AMD de-thrones nVidia with DX12!

[SteelCrysis]

Originally they lumped XBONE and 360 sales together, now everything in consumer hardware gets lumped together.

Yes, just like Valve recently did by lumping the Steam Machines in with the Steam Controller, etc.

[SickBeast]

Here you go, guys, I found some information on this:

http://www.thestreet.com/story/13162108/...-risk.html

Some like taking Microsoft down a peg at every opportunity, and a couple of stories surfaced saying that Microsoft was losing money from Xbox sales. But the figures that Microsoft reported that year were merely increases in revenue and cost of revenue figures, and not the total sales and cost of revenue figures. Microsoft said that the company sold 11.4 million Xbox consoles during the year. Since the company did not break down the actual sales for the different types of consoles, we can only use educated guesses to arrive at actual sales and cost of revenue figures for the platform. My best guess? In the ballpark of $6.8 billion revenue and $5 billion for Xbox cost of sales in 2014. This obviously means that the segment is big enough to have a tangible impact on Microsoft's top and bottom lines, as well as its margins.

[RolloTheGreat]

Don't you get it SickBeast?!

Sony WINS! They have sold more consoles, they are the WINNERS!

By supporting their triumph, I too somehow am the winner!

All hail Sony! All hail GStifford! Champions of the world!

[SickBeast]

Is Stifford supposed to be somehow derogatory?

[RolloTheGreat]

Is Stifford supposed to be somehow derogatory?

No way! It's the champion salute!

[BoFox]

Here you go, guys, I found some information on this:

http://www.thestreet.com/story/13162108/...-risk.html

Some like taking Microsoft down a peg at every opportunity, and a couple of stories surfaced saying that Microsoft was losing money from Xbox sales. But the figures that Microsoft reported that year were merely increases in revenue and cost of revenue figures, and not the total sales and cost of revenue figures. Microsoft said that the company sold 11.4 million Xbox consoles during the year. Since the company did not break down the actual sales for the different types of consoles, we can only use educated guesses to arrive at actual sales and cost of revenue figures for the platform. My best guess? In the ballpark of $6.8 billion revenue and $5 billion for Xbox cost of sales in 2014. This obviously means that the segment is big enough to have a tangible impact on Microsoft's top and bottom lines, as well as its margins.

Whew, Microsoft..  took very little gamble by being such a low-roller (like Bill Gates only playing low stakes Blackjack and never playing over at the high-stakes table - I learned this from watching a biography documentary), so the gains are just as low as the risk taken, by going with an ultra-cheap APU. 

BUT...  $2 billion on Minecraft?  That much $$$$$ for ummm...  excuse me, Minecraft???  Microsoft could've made 50 AAA-quality Kinect games with that money!  Or even 100 remastered top-rated Xbox360 games for XBone with much improved graphics and textures, etc..

[ocre]

What thread is this again?

[RolloTheGreat]

BUT...  $2 billion on Minecraft?  That much $$$$$ for ummm...  excuse me, Minecraft???  Microsoft could've made 50 AAA-quality Kinect games with that money!  Or even 100 remastered top-rated Xbox360 games for XBone with much improved graphics and textures, etc..

http://www.polygon.com/2015/6/30/8872503...ion-copies

At 70m copies of Minecraft sold, and 20m on Xbox, Minecraft is a good franchise to control.

For now MS collects money from Sony customers for Minecraft sales. Next version they make XBone/PC/mobile and kick Sony in the nuts.

Seems reasonable.

[SteelCrysis]

http://www.techpowerup.com/215663/lack-o...tx-12.html
Guaranteed to be the result of more bribes from AMD.

[RolloTheGreat]

http://www.techpowerup.com/215663/lack-o...tx-12.html
Guaranteed to be the result of more bribes from AMD.

Eh, could be the result of NV optimizing for games likely to be played on Maxwell.

Back when I had my FX5800 Ultra, the rallying cry was, "But the 9700Pro has better DX9 and Half Life will be DX9!". By the time Half Life launched, next gen cards were out.

Might be somewhat quicker adoption for DX12 as MS has much interest in making XBone and Win 10 successful, but we'll see.

