I dont buy this Pascal is a facelifted shrunk Maxwell until i see it. What is the source for that?
There is, Nvidia can use hardware accelerated conservative rasterization to their advantage presuming that Polaris doesn't implement the same thing or further improve their geometry shaders in the pass-through case and in vertex data expansion too ...
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
Intel is irrelevant with the pathetically low primitive rate that they push their iGPUs out ...
If people think tessellation is bad on AMD wait till they see how much more poorly it scales on Intel GPUs ...
Fixed function units isn't exactly Intel's strength ...
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
The support is adequate IMO ...
Sure Nvidia doesn't support underestimate conservative rasterization or post-snap degenerate triangles not been culled or 1/256 uncertainty regions but the basic idea is there plus there's tons of applications out there that will work with tier 1 support ...
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
It would be better if you stopped acting like it's some endgame move. It's more a political hey we too.
FYI, I don't have any favouritism towards Nvidia ...
You can't deny that it's an advantage that Nvidia can use ...
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
Ok so you're not biased but you sure have a hard one for it. Again where can we see this advantage?
moonbogg
Lifer
- Jan 8, 2011
- 10,731
- 3,440
- 136
People wondering if Maxwell will fall off a cliff like Kepler did? Didn't I see some benchmarks floating around with a 390x beating a 980ti? Pretty sure I saw that somewhere.
I wouldn't be surprised if Nvidia refused to use Async simply as a matter of pride because AMD did it first. They would rather run their entire company into the ground than admit AMD made a better move and admit that their backs are now against the wall, the room is burning RED all around them and the only way to escape is through that emergency door with a neon sign on it that says "ASYNC COMPUTE THIS WAY BISTCHE".
I wouldn't be surprised if Nvidia refused to use Async simply as a matter of pride because AMD did it first. They would rather run their entire company into the ground than admit AMD made a better move and admit that their backs are now against the wall, the room is burning RED all around them and the only way to escape is through that emergency door with a neon sign on it that says "ASYNC COMPUTE THIS WAY BISTCHE".
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
Ugh, you keep mentioning it as some great advantage but the truth is you can't back that up. Why? Because it doesn't exist yet. CR has potential to be something or not.
you're a bit aggressive there man, buzzkiller is just pointing out that CR could be what nvidia uses to exploit and maintain some kind of win.
think gameworks libraries with only cr support or slow cpu emulation.
There is always DX12 mGPU support that will allow a person to throw in a card from the other guys to take advantage of it's additional benefits. I see no reason why AMD would sabotage it, history shows NVidia doesn't really like the idea. Would seem in todays gaming it's one of the biggest things DX12 will bring....As long as developers are on the boat of course.
poofyhairguy
Lifer
- Nov 20, 2005
- 14,612
- 318
- 126
But where is that pro implemented?
AMD has Ashes to point to as an example of Async. Who has implemented what Nvidia only supports?
ShintaiDK
Lifer
- Apr 22, 2012
- 20,378
- 146
- 106
Do you really believe in a broad mGPU support? I am starting to doubt if there will be more than a handful. You have to convince developers to use time and money on something very few will ever use.
In the perfect gaming world the answer would be yes.
Time will tell in the end no doubt.
It's extremely, extremely early in the DX12 lifespan... we don't even have a single native DX12 game in full release yet. Just early release and whatever Hitman is doing with that episodic thing and some DX9-11 games with DX12 bandaid patch applied. Once AotS is officially out of early release, then we'll finally have 1 single real, fully released native DX12 game. And even AotS has a foot in both worlds since it still can do DX11. Just like how we didn't see the true potential and power of DX11 until we had DX11 only games (that made extensive and deeply embedded use of things like DirectCompute, Tesselation, etc.) we won't see the true power of DX12 until the engines come out that only support DX12, which wont be for at least another year or longer.
Unless y'all are willing to go on record saying that nVidia wont push CR and ROV at all over all of DX12's lifespan, Buzzkiller has a good point.
Just because its not here on the first 2 or 3 DX12 games doesn't mean it won't show up ever...
Last edited:
thesmokingman
Platinum Member
- May 6, 2010
- 2,302
- 231
- 106
exar333
Diamond Member
- Feb 7, 2004
- 8,518
- 8
- 91
nVidia itself
Pascal = Maxwell + Mixed Precision + 3D Memory + NVLink
This makes total sense considering NV hadn't had a new Tesla product in years. They NEED another pro offering and this would match that desire. Very likely, asynchronous capabilities would be a distant priority.
It does, Tom Clancy's The Division uses conservative rasterization for HFTS ...
It's not done using DX12 but I'm pretty sure CR is used in Tom Clancy's The Division using an NVAPI extension ...
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 22K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 411
Facebook
Twitter