Opteron 144 slower than 3.2GHz P4

[Nowhere]

Originally posted by Ray Austin
yeah gize chill out a little...

the opteron is a server cpu... so these tests are bull!

the A64 is different from the opteron less pins plus other things. Although it too can perform 32bit operations.
...

Originally posted by Fluff
Also opteron has been designed to be multi processor! so i wouldn't be too fussed over the single processor benchmarks.
...

Well, practically only differences I can remember from the specs is that first A64 will have only one memory channel instead of two in the opteron and perhaps (not sure about that) will have less L2 cache. So at the same clockspeed they should be even slower...

But that just a guess...and I think Doc has now really harsh time about not beeing able to tell something that would clear things up a little.

[Dr Mordrid]

Grrrr......

Dr. Mordrid

[J1NG]

Can I say that a Dual Processor in ATX form is soon/already (?) available? No? Oh well you fella's probably knew that anyway...

First batch of AMD A64 with Dual Channel RAM? I think I've heard of that too... or was that single Opteron posing as AMD A64? Wish they'd clear that crap up.

Damn it I want that Tyan K8W now!

Sorry, slight rant there. Probably doesn't make much sense either.

J1NG

[Ssin]

Something's odd about AMD behavior. They have never been there for their customers (at least in Belgium) and would only sell to the mass because their products has the performance/price ratio crown (note I didn't say quality/price). Now they are trying to sell PROFESSIONAL solutions to nothing more than anthusiast people to back their sales. I would never buy AMD to put into a server first of all because their customer service is boggus and second because you never know when it's going to fry (notice I didn't say IF). Well then again I'm saying this because mainly in my country (and from the words I heard over the net it's kinda worldwide) AMD won't support you if your CPU is burned, the rubbers are damaged or pins are missing/torned... what else could happen to a CPU? They never understood the sentences where the words "customer" "satisfaction" and "support" where in the same time. That said... I've got nothing personal against AMD

[Nowhere]

Ermmm...do you think that Intel will support you if the plate (in some way it's an analogy of the rubbers...) on the cpu is damaged or the pins are missing? And thermal damage won't be a problem for AMD soon...
Other than this, I know what you mean. AMD doesn't have this...professional feel to it. I think it's even more visible here than at your place.

[Ray Austin]

Originally posted by Dr Mordrid
Can't....violate....damned.....NDA.....CRAPPOLA!!

Dr. Mordrid

easy doc wouldn't want you to pull anything!

[Ray Austin]

Originally posted by Nowhere
Well, practically only differences I can remember from the specs is that first A64 will have only one memory channel instead of two in the opteron and perhaps (not sure about that) will have less L2 cache. So at the same clockspeed they should be even slower...

Well my point was that it is designed to be in a server not a desktop. You have to use the right tool for the job.

The same would occur if you used an Itanium, it would be embarrassed all over the place.

The numbers they show really aren't far off plus we don't know what speed A64 will debut at... so just relax.

AMD runs benchmarks all the time against Intel, they know what they have to do. If they didn't care the A64 would be out already.

I've built 2 AMD machines and several Intel machines, both crash equally the same, the major difference I see is AMD is cheaper on the wallet. So which would you rather have???

IMHO, the biggest factors in whether a machine crashes or not is not the cpu, but the MOTHERBOARD, MEMORY, & POWER SUPPLY!

[Marshmallowman]

read this

http://www.amdzone.com/articleview.cfm?articleid=1329

A single 246 opteron in a nforce3 board @2GHz OC to 2.25GHz

[Nowhere]

Originally posted by Ray Austin
Well my point was that it is designed to be in a server not a desktop. You have to use the right tool for the job.

The same would occur if you used an Itanium, it would be embarrassed all over the place.

The numbers they show really aren't far off plus we don't know what speed A64 will debut at... so just relax.
...

I disagree with first two points. A64 & Opteron are almost the same thing (the same architecture for sure), whereas P4 family & Itanium are totally different, so the thing that Itanium performs poorly with desktop software "too" while desktop cpu Intel family performs good doesn't translate to the same situation for A64 & Opteron. Opteron basically is for A64 what Xeon, not Itanium, is for P4 - the same, but with more juice (at least that is what I remember was announced etc. - that Opteron will have dual, instead of single, channel and bigger L2; something has changed?). Is Xeon geared towards the server? Yes. Is it faster than P4 on the desktop at the same clockspeed? Well, I don't have the exact numbers, but more powerful cache just must do the job.

Of course there's the thing how well new AMD architecture scales with frequency - and what Marshmallowman posted is rather optymistic

[Technoid]

As Bsdgeek said: the opterons isn't slow they are actualy lightning fast as their IPC is able to bridge the gap in 1000MHZ between it and the P4 cpu's.
Amd cpu's arn't slow, they are just clocked slow

[Maggi]

just a few numbers to compare to ...

http://www.amdzone.com/articleview.cfm?articleid=1328

http://www.amdzone.com/articleview.cfm?ArticleID=1313

[UtwigMU]

Check out news on Ace's

http://www.aceshardware.com/forum?read=105029535

Apparently K8 should scale well, how high they'll take it depends.

Apparenlty both K8 and Prescott had/have problems so it's going to be a poor year WRT to CPU scaling.

I predicted 4 to 4.5GHz/PR at the end, we'll be happy if we see 3.8 (I expect 3.6)

[KRSESQ]

I would never buy AMD to put into a server first of all because their customer service is boggus and second because you never know when it's going to fry (notice I didn't say IF).

I've never fried an AMD processor, and I've used a LOT of them. Most frying problems come from incorrectly mounted or dislodged heatsinks. I always use a CPU shim to prevent such problems.

I wouldn't know anything about the quality of AMD customer support because I've NEVER needed it .

Kevin

[Jerry Jones]

Actually, the test results were very impressive:

Compared to *existing* 32-Bit Athlons:

"As we've seen, the Opteron 144 CPU is a highly efficient processor, clock for clock."

"It beats out the Athlon XP 3200+ in a large number of benchmarks."

"Much of this can be attributed to both the larger clock and the onboard memory controller and larger cache size."

"It's probable that AMD's latest baby has been tuned a bit more over the earlier Athlon XP architecture as well."

Compared to Pentium 4 32-Bit processors:

"It's also very impressive that the Opteron 144 can best the Pentium 4 in some of our tests, despite a 1.4GHz clock rate deficiency."

http://www.extremetech.com/article2/0,3973,1217396,00.asp

Jerry Jones

http://www.jonesgroup.net

[Ray Austin]

Originally posted by Marshmallowman
read this

http://www.amdzone.com/articleview.cfm?articleid=1329

A single 246 opteron in a nforce3 board @2GHz OC to 2.25GHz

Wow Marshy... good post! If you think about it 2.2Ghz is pretty slo these days, but if it can beat a 3.2ghz over priced intel sillycon... count me in!!!

I hope they go with 2.4 tho'... hit intel where it hurt$!

I won't even mention apple... oops!