[stir] it worked ;) > >>[0]Supermathie writes "NVidia has finally released drivers for their > >>chipsets and the 2.6 kernel that support 4K stacks.
> Rubbish? Do you have a reason for stating this? Perhaps a reason > backed up some very good objective analysis of why nVidia should > release the source code to their driver? My objective analysis why I want source code for drivers is pretty simple: so I don't have to pussyfoot around and instead have it just work when I boot the kernel. > Just in case anyone else is wanting to take a swipe at companies > like nVidia is it worth me mentioning that this driver release has > bumped performance on my nVidia FX5600 Now we were talking about mobo chipsets, or? Their mobo chipsets are nothing but a PITA, and no better than all the other stuff on the market. > It sounds like Volker has not watched a high-spec nVidia 3D card > doing its stuff on linux... True, but that's beside the point. If their graphics cards are currently the only good ones available, I'd prob buy one too. As soon as the competition has an equivalent card with an OSS driver, nvidia goes West. Please keep these two points apart. > better to have good support and a binary only driver than no 3d at all. True. But the mobo doesn't do 3d. > nVidia have lucrative IP technologies at > risk...protecting that from prying eyes is only protecting their > core business. Mind if I laugh? Gee, I can buy equivalent ethernet stuff for fifteen bucks in town. Driver? What driver? It just goes. Dito for the rest of the stuff on their mobos. Perhaps for their graphics cards, even there it's dubious. For anyone with the right tools (hardware companies have those) and and engineering degree (not rare) it's a question of time to reverse engineer nvidia's chips. The time is in the order of weeks, not years. You're on shaky ground with those arguments. You can't keep preying eyes out. Volker -- Volker Kuhlmann is possibly list0570 with the domain in header http://volker.dnsalias.net/ Please do not CC list postings to me.
