NTMBK
Lifer
- Nov 14, 2011
- 10,322
- 5,352
- 136
Not sure why you post this fallacy?
Try adressing the technical side of things, thx.
You put the "^^" smilie at the end of the sentence, universally recognised as demonstrating glee.
Not sure why you post this fallacy?
Try adressing the technical side of things, thx.
You put the "^^" smilie at the end of the sentence, universally recognised as demonstrating glee.
You call that "technical"?
You call that "technical"?
You got anything to say about low level vs high level API?
User space?
Performance?
BSOD's?
I always thought ^^ meant it peaked your interest. Not necessarily glee.
I always thought ^^ meant it peaked your interest. Not necessarily glee.
Even a canned demo of a mantle implementation to show performance/detail abilities would do wonders.
Tying the ship to EA was ... unfortunate.
Oxide Games AMD Mantle Presentation and Demo
http://www.youtube.com/watch?v=QIWyf8Hyjbg&feature=player_detailpage#t=1582
Oxide show off a game engine, running mantle and compair that with directx.
Maybe if AMD became a top "sponsor", but I would think that they would like to stick with vendor agnostic APIs for comparability reasons. There was more than a little moaning when they used PhysX for 3DMark 06 and Vantage and I doubt they'd want to go through all of that again.Do you think the next 3Dmark will support Mantle?
This so called game doesn't look impressive whatsoever. If AMD wanted to show mantle off, there are better ways to do it.
This so called game doesn't look impressive whatsoever. If AMD wanted to show mantle off, there are better ways to do it.
This so called game doesn't look impressive whatsoever. If AMD wanted to show mantle off, there are better ways to do it.
This so called game doesn't look impressive whatsoever. If AMD wanted to show mantle off, there are better ways to do it.
If you watch the demo, you see Mantle reaching 60-70k batches without having any issues, yet DX according to them starts to hit a wall at 15k batches.
So with DX you can have only about 1/4 of the objects on screen when compared to Mantle. Perhaps even less if they manage to get 100k batches without issues.
Of course we have to keep in mind that this is an alpha engine, a ton of work to do, but it gives a good example of how much more you can get out of your hardware with Mantle.
Also, keep in mind that not every game is done the same way, so we might not see the same performance gains in BF4 as we would in a RTS with thousands of units on screen at one time.
That wall at ~15K batches...that is with AMD's none-multithreaded DX11 driver...right?
So you don't really know?Considering the engine is designed to work well with multy-core CPU's with both DX and Mantle, I would say no.
And besides, how well does DX scale up with cores? Mantle worked with 12, probably even more.
So what then is going on that made Civ V so much faster for NVIDIA? Admittedly I had to press NVIDIA for this - performance practically doubled on high-end GPUs, which is unheard of. Until they told me what exactly they did, I wasn't convinced it was real or if they had come up with a really sweet cheat. It definitely wasn't a cheat.
If you recall from our articles, I keep pointing to how we seem to be CPU limited at the time. Now if you go back to the list of DX11 features Civ V uses, a light bulb should light up: multithreaded rendering. Civ V uses multi-threaded rendering, in fact it uses it quite extensively. Now why do we have multi-threaded rendering in the first place? Half of this is to better mesh with multi-threaded games by enabling additional threads to directly contribute without having to go through a master thread first. But a second purposes is because multi-threaded rendering helps the GPU just as much as it helps the CPU.
If you watch the demo, you see Mantle reaching 60-70k batches without having any issues, yet DX according to them starts to hit a wall at 15k batches.
So with DX you can have only about 1/4 of the objects on screen when compared to Mantle. Perhaps even less if they manage to get 100k batches without issues.
Of course we have to keep in mind that this is an alpha engine, a ton of work to do, but it gives a good example of how much more you can get out of your hardware with Mantle.
Also, keep in mind that not every game is done the same way, so we might not see the same performance gains in BF4 as we would in a RTS with thousands of units on screen at one time.
So now we are down to a subset of games on a subset of gpus? I would say though that if mantle brought a renaissance in RTS games, I would be very pleased with that, as they are few and far between these days, and it is one of my favorite genres.
In any case, as I have been saying all thru this thread, it most likely will take a long time to see whether mantle is a true revolution in games, or just a specialized set of instructions that go nowhere. We really have to see how the engine performs in a wide range of games (finalized games with all features enabled, not a demo to showcase mantle) and whether all the major engines adopt it.
I will admit that it is *possible* that the mantle patch for BF4 will be so spectacular that it will blow everyone away and make it obvious that mantle is the api of the future. OTOH it is also possible that it will show just moderate improvements, and we will have to wait to see the full effect.
Double the perfomance...just from making a driver that is multithreaded...
You say mantle will double the performance? Nice, but shame I lost my %40 increase in performance bet. Thanks for the heads up!