XP 2100+ article - comments?

[FatBurger]

Your thoughts on the article? I'll say something tomorrow, I'm off to bed in a minute.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
Don't step in the sarcasm!

 

[CoOLMaNX]

heh, i just skimmed through it and all i can say is good job AMD, sweet what they've done, but i don't think its gonna be a big seller, even with its nice performance

didnt have one of em electronic pens so ill just type my name,<i>CoOoLMaNX</i>

 

[Schmide]

Just remember while you rest.

All Pentium 4 CPUs with an FSB clock speed of 133 MHz and 533 MHz Rambus memory that are not yet available on the market were marked with a blue-and-black bar and are there only for reference purposes.

Good article except reading that over and over.


All errors are undocumented features waiting to be discovered.

 

[tnadrev]

mmm, they could have at least put it only once on each page...

Engineering is the science of making life simple, by making it more complicated.

 

[dhlucke]

All Pentium 4 CPUs at an FSB clock speed of 133 MHz and 533 MHz Rambus memory that are not yet available on the market were marked with a blue-and-black bar and are there only for reference purposes.

Do they think we're idiots? This was the most annoying review ever.

<font color=red>If you were to have sex with your clone would that be considered incest or masturbation?</font color=red>

 

[tnadrev]

what was with that comment about the thoroughbred not adding anything new, "just a die shrink"
... AMD hasn't finalized anything... well not to the public...

Engineering is the science of making life simple, by making it more complicated.

 

[texas_techie]

Interesting, I didnt think the xp2000+ could compete that well with Intels latest. But I guess it can. I will have to look at other sites before drawing a final conclusion.
The 533mhz bus looks real good on the pentium.. beat AMD is almost all the benchmarks. When is that supposed to hit the market?
Finally, im curious to see what the jump to 1.8 gigs does for AMD on their xp2100+..

Benchmarks are like sex, everybody loves doing it, everybody thinks they are good at it.

 

[Atolsammeek]

I agree its dumb that there three People doing a post about it.

 

[Guest]

Im really annoyed by these copy/paste reviews. "Archiving is a very practical application. WinACE 2.11 was used under Windows XP blablabla". Something tells me we will read that very same line when Tom benches the Hammer, or Precott or whatever next year. They just recycle their last cpu review, change a few comments and publish it. Good thing we still have anand.

As for the conclusions of the review; I cant say im thrilled by a 67 Mhz increase. The difference between Nortwood and Athlon XP are neglectable if you ask me. Office performance really is a NON issue, I mean, how much faster does your spell checker need to run ? Internet performance LMAO.. Quake3.. do we really need 300 Fps ? Who cares about a few percent more or less... maybe im getting too old for this forum ;-)

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[dhlucke]

They need to get rid of some of the old benchmarking games and apps. Use Serious Sam and some of the newer engines instead of Quake3 at the very least.

Maybe I don't understand this, but if a monitor supports 85 Hz at a current resolution then would you even be able to notice anything above 85 FPS?

<font color=red>If you were to have sex with your clone would that be considered incest or masturbation?</font color=red>

 

[Matisaro]

The nv15 demo of quake 3 is much better, the regular quake 3 demo is all about streaming ram bandwith, and not about cpu, when there is alot of cpu intensive action on the screen like in the nv15 demo, the gap is far lessened and a truer picture arises.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

 

[Guest]

So..If AMD can match and beat Intel with DDR ram, what would happen if AMD were to support RAMBUS ram?

Would AMD then really kick Intel's but?

 

[pr497]

So..If AMD can match and beat Intel with DDR ram, what would happen if AMD were to support RAMBUS ram?

you would get a huge heatsink and fan for you chipset...
AMD chipsets already <b>NEEDS</b> a fan on them at 266mhz...imagine the size of the fan for 400+mhz...
btw...you'll get better performance too

<b><A HREF="http://gamershq.madonion.com/compare2k1.shtml?2932776" target="_new">P4 + DDR333</A>=<font color=blue>OK</font color=blue></b>

 

[FatBurger]

AMD wouldn't benefit from RDRAM currently, the extra bandwidth would go to waste.

Well, it's official. Frank and/or Bert is a complete moron. I'll give my reasons if anyone wants, but I think most people read the latest review and know why.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
Don't step in the sarcasm!

 

[tlaughrey]

So..If AMD can match and beat Intel with DDR ram, what would happen if AMD were to support RAMBUS ram?

Would AMD then really kick Intel's but?

Not necessarily. From what I understand, the Athlon can't take full advantage of the bandwidth offered by RDRAM, so you wouldn't see much, if any, improvement. The P4 was designed with RDRAM in mind and it needs all the memory bandwidth it can get in order to perform at its full potential.

