Worst Anandtech review evar

[clarkey01]

He screwed up his makefile on the TSCP benchmark. (It fails to apply the -O2 option during compile) The 3500+ should score about ~300K, not 150K, beating the Nocona by a significant margin.

He copied the wrong data over on the MySql test-- the 3500+ wins both.

It's unclear he built the other synthetic tests correctly.

Some of the other tests are 32bit, not 64bit.

ubench is known to be buggy, as explained at another site.

This "review" is pure garbage.

The fact that he used a 3500+ instead of an Opteron 150 is a minor issue. The major issue is that he doesn't know what he's doing when he attempts to build benchmarks from source.

 

[Vette73]

Yea he went back and "FIXED" a couple things, but will not touch the others AND will not change his conclusion.

Sorry but it needs to be taken down. If you read some of his answers on the frontpage now he keeps saying nothing is wrong. Then someone proves him wrong, he fixes it. and so on. I don;t think he thought anybody would question him and he could slip one by. What BS. I expect this type of garbage from Toms, but not anandtech.

 

[clarkey01]

Toms Hardware/Anandtech on the same page

Never thought id see the day.

 

[ss284]

bdsm, jm0ris0n, lookouthere, Are you all blind or just insanely stupid? Did you not see lxie123's very imformative post about how the xeon 3.6 adds nothing to the 3.6e em64t (which is going to be 416 dollars soon) other than the ability to SMP and use a different package? The processor comparison is sound considering he made note of what he was comparing.

"That Xeon was just the regular 1MB L2 cache Prescott core with EM64T enabled, the same core that Intel just released in FCLGA775 ("P4 3.60F" $637 now, $417 in 2 weeks). The P4 w/EM64T will probably do better than the Xeon tested because 1) it uses lower latency DDR (non-registered), 2) can use DDR2 in i925 (won't help that much) and 3) i925 boards are more tuned for performance than Xeon server boards."

Did you not even see that posted? This sarcasm and conspiracy theory all show how stupid some people can be. Have you ever thought of reading a little more and seeing all sides of the story? Apparently not.

The comparison was fine given the situation and explanation. The benchmarks on the otherhand, do have some serious problems which have already been mentioned.


-Steve

 

[GopherBroke]

Final nail in the coffin for this review:

The John the Ripper test, which is now the only one that favors the Intel chip, uses the invalid CPUID check to see whether it will run optimized code or not. Only a processor that returns a CPUID containint "ntel" gets to use the fast assembler instructions, the Athlon is stuck with some ancient assembler written to the old 486 instruction set.

 

[jose]

Kris, why is there no mention that EM64 WON"T RUN w/ > 4gigs of Ram ?????

You should have used Opetron's vs Xeon's for the review.

Also on a sidenote , why use Linux at all if you won't optimize for 64bit.. Don't you load the latest 32bit drivers for Windows ?
Anyone who would run 64bit Linux would know what he/she is doing and would optimize their configuration...

You should run everything from the Original XP cd w/ no updated drivers whatever for your test..

FWI I've used Intel in my customer's servers for 14+ yrs, but my next recomendation will be AMD Opterons. I'm just waiting for Dell to carry them.

Regards,
Jose

 

[ScrapSilicon]

Originally posted by: jose
Kris, why is there no mention that EM64 WON"T RUN w/ > 4gigs of Ram ?????

You should have used Opetron's vs Xeon's for the review.

Also on a sidenote , why use Linux at all if you won't optimize for 64bit.. Don't you load the latest 32bit drivers for Windows ?
Anyone who would run 64bit Linux would know what he/she is doing and would optimize their configuration...

You should run everything from the Original XP cd w/ no updated drivers whatever for your test..

FWI I've used Intel in my customer's servers for 14+ yrs, but my next recomendation will be AMD Opterons. I'm just waiting for Dell to carry them.

Regards,
Jose

hehe...Dell would probably go defunct before that happens..

 

[Sonic587]

Originally posted by: ss284
bdsm, jm0ris0n, lookouthere, Are you all blind or just insanely stupid? Did you not see lxie123's very imformative post about how the xeon 3.6 adds nothing to the 3.6e em64t (which is going to be 416 dollars soon) other than the ability to SMP and use a different package? The processor comparison is sound considering he made note of what he was comparing.

"That Xeon was just the regular 1MB L2 cache Prescott core with EM64T enabled, the same core that Intel just released in FCLGA775 ("P4 3.60F" $637 now, $417 in 2 weeks). The P4 w/EM64T will probably do better than the Xeon tested because 1) it uses lower latency DDR (non-registered), 2) can use DDR2 in i925 (won't help that much) and 3) i925 boards are more tuned for performance than Xeon server boards."

