Vulkan API instead of DX12 API

Why not give it a try? Is too much trouble?

Well, if major game studios like EA, Sony, Blizzard, Epic and major brands like Intel, Nvidia, AMD are working with Khronos, can't be that bad.

Heck, if even Microsoft is moving close to OpenGL with DirectX 12 (YES, Xbox API is more close from OpenGL than DirectX) this new Vulkan API can't be that bad.

Why not have both (DX12 and Vulkan) so users can chose whats fit better? Is too much trouble?

That is what I have been trying to say this entire time. ^^^

Don't just limit us to DX12. Especially if they are considering adding Linux/Mac support in the future. Why put off what you can do now? And if they do release a Linux/Mac client and they use WINE with DX9....NO ONE will buy the game because Currently, Linux games can and do look freaking amazing and run just fine when written Natively. When games are ported using an emulation system........they can and sometimes do suck. But Valve noticed that if they added OpenGL code to the engine and made some adjustments for the superior Memory Al-locators in the Linux ecosystem, they got on average 20 fps better than the Windows version.

For an example of what games in OpenGL (Vulkan is fresh and is NOT OpenGL) and other Open-standard APIs from Khronos Group look at the last video in my previous post. That was rendered completely in OpenGL. Most likely OpenGL 4.4 or 4.5. Vulkan example is in the Intel video.

Bohemia would have to re-write code not only for Arma 3 but also for Real Virtuality 4 (which was designed for Windows and XBox from the very beginning) to make it happen. So: not going to happen, so why even waste your time to discuss it.

Maybe in Arma 4.

That is what I have been trying to say this entire time. ^^^

Don't just limit us to DX12. Especially if they are considering adding Linux/Mac support in the future. Why put off what you can do now?

Because, as i have said so many times before, no matter how much you love either Linux or OSX, those are never gaming platforms, hence they are low priority/no priority.

Where did you get that BI is considering adding support for either one btw? There is no such desire.

Where did you get that BI is considering adding support for either one btw? There is no such desire.

There is. The devs confirmed themselves in a SITREP that they had employed a third party to investigate the possibilities of porting the client to Linux and OSX.

http://dev.arma3.com/post/sitrep-00104

Because, as i have said so many times before, no matter how much you love either Linux or OSX, those are never gaming platforms, hence they are low priority/no priority.

Where did you get that BI is considering adding support for either one btw? There is no such desire.

I beg to differ. Linux is a gaming platform as much as Windows is a gaming platform. both Operating systems have plenty of games. Sure, windows is the current go to Operating system for gamers because of DX. But, OpenGL is just as powerful as DX can be. In fact, prior to DX being released, gaming was OpenGL based. If Linux had no potential for Gaming, Valve would have never even considered let alone moved their platform to Linux.

As for the Linux/Mac ports? SITREP #104 http://dev.arma3.com/post/sitrep-00104

We are indeed experimenting with 2 ports of Arma 3. An external team has been investigating whether it would be possible to create (non-native) client ports to Linux and MacOS.

Perhaps you should try reading the SITREPs sometime. ^_^

If anyone should be doing the port, it should be Icculus. specifically Ryan C. Gordon of Icculus. His ports are amazing. His resume is astounding. He has worked for :

