DirectX 12: The future of it within the console gaming space (specifically the XB1)

Yeah, I guess that was the point of it.

 

That 20 GFlops was from this PC Perspectives live tweet

 

This is not true, all the XB1 kits had AMD GPUs.

And yeah, it turned out to not be a huge deal after all, especially considering the competition already has lower level access to the hardware.

I do wonder though, how many PS4 devs will start using libGCM over just PSGL.

 

I'm not fully sure what you are alluding to here:

Are you implying that X1 haven't been using their move engines and it will only now be used in DX12?

Or that Move engines are part of the DX12 hardware spec?

 

http://games.yahoo.com/news/xbox-one-see-improved-performance-184529517.html

Unless this is an outright lie, then I don't see how anyone can say that there is no difference between DX12 and the DX API on the Xbox One right now. I'm not expecting any miracle gains. I imagine improvements would be in edge cases, but even small gains in the single digit percentages would be appreciate by developers. And if any of it makes writing code easier, that will go a long way as well. Ease of cross-platform development between PC and Xbox One could also be a big win. It'll be nice to read up as details come out. I'm guessing there's a lot of NDAs blocking talk about the Xbox One API.

 

The improved tools, if they aren't available and in use yet, should be a nice improvement over whatever devs are using currently. After all, anything that makes the jobs of the developers potentially easier is a good thing, right? Also, I didn't get a sense that Microsoft were trying to somehow mislead (don't think anyone is suggesting that, just a general statement) with regards to what is perhaps negligible to the Xbox One, and what will end up arriving on Xbox One later and help performance, whatever that gain might end up being.

They, for example, acknowledge that some of the new things DX12 is doing, are already present on the Xbox One, such as bundles and of course there's already the very tiny resource overhead for D3D. I don't know if the changes in rasterization to improve culling and collision are anything that will help on the Xbox One, or if any of the other new rendering features they touched on barely will benefit the system in any fashion. There's also apparently some improvements to resource and memory management.that, again, may or may not already be happening on the Xbox One. But if Microsoft weren't somehow entirely ahead of the game and already had most of this ready by the time the system launched, then some of these things should realistically benefit the system along with the improved tools.

Now, surely someone else is equipped to answer this a lot better than I am, but what about the way they're spreading their CPU workloads more evenly across the CPU cores, such as things like the User Mode Driver and just, in general, reducing the amount of work that needs to be done by the main CPU thread by more evenly spreading the work across the different cores, in addition to reducing the amount of work required overall? I don't really know, just guessing here, so maybe the level to which work is being spread across the cpu threads as they demonstrated in that slide today is already occurring on the system, as that's already one of the main advantages that consoles have had, but I don't know enough to know if this is exactly how it's been happening, you know?

At any rate, it's clear that not all DX12 improvements are readily available for use on the Xbox One just yet, as they made a clear distinction today between features already present on Xbox One, and ones that will ultimately be made available to developers.

 

No and no but I think there are some improvements.

Phil Spencer:

http://www.onlysp.com/microsoft-has...-theyre-having-to-move-stuff-out-of-the-show/

 

If true, I can only conclude the API on XB1 is fairly unfinished, or inefficient. Or very much 'Direct X' to allow for porting/compatibility. We have been told that XB1's API doesn't allow devs to make as much use of the hardware as PS4's. Maybe we are looking at a fairly thick, slow system level that's holding XB1 back, and DX12 will release that? That'd tie in with what appears to be underperforming hardware relative to the competition. Maybe it'll take DX12 next year to fully release XB1? Quite a lot of business implications if that's how it pans out (not for this thread).

 

I seem to remember that some insiders/devs did suggest so last year.

 

Last year... There have been several updates since then each boosting performance sometimes as much as 10% according to dev sources.

How badly do people want us to believe Microsoft are incompetent?

Small gains may be made with optimizations sure but talk of it making up the deficit with the ps4 with a physical hardware advantage are surely well wide of the mark.

 

Better late than never.

I'll have to wait and see it before I decide to resume working on my 3D engine.
Stopped because current API went in my way and are massive piles of shit. (There I said it. )

 

It's weird how that has become a common ground for both 'teh fanbois' and 'haterz' with the former expecting magic with an SDK change and the latter expecting nothing to improve over time.

