All the more reason to price 380x level of performance, at GTX 960 TDP, at a slight premium to it's current rate.The runaway success of Maxwell demonstrates that efficiency is actually quite important to GPU buyers. If people only cared about performance per dollar, the deeply discounted R9 290 and 290X cards would have flown off the shelves. Instead, it was the GTX 970 that was 2015's big hit. If people only cared about performance per dollar, no one would have bought any GM206 card except for people who needed its HEVC decoding and/or HDMI 2.0 output capabilities.
Moreover, there is really no good reason to believe that a FinFET chip around 200mm^2 is going to be more expensive than an equivalent 28nm planar chip at 438mm^2. The "doomsday charts" that ShintaiDK loves to post don't hold water. Not only will a refined, die-shrunk performance equivalent to Hawaii probably be cheaper to make, it will also be able to use a narrower memory bus due to the introduction of memory compression and possibly GDDR5X. That drives down the total bill of materials.
But if at $350 AMD gains 10% additional marketshare for releasing a GPU first, or if they release at $270 AMD gains 20% additional marketshare, than clearly the $270 is better. It's more money, and more marketshare.
I wouldn't take AtenRa's post as fact; it's opinion the same as anyone else here. For one, his scaling assumption is removing two 128-bit memory controllers from Tonga to save 66mm². Even outside the fact that it's questionable you would save that much die area, Tonga is already slightly gimped vs the 280X with its 256-bit bus. A 128-bit part with 2048 shaders would be ridiculous. Even then, looking at the Tonga die removing 256-bits worth of MC likely wouldn't even get you to 66mm² saved.
If AMD's lead on the market is more than a quarter (*doubtful*), then AMD can even program in a price-drop just before nVidia releases Pascal, playing havoc with nVidia's market positioning and targeting plans, while looking like the good guy to most consumers (so long as they do so before news of Pascal's release is widespread... timing is critical).
Can't both of you see that overall profit is more than just unit profit. Sale volume is the other part of the equation. Simply put, total profit = unit profit X # units sold. Both of you keep getting stuck on maximizing unit profit, when the company is more interested in overall profits. The problem with the sale volume / unit pricing formula is that they are co-dependent variables and so an optimal solution is not as obvious as some may assume.
Actually 2x256 from Pitcairn, which is what is used in Hawaii, is smaller than the 384bit MC in Tahiti. They could actually save a lot of space compared to a cut down Tahiti MC, if that's what Tonga is using.
Here you go. I'm assuming you have switched Hawaii and TahitiNo one really knows what's up with Tonga, outside of AMD. Even then, the area on Tonga taken up by the memory controllers appears to be <66mm², so it wouldn't matter you replace it with.
Can you provide a link that the MC in Hawaii is larger than the one in Tahiti? I've never seen anything official on that.
The result of AMD’s memory interface changes is that between the die space savings from the lower speed controllers coupled with a number of smaller tweaks to improve density, AMD has been able to implement the larger 512-bit memory interface while still reducing the size of the memory interface by 20% as compared to Tahiti. Furthermore these space savings still allow for a meaningful increase in memory bandwidth despite the lower memory clockspeeds, with AMD being able to increase their memory bandwidth by over 10% (as compared to 280X), from 288GB/sec to 320GB/sec. The end result is a very neat and clean (and impressive) improvement in AMD’s memory controllers, with AMD reducing their interface size and increasing their memory bandwidth at the same time. The 512-bit memory bus does have some externalities to it – specifically increased PCB costs and requiring more GDDR5 memory modules than Tahiti (16 vs. 12) – but these are ultimately countered by the die space savings that AMD is realizing from the smaller memory interface.
Here you go. I'm assuming you have switched Hawaii and Tahiti
http://www.anandtech.com/print/7457/the-radeon-r9-290x-review
not saying it would not happen, I am just a skeptics these days. I always wait for the product release. remember some poster here claiming a 380 beating a 980? a 370x beating a 970? that is why I don't do rumors or speculations anymore.Guys, I don't know how you cannot understand this. Its a shrink. GTX980 will become GTX 1060 with new features, and lower TDP.
R9 390X performance/core count will be brought into lower tier of price, power consumption, etc. That is exactly what Roy Taylor implied in his interview page ago posted in this thread.
Based on rumors floating around, it looks like AMD will be out of the gate with new GPUs first on the new node, again. If AMD can offer GTX 980TI +50% as soon as June, I'm in!
No doubt about it. I'd definitely grab one if it had that kind of performance.
If AMD releases something 980Ti+50% in June, they'll sell every unit they can get out of the fab.
Some people are lacking in knowledge of the business and think AMD should just throw money at these foundries for not making wafer allocations. Keep paying for the air they breath. They think AMD should continue to try and leverage their 20 some odd percent market share. These are likely the same people who thought the consoles were a waste because of the slim margins.
They need to move a lot more units than they are for many reasons. Pricing/cost, market share, mind share, etc...
But they're all the geniuses and anyone who disagrees with them are labeled with whatever negative term they can squeeze by the forum rules this week.
Can't they do both with a Halo Titan wannabe and masking the "7950" equivalent as the flagship that's affordable? $330 970 is important to market share. But $1000 Titan X is important to mindshare with the Halo effect. If AMD is first their top GPU should be priced well above the 980 Ti, even if doesn't have to be a silly $1000. They just need a 2nd or 3rd tier that's in the $330 range to get the market to shift and build the market share, imo.
And AMD should try to make every single advantage they get count. More so if their lead is going to be as long as some expect. You can grow the VR segment with your discounted older gen cards. Your new cards should focus on getting you as much back per unit sold as possible.
When NV responds, you can then start a price war. Until then, take advantage of your position.
If I was in charge I'd just shoot for 980Ti + 20% with similar or less power draw for similar price point initially.
Of course I'd be holding the 980Ti +40-50% card for when nvidia countered my product with their new offering.
Time will tell how it unfolds in the end.
Thinking it was more 28nm holding AMD back than knowledge.
$999 - R10 Fury X ~ 980TI+50% HALO CARD in June, Do it!
$699 - R10 Fury ~ 980TI+30% with 175w TDP in August
The runaway success of Maxwell demonstrates that efficiency is actually quite important to GPU buyers.
Guys, I don't know how you cannot understand this. Its a shrink. GTX980 will become GTX 1060 with new features, and lower TDP.
R9 390X performance/core count will be brought into lower tier of price, power consumption, etc. That is exactly what Roy Taylor implied in his interview page ago posted in this thread.
Yup. AMD leveraging 20% marketshare to maximize profits is an INSTANT fail. You need to gain marketshare first, then maximize profits. It's exactly why tons of companies lose money while expanding marketshare first, then work to use that mindshare later to raise prices, and milk their customers for all they're worth later once they've developed a bound with the company.
Ever hear of a little company with tiny marketshare named Apple? Lots of ways to skin a cat and giving away your products for nothing is far from a sure bet.
If AMD can release a bad ass card this summer, they can start building their marketshare with my house. There are a lot of people left that pay for performance and value instead of a logo.