• Valve - Left 4 Dead 2 Port to Linux
  
• Epic Games - UT 3/2003/2004 port to Linux/Mac OS X/win 64
  
• Activision - Ported CoD servers to Linux
  
• Electronic Arts - ported Medal of Honor servers to Linux
  
• Human Head Studios - Ported Prey to Linux
  
• Google - ported Google Earth to Linux
  
• Crytek - Ported Crysis Wars server to Linux
  
• Digital illusions CE - ported Battlefield 1942 server to Linux
  
• Double Fine - Ported Psychonauts and Costume Quest to Linux
  
• Croteam - ported Serious Sam: First and Second encounters to Linux
  
• The Digital Lounge - ported Another World : 20th Anniversary edition to Linux
  
• Aspyr Media - Ported Shrek 2 and Spiderman 2 to Mac OS X
  
• The ArmyGame Project - ported Americas Army to Linux and Mac OS X
  
• Zombie Studios - ported ShadowOps: Red Mercury to Windows 64bit
  
• Running with Scissors - ported Postal 1 & 2 to Linux and Mac OS X
  
• IMVU - ported IMVU client to Mac OS X
  
• Linden Lab - ported Second Life to Linux
  
• Tripwire Interactive - ported Red Orchestra: Ostfront 41-45 and Killing Floor to Linux and Mac OS X
  
• Unreal Tournament Preservation Group - Maintaining UT 1999 for Linux and Mac OS X
  
• Digitalo Studios Inc. - ported Devastation to Linux
  
• Perilith - ported Unrealty client to Mac OS Classic
  
• Pyrogon Inc. - ported Candy Cruncher to Linux, BeOS, and Mac OS Classic
  
• Sprout Games LLC - ported Feeding Frenzy to Mac OS X
  
• PopCap - ported Iggle Pop and Pizza Frenzy to Mac OS X
  
• Trendy Entertainment - ported Dungeon Defenders to Linux
  
• Gaslamp Games - helped develop Dungeons of Dredmor and Clockwork empires
  
• Number-none - ported Braid to Linux and Mac OS X
  
• Klei Entertainment - ported Shank to Linux and Mac OS X
  
• Team Meat - ported Super Meat Boy to Linux and Mac OS X
  
• Bit Blot - ported Aquaria to Linux and Mac OS X
  
• The Chinese Room and Robert Briscoe - ported Dear Esther to Linux
  
• Lazy 8 Studios - ported Cogs to Linux
  
• MinMax Games - ported Space Pirates and Zombies to Linux and Mac OS X
  
• Kranx Productions - ported Hammerfight to Linux and Mac OS X
  
• Mode 7 Games - ported Frozen Synapse to Linux
  
• Oberon Media - ported Candy Crisis to Mac OS X
  
• Reflexive Entertainment - ported Ricochet: Lost worlds and Big Kahuna Reef to Mac OS X
  
• Spiderweb Software - ported Avadon: The black Fortress to Linux
  
• Large Animal Games - ported RocketBowl to Mac OS X
  
• Wolfire Studios - ported Lugaru HD to Linux and Mac OS X
  
• Coffee Stain Studios - ported Sanctum 2 and Goat simulator to Linux and Mac OS X
  
• Playdead - ported LIMBO to Linux
  
• AudioKinetic - ported ( in collab with Playdead) Wwise to Linux
  
• Procedural Arts - ported Facade to Mac OS X
  
• Red Marble Games - ported Magic Match Adeventures, Go-Go gourmet and Go-Go Gourmet 2 to Mac OS X
  
• Gunnar Games - ported Cars: Radiator Springs Adventures and Hidden Mysteries Civil War to Mac OS X
  
• Loki Entertainment Software - Lead dev on Linux port of Descent 3, Heavy Metal: FAKK2, Kohan: immortal Soverigns, Quake 3 Arena, and Eric's Ultimate Solitaire
  
• Icculus.or and open source projects - Sysadmin for 100+ open source devs across dozens of projects, Lead dev on : FatELF, PhysicsFS, MojoShader, MojoSetup, Mac OS X openAL, ManyMouse, SDL_Sound, MojoPatch, mod_offload, and Toby
  
• Designed and taught: visual Basic, Javascript, and computer Literacy classes to secondary and k-12 students.
  
