[Open-graphics] Vision statement still not satisfying

[Nicolas Capens]

Hi Bart,

> It is not. Neither provide full documentation, which is part of the 
> statement.

Yes, it's part of the vision statement as currently written. That doesn't 
mean it should be. Documentation is something for the developers. Our target 
market isn't only developers, is it?

> Extra value is it being fully documented and usable, unlike anything else 
> on the market. This may not be important to you, but it is important to 
> quite a few others it seems.

NVIDIA and ATI cards are not usable? Well they do run immensely popular 
games and professional applications you know. Try shaking a stick at that 
with the Open Graphics card. It just doesn't have the.. oh, right... 
features.

Ah, you probably meant running it on Linux? Then please don't forget to 
mention it. It's important.

> No, running 3d applications is NOT the goal. Having graphics hardware that 
> is fully documented and that can be very well supported on open source 
> platforms is the goal. Good 3D performance is an additional desire, not a 
> goal.

I wanted to express this in a nice way, especially considering your effort 
in your previous e-mail, but...

THIS IS HILARIOUS!

Running 3D applications is not the goal? So you want to buy this graphics 
card and just smell it or something? Your ultimate pleasure is reading the 
documentation? Oh yes, it's going to be very well supported on open-source 
platforms, so we can all watch how well it renders text?

That was mean... sorry, just couldn't help it. I hope it illustrates more 
than ever how important it really is to express the vision of a project, and 
that you just meant it other than I read it.

> A windows driver has been dscussed and considered a good thing. If the 
> card can support the game also then no need to change graphics cards.
If it can't, well.. that might be a problem for some indeed.

There are two options. Either we include into the vision that it should have 
sufficient performance to run some popular (but older generation) games, or 
we don't. In the latter case, which I currently think is the only viable, I 
don't think it should be considered a problem. There's a big difference 
between claiming it has the right performance, while it isn't capable of 
running these games, then just telling people to buy something else because 
the project just doesn't target it. Maybe it can end up being something in 
between, for good or for better, but 'maybe' is not part of the goals we're 
working towards.

> It is however simply unrealistic to expect to be able to develop a high 
> performance 3D graphics card in a short amount of time. It is very 
> realistic to try develop one that will do well for the lower intensiy 3D 
> graphics used in desktops nowadays. It is also quite feasable to develop a 
> graphics card that will do very well for home theatre/media center pcs.

Fully agreed. Don't forget to mention the platform again though, or you're 
working in an already filled market.

> It is also quite possible to do this without limiting ourselves with 
> regards to future 3D graphics development.

Absolutely. Just don't expect to equal NVIDIA or ATI in terms of 
performance, ever.

> So, for as afar as I am concerned, there is no viable way to enter the 
> market on a relatively tight budget and within reasonable time if you 
> focus on 3D graphics. Forgetting them would however be a big mistake, but 
> first priority is catering to markets that can be served on the available 
> budget and within a reasonable time. Gaming/high end 3D is simply not a 
> part of that.

That's what I kept saying. And if that's the project's vision then that's 
fine with me. Go create that card, I'll support it every way I can.

But right now it's not in the vision statement. It's not clear at all, and I 
see discussions about OpenGL 2.0 for the wrong reasons still going. If we 
want to work efficiently everybody has to know exactly what we're going for. 
No more, no less.

> That indeed means that suc a card cannot compete directly with NVIDIA and 
> ATI and the like in their primary markets, that is already clearly stated, 
> and is not a bad thing unless you are aiming for market domination. It 
> seems to me that the ogp is first of all aiming at filling a niche, not at 
> market domination.

Market domination is impossible, I do realize that. But what I and many 
others need is to specify that niche market a little better. Well, this 
e-mail shows your vision about it, and I largely agree with it, but there 
can still be confusion about the text currently on the website.

Best regards,

Nicolas