Athlon XP 3000+ vs. P4 3.06 GHz

i wanna know htf they can justify this then...
http://www.theregister.co.uk/content/3/29252.html

the results on toms hardware speak for themselves... the intel has more throat on it - simply because intel brought out HT and pretty much woulda remade their cpus... whereas amd put more cache in their cpu an said "yeah its better" - it worked with their older cpus... but it certainly didnt this time

but... i noticed that winding the AMDs up helped them a bit in a few of the tests... i think it would be interesting to do the score for the cpus while overclocked also.
 
Originally posted by TechSupport
i wanna know htf they can justify this then...
http://www.theregister.co.uk/content/3/29252.html

the results on toms hardware speak for themselves... the intel has more throat on it - simply because intel brought out HT and pretty much woulda remade their cpus... whereas amd put more cache in their cpu an said "yeah its better" - it worked with their older cpus... but it certainly didnt this time

but... i noticed that winding the AMDs up helped them a bit in a few of the tests... i think it would be interesting to do the score for the cpus while overclocked also.

amd's PR naming is based on performance relative to their older processors (such as the athlon 1400 running @ 1.4ghz...) not on the competitions mhz...

the extra cache does help in many situations as you will see even with the intel cpu's... the extra cache HELPS... certainly with the amd architecture this would be the case... further revisions of the process will help... not hinder the case that more cache is better... some of the benchmarks show a marked improvement with the addition of the extra cache... heck intel's next cpu's will have a lot of extra cache as well :)

amd and intel both have reached the physical limitations of their respective cores... ie they can't push them farther (or much farther) than what we are seeing now...

sure you can oc it to run faster but think of how much lower the yields would be if you were to start producing them @ higher clocks?

another thing to note... if you oc the amd cpu... you wuld have to oc the intel cpu as well... no point in running oc'd v/s stock... that would not be fair @ all..
 
I think he ment to match clock speeds if u could get the 3000+ to 3ghz the p4 couldn't hang and all the intel ppl know it.
 
thats not feasible though :)

the 2 architectures operate completely differently... it would not be a fair comparison at all...
 
never said it would be :) but if it was all those loud intel voices would get real quite I think :eek:
 
thing is though... the intel northwood core was very overclockable compared to their coppermine and katmai predecessors

so intel people might not go too quiet just yet even if overclocked speeds were compared...
 
Originally posted by TechSupport
thing is though... the intel northwood core was very overclockable compared to their coppermine and katmai predecessors

so intel people might not go too quiet just yet even if overclocked speeds were compared...

intel DESIGNED the core to clock high rapidly in order to keep the overall performance crown from amd who had made big strides in cpu technology...

amd's core is designed to do more work per clock cycle hence producing more heat/chatter as it clocks higher... notice the extra layer on the pcb going to t'bred B and barton... this is to eliminate some of the cross-talk or whatever the technical term may be :)

intel's is designed solely around getting more clock cycles since less work is done per cycle.. hence it can be ramped up higher...
 
Well, something that I found really interesting was a site that compared intel and amd processors. In tests of 50 processors each Intel's processors were found to have a much higher mathmatical error rate compared to Amd's. Now this was like 2 years ago that I read this.

Also, for me anyways, speed doesn't mean as much as it used to. Technology and cpu speed move at an average rate of about 1 ghz per 1.25 years. If you compare Intel and amd price wise amd is always cheaper, and are almost always right in line with intel as far as over all performance concerned. Last year I built my pc, and I looked at both intel and amd. The amd 2100+ ($110) and the intel 2.0ghz ($365). Now the Amd runs at 1.8ghz compared to intel's 2.0 but at 1/3 the cost for .2ghz. Well needless to say it was a no brainer. And now there is the Amd 3000+ ($550)and the Intel 3.06ghz ($775).

Now if you also consider the fact that todays software (with just a few exceptions) really don't "need" anything over 1 ghz it all comes down to want. (granted I am a "the faster the better type person") But taking cost to performance ratio into perspective, Amd wins every time.
 
Originally posted by Maveric169
Well, something that I found really interesting was a site that compared intel and amd processors. In tests of 50 processors each Intel's processors were found to have a much higher mathmatical error rate compared to Amd's. Now this was like 2 years ago that I read this.

I would like to see this. Not that I don't believe you but linkage would help to back those claims.

Technology and cpu speed move at an average rate of about 1 ghz per 1.25 years. If you compare Intel and amd price wise amd is always cheaper, and are almost always right in line with intel as far as over all performance concerned. Last year I built my pc, and I looked at both intel and amd. The amd 2100+ ($110) and the intel 2.0ghz ($365). Now the Amd runs at 1.8ghz compared to intel's 2.0 but at 1/3 the cost for .2ghz. Well needless to say it was a no brainer. And now there is the Amd 3000+ ($550)and the Intel 3.06ghz ($775).

