A64 performance @2.0Ghz

[juin]

in 2 or 3 year it will be replace.That a shorth life compare to P6

I dont like french test

 

[imgod2u]

Not really. The P7 core's been out for 3 years. Another 3 years and that's a 6-year life-span. The Pentium Pro was released in 1995 and the architecture was replaced by the P7 core in 2000 with the release of the Pentium 4.
I'm sure that there'll be P7-core based architecture even after the release of the first P8 cores just like there was a P6-based Celeron and P3 right now, but it won't be mainstream.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.

 

[juin]

PPro was not release in 1993

I dont like french test

 

[Mephistopheles]

I am sure I expressed many times by love for it. It's a great promising architecture.

Yes, well, P4 is not without its strenghts at all.

Anyway, from Intel, I think itanium is the most potential-filled architecture. The problem with itanium is that, unlike Opteron, there are only very few options available: the 1.3, 1.4 and 1.5Ghz parts. It's a very conservative scheme. If there was a 1.3Ghz Itanium for single-processor machines which a 20 or 25% lower cost, then things would become considerably more interesting. Maybe Deerfield, with its $700 or so price tag, will change things. They could have done this earlier, however, and then Itanium would have more market acceptance than it has now, at the introductory stages of Opteron.

Like Opteron: there's the 144, 244 and the much more expensive 844. It is a more balanced list of options, that's all.

 

[imgod2u]

I said 1995......

"We are Microsoft, resistance is futile." - Bill Gates, 2015.

 

[nach]

I fail to see how shoving 9 execution units with 9 issueing ports where around 70%+ of that remains idle most of the time counts as "efficiency".

Ah, so the core can be real idle, just like the brain. Rest of the time cpu spends doing bridging. Having a parallel bus to brain at 64 wire capacity is better than a 32 wire bus, this is where A64 would excel.

 

[nach]

I think present trend in CPU is for a fast core speed eg. 2.4Ghz+.

The other problem is a bottleneck called disk IO which is still at around ATA100 or even less, with your CDROM bus.

No high FSB number is going to overcome this one.

 

[juin]

in any case P6 was more succesful that P7.

I dont like french test

 

[juin]

nothing to do with architecture just a plain old 30 year ISA

I dont like french test

 

[imgod2u]

Ah, so the core can be real idle, just like the brain. Rest of the time cpu spends doing bridging. Having a parallel bus to brain at 64 wire capacity is better than a 32 wire bus, this is where A64 would excel.

Erm, the number of traces to the processor itself is dependent on the bus protocol the chipset uses. As for 64-bit bus feed to the processor vs 32-bit, all x86 MPU's since the Pentium 2 (or was it the Pentium) uses a 64-bit bus path to the chipset. The Athlon64, as far as I'm away, will not change this.

in any case P6 was more succesful that P7.

I wouldn't say so at all. The P6 core's release in the form of the Pentium Pro wasn't received much better than the Willamette was. The Pentium Pro was meant for high-end workstations and servers, yes, but it was still considered a flop.
I don't think it's possible to tell just how well the P7 core will do compared to the P6 core until we get to the end of the P7 core's lifespan. Then and only then can we look back in history and look at the relative success of both cores.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.

 

[jmecor]

If opteron will be only for 64-bit computing, it should
also be near equal to that of intel's itanium.

 

[Mephistopheles]

near equal to that of intel's itanium.

Though comparisons are lacking, I think Itanium is a considerably more powerful processor than Opteron. And it's also more expensive... Opteron is probably more capable in 32-bit than Itanium. When Win2003 comes out, we'll be able to judge that...

The only benchmark we have on that is SPEC CPU...

 

[eden]

I thought Windows 2003 was out? Or have I been sleeping in a cave for a while now?

--
<A HREF="http://www.lochel.com/THGC/html/news.html" target="_new"><font color=red><b>Join the Tom's Hardware Guide Community Photo album, send us your pics!</font color=red></b></A>

 

[dwellman]

Windows 2003 Server is definitely out. We have out copies already (MSDN version). I just haven't had the chance to play around with it yet.



__________________________________________________
<b><font color=red>Three great virtues of a programmer are: laziness, impatience, and hubris.</font color=red><b>

 

[dwellman]

Post deleted by dwellman

 

[Mephistopheles]

Ouch... I'm very sorry. I meant Service Pack 1 for Win2003 Server.

 

[dwellman]

Oh, Ok.

It would be nice if Microsoft would release an SP for 2003.

There are 80-something updates, patches, and miscelaneous downloads already for Server 2003. 17 critical

Probably when Microsoft releases XP SP2. Taking their damn, sweet time. Why is that QA is only an issue with Service Packs?

At least it seems that way.



__________________________________________________
<b><font color=red>Three great virtues of a programmer are: laziness, impatience, and hubris.</font color=red><b>

 

[Mephistopheles]

Yes, nothing like upgrades from Microsoft...
"Ooooh, we forgot that critical security issue that allows hackers to incinerate your hardware... oooops."

And something else, Xbitlabs has put <A HREF="http://www.xbitlabs.com/articles/cpu/display/64bit.html" target="_new">this interesting article</A> online. It talks about everything 64-bit related and is quite interesting.