Techspot: AMD Polaris 10 performance will reportedly be on par with Radeon 390/390X

[Drunkenvalley]

It's more or less inconsequential, as "universal" APIs already exist. The only one trying to make it a platform war, as far as I know, is Oculus. And all that did was give the Vive a huge popularity boost.

[SodiumChloride]

It's more or less inconsequential, as "universal" APIs already exist. The only one trying to make it a platform war, as far as I know, is Oculus. And all that did was give the Vive a huge popularity boost.

What is it called?

AMD also has something called LiquidVR ... Of course Nvidia is unlikely to jump on board ... *cough*Mantle*cough*

[Drunkenvalley]

Etc

Firstly, I had SteamVR in mind. It's another API that combines several other VR functionalities into one package, like OpenVR. Basically, a package of APIs. And no doubt there are many other APIs out there, but really, who cares?

Secondly, the existence of several VR APIs does not mean they are competing APIs. The function of these APIs is not to compete with each other. No, the function is much simpler than that: Bringing VR HMDs into the fold. It's not for competition, but to try and have a general installed userbase.

Let's take LiquidVR: AMD doesn't care if an application/HMD is running LiquidVR, SteamVR or whatever. That's irrelevant to it. They don't care if LiquidVR is succesful in and of itself, they just want to have more people wanting VR, because that's a huge step up in hardware requirements, and drives sales of hardware. Similarly, engines want to implement VR support, and some have done so with their own API. Again, that's not to compete. Instead, it is rather that having VR is a selling point. They don't care if you're using Oculus, Vive or another HMD when that comes to market.

Third, newcomers have practical reasons to use existing standards. ...Do I really have to explain that one, though? Okay, let's do it anyway. If a new HMD comes out, they want to get support in existing works where possible, so they're going to try to meet existing standards and functionality. Otherwise, they're basically a shitty console launching without games; they have no support for existing titles, and they can't even afford to have titles develops specifically for them.

All in all, VR is not going to become some monster where game X is only compatible with VR HMD company A's equipment. Only Oculus can really hope to try to get away with it, and that's because they were first, but it's already

[SodiumChloride]

Firstly, I had SteamVR in mind. It's another API that combines several other VR functionalities into one package, like OpenVR. Basically, a package of APIs. And no doubt there are many other APIs out there, but really, who cares?

Secondly, the existence of several VR APIs does not mean they are competing APIs. The function of these APIs is not to compete with each other. No, the function is much simpler than that: Bringing VR HMDs into the fold. It's not for competition, but to try and have a general installed userbase.

Let's take LiquidVR: AMD doesn't care if an application/HMD is running LiquidVR, SteamVR or whatever. That's irrelevant to it. They don't care if LiquidVR is succesful in and of itself, they just want to have more people wanting VR, because that's a huge step up in hardware requirements, and drives sales of hardware. Similarly, engines want to implement VR support, and some have done so with their own API. Again, that's not to compete. Instead, it is rather that having VR is a selling point. They don't care if you're using Oculus, Vive or another HMD when that comes to market.

Third, newcomers have practical reasons to use existing standards. ...Do I really have to explain that one, though? Okay, let's do it anyway. If a new HMD comes out, they want to get support in existing works where possible, so they're going to try to meet existing standards and functionality. Otherwise, they're basically a shitty console launching without games; they have no support for existing titles, and they can't even afford to have titles develops specifically for them.

All in all, VR is not going to become some monster where game X is only compatible with VR HMD company A's equipment. Only Oculus can really hope to try to get away with it, and that's because they were first, but it's already

Problem is, which is THE standard?

That's the point of a standard. So EVERYONE (game developers, VR head set developers, GPU and CPU developers, ... etc.) can get on board and support it universally.

No one wants to deal with potential incompatibility issues.

[Drunkenvalley]

Problem is, which is THE standard?

That's the point of a standard. So EVERYONE (game developers, VR head set developers, GPU and CPU developers, ... etc.) can get on board and support it universally.

No one wants to deal with potential incompatibility issues.

No one wants to deal with them, which is why future HMD creators will look to existing products to match with, and existing tools, to avoid this exact issue. That doesn't require "THE" standard to exist, although it is likely to be SteamVR, as it includes a number of helpful tools for functionality that extends beyond simply the HMD itself.

[Lathais]

pretty sure mainstream in this context is a pricerange (by definition). Budget-mainstream-highend-enthusiast. What exactly those price ranges are can be debated but i'm pretty sure the whole concept has nothing to do with what actually sells the most.