[BoFox]

http://www.techpowerup.com/215663/lack-o...tx-12.html
Guaranteed to be the result of more bribes from AMD.

Well, we shall see.  Oxide made quite a blog about it, which can be read here:
http://wccftech.com/oxide-games-dev-repl...ntroversy/

Then there's a P.S.:

P.S. There is no war of words between us and Nvidia. Nvidia made some incorrect statements, and at this point they will not dispute our position if you ask their PR. That is, they are not disputing anything in our blog. I believe the initial confusion was because Nvidia PR was putting pressure on us to disable certain settings in the benchmark, when we refused, I think they took it a little too personally.
AFAIK, Maxwell doesn’t support Async Compute, at least not natively. We disabled it at the request of Nvidia, as it was much slower to try to use it then to not.

I'm guessing that since Oxide isn't a big dev like Ubisoft, NV won't bother with getting compute to work in a similarly optimized way with their drivers.  If there will be many console ports designed with Async based on GCN arch, Nvidia will have to be putting in a lot of work to get their older GPUs to run well with this DX12 "feature" as promised on their Maxwell cards. 

But then in the end, the dev admits this:

As you’ve pointed out, there does exist a marketing agreement between Stardock (our publisher) for Ashes with AMD.

So, a grain of salt, please!

[ocre]

http://www.techpowerup.com/215663/lack-o...tx-12.html
Guaranteed to be the result of more bribes from AMD.

Wholly smokes!!!!!

did you read all of it. Like the last paragraph?

Given its growing market-share, NVIDIA could use similar tactics to keep game developers away from industry-standard API features that it doesn't support, and which rival AMD does. NVIDIA drivers tell Windows that its GPUs support DirectX 12 feature-level 12_1. We wonder how much of that support is faked at the driver-level, like async compute. The company is already drawing flack for using borderline anti-competitive practices with GameWorks, which effectively creates a walled garden of visual effects that only users of NVIDIA hardware can experience for the same $59 everyone spends on a particular game.

AMD might be hurting, even on their last leg, but they manage to come out with this heavy handed, negative, attack campaign. They are definitely behind this. The whole thing......
starting with the GCN Asynchronous article at anandtech. Then they get with their favorite lowly developer and work up some benchmark. It is so incredibly obvious they rushed this thing out so fast and there is really no other reason why. I mean, come on.....who sends pre beta benchmarks to the press and urges them the publish the results? An alpha stage benchmark, sent to multiple press outlets and results being used in a smear campaign. Clearly their is a real and true motivation behind all this.

Who is even interested in the game that this benchmark is supposed to be based on? The game that is supposed to be coming out yrs from now? This was all rushed out in some master scheme. Really, if you cant see the effort behind this, you just arent looking.

This alpha benchmark sent to the press first finally got released for public download. This came after the fact. It was much more important to spread the negative word than for people to actually check out the work the developer should want to showcase. Has any of you guys even looked at Ashes? I mean, impressive? Anything but. Its not anything special at all.

The DX 12 games i will play will be very very different than this. I wouldnt expect the ones that start up with an AMD logo to perform as well, of course AMD GPUs will get better results than Nvidia....at least an launch.

Nvidias architecture is different than GCN. They are not the same. I cant wait till nvidia responds, I full expect them to. But the thing is, nvidia has tried to take the high road as of late when it comes to these AMD smear campaigns. Certainly Nvidia can gain performance in this benchmark but it appears this developer may be more interested in drama than anything else. It makes sense to me that a lower level API would and could easily favor one HW over another. GCN and maxwell are different, so it is not so strange that their would be issues. That nvidia would want to do things differently.

I really cant wait to see how Nvidia responds to this. The star swarm busted right in their faces but this time with DX12 it may take more than a driver. But i really doubt this developer would be willing to let nvidia have specific code in their benchmark. But, i dont imagine nvidia will take this sitting down.

[BoFox]

http://www.techpowerup.com/215663/lack-o...tx-12.html
Guaranteed to be the result of more bribes from AMD.

Wholly smokes!!!!!

did you read all of it.  Like the last paragraph?  

