Re: [oc] Interested in helping out

[Victor the Cleaner <jonathan@canuck.com>]

On Sun, Mar 26, 2000 at 06:22:37PM -0800, Andrew Martens wrote:
> 
> Well, I'm looking through Xilinx's software offerings, and there's quite a
> bit to choose from... Unfortunately, I'm not a bit fan of time-based
> licensing, especially if the recommended software turns out to be something
> other than Foundation Base ($99 USd) - since the next offering up the ladder
> is Base Express ($495 USd and out of my price range).  Their student edition
> seems as though it may be of questionable usefulness, as it only will
> support devices up to 20k system gates.
> 
> Now I realize why our digital systems design labs went with Altera FPGAs and
> development boards - they have a _free_ student version of their software.
> Well, I suppose that Xilinx's software does seem a tad more sophisicated...
> 
> Is there someone out there who is experienced with Xilinx's software who
> could suggest what I should buy to help out with OC?  Feel free to mail me
> in private at amartens@interchange.ubc.ca

I've got a fair amount of experience with it - I've been using Foundation 
2.1i for the last few months (I *think* it's tricked-out, but to be honest, 
I really haven't cared to learn about the various silly "base", "express",
and "geewhiz lights a-blinkin'" versions.  All I can think about is getting 
the hell off of this revolting windoze pissy and making the switch to Alliance
under HPUX.  But that's just me getting off on a tear...

Yes, I agree that Xilinx charging for development tools is stunned.  I 
feel that if they want me to buy+use their sililcon - especially in volume -
they should give me the development software for nothing.  The rep and 
disty up here (Calgary - same outfits service BC, Electrosource and Insight 
respectively) have been incredibly helpful, but they've been unable to make
the software a freebie because (they claim) of the per-copy terms of the 
license agreement they have with Synopsys, whose compiler is used in
Foundation.  I think it's bogus and that Xilinx should just eat that fee,
but one can only spend so much time arguing over this stuff.

As it happens, there's a big Xilinx seminar being put on here tomorrow, so
I'll see if I can browbeat some kits out of them.  Can you give me a pointer
to the current list of flavours so I know what to ask for?

> I'm quite tempted to sign up for a fully assembled OCRP-1 (if the cost will
> be in the low hundreds of $), but first it seems that I'm going to have to
> get some Xilinx software - which is starting to look prohibitively
> expensive.  Naturally, if I pursue a 68k-core design route (see below), I
> may not get one.  Well, unless someone thinks I could rig up a 68k core to
> work on the OCRP-1 board...
> 
> > Regarding your work: you can help with existing design, start something
> new,
> > ... It depends what you would like to do.
> 
> For a while I was toying with the idea of making a Motorola 68k-compatible
> core, although there's a few issues I have yet to look into:  whether (a)
> Mot would be really hostile to the idea (as ARM appears to be), (b) whether
> there would be any demand for such a core, and (c) the approximate number of
> gates required.  One of the key reasons for doing a 68k core is that I've
> got a lovely set of 68k manuals that Motorola gave me, and because they seem
> to be used all over the place.
> 
> If people are really keen on a 68k-compatible core, I'll investigate it
> further and possibly even come up with a development plan.  "Keen" meaning
> "Someone should speak up if they think it's a good idea, and let me know if
> it's a bad idea."  If not, I'd be happy to jump on board any other OpenCores
> project that needs additional manpower (assuming I manage to pick up the
> software I need).

Hell, *YES* there should be a 68K.  That and an MMU will get us very close the 
point at which NetBSD can be easily ported, giving us a mature, trustworthy OS
instead of the penguix that's been discussed thus far.

Jonathan