Let the Flame Wars begin!

bum_jcrules · Apr 22, 2003

So who has the better processor?

I think that AMD has put together a good product and it will only get better.

<A HREF="http://www.millionmanlan.com/MMLDefault.asp" target="_new">Million Man LAN 2 is June 25-29, 2003 in Louisville Kentucky... Be there!</A>

rain_king_uk · Apr 22, 2003

Pricing is too close to the Xeons for the Opteron - why switch all your server infrastructure for a CPU that is slightly better in a few areas when the price is almost the same? Again AMD are guilty of over-rating their products - a 1.8 at 1.6 prices would've made things interesting.

As for the Athlon64, unless they drop the single channel plans and go with dual channel then I'm afraid Prescott will eat it for breakfast. And no doubt the prices will be quite similar too.

Overall I am an AMD fan but I am more inclined to switch to an Intel setup next time I upgrade.

reever2 · Apr 22, 2003

What about people buying new servers or upgrading?

vk2amv · Apr 22, 2003

I would believe that is why everyone will buy them. I agree with rain_king when companies arnt ging to scrap their xeon setups just to go opteron for only a slight performance increase. Unless they have a specific need for 64Bit for something.
AREA_51

'It's only when you look at an ant through a magnifying glass on a sunny day that you realise how often they burst into flames'

sabbath1 · Apr 22, 2003

I don't understand this reasoning. Sure, servers that already run a very heavy Intel setup wouldn't benefit as much as newcomers would, but realize that the performance gain is actually quite substantial in some cases, and that the Opteron offers more of almost everything than the Xeons does (It offers, 64 gb of max system memory for instance.)
I'd say AMD is definitely ahead of Intel this time.
But that's just my opinion. You sure can have yours.

Prescott may be faster than the Athlon 64 by the time of release, who knows? Or maybe it'll be slower. Interesting battle for sure.

BlatantAMDFanboy · Apr 22, 2003

Any company that buy intel processors, particularly Xeons, are stupid. Xeons are so pathetic they had to name them after a 'Warrior Princess' to get anyone to buy them.

Intel is Stupid. 'nuff said.

jaythaman · Apr 22, 2003

OH JOY! another fake user.

Algebra was easy for the Romans because "X" was always 10 :lol: 
Jay Kay

BlatantAMDFanboy · Apr 22, 2003

I'm not fake. It's intel's 'Superiority' that's fake!!

Intel is Stupid. 'nuff said.

Lonemagi · Apr 22, 2003

Good product, needs optimized. I like the fact that it allows a company to stick with current software, and migrate forward, and I like the fact that it can keep on a level of a cpu that is 1000 mhz or more ahead in clock speed (then again, we have seen stuff like that for a while).

AMD needs to get moving with performance, needs to get the A64 at a pretty high caliber by September.

They are on the road, but can they get to the destination?

***
... And I'm proud to be an American, where at least I know I'm free, and I won't forget the men who died, who gave that right to me.

jaythaman · Apr 22, 2003

The thing about it performing equal/better than a proc iwth 1ghz advantage is good really good. But IMHO it'd be useful if they had a higher clock speed! I mean whats the use of being better performing clock to clock when you top model only performs equal or only a bit better than the rivals top model. I want to see a 3 gig AMD which will tear the 3 gig Intel to pieces, then they can truly claim to be the best.

Algebra was easy for the Romans because "X" was always 10 :lol: 
Jay Kay

Lonemagi · Apr 22, 2003

I doubt we will see a 3ghz cpu from amd soon... at least at this size of fabrication.

I would like to see some benches against the Itanium cpus in this 8-way config, might be interesting.

***
... And I'm proud to be an American, where at least I know I'm free, and I won't forget the men who died, who gave that right to me.

jaythaman · Apr 22, 2003

I havent really read the article coz well the site is being bombed. But i'm surprised that it doesnt feature the Itaniums. What the hell are they comparing it with then??? You have to have a 64 bit Proc.

Algebra was easy for the Romans because "X" was always 10 :lol: 
Jay Kay

bum_jcrules · Apr 22, 2003

I think that the AMD will do well. So far it does not look good as a workstation but it is still way to early in the game. P4 looked like a piece of junk back when it first hit the street. It got the first couple of rounds under its belt and came out swinging with higher clock speeds, a better FSB, and tweaked out toys.

