AMD XP 3000+ Benchmarks

[BlueHornet]

http://www.detonators.net/

to

http://uhaweb.hartford.edu/ozarca/amd/index.htm



peace,
-=<font color=blue>i3lue}{orneT</font color=blue>=-

 

[eden]

If AMD kept lower the clock speed each time, obviously the XP3000 won't compete well. Cache doesn't play a role everywhere. If they had used an XP3000 at 2.25GHZ like the XP2800, maybe it'd help, otherwise jump even further in clock speed. Those achieved 2.4GHZ, should be enough to compete the 3.06GHZ. IMO there should always be a 400MHZ gap between the CPUs to compete equally.

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden. <P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 02/09/03 03:56 PM.</EM></FONT></P>

 

[Makaveli]

I agree with u eden. I didn't think it was a good idea or give it a higher XP rating and lower clock speed cause of the Extra cache. Most people know that cache doesn't always give extra performance depending on the application.

Cause of this I will be getting a Higher clocked Athlon XP2600 over an XP2500 barton.

More clock speed is more important to me than extra cache.
Plus I don't think athlons will benifit much from 512 L2 as the P4 did. I might be wrong but Athlons are far better dealing with less cache than P4's.

I do agree the XP2500 will be a great overclocker.
AMD needs more clock speed not more cache to be competitive with intel. I don't mean to having the same clock speed, but just to keep the gap under 400-500mhz where the higher IPC of the athlons keeps it right there with P4s.

 

[Spitfire_x86]

IMO, AMD should clock Barton 3000+ higher than AXP 2800+, at least 83 MHz higher.

<b> "You can put lipstick on a pig, but hey, it's still a pig!" - RobD </b>

 

[HardWareBoss]

Spitfire your right. Barton should be a lot faster than XP2800. Why pay $600 for a XP3000 when you can buy XP2800 $380. Makes no sense too me. P4 3.06 at lot faster than XP3000 for $580.

 

[Schmide]

Judging from the rendering benchmarks, the extra cache definitely shows some serious crunch ability. I think there were some minor anomalies in those benches, as the XP2700 was surprisingly show worthy at its lesser speed. I look forward to seeing the actual thermal output at this speed, it could prove to be a power efficient floating point titlist.

Dichromatic for your viewing plesure...

 

[Schmide]

I didn't realize that the XP3000 ran at XP2700 2.17ghz speed. I guess the XP2700 was duly show worthy.

Dichromatic for your viewing plesure...

 

[Quetzacoatl]

not a bad show though, has a little better scaling, so we might see those 2500+ Bartons do 1.83Ghz to a good 2.7Ghz on extreme air, and 2.55Ghz or so on decent air cooling. Otherwise, just another step up for the Athlon till the Hammer takes over. Let's see some Barton 1700+ action!

Instead of Rdram, why not just merge 4 Sdram channels...

 

[Schmide]

I've been lurking around the various sites and it seems that <A HREF="http://www.aceshardware.com" target="_new">Ace's</A> was first out the door with their article. Must be a time difference/sleep thing. Their spin definitely showed the 3.06's shine.

Dichromatic for your viewing plesure...

 

[Schmide]

It seems <A HREF="http://www.extremetech.com" target="_new">Extreme Tech</A> has followed (they could have been first as I have no search order and may not include some sites that contain reviews that should be included)

Dichromatic for your viewing plesure...

 

[Schmide]

It seems a bunch of sites have entered the fray. Still waiting for <A HREF="http://www.tomshardware.com" target="_new">Tom's</A> front page to change.

<A HREF="http://www.amdmb.com/article-display.php?ArticleID=224" target="_new">AMDMB</A>
<A HREF="http://www.amdworld.co.uk/bar3000.htm" target="_new">AMDWorld</A>
<A HREF="http://www.sudhian.com/showdocs.cfm?aid=324" target="_new">Sudhian</A>
<A HREF="http://www.bit-tech.net/review/180/" target="_new">BitTech</A>
<A HREF="http://www.hardwarezone.com/articles/articles.hwz?cid=2&aid=628" target="_new">HardwareZone</A>
<A HREF="http://tech-report.com/reviews/2003q1/athlonxp-3000/index.x?pg=1" target="_new">TechReport</A>
<A HREF="http://www.xbitlabs.com/cpu/athlonxp-3000" target="_new">Xbit Labs</A>
<A HREF="http://www.hothardware.com/hh_files/CCAM/amd_axp_3k+.shtml" target="_new">Hot Hardware</A>

Ok I'm board of reading about this processor.

