Intel 3.2E vs. 3.2EE vs. 3.2C: Comparing Baseline Performance
by Wesley Fink on February 12, 2004 2:46 PM EST- Posted in
- CPUs
With the introduction last week of the Socket 478 Prescott, we now have no less than three 3.2 GHz Pentium 4 processors from Intel. As we move toward the expected April 2004 introduction of Socket 775 and a new revised Prescott, we will be including compatibility testing of the 3.2E and 3.2EE, in addition to the 3.2C in future Pentium 4 motherboard and system reviews. With 3 flavors of 3.2GHz P4, 3 different architectures, and potentially different performance, all benchmarks needed to be updated to allow a reasonable comparison of current and future benchmark results. The results of the updated tests were so interesting that we decided to publish the results. Some of the results are already well-known, but other results like Workstation Performance were a bit of a surprise.
This is not intended to be an in-depth review of the new Prescott processor; Anand published a definitive review of Prescott at launch. For an in-depth look at Prescott and a comparison to AMD processors, you should read Anand's Intel's Pentium 4 E: Prescott Arrives with Luggage. The scope of this article is much more basic - to update our standard suite of Motherboard Benchmarks with the 3.2C, 3.2EE, and 3.2E on the same Asus P4C800-E motherboard.
While the three 3.2GHz processors from Intel all run at the same clock speed, they are quite different under the hood.
| 3.2GHz Pentium 4 Processors | |||
| 3.2C | 3.2EE | 3.2E | |
| Name | Northwood | Extreme Edition | Prescott |
| CPU Speed | 3.2GHz | 3.2GHz | 3.2GHz |
| L1 Cache | 8k Data Cache + 12k instruction cache |
8k Data Cache + 12k instruction cache |
16k Data Cache + 12k instruction cache |
| L2 Cache | 512kb | 512kb | 1MB (1000kb) |
| L3 Cache | - | 2MB (2000kb) |
- |
| Core | .13µ | .13µ | .09µ |
| Socket Type | 478 | 478 | 478 |
| Memory Type | Dual-Channel Unbuffered DDR | Dual-Channel Unbuffered DDR | Dual-Channel Unbuffered DDR |
| Bus Speed | 800 | 800 | 800 |
| Web Price 2/10/2004 | Retail $289 | Retail $957 | ? (Projected $289) |
We could not yet find the new 3.2E available for sale anywhere, but Intel has priced the 3.2E the same as the 3.2C. When it does arrive, the prices should be about the same as Northwood. To keep value in perspective, the Prescott and Northwood will sell for about the same price, while the P4EE is more than 3 times that price.
Facebook
Twitter
RSS
19 Comments
View All Comments
mindless1 - Sunday, April 18, 2004 - link
WOnder of all wonders... we have evolved to supporting INDUSTRY STANDARD image file formats yet the review uses proprietary pics. I (and I suspect others) are "a rock" on this point, won't run flash. Using such pics in an article simply causes readers to go elsewhere, which in an overall elite-snob sort of way may not matter but in the grand scheme of things, losing readers beside of abandonment of industry standards is just a dumb idea.Aervires - Saturday, February 14, 2004 - link
Are you sure you bought both processors retail?The P4EE is $955 AND THE A64FX-51 is $745
The total cost a the P4EE will run you more however both are a waste of money. Also the when socket 939 comes BEWARE.
Aervires - Saturday, February 14, 2004 - link
Pumpkinierre - Saturday, February 14, 2004 - link
#9 Trog, if it was only pipeline related then the p4EE comanche4 results should have the same as the northwood. But clearly it is better because of the extra cache 2Mb L3. The Prescott has 1Mb L2 and 16K L1 so it should have results in between the 3.2EE and 3.2c (if not close to the P4EE) if it had the same core as the n'wood and cache latency. In fact it is markedly down in commanche4 cf. with N'wood. Given that other games benchmarks/demos and tests have the two processors within 10%, I dont think this failure is solely as a result of the 31 stage pipeline and branch predictor etc. (remember intel also improved its efficiency with a lot of tweaks). The main problem is the cache latency has slowed down and this obviously has a marked effect on comanche4 as it would in real time gaming where memory subsystem latency is paramount. See here:http://www.xbitlabs.com/articles/cpu/display/presc...
