Sign in with
Sign up | Sign in
Your question

Itanium 2 vs Opteron

Last response: in CPUs
Share
August 13, 2003 1:57:02 AM

I was wondering if there any benchmarks out there comparing Itanium 2 to opteron. Seems like with Itanium 2 at 1.5 ghz only opteron should be the top dog in 64 bits now. Really haven't heard about how amd is doing with sales vs intel in the 64 bit server arena though. Just wondeirng whats going on there.

More about : itanium opteron

August 13, 2003 2:29:51 PM

itanium 2 better
August 13, 2003 3:18:14 PM

I don't know a lot about this but some evidence to back up your claim would be nice. I'm not saying you are wrong but I'm not really convinced either.

"Don't question it!!!" - Err
Related resources
August 13, 2003 4:44:24 PM

lol I was asking for benechmarks comparing those 2 cpus, I wasn't giving any evidence about it. And it meants it just seems like opteron might be better because its got .5 ghz more than itanium. Anyhow like I said before I can't find any benechmarks comparing 64 bits cpus anywhere.
August 13, 2003 6:28:50 PM

The itanium will rip the opteron in just about any task...

I think the opterons main competitor is the xeon...not IA...

Either way..the opteron is good for server tasks

xeon is better for workstation tasks...


Proud owner of DOS 3.3 :smile:
August 13, 2003 6:49:23 PM

And Tommunist post wasnt directed at you, Maybe you should look in the reply to line.

The Rig
2.4c @ 3.0
OCZ pc3500
Asus P4C800-E
4 Maxtor 250gig serial ata drives
Enermax 465watt
Viewsonic 17" tft
ATI 9800Pro @ stock
August 13, 2003 7:35:13 PM

SPEC benchmarks:

CINT2000
Itanium 2 1.5 GHz - 1322
Xeon 3.06 GHz - 1242
Opteron 1.8 GHz - 1095

CFP2000
Itanium 2 1.5 GHz - 2119
Xeon 3.06 GHz - 1173
Opteron 1.8 GHz - 1122

<A HREF="http://www.spec.org/cpu2000/results/cint2000.html" target="_new">http://www.spec.org/cpu2000/results/cint2000.html&lt;/A>
<A HREF="http://www.spec.org/cpu2000/results/cfp2000.html" target="_new">http://www.spec.org/cpu2000/results/cfp2000.html&lt;/A>

Itanium is in a completely different class as Opteron. There is no competition. Opteron clearly is meant to compete with Xeon though not quite as fast.

Wanted: Large breasted live-in housekeeper. Must be a good cook, organized, and willing to pick up after me.
August 13, 2003 9:20:15 PM

Ironic when you say…
Quote:
Itanium is in a completely different class as Opteron. There is no competition. Opteron clearly is meant to compete with Xeon though not quite as fast.

You took old results for the Opteron and new results for the Xeon and Itanium. Time to update your Bias (BIOS)

See

CINT2000
Rioworks HDAMA Motherboard, AMD Opteron 246 1 1248 1317
HP Integrity Server rx2600 (1500 MHz, Itanium 2) 1 1322 1322

CFP2000
Rioworks HDAMA Motherboard, AMD Opteron 246 1 1209 1293
HP Integrity Server rx2600 (1500 MHz, Itanium 2) 1 2119 2119

Dichromatic for your viewing plesure...
August 14, 2003 12:08:51 AM

Indeed... Opteron is a good processor if stacked against Xeons...

However, Itaniums can still chew through a ridiculous amount of floating-point calculations at any given amount of time...

:evil:  <font color=red><b>M</b></font color=red>ephistopheles
August 14, 2003 12:24:05 AM

Sorry, I didn't see that one.

But my point is still valid though. Opteron does compete well with Xeon, but Itanium 2 destroyes it.

