AMD Nano Blacklist Situation

[RussianSensation]

Agreed. People hating the $650 price need to ask themselves: what's the alternative? Price it lower and continue the epic fail bang-for-back strategy? What will that achieve? How will that help AMD?

We don't even need to go so far back to $199 HD4850/$299 HD4870 days but looking at recent times when AMD had $399 R9 290 against $499 780 or $549 290X vs. $699 780Ti or $299 R9 280X vs. $379 770 2GB/$449 770 4GB, that hardly produced massive wins in market share OR profits. The price/performance strategy barely worked with the 5850/5870 despite AMD essentially having the $250+ market all to itself for 6-9 months before GTX460/470/480 dropped. Since AMD cannot expect to beat AMD to market by 6+ months every generation, this strategy is not sustainable.

It's better for AMD to sell out a low-volume part for $650 rather than $400. As Apple has shown in the smartphone market, marketshare is virtually irrelevant. All that matters is profit.

I agree with you.

"Report: Apple takes 92% of smartphone market profits on just 20% of sales"

If AMD makes profits with 18-20% market share, it's actually a more sustainable going concern than having 40-50% market share with quarterly losses. We'll have to wait for Q3-Q4 2015 results though to see what's happening in their GPU division. I think the shortage of supply for Fiji cards is going to hold them back.

Surprisingly, even at $650-670, XFX Nano or Gigabyte Nano are currently OOS on Newegg.

The price of the Fury X on Newegg has actually risen from $649 MSRP to $680 and $700. Both cards OOS.

Amazon has PowerColor Fury X for $900, XFX Fury X for $1000, Sapphire Fury X for $1531. Are those prices crazy? Absolutely but if there are enough people paying them, then I guess AMD's CEO has made the calculated bet that it's better to maximize profits on early adopters (not surprisingly a similarly NV uses with its Titan series).

Again, that might not be good for customers, but Apple/nVidia/Intel have historically engaged in many anti-customer practices. But these companies are profitable while AMD is not. Customers still open their wallets so they tacitly accept being treated this way, despite complaining on the forums.

True. Also, certain brands have so few sales until the existing generation is discontinued/becomes very old in relative terms of its product cycle (Louis Vuitton, Hermes, Apple, NV), that because the consumers know that dramatic price drops are unlikely to occur in the short term, they are conditioned to not wait for months at a time for some future sale, because what's the point?

Perhaps, as a long-term strategy, the new CEO is trying to establish some brand equity and send a similar message that AMD is not a budget brand? Without a doubt some consumers do correlate price with quality whether it's warranted or not. That means for some consumers the fact that AMD constantly prices their cars for less is automatically correlated that AMD's products are never on the same level as NV's. Should AMD miraculously manufacture a superior flagship in the future, this perception won't just go away either.

Also, some gamers might not bite on the Fiji cards this generation, but let's say in 2016-2018 when it's time to upgrade, they'll recall that AMD didn't bother going after price/performance or didn't bother dropping prices quickly post-launch as they often have done in the past. As a result, should they find some other next gen AMD GPU appealing at launch, they won't bother waiting for months for the typical/expected price drops that AMD's cards have had in the past. I am just speculating but there are larger implications for the product/brand that result from keeping prices high and maintaining them longer. The new CEO can try new strategies, look at the data over 3-4 quarters and see what works and was hasn't worked. I think it's a lot better than doing the same thing over and over that hasn't worked for 5 generations as you and I both agreed.

Another implication is that if start with a low price, you don't have a lot of room to cut prices in case your competitor releases newer products. In this example, should NV beat AMD to market with Pascal, as a last resort, AMD could drop prices on Fiji products to lower levels to give them some time to respond with 16nm HBM2 GPUs. Had AMD started off with $399-499 pricing on Fiji cards, then they would have had no room to respond really. Since undercutting NV right off the bat wasn't really successful in converting market share/loyal NV customers, it seems futile especially after seeing $200 GTX960 outselling after-market R9 290s or $550 GTX980 absolutely clobbering R9 $300-350 290X. There is no way Lisa Su isn't paying attention to these trends so she probably consciously gave up on trying to convert the most loyal customers of NV and focused purely on profits.

