NVIDIA Tegra Architecture

Screen_Shot_2014-06-25_at_12.58.53_PM.0_cinema_960.0.png
Two Ls! :runaway:
 
I can't tell for sure, but it looks like it was rendered at < 720p and upscaled.

Youtube H.264 super-duper-blur (TM) ?

There is an article on either Arstechnica or Engadget that Google also "invented" its own low-level Graphics API (like Mantle, Metal, DX12).
Is there any proof of that ?
 
Last edited by a moderator:
One have to wonder how much a single gimped Kepler SMX can achieve in balancing the performance and visual fidelity. Render resolution (and battery drain) aside, I bet most of the assets in that demo were carefully pre-baked and sized for the occasion. I want to see how this platform would fare in a more open and dynamic 3D environment.
 
Youtube H.264 super-duper-blur (TM) ?

There is an article on either Arstechnica or Engadget that Google also "invented" its own low-level Graphics API (like Mantle, Metal, DX12).
Is there any proof of that ?
I don't think it's YouTube. Looks like sub-HD and a fairly annoying chromatic aberration pass despite not really being shot through a lens. Nice art, lighting (even though it's mostly baked) and a stable framerate, which are all things for me to point my demo team at, to achieve in their work :cool:
 
I don't think it's YouTube. Looks like sub-HD and a fairly annoying chromatic aberration pass despite not really being shot through a lens. Nice art, lighting (even though it's mostly baked) and a stable framerate, which are all things for me to point my demo team at, to achieve in their work :cool:

I'll tip my hat off if you manage to get your demo team to create as stunning visuals. The average viewer hardly cares how it's been realized and I honestly don't see why he should care either.
 
On that note, I think you'd similarly agree that if there was a 50% overhead due to language for GPU drivers it'd be a big deal.
Incidentally, driver overhead due to API design (say, OpenGL ES compared to Metal) is huge. It's an issue in certain circumstances.

I'm not saying optimize everything blindly here or that less efficient languages aren't fine for plenty of tasks, just that blanket statements about 50% performance not mattering make as much sense as saying that CPU perf dropping 50% doesn't matter. Good programmers will be able to determine when optimization is and isn't suitable. Thing is, a lot of programmers today don't have the faintest idea of performance implications of what they program and are neither aware of performance issues with the language they use nor how to optimize at all.
Yes, but I don't think anyone made such a blanket statement.

As for your claim about writing faster code with slower languages because it saves budget time that you can use on better optimization, I could see that being true if optimization with the "faster" languages would mean more platform specific branches. Otherwise I'm skeptical, especially if it comes down to Java vs C++.
I don't see why that would be limited to cases where you need platform specific branches.
 
Incidentally, driver overhead due to API design (say, OpenGL ES compared to Metal) is huge. It's an issue in certain circumstances.

Yeah, and that's getting a ton of attention right now :p

Yes, but I don't think anyone made such a blanket statement.

Well, the original statement was that it should apply to all but a narrow class of games. I think it's broader than that.

I don't see why that would be limited to cases where you need platform specific branches.

I get the idea behind what you're saying, that if you save development time using a slower but easier language you could use that saved time to better optimize your program's higher level design/algorithm choices. But I think in practice it'll rarely play out this way, that the people looking for quicker and easier development in an easier language are not going to tend to be the ones who will turn around and spend more time on optimization. And since we're specifically talking about Java here, I don't think Java saves you that much effort vs C++.
 
http://evleaks.at/2014/07/02/htc-vo...-final-8-9-1680x1050-ship-2560x1600-64gb-5mp/

I'd be perfectly fine with 2GB for it; it just doesn't sound like availability before early 2015 :cry:

This is the first rumour I see claiming a 16:10 screen. All others are mentioning the same 8.9" 4:3 screen as the ipad mini and xiaomi mipad.

I'm glad that HTC has finally landed a design win for a [strike]Nexus[/strike] Android Silver device. Despite a few bad decisions (camera, mostly), their hardware, execution and software updates have been excellent IMO.
Plus, they're back into profitability (largest profit since 2012, actually).
 
Back
Top