Another MX fiasco?

[Ailuros]

Hmmm how does the theory sound that some algorithms might get ported to lower end parts, much like GF4 MX got Accuview and LMA?

 

[Dave Baumann]

Hmmm how does the theory sound that some algorithms might get ported to lower end parts, much like GF4 MX got Accuview and LMA?

That would indicate increased performance, which the graph does not represent (on the basis that the y axis is relative performance).

 

[Ailuros]

That doesn´t sound like anything else but GF4MX, Ti4200, Ti4600 with an AGPx8 protocol. A dx7+ compliant low budget line this late?

 

[kid_crisis]

This being the case doesn’t 81M transistors for the NV18 seem a bit too much? We know that NV30 is the only .13um product so NV18 is going to be .15um; ATi could only scale down to the low end with Radeon 9000 by scaling back many of 8500’s performance features to cut down on the 60M transistors it uses.

What you said about the NV18 makes perfect sense. I am just saying that WAS what the chart said. You've almost got me convinced it was a typo.

And if that is true, then did they also have a typo for the NV28 number, which was 86 million? Ahh, well, should be an interesting next week....