[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [f-cpu] Something new to play with :)


and btw, thank you for the log in the other message but i don't
catch the meaning :-) if a graphical output was available...

Kim Enkovaara wrote:
> On Mon, 24 Sep 2001, Michael Riepe wrote:
> > > > Utilization: 30%
> > > wow :-/
> > > How big is it btw ?
> > The multiplier also took something like 30%.  The SHL EU is very
> > space-consuming...
> I think I compiled the multiplier to a bigger chip (XC2V6000). It is quite
> much bigger chip. It was so big that the routing delays etc. become big
> factor if I try to compile too small things to it without constraining.

I have no "education" with P&R tool design, but from what i have seen,
what you describe is normal. Unlike humans which tend to optimize and
make "round" things, synthesisers for this kind of "devices" will use up
whatever ressource is available, and then start to "think" when it gets limited.
Constraining the design, if you want to use the rest of the chip, is important.
It will be interesting to compare individual units with the result when they
are together. We might see density increases, for example.

did you try the "new" ROP2 ?

To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/