Athlon XP at Anandtech

[lozina]

Now I just need ASUS to release their KT266A mobo and I'm ready to build my new system! I got the money waiting It's all up to you now ASUS! *is gonna have a great winter*

 

[lozina]

<<

<< Anand should've named the article, "AthlonXP: Intel gets its ass kicked." >>

but I wish if they're so for this, they would not be so conservative. Like the 1800+ should be 2000+, 1700+ should be 1900+, and so forth. >>



Definately... since the average user out there 'don't-know-jack' about cpus and always try to look for a single numeric statistic that they can try to base a product's performance by... They should name it "Athlon XP 3000" =P

 

[wfbberzerker]

take that intel! *pow* *kablam*

 

[Spook]

I was just looking up at Newegg.com, and they have them in stock allready... at a great price for $220 for the 1.8/1.53Ghz....

 

[spikegifted]

mmm... need a little bit of suggestin here... as you can see in my sig (as well as my rigs), i've a t'bird 1.2ghz oc'ed to 1.4ghz on a kt7a and 512mb of pc133... do you think it is worthwhile for me to upgrade to the xp 1.53ghz with ddr board (amd760 or kt266a board), with, of course, ddr ram??

 

[Shmorq]

<< Just for fun, I tallied up the benchmark wins for each processor:

Wins for AMD Athlon XP 1.53Ghz (1800+): 9
Wins for Intel Pentium 4 2.0Ghz: 4
The Athlon XP 1800+ and Pentium 4 2.0Ghz tied in 1 benchmark. >>

You forget to mention that the P4-2.0GHz barely edged out the XP 1800+ in the benchmarks it won whereas the XP 1800+ (and even sometimes the slower XP's) beat out the P4 by a large percentage.

I was waiting for Northwood to come out to upgrade my 1 year old P3-733MHz, but now seems like a good time to get the XP. It's hard to imagine a brand new high-end processor debuting for so cheap. I just can't wait for my favorite vendors to get it in stock.

[Edit] Didn't know that newegg already has it stock. They had it on their site yesterday but quickly pulled it...

 

[wfbberzerker]

why the heck cant newegg put the processors in order?

 

[lozina]

<< why the heck cant newegg put the processors in order? >>



we're overloading their servers with simultaneous orders

 

[Megatomic]

WOW!!

First off, great review. Very readable and the content is presented quite well.

Second, way to go AMD. Those are very respectable scores. I will now have a hard time restraining myself from upgrading.

 

[Adul]

dasm, there goes my savings .


I pick up the cpu for now and a ddr board later on. I also need a 8500. When is the 8500 showing up!?

 

[Macro2]

Bapco Sysmark 2001 and the rest of their trash benchmarks should never be used again. Thanks to Anand for exposing their fraud. Imagine including a test in Sysmark2001 where SSE is not used unless the chip is identified as "GenuineIntel". (WME7)
These hardware sites must all feel like they have been used to perpetuate the P4 joke...

 

[Megatomic]

<< Bapco Sysmark 2001 and the rest of their trash benchmarks should never be used again. Thanks to Anand for exposing their fraud. Imagine including a test in Sysmark2001 where SSE is not used unless the chip is identified as "GenuineIntel". (WME7)
These hardware sites must all feel like they have been used to perpetuate the P4 joke... >>



Agreed.

 

[SpecialEd]

awwwwww.... yeah... AMD sporting its new muscle!!!! Once again... Intel and AMD duke it out in a brains vs brawn match!! I love these battles!!

great article!

 

[ST4RCUTTER]

Bapco Sysmark 2001 and the rest of their trash benchmarks should never be used again. Thanks to Anand for exposing their fraud. Imagine including a test in Sysmark2001 where SSE is not used unless the chip is identified as "GenuineIntel". (WME7)
These hardware sites must all feel like they have been used to perpetuate the P4 joke...


Macro2,

Dude, the problem is with WME7, not Sysmark2001...the real affront is that Microsoft has yet to fully realize AMD as a competitor to Intel. Microsoft needs to get on the ball and give AMD its due. It's foolish to limit yourself to one manufacturer. Gateway should have learned this a while ago when they dropped AMD and then Intel had a huge processor shortage...it cost Gateway a ton of money when the dust settled. Gateway claims they need to streamline their offerings, but a Gateway computer with the ECS 8K7SA, a Palomino CPU, and 256MB of DDR would be one powerful and inexpensive solution. I wonder if the Intel Inside program had more to offer...

 

[Athlon4all]

I disagree Macro and seekingTao. Let me quote Anand:

<< Normally how you're supposed to detect whether or not a processor supports SSE is by looking at the SSE bit in the standard features register. If the SSE bit is set to true, then the processor you're dealing with supports SSE; if it's set to false, then your processor doesn't support SSE. The Athlon has this bit set to false since it doesn't have a full SSE implementation, however the Mobile Athlon 4, Athlon MP and the Athlon XP all have this bit set to true. The problem with SYSMark isn't really a problem with SYSMark, it's a problem with Windows Media Encoder 7. Windows Media Encoder 7 doesn't look at this bit to determine whether to use SSE or not, instead it reads the manufacturer code in the CPUID string of the processor. If the manufacturer code reads 'GenuineIntel', as all Intel processors report, and the SSE bit is enabled then WME7 uses SSE. This wasn't a problem until recently since AMD processors didn't have SSE support, but now with the Palomino core they do have SSE support but according to the WME7 SSE detection algorithm they will never run WME7 with SSE enabled. The reason behind this being that the manufacturer string outputted by all AMD processor is 'AuthenticAMD' which doesn't pass WME7's test for SSE compliance. Microsoft is aware of the problem and will fix it in the next version of Windows Media Player, but that doesn't change the issue with SYSMark since it relies heavily on WME7 for determining performance. >>