• Has written: "Porting MFC to GTK+" and "The Whys and Hows of Porting Software"
  
• Proficient in: C,, C++, x86 Assembly, Lua, Java, Perl, PHP, UnrealScript, etc, SDL, DirectX, OpenGL, OpenAL, 64-bit, BSD Sockets
  

As you can see, Ryan C Gordon is the perfect guy for the job

I can imagine that for BISim it might be more interesting to look at porting the VBS codebase to GNU/Linux and OpenGL(ES/Vulkan). Mainly because of the Linux-based embedded systems used by the many simulators NATO armed forces use.

Interesting news, thought it was related: http://www.pcworld.com/article/2940470/hey-gamers-directx-11-is-coming-to-linux-thanks-to-codeweavers-and-wine.html

Anyone played any fully fledged dx12 games yet ? ... I think this joke of dx12 fixing everything in broken games is a myth and until its proven in the field then take it all with a pinch of salt. A3 add-on and dx12 is not yet proven and until a preview build is out then its all marketing bull tbh.

Interesting news, thought it was related: http://www.pcworld.com/article/2940470/hey-gamers-directx-11-is-coming-to-linux-thanks-to-codeweavers-and-wine.html

You want something related? Here you go:

http://www.extremetech.com/gaming/207767-apple-brings-its-metal-api-to-os-x-10-11-kicks-vulkan-to-the-curb

A Vulkan version is hardly coming at this point. That makes it almost certain that there will be no Llinux version. Get over it.

Anyone played any fully fledged dx12 games yet ? ... I think this joke of dx12 fixing everything in broken games is a myth and until its proven in the field then take it all with a pinch of salt. A3 add-on and dx12 is not yet proven and until a preview build is out then its all marketing bull tbh.

We are all hoping for a miracle and a heavy reworked engine. :D

Anyone played any fully fledged dx12 games yet ? ... I think this joke of dx12 fixing everything in broken games is a myth and until its proven in the field then take it all with a pinch of salt. A3 add-on and dx12 is not yet proven and until a preview build is out then its all marketing bull tbh.

This is not a joke. This is just you being ignorant.

AMD's Mantle has more than proven itself in several games. Dx12 and Vulkan are just trying to do the same thing that Mantle has been doing for well over a year, and there's no reason why they should not succeed.

As I have said a thousand times on this forum before, I played Thief using Mantle, and it literally doubled my frame-rate, going from an average of 45 to around 90 frames per second. That is the best case scenario I have encountered. I have also played other games with Mantle and the results have varied. In BF4 I saw maybe 10% increase in average frame-rate, but the experience was still much better because it completely removed the microstuttering I had on Dx11. In Dragon Age: Inquisition it was about the same performance in Dx11 and Mantle I think, but again it seemed to microstutter less with Mantle which is always good.

The only publicly available Dx12 software currently is a benchmark tool from Futuremark that tests how many draw calls per second you can manage with different API's. On AMD cards you will see about 15 times as many draw calls per second than Dx11.

You are right about the marketing campaigns though. Just ignore them, but open your eyes, look elsewhere besides marketing material, and you will find that there is good evidence that the technology is going to be spectacular. It's going to vary a lot from game to game though. I don't expect Dx12 or Vulkan to do much good in Arma. I do however expect it to do some amazing things in some big games coming out later this fall, and early next year, like Deus Ex Mankind Divided, which will run Dx12.

This is not a joke. This is just you being ignorant.

AMD's Mantle has more than proven itself in several games. Dx12 and Vulkan are just trying to do the same thing that Mantle has been doing for well over a year, and there's no reason why they should not succeed.

As I have said a thousand times on this forum before, I played Thief using Mantle, and it literally doubled my frame-rate, going from an average of 45 to around 90 frames per second. That is the best case scenario I have encountered. I have also played other games with Mantle and the results have varied. In BF4 I saw maybe 10% increase in average frame-rate, but the experience was still much better because it completely removed the microstuttering I had on Dx11. In Dragon Age: Inquisition it was about the same performance in Dx11 and Mantle I think, but again it seemed to microstutter less with Mantle which is always good.