Quote:
"Yes, unfortunately, we have to state that not only the size of Prescott’s cache-memory has grown bigger, but also its latency. And the latency grew up a lot, I should say: for L1 cache the latency doubled! As a result, Intel will no longer be able to boast the extremely low latency of its L1 data cache. From the temporal point of view, the latency of Pentium 4’s L1 data cache got close to that of Athlon 64 L1 cache, though the latter is four times larger. However, the increase in the L1 cache latency is another forced measure, so that the new Pentium 4 processors on Prescott core could go beyond 4GHz core frequency.
Similar changes were made to the L2 cache, too. In terms of L2 cache latency, the new Prescott processor yields to Northwood, as well as to the competing CPUs from AMD Athlon 64 family."
That, plus poor fpu plus heat- not a good gaming cpu scenario. Intel have blundered big time and they must be rushing to get something out soon to relegate this doozer to the dustbin of history.
truApostle - Friday, February 13, 2004 - link
The P4 extreme is a total waste of money, IMHO. To spend $550 more dollars only to gain 3-5 fps in some of the new(er)games is NOT worth it. OF course if you are still playing Quake III then you will realize some smokin fast frames in which case it's all overkill anywho. In every bench from content creation to encoding to gaming the P4EE was marginally ahead, at best. I guess if your a Rockefeller or a Rothschild then it's okay to build a system based upon that proc.I think the best deal out right now is a 64 3400. Benchmarks are really close to FX-51 and are matched too (content creation, encoding) or better (in gaming benchs) than most Pentium archrivals. Just my humble opinion.
Either way isnt it a great time to be human, with all this technology to play with.
truApostle
-when you aim for perfection, you find that it's a moving target.
sipc660 - Friday, February 13, 2004 - link
we seem to go down the same road as with the previous article.#1 was right you should overclock both 3.2C and 3.2E to show the overclocked performance delta with the measured temperature.
that should give people who are interested in buying either a better comparison since the chips are going to be the same price at launch
what you also should drain out of intel is wether or not it has built in 64 bit circuitry. and can it be unlocked through bios at a later stage.????
right now at the moment it seems 3.2C is a clear winner regardless of the high end workstation performance.
overall i would buy barton before i'd buy any of the intel chips. i don't beleive anyone here would buy any of these chips till they seriously come down in price.
just keep it cool (hehehehe)....
go amd
P.S: cramitpal is a passionate joker, if no one can c that than you are blind. personally i'd give him a nice friendly hug if i met him on the street.
LOL
Superbike - Friday, February 13, 2004 - link
I think I'll pick up a Prescott I need a computer/space heater.edub82 - Friday, February 13, 2004 - link
Well this article clinches it for me...The 3.2EE must be mine, the increased performance over that pathetic 3.2C is well worth the 600 extra dollars I'm going to have to spend.
kmmatney - Thursday, February 12, 2004 - link
Yep...Athlon 3400+ $270
2x512Mb PC3200, Corsair: $155
MSI KT8 NEO $97
Total Price $522.
That's the system we just ordered at work. We do a lot of image processing and image compression, and fpu intensive image manipulations and the Athlon is faster than a P4 - at least with our particular applications.
vedin - Thursday, February 12, 2004 - link
That's true johnsonx but..Athlon 3400+ $404
2x512Mb PC3200, Corsair: $155
MSI KT8 NEO $97
Total Price 656.
And it's like, what? .5% slower than the Athlon FX-51?
I see your point about the FX though, as do most here. It's kinda useless with a much cheaper AMD option right next to it performance-wise. But either way, the system I just described makes BOTH systems look silly for buying, unless you just like spending money.