Apple Deprecating OpenGL

[LAwLz]

2 minutes ago, djdwosk97 said:

Umm, Microsoft does force you to use DX12 for UWP apps, which they're pushing for.

Yes, and it is one of the many reasons UWP is a cancerous framework that should be condemned.

It's a clear attempt vendor-locking from Microsoft. It's disgusting.

[DrMacintosh]

1 minute ago, valdyrgramr said:

It's like lately all Apple is good at is pissing people off.  Like them, if you want, but this is just going to piss people off more.

Correction, it's only going to piss off devs...and you might not like this, but Devs don't really have a choice. Their livelihoods are entirely depending on companies like Apple and Microsoft and they will always give into what these companies say, they have no other option. 

 

The average end user of a Mac is not running legacy 32Bit OpenGL software anyway, so the end user is only going to get benefits from a move like this. 

[79wjd]

Just now, valdyrgramr said:

In that one scenario, but I can still use OpenGL, Vulkan outside of it.

 Until they start forcing UWP and drop support for win32. (Assuming they don't figure out that literally no one will use Windows for anything without it's massive compatibility first)

[DrMacintosh]

1 minute ago, valdyrgramr said:

The end user is not going to have anything to open if they push bs like this.

I'll believe that when I see it. 

[LAwLz]

3 minutes ago, DrMacintosh said:

Correction, it's only going to piss off devs...and you might not like this, but Devs don't really have a choice. Their livelihoods are entirely depending on companies like Apple and Microsoft and they will always give into what these companies say, they have no other option. 

You're describing a dictatorship as if it's a good thing. I find it genuinely scary how you can be so docile and defend anything Apple does.

"Hooray, Apple can force developers to do whatever they want because if they don't comply they will starve to death! Yay for Apple!"

 

1 minute ago, DrMacintosh said:

so the end user is only going to get benefits from a move like this. 

Please explain how this will benefit end users.

You keep saying there will be benefits but so far you have given us 0 arguments for it.

And no, comparing Metal vs OpenGL is NOT a benefit.

Again, developers can already use Metal if they so desire. Nothing is stopping them. Apple is strictly REMOVING things here, not adding anything. They are limiting developers.

[DrMacintosh]

3 minutes ago, LAwLz said:

Please explain how this will benefit end users.

Just look at macOS before and after Metal was implemented in areas like Window management, Finder, rendering, decreasing overhead, and it gives Apple the benefit of full control over their platform. 

 

Granted that last point is good for Apple, but all the others have made macOS an objectively better OS because of the implementation of Metal. 

[LAwLz]

3 minutes ago, DrMacintosh said:

Just look at macOS before and after Metal was implemented in areas like Window management, Finder, rendering, overhead, and it gives Apple the benefit of full control over their platform. 

 

Granted that last point is good for Apple, but all the others have made macOS an objectively better OS because of the implementation of Metal. 

Again, you're comparing OpenGL vs Metal and going "see how much better it performs!".

Apple did all the things with Finder BEFORE they dropped support for OpenGL. You seem to think that you can only support one set of APIs, which is wrong. You can easily support both.

 

Again, all this is doing is removing options. So please explain to me how the removal of OpenGL (not the addition of Metal) benefits end users.

 

 

And I'll repeat this again so that you fully understand.

SAYING THAT METAL IS BETTER THAN OPENGL IS *NOT* AN ARGUMENT FOR REMOVING SUPPORT FOR OPENGL. YOU CAN SUPPORT BOTH, WHICH IS WHAT APPLE HAS DONE SO FAR WITH EASE.

 

I am not asking how Metal is better than OpenGL. I already know the answer to that.

What I am asking is how removing OpenGL is better than supporting both Metal and OpenGL.

[DrMacintosh]

7 minutes ago, LAwLz said:

So please explain to me how the removal of OpenGL (not the addition of Metal) benefits end users.

I think you are not recognizing the implications behind my posts. So I guess I'll spell it out for you. 

 

Before Metal, OpenGL was all Apple used, and macOS performed the way it did because of the limitations of those APIs......

 

After Metal, the end user experience has significantly improved because macOS is no longer being held back. 

 

So yes, you could leave OpenGL around to hang out in the back ally of macOS, but why would you do that when you can make sure developers are making the best applications that they can for your platform and benefit your people. 

 

Some apps are going to stop working, and some developers are going to be pissed, but you know what? That's just too bad. People are going to move on and developers are going to move on, and most importantly, macOS is going to move on. 

[DrMacintosh]

1 minute ago, valdyrgramr said:

You can keep arguing "nu uh!  I got that feeling!"  But, in reality, no dev enjoys restrictions.

Then devs can stop complaining about it and actually start doing something about it. 

