After the difficulties in navigating the Windows WDDM and realizing Direct-to-Rift mode this doesn't come as a surprise. When it was announced the other day I had a suspicion there would be targeted improvements on Windows 10 for this very purpose, so I'm glad to hear this is happening, and not just for Oculus, but for all VR development on Win10.
The Ubisoft Quebec team were responsible for the WiiU port, not the main development. It was primarily Ubi Montréal (in particular, one of the several teams working independently and in parallel under the 'Montréal' moniker), but with assistance from other Ubi teams from around the world.
There are a few devs from Valve whose work I've followed over the years, in particular: Tom Forsythe, Michael Abrash, Rich Geldreich, Bruce Dawson, Elan Ruskin. The first two now work at Oculus, one's gone solo with his own company, and the others at Google and Insomniac Games respectively.
They were all what I'd consider 'senior' guys (though being a 'junior' there is unlikely), so I look at their leaving as more than just your typical kind of ind...
Edit: Having re-read my "gpu count" paragraph I literally have no idea what I was trying to say there :P
*** There is a cost here, and it comes from having to queue up frames after each other whilst rendering is on-going on other system gpus, thus increasing the likelihood of latency introducing micro-stuttering issues due to frame-pacing inconsistencies, and still not lowering the time it takes a gpu to render a single frame. Now, with SFR used instead we can queue...
'Split Frame Rendering' (SFR) is not new by any means, and nor is the concept of hybrid/heterogenous rendering. You'll be able to find mention of it dating back years.
In any case, by "2x gpu means 2x performance" I assume you're referring to the article's "2-3 times faster than typical SLI or Crossfire AFR setups" sentence? I think they worded that somewhat poorly and is much clearer in AMD's slide:
"Frames no...
That's because modern multi-gpu usage relies on buffering more frames the more gpus are used - thus increasing latency - along with duplicating resources over each gpu for driver enabled alternate frame rendering or similar. However, with D3D12 developer's will be able to choose which resources and commands to use across different gpus, and so allow for more tailored solutions rather than merely using each additional gpu in near-complete isolation to each other.
The m...
"Lighting on this looks dull and flat, especially when compared to the storm trooper model shown on the E3 trailer"
It just looks like differences in image-based lighting, so it's hardly unexpected. Those materials, being PBR, still look the same though. If you've any doubts just lookup Marmoset's Toolbag software to see how big a difference it makes when ambient lighting is included through IBL.
edit: moved
It seems unlikely.
This is a synthetic test designed to measure only api performance. To best avoid other bottleneck locations affecting the result, the gpus really aren't doing much work in comparison to typical usage (eg. simple shaders, low resolution, minimal vram bandwidth usage, etc.). Even though I'm not certain why the api scores differ (especially when looking at the gtx 960 stock/oc'ed numbers), reaching that point again in games will be unlikely.
Go to the source itself on the GameDev.net forums:
http://www.gamedev.net/topi...
If you're interested in the subject you may want to read the whole thread in context too then.
It's just an update to the 'Enlighten', so of course it's going to do much of what we've seen in the past. With this release it's all about keeping consistency in dynamic indirect lighting with destruction and translucency, as was highlighted in the video, plus a few extras regarding materials support, optimization and tools.
All Frostbite games use Enlighten, and it's supported on Unity and Unreal Engine too.
Seeing as they want t...
That or 'Vulcan' from Greco-Roman mythology, just as 'Khronos' appears to be.
Aside from any independent game journo coverage, Gamespot will be streaming the awards and a few of the scheduled sessions:
http://www.gamespot.com/art...
Nvidia also have a live press conference on, but people aren't too sure why:
4120d ago 1 agree0 disagreeView comment
Yes, SCE are part of the board; however, that in no way detracts from the points above.
The Playstation APIs are developed and maintained independent of OpenGL. The development paths are different. SCE being in the consortium doesn't inherently mean it will have a direct affect on Playstation development, though as the intention is to improve the general state-of-affairs in the industry, indirect benefits are a reasonable enough assumption. That's about the only reaso...
Evidently some people are struggling to cope with my above comment.
Answer me this then: Why do people here think anything the Khronos Group does to OpenGL will affect the PS4, when even the PS3's libGCM API wasn't running it?
Just because higher-level interfaces are provided that try to mirror what OpenGL (or D3D11) does, that doesn't mean Khronos' OpenGL itself is being used. The PS3's PSGL may have been based on OpenGL ES, but it wasn&#...
@Nekroo91,
Are you sure you're not mixing up feature-set level with the API?
@Saigon,
The PS4 has two available APIs, neither of them OpenGL. There's a low-level one called 'GNM', and a higher-level one called 'GNMX'.
Granted, your point is factually valid where light transportation enters and leaves a front-facing surface underneath the same pixel. However, it's worth remembering this is about translucency, so light can enter back-facing surfaces, transmit through it, and then exit on a front-facing surface. In real-life this will occur with most dielectric surfaces, and so can drastically change the aesthetics/looks/accuracy of a scene's lighting in various situations. Look at plant leaves exampl...
One way or the other binaural audio computed from HRTFs will be needed if immersion in VR is the end goal, so headphones will be necessary. That being said, I can't see them limiting the option to just the exampled headphones though - I'd bet it's more a case of having some headphones available is better than none at all.
The lower the frequency between ticks the more likely you are to see integration errors and collision detection complications. It's all dependent on what we can get away with on the hardware whilst maintaining a suitable level of precision for what type of interactions we expect to occur and the quality of the modelling of those interactions.
In general the norm is around 60Hz, though some games may opt for 30Hz, say, where simulation accuracy allows. The same goes vice v...
Keep in mind that most games don't have nearly as many programmers on the job as there are artists, modellers, and sometimes designers.
In this particular instance porting the game should rely mostly on programmers, as the main task is re-purposing an existing code-base and assets to work with PC (should be abstractions in place, especially as Arkham Knight is built on a modified UE3), it shouldn't really require a large team size. Even when adding new features by way...