Aceshardware

(not so) temporary home for the aceshardware community
 FAQ •  Search •  Register •  Login 
It is currently Sun Sep 21, 2014 11:06 pm

All times are UTC + 1 hour



Welcome
Welcome to <strong>Aceshardware</strong>.

You are currently viewing our boards as a guest, which gives you limited access to view most discussions and access our other features. By joining our free community, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content, and access many other special features. Registration is fast, simple, and absolutely free, so please, <a href="/profile.php?mode=register">join our community today</a>!


Post new topic Reply to topic  [ 10 posts ] 
Author Message
 Post subject: SUn/Intel slides have Nehalem performance projections
PostPosted: Sat Feb 23, 2008 10:11 am 
Offline

Joined: Wed Aug 22, 2007 9:24 am
Posts: 36
removed broken link


Last edited by jokerman on Mon Feb 25, 2008 11:02 am, edited 1 time in total.

Top
 Profile  
 
 
 Post subject: Re: SUn/Intel slides have Nehalem performance projections
PostPosted: Sat Feb 23, 2008 2:13 pm 
Offline

Joined: Fri Aug 17, 2007 2:55 pm
Posts: 369
jokerman wrote:
http://at.sun.com/sunnews/events/2008/jan/sun_intel_breakfast/pdf/Sun_Intel_Road%20Show_Austria.pdf

also has AMD Shanghai projections ... Intel expects a pretty decent fight from Shanghai.


In FP it looks relatively competitive but if the Integer performance is even close to what is realized this would be very bad news for Shanghai, with Nehalem having a 40% lead (looks like about 2 for SH versus 2.85ish for Nehalem).


Top
 Profile  
 
 Post subject:
PostPosted: Sat Feb 23, 2008 7:34 pm 
Offline

Joined: Wed Sep 26, 2007 11:11 pm
Posts: 38
Working the numbers for Nehalem I get about 164/167 for SPECint/fp_rate_base2006 (assuming two socket configurations, which seems to fit with the relative numbers).

Current top of the line Harpertown is at 121/80.4, while the (now invalidated) 2.5 GHz Barcelona submissions were 102/86.3. It looks like Intel/Sun are expecting a 2.8 GHz Shanghai to come in at about 122/126.

Edit: switched int/fp for Barcelona. Also those are peak numbers, not base.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Feb 24, 2008 10:50 am 
Offline

Joined: Sun Sep 30, 2007 12:27 am
Posts: 10
I posted a link to here on my blog.
http://blogs.zdnet.com/Ou/?p=1025

I ran some numbers of my own and I had to make some adjustments to the numbers since I don't know what Intel's reference point is. The following are the adjusted numbers I came up with.

I estimated 176/163 for Nehalem on SPECint/SPECfp rate base 2006.
I estimate 123/121 for Shanghai 2.8 SPECint/SPECfp rate base 2006.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Feb 25, 2008 7:23 am 
Offline

Joined: Mon Feb 25, 2008 7:22 am
Posts: 3
The slides are down, anyone got a mirror or can put one up please? Thanks.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Feb 25, 2008 8:03 am 
Offline

Joined: Sun Sep 30, 2007 11:01 am
Posts: 22
This was the interesting one.

Image


Top
 Profile  
 
 Post subject:
PostPosted: Mon Feb 25, 2008 8:43 am 
Offline

Joined: Mon Feb 25, 2008 7:22 am
Posts: 3
Thanks wolf, but I was interested specifically in something I saw on the inq, they say the 2 core, 4 core and 8 core are coming "in late" and that's it, is there a date in the slides? Specifically I want to know about the 8 core desktop/enthusiast chips.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Feb 25, 2008 10:41 am 
Offline

Joined: Sun Sep 30, 2007 11:01 am
Posts: 22
Makes you wonder why companies bother "pulling" leaked documents. Once the horse has bolted as they say...


Top
 Profile  
 
 Post subject:
PostPosted: Mon Feb 25, 2008 1:04 pm 
Offline

Joined: Fri Sep 07, 2007 8:41 am
Posts: 23
JamesG wrote:
Thanks wolf, but I was interested specifically in something I saw on the inq, they say the 2 core, 4 core and 8 core are coming "in late" and that's it, is there a date in the slides? Specifically I want to know about the 8 core desktop/enthusiast chips.


Hiroshige Goto says:
Bloomfield/Lynnfield Nehalem-EP (Native Quad-Core) Q3 08
Dunnington (Native Hexa-Core for old FSB) Q3 08
Nehalem-EX Beckton (Native-Eight-Core-Nehalem) H2 09

http://pc.watch.impress.co.jp/docs/2007 ... gai394.htm

Dunnington only for MP-Boards (4 Sockets)
Nehalem-EP is for Single- and Dual-Socket-Configurations.
There is no Beckton planned for DP-Boards or Desktop, big Server-Iron only.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Feb 25, 2008 6:17 pm 
Offline

Joined: Mon Feb 25, 2008 7:22 am
Posts: 3
The_Wolf_Who_Cried_Boy wrote:


Hehe, yep. Thanks, but you were right, nothing much else of interest in there and no dates for nehalem, not sure what the inq was talking about. oh well.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 10 posts ] 

All times are UTC + 1 hour


Who is online

Users browsing this forum: No registered users and 0 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
suspicion-preferred