[Thaldor]

After looking some specs and showreels, I think I might have found why Apple forces Metal and don't want to support Vulkan. There really isn't any performance differences, but in computer graphics Vulkan is few steps ahead, few quite big steps. Those steps are mainly compatibility and usage, while Vulkan is firstly made to compete with Direct3D and to be universal successor of OpenGL, Metal is just made to be successor of OpenGL in Apple products firstly in mobile and even still mainly Metals 3D part is done iOS in mind. This means that in the high-end of spectrum Vulkan seems to eat Metal as breakfast.

 

Looking at the Rise of the Tomb Raider, there isn't much difference between Linux and Windows versions, but Windows and Mac OS have some minor deffirences. Mac version has somewhat softer geometry, it has increased fog effects (just for the show or to lower the drawing distance?) and lightning is diluted (shadows are more grey than black). These might be because of bad developer, but they can as well be the limits of a API that is mainly made for mobile gaming and GUI. On a mobile game you wouldn't really need that precise tesselation and you can sacrifice a lot of geometry just because no one sees that well, it's a very old trick to add fog to hide that your graphics engine cannot draw any longer and on mobile devices no one is really enjoying the scenery and on mobile devices you would probably want to dilute the shadows to be grey so you don't need to worry about screen lagging (black to white is always longer transition time than grey to grey/white) and phone screens, especially LCD screens, are not there yet to give fast reaction times (OLED is there, but Apple uses IPS). So, the problems can be in bad translation or just that the Metal API is just not there yet for computer gaming.

 

So, basicly, Apple wants OpenGL away because it is old and no one really uses it to it's limits anyway, but refuses to support Vulkan, because it's better than their own and those few developers who still care about Mac gaming wouldn't otherwise even look twice at the Metal as their graphics API.

 

Also then there's that even if your company is only developing games for Macs and there was a chance to choose between Vulkan and Metal, would you choose the graphics API that over 99% of PC gamers can use or invest to learn or hire someone who can work with graphics API that less than 1% of "PC" gamers can use?

[Trixanity]

1 hour ago, LAwLz said:

Please explain how this will benefit end users.

You keep saying there will be benefits but so far you have given us 0 arguments for it.

And no, comparing Metal vs OpenGL is NOT a benefit.

Again, developers can already use Metal if they so desire. Nothing is stopping them. Apple is strictly REMOVING things here, not adding anything. They are limiting developers.

This is eerily similar to the headphone jack debate which Apple is also in the center of.

 

Removing features as if they haven't already coexisted and worked perfectly.

 

If the 'new' feature/technology was superior in every way then it would make the predecessor disappear without forced deprecation. 

 

Apple is just stubbornly trying to make their investment in their own API bear fruit (pun not intended but I suspect a subconscious effort) despite it being a stupid idea to begin with considering they could have contributed to Vulkan instead. However Apple wants things to be unnecessarily exclusive and proprietary for maximum obnoxiousness.

[Sniperfox47]

4 hours ago, Nicholatian said:

The performance impact is negligible, and you’re more than welcome to show me your prowess in engineering and demonstrate otherwise. Honestly you sound so full of yourself I’m sure you’ll try anyway. You’ll want to leave out the melodrama and show some real research about that, though!

https://www.phoronix.com/scan.php?page=article&item=dota2-mac-vulkan&num=2

 

I would hardly call a 20% performance differential "negligable". This is not a *strictly" 1-1 comparison because MacOS doesn't support Vulkan and Ubuntu can't support Metal, but it's as close a comparison as you can get.

[79wjd]

1 hour ago, Trixanity said:

This is eerily similar to the headphone jack debate which Apple is also in the center of.

 

Removing features as if they haven't already coexisted and worked perfectly.

 

If the 'new' feature/technology was superior in every way then it would make the predecessor disappear without forced deprecation. 

That's true over long scales; people don't like change. 

[Blademaster91]

2 minutes ago, djdwosk97 said:

That's true over long scales; people don't like change. 

Like the 3.5mm jack removal,people don't like change when said change isn't necessary.

[79wjd]

4 minutes ago, Blademaster91 said:

Like the 3.5mm jack removal,people don't like change when said change isn't necessary.

Except it's already driven bluetooth headphones forward quite substantially -- there were far fewer options both at the ultra cheap and more premium ends before they dropped the headphone jack. Now, bluetooth headphones are far more ubiquitous because someone forced the issue. 

[Drak3]

1 hour ago, Sniperfox47 said:

This is not a *strictly" 1-1 comparison because MacOS doesn't support Vulkan and Ubuntu can't support Metal, but it's as close a comparison as you can get.

If Vulcan can be ran on PureDarwin natively, it'd be a perfect Apples to Apples comparison.

[Blademaster91]

6 minutes ago, djdwosk97 said:

Except it's already driven bluetooth headphones forward quite substantially -- there were far fewer options both at the ultra cheap and more premium ends before they dropped the headphone jack. Now, bluetooth headphones are far more ubiquitous because someone forced the issue. 

