Searching \ for '[EE]: What I/O standards --> consumer computers' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/ios.htm?key=i%2Fo
Search entire site for: 'What I/O standards --> consumer computers'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: What I/O standards --> consumer computers'
2005\06\22@175715 by Mike Hord

picon face
> PCIE may also be very appealing for hobbyists some day. There are
> already FPGAs out there with the resources to do PCIE, and creating a
> PCIE IP core is certainly doable for a group of hobbyists. The beauty of
> course is the small number of wires (for a 1X config 7 wires, 5 if you
> have your own clock), and the sheer bandwidth available. Very exciting
> times IMHO.

That's an encouraging thought...I've been worrying that by the time
I'm middle-aged, technology would have passed beyond what I could
tinker with at home without tens of thousands of dollars of equipment.

Although that has probably been the worry of the electronics hobbyist
for the last 300 years, ever since Ben Franklin first flew a kite in a
thunderstorm (yes, I know it probably didn't happen!).

Mike H.

2005\06\23@084959 by alan smith

picon face
But.....they all have to be matched lengths in order to comply with the spec.

Mike Hord <spam_OUTmike.hordTakeThisOuTspamgmail.com> wrote:> PCIE may also be very appealing for hobbyists some day. There are
> already FPGAs out there with the resources to do PCIE, and creating a
> PCIE IP core is certainly doable for a group of hobbyists. The beauty of
> course is the small number of wires (for a 1X config 7 wires, 5 if you
> have your own clock), and the sheer bandwidth available. Very exciting
> times IMHO.

That's an encouraging thought...I've been worrying that by the time
I'm middle-aged, technology would have passed beyond what I could
tinker with at home without tens of thousands of dollars of equipment.

Although that has probably been the worry of the electronics hobbyist
for the last 300 years, ever since Ben Franklin first flew a kite in a
thunderstorm (yes, I know it probably didn't happen!).

Mike H.

2005\06\23@122655 by Gus Salavatore Calabrese

face picon face
Where does one get the specs for PCIe ?
Or possibly some FPGA  IP that supports it ?





Gus Salvatore Calabrese 720.222.1309  GSC
http://www.omegadogs.com   Denver, CO

2005\06\23@144641 by Herbert Graf

flavicon
face
On Thu, 2005-06-23 at 05:49 -0700, alan smith wrote:
> But.....they all have to be matched lengths in order to comply with the spec.

When it comes to the PHY layer I'm certainly no expert, but with async
clocks I thought the only length matching necessary is with regards to
the pairs.

PCIE deals with differing clocks and clock skews with the "comma pad"
set sent every certain number of clock cycles.

Now, with that said, my knowledge (and reading of the spec) of PCIE
deals mostly with the protocol layers and link training (ltssm design).
While I have read some of the physical sections of the spec I'm
certainly nowhere near an expert, so it's entirely possible the spec
does require something more then what is strictly necessary.

TTYL


-----------------------------
Herbert's PIC Stuff:
http://repatch.dyndns.org:8383/pic_stuff/

2005\06\23@145228 by Herbert Graf

flavicon
face
On Thu, 2005-06-23 at 10:26 -0600, Gus Salavatore Calabrese wrote:
> Where does one get the specs for PCIe ?

I'm not sure if you have to pay for the actual spec or not. A good book
I can recommend is "Introduction to PCI Express" from the Intel Press.
It's about $100USD, but it really describes the elements of PCIE in a
way that's easier to understand then the spec.

Unfortunately if you want to design your own core you'll have to get the
spec.

> Or possibly some FPGA  IP that supports it ?

Check out Xilinx's site, they have a PCIE end-point core there. There
are other cores out there (RC cores for example).

At the moment I don't think there are any FREE cores out there, PCIE is
still too new for that.

TTYL

-----------------------------
Herbert's PIC Stuff:
http://repatch.dyndns.org:8383/pic_stuff/

2005\06\24@042459 by Alan B. Pearce

face picon face
>At the moment I don't think there are any FREE cores
>out there, PCIE is still too new for that.

The place to look is probably opencores.org. I haven't checked, but know
there is a heap of stuff there (including a PIC core).

2005\06\24@085004 by alan smith

picon face
I think...to get the real spec....you have to be a member of PCISIG ? Dunno..we are...so we get the specs, still have to pay tho...

Herbert Graf <.....mailinglist2KILLspamspam@spam@farcite.net> wrote:On Thu, 2005-06-23 at 10:26 -0600, Gus Salavatore Calabrese wrote:
> Where does one get the specs for PCIe ?

I'm not sure if you have to pay for the actual spec or not. A good book
I can recommend is "Introduction to PCI Express" from the Intel Press.
It's about $100USD, but it really describes the elements of PCIE in a
way that's easier to understand then the spec.

Unfortunately if you want to design your own core you'll have to get the
spec.

> Or possibly some FPGA IP that supports it ?

Check out Xilinx's site, they have a PCIE end-point core there. There
are other cores out there (RC cores for example).

At the moment I don't think there are any FREE cores out there, PCIE is
still too new for that.

TTYL

-----------------------------
Herbert's PIC Stuff:
http://repatch.dyndns.org:8383/pic_stuff/

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...