• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Directx and OpenGL getting low-level access soon (GDC 2014)

Zyae

Member
More options are great, but I have zero faith in Microsoft. The sooner PC gaming can move away from their platform the better, and unfortunately this might just keep Windows trucking along a wee bit longer. Still, that aside - I don't mind DX being improved in the meanwhile.



lol. You really think that this has any effect on Windows health, and that Windows is just "trucking along"
 

H6rdc0re

Banned
Good for PC gamers and bad for Nvidia because without the CPU bottleneck their $1000-1300 GPU's won't be that attractive anymore.
 
Good for PC gamers and bad for Nvidia because without the CPU bottleneck their $1000-1300 GPU's won't be that attractive anymore.

As more and more games (hopefully) adopt a lower overhead system, more games will (again hopefully) also take greater advantage of that "free CPU" to do more interesting things. (I'm personally still waiting on that "good AI" that game developers mention from time to time)
 

H6rdc0re

Banned
If it is GPU limited... then yeah, it may. Freeing up a CPU bottleneck does not change the reality of GPU bottlenecked scenarios.

With Multi GPU's there's always a CPU bottleneck because of the API overhead. Something Mantle already clearly has shown with BF4.
 
With Multi GPU's there's always a CPU bottleneck because of the API overhead. Something Mantle already clearly has shown with BF4.

...

If there is a CPU bottleneck, then wouldn't it be Intel and AMD who stand to lose the most by having it eliminated? You know, companies which make CPUs?
 

Seanspeed

Banned
Good for PC gamers and bad for Nvidia because without the CPU bottleneck their $1000-1300 GPU's won't be that attractive anymore.
Quite the opposite. This actually helps high-end graphics card users by eliminating the CPU bottlenecks they were previously facing. If anything, this is fantastic news for Nvidia.
 

H6rdc0re

Banned
...

If there is a CPU bottleneck, then wouldn't it be Intel and AMD who stand to lose the most by having it eliminated? You know, companies which make CPUs?

It's not a CPU bottleneck when the CPU is waiting for the API.

Better CPU's will always be needed. When developers start to push better lighting (ray-tracing) and physics (soft boddies, fluids). Those need a lot of CPU power so bigger and better CPU's will still be needed. Bad for the POS Nvidia PhysX.

Im kinda sure that I can run Crysis 3 at thas res and fps in my pair of 680

With API overhead I'm sure they can but with the current mess that's Direct X no way. My overclocked GTX780TI can't even run 60fps locked @1920x1080.

Quite the opposite. This actually helps high-end graphics card users by eliminating the CPU bottlenecks they were previously facing. If anything, this is fantastic news for Nvidia.

Running a game on a Single GPU @60fps instead of needing Multi GPU's benefit Nvidia.
 
It's not a CPU bottleneck when the CPU is waiting for the API.

Better CPU's will always be needed. When developers start to push better lighting (ray-tracing) and physics (soft boddies, fluids). Those need a lot of CPU power so bigger and better CPU's will still be needed. Bad for the POS Nvidia PhysX.

Draw call overhead causes a CPU bottleneck which mostly affects slower CPUs, so if it's eliminated then the need for a faster CPU is diminished. And PhysX is just an API which exists in a hardware implementation on Nvidia cards, it falls back to the CPU if no hardware is found.

Clearly you're not a fan of Nvidia, but that's no reason to spout gibberish.
 

H6rdc0re

Banned
Draw call overhead causes a CPU bottleneck which mostly affects slower CPUs, so if it's eliminated then the need for a faster CPU is diminished. And PhysX is just an API which exists in a hardware implementation on Nvidia cards, it falls back to the CPU if no hardware is found.

Clearly you're not a fan of Nvidia, but that's no reason to spout gibberish.

Limiting draw calls in current API's limits performance of every CPU. Next generation games will see an increase in lighting and physics which with better CPU's can and should be done by the CPU.

PhysX in it's current state is terrible and you know it is. I always buy Nvidia GPU's but I hate their current way of doing business.
 

Nzyme32

Member
Really rooting for OpenGL to gather momentum. Its about time gaming moved to a multiplatform open API. Its a win for everyone
 

Orcus

Member
This is what's been going on just fine since the demise of 3dfx until AMD decided that they know better.


