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
Final Words
The only goal of testing three different Intel 3.2 processors on the same reference motherboard was to establish benchmark results for future motherboard testing. However, it is hard not to compare the CPU performance differences in these tests. What perhaps stands out the most is that for all the negative press Prescott has received at launch, it is really a decent performer that is very close to Northwood, at least at the 3.2GHz speed. As Anand found in Intel's Pentium 4 E: Prescott Arrives with Luggage, Prescott scales faster than Northwood as speed increases. At 3.2GHz, performance is closer than we first thought it would be, and Prescott is certainly ahead of Northwood by the time you reach 3.7 to 3.8GHz.Looking at individual performance categories from our standard benchmarks, it is easier to understand why Intel introduced another EE chip in the 3.4 speed. The one area where Prescott is poorest is gaming performance, an area already dominated by the Athlon 64. Prescott is slower compared to Northwood in this area, but the real difference is 0% to 5% in most cases. Northwood is not a great game chip either compared to A64, so the 3.2EE and 3.4EE fill that void. Unfortunately, the cost of bringing Pentium 4 gaming performance to Athlon 64 levels is very high, with the EE chips selling for premium prices.
For our standard "real application" benchmarks, Multimedia Content Creation Winstone and Office Winstone, Prescott and Northwood performed virtually the same in these baseline tests. You will not notice the difference between these similarly-priced chips. P4EE performs best in these benchmarks, as expected, but at 3 times the price, the P4EE is not really in the same league with Prescott and Northwood.
The areas that most surprised us were Media Encoding and Workstation Performance. We expected P4EE to lead in these benchmarks, but instead, Prescott was the top performer. Intel tells us that Divx 5.1.1 provides support for new SSE3 instructions, and at least in our configuration with an XMpeg 5 front end, Prescott leads in Media Encoding. Other Media Encoding benchmarks that use older codecs have been showing Prescott about the same as Northwood in encoding tests. Workstation Performance was generally dominated by Prescott, so those applications that depend on the types of operations tested in SPECviewperf will perform best with the Prescott 3.2E. The SPECviewperf 7.1.1 benchmarks were repeated on a VIA PT880 motherboard with all three 3.2 processors, and Prescott again dominated these benches, providing the top performance in all the 7.1.1 suite including UGS.
Prescott and EE compatibility will be a part of all future testing of Pentium 4 motherboards. With the testing that went into this comparison, you should have a better idea of how the 3 flavors of Pentium 4 compare at the same speed on the same reference motherboard. Prescott is not a very good gaming chip at 3.2 GHz, but it is closer to Northwood than many expected. If you can afford the price tag, P4EE is still the best choice for gaming among Intel processors. Northwood and Prescott are virtually interchangeable as measured in Content Creation and General Usage performance. For Media Encoding and Workstation Performance, Prescott at 3.2GHz appears to have an edge over both Northwood and the much more expensive 3.2EE.
Facebook
Twitter
RSS
19 Comments
View All Comments
TrogdorJW - Thursday, February 12, 2004 - link
Pumpkinierre, I think you are totally wrong on the cause of the Prescott slowdowns. Without getting technical, let me just assure you that it has pretty much everything to do with the 31 stage pipeline in the Prescott compared to the 20 stage pipeline in the Northwood (and P4EE).The larger cache can help the Prescott overcome the effects of the long pipeline, but in certain types of code, you're basically screwed. Even the branch prediction can't help in some instances. Say a program has a lot of branches, and they're spread over a large enough area that the predictor can't track all of them. If you "overflow" the size of the branch prediction table, then the penalties of the longer pipeline are going to become very apparent.
