> You've taken the words out of mouth.  The OGC1 is going to be designed and=
> =20
> marketed to be a general-purpose graphics card, to compete in a field that =
> is=20
> dominated by nVidia and ATi.  In order to succeed well enough to be able to=
> =20
> create an OGC1.1, OGC2 or whatever, we need to sell lots of them and make=20
> lots of money.

A general-purpose card to compete against nVidia and ATi?  Are you serious?

PCI only.  No AGP.  No PCIe.  No laptops.  No STB.

No high-end CAD.

No "l00t" high-end gaming.  (All important if you believe the reviews.)

No video.  (There goes a huge number of sales.)

And support for serious monitors is now in question.

On the plus side there will be documentation, allowing high quality drivers
to be written for BSD/Linux/Plan9/Mach/Minux/BeOS/OS-X/etc.

> Putting things like jumpers on a board cost money.  Each jumper will increa=
> se=20
> the cost of each board by about =A31

I wasn't suggesting that the pins need to be lovingly installed by Swiss
watchmakers.  If a pin costs a pound the rest of the board must cost thousands.

> extra testing time per board

Board test was automated 25 years ago.

> It's important to realize that the OGC1 is not supposed to be everything to=
> =20
> everyone.

Agreed.  I don't see anyone lobbying for high-end gaming.  Way too much to
bite off.

> What the target market need is a 2D/3D graphics card which is fully support=
> ed=20
> out of the box on free operating systems, assuming typical hardware.

I don't see what the target market for this chip is.  It has too little
for a general purpose computer, and too much (meaning too expensive)
for a kiosk/pda type application.
_______________________________________________
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