Gamespy vient d’interviewer John Carmack, le dieu vivant qui a travaillé sur tous les moteurs 3D des jeux d’ID Software. L’interview en elle-même n’est pas très inintéressante mais WTF! it’s Carmack!!
Comme dans toutes les interviews de John Carmack, les passages que je supposent être les plus interessants sont ceux auxquels je ne cromprends strictement rien:
John Carmack: In the next five years, all of the flexibility issues should be worked out of hardware acceleration, so we really should be doing renderman-like content creation without the hard limitations that we live with today. Ten years is harder to say. With all the training currently tied up in it, I don’t expect the basic modeling actions to be changing fundamentally, but there may be some radically different rendering models that use that data. Dealing with all the analytically difficult issues with something like distributed ray tracing solves lots of problems, but isn’t very compatible with existing hardware acceleration methods. More important would be breakthroughs in input/output devices. A truly good VR headset would be a big deal, but there are a lot of challenges still to be addressed there. Now that we have consumer cards with lots of 3D muscle and dual monitor outputs, I am rather looking forward to some more attempts at it, but attitude sensing is still not really nailed down well. When they get to the point of « putting on sunglasses » and they work like the movies always showed them, it will be significant.
Amen…
