Pentium IV Problems?

zottl writes: "German tech site computerchannel.de has an article about various problems concerning Intel's Pentium IV. It says that the new processors will draw lots of power (66 watts for the 1,4 GHz version), need special copper-core coolers, might need radiation shields for the socket pins for ECC compliance and will remain expensive for quite some time. It also says that the P4 will only get mass-market appeal with the introduction of the slimmed-down 0.13 micron version. Oh, and best of all, it seems to be slower for certain apps than a P3 of the Mhz. Seems like a repetition of the problems the P6 architecture had when the Pentium Pro was first introduced" Isn't this pretty much what they say about every generation of Intel chips when first released? Anyway, the article is in German, so you'll need to feed it to the fishy until translations crop up.

Comments

When will the world accept that x86 is a deadend?

[Anonymous Coward]

I'm not going to say that it has anything to really do with x86 in particular but I know the architecture and it is the most antiquated POS out there. Intel clearly has to put more effort in to producing processors than say, Compaq or Motorola, that translates in to more expensive processors and less performance. Moore's law isn't still in effect unless you look at Hz, the actual performance isn't going up that fast.

I think the best thing would be for us to dump the old architecture. If you're not going to get the continuous speed improvements and dirt cheap costs (although AMD and Intel have been driving them down) at least you should get a clean and easy to program architecture. Hopefully in 5 years IA-64 and PowerPC will be where it's at, since I'm on Linux and staroffice is going GPL I won't have anything I use regularly that I won't have code to. I also think AMD should wake up and smell the coffee, they need to look at how much they have invested in trying to emulate Pentiums, honestly, I think they should work a deal with Sun or Motorola or somebody and more their 64bit plans to something sane.

Let x86 die.

I'm sorry but...

[Anonymous Coward]

--It seems to me there has been a lot of whining about P4 power consumption. This isn't meant as a flame, but the P4 is supposed to be a high performance microprocessor. Honestly, 66W is not a lot in the performance arena. Take a look at the power consumption of the alpha 21264 @ 550Mhz... 100W.

ftp://ftp.digital.com/pub/Digital/info/semiconduct or/literature/21264ds.pdf

I think a little perspective is required before jumping on these "P4-consumes-too-much-power/generates-too-much-hea t" bandwagons.

-just my opinion.

ECC?

[Detritus]

It's EMC (electromagnetic compatability). not ECC (error correcting code).

Slower at same MHz no surprise

[Jah-Wren Ryel]

Intel made a lot of changes to the architecture in the P4, the primary goal was scalability, not efficiency. So, it is no surprise that they are slower than a P3 at the same clock frequency. But the P4 is expected to scale to the 3-4GHz range, while it is doubtful that the P3 will ever even make it to 1.4GHz. So, Intel has given a little at the low end as the cost for being able to go much, much faster in the long run.

Not currently seriously considering sledgehammer

[Greyfox]

I know Intel's out there working with everyone to make sure compilers, apps and OSes are available for their product. I haven't heard a peep out of AMD. If they're not careful here, they might lose the industry back to Intel.

Clock speed is everything in the mass market

[ahg]

It's easy to see why Intel is taking the road to higher clock speeds even at the risk of (initially) slower performance. It's all too readily evident that joe-sumer (average joe consumer) is hooked on having "more megahertz" in their box.

Intel still dominates the mass market, and I'm sure it's an important lead for them to maintain. Given the recent 1.1 GHz P3 debacle - it's reasonable to assume that their marketing people have told them that they must stay on top of the GHz war even if it's at the cost of better technology.

While it may take some time for the P4 to hit the mass market, that's where it's headed in the long term. (The P4 is not like the PPro that was only intended for use in high end workstations)

Apple tried to transistion the mass market in thier TV ads to thinking in Gigaflops, a better indicator than MHz. - See any store ads pitching Gigaflops recently? neither have I.

Re:Does anyone really still trust Intel?

[be-fan]

So AMD's total screwup of the K5 and original K6, it's it's mess-ups with the 2/5 clocked cached on the previous Athlons didn't make you lose trust in THAT company? Seriously though, people who base their opinions of a company on one or two debacles are stupid. Wait for the P4 to come out. See how it reviews. Try one out. Ask your friends how they like theirs. Compare it with an Athlon. Buy whichever you like best.

Battle Plan

[NoWhere Man]

Intel has been having a hard time with chips lately. I don't trust any PIII beyond the 850 mark (okay alot of ppl are prolly going to flame me for that, but that is a personal feeling). The PIII process is out of date and should be retired and the P4 is just being shoved in our faces like an empty promise.
Intel needs to take a step back and create a new product that'll help em take back the market. And yes I think AMD should take the market for awhile, they have not only earned it but deserve it. This is not to say that I think Intel should crash and burn, quite frankly, the race for speed has forced the 2 companies to develop new designs and think differently than previously.
But Intel just seems to be thinking like Homer Simpson.
"I don't know Herb, people are afraid of new things. You should have just taken an existing product and added a clock or something.."

Maybe Intel's recent problems with the 1.13 Ghz PIII will help them realize that they need to concentrate more on the products for awhile, rather than the market and $$$.
One can only hope.

Re:CPU wattage question

[VAXman]

The 500 MHz Pentium III draws about 30 Watts.