You do know that internally 7 is 6.1 while 8 is 6.2 and they both are nearly identical to Vista which is 6.0 in pretty much all underlying technology, right?
The only reason why something is exclusive to 7 or 8 is because of marketing.
Actually, the reason they didn't change the major version number was due to legacy software and drivers, which was a lesson learned from Vista's compatibility issues. Basically a lot of software was just checking to see if windows was version 5.x, so when they upped the version to 6 a lot software no longer worked and it was impossible to get everyone to update old software. It had nothing to do with marketing.
 

dr_rus

Member
Actually, the reason they didn't change the major version number was due to legacy software and drivers, which was a lesson learned from Vista's compatibility issues. Basically a lot of software was just checking to see if windows was version 5.x, so when they upped the version to 6 a lot software no longer worked and it was impossible to get everyone to update old software. It had nothing to do with marketing.
Right, and we don't have compatibility mode since XP to fix this awful "problem".
The reasons why they haven't used 7.0 and 8.0 are because both 7 and 8 are an extension of Vista in technology. XP was an extension of 2000 (that's why it was 5.1 and not 6) and Vista was a fresh start in many ways including a new driver model - that's why DX10+ being exclusive to Vista made sense and why nothing being exclusive to either 7 or 8 makes any sense beyond marketing.
11.1 being exclusive to 7 makes zero sense because 11.1 is a minor update to 11 which was brought to Vista as an update. Same goes for 11.2 being exclusive to 8.
These internal numbers tells much more about the underlying technology than you think.
 

Serandur

Member
Do we have anything to worry about here with regards to backwards compatibility issues?

None whatsoever, at least not from the APIs themselves (Windows and legacy compatibility issues are another story). New features aren't likely to threaten existing and continually-supported feature-sets, nor will low-level access improvements be likely to improve older software not designed for it or with the latest API revisions either.
 
Right, and we don't have compatibility mode since XP to fix this awful "problem".
The reasons why they haven't used 7.0 and 8.0 are because both 7 and 8 are an extension of Vista in technology. XP was an extension of 2000 (that's why it was 5.1 and not 6) and Vista was a fresh start in many ways including a new driver model - that's why DX10+ being exclusive to Vista made sense and why nothing being exclusive to either 7 or 8 makes any sense beyond marketing.
11.1 being exclusive to 7 makes zero sense because 11.1 is a minor update to 11 which was brought to Vista as an update. Same goes for 11.2 being exclusive to 8.
These internal numbers tells much more about the underlying technology than you think.
What do you want, them to spend the development resources to back-port all their APIs and software forever? You're also being ridiculous in saying that the work they put into 8 and 7 were only minor refinements to Vista. The kernel version isn't everything, either...

Fact of the matter is, yeah, I do agree with you about 11.1 and 11.2, not that last I checked, developers were clamoring to use the new features added in those APIs in 7, but at a certain point, just like the Internet Explorer back-ports, you're asking for them to do a lot of work to back-port some fundamental frameworks of the Windows OS to 7, and probably in a solution that's not nearly fleshed out as the implementation in the later OSes.
 
I'm glad this is coming to both DirectX and OpenGL, since the most people benefit that way, but I really do want a huge push for broader OpenGL support.
 

DieH@rd

Banned
AMD sent this to PCPER:

"AMD would like you to know that it supports and celebrates a direction for game development that is aligned with AMD’s vision of lower-level, ‘closer to the metal’ graphics APIs for PC gaming. While industry experts expect this to take some time, developers can immediately leverage efficient API design using Mantle, and AMD is very excited to share the future of our own API with developers at this year’s Game Developers Conference."
http://www.pcper.com/news/Graphics-Cards/DirectX-12-and-new-OpenGL-challenge-AMD-Mantle-coming-GDC
 

DieH@rd

Banned
ExtremeTech:

We’ve spoken to several sources with additional information on the topic who have told us that Microsoft’s interest in developing a new API is a recent phenomenon, and that the new DirectX (likely DirectX 12) will substantially duplicate the capabilities of AMD’s Mantle. The two APIs won’t be identical — Microsoft is doing its own implementation — but the end result, for consumers, should be the same: lower CPU overhead and better scaling in modern titles.

This has already been read in several circles as to be the death knell for AMD’s custom API, but such claims are short-sighted, for multiple reasons.
First, there’s the fact that DirectX 12 is almost certainly 12-18 months away. Second — and equally important — there’s the fact that Microsoft has been locking DirectX releases to Windows versions. DirectX 11.1 is only available on Windows 8; DirectX 11.2 is only available on Windows 8.1. Microsoft has every reason to synchronize the launch of DX12 with its next version of Windows — and if it follows its typical MO, that means DirectX 12 will be most likely be a Windows 9 exclusive.

Microsoft, of course, doesn’t have to play its cards that way, but if it does, Mantle may remain relevant as a cross-platform alternative for extending close-to-metal benefits across the gaming stack. Of course, OpenGL may or may not be capable of performing the same tasks in the same time frame — but less is known on that front. Hardware support could also complicate matters — if DirectX 12 closely maps to Mantle, it’s possible that today’s GCN GPUs will still support it. Alternately, if it doesn’t, then Mantle may become the preferred option for ensuring broad backwards compatibility.

And if Mantle is ultimately subsumed by DirectX — so what? When I first talked to AMD about the next-generation API at APU13, the developers candidly told me that the long-term goal was to get Microsoft and the Khronos Group in charge of OpenGL to adopt a Mantle-like architecture. The entire point of Mantle was to spur game development and drive the adoption of a better standard.

http://www.extremetech.com/gaming/1...rectx-12-will-imitate-and-destroy-amds-mantle
 
Top Bottom