Given its growing market-share, NVIDIA could use similar tactics to keep game developers away from industry-standard API features that it doesn't support, and which rival AMD does. NVIDIA drivers tell Windows that its GPUs support DirectX 12 feature-level 12_1. We wonder how much of that support is faked at the driver-level, like async compute. The company is already drawing flack for using borderline anti-competitive practices with GameWorks, which effectively creates a walled garden of visual effects that only users of NVIDIA hardware can experience for the same $59 everyone spends on a particular game.

AMD might be hurting, even on their last leg, but they manage to come out with this heavy handed, negative, attack campaign.  They are definitely behind this.  The whole thing......
starting with the GCN Asynchronous article at anandtech.  Then they get with their favorite lowly developer and work up some benchmark.  It is so incredibly obvious they rushed this thing out so fast and there is really no other reason why.  I mean, come on.....who sends pre beta benchmarks to the press and urges them the publish the results?  An alpha stage benchmark, sent to multiple press outlets and results being used in a smear campaign.  Clearly their is a real and true motivation behind all this.

Who is even interested in the game that this benchmark is supposed to be based on?  The game that is supposed to be coming out yrs from now?  This was all rushed out in some master scheme.  Really, if you cant see the effort behind this, you just arent looking.

This alpha benchmark sent to the press first finally got released for public download.  This came after the fact. It was much more important to spread the negative word than for people to actually check out the work the developer should want to showcase. Has any of you guys even looked at Ashes?  I mean, impressive?  Anything but.  Its not anything special at all.  

The DX 12 games i will play will be very very different than this.  I wouldnt expect the ones that start up with an AMD logo to perform as well, of course AMD GPUs will get better results than Nvidia....at least an launch.

Nvidias architecture is different than GCN. They are not the same.  I cant wait till nvidia responds, I full expect them to.  But the thing is, nvidia has tried to take the high road as of late when it comes to these AMD smear campaigns.  Certainly Nvidia can gain performance in this benchmark but it appears this developer may be more interested in drama than anything else.  It makes sense to me that a lower level API would and could easily favor one HW over another.  GCN and maxwell are different, so it is not so strange that their would be issues.  That nvidia would want to do things differently.  

I really cant wait to see how Nvidia responds to this.  The star swarm busted right in their faces but this time with DX12 it may take more than a driver.  But i really doubt this developer would be willing to let nvidia have specific code in their benchmark.  But, i dont imagine nvidia will take this sitting down.

Wow, just wow.  That's why I said a grain of salt when Oxide admitted that they are in a marketing agreement with AMD.  That says quite a lot.  I'm just hoping NV will respond with something substantial about this real soon - if they didn't already?  NV should just go ahead and put a stop to whatever momentum the publicity stunt this has gained for AMD, rather than remain silent and not even bother to publicly deny Oxide's claims.

[BenSkywalker]

This talk about nVidia hardware lacking support for async compute is truly disturbing. It must be the absolute truth or nVidia would have done something to refute it by now. At this point, the only thing that I can think of that would give any credibility to their stance would be for them to travel back in time, several years, and build something into their GPU that utilized async compute in a manner that would truly stand out to the end users, given the API limitations of the time this would be tough, maybe they could do.... I don't know..... how about physics. They could come up with some dumb marketing name, we'll call it PhysX for the sake of argument- and it would use async compute to do things like generate large amounts of particles, handle things like cloth and fog.... stuff like that. They could do this on older DX11/10/9 games and add physics based calculations that ran concurrently with the graphics engine on the same GPU to prove that their parts were built from the ground up, for many years, for async compute.

Of course, something like that is way outside of the realm of possibility.

[ocre]

Here is my take.......

Maxwell and GCN are two different designs. AMD has HW ACEs that could be exploited for more performance in workloads that utilize async. I believe that there is a reason they exist, to utilize more of the HW and get more done. There was a problem that the ACE is designed to address.

Nvidia has been trying to get more performance from emir designs as well. They have been trying to solve the same problems. Except, they went a totally different route.

All you have to do is look at the paper specs of any maxwell card vs the paper specs of the cards from AMD they compete with. Nvidia seems to get a lot more performance out of their theoretical than AMD. Obviously, they are doing something that works very very well.

