Is anyone following me into Silicon Graphics workstations and wants to chat about how rad they are?pic.twitter.com/4KYba3Vmvd
-
Depth buffering was faked on XL/Newport boards on the Indy by loading a 32-bit pixel mask register, meaning any 3D geometry had to be drawn in horizontal 32-pixel spans, max. That's all you got unless you were using painter's algorithm and ate the overdraw cost.
Things weren't all sunshine and roses even with the XZ board on the Indy that provided accelerated 3D. Anecdotally, a top-end R5000 CPU module was better off with a 24-bit XL board, since the CPU could transform and light geometry faster than the XZ board itself.
-
-
The Indigo 2 and Octane had similar rough aspects, graphically: The boilerplate SI/SE and SSI/SSE lacked hardware-accelerated texturing. On the I2, only the High IMPACT and Maximum IMPACT boards have hardware-accelerated texturing.
-
On the Octane, it gets even more weird, as you could turn an SI/SE board into an SI+T/SE+T board, with hardware-accelerated texturing, by adding a TRAM card. Or you could go with an MXI/MXE board. To that end, the first-gen Mardigras (MGRAS) pipe was the same as in the I2.
- Näytä vastaukset
Uusi keskustelu -
Lataaminen näyttää kestävän hetken.
Twitter saattaa olla ruuhkautunut tai ongelma on muuten hetkellinen. Yritä uudelleen tai käy Twitterin tilasivulla saadaksesi lisätietoja.