That's page is interesting.
However...http://wccftech.com/nvidia-planning-announce-price-cuts-geforce-gtx-660ti/
Perhaps 660ti isn't better than 7950 overall Lol
However...http://wccftech.com/nvidia-planning-announce-price-cuts-geforce-gtx-660ti/
Perhaps 660ti isn't better than 7950 overall Lol
How convenient that you forgot the $60 game coupon Nvidia included with each GTX660Ti. That made the cost to be more like $240.
So if the price drop to $279 (a $20 reduction) does not include the game coupon then Nvidia actually makes $40 more per sale.
Let me see...why when Amd launched 4870 Nvidia cuts immediately gtx280's price? The brand Nvidia wasn't strong enough?
A stronger brand can command a higher price. Say 10-20%? In the case of 4870 vs. GTX280, the price difference was WAY higher than that.
While still being inferior in every way to the 7850.
Except ; OGL, LINUX, Actual College CS work with CUDA cores...While still being inferior in every way to the 7850.
Except ; OGL, LINUX, Actual College CS work with CUDA cores...
Considering how many people across gaming forums list their new PC's with 560 Ti's, it's pretty obvious that the average gpu buyer doesn't have a clue.
So if the price drop to $279 (a $20 reduction) does not include the game coupon then Nvidia actually makes $40 more per sale.
I really strongly prefer this way of presenting benchmark information, rather than the usual case of just posting one game at a time. But it would be nice if they presented a bit more information about what the precise settings were (or, if they do, made it easier to see in the article). Showing absolute performance as well as relative performance would be nice as well.
These are the reference MSRPs.MSRP's appear to be becoming a joke. Quite a few AMD cards that were excellent values have now risen in price (disregarding rebates).
(newegg)
Used to be 299 7950's, now they are 319 to start.
Was 220-ish 7870's, and at least 249, now they are 259.
I really strongly prefer this way of presenting benchmark information, rather than the usual case of just posting one game at a time. But it would be nice if they presented a bit more information about what the precise settings were (or, if they do, made it easier to see in the article). Showing absolute performance as well as relative performance would be nice as well.
I really strongly prefer this way of presenting benchmark information, rather than the usual case of just posting one game at a time. But it would be nice if they presented a bit more information about what the precise settings were (or, if they do, made it easier to see in the article). Showing absolute performance as well as relative performance would be nice as well.
I really strongly prefer this way of presenting benchmark information, rather than the usual case of just posting one game at a time. But it would be nice if they presented a bit more information about what the precise settings were (or, if they do, made it easier to see in the article).
I half suspected that, though I didn't look closely at it. I know that nVidia has been presenting their information in this way for many years now (though usually they say a bit more about the precise settings). It just baffles me that reviewers have yet to catch on, as it's a tremendously better way to present relative performance information.It's (from) nvidia slides, so videocardz don't really now, as they also write.
Oh just realized this is GTX 560 non-Ti there so indeed nvidia managed to skew the results by some 25% which is indeed very impressive. Unfortunately for them it also means the GTX 660 is in reality slower than I thought, should end up just barely above 7850.It's (from) nvidia slides, so videocardz don't really now, as they also write. Expect the settings to vary for the best-case result in all games
We can also read from those slides that gtx560 is slightly faster than hd7850 just like gts450 is slightly faster than hd7750...
(so job well done in the benchmarketing department)
I'm not sure though even the nvidia-only results don't make much sense to me, the GTX 560 is up to 32% faster than the GTX 460 there, even though the former is pretty much just a 20% higher clocked (core clock, less difference on the memory side) version of the latter.
Wrong. Beyond clocks, the GTX560 enables one more SP cluster. While GTX460 has 336 cuda cores, GTX560 has the full 384.