The benches we see today are only the start of it all for the Hammer. Take STREAM 5.1 for example. STREAM was never written to handle extra bandwidth through the X-Bar and Hypertransport. Notice how it makes even the dual Xeon systems look bad compared to a single P4 3.0 GHz. Hammer systems can use info from all of the other DIMMs in the entire system. If it looks only to the DIMMs attached to the immediate MCH then of course you are going to see poor results on the memory benches. The bandwidth is much higher than what was shown. HT would allow for an additional 19.2 GB/s on top of the memory in the DIMMs under the immediate MCH. Take a request from CPU1 that draws from the memory as a whole. It can pull info from the DIMMs attached to CPU2. Limit it as much as you want. Even if there were two CPUs with two channels a piece.

DDR333 = 500/3 or 166.667 MHz ~ 2666.667 MB/s or 2.67 GB/s

Take that and multiply it by 4 (two sets of dual channels) = 10666.667 MB/s or 10.67 GB/s

The benches returned only 1.7 GB/s.

Part of that could be because of the motherboard. There is no way to tell how the memory buses were allocated on the motherboard. Maybe it only used two channels for both CPU's? That would be a possible solution, in theory, since you could have CPU1's MCH run the DIMMs and CPU2's be disabled. "That is a SPECULATION."

The bandwidth is much better than 1.7GB/s and when a single P4 got a score of over 700 MB/s better there has to be a problem with the benchmark. Just to show haw bad some of these benches are look at the <A HREF="http://www.tomshardware.com/cpu/20030422/opteron-27.html" target="_new">Sisoft Sandra 2003 Memory bench</A>. The single Xeon did better than a dual Xeon.

I think one reason there is some improvement in the scores of the Dual Opteron verse the single XP 3000+ is the simplicity of incorporating the MCH into the die. No FSB bottleneck between the MCH and the CPU. Both are operating at the same clock frequency.

There are a lot of things that need to change in the world of benchmarks. There has never been a system like this before and now that it is here there needs to be a way to rate the two against each other.

There are my two cents.

Why is it a penny for your thoughts but you have to put your two cents in? Somebody is making a penny. - Steven Wright

<A HREF="http://www.millionmanlan.com/MMLDefault.asp" target="_new">Million Man LAN 2 is June 25-29, 2003 in Louisville Kentucky... Be there!</A>

Lonemagi · Apr 22, 2003

CPU benched:

Opteron 1.8 GHz (1800 MHz / 128/1024 kB)
Athlon XP 3000+ (2166 MHz 128/512 kB)
Athlon XP 2800+ (2083 MHz 128/512 kB)
Athlon XP 2500+ (1833 MHz 128/512 kB)
Athlon XP 2700+ (2166 MHz 128/256 kB)
Athlon XP 2600+ (2133 MHz 128/256 kB)
Athlon XP 2400+ (2000 MHz 128/256 kB)
Xeon 3.06 GHz (3066 MHz 12-8/512 kB)
Xeon 2.8 GHz (2800 MHz 12-8/512 kB)
Pentium 4 3.00 GHz (3000 MHz 12-8/512 kB)
Pentium 4 3.06 GHz (3066 MHz 12-8/512 kB)
Pentium 4 2.80 GHz (2800 MHz 12-8/512 kB)
Pentium 4 2.66 GHz (2800 MHz 12-8/512 kB)
Pentium 4 2.53 GHz (2533 MHz 12-8/512 kB)

No itanium.... Dissapointing. Im gonna look around for other reviews later tonight.

***
... And I'm proud to be an American, where at least I know I'm free, and I won't forget the men who died, who gave that right to me.

soulprovider · Apr 23, 2003

I think the tests were to show not just the 64 bit performance but also the backwards compatible 32bit performance of the chip. When the Itanium tries to do this it sucks to the extent where it's nearest rival is a Pentium 1.

There was therefore no need to include this chip in the test as placing it in such a poor light would have given wrong impressions. Also theres not much point including a chip in a group test where you're benchmarking the good bits and leaving out the bad as this would be deemed favourable or inconclusive.

Therefore I'd only expect to see the Itanium next to the Opteron in strictly 64bit mode. It's what the two processors are designed for.

The case with the Opteron is that it will run your 'old' legacy 32bit applications just as good as current 32bit CPU's but not better, it just does that to allow you a stepping stone to 64bit computing where it's real power lies.

Vorsprung durch Dontwerk.....as they say at VIA

reever2 · Apr 23, 2003

The thing about it performing equal/better than a proc iwth 1ghz advantage is good really good. But IMHO it'd be useful if they had a higher clock speed! I mean whats the use of being better performing clock to clock when you top model only performs equal or only a bit better than the rivals top model. I want to see a 3 gig AMD which will tear the 3 gig Intel to pieces, then they can truly claim to be the best.

