Dude, I haven't bought an AMD card for my own use in years, but if 390X had been in the market at the same time as the Titan X, and the 390X was 25% faster
and $200 cheaper, then I'd have gone with the R9 390X.
But the reality is, AMD didn't show up to the fight, and NVIDIA won by default. I think it was AMD's own people who stressed how important showing up to the fight is:
http://www.anandtech.com/show/2937
Good post. AMD did technically lose this generation because NV used a new gen architecture to beat up AMD's old architecture uncontested since GTX750/750Ti launched last summer. At this point R9 390X could beat Titan X by 50% at $499 and you can't take away all the profits and positive PR and millions of sales NV already got with GM206 and GM204 and now the Titan X. Every day AMD is behind, people keep buying Maxwell cards. That means like you said if you are late, the next time AMD will have a chance to attract those consumers is maybe in 2-4 years from now when their upgrade cycle comes up again. That's the point - if you miss the upgrade cycles, it doesn't matter how much better your card is than 960/970/980/Titan X because 970/980/Titan X owners aren't going to sell their cards now.
Unlike NV that managed to win the Fermi generation despite being 6-9 months late, and yet again was 2.5-9 months late with Kepler top-to-bottom roll-out, AMD can't do the same since most people buy AMD cards not because they are loyal to AMD, but because the cards offer good value. Someone who had any of the 4850--->7970 series eventually bought Dying Light or GTA V or AC Unity, got bad performance and just said forget it and bought 960/970/980. That sale is now lost because that gamer is not going to care now about R9 300 series. Once the user switches to the NV eco-system, there is also an added risk to AMD that they will like NV's features more, for example. That means in 2-4 years when their upgrade cycle comes up, they might not even consider AMD. Now you've permanently lost your customer. Oh oh.
If we look at historical market share, once many gamers left ATI during HD2900 series, they didn't really come back to ATI and stuck with NV since GeForce 8. AMD is making this exact same critical mistake for the 2nd time. This is such a risky move that it could cement AMD's GPU market share in the 20-25% range for 5-7 years to come just because once a user switches, it's very difficult to get them back. Marketing studies have shown for years that it's ~6X more expensive to get a new customer than to retain a current one. AMD's short sighted strategy with late R9 300 series launch is not just an engineering milestone miss, but a business failure that goes beyond hardware design. They are making a critical business mistake when it comes to customer retention and brand value.
More so, if we take a loyal AMD customer who waits and waits and finally switches to NV, they are going to be even less likely to return to AMD since they will feel betrayed and let down by AMD. The brand agnostic customer will simply upgrade when the game he wants to run needs more performance. It's like everything AMD did well with graphics during HD4800/5000/6000 series is what NV is doing now, but with much better marketing too! :biggrin:
The sad part about this is that with less market share, AMD has less earnings, less ability to invest into future GPU designs, thus less chance to make impressive products. Ultimately that leads to higher prices and lower level of competition when NV is in full control. The end result is consumers end up paying $550 for a mid-range GTX960 that NV can now call 980.
The only people shifting goalposts are the ones comparing GTX 750 Ti to 285(LOL?) and GTX 960 to 280X/290/290X(LOL?), they know who they are.
That's because we are consumers/gamers not graphics engineers. Since I am not an AMD or NV engineer, I could care less if a 1mm2 chip with 100W of power beats my 600mm2 250W chip in perf/watt. When I am playing a game, it makes no difference to me how large or small the chip is. As long as it runs cool and quiet and gives me stable X performance for Y dollars, that's what I care about. The reason people compare GTX750Ti to R9 270/270X or 960 to 280X/290 is because they are
closely priced. I know this concept is hard for you to understand, as well as the concept of price/performance. See, your perf/watt concept that you keep cherry-picking has a major gap in it because you compare perf/watt on a card vs. card basis but since I cannot play videogames without the rest of my components, as ga gamer if I truly did care about perf/watt for gaming, the ONLY proper comparison is Total System Power perf/watt. Why? Because it's about the total amount of power requires to generate a certain level of IQ and performance on the screen. Since the graphics card cannot do it without other PC parts, comparing perf/watt on a GPU vs. GPU basis is an engineering basis, but comparing perf/watt on a Total System power usage is a gamer's basis!
But of course you never want to talk about this style of perf/watt comparison because your entire argument falls apart. An i5/i7 system with an R9 290 is 50-60% faster than the GTX960 system and it uses well about 50-60% more power. Therefore, that 960 system is not really more efficient.
http://www.techspot.com/review/946-nvidia-geforce-gtx-960/page7.html
Obviously, the exact same with R9 270X that beats 750Ti by 40-43% and the entire system might use just 20-30% more power. There goes your perf/watt argument.
Your argument is akin to some German fanboy arguing that BMW's engineers are amazing because BMW M4 accelerates 0-60 mph in 3.9 seconds with "only" a 3.0 twin-turbo V6 engine but Cadillac needs a
3.6L Twin-Turbo V6 in the Cadillac ATS-V to get 3.8 seconds. If your #1 priority in life is perf/Litre or Perf/watt, by all means keep hyping up 750ti and 960.
Maxwell embarasses AMD in performance per watt and AMD had to cut the 285 from $249 to $199 to match the GTX 960's price, losing money because their chip is much bigger and requires more expensive power components and PCB.
Umm....AMD and NV cut prices of GPUs all the time when there is healthy competition.
Welcome to the GPU industry. Also, unless you work for AMD, you cannot possibly say if AMD is losing money selling the 285.
So we have two R9 300 cards which show signs of a much better efficiency than current cards. That must come from some place, and who knows what it is. Whstever it is, it must be good since TweakTown yesterday said that the R9 300 cards will have something secret that no rumor have spoken about yet which didnt involve HBM but something spec wise..
AMD has improved perf/watt many times before. For some reason this generation though many posters on our forum think it's no longer possible. All those HD4000-7000 generations must have been just magic or pure luck that can't be explained by engineering skills. It makes sense why all the re-branding rumours popped up. Since people can't explain to themselves how AMD can improve perf/watt, and since AMD is stuck on the same 28nm node, the only logical conclusion is AMD will just release the exact same products again, just call them R9 300 series and throw in HBM1 for fun cuz they are bored.
The comparison was meant to align the respective architectures GCN 1.2 vs GM107 from a generational/technological standpoint not the cards specifically. Hence Maxwell being vs ??? due to AMD being late to reply to it. The point wasn't what was competing with what in the market at a given price.
:thumbsup::thumbsup:
Right now we are comparing Maxwell which is 1 generation ahead of anything AMD has. Why would it be surprising that NV's Maxwell is pounding old GCN cards? The only way to gauge just how amazing Maxwell really is in its respective generation is to compare it against AMD's respective generation's architecture. Since no such architecture has been released yet, we are not comparing apples-to-apples when we compare 290/290X against 970/980 and Titan X. Unless people think R9 390 series is a generational competitor to Pascal, then we have to conclude that R9 390 series is a Maxwell generation competitor.