SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?


To: Tenchusatsu who wrote (102994)4/28/2000 8:43:00 AM
From: Rob Young  Read Replies (1) | Respond to of 186894
 
"Rob, if 4-way Foster is going to beat 4-way Itanium, aren't you worried about that same 4-way Foster outperforming
just about every other server out there, including Compaq's much-delayed Wildfire?"

---

I'm sure you see what you are comparing... you are
comparing CPU to Server. No, I'm not worried at all.
Being in the server division , you realize there is much
more to a server than a CPU. AlphaServers have been
somewhat hamstrung of late compared to the rest of the
industry. Lagging in MHz and lacking on-chip L2. In the
next year or so the story gets much better as L2 comes
on-chip with one version of EV68 , the clock gets much
faster and we already see DDRSRAM in today's latest
rev of ES40 (4 cpu server). The follow-on to the 4 CPU box
is a nice one from what I hear.

If you look at the great box that Compaq will be
selling in the 32 processor Intel space, the Unisys
ES7000, it isn't in the same league as Wildfire. Don't
get me wrong, more than enough for most folks but Compaq
anticipates selling a Billion $$$ worth of Wildfire
this year so it is a very good story. How good?

Look how well it blows away an S80:

ideasinternational.com

60% faster than an S80.

Here is where I agree with you fully... the 4 processor
Foster and equivalent 4 processor Athlon follow-on will
give Itanium a dickens of a time in that space, something
I've been saying all along. With 1.2 GHz EV68 later this
year and copper-SOI to follow shortly after (whenever) the
.18 21264 (EV68) should be a more competitive part than
it is currently. Even today, the .25 667 MHz 21264 is
leading in SPEC2000. High-bandwidth helps... something
Foster and Willamette will be doing better than Itanium
so I don't expect Itanium @700 MHZ to do nearly as well at SPEC2000 as EV68 at 1.2 GHz and after July 1st SPEC95 results will no longer be accepted.

Itanium not good at SPEC2000
Itanium as good as or less than Foster at tpmC

So we wait for McKinley.

Rob