| View previous topic :: View next topic |
| Author |
Message |
iiB
Joined: 20 Aug 2007
Posts: 66
|
 Posted: Thu Sep 20, 2007 12:30 am Post subject: |
 |
|
| Paul DeMone wrote: | | Travelmug wrote: | | Paul, you are unbelievable. |
No, IPF bashers are. They have already killed off it off a thousand
times yet sales just keep growing. Talk about stupid set in stone. |
I don't think many IPF bashers really wthink it will 'die'. Obviously being set to replace Alpha and PA-RISC it will vitel enougth to sustain itself for a decade or so.
The fact is that IPF is shaping up to be not much more then another properity architecture in the RISC space.
I think Intel better try to overtake some high-end risk market through x86 rather then IPF.
It might change if AMD dies or completly fails to compete, but even if it happens, IPF prospects as more then just a RISC replacment are slim. |
|
| Back to top |
|
 |
dkanter
Joined: 20 Sep 2007
Posts: 59
|
| Posted: Thu Sep 20, 2007 10:48 am Post subject: |
|
|
| dougSF30 wrote: | | Slide 25 of the press briefing has a decent die photo of Nehalem. |
Does anyone have a copy of the presentation where the die photo is included?
I think it has been scrubbed out of the one that is currently online.
DK |
|
| Back to top |
|
|
Markace
Joined: 31 Aug 2007
Posts: 11
|
| Posted: Thu Sep 20, 2007 4:00 pm Post subject: Die Photo |
|
|
Check here:
http://download.intel.com/pressroom/kits/45nm/Penryn%20Die%20Photo_300.jpg
| dkanter wrote: | | dougSF30 wrote: | | Slide 25 of the press briefing has a decent die photo of Nehalem. |
Does anyone have a copy of the presentation where the die photo is included?
I think it has been scrubbed out of the one that is currently online.
DK |
|
|
| Back to top |
|
|
TacoBell
Joined: 17 Aug 2007
Posts: 285
|
| Posted: Thu Sep 20, 2007 4:05 pm Post subject: Re: Die Photo |
|
|
This is penryn, not Nehalem.
|
|
| Back to top |
|
|
mocad_tom
Joined: 07 Sep 2007
Posts: 12
|
| Posted: Thu Sep 20, 2007 8:03 pm Post subject: |
|
|
@dkanter
you have a pm in your inbox.
If Nehalems L2Cache is really shared and all requests are handled through one central port - isn't this very slow.
greets, tom
|
|
| Back to top |
|
|
Hans de Vries
Joined: 07 Aug 2007
Posts: 89
|
| Posted: Thu Sep 20, 2007 8:11 pm Post subject: |
|
|
| mocad_tom wrote: | @dkanter
you have a pm in your inbox.
greets, tom |
Nice! Can I get one as well?
Thanks, Hans. |
|
| Back to top |
|
|
Paul DeMone
Joined: 29 Aug 2007
Posts: 528
Location: Great white north
|
| Posted: Thu Sep 20, 2007 9:05 pm Post subject: |
|
|
| mocad_tom wrote: | @dkanter
you have a pm in your inbox.
If Nehalems L2Cache is really shared and all requests are handled through one central port - isn't this very slow.
greets, tom |
Who says there is one central port? The L2 probably supports concurrent
accesses by being pseudo multi ported. That is, each core can access the
L2 at the same time as long as it is to separate banks. |
|
| Back to top |
|
|
Markace
Joined: 31 Aug 2007
Posts: 11
|
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
