Pentium 5? AMD has Hammer in production already

[jbond04]

Originally posted by: SSXeon5


Yup Intel 0wnz j00!

SSXeon

Hey...don't worry about that. I'm an Intel fanboy, too (I'm just not rabid). I own a 2.53GHz P4 that's running at 2.85GHz with 1GB of PC1200 RDRAM... The last thing I am is an AMD fan.

 

[paralazarguer]

How about that new benchmark created by Van Smith, COSBI (that is a pure mathematical benchmark, so pick the winner.....)???

That benchmark is amateurish. A 1.7ghz celeron beats out a 2.53ghz Northwood using COSBI. Are you telling me that the celeron has more pure mathematical power then them northwood?

 

[gennro]

see the thing is a 2.7 ghz p4 can't do more MIPS then a Tbred 2200, why cause the p4 has 20 pipes and the amd has 10

 

[imgod2u]

Originally posted by: gennro
see the thing is a 2.7 ghz p4 can't do more MIPS then a Tbred 2200, why cause the p4 has 20 pipes and the amd has 10

What does that have to do with MIPS?

 

[paralazarguer]

see the thing is a 2.7 ghz p4 can't do more MIPS then a Tbred 2200, why cause the p4 has 20 pipes and the amd has 10

Firstly, they don't have 10 or 20 pipes. The p4 has a 20 stage PIPELINE. It's like an assembly line. A 10 stage pipeline is much more effecient than a 20 stage pipeline. The 20 stage pipeline requires higher clock speeds to achieve the same performance. Your argument doesn't make any sense.

 

[alexruiz]

Originally posted by: 7757524

How about that new benchmark created by Van Smith, COSBI (that is a pure mathematical benchmark, so pick the winner.....)???

That benchmark is amateurish. A 1.7ghz celeron beats out a 2.53ghz Northwood using COSBI. Are you telling me that the celeron has more pure mathematical power then them northwood?

Did you check the update??? Settings have to be EQUAL, which they weren't the first time the benchmark was run..... Go and check the new data, and then draw your conclusions.

 

[SSXeon5]

Originally posted by: jbond04

Originally posted by: SSXeon5


Yup Intel 0wnz j00!

SSXeon

Hey...don't worry about that. I'm an Intel fanboy, too (I'm just not rabid). I own a 2.53GHz P4 that's running at 2.85GHz with 1GB of PC1200 RDRAM... The last thing I am is an AMD fan.

NICE!!!!!!!!! Im jelious now lol ...... yeah i know .... intel is the king .... eat it amd

SSXeon

 

[Duvie]

genro, WTF are you talking about....Sandra 2002 SP1 said my 1.6@2.66ghz (666fsb) was faster then a T-bred 2200+...At 2.736ghz I am even faster...


Rainsford....

POint taken on the tbird and lack of sse....but unless it gave like a sh^t load in boost which I don't think it does then it still is far behind....

I would think the 1900+ xp (1.6ghz?? stupid rating)...would take about 3 hrs and 10min....I mean the 1900+ stock should be comparable but slightly behind my 1.8 when it was running 2.2ghz and 400mhz ddr and that is roughly what it got....The 1.6 amd is 32% faster then 1.6ghz p4a but that is no surprise to many....

Notice the 1.6ghz xp would have a 31-32% gain on 1.4tbird@1.5 in performance but only equate to 7% in speed increase. Couple that with the fact i was running a sis735 chipset, then throw in advancements in hardware prefetching and sse encodings and that would be about right....


Go to Tomshardware and look at where the athlon relates to p4 in divx encoding and in that program (which I don't run) it shouldn't be sse2 optimized either...so no favoritism should be being played....

 

[imgod2u]

Firstly, they don't have 10 or 20 pipes. The p4 has a 20 stage PIPELINE. It's like an assembly line. A 10 stage pipeline is much more effecient than a 20 stage pipeline. The 20 stage pipeline requires higher clock speeds to achieve the same performance. Your argument doesn't make any sense.

Yes and no. While branch mispredicts do cost loss in efficiency, or rather, more clocks wasted. Prediction methods nowadays on the P4 are up to 97% accurate. With enough memory bandwidth to feed the pipeline, it could achieve nearly the exact same throughput per clock as a 10-stage design. This is ignoring the other limitations in the P4 such as the limited issuing ability of the trace cache, the weak FPU, etc. But generally, everything else being equal, a 20-stage design compared to a 10-stage design would only have slightly less throughput per clock with modern prediction methods.

 

[Rainsford]

Originally posted by: Duvie
genro, WTF are you talking about....Sandra 2002 SP1 said my 1.6@2.66ghz (666fsb) was faster then a T-bred 2200+...At 2.736ghz I am even faster...


Rainsford....

POint taken on the tbird and lack of sse....but unless it gave like a sh^t load in boost which I don't think it does then it still is far behind....

I would think the 1900+ xp (1.6ghz?? stupid rating)...would take about 3 hrs and 10min....I mean the 1900+ stock should be comparable but slightly behind my 1.8 when it was running 2.2ghz and 400mhz ddr and that is roughly what it got....The 1.6 amd is 32% faster then 1.6ghz p4a but that is no surprise to many....

Notice the 1.6ghz xp would have a 31-32% gain on 1.4tbird@1.5 in performance but only equate to 7% in speed increase. Couple that with the fact i was running a sis735 chipset, then throw in advancements in hardware prefetching and sse encodings and that would be about right....


Go to Tomshardware and look at where the athlon relates to p4 in divx encoding and in that program (which I don't run) it shouldn't be sse2 optimized either...so no favoritism should be being played....

I agree, and I've looked at the Tomshardware page. Like I've said before, I'm not a big overclocking guy, so that's why there isn't as much draw for me to the P4. However, to each his own and I think both are good choices right now, just different strokes for different folks or something.

 

[jbond04]

Originally posted by: gennro
see the thing is a 2.7 ghz p4 can't do more MIPS then a Tbred 2200...

Hahahaha...

You're joking, right? Running my 2.53GHz at 2.7GHz, I get 5236 MIPS. You're telling me a 2200+ Tbred can beat that? Show me...

 

[jbond04]

Originally posted by: gennro
all i have to say who ever can offer top performance at the lowest price i'll buy it first.... and right now AMD seems to be the winner
MIPS (Million instructions per second) a my 1.3 ghz t-bird does the same MIPS as a 2ghz p4

Yeah...a P4 Willamette or a P4 Northwood? Is this what you're basing your "2200+ Tbred is faster than 2.7GHz P4 in MIPS" claim on?