---

It seems PC gamers keep voting with their wallet that they aren't concerned with how NV achieves the performance optimizations in games, all they care about is the end user experience. If that's the case, AMD must do the same or it's literally giving away all of its competitive advantages. I would really hate to see the PC industry turn to this, but in line with NV's GW's strategy that PC gamers seem to be embracing/have no problem with on the whole, the next step is for AMD to work directly with major developers to optimize DX12 code-path for its Asynchronous Compute Engines and lock out the source code from being optimized for NV cards wrt to features like TressFX 3.0. Since the free market failed to be efficient and didn't boycott GW's titles, which is clearly a conflict of interest of PC hardware developer directly influencing software development in exchange for co-marketing (i.e., monetary benefits), it's survival mode now for AMD. They can't continue being Mr. Nice guy.

 

[antihelten]

You're not getting it. Why don't we see retail samples of nVidia cards to see if they boost like the review cards (and anyone who thinks a 1% difference over 22 games proves anything needs to go to school for statistical analysis)? Why don't we see TPU give us frequency charts of cards that we know the clocks reduce over time like Titan, 780/ti, Titan-X, 980/ti if they are solely interested in giving us more accurate info? They changed up their review routine purely trying to get a gotcha on Nano.

Just to expand upon your point, these are TPUs reviews of the mini-ITX versions (MSI and Asus) of the GTX 760 (they never reviewed a mini-ITX version of the 970 as far as I can tell):

http://www.techpowerup.com/reviews/MSI/GTX_760_Mini_ITX_Gaming/
http://www.techpowerup.com/reviews/ASUS/GTX_760_DirectCU_Mini/

Differences versus Nano review:
No fan noise test in SFF case.
No clock frequency analysis.
No heat and noise video.
No temperature test in SFF case.

 

[flopper]

---

It seems PC gamers keep voting with their wallet that they aren't concerned with how NV achieves the performance optimizations in games, all they care about is the end user experience. If that's the case, AMD must do the same or it's literally giving away all of its competitive advantages. I would really hate to see the PC industry turn to this, but in line with NV's GW's strategy that PC gamers seem to be embracing/have no problem with on the whole, the next step is for AMD to work directly with major developers to optimize DX12 code-path for its Asynchronous Compute Engines and lock out the source code from being optimized for NV cards wrt to features like TressFX 3.0. Since the free market failed to be efficient and didn't boycott GW's titles, which is clearly a conflict of interest of PC hardware developer directly influencing software development in exchange for co-marketing (i.e., monetary benefits), it's survival mode now for AMD. They can't continue being Mr. Nice guy.

the async maxwell2 issue for dx12 will hit Nvidia hard.
AMD planned fully for dx12.

AND are engineers that innovate. Not sure what Lisa Su is going to do about that yet.
while keeping an open standard and thanks to that its a lot better they also need to borrow a few things from salespeople and marketing.

 

[Cookie Monster]

Just want to mention that coil whine has almost nothing to do with components. If there is coil whine, it literally means that the engineers who designed the DC-DC converter stage (or VRM) for the GPU power just didn't do a good job (OR sometimes its actually intended, a by product of the design). Making sure the inductors (chokes) dont whine by encapsulating the coil/inductor in an adhesive or some material so it cant vibrate is just masking the issue.

Most importantly it mostly means it ain't stable so the GPU voltage during the whine could be bad in terms of voltage ripple e.g. 1.1V +/- 0.01V to 0.05V. This could also affect your GPU clocks perhaps and most definitely would affect overclocks.

It is shocking to even get the whines on such an expensive product. Like they got lazy and didn't validate what should be a rock solid robust part of the video card. Not just for power delivery and efficiency but also stability. Companies like Asus, MSI harp on about some super alloy inductors etc (all marketing stuff that doesn't really mean anything) yet they get away with high pitched whining from coils which can drive people nuts! unless your hearing range is outside that is

 

[Flapdrol1337]

the async maxwell2 issue for dx12 will hit Nvidia hard.

I don't think so, they might have to lower prices a little, but that's only fair, considering the huge margins on maxwell currently. If you look at die size, memory bandwidth and powerconsumption a 390x should be quite a bit faster than a 980.

 

[antihelten]

the async maxwell2 issue for dx12 will hit Nvidia hard.

The Async issue won't hit Nvidia at all. By the time that we can actually buy any significant number of DX12 games using async, Nvidia will most likely be selling Pascal.

The people who will (potentially) be "hit hard" are Maxwell 2 owners, not Nvidia themselves.

 

[3DVagabond]

The Async issue won't hit Nvidia at all. By the time that we can actually buy any significant number of DX12 games using async, Nvidia will most likely be selling Pascal.