All other server processors have well over 1 ghz disadvantage from xeon, most of which are chosen by companies over xeon, which is considered just as much as a joke as the Athlon MP is. opteron is just about the only one aside from Athlon MP that even gets close to the clockspeed gap. Opteron would have been better compared to Itanium, Ultraparc, power4+ and the like

Twitch · Apr 23, 2003

I think the only thing that could hold the Opteron down is if it proves to have instability issues. Considering the pains AMD went to in order to avoid any thermal issues, I would think the only potential problems might lie in core logic instabilities. It seems as if the integrated memory controller will help avoid some of the ways core logic might screw it up.

Still, knowing that VIA will be a major player in Opteron's core logic makes me a bit nervous. VIA doesn't have the greatest reputation for stability.

I think if Opteron proves to be rock-solid stable, there is no reason it won't be a very succsessful product. After all, it is unique in the computer world. If you want high-end 32-bit performance and very good 64-bit performance, there will only be one real choice.

Those who live in glass houses shouldn't take showers. :tongue:

eden · Apr 23, 2003

Note that finally AMD has integrated on-die thermal shut-off, à la Intel. I think instability is no longer an issue at AMD for Opteron.

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden.

Twitch · Apr 23, 2003

Yeah...that was a very good move by AMD. That's why I say the only real worry I have is in the chipsets designed by VIA, nVidia, SiS, etc. As long as those are stable, I think Opteron will have its niche. And I think the integrated memory controller on the Opteron takes away about 75% of the opportunity for VIA and others to mess up the core logic.

Those who live in glass houses shouldn't take showers. :tongue:

Edited by Twitch on 04/23/03 04:36 PM.

eden · Apr 23, 2003

Yeah, now it's more relying on chipset features. Which is pointless IMO, since all servers need is a reliable CPU connection to others, they won't bother checking and using individual racks to tinker with USB 2.0, Firewire or even integrated 5.1!

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden.

flamethrower205 · Apr 24, 2003

Ok, here's my flame:

you ready?

sure u r?

ok, here goes:

YO MOMMA!!!

Hilbert space is a big place.

Kemche · Apr 24, 2003

But Wait, what if the integrated memory controller has a Bug. For Athlon MPs and Xeons you can just go with another verdor for the chipset. But Opteron you can't even do this because you will have same issue on all the chipsets.

So instead of just replacing a MB you have to replace the whole processor.

KG

"Artificial intelligence is no match for natural stupidity." - Sarah Chambers

Twitch · Apr 24, 2003

Ummm...well you could say the same thing about any type of CPU bug. I haven't heard anything about AMD's IMC having any bugs though.

The Netherlands is where you go when you're too good for heaven. :tongue:

eden · Apr 24, 2003

Plus, the fact we got many websites, testing the Opterons in real-world applications, even their own servers, stress-testing it, to me proves that for a new product, it is not only high-performing, but stable as well.

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden.

Spitfire_x86 · Apr 24, 2003

P4 looked like a piece of junk back when it first hit the street.

P4 is still a piece of junk. Let's see why people told initial P4's junk.

1) It was beaten by P3 and T-bird at equal clock.
2) Clock speed difference from AMD was not big
3) No support for SSE2

Now more softwares have SSE2 support, and AMD introduced PR system, so it looks good. But the processor itself is sill a piece of junk. 533 MHz FSB looks good when you compare it vs. a 266 MHz FSB AXP. 533 vs. 333 (AMD) is equal to 400 vs. 266 (AMD). Clock vs. clock, Barton vs. Northwood "B" is equal to T-bird vs. Willamette. And latest 800 MHz FSB is another big joke. With 50% more memory bandwidth, it's unable to show significant performance boost in most benchmarks, in many cases 3.06 GHz (533 MHz FSB) can beat the new 800 MHz FSB crap. And all P4's are inefficient as hell. AXP 333 MHz FSB catches 400 MHz FSB P4s in terms of memory bandwidth, 400 MHz will catch 533 MHz FSB. And new 800 MHz FSB crap is FAR AWAY from 6.4 GB/s memory bandwidth.

----------------
<A HREF="http://geocities.com/spitfire_x86" target="_new"> My Website</A>

<A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new"> My Rig</A>

Let the Flame Wars begin!

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Champion

Distinguished

Champion

Illustrious

Distinguished

Distinguished

Champion

Splendid

Share this page