The ACE is an AMD thing, nvidia doesn't have them. But here is another thing to consider, the way async works. The more gaps you have, the better potential performance boost from async functions. Nvidia has been working to minimize gaps and wasted cycles. There just may not be a lot of potential anyway.

Async is being used in a propaganda campaign right now. Of course if a developer builds a special and specific engine, you could see AMD have great performance. But there just isn't much to go on right now. Async in theory sounds great but we have no examples. We need to see real games and results

[SickBeast]

This wouldn't be the first time that AMD/ATi worked closely with Microsoft on the newest DirectX standard. Let us not forget DX9 and their utter domination of nVidia at that time for years.

I think that there may be something to all of this. We could see the GCN cards massively outperform the comparable nVidia cards like the GTX 680 and 780.

[ocre]

async and ACE are two different things. Nvidia may not have HW ACEs in maxwell but this doesn't mean that they can't do async compute.

It's also silly to assume that this is some super all important feature for dx12. This is a feature that may help get extra performance out of GCN. AMD has been talking it up and it could be way more useful to their architecture.

I don't think nvidia has put a lot of time or emphasis on this because they already get high utilization out of their architectures. There may not be the opportunity to gain on maxwell but this may not be the bad thing that people are trying so hard to make it out to be. Nvidia seems to be able to get a lot more performance out of their architectures than AMD, just look at the theoretical performance figures. At the end of the day, it is all about how well you can utilize your HW, that is all that ultimately matters. And so far, nvidia has done this very well.

We will see if async is this magical ability that makes GCN amassing but its always wait and see with AMD......and it seldom lives up to the hype. Why do you believe this will be any different? Cause I am a believer when I see it. And this small time developer with close ties to AMD, their ashes benchmark was once used by AMD to show off mantel. Look up the YouTube video. This thing has been converted to dx12 but no wonders.......it runs great on GCN. This developer has a marketing agreement with AMD. This should be all you need to know.

We will see how much asyn helps GCN when real games come out

[googoo24]

Meh.......By the time Dx12 games are prevalent, most of these cards will be ridiculously outdated anyway. IMHO, AMD' days of domination are over.

I think that there may be something to all of this. We could see the GCN cards massively outperform the comparable nVidia cards like the GTX 680 and 780.

I think I recall benchmark showing one of these cards performing well. I can't remember.

[ocre]

AMD has done a great job over the years with GCN. One can only commend them

[BenSkywalker]

MD has HW ACEs that could be exploited for more performance in workloads that utilize async. I believe that there is a reason they exist, to utilize more of the HW and get more done.

They exist because the Jaguar CPUs are a fucking joke compared to Cell. Being quite serious.

I think that there may be something to all of this.

Think the GameWorks rants are bad now, just wait.

Let us not forget DX9 and their utter domination of nVidia at that time for years.

Well, not exactly-

http://www.anandtech.com/show/1293/11

DX9 launched in December of '02. Two things about the above review, it was a bit over a year from DX9 launch, and nVidia was sodomizing AMD at that point- even if we look at the prior generation part from nV, it is still neck and neck with the top dog from AMD at the time. I know people like to remember things very differently because the 5800 was notorious, but in terms of actual performance, ATi being competitive at all was so shocking, it slants people's recollection of how it actually was at the time.

async and ACE are two different things.

Yes, but the viral marketing team from AMD has been given their orders, and they do as they are told. nVidia has been doing async computing on their GPUs for several years now, and in far more elaborate situations than what we are talking about with these benchmarks.

DX12 is a low level API. If you write close to the metal for one GPU architecture, it is going to perform *horribly* on other architectures. That is inherent to low level APIs. One company has ~80% market share and is widely known for their amazing dev relations. One company is AMD. Like I said, you think the bitching about GameWorks is bad now, wait a little while.

[ocre]

And only a fanboy or red team shills can talk about this while ignoring that the ashes benchmark was developed by a company that AMD is invested in. That this benchmark was first used in an AMD promotion of mantle and then now it's ported to DX12 and being used specifically in an AMD campaign.