The people who will (potentially) be "hit hard" are Maxwell 2 owners, not Nvidia themselves.

Do we know that Pascal addresses async compute in hardware? Or are they going to continue to do software scheduling in the drivers? I haven't seen anything one way or the other. Considering that we just found out Maxwell doesn't have hardware support though, how would we know?

 

[Hitman928]

Just want to mention that coil whine has almost nothing to do with components. If there is coil whine, it literally means that the engineers who designed the DC-DC converter stage (or VRM) for the GPU power just didn't do a good job (OR sometimes its actually intended, a by product of the design). Making sure the inductors (chokes) dont whine by encapsulating the coil/inductor in an adhesive or some material so it cant vibrate is just masking the issue.

Most importantly it mostly means it ain't stable so the GPU voltage during the whine could be bad in terms of voltage ripple e.g. 1.1V +/- 0.01V to 0.05V. This could also affect your GPU clocks perhaps and most definitely would affect overclocks.

It is shocking to even get the whines on such an expensive product. Like they got lazy and didn't validate what should be a rock solid robust part of the video card. Not just for power delivery and efficiency but also stability. Companies like Asus, MSI harp on about some super alloy inductors etc (all marketing stuff that doesn't really mean anything) yet they get away with high pitched whining from coils which can drive people nuts! unless your hearing range is outside that is

I'm interested to hear what design changes you feel need to be made that would allow the engineers to eliminate coil whine from their designs.

 

[Gikaseixas]

Just want to mention that coil whine has almost nothing to do with components. If there is coil whine, it literally means that the engineers who designed the DC-DC converter stage (or VRM) for the GPU power just didn't do a good job (OR sometimes its actually intended, a by product of the design). Making sure the inductors (chokes) dont whine by encapsulating the coil/inductor in an adhesive or some material so it cant vibrate is just masking the issue.

Most importantly it mostly means it ain't stable so the GPU voltage during the whine could be bad in terms of voltage ripple e.g. 1.1V +/- 0.01V to 0.05V. This could also affect your GPU clocks perhaps and most definitely would affect overclocks.

It is shocking to even get the whines on such an expensive product. Like they got lazy and didn't validate what should be a rock solid robust part of the video card. Not just for power delivery and efficiency but also stability. Companies like Asus, MSI harp on about some super alloy inductors etc (all marketing stuff that doesn't really mean anything) yet they get away with high pitched whining from coils which can drive people nuts! unless your hearing range is outside that is

Many many quality GTX 970, 980 had coil whine, not once i saw some users here using it against NV. I guess there's not much bad points to talk about Nano so people focus on these things, lol

My 7970 has coil whine, so does my GTX 980TI but i only hear it if i'm playing games with my speakers / headphones off and with the case close to my head.

 

[sm625]

We don't even need to go so far back to $199 HD4850/$299 HD4870 days but looking at recent times when AMD had $399 R9 290 against $499 780 or $549 290X vs. $699 780Ti or $299 R9 280X vs. $379 770 2GB/$449 770 4GB, that hardly produced massive wins in market share OR profits. The price/performance strategy barely worked with the 5850/5870 despite AMD essentially having the $250+ market all to itself for 6-9 months before GTX460/470/480 dropped. Since AMD cannot expect to beat AMD to market by 6+ months every generation, this strategy is not sustainable.

It is the sad truth. AMD has held the performance per dollar crown for most price segments going back several years. It simply hasnt mattered one iota, and will probably never matter. To the uninformed consumer drowning in debt, brand image is the only thing that is ever going to matter. Raising prices could actually help with brand image.

 

[railven]

Considering how easy it is to buy an R9 Fury Nano right now versus a R9 Fury X, something tells me this isn't a low volume product and more akin to my original thought - Nano is gonna replace Fury X as the leader.

Higher margins, similar performance, smaller foot print, most likely higher desire due to no radiator for some users, lower power requirements.

If I recall correctly, the other two Furies didn't have this much availability this far after launch. Quick glimpse at Newegg, there are currently more Nano versions available to buy than Fury and Fury X is out of stock and down to two vednors.

EDIT: Or, I guess, Nano isn't as popular and hasn't gone through it's initial production? Either way, Fury X is pretty much a ghost at this point.

