[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: PCI 2.2
- To: Mailing List Recipients <pci-sig-request@znyx.com>
- Subject: Re: PCI 2.2
- From: png@woof.net (Peter N. Glaskowsky)
- Date: Sun, 4 May 1997 22:47:57 -0700
- Resent-Date: Sun, 4 May 1997 22:47:57 -0700
- Resent-From: pci-sig-request@znyx.com
- Resent-Message-Id: <"Iu-Uh1.0.Ti3.1JNRp"@dart>
- Resent-Sender: pci-sig-request@znyx.com
> In terms of new
> form, fit, or function, however, you should look to the Hot-Swap
> PCI, Compact-PCI, PMC-Mezzanine, or Cardbus specifications.
This is certainly true; the originators of the PCI specification at Intel
are now working on AGP and other projects, so it's hard to see where new
design effort will come from on PCI itself.
It's a pity, too. There are some things we will need very soon:
* A 2.5V signalling environment for desktops
* A 1.8V signalling environment for mobile systems
* A 533-MBps version of PCI capable of connecting to three or four card slots
- preferably 32 bits wide, 66 MHz, DDR data transfers
* A 266-MBps version of the same interface that supports six to ten slots
PCI is conceptually excellent and had a good roadmap when it came out but
let's face it-- most PCs STILL aren't getting as much bandwidth out of PCI
as some machines had with VL-Bus three years ago. I know, it's an unfair
comparison, but still...
. png
Peter N. Glaskowsky, Senior Analyst, MicroDesign Resources
Publishers of the Microprocessor Report and Technology Roadmap
f 8 '