The story plots out as so

1) First time we see this demo, it's in a promotional video with Huddy "interviewing" in a staged event to showcase mantle.

2) AMD sponsors ACE articles to large marketing tech sites.

3) this ashes benchmark is sent to tech sites, conveniently enough it shows very promising gains for GCN.....kind of just like it was used to show off mantle many months prior.

AMD has been using this demo for some time. It's purpose has been to try to showcase their HW and capabilities. This is not an engine like any other. No games exist that use it

[SteelCrysis]

http://www.extremetech.com/gaming/214834...-benchmark

The results for these two cards show the Fury X with a marked advantage at 720p, but no one buys a GPU like this to play at a resolution that low. The 1080p results are the best balance between image quality and playability, and they show the Fury X and GTX 980 Ti tied in both cases — the Fury X may lead by 2% at 1080p and the 980 Ti by 3% in 4K, but the amount of variation between runs makes this an equivalent showing.

For readers interested in seeing how other cards compare, we’ve taken the step of including AMD’s results from its own testing. Nvidia did not provide us with an equivalent data set. Note that while our benchmark figures are actually lower than AMD’s, the performance we see between the Fury X and the GTX 980 Ti mirrors their own.




If other analysis confirms these figures, it implies that AMD’s GCN-based cards could have real staying power in the DX12 API. For the past few months, there’s been a great deal of chatter back and forth as users debated whether or not DX12 feature levels were likely to deliver significant performance advantages or disadvantages. These types of questions are always best answered by games, and the (very) early DX12 data is looking very solid for Team Red. The fact that the R9 390, at $329, outperforms a GTX 980 with an MSRP of ~$460, is a very nice feather in AMD’s cap.

As with Ashes of the Singularity, the usual caveats apply. These are pre-launch titles and early drivers on a still-young operating system. So far, however, the DX12 results we’ve seen have been very positive for AMD — lending credence to the company’s longstanding argument that GCN would fare well under the new API.

[ocre]

The results do favor GCN. There have been a few games over the yrs that tend to. I dont think nvidia is in trouble or anything. They seriously could afford to loose some market share. It would actually be better for us all. But, honestly..........if this really is the trend, nvidia doesnt have to do much but lower prices on the gm204s and such and all is well. The 980ti competes fine at its current price point.

But this is a canned benchmark, is it not? We really need to see what the DX12 games will be like. The first few are supposed to be AMD (evolved) titles. So there might be a wave of games that do well on GCN.

[SickBeast]

AMD needs this boost desperately from DX12.

What they really need, however, is an answer to Pascal. I really hope that they have a good 16nm card coming soon. I'm not holding my breath, though. AMD recently said that it cost them around $1 billion to develop Fiji. It's not like they have several billion dollars laying around. We will see what they do next.

[googoo24]

Strange. Other sites show the 980ti dominating in most Fable benches.

http://wccftech.com/directx-12-enabled-f...ics-cards/

Never mind...It appears to be within the margin of error. I think....

[RolloTheGreat]

This talk about nVidia hardware lacking support for async compute is truly disturbing. It must be the absolute truth or nVidia would have done something to refute it by now. At this point, the only thing that I can think of that would give any credibility to their stance would be for them to travel back in time, several years, and build something into their GPU that utilized async compute in a manner that would truly stand out to the end users, given the API limitations of the time this would be tough, maybe they could do.... I don't know..... how about physics. They could come up with some dumb marketing name, we'll call it PhysX for the sake of argument- and it would use async compute to do things like generate large amounts of particles, handle things like cloth and fog.... stuff like that. They could do this on older DX11/10/9 games and add physics based calculations that ran concurrently with the graphics engine on the same GPU to prove that their parts were built from the ground up, for many years, for async compute.

Of course, something like that is way outside of the realm of possibility.

I remain undisturbed.

Last I checked, my 980Ti is a faster card than a Fury X pretty much across the board. When DX12 games matter, I'll have a Pascal chip that will likely be faster than Fury X2.

Back in the day people yelled a lot about how DX9 was faster on the ATi cards, which was true.

