• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Very interesting P4 benchmarks

Whoa, looks like the new SSE2 Intel compiler is really kickin' in the Spec2000, which is quite an optimal situation for P4 anyway for it's incredible hunger for bandwidth. However, the horrid RC5 score shows that P4 has some serious problems in certain areas. Anyone know how much typical RC5 client code has "difficult", hard-to-predict branches?
 
I am not trying to 'defend' intel here
but as far as I know, the current RC5 client do not recongise the P4, it doesn't mean much really 🙂 until we have one optimised.

 
Yes the article points out that the core for the AMD is totally optimized as is the PPC/Motorola core. When a Optimized P4 core comes out you'll see some great numbers.

Thorin
 
The SPECfp2000 looks really faked. Doesn't the alpha chip has about 550????????? Correct me if I'm wrong
 
those spec numbers are quite an eyeopener, but id like to see a direct clock for clock comparison to really guage the benefits of the Pentium 4
 
i am not 100% sure if they are fake but from what i remember

SPECfp runs a LOT better with high memory bandwidth

P4 has 3.2GB/s
P3 has 1.066GB/s

~550 isnt 'impossible' to me...

 
Wow, I really like that (p)review. I don't disagree with a word the guy says, so I encourage you to read it (I'm never wrong! *cough*).
 
Ok lets say they got a 30% performance increase.... but they have a 50% speed increase.... thats a 20% performance loss... with Palomino we should expect around a 30% performance increase.... so at 2Ghz AMD could have 1.5 and woop the crap out of Intel. And I do expect atleast a 30% performance increase from Palomino.... read the articles on this stuff lots of people said that the Athlon has tons of potential and that AMD had to rush it.... but now they had time to tweak it out big time.
 
People always over estimate the speed increase for the Athlon.

First it was going to be 30% speed increase when moving the cache on-die.

We saw a 10-15% increase.

It was going to be a 30% speed increase when DDR was introduced.

We saw a 10-15% increase.

What is this to say about your prediction of a 30% increase?
 
IBMer you need to read articles. There have been plenty of them on how AMD rushed the Athlon out and that it is not near the limit of its potential.
 
Back
Top