Info from John Carmack (6)

Bernd Kreimeier (Bernd.Kreimeier@NeRo.Uni-Bonn.DE)
Mon, 3 Jun 1996 19:45:57 +0200 (MET DST)

Date: Mon, 3 Jun 1996 19:45:57 +0200 (MET DST)
From: Bernd Kreimeier <Bernd.Kreimeier@NeRo.Uni-Bonn.DE>
Message-Id: <199606031745.TAA05060@marvin.nero.uni-bonn.de>
To: quake-dev@gamers.org, bernd@marvin.nero.uni-bonn.de
Subject: Info from John Carmack (6)

In march, following the CeBit fair, I asked about support
of the S3 ViRGE and ViRGE/VX. I add that one short reply
to the discussion of 3D hardware in the more recent
posting.

b.

----- Begin Included Message -----

From: John Carmack <johnc@idcanon1.idsoftware.com>
Date: Sat, 16 Mar 96 15:00:50 -0600
To: Bernd Kreimeier <Bernd.Kreimeier@NeRo.Uni-Bonn.DE>
Subject: Re: Verite premier platform

We are supporting both S3 and rendition for release.
Rendition should be a bit faster.

John Carmack

----- End Included Message -----

----- Begin Included Message -----

>From johnc@idnewt.idsoftware.com Fri May 24 22:32 MET 1996
Mime-Version: 1.0 (NeXT Mail 3.3risc v118.3)
Content-Transfer-Encoding: quoted-printable
X-Nextstep-Mailer: Mail 3.3 (Enhance 1.0)
From: John Carmack <johnc@idnewt.idsoftware.com>
Date: Fri, 24 May 96 15:32:46 -0500
To: Bernd Kreimeier <Bernd.Kreimeier@NeRo.Uni-Bonn.DE>
Subject: Re: q&a (4)
Content-Type: text/plain; charset="us-ascii"
Content-Length: 3183

You wrote:
> Q: Brush.h in the QuakeEd sources is obsolete, it refers to
> EditWindow.h and is not imported at all. Is this part of something
> new (Brush Editor) that is not released yet, or a leftover?
>

It was a leftover. Delete it.

>
> Q: qbsp, vis, light and qlumpy are written in C? ObjectiveC? If the
> latter, using only common classes (in gcc libobjc.a), or anything
> more NS specific?

Pure C. ( Pure ANSI, in fact, except for cwd() ) They compile on
practically anything.

>
>
> Q: You wrote:
> >I am eagerly looking forward to writing a brand new editor for
> >windows NT + open GL as soon as Quake ships.
>
> Reminds me of two drawbacks I foresee with the Verite cards:
> a) Rendition never indicated OpenGL support (but supports HEIDI)
> b) judging from their zero-response, XFree86 support will suffer from
> NDA and Matrox-style attitude. Am I wrong? I'd rather be wrong: the
> Verite looks very interesting, esp. if programming the RISC would
> be an option for pedestrians, too. The Technical Specs on the
> Chromatic pages are much more promising in that regard, i.e. open
> to the public.

I doubt we will see a verite GL driver, but I did mention it to them.
They are dram based, so it isn't a real workstation class card anyway.

They are pretty decent folks, so I don't think XFree86 support is out
of the question. I'll mention it next time I talk with them. I
suspect they are just in crunch mode and don't have the time to
devote to freeware projects.

I think being open with the RISC programming would be to their best
advantage, becaue demo-coder types would be able to do incredibly
impressive stuff with it.

> Q: In addition to S3 and Verite: have you had the time to evaluate
> other cards/chips? Pyramid3D or Chromatic? Not the GLINT300SX for
> obvious reasons. Too bad 3D Labs targets a different market, they
> already support a Mesa driver project.

GLINT300TX + GLINT delta will be a kick-ass workstation setup.

3DFX has the best fill rate around.

ATI is no good, because it doesn't have a Z buffer at the moment.

I haven't evaluated Pyramid3D or Chromatic.

I have an intergraph GL workstation, which is a solid performer,
and they have new technology coming out in a few months, but it
isn't consumer level stuff.

BTW, I just became aware of Mesa last week, and I am very impressed
with it. I hope to contribute something to the project when I have
more time. I may compatability check my new editor against it.

>
> Q: Idle curiosity: Michael Abrash mentioned a "Vertex-free
> surfaces" approach to Visible Surface Determination that sounded
> like using a MAP-like representation for rendering as well. Have
> you looked at the NEC/VideoLogic PowerVR chip, which, to my
> understanding, uses a similar representation?

It was more like using an ultra simple bsp, like the clipping hull,
to directly generate the scene. It was cool, and very very space
efficient, but not fast enough. It would run great on the PowerVR
chip, but because that is such a strange (not mainstream) paradigm,
I don't predict great acceptance of PowerVR.

John Carmack

----- End Included Message -----