Interesting Orton Interview

[Pete]

NV30 peformed pretty well against R300 in then-current games; it was next-gen, SM2 benchmarks that exposed its weaknesses. I don't believe R520 will be weaker than G70 in SM3 games or benchmarks to the same degree NV30 was weaker than R300 (1:2), so really all we have is a pretty significant (because G70 is already here) delay, but still not NV30 length (which was six months, IIRC).

 

[ANova]

Or maybe pulling an "NV30" is now a noun to describe a product that arrives very late to market and underperforms to the expectations of the enthusiast community.

Well the R520 is not as late as the NV30 was, yet. Late does not automatically equal bad performance either.

 

[Bjorn]

Well the R520 is not as late as the NV30 was, yet. Late does not automatically equal bad performance either.

No it doesn't. And it sure sounds as Orton have some faith in the R5xx series. But on the other hand, NV was pretty positive about the NV30 series also back in those days.

 

[phenix]

I wonder how Nvidia spent the time advantage they gained by ATI's delay. I am guessing that ATI will attack Nvidia on image quality front as soono as they launch R520 and if Nvidia cannot solve their texture shimmering problem soon ATI may gain some ground (at least among the purists such as myself).

 

[incurable]

I wonder how Nvidia spent the time advantage they gained by ATI's delay. I am guessing that ATI will attack Nvidia on image quality front as soono as they launch R520 and if Nvidia cannot solve their texture shimmering problem soon ATI may gain some ground (at least among the purists such as myself).

Personally, I think history shows that you shouldn't even try the IQ angle unless your part is considerably faster in the majority of benchmarks. Yes, you might gain some credit with us die-hards, but the general population is simply going to ignore it. Even just as fast and better IQ wont cut it, mostly because you probably have to see the difference side-by-side, only FASTER and better IQ might give you some leverage.

Also, price is always up there with performance (and generally much more important than IQ) when it comes to making a purchase decision, and the GTX is available below 450 Euros already. It's going to be interesting how R520 will fit in there.

 

[phenix]

Personally, I think history shows that you shouldn't even try the IQ angle unless your part is considerably faster in the majority of benchmarks. Yes, you might gain some credit with us die-hards, but the general population is simply going to ignore it. Even just as fast and better IQ wont cut it, mostly because you probably have to see the difference side-by-side, only FASTER and better IQ might give you some leverage.

Also, price is always up there with performance (and generally much more important than IQ) when it comes to making a purchase decision, and the GTX is available below 450 Euros already. It's going to be interesting how R520 will fit in there.

From what I read over these boards I expect nothing new feature-wise (at least not earth shatteringly new) from R520. SM3 is already there, nobody is expecting a breakthrough in performance against Nvidia, and I seriously doubt ATI would get into a price-war against Nvidia although they may have a slight advantage due to a more advanced process tech. What is left for ATI to use against the competition in order to sell their product: best image quality and better/wider multi GPU application. They will try to win the heart of geekdom and expect the rest of the world to follow. I am not suggesting that they will win this bet I just predict that this may be their fighting plan for this generation.