<html>
<body>
<font size=3>At 04:09 PM 1/12/2005, Ronald G. Minnich wrote:<br><br>
<br>
<blockquote type=cite class=cite cite="">On Wed, 12 Jan 2005, Grant
Grundler wrote:<br><br>
> <br>
>
<a href="http://cmclab.rice.edu/projects/giganic/datasheets/PCI/SPECS/Pci22.pdf" eudora="autourl">
http://cmclab.rice.edu/projects/giganic/datasheets/PCI/SPECS/Pci22.pdf</a>
<br>
> and<br>
>
<a href="http://www.singlix.org/trdos/PCI22.pdf" eudora="autourl">
http://www.singlix.org/trdos/PCI22.pdf</a><br><br>
yikes. That's cool. They used to charge for it, I think all my copies are
bootleg. <br><br>
Wait, was the microphone on when I said that? I was taken out of
context!</blockquote><br>
The PCI-SIG provides specs free to its member companies and charges for
physical copies (the membership fees for the PCI-SIG are substantially
less than for the IBTA and there are other benefits of membership beyond
free specs). There are special discounts in place for academic
purposes. Given there are over 900 companies in the PCI-SIG and I
suspect a large percentage of the people on this reflector work for
member companies, you should have free access to the specs (Grant has
full access via an internal website I manage for distribution of such
material within HP). I would also note that the latest specs
are:<br><br>
PCI 3.0 for conventional PCI<br>
PCI-X 2.0a for PCI-X<br>
PCIe 1.0a for PCI Express (will soon be 1.1 which incorporates nearly all
of the ECN / errata generated over the past year as well as the corrected
error handling that will be important for people to understand since the
1.0a spec is broken in this regard - a final ECN for 1.0a should also
become available shortly).<br><br>
BTW, MSI-X was technology originally conceived within HP many years ago
(~15) and developed within the PCI-SIG as an extension to MSI a couple of
years ago (I first proposed it during the development of PCI Express and
we decided to make it applicable to all PCI technologies - ECN exist for
all three types).<br><br>
Mike </font></body>
</html>