<i>I made you look. But I can't make you see.</i>

 

[Makaveli]

I was impressed with the performance of the CPU, and would still buy an XP over a P4. Regardless of the improved performance in Northwood Pentium 4 are still to expensive.
The biggest thing holding back the athlon I believe is it's FSB. I would have liked to have seen 166FSB with a 512L2.

 

[zengeos]

Perhaps it was Anand in their recent 2100 review that suggest a FSB increase to 166mhz won't have a major impact on Athlon's performance compared to P4's performance. They suggest that AMD would be better off increasing the on chip cache to 512 k rather than upping the FSB if it came down to a choice between the two.

Mark-

When all else fails, throw your computer out the window!!!

 

[xxsk8er101xx]

i find it funny that he did that ... we all know there are a few who can't read and takes things completely out of context. i simply skipped it .. why did you read it over and over anyway? lol just skip it

<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>

 

[xxsk8er101xx]

why not just skip it

<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>

 

[slvr_phoenix]

Of course with a die shrink from .18 to .13, they <i>should</i> be able to do both <i>and</i> still have a smaller die with better electrical and thermal performance. To do anything less is, in my opinion, no better than what Intel did to the P3. (I.E. Turning it into a Celeron so that it can't out-perform their P4s.)

<pre>If you let others think for you, you're the
only one to blame when things go wrong.</pre><p>

 

[zengeos]

With double the engineers AMD probably would. But, like most any business, AMD tries to budget limited resources more in line with actual revenue. So, AMD seems to do things in more gradual steps.. 6-9 month smaller upgrades instead of 12-18 month larger upgrades. It seems to work well for them and has allowed them to more rapidly qualify enhancements from what I can tell.

Intel for it's part has more chip families in concurrent production:

P3, Celeron, P4 and Itanium. All of these have split off resources. BUT Intel does have substantially more engineering resources. I suggest you give AMD the millions it would cost them to hie more engineers to upgrade Athlon.. AGAIN and still maintain the Hammer development which is a very aggressive schedule in and of itself.

Mark-

When all else fails, throw your computer out the window!!!

 

[zengeos]

slvr

Put yourself in the AMD engineers shoes.

They have just so many engineers. So, they have to balance the continued development of Athlon which is now reaching and perhaps exceeding the original design specifications; exceeding their original end of life performance expectations, I bet, OR transition engineers to the new Hammer processor, which by those same expectations will perform significantly better AND scale better as well.

AMD and Intel BOTH must continually analyze where things are today and where their opposing CPUs will be months and years from now and plan accordingly.

Look at Intel's roadmap.. Prescott coming in around a year; increased FSB coming in a few months. Athlon probably will NOT compete well against these and no amount of core tweaking would improve performance enough to make it worth the engineering effrort. The CPU is nearing EOL by any standards. It will be 4 years old this coming winter and that's getting long in the tooth in CPU terms.

Hammer is their logical next generation CPU. It is a year ahead of the next closest competition from Intel and promises, as I mentioned, better speed scaling than Athlon can provide. At the same time the die size is only marginally larger than Athlon but it offers SSE2 support along with the SSE support Athlon added so well.

You can give this or that rebuttal, but the bottom line is AMD has to make engineering decisions based on their own inside information. I suspect AMD is far more in the know about where Hammer is in development and how well it will perform already. Furthermore, IF Hammer was going to underperform you can be sure AMD would be making major improvements to Athlon as we speak to fill the void IF those improvements woul significantly improve performance.

Mark-

When all else fails, throw your computer out the window!!!

 

[slvr_phoenix]

How many engineers does it take to change a few bridges? The move to a 166MHz FSB shouldn't require <i>any</i> new engineers to implement.

They're already doing a die shrink, which requires planning, implementation time, and testing. Then they're doing a switch over to SOI which requires at the very least testing. They're already going to be changing the bridges anyway just for the new clock speeds. So at the very most if AMD moved to a 166MHz FSB, it would still require no more man-hours than what they are already doing.

True, the cache would be nicer to see, but even still, anything is better than nothing, which is what AMD has planned.

Speaking of cache, it would take a whole 1 additional engineer at most to implement.

Millions? I'm not talking about any major change. Just up the cache a little and set it to a FSB default of 166MHz, which most overclockers can already do anyway, and which VIA already officially supports a chipset for.

<pre>If you let others think for you, you're the
only one to blame when things go wrong.</pre><p>

 

[zengeos]

LOL

I'm glad you are so intimate with CPU engineering Slvr. I look at things like upping the FSB or adding more cache and see...more engineering time and work required to implement.

But you seem to know more about it than me so I defer to your magical no engineering resources required statement

Mark-

When all else fails, throw your computer out the window!!!

 

[sjonnie]

I thought it was awsome, I wonder when there will be a CPU which uses >100W of power, then I will be able to heat my house with it. Kinda incredible considering there are P3s that use 8.2W of power. Don't even consider putting a UPS on your AMD system unless its BIG.