I think at the time the 6800 series launched and beat ATi at DX9 there were around three games that used DX9 and they weren't exactly the big sellers. (Dirt car racing and one of the Tomb Raiders?)

AMD is doing what AMD always does- trying to create fear that future games will favor their hardware. (see 8m Advocate threads saying "When future games are coded properly, you will need moar corez and a Faildozer!")



The one hope they have this time is a MS (a company that matters) is probably pushing DX12 hard for XBone. AMD graphics can only influence devs who work in their basement with their 18% discrete market share.

[BoFox]

This talk about nVidia hardware lacking support for async compute is truly disturbing. It must be the absolute truth or nVidia would have done something to refute it by now. At this point, the only thing that I can think of that would give any credibility to their stance would be for them to travel back in time, several years, and build something into their GPU that utilized async compute in a manner that would truly stand out to the end users, given the API limitations of the time this would be tough, maybe they could do.... I don't know..... how about physics. They could come up with some dumb marketing name, we'll call it PhysX for the sake of argument- and it would use async compute to do things like generate large amounts of particles, handle things like cloth and fog.... stuff like that. They could do this on older DX11/10/9 games and add physics based calculations that ran concurrently with the graphics engine on the same GPU to prove that their parts were built from the ground up, for many years, for async compute.

Of course, something like that is way outside of the realm of possibility.

I remain undisturbed.

Last I checked, my 980Ti is a faster card than a Fury X pretty much across the board. When DX12 games matter, I'll have a Pascal chip that will likely be faster than Fury X2.

Back in the day people yelled a lot about how DX9 was faster on the ATi cards, which was true.

I think at the time the 6800 series launched and beat ATi at DX9 there were around three games that used DX9 and they weren't exactly the big sellers. (Dirt car racing  and one of the Tomb Raiders?)

AMD is doing what AMD always does- trying to create fear that future games will favor their hardware. (see 8m Advocate threads saying "When future games are coded properly, you will need moar corez and a Faildozer!")



The one hope they have this time is a MS (a company that matters) is probably pushing DX12 hard for XBone. AMD graphics can only influence devs who work in their basement with their 18% discrete market share.

Yo wassup - glad you still kicking nuts around here! 

Yeah, and what was all of this nonsense talk about NV having their Fermi GPUs ready for DX12?  By then, I'd be giving away Fermi cards for free! 

I expect AMD to beat NV to the punch with 16nm FINFET (as usual, once again, like with being first to 90nm, 80nm, 55nm, 40nm, and 28nm).  Hopefully it won't be like Intel's Skylake which is completely unimpressive with hardly any power savings over Haswell at the same 4.4GHz clock (14nm vs 22nm), but at least 16nm is a full-node shrink from 28nm. 

BTW, it was DX9.0c that gave NV an edge with the 6800 Ultra, when AMD was still limited to DX9.0b.  Games like Bioshock that came out in 2007 required DX9.0c as a minimum, and was rather unplayable with the X800XT.  Still, I absolutely loved my X800XT AIW!!!!  It was great with Far Cry, one of the first DX9 games ever - released in like 2003 or 2004 (and perhaps the most impressive one until Crysis came out).  What really helped NV with its 6800GT/Ultra cards was the SLI feature introduced with PCI-E, with Doom3 as the killer app as a single 6800Ultra blew the X800XT out of the water.  Then with 7800GTX SLI, I started playing everything in 3D, and from that point on, NV was the leader for me with gaming experience, with superior 3D driver support.  Good days.  Now, boring days until 16nm.

[RolloTheGreat]

I think you give AMD way too much credit these days. They're just stretched too thin, too broke.

Look at last two launches:

290X they slap some poor dick broke dust buster on it and reincarnate the FX5800 Ultra. You think they didn't KNOW that fan was as shitty as it gets?

You can just imagine the meeting:

Engineers: Boss, the 290X is a REALLY strong chip! If we put a good fan on it, the press will treat us like the second coming! NVIDiA high end sales will CEASE!

Mgmt: We have a TON of 7970 fans left. Use those, and make a "dual mode" so the schmucks think we're doing it to help them overclock!

Fast forward two years and we get a tweaked Tonga with a new kind of memory and a cheap pump.