Did you not even see that posted? This sarcasm and conspiracy theory all show how stupid some people can be. Have you ever thought of reading a little more and seeing all sides of the story? Apparently not.

The comparison was fine given the situation and explanation. The benchmarks on the otherhand, do have some serious problems which have already been mentioned.


-Steve

I don't see where some people are going with these opinions. Are you happy with shotty journalism and skewed results? Is it now the norm to have stand-ins for yet to be released hardware? Maybe AT should OC a A64 3800+, bench it, and title the review: "Benchmarks of AMD's new A64 4200+"? Is that the way we do things here?

It's clear that the Intel results were inaccurate as well. With the possibility of better RAM and a faster chipset, it should give the desktop version a boost. That leaves this article and it's conclusion worthless on BOTH sides. IMO, the 3.6F should have been compared to other Intel desktop/server CPUs. Leave AMD out of it until you can do it right :disgust:

It's confusing and misleading. A 3.6F Xeon is a stand in to bench against it's desktop counterpart's competitor. Got that part. We're going to be looking at things like gaming performance, encoding, right? WRONG! What we get is Linux and strange, unfamiliar tests. SuperPI scores, Jack the Ripper...WTF? This will indicate destkop performance? In what? And how would the tests relate to the average PC user? It's just not up the usually very high standards of the AT review team.

 

[clarkey01]

Sonic587 I couldnt have put it better.


everyone check the URL, it doesnt say tomshardware.com does it ?

 

[Accord99]

Originally posted by: jose
Kris, why is there no mention that EM64 WON"T RUN w/ > 4gigs of Ram ?????

Because that is false. Even pre-EMT64 Xeon platforms run with >4 gigs of RAM.

 

[BDSM]

SS284.. All I have to say to YOU is: Suck my hairy balls!!!!!

 

[clarkey01]

ss284 is pro Intel, Kill him lol


"open mind for a different view"


Nope, i dont like your view, its not true, so yoooooooooou"

whoo i can sing.

 

[Sid59]

Originally posted by: woodscomp
All I can say is 40%..... Wait till Intel releases there P4F 64 bit line of processor and the all the PR fools will crying again.

I hated the PR system when they used it before and I hate it now. It is so deceptive, and truly is inaccurate to anything about the processor. AMD could have come out with a different method of branding there processors than the PR.

Ohhh btw who invented the coin term Pentium?

Could it have been Intel...


Have a nice day boys!

Sorry for the thread crap, I was going to start my own thread but figured I would have gotten flak from the AMD fanboys for a double post type of thing.

and ...

what about intel for changing their scheme to a PR rating too?

 

[ThunderPC]

LOL, he further shows how much of a moron he is here. Since when did "unconstructive criticism" become SPAM. He is writing about computers and doesn't even know what SPAM is:



Jason,

I appreciate your unconstructive and unhelpful spam^H^H^H^H criticism? Seriously was there something here you wanted to tell me?

Kristopher

---------- Original Message ----------------------------------
From: "Jason Cothran" <edit>
Date: Mon, 9 Aug 2004 22:10:41 -0400

>After the lack of quality of that review, I see know why you have never done a CPU review before and hopefully will never attempt another. Stick to your optical drives. Do you know how many boards, forums, and usenet groups you are now the laughing stock of?
>

 

[Sonic587]

Originally posted by: ThunderPC
LOL, he further shows how much of a moron he is here. Since when did "unconstructive criticism" become SPAM. He is writing about computers and doesn't even know what SPAM is:



Jason,

I appreciate your unconstructive and unhelpful spam^H^H^H^H criticism? Seriously was there something here you wanted to tell me?

Kristopher

---------- Original Message ----------------------------------
From: "Jason Cothran" <edit>
Date: Mon, 9 Aug 2004 22:10:41 -0400

>After the lack of quality of that review, I see know why you have never done a CPU review before and hopefully will never attempt another. Stick to your optical drives. Do you know how many boards, forums, and usenet groups you are now the laughing stock of?
>

Is this a PM or email you sent to KK? If it is, please STFU and leave Anantech for good.

Regaurdless, I am seeing a lot of unwarranted hate and bashing on KK. What is wrong with some of you people? Yes, there are errors in the article as we have pointed out, but Kristopher has come clean that he made mistakes, and is working hard to correct them. Since when is everyone perfect? He's new to this! Some bad choices were made, and IMO, the review should be taken down until it's done right, but I see people wanting him to lose his job over this. Huh? Not to mention we are being plastered on the Inq and other tech websites for making ONE erroneous review. :| Do we and KK really deserve to be shot down this badly? The AT staff works hard to provide hundreds of fantastic reviews/previews/articles/ect that they give to us for free . One bad apple, and people seem to be out for blood. It just isn't fair.