Processor speed doubling about every 18months according to moores law, not 1ghz every 1.25 years. If what you said was true we would only be up to 7ghz in 2008. I would hope we would be higher than that in 2008 :)


Now if you also consider the fact that todays software (with just a few exceptions) really don't "need" anything over 1 ghz it all comes down to want. (granted I am a "the faster the better type person") But taking cost to performance ratio into perspective, Amd wins every time.

Don't want to start a flame war, but I'll let my opinion be known that I disagree :)
 
Originally posted by jumpy
I would like to see this. Not that I don't believe you but linkage would help to back those claims

I haven't been able to find it again. The one site that I found it on no longer exists. But like I said it was 2 years ago, not sure if that would still be the case now.



Processor speed doubling about every 18months according to moores law, not 1ghz every 1.25 years. If what you said was true we would only be up to 7ghz in 2008. I would hope we would be higher than that in 2008 :)

Well I didn't use moores law, I was strickly speaking from my own experience, hence why I said "about every 1.25 years". I hope it moves faster than that as well. But unfortunatly moores law hasn't exactly proven true.



Don't want to start a flame war, but I'll let my opinion be known that I disagree :)

No flame war here. But I am courious as to your opinion and reasoning.
 
Originally posted by Maveric169
I haven't been able to find it again. The one site that I found it on no longer exists. But like I said it was 2 years ago, not sure if that would still be the case now.
Yeah, no problemo :)


Originally posted by Maveric169
Well I didn't use moores law, I was strickly speaking from my own experience, hence why I said "about every 1.25 years". I hope it moves faster than that as well. But unfortunatly moores law hasn't exactly proven true.
Not exactly proven true, but a good approximation


Originally posted by Maveric169
No flame war here. But I am courious as to your opinion and reasoning.
Believe me, it will become a big flame war if I start this (not necessarily with you), so I'll just keep my mouth shut. :)
 
you sure that the mathematical errors were not m$ related ?

I can't for the life of me see intel making cpu's that did "wrong" calculations...

the only thing that the article may have hinted @ would be precision in terms of floating point apps depending on how long these registers are... it is HIGHLY unlikely... nay... totally impracticle to say intel makes less precise hardware... they have very good solutions that do exactly what is expected and more... furthermore the SSE and SSE2 instruction sets allow for many apps to be written to specifcally take advantage of the northwood core's abilities and perform faster in apps...

when the hammers debut they are also supposed to support SSE and SSE2 fully and therefore also give an added advantage when it comes to certain apps written to take advantage of these instructions...

all in all a bumper year for cpu'd and gpu's shaping up with lots of new tech in the hard-drive and other storage areas as well as possible VDU development that will give us all low cost "lcd" type screens...

:)

I love technology...
 
That math error thing sounds really strange. If they counted wrong there would be tons of problems. It's not like Intels are known to crash...

The only thing I know was that the old Penitum 60 and 66 MHz had a calculation error with floating point numbers. It was removed from the P 75 MHz and up.
 
Originally posted by Zedric
That math error thing sounds really strange. If they counted wrong there would be tons of problems. It's not like Intels are known to crash...

The only thing I know was that the old Penitum 60 and 66 MHz had a calculation error with floating point numbers. It was removed from the P 75 MHz and up.

well that is more likely than something a year or two ago... because I have not heard of it...

like you said... intel's are not known to crash... and they are very solid cpu's... thats why reading that was strange...
 
If I remember correctly, it was 2 years ago. It was speaking strictly on the cpu mathmatical calculation, I don't think it was refering to the floating point errors of the past. I remember it was something like Intel had 50 something math errors for every 1 million(might have been billion) calculations, and Amd had like 20. But I haven't found anything like it since. So I really don't know. I haven't found anyone else that has done testing like that.
 

Members online

No members online now.

Latest profile posts

Also Hi EP and people. I found this place again while looking through a oooollllllldddd backup. I have filled over 10TB and was looking at my collection of antiques. Any bids on the 500Mhz Win 95 fix?
Any of the SP crew still out there?
Xie wrote on Electronic Punk's profile.
Impressed you have kept this alive this long EP! So many sites have come and gone. :(

Just did some crude math and I apparently joined almost 18yrs ago, how is that possible???
hello peeps... is been some time since i last came here.
Electronic Punk wrote on Sazar's profile.
Rest in peace my friend, been trying to find you and finally did in the worst way imaginable.

Forum statistics

Threads
62,015
Messages
673,494
Members
5,621
Latest member
naeemsafi
Back