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.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?


To: Cirruslvr who wrote (89965)1/27/2000 11:48:00 AM
From: milo_morai  Respond to of 1572542
 
Cirruslvr some info you make enjoy

LDT bandwith

amd.com
Unidirectional point-to-point links in each direction
Differential signaling with source synchronous clock forwarding

Variable widths negotiated at initialization
Upstream and downstream links can be of different size

16/16-bit link provides 6.4 GB/sec each way

Multiple logical channels in each link
Guaranteed isochronous bandwidth

In-band system management and legacy signal transport

PCI like configuration mechanism


LDT provides the bandwidth I/O and SIO require
LDT simplifies and improves MP solutions

And we all know EV6 is a pt. to pt. bus.

Milo



To: Cirruslvr who wrote (89965)1/27/2000 1:26:00 PM
From: Tenchusatsu  Respond to of 1572542
 
Cirrus, re: Foster bus,

That article on CNET is correct. Foster's bus will indeed have a bandwidth of 3.2 GB/sec. Interestingly enough, the Foster bus is very similar to the P6 bus, i.e. shared multiprocessor bus, not point-to-point interface.

Yes, you are right when you say that Athlon's P2P interface can also do 3.2 GB/sec if (or when) it's clocked up to 400 MHz. Then again, I'm not really sure that Athlon's P2P interface is necessarily superior to a shared multiprocessor bus, considering that Intel is still sticking to the latter even for Itanium and Willamette/Foster.

Also, remember that Coppermine with its 1.0 GB/sec bus is doing very well against Athlon, even though Athlon right now has a 1.6 GB/sec bus. It all depends on your architecture and where the performance bottlenecks lie.

Tenchusatsu



To: Cirruslvr who wrote (89965)1/30/2000 10:37:00 AM
From: steve harris  Read Replies (3) | Respond to of 1572542
 
cirruslvr,

You have posted some of the information before, therefore I have responded to your earlier post concerning Foster.

It looks like AMD has chosen to move up LDT on a SMP chipset to go with the Mustang Athlons.

www1.amd.com

It does appear AMD has decided to bypass a quick solution for now, but then again, it is only 1Q2000.

Bring Willy and Foster on................

steve

More reading...
ebnews.com

It is kinda neat that no one outside of AMD really
knows what AMD is working on.

Cool.