Athlon XP 2000+ Review

[Dr Mordrid]

And MSPro 7 will make SSE2 even more advantageous. I can't say why...but it'll be a necessity. Unfortunately SSE2 isn't likely to show on AMD's 'til the Hammer core hits....

Dr. Mordrid

[Jerry Jones]

Indiana,

I *have* the Athlon 1.4gHz Thunderbird.

You seem to consider the ECS motherboard, the K7S5A, as "questionable" where the Anandtech site's tests were concerned.

I'm running my Thunderbird without *any* stability problems and it's plenty fast. In fact, this SiS 735 chipset has been proven to be faster than the AMD chipsets. So I am surprised - no - APPALLED you would make a remark like that when it flies in the face of the tests that have been done here:

http://www.amdmb.com/article-display.php?ArticleID=112

"ECS and SiS have received lots of recognition from other publications for the speed and performance of this chipset and motherboard."

...and here:

http://www6.tomshardware.com/mainboard/01q3/010924/

"Our SiS735 chipset review back in June revealed that the single-chip solution is able to compete with strong competitors such as AMD's 760 chipset. As a matter of fact, the SiS 735 clearly beats both the Ali MaGiK-1 and VIA's KT266 chipset, and still outperforms the AMD760 in the majority of our benchmarks."

When it comes to my *first hand* observations of the 1.4gHz Thunderbird, I know for a fact it is not as fast as even a 1.7gHz Pentium 4 chip as I use two systems with both and I can compare them first hand.

So I didn't really want to be mean and rub it in to the "moderator," but he kept insisting on stating something that know - based on first hand experience - to NOT BE TRUE.

(And let's not even deal with the issue of his disrespect and his ridicule toward me - something I certainly have done nothing to deserve.)

My Dell Dimension 8100 - with 512mb of RAMBUS memory - outperforms my AMD Athlon machine with 512 of 2100 DDR memory in

a. DV rendering

...and...

b. mpeg encoding

...using Doc's test.

If one visits this Web page to see the table that many of us have worked together to create, one will observe that my 1.7gHz machine still has the fastest time for DV file creation (DV rendering)... even when pitted against an AthlonXP 1700+.

http://www.hyperactivemusic.com/MS%20Pro%206.0.htm

However, the fastest Athlon machines outperform the Pentium 4 when it comes to MPEG encoding.

The point I made in my original post, which - by the way - did not deserve the kind of remark that was posted by the "moderator" - merely called attention to the fact that the new Athlon XP "2000+" truly *is* faster than the Pentium 4 2gHz processor in virtually *all* benchmarks.

In that respect, my main issue with Greebe's remark was that the "Thunderbird" is...

a. not faster than the Pentium 4 2gHz processor - and I quote - IN ALMOST EVERY BENCHMARK.

b. the Thunderbird core is now outdated and clearly not as fast as the newer Athlons.

The fact is - the Pentium 4 running at 2gHz will even outperform the Athlon 1.4gHz Thunderbird in a majority of benchmarks "important to the average user" assuming gamers probably outnumber those of us doing non-linear video editing.

I counted five wins on the Anandtech site:

1. Content Creation Winstone 2001
2. Internet Content Creation SYSMark 2001
3. Overall Application Performance
4. Quake III (Yes - this is the big consumer category)
5. DroneZ (Yes - this is another big consumer category)

Also, Indiana, do you base your claim, quote: "people I know that are into video-capturing/-editing are NOT using WME at all, but DivX, TsunamiMPEG and various other encoders" on a scientific survey?

Or is this claim based on your own personal observations (in which case it is not scientifically valid)?

"Analtech" is really an editorial comment you made that I feel was in poor taste.

But if you don't feel that site has any credibility, then visit this site:

http://www.tomshardware.com/cpu/01q3/010827/p4-06.html

Again, the benchmark tests revealed the 2gHz Pentium 4 to be faster than the Athlon Thunderbird 1.4gHz chip.

http://www.tomshardware.com/cpu/01q3/010827/p4-12.html

"Conclusion
Benchmarking Pentium 4 against Athlon is a controversial thing. The right choice of applications can either make the one or the other look better or worse. Intel's new Pentium 4 at 2 GHz however is beating AMD's fastest Athlon in the majority of frequently used applications and deserves therefore the title 'fastest PC processor'. Intel won the 2 GHz race and it also retook the crown of the maker of the fastest CPU. I am hesitant, but in a way we have to respect those facts and congratulate Intel."

Greebe, you are a disrespectful little fellow, aren't you?

[Greebe]

"Little fellow" HAHAHAHA
You certainly don't know me, even more so have no idea where I have been and why I say what I do (obviously).

As far as quoting anything from Anand and Ol Tommy boy... I've been on the scene pushing the limits and knowing what is better than both of them combined... maybe you should realise that their words aren't gospel. To be honest many on the forums outside the realm of DV forum (did you know that there was more to the MURC than Desktop Video?) that'll back me up on that 100%

Like on the X-Files... the Truth is out there... you just gotta dig through a whole lotta shit to find it... especially @Anandhype and Tom's Fubar Guide

[Jerry Jones]

My, my, my.

I had no idea you were so smart.

I'm sure glad you told me.

Otherwise, I never would've realized you know so much more than two of the most popular motherboard Web site authors on the internet.

I'm sure they would be very grateful if you would grant them the honor of "digging through all of your shit" so they could be just as smart as you are.

Wow.

I'm so impressed.

