Patrick McNamara wrote:

I am quite surprised at the tack the whole discussion took.  I was
expecting more discussion on the fact that we actually had a company
release a complete, commercial processor under the GPL.  Regardless, let
me remind everybody of one thing.  Somebody has already done everything
we are trying to do, before we have.  Right now, the Sun processor is
about the only large scale body of work from which we can use code.  It
doesn't matter what architecture approach we use or what we are doing. The first question in everybody's mind should be "Can I find someone
else's code that does this so that we don't have to do it from
scratch?"  The second question should be  "Can I legally use it?",
followed by "If it doesn't work, why and is it easier to do it from
scratch or change what I doing so it does work?".  So maybe we don't use
a full SIMD unit.  Has anyone stopped to think that we now have access
to a professionally developed hardware multiplier that we can study? Same thing for the adder. We don't have to take the whole SIMD unit. For that matter, we don't have to take all of any module in the
processor.  We have the ability to re-use any code we can.  This is one
of the biggest arguments for Open Source RTL code.  We don't have to
re-invent the wheel just because we want to build a car.  We can use
someone else's wheel and spend more time on making our care go better.

Exactly. When I saw the announcement I immediately looked to see if the SIMD processor was included this time.

Sorry if I got way off the track, but when people start attacking what I say rather than discussing things, I don't usually take it well. I especially don't like it when the attacker doesn't know what they are talking about. In this case, don't appear to know what vector and matrix arithmetic are (the difference between scalar and vector).

--
JRT
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to