[fpu] cooperation or not?

[Damjan Lampret <damjanlampret@yahoo.com>]

Guys,

I don't read my email for a day and you already start fighting. ;-)

Seriously, I don't really see why you two started to fight in the first
place. The only thing I see is that this could be blamed on my since I
promised to write 'C' generator for test vectors and I didn't write it
yet. So since you don't have test vectors you really can't say if
Rudi's code works properly or not. But it is not important whether
works properly or not. Important is that Rudi started working on
something. Jamil's way is a bit different: he likes to define things
before he starts working on something. I think both ways can be right.
I personally prefer to start coding and fix problems later (often it
takes me more time to fix problems and errors later than if I would
first define what I what to do and how). So I really can't figure out
why you two started to fight in the first place. Don't get me wrong. I
ain't dumb and I know how to read. But there was not important issue to
fight about. If something does not work properly then we can fix it. Of
course we won't have Pentium type of FPU right away. I am quite sure we
never will (maybe in a couple of years but not if we will fight all the
time).

Maybe we should resolve how to update the web page with new information
and latest decisions and how to put code in the CVS (basically who are
the maintainers etc.). In case you didn't know: there is a web
interface to update web pages through web
(http://www.opencores.org/editor)(you need account on opencores server
to do that and EVERYBODY interested can get his own account !!!!). And
also I believe that is important to put everyone's name in FPU section
if they contribute something (not necessary just HDL code but also if
they contribute ideas, specs, software code etc etc.).

Nobody pays us to do this anyway. So why should we fight?

I'll write 'C' generator right away. But not to verify who is right
(Jamil or Rudi whether Rudi's code works ok or not) but to verify if
the code works ok. Can you understand the difference. It is not
important who is right. Important is to have working code eventually
(not necessary today since we don't have a serious deadline). There is
a soft deadline that most of you don't know (I think only Rudi knows
it). I have a chance to get silicon. I would prefer not to talk about
this yet since the company who will sponsor this didn't give me 100%
gurantee. Once I get it we (opencores people) prepare joint press
release, update our web page and get ready to get our first silicon. We
need to be careful about this since one spin costs $50000 and more. At
this point I can tell you that it looks we will have quite cool
technology (0.25u or maybe more !!!). Expect more info soon (I will
certainly provide more info to people who actively work on the projects
(not just FPU) via direct email (there is 90% of people just reading
mailing lists w/o syaing anything). I don't say it is wrong to just
read but you never know who reads it anyway so I want to be sure that
this news doesn't come to reporters before we want that). This could
terminate sponsors BTW. BTW there is about 50 people reading FPU
mailing list (and about 300 cores mailing list).

So I hope we are all friends again? (this is not meant as a retorical
question BTW).

regards,
Damjan

PS I am always open for criticism (I think I didin't spell this word
right or did I?).



__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail – Free email you can access from anywhere!
http://mail.yahoo.com/