
The P4 just simply isn't worth what Intel is charging for it. And really I can't understand what the big deal about it is... It seems to me that it's a FAILED design, after all, hasn't every Intel chip since the 8088 out performed the prevtious generation at the same clock?
The only thing the P4 has going for it is that it's a new core, and probably can end up at higher clock rates than the Thunderbird Athlon core. But the P4 is like a school bus racing against a Porsche, it's got to have a much bigger engine running at a much higher RPM to equal the speed.
Plus, the P4 can't do SMP yet, and likely won't be able to before the Thunderbird Athlon (and the upcoming new core) can.
Intel will market the higher Mhz, but hopefully people will see thru it. After all, would you rather have a 1 GHz Celeron, or a 900 Mhz Athlon? The P4, like the Celeron, would have to run considerably FASTER than 900 Mhz to equal a 900 Mhz Athlon.
=== The price of freedom is eternal vigilance
There may have been a time about four or five years ago when Tom's opinion on some piece of hardware was actually worth something. Frankly, he's reached the point now where not only do I (and no doubt countless others) refuse to take any of his reviews seriously, I flat out refuse to even visit his site, period.
Ever since he started pimping for Nvidea, his opinion has been about as valuable as a truckload of dead rats in a tampon factory. He may have had something important to say, he might even have an intelligent opinion but his credibility is shot. My guess is he's probably dumping on the P4 because Intel refused to give in to his demands for free stuff.
Face it, Tom Pabst would call his own mother a syphillitic crack whore if someone else's mother gave him an apple pie and told him there were more if he kept it up.
ChodaBoy
- The preceding statement is the product of a deranged mind and the sole property of the voices in my head.
You said that pretty perfectly... Caveat Emptor!! (Almost certainly spelled wrongly, sorry bout that). I have nothing to add.
;-)
Thank you, and yes, its spelled right
-DP
What could possibly hurt the security of the American people more than giving our own government the ability to hide its
Since it was widely speculated and pretty much known that the P4 would accomplish less per clock than the P3, Intel should have simply milked the P3 for a couple more months until they could ramp up production of 2 or 2.1 GHz P4's which would hopefully outperform P3's at all tasks...
Why would anyone want to spend the fortune to upgrade their system to a P4 when they could save a bundle and get equivalent performance from one of Intel's own P3's, and that's forgetting all about AMD for the moment...
Intel had best be extremely aggressive about upping the clocks on it's chips... I read (I think on the Register, I'm not possitive) that AMD's roadmaps don't forecast the Athlon going much past 1.5 or 1.7 GHz this year. So, if Intel can go strong with the P4 and AMD does perhaps stumble under the weight of their recent successes, they might be able to be king of the hill once again...
Just not right now.
tom says his favourite computer magazine is PC World...yeah, i wanna get my hardware advice from this guy.
I have always liked intel more, however seeing these benchmarks disappoint me... then after hearing about all the other bugs, i switched my view of "intel is the best of the best" to intel.
After seeing a lot of intel's slip ups(0xf0 0x0f, and rambus to name a few) its amazing that they have a large percentage of the processor market. What I dont get is why they dont bother to do something good with their power and create a carefully crafted processor?
Personally, I think intel and amd should just slow down, especially in the years to come. In Wired 8.11, a 5 qubit calculator was made. While this innovation is small, it signifies a definite path of where computers of the future will go.
>Maybe I live in a cave, but I personally have never seen an AMD advertisment on TV or in the trade mags.
You probably lived in a cave a year or two ago, when they launched some SERIOUSLY popular ad's about their brand name, and nothing else.
I have mirrored some of the funnier ones at http://iamsure.psychasia.com/html/funny.html
>Besides, someone needs to take a shot at those blue Intel whatever they ares.... everytime I see those ads I want to puke
Those 'blue Intel whatever' are The Blue Man Group, a world famous perfmormance group on par with Stomp!. The Intel ads capture a glimpse of their humor and style, but doesnt really do it justice.
GPL'd web-based tradewars themed space game
Another reviewer (local to us Aussies), basically comes up with the same opinion. The review has some nice pictures, a good discussion as to the pro's (negligible) and con's (many) of RDRAM,
why Rambus is evil, and a lot of links for further info as to why.
http://www.dansdata.com/p4.htm
Actually, until this revision, Tom's was one of very FEW sites that actually had done a positive P4 review... Tom wanted to like the P4, but it turns out he can't.