The only publicly available Dx12 software currently is a benchmark tool from Futuremark that tests how many draw calls per second you can manage with different API's. On AMD cards you will see about 15 times as many draw calls per second than Dx11.

You are right about the marketing campaigns though. Just ignore them, but open your eyes, look elsewhere besides marketing material, and you will find that there is good evidence that the technology is going to be spectacular. It's going to vary a lot from game to game though. I don't expect Dx12 or Vulkan to do much good in Arma. I do however expect it to do some amazing things in some big games coming out later this fall, and early next year, like Deus Ex Mankind Divided, which will run Dx12.

Ignorant ? I'll take that as not an insult but your over excitement for a new DX component however I stick by my words, until its a fully fledged game, working, not some video showing 'features' that may or may not exist in game then like A3 still not running like the videos (sure its all our fault for nor designing missions or servers correctly lol...) then lets all be a little reserved. I would love the game to get its boost like your single player Unreal Engine game Thief does but this is not that game nor is it an engine that works in the same corridor way, happy to eat my words when it comes out in 201x <-- we hall see. xx

I would love the game to get its boost like your single player Unreal Engine game Thief does but this is not that game nor is it an engine that works in the same corridor way

No one said we are going to see that kind of boost in Arma 3 from Dx12. Personally I would be happy if they manage to increase avg. fps by just a few percent, which I think is reasonable expectations.

You want something related? Here you go:

http://www.extremetech.com/gaming/207767-apple-brings-its-metal-api-to-os-x-10-11-kicks-vulkan-to-the-curb

A Vulkan version is hardly coming at this point. That makes it almost certain that there will be no Llinux version. Get over it.

Apple is kicking Vulkan to the curb because like in everything Apple does, they want to do their own thing. If everyone else is doing it one way, Apple will without fail go their own way and alienate everyone else. Hell, take for example FLAC audio. Apple said, screw the standard, we are going to release the EXACT same thing only call it ALAC or Apple Lossless Audio Codec. What is Apple's Metal API? Vulkan and DX12 Apple style. in other words, make it look pretty and nothing else.

The fact that Apple is releasing their own version means nothing to the whole Linux/Mac Port. In reality, a Linux port is more likely than a Mac port simply because of Valve and SteamOS. So get off your high Microsoft horse and move along.

Bohemia would have to re-write code not only for Arma 3 but also for Real Virtuality 4 (which was designed for Windows and XBox from the very beginning) to make it happen. So: not going to happen, so why even waste your time to discuss it.

Maybe in Arma 4.

I don't think the things are like you are saying.

The Core Engine (VR4) is one thing, the Game Code is another thing, APIs are completely different things and external from Engine or Game Code, usually integrated when coding a game.

With whatever Engine, when you Code a game you can integrate whatever API, from rendering APIs (DX12/Vulkan) to sound APIs (Wwise, FMOD) or similar, it is just a matter of integration methods.

The living proof is that BIS is going to implement DX12 which basically is a completely new API and probably it will require the same effort of implementing Vulkan.

Also, AMD is going to release TressFX 3.0 in short time which will support grass, foliage and real time physics. Looking at what 2.0 did with hair effects, we can expect a massive quality and detail increase at cost of basically zero GPu workload increase (with AMD cards at least).

Now, with these three combined (Vulkan/DX12 and TressFX) I believe there could be a real performance benefit for Arma 3 and also better graphics.

I know that BIS is a bit stuck at Nvidia with Physx but let's be honest, this technology is not being used by the game, so why do not give a try to AMD since what is being presented seems to have some juice? We know the outcome with Nvidia..

I don't think the things are like you are saying.

The Core Engine (VR4) is one thing, the Game Code is another thing, APIs are completely different things and external from Engine or Game Code, usually integrated when coding a game.

With whatever Engine, when you Code a game you can integrate whatever API, from rendering APIs (DX12/Vulkan) to sound APIs (Wwise, FMOD) or similar, it is just a matter of integration methods.