 

[4K_shmoorK]

It is the sad truth. AMD has held the performance per dollar crown for most price segments going back several years. It simply hasnt mattered one iota, and will probably never matter. To the uninformed consumer drowning in debt, brand image is the only thing that is ever going to matter. Raising prices could actually help with brand image.

I'd like to know the percentage of "uninformed consumers drowing in debt" that buy and build their own PCs. I'd wager the number of uninformed and clueless consumers of enthusiast and mid-range PC buyers/builders is relatively small. And I'm not sure individual debt ratio really plays a part in brand preference.

Uninformed and debt ridden is more indicative of console buyers, no?

 

[railven]

I'd like to know the percentage of "uninformed consumers drowing in debt" that buy and build their own PCs. I'd wager the number of uninformed and clueless consumers of enthusiast and mid-range PC buyers/builders is relatively small. And I'm not sure individual debt ratio really plays a part in brand preference.

Uninformed and debt ridden is more indicative of console buyers, no?

As an uninformed and debt ridden PC Enthusiast, you're wrong. I always buy a GPU I can't afford by charging it on my 28% APR interest credit card, and then make ONLY monthly minimum payments until the next card I want comes out.

Clearly you're doing it wrong.

 

[Unoid]

The Async issue won't hit Nvidia at all. By the time that we can actually buy any significant number of DX12 games using async, Nvidia will most likely be selling Pascal.

The people who will (potentially) be "hit hard" are Maxwell 2 owners, not Nvidia themselves.

As someone who just bought a 980TI I want this beast of a card to excel for at least 2 years. If it doens't, this WILL reflect badly on Nvidia. If I see AMD 290's and fury's kicking butt in DX12 in 2 years and my 980ti is a low performer, Then I will go AMD again.

If Nvidia has pascal why would I reward a company for planned obsolescence style shenanigans.

 

[4K_shmoorK]

As an uninformed and debt ridden PC Enthusiast, you're wrong. I always buy a GPU I can't afford by charging it on my 28% APR interest credit card, and then make ONLY monthly minimum payments until the next card I want comes out.

Clearly you're doing it wrong.

Who pays credit card bills? My gmail sends my statements straight to spam.

Out of sight, out of mind free GPUs.

 

[LTC8K6]

Amazon has PowerColor Fury X for $900, XFX Fury X for $1000, Sapphire Fury X for $1531. Are those prices crazy? Absolutely but if there are enough people paying them, then I guess AMD's CEO has made the calculated bet that it's better to maximize profits on early adopters (not surprisingly a similarly NV uses with its Titan series).

Hasn't AMD already made it's money from the mfgs before the cards are sold to the consumer?

Those prices don't reflect AMD profits, do they?

Or does AMD set consumer prices and not make money until a card is sold to the end user?

 

[railven]

As someone who just bought a 980TI I want this beast of a card to excel for at least 2 years. If it doens't, this WILL reflect badly on Nvidia. If I see AMD 290's and fury's kicking butt in DX12 in 2 years and my 980ti is a low performer, Then I will go AMD again.

If Nvidia has pascal why would I reward a company for planned obsolescence style shenanigans.

Scenario: Pascal is out and it's faster than whatever AMD has, at the same price point? Would you reward AMD, instead?

Who pays credit card bills? My gmail sends my statements straight to spam.

Out of sight, out of mind free GPUs.

That's how it is done! I haven't "paid" for a GPU since 1998!

 

[Madpacket]

To the uninformed consumer drowning in debt, brand image is the only thing that is ever going to matter. Raising prices could actually help with brand image.

This +1000

This works extremely well for Apple and other popular brands. AMD's major problem is an image problem, it always has been. This is how BlackBerry died in the smartphone market. Perception's a killer, when people think BlackBerry they think of old crusty work phones. By the time they released a modern smartphone (Z10 / Z30) it was too late.

As stated AMD's generally always sold better price to performance cards, now it's time to change that strategy. Lisa Su needs to get aggressive and be more cut throat like JHH. They've always had the engineering talent (at least for GPU's) so I agree with RS, time to shine a light on DX12 and ACE and get game developers (pay them, sit with them during development) to ensure they are using every ounce of power out of GCN tech. Luckily for AMD the graphic card market is much less competitive than the smartphone market and they have the consoles to help push along game development in this direction. Pascal better be built with ACE's or they're going to keep having to optimize for each major game title (this is much more expensive long term however one Nvidia can currently afford) and prematurely obsolete Maxwell cards as engineering resources shift to Pascal optimization.