=== The price of freedom is eternal vigilance
The P4 is a new core actually, its the first new core since the PPro. Thats the reason why its slower right now really, Intel always has this problem with its new cores at first. The Pentium was slower at first, and so was the PPro.
Much like those, give the P4 some time. As the clockrates go up and SSE2 enabled software comes out, it will start to look better.
Actually, The Register did some SSE2 enabled benchmarks, and the P4 was rather impressive in them.
http://www.theregister.co.uk/c ont ent/3/14922.html
-- "So they told me that using the download page to download something was not something they anticipated." - Bill Gates
I think Dr. Tom is biased against BAD HARDWARE, not necessarily Intel. You should have read him some time ago, he was always endorsing the P3/Celeron, until AMD just simply came out with a better product.
And any objective reviewer would have to conclude, that unless you need SMP, AMD's top processor is better than Intel's top processor. And Intel's top processor IS NOT the P4, it's the 1 GHz P3...
I hadn't owned an AMD based machine (since my original `286, circa 1990) until I recently replaced my P3 with a Duron 700. And I'm very happy with it and plan to replace my Duron with a 1-1.2 GHz Thunderbird Athlon. Not a P4.
Keep in mind, these benchmarks are RAMBUS based P4's going up against PC133 SDRAM Athlons... And the Athlons win. When the DDR based Athlon system is available, the gap will WIDEN.
=== The price of freedom is eternal vigilance
This design is still in beta and not meant for the general public. Imagine the horror if intel benchmarked an early buggy athlon against a production p3.
Only the State obtains its revenue by coercion. - Murray Rothbard
How likely is it that Intel will have used Quake 3 as a benchmark internally, and actually optimised their design for it?
Think about it, it's not completely unlikely, as the percentage of people and websites out there that rates Quake 3 as the most important benchmark is very high (I was one of them until the P4 came out), so it would make more sense commercially than to optimize for SPEC as they do traditionally.
Most reviews explain the "Quake 3 discrepancy" by its large hunger for memory bandwidth, which certainly is true. However within the large range of benchmarks I saw (for wide selection see the one on Ace's for example) there were plenty more memory-bound benchmarks, all of which the Athlon did equally well or better on.
Also, Quake 3 is VERY hungry for float performance, which the P4 appearently hasn't got. Does Quake3 use SSE2, either itself or in nvidea drivers which would allow them to mask this? Maybe it would be fun to test Quake 3 on a P4 with a video card that doesn't have T&L (to reduce the floating point load) and has no SSE2 optimisations (voodoo5 maybe?).
I know this will be considered flamebait, but someone has to say it. Tom pretty much hates intel. I wouldn't go to Toms to hear about how good the newest Intel Chipset is, ever.
i would take his 'reversal' with a grain of salt. i'm not trying to say anything mean, just don't take it at face value. would you trust a microsoft report on the bad side of open-source?
The company that people were expecting to dominate the market (well, they have been...) finally screwed something up under the legitimate pressure of AMD. A two-chip world, what a wonderful place!!
Cogito, ergo sum.
Maybe I live in a cave, but I personally have never seen an AMD advertisment on TV or in the trade mags.
That AMD is gaining so much market share from Intel is indeed a tribute to their price and performance but I'm sure they could do even better if they would launch a real advertising campaign.
Besides, someone needs to take a shot at those blue Intel whatever they ares.... everytime I see those ads I want to puke.
I know I am not the only one who can see that when all your applications are being run on a remote server you do not need a powerful processor. You just need a good marketing campaign. A year from now the P4 will be smoking along at 2-3Ghz while the better processor, Athalon (Palamino version), will be at 2 Ghz. Intel will say "We have the fastest processor and the best for Microsoft.Net". Running applications from a server optimally would require huge bandwidth but a poor FPU. That describes the P4 perfectly I believe. Read Red Herring's take on P2P and Intel.. Welcome to "Wintel" reconsolidating their shared monopoly.
If voting were effective, it would be illegal by now.
That's really what we're waiting for, right? I mean, right now a 450 mHz Macintosh probably beats the x86 processors across the board on mpeg4 compression. Based on Intel's stated goal of making the P4 a powerful chip for media processing in particular, I don't think this will be a long wait.
If you are modding me down because you disagree with me, use the "Flamebait" category, not the "Troll" one.
The FPU in the P4 is there for x86 compatability. Intel is betting that software developers will use some of the P4's 144 new instructions to accomplish floating point operations. The new instructions, if used properly, could realize significant speed increases.
