Toms super cpu comparision

[Matisaro]

I think im the first one to post on it so here goes.

"However, you mustn't forget that if you're in the market for an AMD processor, be prepared to shell out at least 50 dollars for a powerful cooler. Cheap run-of-the-mill coolers costing as little as 15 dollars can only be used for speeds up to 1200 MHz."

So all those people who got volcanos for 10 bucks are screwed? What about the default heatsink, this line is utter bs!

"At the low resolution of 640 x 480, the Pentium 4/2000 is in the lead, while the AMD Athlon XP 1800+, boosted by the enhanced Nvidia driver, takes the lead at the high resolution. "

So the p4 wasnt boosted by the same driver? Sounds a tad biased to me, what does everyone else think?


"A comment on Sysmark 2001 - despite the fact that AMD sent us a patch activating the recognition of the SSE command set in Media Encoder 7.1, which is included in the benchmark suite, we didn't use it. The reason for this was that users can't update the Media Encoder in reality. There is no official update from BAPCo yet. "

LOL, ok tom, whatever you say.


Another one of the reasons is Windows XP, being forced as the successor to Windows 98/ME and Windows 2000 by Microsoft. In comparison to the past, the Athlon XP can profit from Windows XP, probably more than Intel CPUs. It seems that AMD helped Microsoft get the optimum out of its new CPUs. "

So much for those sse2 optimisations.



Thats pretty much what I gathed from that review, it seemd as though tom had to grudgingly admit the xp is faster, but had to add those small shots at amd in there. Maybe I am reading into the comments a tad too much.

Anyways, decent review, if not way late(been done weeks ago).



~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[charliec2uk]

What are you trying to say exactly?

Woke up in Borneo one day, stuck in the pouch of a big marsupial, but the boat was made of marsipan!

 

[jlbigguy]

There were no real surprises in the article, other then the PR ratings adopted by AMD being a bit conservative when compared with P4 clock speeds.

<font color=blue>This is a Forum, not a playground. Treat it with Respect.</font color=blue>

 

[Matisaro]

Well, the heatsink comment struck me as utter bs, and while I read the article I posted all of the biased/untrue comments in the review. A highlights of sorts.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[bront]

The HS thing isn't 100% true, however, a better heatsink can lengthen the life of other parts, as heat plays off itself. A hot running Video card heats up the MB, and in turn heats up the CPU.

I haven't seen any heat tollerances for any components listed in many of the specs, how hot can a CPU run without problems? And other components for that matter?

60 FPS, 70 FPS, 80 FPS Crash!
Daylight comes and I have to go to work :frown:

 

[bront]

What I found interesting, is that the one benchmark where AMD got it's butt kicked, was the one that was optimized for SSE2, and yet the review didn't address that.

However, I thought that in general, it was a thorough investigation into CPUs and how they work.

As far as the driver issue, Nvidia's first set of drivers weren't optimized for the AMD chips, so that is where that reference comes from. It was a short lived problem though.

60 FPS, 70 FPS, 80 FPS Crash!
Daylight comes and I have to go to work :frown:

 

[girish]

Yes, Tom has done it again!

After the dubious overheating test, he is now shooting at AMD again! Tom has been pretty neutral regarding processor brands until recently. Now surprisingly he seems to be inclined towards Intel.

I had browsed through the article, and was disappointed at exactly the same points he raised. Well, i was disappointed by his CPU heating review too, and I had too much to say so I write a complete <A HREF="http://www.64bits.org/cpuheat/cpuheat1.htm" target="_new">article</A> to express it.

This CPU comparison could also take a article, but Mat made it short and to the point. I agree with all his comments.

girish

<font color=red>No system is fool-proof. Fools are Ingenious!</font color=red>

 

[FatBurger]

"A comment on Sysmark 2001 - despite the fact that AMD sent us a patch activating the recognition of the SSE command set in Media Encoder 7.1, which is included in the benchmark suite, we didn't use it. The reason for this was that users can't update the Media Encoder in reality. There is no official update from BAPCo yet. "

That makes perfect sense to me. If the consumer can't have it, it's an infalted benchmark.


I think the first comment (about heatsinks) is pretty off, but I'll re-read the article and wait for Frank to return my email before commenting further.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>

 

[Matisaro]

Burger, the benchmark does not enable sse2 due to an admitted flaw(using the string genuine intel to detect sse support). I see that as a glaring flaw especially on a benchmark, if he had such moral objections to using the patch to level the playing field, he should not have used an obviously broken benchmark in the first place.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[Guest]

Actually, if anything, I felt the wording of the review was skewed towards AMD. So I think you could probably see it either way, depending on your preferences.

Which means it probably isn't biased at all.

 

[sakattack]

Great benchmark:

"-Are you a Genuine Intel CPU?
-No? Then you DO NOT have SSE!
-Proceed without enabling SSE support on that poor non-Intel bastard."

THG: "Well, we are benchmarking all CPUs for SSE performance with a benchmark that does not enables SSE on non-Intel CPUs" !!

Woow, that's so professional !!!

