[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
3.3V - when?
Alan Simmonds writes:
-> Anyone know when is this migration to 3.3V is supposed to occur? As
-> far as I can tell, only the ATX PCI PCs in our company have even the
-> PCI 3.3V connected let alone 3.3V signalling.
->
-> I have a board design that requires both 5V and 3.3V - I guess I'm
-> stuck with using an on board regulator to get my 3.3V. It also looks
-> like most manufacturers are still producing 5V signaling cards - I
-> haven't even seen a universal card. Is 3.3V ever going to hapen?
-> Page 121 of the bus spec appears farcical unless the timescale is 10
-> years per inch.
You should be providing 3.3V from the 5V rail on your card, as nothing
requires 5V PCI bus connectors to provide 3.3V. They are required to
provide the leads, in case the user wants to hook up his/her/its own 3.3V
source, but not the source itself.
HP, Sun and perhaps a few other workstation vendors have 3.3V signalling
PCI backplanes on the market. To go into those boxes you should design a
universal (i.e. capable of living in either 3.3V or 5V signalling
environment) card. There are universal cards on the market, but they
require searching. Whether 3.3V signalling will follow into the PC market
is an open question.
The incentive to move to 3.3V signalling was to use bus clocks faster than
33 MHz. This was driven by graphics. This was shot down by Intel's AGP
proposal, which now appears to be the target for PC graphics cards
designers.
Note that the issue of 3.3V v 5V signalling is entirely separate from the
issue of using components which require 3.3V power on your card. You may
do either or both.
-- C^2
___ __ /___ ___
| / /_ _ \ |
| / / // / \ |
| \ / //_/ / |
|___\_ /____/___|
/
H E W L E T T
P A C K A R D
Charles Curley is not an employee of Hewlett Packard.
This message does not necessarily represent the opinion(s) of HP.
˜ H 8