Re: Fastcode Benchmarking

Hi Anders

> > You should not, but you should start finding solutions. I expect to
> > see a little progress every day. That is why I am pushing.
>
> To find solutions, first the problems have to be defined. Are you saying
> that the B&V:s really haven't worked previously? No, you're not saying
that,
> so what are the problems?

I am not talking about B&V's.

> As I see it, the main problems are
>
> - Coordination - there must be a driving force keeping the lambs in the
> fold.
> - Creation - we must still be interested in creating functions
> - Verification - we must ensure that we are measuring the right thing
> - Execution - we must verify and benchmark on all targets
>
> You have been running all of that (among other things), and now we need to
> delegate. This should be priority one, especially the Coordinator.

I agree.

This is the problems I want to see some solutions for.

> >> Do you have any notes on how much time you have spent on Fastcode?
> >
> > Full time from 800 - 2200 every day including weekends.
>
> Ugh!

But also taken time out to eat and rest ;-)

> >> When work is given over to others, it's always best to note down 'use
> >> cases' - how it has been done historically, and then start the
> >> discussion around this document. Among other things it will be
> >> necessary to find natural 'compartments' in the process which can be
> >> done (more or less) independently by different people.
> >
> > See the list I made.
>
> Yes, the list is a start - it would probably be best to have that put in a
> Wiki so there's only *one* document, not 200 different copies.

Do it - somebody

> > I made a list for that a few days ago. Dennis Landi signed up for
> > webmaster.
>
> Were all the names there volunteers, or were some of them suggestions by
> you?

I put some on to provoke a little, but they did not really respond. It is up
to you now.

> >> It is probably also best to have some central repository
> >> (sourceforge or any other like that) for all documents and code.
> >> This way, everything is available to all (entrusted) parts.
> >
> > And nobody is responsible - it is your decision. Try it.
>
> How do you think Wikipedia works? Would you call that success or failure?

I know nothing about it. What is it?

Regards
Dennis


.

Relevant Pages

  • Re: Fastcode Benchmarking
    ... Running on different targets? ... but you should start finding solutions. ... No, you're not saying that, ... - Coordination - there must be a driving force keeping the lambs in the ...
    (borland.public.delphi.language.basm)
  • Re: OT smoked salmon
    ... We were somewhere around Barstow, on the edge of the desert, when the ... drugs began to take hold. ... I remember Work in progress ... saying something like: ...
    (uk.rec.motorcycles)
  • Re: Bathing in Pain
    ... you' to all these attractive girls that seem to value that stuff ... much that I could "fuck ten other women" (that's the standard advice ... The problem is I can't just make Ben progress in the way I've ... Not having a dig at you, just saying I personally disagree with the ...
    (uk.people.support.depression)
  • Re: Do you think?
    ... What progress are you looking for? ... accept supernatural explanations for natural events? ... Saying anything else is saying the thousands of ancient texts handed down ... And God is your master k00k. ...
    (talk.origins)
  • Re: OT Pissed discussions FOAK
    ... We were somewhere around Barstow, on the edge of the desert, when the ... drugs began to take hold. ... I remember Work in progress ... saying something like: ...
    (uk.rec.motorcycles)