The living proof is that BIS is going to implement DX12 which basically is a completely new API and probably it will require the same effort of implementing Vulkan.

Also, AMD is going to release TressFX 3.0 in short time which will support grass, foliage and real time physics. Looking at what 2.0 did with hair effects, we can expect a massive quality and detail increase at cost of basically zero GPu workload increase (with AMD cards at least).

Now, with these three combined (Vulkan/DX12 and TressFX) I believe there could be a real performance benefit for Arma 3 and also better graphics.

I know that BIS is a bit stuck at Nvidia with Physx but let's be honest, this technology is not being used by the game, so why do not give a try to AMD since what is being presented seems to have some juice? We know the outcome with Nvidia..

in one of the recent SITREPs, it was announced that they were either testing or had released an OpenSource version of PhysX. Since Nvidia has been shafting the Linux side by not giving them PhysX with their drivers........possibly cause the Linux guys would have reverse engineered it for the Open source driver Nouveau.

A truly Open Source PhysX implementation can only mean good things. For example, devs dont have to wait for Nvidia to fix bugs or add features.... not that Nvidia does that anyways without lots of pressure from game devs.

Here is an experimental build demo of Vulkan being used with PowerVR.

I would love the game to get its boost like your single player Unreal Engine game Thief does but this is not that game nor is it an engine that works in the same corridor way

You see pretty nice gains in BF 4 as well and that's by no way "corridor shooter". And the best part, not only that the FPS increased dramatically, but also the frame time was far more stable.

Apple is kicking Vulkan to the curb because like in everything Apple does, they want to do their own thing. If everyone else is doing it one way, Apple will without fail go their own way and alienate everyone else. Hell, take for example FLAC audio. Apple said, screw the standard, we are going to release the EXACT same thing only call it ALAC or Apple Lossless Audio Codec. What is Apple's Metal API? Vulkan and DX12 Apple style. in other words, make it look pretty and nothing else.

The fact that Apple is releasing their own version means nothing to the whole Linux/Mac Port. In reality, a Linux port is more likely than a Mac port simply because of Valve and SteamOS. So get off your high Microsoft horse and move along.

I thought about explaining my points, but maybe I'll keep it for myself. I guess I have no chance to make myself heard. But I'm happy to show you how clueless you prove to be in your last statement:

http://forums.bistudio.com/showthread.php?189433-The-OpenGL-answer-to-DirectX-12-Vulkan

http://forums.bistudio.com/showthread.php?168308-Games-for-Linux

I am just a realist. So I'm off my horse, but you can stay on your unicorn. I'm out of this thread.

I thought about explaining my points, but maybe I'll keep it for myself. I guess I have no chance to make myself heard. But I'm happy to show you how clueless you prove to be in your last statement:

http://forums.bistudio.com/showthread.php?189433-The-OpenGL-answer-to-DirectX-12-Vulkan

http://forums.bistudio.com/showthread.php?168308-Games-for-Linux

I am just a realist. So I'm off my horse, but you can stay on your unicorn. I'm out of this thread.

I have confirmation from Korneel vand 't Land that they are working on a Linux port via third-party contractor.

You are a realist......I am an optimist.

I have confirmation from Korneel vand 't Land that they are working on a Linux port via third-party contractor.

It'll be a wrapper-based Linux port I assume. I do hope that the new engine iteration will have native Linux binaries.

Correct, it was a look into feasibility of a non-native Linux port.

Correct, it was a look into feasibility of a non-native Linux port.

Still beats having no Linux support, big time.

Agreed, I'm just specifying that the Bohemia report specifically declared external team (read: not taking away from the Expansion) and non-native, no mention of interest in a native port.

well, the new APIs may make the option of a native port a realistic option.

Totally agree with main ide, It no matter its windows or linux - it will work same as on DX 12, so why lock only for windows 10, if we can have all features of DX 12 on older systems?