Given that Apple is buying in pairs. if the price is $200 higher per card, that is $400 higher in component costs.
A few things:
- Apple is buying in bulk so they're not paying $200.
- This is an upgrade cost we're talking about, so you have to remove the cost of the baseline card (unless this is both the baseline and max card, which would be a little strange.) Apple's upgrade cost would probably only be around $350, so $500 would be a reasonable upgrade price.
- I'm not really talking about Apple's pricing here, I'm talking about the end price to the consumer. The Mac Pro sells to a market that doesn't really care as much about price. Apple could sell a GeForce 1080 dual GPU upgrade for $2000 and people would probably buy it (although they'd complain a bit.) Price is totally the opposite optimization direction you'd want to go for the nMP. Performance per watt is where you'd want to go, which is where Polaris seems to be stumbling.
Price isn't the only factor (just like gamer benchmark 42 isn't the only factor), but it is in there. At the top end of the Mac Pro market ( folks who regularly buy full tricked out D700 and eye-popping 12 core BTO options)? Yeah that is pretty elastic. Another $500-800 won't do much damage. At the lower end though... even Apple's move suggest there is a difference.
Right, but remember that extra $500-$800 is putting it in direct competition with both the 1070 and possibly the 1080. The 1070 is reasonably faster (with the same wattage), and the 1080 is much faster. That's why taking a $200 card that doesn't perform that great, marking up the price to something ridiculous, and then shipping it would be suicidal.
If we were looking at a more efficient card the story would be different. But if they're going to charge 1070 level prices they need to deliver 1070 level performance. If Polaris was more efficient Apple could at least overclock to try to gain ground back.
Again, I think Vega is the only way out here on the AMD side. Polaris has really bad performance per watt compared to Nvidia which is critical to the Mac Pro. At the very least, you could hope that Polaris is just an early sample and that Vega will be much better. But if this is Apple's top end option, they're in trouble.
(It also worries me about the Macbook Pro. Hopefully Polaris 11 is a better binning or something. But at this point almost anything would be better in the Macbook Pro.)
Personally I don't care as long as they finally provide proper OS X drivers. GTX 980 performance for just $200? Would be a neat choice for my Hackintosh.... Guess it's finally time to track down those Hackintosh-related Radeon bugs...
GTX 980 level performance is a problem with the 1080 out now. Apple doesn't have to match the 1080 (especially in since we knew Polaris would not be going head to head with the 1080 in raw horsepower), but they at least have to be somewhere around the performance. I was much more expecting something in the 7 gigaflop range, not 5 to 5.5.
Apple isn't competing with the workstations that were out last year, they're competing with what is out this year.