Dichromatic for your viewing plesure...

 

[Schmide]

And it hits the front page. <A HREF="http://www.tomshardware.com/cpu/20030210/index.html" target="_new">Tom's XP3000</A>

Dichromatic for your viewing plesure...

 

[Schmide]

Not a bad article, nice color and technology. Cool information on the XP multipliers. It was nice to see the scalability of Intel/AMD. I thought the cute kid at the end was a shameless ratings ploy, but still in good taste.

Dichromatic for your viewing plesure...

 

[baldurga]

In response to Makaveli:

Plus I don't think athlons will benifit much from 512 L2 as the P4 did. I might be wrong but Athlons are far better dealing with less cache than P4's.

You are right. Check it at this page on <A HREF="http://www.anandtech.com/cpu/showdoc.html?i=1783&p=5" target="_new">anandtech</A> review. Interesting to see and well explained why AMD and Intel show different performance gain.

About price/performance, IMO Barton 2500+ is what I would buy and overclock it as much as I can. Well, it's also clear it depends on what are you run usually in your computer. If cache is not involved, forget Barton.


Still looking for a <b>good online retailer</b> in Spain :frown:

 

[slvr_phoenix]

Not a bad article, nice color and technology.

Not an entirely bad article, but again, typical THG biasing towards AMD. Note how very downplayed the fact that the Bartons couldn't even begin to challenge equally-rated P4s was. Heck, notice now downplayed Bartons being beaten out by lower-rating T-Breds was. It was barely talked about at all.

And yet what kept popping up again and again? Statements like "<font color=blue>putting the overclocked Athlon with 2500 MHz at the head of the pack</font color=blue>" did, even though the Barton was severely OCed to get there. And yet almost every single time that these statements were made, what processor was <i>actually</i> at the top of their benchmark charts? <b>The OCed P4 was!</b> Proving once again that no matter what the benchmarks say, THG staff will stretch logic and reality to great lengths just to say good things about AMD while ignoring Intel. They couldn't possibly just say "Even the best OC we could get with Barton wasn't able to match our Intel OC." No. A statement like that wouldn't be AMD-biased, so THG can't say it.

Cool information on the XP multipliers. It was nice to see the scalability of Intel/AMD.

I agree here. They did a pretty good job, even if they did put in so many Barton OCs that it made AMD look much better than they actually did. If you redid those charts without any OCs, AMD would have looked much worse.

Yet I still have to wonder why the AXP ratings went from 1800+ to 2000+ to 2400+ to 2500+. The 2100+ and 2200+ were complete skipped! Oh wait, sorry. How silly of me. I know why. It was to keep from having to show the shame of the Barton 2500+ occasionally being tied by a T-Bred 2200+!

I thought the cute kid at the end was a shameless ratings ploy, but still in good taste.

I think it was a conspicuous attempt at saying "AMD, Intel, it's <i>all</i> good." in a way that readers can't get mad over.

Back to the Barton discussion though... The one thing that I have to say after looking at various Barton reviews is: "What the hell was AMD thinking?" The cache <b>clearly</b> doesn't give it an extra 300 rating points. Worse, they've duplicated two rating numbers now. The 2600+ with both 2x133FSB and 2x166FSB was bad enough, but now to have two different chips for 2800+ as well? (And have the Barton 2800+ perform noticably worse than the original T-Bred 2800+?) AMD is clearly f-ing things up big time. I'd even go as far as to say that AMD is thumbing their nose at the customer. This just isn't right.


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?

 

[Mephistopheles]

I agree with slvr_phoenix. THG is quite often AMD-biased, and often contrary to reason.

It´s annoying to read again and again in CPU benchmarks that little "comment" telling us "AMD CPUs lack enhancements or optimizations" in EVERY review for months now, as if explaining why AMD "lost" in that benchmark. Well, if Intel CPUs get better software support, then that´s a big credit for them, isn´t it? The reviews are irrational, indeed! It sounds like they´re saying, "Those poor AMD Cpus lack support and trail behind P4s, with their "unlawful" competition from optimized software. Pity on them" Unlawful in what way? It´s quite absurd, what they say.