If DX12 is coming to XB1 in the future I would have to imagine it's more akin to better alignment between the capabilities of XB1 API and DX12 making porting easier from PC to XB1 (and vice versa). Of course in developing any API you are going to work out better ways of doing things today than you did several months/years ago so I can see that adding speed gains to the XB1 API but I have a hard time as seeing that as more than single digit % gains for the same code base. As MS works out the kinks I can see them offering better advice and guidelines on how to optimise data structures for best performance with ESRAM et al which may offer larger gains than 'simple' API tweaks.

Still not sure how relevant DX12 is to the XB1 though beyond the kind of cross pollination between APIs that people not under NDA are never going to get enough detail on to comment on authoritatively.

 

I can't seem to find a link, but the timing here does not line up well with the supposed mono driver which was released very close to launch date.

I assume that X1 API is just a pared down version of DX11. While DX12 is something else entirely - as suggested above 4 man months to port over just the renderer sounds rather long if X1 API is just DX12.

 

Only loco cycle I think, which was an indie game that was always on PC, so hardly surprising.

People at a show saw a loco cycle display labeled Xbox One and opened a cupboard to see it running on a GTX 780 PC, and took pics also. Given the nature of the game, the early state of Xbox One hardware at the time, and the irrelevance of what platform it runs on, not really surprising.

Of course that quickly became "all Xbox games are really running on 780's Microsoft is teh liar!!". But to answer your question, no, Xbox games never ran on Nvidia cards.

 

It's not about incompetance. It's about the choices of the software layer and how that's impacting games. Up to this point, everyone has been working on the assumption, founded in decades of consoles, that the abstraction of the hardware from the system API's was fairly low overhead and efficient. There was no apparent need for MS to use a full-fat Direct X when an API using the same calls but much more efficient, hardware specific implementations to eliminate the overhead. Now that we're told DX12 is all about efficiency and low overhead, and we're told it's going to speed up XB1, that suggests that XB1 has the same sort of high overhead as PC. Why is that? Is this true, and XB1 isn't as low-level as typical consoles to the point it notably impacts performance? Or is the talk of DX12 speeding up XB1 inaccurate, and the gains will be minimal as XB1 already has a fairly efficient software layer?

 

I don't think it's likely that the first option is true. Maybe if this was their first console and they were blagging it but this is their third.

There has been no talk of how this is going to improve the Xbox with any sort of conviction or statistics so I am not expecting miracles (apart from on my pc when they finally get round to releasing it)

I think with the performance increases we've been told about pre release and post release we can reasonably assume that it's not a high over head api any more even if it was on the first place (which is just add unlikely in my opinion)

Xbox one's involvement in this had been a smoke screen in my opinion to keep the pr going. We were assured to "watch this space" at gdc to find out the benefit to Xbox one by a Microsoft mouth piece on Twitter yet gdc rolls around and there was nothing.

They have been very specific about what's happening on the pc and very vague about Xbox.

This is all about bringing the pc api closer to the console much like mantel. Improvements will be made in the process but we are not talking
About drastic performance increases here.

The biggest boost will come when they release some of the system reservation.

 

Guessing the bolded is correct.

It would be sad if MS can only get X1 API to a PC API, even one designed for efficiency as DX12, level of efficiency.

There may be a few things they can gain on X1 with DX12 features is my guess...but I'm guessing the X1 API is already efficient/low level.

 

Or maybe DX12 on XBO will just bring ease of use for those that are not familiar with the XBO API.
DX12 on XBO could also just favor/allow even more a seamless shift from PC development to console development for indies and multiplat studios.

DX12 speeding up XBO could very well be a PR stunt devised to put XBO in a better light, which never hurts.

 

http://www.igameresponsibly.com/201...es-the-xbox-one-gpu-and-drops-in-mono-driver/

This was August 2, 2013

 

A couples of things leave me with strange feelings:
* First time line why wait that long? Nvidia has a neat demo running already.
* What i the XB1 running now so jumping to Directx 12 will actually improve performances? Certainly not a really low level API.
* late 2015, the XB1 will turn into a PC more or less from developers pov.

Overall it doesn't compute, hardware companies have been knowing about Dx 12 for a long while it seems, with demo running now I would think that it could be launched fall 2014.
MSFT may also need that boost on its mobile devices. Why the wait, 2 years is a lot of time.

I've my idea, I think they have greater plan. It is not directx 12 that is going to take 2 years but their next OS. They have stated that they want to have a single OS across all their platforms, so I think that by fall 2015 they are going to release a new OS working on both X86 and ARM CPU.
I think they are to push it aggressively everywhere, along with their Live network.
I expect a great change by that time and a leveled played field wrt to was used to be PC, mobile, console gaming.