Wanted: Large breasted live-in housekeeper. Must be a good cook, organized, and willing to pick up after me.
August 14, 2003 12:51:45 AM

Itanium 2 will destroy it in 64-bit tasks. Itaniums architecture is made for 64-bit and more optimizations can be made which will make it a lot faster than the opteron
August 14, 2003 2:00:54 AM

The only thing that will determine if they compete against each other is who uses them for what.

It’s kind of an unfair comparison. The Opteron has a die size of under 120mm^2 while the Itanium 2 has a die size of 374mm^2. That’s a 9.8x area difference.

Madison I2 has a 6mb L3 cache.

Itanium compiler development is years ahead of Opterons.

The results for the Opteron were made on Microsoft Windows Server 2003 and most likely were not using 64 bit code.

When (if) the Opteron comes of age and gets good software support, it could possibly enter the league of the I2.

Dichromatic for your viewing plesure...
August 14, 2003 2:52:25 AM

Thats the stuff you play with. Itaniums an Xeon.
August 15, 2003 5:35:32 PM

Itanium compiler development is years ahead of Opterons.

typo error

Itanium compiler development is years behind of X86.


I dont like french test
August 15, 2003 6:05:09 PM

I think that’s an unfair assumption.

You can’t apply the age of the instruction set to the age of the chip, and for that matter, the Opteron’s 64 bit instructions are new, although similar to 32 bit instructions.

Intel’s been tweaking code (64 bit code) running on Itanium for 3 years now.

The same can’t be said for the Opteron. At most there has been 2 years of silicon.

Dichromatic for your viewing plesure...
August 15, 2003 7:03:59 PM

Considering the similarities between the K8 and K7, I'd say that plenty of the optimizations and tweaks that have gone into x86 compilers over at least the past 3 or so years (since the release of the K7 at least) will benefit the K8. Add on top of that how trivial it is to adjust a compiler to take advantage of the extra registers and I don't think IA-64 has any advantages in terms of compilers than the K8.

If anything, IA-64 is at a disadvantage because the K8 is a much more traditional OoOE core with high-frequencies and a 3-way superscalar design. Compilers have been designed and made with such architectures in mind for the past 10+ years.
Compiling methods and principles for an in-order VLIW architecture just recently came into focus because of Intel's push of IA-64.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 15, 2003 7:38:17 PM

I agree with the history of VLIW, and 300+ million dollars sent to <A HREF="http://www.intergraph.com/press02/intel10-10.asp" target="_new">Intergraph</A> for their <A HREF="http://www.intergraph.com/intel/indexpic.asp" target="_new"> 1992 chip design</A> prove it is not a recent idea.

However, the recorded benches speak for themselves. The fact is that the Opteron was running on a 32 bit OS most likely executing 32 bit code. Whether or not running in 64 bit code actually improves the Opteron, the Itanium is currently running native 64 bit code.

Dichromatic for your viewing plesure...
August 15, 2003 8:42:45 PM

I you can remember well that was a co production project made on P7 at the 1 place and legitation have appear after that intel sperate from co-project that far from be anything relate to Itanium wich have start in 89 lab at HP

I dont like french test
August 15, 2003 8:44:42 PM

PG1 5.01 a brand new compiler for Opteron according to most AMD eng and The inquirer this was suppost to boost AMD score to a unreachable level.So i guess i was right 60 point int 40 FPU.

I dont like french test
August 15, 2003 8:49:23 PM

3/4 at aceshardware was running slower with 64 bit mode and the only that was faster was using highly complex FPU code and very high precision.

I dont like french test
August 15, 2003 9:40:58 PM

Anyone actually know a website that benchmarks Itanium and opteron in 64 bit mode?
August 15, 2003 9:47:57 PM

I don’t know the aceshardware article you’re referring to. Either way it just reaffirms what I am saying.

A compiler that just started using the new extensions, released a month ago, is not considered mature.

All your figures seem to come out of a crystal ball, where are these 60int 40fpu numbers coming from. Please document your numbers!!!

Dichromatic for your viewing plesure...
!