[Indiana]

The ECS board surely is not the fastest SocketA board, both the KT266A and the NForce are quite a bit faster. If you really want an anandtechlink, here for the KT266A, NForce has about the same speed.
And a lot of users especially of the ECS board (this seems not to be SIS' fault) had/have MAJOR problems with TBirds@1.4GHz if you have read e.g. the AMDMB and Planet3DNow! fori. For some it works, for other it simply won't (looks like bad quality control, like ABit). At least some revisions of the ECS seem to have major problems with 133MHz FSB / C TBirds. Then there are those cold-boot issues unless you connect the L7-bridges, the board seems to need a really strong Powersupply, there are compatibility issues with Infineon-RAM, and the non-functional ACPI/powersaving,,...
There even is a poll/thread in the amdmb ECS-mainboard forum which was meant to prove that there are not that many problems with those boards. Well I guess the result of only 62% claiming no problems and 38% voting for "problems with ECS" speaks for itself.
This page lists the K7S5A as a relabled PCChips mobo (M830LR)

The P-IV wins you claim (in this specific review):

1. Content Creation Winstone 2001

Not really a win, it's a draw (like e.g. the Business Winstone2001) - and it's rated as a draw in this review, either.

2. Internet Content Creation SYSMark 2001

This is the IMO "flawed" one (one single subtest dominating the whole result).

3. Overall Application Performance

It's not a test on it's own, just the average of two test-suites (with one being the flawed one above, the other one won by the Athlon)

4. Quake III (Yes - this is the big consumer category)

Agreed, but the KT266A and NForce tests already show another picture

5. DroneZ (Yes - this is another big consumer category)

Ah come on, not DroneZ. Yes, it's one of the newer titles, and thus often tested but how many people play it? The game is crap, btw (IMO, of course).

Also, Indiana, do you base your claim, quote: "people I know that are into video-capturing/-editing are NOT using WME at all, but DivX, TsunamiMPEG and various other encoders" on a scientific survey?

Or is this claim based on your own personal observations (in which case it is not scientifically valid)?

Well I said "people I know", I guess this should be clear. I never claimed to have done a scientific research on this and I guess you would be surprised how many of those "scientific surveys" actually are "scientifically valid"

"Analtech" is really an editorial comment you made that I feel was in poor taste.
But if you don't feel that site has any credibility, then visit this site:

http://www.tomshardware.com/cpu/01q3/010827/p4-06.html

Sorry if you are disturbed that much by this comment, but like already said I believe that Anandtech is not as bad as THG (=Toms Hardware Guide).


You're right in that the PIV 2GHz is faster for most things as the old TBird1400 (it would be ashaming if it weren't, or?), I thought you were speaking of the Palomino. Here, a 2GHz PIV is naturally faster in some areas, but the Palomino will still be the overall faster for most used older apps - especially when paired with a decent mainboard.

[Jerry Jones]

Jerrold Jones

All I know is my ECS K7S5A performs flawlessly...

...and I know of one other person who is running one and he's not having any problems either.

On the other hand, I *did* buy the 430 watt power supply from Enermax when I built it, knowing that the 1.4gHz Thunderbird needs a lot of juice.

I suspect some issues that might be attributed to the board are - in reality - power supply issues.

Lots of people think they can get away with a skimpy power supply.

The first book I read on building PCs was written by an author who insisted on going the extra mile with the power supply. In fact, he considered it the #1 weak spot in a computer.

The Thunderbird 1.4 also can run fairly hot - which is why it's important to make sure one's case is properly ventilated and fanned.

It's also important to use good thermal grease (the best is silver paste) when installing the processor heat sink to bring the temp down as much as possible.

[Greebe]

Actually you don't neccessarily need a PS which is that large unless you have other componets which demand alot from it. Example: I'm still running an Athlon500@850 (for the last two years none the less and it's power requirements (tho OC'd to the hilt) is the same as a stock 1.4gHz Tbird... all on a Powerman 300watt PS (AMD approved) while having two HD/dual nics/WinTVpci/sound card/G550/CD burner/DVD.

The heat these cpu's produce (~85watts) is extremely stressful and somewhat difficult to manage unless you do improve case cooling and use a very good HS/fan setup as Jerrold mentions. I've obfuscated the noise and heat problems associated by building a custom water cooling system for mine... much quieter due to only needing 2/80mm standard case fans and much much more effecient in dissapating the heat (current ambient 23c/cpu 31.5c). The best HS compound on the market today is Arctic Silver 2... tho the standard white silicone zinc oxide type normally used works fine and substantially cheaper(what I'm currently using).

[Indiana]

Re: Jerrold Jones

Originally posted by Jerrold Jones

On the other hand, I *did* buy the 430 watt power supply from Enermax when I built it, knowing that the 1.4gHz Thunderbird needs a lot of juice.

I suspect some issues that might be attributed to the board are - in reality - power supply issues.

Lots of people think they can get away with a skimpy power supply.

This is surely right. If you follow those >1.4GHz problem threads, they can be quite often resolved by using certain brand powersupplies and RAMs.

Still other boards are not that picky about PS and used RAMs.

[Greebe]

Which is why I always suggest people who are building a new AMD system checkout AMD Athlon™ Processor System Configuration Information page.

[This_Idiot]

Quote "(did you know that there was more to the MURC than Desktop Video?)"

Oh Gawd, here we go again. Greebe and his soapbox. Well since this is more of a general hardware problem, why don't we just move the thread over the "general" forum and keep the atmosphere over here a bit more breathable.

And did you know, Greebe, that that there is more to Desktopvideoworld than this forum? And it costs me plenty to have it hosted. If ya don't like it, butt out. Or at least have the respect to keep your tongue secured behind your teeth.

To the rest of you, Greebe is not the moderator of this forum. He is, however, a MURC moderator, and as such has moderator staus and privelege over all MURC fora. As I do. Except I don't go interferin.

Chris
(T_I)