Oh, and I'm sure Kris is losing a LOT of sleep over some fanboy tech nerds whining over at usenet. Apparently, they are all perfect gods and incapable of commiting a mistake.

 

[Zebo]

Originally posted by: GopherBroke
Final nail in the coffin for this review:

The John the Ripper test, which is now the only one that favors the Intel chip, uses the invalid CPUID check to see whether it will run optimized code or not. Only a processor that returns a CPUID containint "ntel" gets to use the fast assembler instructions, the Athlon is stuck with some ancient assembler written to the old 486 instruction set.

Not the only one.. But shows just how lame this review was. Now The midrange AMD chip wins every single real world bench and more than half the synthetics and he's still praising the prescott 64. What a sham.

Thank fully there are plenty of other review site out there which are engaging.. HardOCP, Hexus, aceshardware, X-bit labs, Firing squad, and Tech Report... in no particlar order.

 

[Markfw]

I agree the review is bad, but as others have said, Kris admitted a mistake was made, and there will be a revised review up. Please give them time to fix the errors. Are you perfect ? I am NOT !

 

[Yanagi]

So what? Anand did 1 bad review and it is now being worked on. C'mon AT is still top notch on my book. Everyone's entitled to make misstakes. We're only humans ya know. Bleh...

 

[ThunderPC]

Is this a PM or email you sent to KK? If it is, STFU and leave Anantech for good.

Wow, it must have been bed time when you posted that or perhaps you are just the biggest lamer fanboy in the world.

 

[3chordcharlie]

Originally posted by: clarkey01
"open mind for a different view"

hahaha James the cowardly lion

 

[alexruiz]

I don't think it was that bad..... Keep in mind that many of those benchmarks LOVE big caches (super pi for example). Saying that a P4 core beats a K8 in math performance is quite daring..... but those cache friendly benchmarks suggest so. A 3400+ would have been a very good reference point. Yes, quite incomplete, but valid after all

As far as bad review, even the worst Anand review would be top at many other sites. While I personally prefer techreport.com or aceshardware.com, Anand is still one of the top guns.


Alex

 

[Sonic587]

Originally posted by: ThunderPC

Is this a PM or email you sent to KK? If it is, STFU and leave Anantech for good.

Wow, it must have been bed time when you posted that or perhaps you are just the biggest lamer fanboy in the world.

Fanboy? I am in favor of AMD not being in those benches at all. Maybe you should have read a little harder. Those tests were shabby at best on both sides, IMO.

I apologize for blowing up, and I will remove some of my post, but it really angers me when I see unproductive personal attacks on Kris after he has admitted he's wrong. Where do you get off telling him "Stick to your optical drives."? Do you somehow think such a email/PM would help Kris correct his mistakes? You are only demonstrating that you take the staff and it's hard work for granted. If all you can do is lunge at Kris with nothing more than a fumbling attempt to tell him "U SUX", then you should by all means leave this website.

 

[DieHardware]

I agree the review is bad, but as others have said, Kris admitted a mistake was made, and there will be a revised review up. Please give them time to fix the errors. Are you perfect ? I am NOT !

So what? Anand did 1 bad review and it is now being worked on. C'mon AT is still top notch on my book. Everyone's entitled to make misstakes. We're only humans ya know. Bleh...

Anyone remember that power supply review?

I wonder if KK still has his job?

OK a little harsh perhaps, but after the PSU review you'd think he'd have learned to have someone check his reviews before posting them.

 

[DAPUNISHER]

Originally posted by: Markfw900
I agree the review is bad, but as others have said, Kris admitted a mistake was made, and there will be a revised review up. Please give them time to fix the errors. Are you perfect ? I am NOT !

What else could he do? Tell the whole world they are wrong and he is right? As to everyone making mistakes, I agree, in fact I said almost the same thing verbatim in a video forum thread about issues with a vid card review here. The problem is that the mistakes are coming more and more frequently, and when do you stop giving them a pass on it? For me this review was where the pass expired.

Besides, there are mistakes, and then there are mistakes, this was a complete train wreck, and there wasn't one mistake, but a whole slew of them. I can't reiterate enough that I think Anand is da man, or that the staff here is top notch, but I think this amounts to having your shortstop pitch.

 

[carlosd]

The review was crap. Those who says that it was fair don't have any idea of what they are saying, so won't waste time with them. Just let them learn. Then they will be able to say something valid.

That Xeon processor is the optenron competition, and have nothing to do against it:

http://forums.anandtech.com/messageview.cfm?catid=28&amp;threadid=1370368&amp;STARTPAGE=5&amp;FTVAR_FORUMVIEWTMP=Linear