Perhaps look up the definition of mainstream?

main·stream
ˈmānˌstrēm/
noun
noun: mainstream

1.
the ideas, attitudes, or activities that are regarded as normal or conventional; the dominant trend in opinion, fashion, or the arts.
"companies that are bringing computers to the mainstream of American life"
jazz that is neither traditional nor modern, based on the 1930s swing style and consisting especially of solo improvisation on chord sequences.
noun: mainstream jazz; plural noun: mainstream jazzes

adjective
adjective: mainstream

1.
belonging to or characteristic of the mainstream.
"mainstream politics"
synonyms: normal, conventional, ordinary, orthodox, conformist, accepted, established, recognized, common, usual, prevailing, popular
"the mainstream audience may not be ready for these graphic sculptures"
antonyms: fringe
(of a school or class) for students without special needs.
"children with minor handicaps would be able to attend mainstream schools"

verb
verb: mainstream; 3rd person present: mainstreams; past tense: mainstreamed; past participle: mainstreamed; gerund or present participle: mainstreaming

1.
bring (something) into the mainstream.
"vegetarianism has been mainstreamed"

The dominant trend is obviously the 970, therefore, the price of the 970 is the mainstream price. It's exactly why AMD said they are targetting minimum VR spec at $349, because it's the mainstream price. In turn, that makes me think that there is no way this is the 480/480X, because the prices for those should be similar to the 380/380X and that's far below the price AMD says they are targeting.

[Vash The Stampede]

PS VR is probably what will push VR into the mainstream - if it catches on.

On PC ... is there even a standardized API for VR yet?

Given the PS4 can provide the frame rates and detail. The reason why a 970/390 performance is needed is to have a good 60-90 fps with a resolution slightly higher than 1080P. Right now the PS4 can't play games at 1080p with 60fps. If you don't have this, you break immersion and the VR isn't doing its job.

Valve has the API thing covered with OpenVR. And when games are actually made for VR, (most likely Valve), then they'll likely use OpenVR.

OpenVR is an API and runtime that allows access to VR hardware from multiple
vendors without requiring that applications have specific knowledge of the
hardware they are targeting. This repository is an SDK that contains the API
and samples. The runtime is under SteamVR in Tools on Steam.

[Lathais]

Given the PS4 can provide the frame rates and detail. The reason why a 970/390 performance is needed is to have a good 60-90 fps with a resolution slightly higher than 1080P. Right now the PS4 can't play games at 1080p with 60fps. If you don't have this, you break immersion and the VR isn't doing its job.

Valve has the API thing covered with OpenVR. And when games are actually made for VR, (most likely Valve), then they'll likely use OpenVR.

I thought the PS4 VR was going to have it's own separate processor and whatnot so that it would be able to take some of the work from the PS4 itself so it could achieve the frames needed to make VR possible.

Edit: Well, it kinda does, but not really:
http://www.eurogamer.net/articles/di...essor-revealed

However, despite the active cooling, we shouldn't expect too much from the external processor - core graphics rendering for VR all takes place within the PlayStation 4 hardware itself and it's going to be down to developers to work within existing limitations to get the best results. But from our perspective, the idea of computationally expensive, latency-sensitive tasks being offloaded to the external box is a good one, and should produce a better overall experience. It certainly represents a sizeable investment from Sony - and we can't help but wonder what this may mean for the basic build cost, and by extension how much PlayStation VR will cost at retail.

It is offloading some tasks, but not the actual rendering itself. Still, it has an extra box with extra power to help out, since the PS4 itself is not really enough.

[Deleted]

Firstly, I had SteamVR in mind. It's another API that combines several other VR functionalities into one package, like OpenVR. Basically, a package of APIs. And no doubt there are many other APIs out there, but really, who cares?

Secondly, the existence of several VR APIs does not mean they are competing APIs. The function of these APIs is not to compete with each other. No, the function is much simpler than that: Bringing VR HMDs into the fold. It's not for competition, but to try and have a general installed userbase.

Let's take LiquidVR: AMD doesn't care if an application/HMD is running LiquidVR, SteamVR or whatever. That's irrelevant to it. They don't care if LiquidVR is succesful in and of itself, they just want to have more people wanting VR, because that's a huge step up in hardware requirements, and drives sales of hardware. Similarly, engines want to implement VR support, and some have done so with their own API. Again, that's not to compete. Instead, it is rather that having VR is a selling point. They don't care if you're using Oculus, Vive or another HMD when that comes to market.