It appears that games are quite capable of doing this, and Comanche 4 in particular seems to have a lot of unpredictable branch code. Really, though, who cares? Comanche 4? I tried it, and thought it was pretty lame. At least UT2K3 and Q3 are pretty fun to play, even if they're old now.
johnsonx - Thursday, February 12, 2004 - link
Cram says the P4 EE (aka the Piss-4 Enema Edition, according to Cram) is a $1000 pipe dream. Lets compare, shall we, using prices from NewEgg:Socket 478 Mainboard, ABit IC7: $119
Intel P4 EE 3.2Ghz: $880
2x512Mb PC3200, Corsair: $155
TOTAL PRICE FOR P4EE: $1154
-------------------------------------
Socket 940 Mainboard, ASUS: $205
AMD Athlon64 FX-51: $733
2x512Mb PC3200 Registered, Corsair: $258
TOTAL PRICE FOR A64 FX-51: $1196
Why is the P4EE more of a "$1000 pipe dream" than the FX-51? The P4EE is actually a touch cheaper, plus the board and ram are both standard types which many people may already have; with the FX-51, it all has to be purchased new yet will soon be obsolete.
Seems to me the FX-51 in it's current form is just silly... perhaps even a $1200 pipe dream.
None of this has much to do with the article these comments are supposed to be referring to, but Cram's comments never do either...
Pumpkinierre - Thursday, February 12, 2004 - link
The p4 northwood IS a good gaming chip. The equivalent a64 doesnt beat it in some games and only by < 10% in the others. If AT had included a stock cooled o'clocked 2.4c@3.2 in these benchmarks (and I believe they've got a good one in the cupboard) the story would be a lot different. The big problem here is Prescott. It's too hot to o'clock sanely with standard air and stock, it underperforms Northwood which has half the cache. Why? Is the fpu a dog? Is the double sized cache slower latency-wise? From all reports its both.Comanche4 gives it away- generally regarded as a cpu intensive benchmark. The p4 EE is the same as the Northwood up to the L2 cache but the P4EE has 2Mb of L3 which lowers memory latency in some apps. cf. to Northwood. So the difference in performance, 3.2c vs 3.2EE, cant be fpu related but must be dependent on memory sub system latency The same lowering of memory latency can be achieved by o'clocking a 2.4c. In the case of the Prescott you cant do this because of the heat, you are just stuck with a dog!
I think 512K L2/8K L1 caches is the optimum for gaming with the P4. Northwood is the better gaming and enthusiast P4 as well as still the best all round cpu (coolness, reliability, compatibility and M'board support). Prescott is relegated to workstation/encoding where sse3/HT play a bigger part and o'clocking/tweaking/ reliability are'nt of great importance. A poor fate for the P4 where Prescott should have been its crowning glory.
PrinceGaz - Thursday, February 12, 2004 - link
The article pretty much confirms my gut feeling about Prescott after reading all the reviews on NDA day, which is that its every bit as good as the Northwood for non-gamers if you aren't going to overclock it and therefore not concerned with heat.As the Northwood is still considerably behind the A64 in all gaming type applications, its a non-issue that the Prescott is slower than the Northwood for games as no gamer would consider buying either anyway. As for the P4EE, systems built using it will only be purchased by those where cost isn't a concern -- I can't see many people who build their own box actually opting for it over an A64 (gamers) or P4 'E' (DVD-rippers).
Please don't mock Cram..., he may actually believe some of what he spouts and it sure gives me a good giggle whenever I get to read any of it (before mods delete it) :)
Icewind - Thursday, February 12, 2004 - link
Athlon 64 is just looking better and better every day.Can't wait to upgrade this summer.
cliffa3 - Thursday, February 12, 2004 - link
when anand publishes his article on how his mac experience goes, that's still going to be the "bottomline" according to cram...can we get a feature that auto-posts that as a comment to each new article to save him the trouble?.CRAMITPAL - Thursday, February 12, 2004 - link
Bottomline:Prescott SUCKS and PEEEEEE is a $1000 Pipe Dream.
Only a fool...
araczynski - Thursday, February 12, 2004 - link
Great article, very usefull and to the point.Good job :)
tfranzese - Thursday, February 12, 2004 - link
Nice article and I'm glad you guys went ahead and published it as is. I just wish I did encoding - then I'd have a good reason to put together an Intel system. I may just do it for fun someday to compliment my other boxes.Can you guys please included distributed.net RC5 crunching benchmarks? I would like to see that and benchmarks of overclocked Prescotts vs. overclocked Northwoods to better see the scaling @ 3.4/3.6 GHz and how big a gap will start to form.