It'll take a few years to change the majority of market perception but there will always be stragglers that spew idiotic statements like "don't AMD CPU's run hot and burn up?" or "I only buy Nvidia because ATI has has terrible drivers".

My theory on why Fury-X is non existent is simply due to margins. The Nano sells for the same price but likely costs much less to manufacture so why rush our more Fury-X's? If Nano's fails to sell as well as predicted they have the margins to drop the price and still make money, and then bring back the Fury-X's.

 

[4K_shmoorK]

This +1000

This works extremely well for Apple and other popular brands. AMD's major problem is an image problem, it always has been. They've always sold better price to performance cards, now it's time to change that strategy.

Yes. Except the people who buy apple products, beats headphones, and IO Hawks aren't the same people who spend time deliberating and researching whether to choose AMD or Nvidia for their next GPU. Or would even purchase a discrete GPU.

Two fruit come to mind, an apple and an orange.

Show me an AMD GPU that doesn't just get 1-10 FPS on average more than its Nvidia counterpart at the expense of power, noise, and heat and the masses will flood to team red.

I'd argue this theoretical GPU wouldn't even have to be cheaper, the Nano has had great sales and reviews despite its price to performance ratio.

Just look at the PS4 vs XBONE scenario. I'd say there were/are some pretty intense brand loyalties that exist between M$ and Sony fanboys. However, once it was reported that the XBONE was going to have always on DRM people flocked to Sony. And Sony didn't have to do a thing.

That is directly reflected through Sony's relative dominance in the console market, steadily beating M$ month after month in units sales/shipments.

I do think an ATI 'revival' would be awesome for AMD. Would do well to improve their image. :thumbsup:

 

[railven]

Yes. Except the people who buy apple products, beats headphones, and IO Hawks aren't the same people who spend time deliberating and researching whether to choose AMD or Nvidia for their next GPU. Or would even purchase a discrete GPU.

Two fruit come to mind, an apple and an orange.

Show me an AMD GPU that doesn't just get 1-10 FPS on average more than its Nvidia counterpart at the expense of power, noise, and heat and the masses will flood to team red.

I'd argue this theoretical GPU wouldn't even have to be cheaper, the Nano has had great sales and reviews despite its price to performance ratio.

Just look at the PS4 vs XBONE scenario. I'd say there were/are some pretty intense brand loyalties that exist between M$ and Sony fanboys. However, once it was reported that the XBONE was going to have always on DRM people flocked to Sony. And Sony didn't have to do a thing.

That is directly reflected through Sony's relative dominance in the console market, steadily beating M$ month after month in units sales/shipments.

I do think an ATI 'revival' would be awesome for AMD. Would do well to improve their image. :thumbsup:

If AMD brings back the ATI name, I would buy that GPU without even looking at charts. Frankly put, the AMD name to me has always been garbage (Intel fanboy) and I have no love for it. When Bulldozer flounder the stigma/stank of their CPU division stuck to the GPU. Then add in the overpriced AMD RAM and AMD SSDs, and well...

Image is definitely an issue for AMD.

 

[tential]

Considering how easy it is to buy an R9 Fury Nano right now versus a R9 Fury X, something tells me this isn't a low volume product and more akin to my original thought - Nano is gonna replace Fury X as the leader.

Higher margins, similar performance, smaller foot print, most likely higher desire due to no radiator for some users, lower power requirements.

If I recall correctly, the other two Furies didn't have this much availability this far after launch. Quick glimpse at Newegg, there are currently more Nano versions available to buy than Fury and Fury X is out of stock and down to two vednors.

EDIT: Or, I guess, Nano isn't as popular and hasn't gone through it's initial production? Either way, Fury X is pretty much a ghost at this point.

Fury x being a ghost is idiotic of them. That's an amazing card and platform to continue to build the flagship amd card.

It was a card so nice looking I wanted it end of discussion. Amd just had to go and say it was an ocers dream. I hate blatant lies.

 

[tential]

I'd like to know the percentage of "uninformed consumers drowing in debt" that buy and build their own PCs. I'd wager the number of uninformed and clueless consumers of enthusiast and mid-range PC buyers/builders is relatively small. And I'm not sure individual debt ratio really plays a part in brand preference.

Uninformed and debt ridden is more indicative of console buyers, no?

Lol.... I dunno why you think pc users are some superior race. Most pc users no nothing about it.

I bought an r9 290 used.