Third, newcomers have practical reasons to use existing standards. ...Do I really have to explain that one, though? Okay, let's do it anyway. If a new HMD comes out, they want to get support in existing works where possible, so they're going to try to meet existing standards and functionality. Otherwise, they're basically a shitty console launching without games; they have no support for existing titles, and they can't even afford to have titles develops specifically for them.

All in all, VR is not going to become some monster where game X is only compatible with VR HMD company A's equipment. Only Oculus can really hope to try to get away with it, and that's because they were first, but it's already

In regards tot his particular detail, I strongly recommend that you look at the currently-ongoing Oracle vs. Google case. Which set horrific precedent in this area.

[Drunkenvalley]

In regards tot his particular detail, I strongly recommend that you look at the currently-ongoing Oracle vs. Google case. Which set horrific precedent in this area.

The Oracle vs Google case is mostly irrelevant to this conversation. These APIs exist for the purpose of selling VR. If someone comes along and basically copy-pastes LiquidVR, puts some of their own stuff in at the end, and repackages it and sells it as if they made it all? Yeah, that's a problem, but that's not really what I expect to happen in all this.

[Life-Binder]

[Remilia]

You're going to have to do better than that... like seriously a lot better than that. I've seen various kinds of rumors and leaks, but at least they were more amusing or some tiny shred credibility, but... this one is just dumb. A random copy of a NDA clause that anyone can make and sign is being used as fact for a date?

[Noctifer616]

You're going to have to do better than that... like seriously a lot better than that. I've seen various kinds of rumors and leaks, but at least they were more amusing or some tiny shred credibility, but... this one is just dumb. A random copy of a NDA clause that anyone can make and sign is being used as fact for a date?

Even if we assume the document is real, it doesn't say anything about any product so it wouldn't even have to be related to graphics cards.

- - - Updated - - -

Here is the source btw:

http://www.24liveblog.com/live/1331452

[Life-Binder]

You're going to have to do better than that... like seriously a lot better than that. I've seen various kinds of rumors and leaks, but at least they were more amusing or some tiny shred credibility, but... this one is just dumb. A random copy of a NDA clause that anyone can make and sign is being used as fact for a date?

no need to get your panties in a twist over me simply posting an image


and yes its from the Macau thing so it may well be related to Polaris

[Noctifer616]

no need to get your panties in a twist over me simply posting an image


and yes its from the Macau thing so it may well be related to Polaris

You should have posted the source with the image, not just he image.

[Life-Binder]

source doesnt change or add anything

[Noctifer616]

source doesnt change or add anything

Yes it does. The picture alone can be just a random document, the source means it's actually from the AMD event.

[Denpepe]

I'll leave this here as food for thought

http://www.hardocp.com/article/2016/...y#.V0hXNo9OI_U

[Lathais]

I'll leave this here as food for thought

http://www.hardocp.com/article/2016/...y#.V0hXNo9OI_U

They state right at the beginning, full disclosure, we were not invited to Macau. That means, they have no official info at all. Nothing. We do have official info, from seeing Polaris 10 running Hitman. Hmmm, who do I believe, site that is bitter they did not get invited or what was actually shown? I mean, that actually is kind of tough, because they did not show anyone the actual card itself, so they could be pulling an nVidia move and lieing about what they showed us. We are also not 100% sure about the settings used, as AMD says that no one saw that. A couple media outlets did claim they got a sneak peak of it running at Ultra, but AMD would neither confirm nor deny. So it's really tough to decide who to trust here. I normally don't trust bitter people though, lol.

On the other hand, if it really is the 480/480X, which is entirely possible though I hope it's not, all that article is really saying is that budget cards will not compete with higher end cards, and painting that in a really negative way. I mean, if AMD IS releasing the 480/480X, they are obviously not even TRYING to compete with the 1080. It's a different tier card, derp.

[Bigvizz]

I'll leave this here as food for thought

http://www.hardocp.com/article/2016/...y#.V0hXNo9OI_U

#Drypie < no will get this. lol

- - - Updated - - -

On the other hand, if it really is the 480/480X, which is entirely possible though I hope it's not, all that article is really saying is that budget cards will not compete with higher end cards, and painting that in a really negative way. I mean, if AMD IS releasing the 480/480X, they are obviously not even TRYING to compete with the 1080. It's a different tier card, derp.

If it is then we have another AMD not wanting to compete and just aiming for the middle price market.
Which did so well on the CPU front. /sarcasm