So you see, it had nothing to do with Bapco, it had to do with WME7. Plus I will point out that this didn't change the performance on T-Bird or Spitfire. Just Palomino and Morgan. So this wasn't a conspiracy or anything. Microsoft just didn't realize this error would effect the numbers on AMD CPU's. It's a good benchmark IMHO, and I'm glad to see that this was corrected.

EDIT: You read my mind ST4Cutter

 

[Megatomic]

<< So you see, it had nothing to do with Bapco, it had to do with WME7. Plus I will point out that this didn't change the performance on T-Bird or Spitfire. Just Palomino and Morgan. So this wasn't a conspiracy or anything. Microsoft just didn't realize this error would effect the numbers on AMD CPU's. It's a good benchmark IMHO, and I'm glad to see that this was corrected. >>



True statement, I can't really argue against it except for this. Bapco is pushing this SysMark2001 as a package. True, they are using WME 7 as a module in this package, and M$ is entirely responsibe for it's own product, but overall they (Bapco) are responsible for the finished/complete product and it's overall merit.

I had to deal with this kind of situation in my work as a controls engineer when I was a Systems Integrator. It sucks when you get bit for not catching a flaw with someone else's product when you use it in conjunction with your own product, but that's life and business.

 

[Macro2]

StaCutter,Athlon4All,
RE:"Dude, the problem is with WME7, not Sysmark2001..."

I hear you but it's Bapco that created sysmark 2001 and included WME7 and overwieghted it in their bench mark. A fair benchmarker would have tossed WME7 out of the mix because it was rigged to "GenuineIntel"...
Not only did Bapco choose to include it...they overwiegted it...like we sit around and encode media all day long.
Something stinks...but we already knew that...it's the P4.
You also forget that hardware sites have bee benching the Athon4/MP for 4+ months now and as a comparison to the forthcoming XP...all the time happily reporting the Athlon as inferior to P4 in Sysmark 2001 and WME7...all the while Bapco and Intel knew the real reason...
Thank goodness Anand finally figured it out...but if I was a hardware site I'd feel a little "duped" right now...

Mac

 

[Texmaster]

<< Shiznit!! Athlon XP smokes!!! Buying a Pentium 4 now doesn't make much sense to me ($225 for the XP vs $530 for the P4, not to mention the cost of RDRAM), however........

Unless AMD can ramp Athlon XP to 1.7-1.8GHz within the coming months, Northwood is gonna start bitch slappin' >>



I agree on both points.

AMD is clearly king of the castle right now.

I am curious however, how far they can go with the same archecture.

Looks like AMD is gonna keep it interesting

 

[Athlon4all]

See now Macro2 that's where you're wrong. WME7 isn't "rigged" to Intel, at the time, Intel processors were the only CPU's that had SSE support, and apparantley MS didn't see AMD using SSE in the near future. This wasn't a conspiracy, MS just didn't realize that AMD would be using SSE soon.

 

[joohang]

I'm so tempted to get a couple of these babies.

 

[Macro2]

Athlon4All,
I'm not blaming MS...although the coding was a little too quick and dirty...
The joke was that Hardware sites were using this bench to compare P4 to Athon MP/4 all the while not knowing that WME7 didn't even recognize the AMD chips SSE capability...
I agree with you on one count...Bapco2001 is a whole lot fairer bench now that AMD has provided a patch...duh...
Maybe Bapco Sysmark 2001 is just one bench but a lot of hardware sites use WME7 as a bench too...up 'till the rig was exposed that was 2 easy wins for P4...add a couple synthetic memory bandwidth benchs, a couple versions of Q3 benches and the P4 actually starts to look good and we know that can be deceiving...
Mac

 

[dejacky]

hmm the XP1500 (1.3Ghz) looks very tempting..i wonder if my 300watt Antec303x power supply would be enough for this cpu?
-dejacky

 

[GTaudiophile]

Haha. I am kinda like the IT guy for the small company I work for. I just received a call from Micro$oft, asking me if I wanted to attend their official unveiling of Windows XP. The M$ rep said that installation and operation of WinXP Pro would be demonstrated. He also said that Intel would be on-hand to demonstrate how the P4 is optimized to run WinXP. I was offered WinXP RC1 as an incentive to come. Needless to say, I declined the invitation, and when the M$ rep wanted to list my company in his OEM database, I refused to give him the information. Besides, I already have an official copy of WinXP RC1, AND I have a not-so-official copy of WinXP Pro Corporate, Build 2600

 

[Spook]

I thought it was a good article, but doesn't Anand usually put some kind of overclocking results in his articles?

I would like to know how well it overclocks....

 

[vampiro]

Finally, the P4 killer has arrived! All hail the Athlon XP.