![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
The 1050ti should also be VR Ready. ASW (not asynchronous as far as I've seen) seems to be providing some interesting benefits. Designing a VR game around the capability doesn't seem like something to scoff at. Still wish someone would actually do a review on that with frametimes and IQ impressions. I know it's still early, but that feature seems somewhat fundamental to VR in the future. Especially more direct implementations, Doom being the only title that comes to mind with a similar approach. They already had the motion vectors and separate render targets to access occluded details.
I wouldn't quite say the product stack isn't there. Most of the hardware seems good, but the software needs some work. PSVR is arguably the best VR experience right now followed by Oculus. PSVR obviously takes advantage of async compute as do all the consoles. On the PC side we're only starting to see games with DX12/Vulkan and most of those have it tacked on. None of those were designed with VR in mind, save for maybe Doom which is still experimental.
Zen already seems successful based on one significant feature. The encrypted memory which is somewhat indispensable for the VM market. Even if slower than Intel that feature will be a win for some time. Even if Vega sucks for gaming it will be significant to the enterprise markets alongside HPC Zen. The SSGs with Vega also seem like they will have good enterprise results.
I could see Microsoft as a possibility in an attempt to get back into the IoT market. Get APUs into TVs running Windows. Apple maybe, but not sure they have enough of a processor market to warrant it. Samsung would seem a strong possibility. They'd have TVs that could benefit from APUs and that FPGA board. AMD also has that new Dracarys board and Samsung certainly likes things that breathe fire.
The current implementation yes, but we haven't seen the effect directly integrated with a LLAPI. Nor have we seen anyone perform an in depth writeup on the feature. Theory being AMD, if using async, should actually have less input lag as they could push an asynchronous frame faster. The effect could also affect performance differently.
For some markets CPU performance may not be as critical as security. Considering AMD's current share in the server market, any progress would be significant. I know someone whose company would definitely use the stuff I'll ask. He's the head of product development so would definitely know, assuming he could answer it.
They've attempted it in the past. It would be a rather bold move, but I could see them attempt to go vertical to gain traction. I'm still doubtful they'd do anything, but if they really try to push Cortana or Surfaces maybe.
Microsoft would lose the integration since buying RTG would only get you 1/2 of the tech. This would be detrimental to Microsoft Surface Book future. They can't buy AMD unless they don't mind losing the x86 license meaning it would be worthless to buy. They need the whole APU which Intel already provides, if Zen and Vega in an APU is better Microsoft would just shift gears and maybe even work the next Xbox and Surface together (controller, play, VR/AR etc.)
.(I can't tell if she still is), Layla Mah, and Yuval Boger. She talks first about APIs, including AMD and Nvidia, and then a bunch of other VR technical stuff...im not that smart. :\
As long as that hardware is superior to others - it's not - plus they are able to maintain that superior position makes the investment poor since already it is barely breaking even. Plus it would only be the GPU and not CPU. Too costly to separate the two for something like the console.
I can't see any reason to split RTG out of AMD atm. The x86 license is tricky, but it ultimately would need to be transferred. Intel simply can't stand as the only license holder without anti-trust issues.
That's hardly the market I'd see them going after. Sure PC and discrete graphics are nice, but like you said the market is dwindling and there are far larger pieces of emerging pie to pursue. My move would be pushing XBs and embedded windows cable boxes using the AMD hardware to provide VR and home entertainment. Do battle with the FireTVs and Chromecasts. Not to mention the home assistant market that is currently booming. Gaming is great, but there is a lot more to the entertainment sector. Unseating Android and iOS in mobile is unlikely, the home media market is another story though. It's also a crapload of devices.
There's a lot more to VR than gaming. I'd expect video streaming to be a more significant revenue stream to chase.Only thing RTG has going for it is a future expected market and IP as in VR expectations in the next 10 years with bloated numbers and a design that will win. AMD making a VR headset and show casing it and a few other VR projects that will probably net them zero but it can be used to persuade potential investors. Of course when Kyle demonstrates a total lack of performance out of RTGs VR stuff (being goody two shoes in Raja's eyes but telling the truth to his readers) can make some weird vibrations emanating from AMD
lol, I agree. Now if behind the scenes Intel is also involved for the x86 licensing making it viable for purchase then that could up the chances. I just don't see it happening and basically AMD is more on their own with joint ventures to tide the weather over. If Zen is good -> AMD will probably fare well, if not then I expect RTG to be sold off or attempted to be since AMD would not be able to put enough funds to keep it viable -> worth would steadily go down.
If AMD sells RTG - it could be AMD and Nvidia. There would be no reason for AMD or Nvidia to compete against each other. For example an AMD/Nvidia APU. Intel and Nvidia are not the best of pals in the end. How about Nvidia chipsets for AMD cpu's? There is potential
Oh ye of little faith in AMD. Zen will be fine and could be quite good for performance/$. Also Vega may pack a punch as well but don't assume one way or another. Not as if 2-3 months is very long to see how this shapes up.
Wouldn't have to be remote hacks. Corporate espionage and state actors would need only insert a USB stick. One bad tech and your system gets owned. There are also forms of non-volatile memory showing up.
In which case this memory encryption is still useless, because physical access is present.
Is it even confirmed to be in Zen? But ye, the idiotic hype was easy to detect. And Intel products already support it in Skylake CPUs.