Poverty = death in chip design industry

[BenSkywalker]

The one hope they have this time is a MS (a company that matters) is probably pushing DX12 hard for XBone.

Amusingly, this isn't going to help them nearly as much as they like to think. The XBone has 3 ACEs, Sony made AMD add additional units for their hardware because the CPUs were so stupidly weak they couldn't come close to matching the PS3's CPU. Games that push ACEs hard on the XBone, are going to be using less than half the available resources on the PC hardware.

I remain undisturbed.

Was I too subtle in my facetiousness?

Fermi supports 16 async threads-

http://on-demand.gputechconf.com/gtc-exp...ebinar.pdf

Their focus group is downright painful to read, they have no fucking clue what they are talking about in terms of any of the technology. I swear, they get a memo written by a guy at AMD who also doesn't understand what the fuck he is talking about and then manage to mangle it even further.

[RolloTheGreat]

The one hope they have this time is a MS (a company that matters) is probably pushing DX12 hard for XBone.

Amusingly, this isn't going to help them nearly as much as they like to think. The XBone has 3 ACEs, Sony made AMD add additional units for their hardware because the CPUs were so stupidly weak they couldn't come close to matching the PS3's CPU. Games that push ACEs hard on the XBone, are going to be using less than half the available resources on the PC hardware.

Feeble AMD APU is feeble? I'm inclined to agree with you that DX12 won't bring parity with PS4. Somewhere here I posted that my thought is what it might do for XBone is add some games to the 1080p list that wouldn't have been and bring it closer to PS4 on that metric.

I remain undisturbed.

Was I too subtle in my facetiousness?

Fermi supports 16 async threads-

http://on-demand.gputechconf.com/gtc-exp...ebinar.pdf

Their focus group is downright painful to read, they have no fucking clue what they are talking about in terms of any of the technology. I swear, they get a memo written by a guy at AMD who also doesn't understand what the fuck he is talking about and then manage to mangle it even further.

When I was in NVIDIA's focus group they'd ask me for ideas on who to recruit as people would come and go, seemed like they wanted to have 5-7 members. Then we'd chat about candidates and whether they seemed knowledgeable and able to put a professional image of the company forward. (you may remember I tried to get you in back in the day, but you declined)

The AMD path seems to be "Let's recruit as many dumbasses as we can with posting contests. We'll award them points for the number of times they post "Arf arf arf! I love AMD and you should too!" or "Arf arf arf! NVIDIA is the devil!" and then give them games and hardware based on the number of posts.".

With us it was always,"Here's our new stuff. Use it, tell us if you find any bugs or things that aren't user friendly and post about your experiences online, good or bad.". When you're using what is usually the highest end stuff in existence, you don't usually have too much negative to say, because what you have is better than anything else or at least as good usually. So they got free advertising and QA out of the deal, because it's pretty easy to get excited when you're building quad SLi rigs and seeing you can do things like ginormous AA.

[BoFox]

The one hope they have this time is a MS (a company that matters) is probably pushing DX12 hard for XBone.

Amusingly, this isn't going to help them nearly as much as they like to think. The XBone has 3 ACEs, Sony made AMD add additional units for their hardware because the CPUs were so stupidly weak they couldn't come close to matching the PS3's CPU. Games that push ACEs hard on the XBone, are going to be using less than half the available resources on the PC hardware.

Feeble AMD APU is feeble? I'm inclined to agree with you that DX12 won't bring parity with PS4. Somewhere here I posted that my thought is what it might do for XBone is add some games to the 1080p list that wouldn't have been and bring it closer to PS4 on that metric.

Some more games to the 1080p list, when the PS4 already does many games at 900p or even 720p?!?? 

Maybe just 10% faster with DX12 (or 10% more games running at 1080p), YAYAYAYAYA!!!!

XBone doesn't even have more than 1 good Kinect game (while Xbox360 has a few).  PS4 doesn't even have the Move controllers (I absolutely love them for my Sharpshooters - it's my favorite part of the PS3..  about 20 great games that work with the move controllers). 

This is by far the most boring, unexciting console generation since Sega Genesis vs Super Nintendo when everything else was fail.