What's funny is that the Athlon also draws 60 watts. That chip was released over a year ago, but nobody thought that 60 watts was a lot until the Pentium 4 is about to come out.

Also, Willamette is expected to be a stop-gap to get the Pentium 4 in the marketplace, while the Northwood is going to be the real deal. That's going to be at 0.13 and there is even expected to be a laptop version, so the power is going to be much less. Much like Pentium Pro, where only about one million parts where shipped, and the Pentium II shipped umpteen millions of parts were shipped.

Re:Chip Quality Formula

[Mike1024]

Hey,

How about:

([Speed in Mhz]/(([Highest speed in contention]+[lowest speed in contention])/2)) - ([Price]/(([Highest price in contention]+[lowest price in contention])/2)) + ([Power in Watts]/(([Highest wattage in contention]+[Lowest wattage in contention])/2)))

Basically, each property is reduced to an integer denoting it's performance relitave to the average of all the processors under consideration, then the numbers are added and subtracted, depending on wheather each number should be high or low. The processor with the highest number would be the best.

To find the best value for money, a far easier formula to use would be:

[Price]/[Speed]

That would give you the pounds-per megahertz value for each chip. Personally, I'd sooner judge it with:

[Price]/[Speed in FLOPs/s (Floating point operations per second)]

Because FLOPs/s is a better judge of speed that Mhz, in my experience.

Michael

...another comment from Michael Tandy.

Moore's Law speaks only of transistors!

[Mr Z]

Folks, get it right. Moore's Law simply states that the number of transisitors on a chip doubles every N months, where N = 24 in the first statement of the "law", and was revised shortly thereafter to N = 18 .

Typically, performance scales with number of transistors, but that is not always true! There are three main reasons performance does go up roughly by the same ratio as the number of transistors:

• Some of those transistors can be used for new functions. For example, additional functional units (such as the three-way issue pipeline on PentiumPro/PentiumII vs. the U-pipe and V-pipe on Pentium vs. the single-issue pipe on 486). This is a direct application of transistors to performance, but it only addresses computation-bottlenecked applications. Additionally, some of those transistors can be used to build wider pathways on the chip, leading to improved bandwidth to help bandwidth-starved applications.
• Smaller transistors switch faster, and so can operate at higher clock rates. This has the dual effect of increasing the number of computations per second (again helping compute-bottlenecked applications), as well as increasing bandwidth--at least on the die. Going off-chip can still be a bottleneck. That brings us to the third bullet:
• Smaller transistors can be used to build a bigger cache, so that the clock rate and on-chip bandwidth benefits can be used to greater effect.

Sounds great, but what's bad?

Well, one big thing that is not addressed by faster transistors is latency. As transistors get smaller and the wires that connect them get smaller, communication between transistors starts to become the true bottleneck. In the "Good Old Days", you could send a signal anywhere on the die in a single cycle, and you could treat a wire as an instantaneous link. In these smaller technologies, though, transport time for signals burns a significant portion of the time for any computation. This is why pipelines get deeper and deeper with each generation. Essentially, you can only make effective use of all of those transistors if you can minimize the amount of communcation between them, and that's what pipelining is all about. Unfortunately, this limits how much you can speed up many applications, especially general-purpose compute problems.

Newer architectures address latency problems by exposing their pipeline (see EPIC or VLIW), or providing extensive resources for dealing with it. The Alpha CPUs, for instance, have an aggressive cache and reorder buffer that allow many pending cache misses to be services while non-dependent instructions are executed happily. (IIRC, the 21264 allows up to 4 hits under miss in the cache -- that is, you can have up to four misses outstanding and still take hits in the cache and allow instruction execution to proceed. I don't have Hennessey and Patterson handy to check though.) The reason this is even conceivable is that the Alpha provides a huge bank of architecturally-visible registers, and an even larger bank of rename registers for rescheduling code. Since compiled code spends most of its time moving data between registers, the architecture can easily determine which instructions are dependent on each other and very effectively hide the latency of the pipeline by reordering instructions and renaming registers.

In contrast, the x86's highly bizarre and rather small register file create a huge bottleneck to reordering, since compiler ends up spilling many intermediate values to the stack or other memory locations. As a result, the CPU can't use register names to determine instruction dependencies as often, and so it cannot aggressively reorder instructions. As a result, it cannot hide the latency in the pipeline as effectively, and gets bitten with poor performance. All those transistors sit idle more often. (This, BTW, is why the Alpha can beat the Athlon on some apps, despite a 2x clock-speed advantage on the Athlon's part.)

There are plenty of other reasons why x86 can't keep up performance-wise, but this is not the forum to discuss them. Just remember, x86 is keeping up with Moore's Law just fine. Don't expect its performance to keep scaling at the same rate.

--Joe
--

Like the PPro?

[be-fan]

It seems that the pundits spend most of their time doubting Intel, while Intel becomes the de-facto standard with their new chips. Take the Pentium. Soon, everyone (AMD & Cyrix) moved to the super-scaler design. Intel added MMX, and AMD and Cyrix added 3DNow!. People originally thought that the PII would be a failure (it's slower than a PPro at the same clock-speed) but it became THE high-end standard for years. People thought that the Pentium wouldn't make it because it ran 486 optimized code slower than a 486. Instead, people just reoptimized their code. All these chips had quirks. Just like the P4 has quirks. However, the software industry will work around these quirks, just like they have for all the other Intel chips.