 

[girish]

strange method to ascertain SSE capability, look for "GenuineIntel" and thats it! as if <i>ALL</i> intel processors right from the 486 (that returned CPUID) to P4 support SSE!

I remember a software called SolidWorks99, it dint work on <i>any</i> AMD processor, you wouldn't imagine what error it would return!

<font color=blue>AMD processor detected!</font color=blue> and did a exit!! maybe allergic to AMD processors, all of them - I tried it on K6, Duron and Slot-A Athlon.

girish

<font color=red>No system is fool-proof. Fools are Ingenious!</font color=red>

 

[Matisaro]

The results were favoring AMD because the xp1800 is the faster chip, but the words of the review seemed to take every possible shot at AMD they could.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[Guest]

...well, our pal Tom actually admitted that XP1800 was the fastest CPU, he crownded the XP1500 "Best value".

It is hard to admit the truth for a man with Toms "Pride", so he had to give AMD some punches, I think it is plain pathetic to do so... And he even left the XP1900 results out, because Intel had looked plain silly if P4 hadn't won som benches...

But, it is quite good to look at the numbers anyway, and they will surely annoy some people! LOL

***A-Man***

...is it?...NO, it's AnotherMan...

 

[caater]

Great article, but i was hoping to see p!!! in it.. Or will 1.26 Tualatin produce such pathetic scores..?

 

[AmdMELTDOWN]

>The results were favoring AMD because the xp1800 is the faster chip, but the words of the review seemed to take every possible shot at AMD they could

boo hoo, it seems to me that the P4 systems Tom uses are less prone to headaches! than a AMD system, btw where you around when Tom would take a shot at Intel almost every day?

ooh, the AXP wins some non-SSE2 tests, ooh, I should be afraid, oooh! AMD trounces the P4, ooooh, I must tremble, oooh.

"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"

 

[Matisaro]

Yeah lol, I didnt even notice that he left out the results of his pre release chip he tauted so much.

PS: Tom, just because the faster athlons have deeper burn makes dosent mean that they came off of different production lines, it just means they ran on a different lazer tool(AFTER PACKAGING)!

Meltdon: LOL, poor guy, must hurt you like it was you getting stomped, and not intel.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[AmdMELTDOWN]

>strange method to ascertain SSE capability, look for "GenuineIntel" and thats it! as if ALL intel processors right from the 486 (that returned CPUID) to P4 support SSE!

girish, sse has never been available to Athlons until the palamino, bapco sse support for AXP does not magiclly appear, especially when sse was an Intel only optimisation.

bapco suite can not enable a sse flag on a future processor it knows nothing about.

btw, nothing in the 486 days had sse or any simds, you guys are just f'n babies.

"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"

 

[FatBurger]

nothing in the 486 days had sse or any simds

That's the point they're making.

bapco suite can not enable a sse flag on a future processor it knows nothing about

It should be able to look for SSE implementation, not 'GenuineIntel'. Someone correct me if I'm wrong.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>

 

[Matisaro]

You are right burger, instead of doinga routine to determine if sse was available they took the cheap and easy way of having the genuine intel string enabler, which of course was fine when intel was the only game in town.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[micronot]

I liked the CPU comparison that Toms Hardware did but...

If they wanted to "only" compare the cpu's, would it not have been a fairer comparison to have both systems running DDR-SDRAM??? Perhaps using a KT266A and P4X266 motherboard from the same manufacuter would have done a better job isolating the CPU.

another point that I thought was a bit odd was:
"The Lame MP3 Encoder...the Intel Pentium 4/2000 makes a clean sweep of the competition". What do you mean a clean sweep? P4=185, XP=186... that is not a clean sweep, that is practicaly a tie.

A question: Why did the P4 do so much better in the WinAce test????

 

[AmdMELTDOWN]

>It should be able to look for SSE implementation, not 'GenuineIntel'. Someone correct me if I'm wrong.

yo, smartass you think it's easy to code a benchmark suite? huh? what, what? you think you're so smart why don't you get a job with bapco, since you are a better coder than them!

sse is owned by Intel, I'm sure when checking for 3dnow it will prob look for "Authentic AMD".


"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"

 

[Matisaro]

I am betting win ace is sse2 enabled, but the p4 has always been good at compression/decompression tasks.

Yeah less than 1 point is hardly sweeping, but that was just par for the course regarding the textual portion of that comparision.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~

 

[Guest]

I also would have liked to see the latest PIIIs - not to mention the Celerons and Durons. As it is, this is a slightly less than comprehensive overview.

 

[Copenhagen]

Actually I was surprised that the Athlon XP1800+ didn't do better agaist the P4 2000 Willamette.

Overall I share the same feling as Matisaro when he speeks about Tom being biased towards Intel at the moment. I think the bias is due to an underlying belief in the P4 design, which will allow it to out-gun the Athlon XP in the long run, probably already in 2-3 month time.

Regarding his remarks on shelling out an additional 50$ on a CPU cooler, I think it is true at least when an effective low noise solution is required.