Further, as was mentioned earlier, Intel has always released new generations of CPUs that didn't exactly take the benchmark world by storm. Wait 'till software emerges that takes advantage of what the P4 has to offer. Then you can try to complain.
Didn't anyone read what Paul DeMone had to say? Or Ace's review of the P4?
Im not buying the new p4 anytime soon. I like SMP since I run Windows 2000 server at my house. A Dual p2 400 with 256 Mb ram. My next computer will be a single procressor and Im heavily leaning towards AMD since theyre not as expensive as the p4. I do agree with Intel that they made the p4 to show the world the 'next' generation with their new instruction set, I just hope that it doesnt flop like MMX did. Everyone said MMX was the greatest thing, too bad no apps supported it. I still hear infomercialls hyping MMX makes me wanna shoot em for being morons. We here know that they are full of crap but the public has no idea and they are getting ripped off. I just hope that doesnt happen with the P4. The P4 will one day be a good chip but it seems that now their prices are way too high, and the only explanation is that they gotta pay their thousands of employees, good thing AMD is smaller and makes better AND cheaper CPU'S.
stainless steel
He does a good job of writing interesting articles, with the right ammount of technical detail and a "easy to read" language.
And his alleged "anti intel bias" is bull i.m.h.o. I've followed his writing for a while. The P4 article actually did it's fair share of Intel praising and AMD bashing in areas such as build quality and system stability where Intel really DOES shine. The fact that Intel's processor's just aren't as good as AMD's at the moment doesn't mean he has a bias anywhere. Even despite this he does mention, over and over again, that the P4's performance will likely be a lot better a few months down the road as SSE2 gains ground. That's really what you'd but in the words "bias" and "bashing" isn't it?
I mean, someone should have been around long enough that they remember the old AMD processor articles he ran, which weren't all that positive. And someone should remember his praise of Intel's information and business practises and how rude he thought AMD was.
-
-
Listen. Strange women lying in ponds distributing swords is no basis for a system of government.
Um, IRQ's are not related to a the CPU arcitecture, but rather to the origional design of the IBM board on the XT. Everything in a "PC" excpects things that way, so trying to change it would break every card that wants to use an IRQ.
And what does SCSI have to do with it?
ReadThe ReflectionEngine, a cyberpunk style n
That's exactly the point. Intel will be able to get massive MHz out of this core, and it'll leave Athlons in the dust(unfortunatly). Athlons will only be able to clock so high, and then that's it. They won't get any faster, without staying an Athlon.
:).
While the P4 will always be able to run at a higher clock at a given linewidth than an Athlon, this differential won't just keep growing - it levels off at a fixed ratio and stays there, regardless of how many linewidth shrinks each architecture gets.
This is just a reflection of how pipelining works. To do, say, a floating-point multiply, you have to run inputs through a block of logic. This logic block has a certain internal delay, that won't change. A pipelined processor breaks this block into smaller blocks, but the _total_ delay remains the same (actually, it gets worse, due to pipeline register overhead). Sell a case of Coke as four 6-packs or 24 individual cans, you still have the same amount of Coke.
The relative clock ratio between the machines is simply the ratio of the durations of the individual pipeline stages in each architecture (actually of the longest pipeline stage in each architecture). This ratio doesn't change as the design shrinks; the P4 still has j stages, and the Athlon still has k stages, and a multiply still takes a fixed amount of time to perform at that linewidth.
Now, how many stages you should have is an interesting optimization problem, but it's not relevant to this discussion (scaling of each of these two existing cores).
Linewidth shrinks can be applied to any processor, and will speed up each processor by the same factor (more or less). Differences are the result of different low-level implementations (static vs. dynamic logic, one-sided vs. differential signalling, manufacturing process differences, etc); not the result of large-scale architecture.
So, I have a lot of trouble with these claims that one design can scale farther than another; as far as I can tell, there's no reason why both couldn't scale as far as they wanted to.
The reason why we don't just keep shrinking the same design forever is that as more transistors become available, you have space to implement more complicated structures. Things like schedulers and branch- and data-predictors can work in any of a large number of ways, and some of the more interesting versions take huge numbers of transistors. Similarly, adding SIMD support only became practical recently. This is why designs change - not because of some magical clock speed beyond which the old ones can't run.
I have a BASc in Computer Engineering and am working on my Masters, so I've been studying this a fair bit
beacuse 450 Mhz is so much faster then 1.2ghz...
ReadThe ReflectionEngine, a cyberpunk style n
AMD were the ones who made Intel's chips. It was not that long ago when AMD was able to make it on there own, and Intel as well. AMD didn't steal anything.