The person was using amd raptr, no idea how to oc, I actually helped them remove the gpu, they called me asking if I can install a new cpu for them sometime, and the firmware version was 14.4

The vast majority of gpu users I take to (since I don't know enthusiasts) know nothing about gpus compared to users on here, and have $300+ gpus....

On the forums I go to that aren't tech but have a lot of pc gamers? It's even worse.... People go "I have 2000 dollars and want to get into pc gaming what should I get?" and will build around ridiculous cards when they have no idea what they're doing. I had one guy tell me he couldn't play asseto (racing game blanking on name). He had every setting on maximum and couldn't understand why it wouldn't work. It's a 2000 pc it can play everything at maxed settings because it's expensive. He had ko idea about gpu settings or how fast his gpu is or wasn't.

Pc gaming does not mean you're a hardware enthusiast.... Just because yall care don't think people who are gamers first care. Their hardware is just a means to an end, which is why they are such massive fanboys...

 

[ocre]

I'd like to know the percentage of "uninformed consumers drowing in debt" that buy and build their own PCs. I'd wager the number of uninformed and clueless consumers of enthusiast and mid-range PC buyers/builders is relatively small. And I'm not sure individual debt ratio really plays a part in brand preference.

Uninformed and debt ridden is more indicative of console buyers, no?

You misunderstood what he was saying. We could blame this on the grammar

Anyway,

To the uninformed consumer drowning in debt, brand image is the only thing that is ever going to matter. Raising prices could actually help with brand image.

My translation,

"To the uninformed consumer, [AMD] drowning in debt and brand image is the only thing that is ever going to matter. Raising prices could actually help with brand image."

 

[ocre]

It is the sad truth. AMD has held the performance per dollar crown for most price segments going back several years. It simply hasnt mattered one iota, and will probably never matter. To the uninformed consumer drowning in debt, brand image is the only thing that is ever going to matter. Raising prices could actually help with brand image.

Now my take on this.............

I totally disagree, its way more complicated. AMD dropping ATI brand name was a huge mistake. It was done in a time when they were on top, the glory 5000 series. After the legendary Nvidia 8800 series, Nvidia had gained a huge chunk of mind share. Some of you guys may not be able to admit it but this was a major pivot point for nvidia. The buzz lasted for years.

AMDs sweet spot strategy was a strategy that initially worked and was quiet effective. It was when they gained back a large chunk of marketshare and were flying high, AMD drops the ATI brand name.
My personal belief is that the timing was no coincidence. I believe AMD thought this would prop up their damaged CPU image/brand. In actuality, it did the reverse.

It took awhile for the vendors to replace the ATI with AMD. I believe this was a huge massive mistake. The brand name was just not strong and the damage done by dropping ATI is irreversible.

I believe ultimately, this hurt them more than anything else.

 

[ocre]

the async maxwell2 issue for dx12 will hit Nvidia hard.
AMD planned fully for dx12.

AND are engineers that innovate. Not sure what Lisa Su is going to do about that yet.
while keeping an open standard and thanks to that its a lot better they also need to borrow a few things from salespeople and marketing.

you would say something like this
I seldom speak in such absolutes. Especially when there is absolutely no way you could possibly know.
For the record, can you admit this is what you want to see and that you really dont know.

See, what about the million other possibilities?

There is this Ashes benchmark and then a whole lot of FUD.

Your post, it is just FUD.

There will be games in the future, I am pretty sure of that. There will be DX12 games too. But i am quiet sure that very very few DX12 games will be anything like Ashes of Singularity. Very few will have so many things happening on such a large scale. AMD seems to gain more in that benchmark than nvidia, ATM. This could be the case for other DX12 games. But it is a huge stretch to say "maxwell2 issue for DX12 will hit Nvidia hard".....massive stretch.

I dont see many games at all being like Ashes. I surely dont expect there will be huge AAA titles in that style. Most big blockbuster games are very different, in every way. See, this game engine will not be the game engine for all DX12 games.

You are assuming so much and speak in absolutes, i hope you are just joking or being sarcastic. Unless maybe you have a crystal ball or some amassing gift to predict Nvidia's doom. Perhaps this is your first time making predictions like these, I dont know you, but my gut makes me think you could have made some grand predictions like this in the past.
When it comes to AMD, for a long time now people have been saying, "just wait till <insert future circumstance> then <inset competing brand> will be helpless"..........or we can switch out helpless to "hit hard" if you like.........oh wait, you already did.