Page 1 of 2

Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Jun 06, 2018 4:03 am
by Deceiver
Well, we survived the end of 32bit apps but seems Apple will end OpenGL/OpenCL and want people to move to Metal. What do we do??

"Deprecation of OpenGL and OpenCL
Apps built using OpenGL and OpenCL will continue to run in macOS 10.14, but these legacy technologies are deprecated in macOS 10.14. Games and graphics-intensive apps that use OpenGL should now adopt Metal. Similarly, apps that use OpenCL for computational tasks should now adopt Metal and Metal Performance Shaders.
Metal is designed from the ground up to provide the best access to the modern GPUs on iOS, macOS, and tvOS devices. Metal avoids the overhead inherent in legacy technologies and exposes the latest graphics processing functionality. Unified support for graphics and compute in Metal lets your apps efficiently utilize the latest rendering techniques. For information about developing apps and games using Metal, see the developer documentation for Metal, Metal Performance Shaders, and MetalKit. For information about migrating OpenGL code to Metal, see Mixing Metal and OpenGL Rendering in a View."

https://developer.apple.com/macos/whats-new/ <--- at the bottom of the site there's more info

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Jun 06, 2018 5:28 pm
by Melekor
This presents a problem for us since all our graphics devs are on Windows and the OpenGL code that's powered the Mac build has mostly been developed as a consequence of the Windows efforts...

This is obviously the case for tons other games as well. Generally speaking if you have small market share, you absolutely want to support cross platform APIs to incentivize developers, otherwise a lot of them won't bother or even be able to afford to support your platform.

Apple is doing the exact opposite here and it seems, frankly speaking, completely insane. I think for now we just wait and see. I think there's at least 50% chance of them backtracking on this just like Microsoft did when they said they would kill off OpenGL.

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Jun 06, 2018 11:15 pm
by Myrd
I don't think OpenGL will stop working any time soon. It will just stop getting updates, might get buggier, etc.

I think we should wait and see what happens. I suspect at some point there will be an OpenGL -> Metal library that we could just use. Or we port to Vulkan and use a Vulkan -> Metal one.

Anyway, I don't think it's disappearing in the next couple years, so we have time.

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Thu Jun 07, 2018 12:39 pm
by Deceiver
Myrd wrote:I don't think OpenGL will stop working any time soon. It will just stop getting updates, might get buggier, etc.

I think we should wait and see what happens. I suspect at some point there will be an OpenGL -> Metal library that we could just use. Or we port to Vulkan and use a Vulkan -> Metal one.

Anyway, I don't think it's disappearing in the next couple years, so we have time.
Good to hear this!!

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Mon Jun 11, 2018 12:25 pm
by Deceiver
So I have installed macOS Mojave beta 1 on my MacBook Pro -- seems to be working fine: https://vimeo.com/274537909

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Sun Aug 26, 2018 9:18 am
by Eddaweaver
Won't you still have software graphics mode after OpenGL becomes unavailable?

Many big game houses gave up on Mac around 2005-2008. At least that's what I noticed by looking at platform releases for major games. Apple doesn't sell "rigs" anymore and their Mac Pro has morphed into Darth Vader's Teapot. :facepalm:

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Sun Aug 26, 2018 10:42 pm
by Melekor
The software renderer is highly undesirable -- ugly and slow. We would work hard to avoid this being the only option on a platform.

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Sep 12, 2018 5:54 pm
by GimlisGrenade
Have just got back into playing Myth again after a several year hiatus. I have been aware of thetain/totalcodex etc and their previous iterations since they were first around.
I just want to say how appreciative I am of everyone responsible for keeping Myth 2 still playable.
I find it incredible that I can still play the game I played when I was a teenager (and remember buying the game brand new), but now on my retina MacBook Pro running the latest macOS. It is still just as addictive, plus I now have probably all the hi-res effects that I wanted back when a 3Dfx card was the holy grail of hardware upgrades.

The other day I played my first LAN game of Myth 2 in over 20 years. Playing with my stepson, over wifi, with my MacBook Pro and a late 2009 iMac, both ran flawlessly. Sadly he completely wrecked me, but is at least keen on exploring Fear & Loathing and making his own maps :)

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Sep 12, 2018 6:09 pm
by Melekor
GimlisGrenade: Reading that made me smile. Thanks for commenting :)

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Fri Jun 12, 2020 12:26 am
by Reiginoob
Myrd wrote:I don't think OpenGL will stop working any time soon. It will just stop getting updates, might get buggier, etc.

I think we should wait and see what happens. I suspect at some point there will be an OpenGL -> Metal library that we could just use. Or we port to Vulkan and use a Vulkan -> Metal one.

Anyway, I don't think it's disappearing in the next couple years, so we have time.
Rumour has it Apple will announce that they’re going to transition Macs from Intel processors to beefed up versions of their ARM CPUs that are in iPhones and iPads at WWDC in two weeks. According to this there’s a good chance that Macs will stop supporting OpenGL and everything will need to use Apple’s Metal API instead.

All this is speculation but do I need to start worrying that Myth will stop working on macs at some point soon?

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Fri Jun 12, 2020 4:57 am
by Eddaweaver
I'd expect you'd still be able to run X86 apps through virtualisation much like the PPC/x86 fat binaries of old though knowing Apple they may discontinue that at some point.

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Fri Jun 19, 2020 8:04 pm
by GodzFire
We've already reached the end of Myth essentially as 10.14 broke Oak and 10.15 killed off Fear and Loathing. There are no devs left to update them and the remnants of the ProjMag team left (1-2 people) don't seem willing to allow new devs who want to help in. It's really unfortunate.

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Tue Jun 23, 2020 7:30 am
by Melekor
GodzFire wrote:We've already reached the end of Myth essentially as 10.14 broke Oak and 10.15 killed off Fear and Loathing. There are no devs left to update them and the remnants of the ProjMag team left (1-2 people) don't seem willing to allow new devs who want to help in. It's really unfortunate.
I am going to be open sourcing Oak soon which should help here. It can also replace fear with a bit more work.

Myth will get another update eventually, but it's in a decent state for now.

Loathing is the difficult one, I don't see an easy path to get it upgraded for Mac.

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Jun 24, 2020 3:25 am
by Jon God
Wasn't Idles or something working on a new version of Loathing? Whatever happened to that?

Also, due to the way it's coded, I am assuming Loathing functionality in Oak is pretty much impossible?

Re: Is Apple's move to end OpenGL / OpenCL the end of Myth?

Posted: Wed Jun 24, 2020 9:36 am
by Melekor
Good memory. Yeah, Idles was porting it to wxWidgets. I don't recall exactly how far it got but I believe it was more than 50% done (should check it out at some point and see where it's at). Finishing that up would be the best bet for loathing on modern OSX.

Loathing in Oak: Loathing uses the Myth code so putting it inside Oak wouldn't work with my plan to open source Oak. And rewriting all the loathing functionality from scratch would be a waste of time. Either way it would also need UI which is also a ton of work. While I agree it would be cool in theory, I don't think it can be justified.