Plus, of course, that the reviewers sometimes use their own graphics to get to unreasonable conclusions regarding Intel/AMD competition. Hope it doesn´t get to that point with Nvidia/ATI.

 

[slvr_phoenix]

Not just that, but Intel is hardly preventing AMD from implementing SSE2. AMD <i>could</i> support SSE2 if they wanted to.


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?

 

[ctbfalcon]

Everyone is talking about comparing the latest athlon to the latest P4. but you can't do that, how can you compare two cpus that are 894Mhz apart!! The only thing that it is good for is to see who has the fastest CPU. But who cares? Thats not the point. Who make a better cpu? is the question

Overclocking has no impact other that to see what future cpus are capable of doing. Frankly i dont want to void my warrenty to make more cpu power that i dont need anyway. I run a "outdated" athlon 1ghz, and every game I run is still playable by far (ut2k3, mohaa, quake3 etc).

With that being said, you can only compare cpus that are as close in mhz as possible. Apples to apples people, not apples to oranges. So MHZ for MHZ athlon beats intel! And with the fact that amd is cheaper it is a no brainer.

 

[cowpower]

It´s not about MHz to MHz but PRICE to PRICE (while regarding performance). I´m not completely sure but I guess even there Intel plays the leading role (at least in the high end class)

 

[ctbfalcon]

Also how can anyone say that THG is amd biased? In every benchmark not testing CPUs, they use an Intel! So the way i see it, if they were biased they would use amd for those benches.

 

[Mephistopheles]

But does that really make sense? OK, apples to apples (Mhz to Mhz) but if you buy an intel CPU, you get "more apples"! Ever though of that? Mhz for Mhz, amd is of course better, but the latest CPUs from intel are (as you pointed out) 900Mhz ahead of amd. The net result? 3.06Ghz P4s are the fastest.

I believe that, for the end user, the issue might be "I want this or that level of performance". So, I think you CAN compare CPUs that are NOT as close in mhz (if you couldn´t, how the hell would you compare CPUs from AMD to intel?). What you can also do is compare the performance of the latest Intel CPU to the latest AMD Cpu. You can also compare price/performance.

But comparing the processors Mhz for Mhz is pretty useless. The net result of (efficiency per clock cycle)*(clock), along with software support, should be considered. And price/performance, where AMD is very good from 2000-2800+, of course.

Oh, and by the way: I agree with you in regard to OC!

 

[ctbfalcon]

When you are talking about value PRICE to PRICE matters. But not in performance. Just look at the bench AMD:
P4 2GHZ in quake3 is 251fps
AMD 2400+ 2GHZ 272

Without going OCed lets look at the highest AMD 3000+(2166)
and P4 2.4, 234MHZ behind but still amd takes the lead.

It is clear to me. Price may be the deciding factor. But you need to take into concideration intels history. I dont want to buy a new mobo just because intel decided to change the socket. so theres another $150 plus cost of cpu. Every amd athlon has used the same socket. so up till the barton i can use the same mobo.

 

[ctbfalcon]

Yes it does make sence, but it dependes on your goal.
If all the end user wants is the latest and greatest, yes the fastest CPU will do.

But my point was that this end user will for example, talk to his buddy with the fastest amd and say "i have the better cpu" well no he does not. He has the fastest CPU but not the better CPU.
Because if amd made a athlon at 3ghz it would be faster.

MHZ for MHZ AMD is a better CPU.

 

[garfieldDK]

Is it just me, or are Frank Völkel and Bert Töpelt really smoking crack when they say that the Athlon is almost five years old? First Athlon review at THG is dated August 9, 1999 and I remember my purchase of my Athlon Classic 500MHz in the autumn 1999. Also the article says that the Thunderbird is from 1999 (acording to the picture with the Socket-A overview of the Athlon). Damn, the articles at THG aren't what it used to be.


Simon

 

[dhlucke]

6-23-1999 Athlon release date

So, a bit more than 3.5 years. The design is what's probably 5 years old though.

<font color=red>
<A HREF="http://kevan.org/brain.cgi?dhlucke" target="_new">If you were to have sex with your clone would that be considered incest or masturbation?</A></font color=red>