Except that it's only because they forced it on everyone, which isn't necessarily a good thing when people should be for choices not artificial limitations. It always works for Apple though,everyone still lines up for the newest thing that requires a pile of dongles.

And "premium"  bluetooth headphones still don't touch even midrange headphones i can plug into everything else without needing to charge it.

 

[79wjd]

5 minutes ago, Blademaster91 said:

Except that it's only because they forced it on everyone, which isn't necessarily a good thing when people should be for choices not artificial limitations. It always works for Apple though,everyone still lines up for the newest thing that requires a pile of dongles.

And "premium"  bluetooth headphones still don't touch even midrange headphones i can plug into everything else without needing to charge it.

 

No one forced anyone to buy an iPhone 7 or an 8/X. They could've kept what they had, gotten a 6s, or an entirely different phone. The fact is that for most people it's completely irrelevant to not have a headphone jack.

 

And no, most people don't need a pile of dongles, because most people don't plug all that much into their computers. 

 

Most people still aren't using audiophile headphones. Nor do most people care. And those that do can leave the 3.5 to lightning adapter, that's included, on their headphones -- as even fewer people switch between devices frequently.

[Sniperfox47]

1 hour ago, Blademaster91 said:

Like the 3.5mm jack removal,people don't like change when said change isn't necessary.

Except it's even worse than that.

 

The headphone jack removal, at least from a developer side has a legitimate reason. The extra traces and isolation for the audio lines to the jack, as well as the jack itself take up precious real estate in the phone. You as a consumer may not agree that that's a good reason but it is a valid reason.

 

This has really no reason since the "support" that they're depreciating never existed in the first place. All the work for OpenGL is handled by the driver that AMD, not they, make.

 

The only reason I can see being this is that they're planning an ARM MacBook and want to be able to say "Well the software that doesn't work with it was already depreciated! We told developers! It's their fault!". Apple's PowerVR GPUs for their ARM SoCs don't support full OpenGL, just OpenGL ES.

 

 

[LAwLz]

7 hours ago, DrMacintosh said:

After Metal, the end user experience has significantly improved because macOS is no longer being held back.

I'm gonna do you a huge favor and give you the benefit of the doubt that that's true.

From what I know, those things you mentioned are all 2D rendered, and for that there is not a big difference between Metal and OpenGL, from what I know.

But I will not ask you to provide any evidence for that claim because my guess is that you have no evidence for it.

 

 

7 hours ago, DrMacintosh said:

So yes, you could leave OpenGL around to hang out in the back ally of macOS, but why would you do that when you can make sure developers are making the best applications that they can for your platform and benefit your people.

Because Metal is not always the best. If it was just flat out better in every way then developers would already have moved over on their own, without being forced.
One of the biggest drawbacks of Metal is that the code is not portable. So no, it is not better in every way and that's why you need to give developers options, so that they can use the right tool for the right job.

Here Apple is going "screwdrivers are better than hammers, so I will take your hammer away from you". The builder already had both a screwdriver and a hammer, and used them for different things. The builder having a hammer did not prevent them from using the screwdriver to screw things in, right?

 

 

7 hours ago, DrMacintosh said:

Some apps are going to stop working, and some developers are going to be pissed, but you know what? That's just too bad. People are going to move on and developers are going to move on, and most importantly, macOS is going to move on. 

That is not a justification. Saying "get over it" is not an excuse for shitty behavior.

 

 

7 hours ago, DrMacintosh said:

Then devs can stop complaining about it and actually start doing something about it. 

Start doing what exactly? On the last page you proudly said that developers are forced to do whatever Apple dictates. Here is the quote:

7 hours ago, DrMacintosh said:

Correction, it's only going to piss off devs...and you might not like this, but Devs don't really have a choice. Their livelihoods are entirely depending on companies like Apple and Microsoft and they will always give into what these companies say, they have no other option.

 

 

[Christophe Corazza]

5 hours ago, Blademaster91 said:

Except that it's only because they forced it on everyone, which isn't necessarily a good thing when people should be for choices not artificial limitations. It always works for Apple though,everyone still lines up for the newest thing that requires a pile of dongles.

And "premium"  bluetooth headphones still don't touch even midrange headphones i can plug into everything else without needing to charge it.

 

 

5 hours ago, djdwosk97 said:

No one forced anyone to buy an iPhone 7 or an 8/X. They could've kept what they had, gotten a 6s, or an entirely different phone. The fact is that for most people it's completely irrelevant to not have a headphone jack.

 

And no, most people don't need a pile of dongles, because most people don't plug all that much into their computers. 

 

Most people still aren't using audiophile headphones. Nor do most people care. And those that do can leave the 3.5 to lightning adapter, that's included, on their headphones -- as even fewer people switch between devices frequently.

 

I actually did the opposite. I have wired headphones and a 3.5 mm bluetooth receiver for the devices that don't have the 3.5 mm jack anymore. 

[normality]

So here's the crux of an issue as it affects our business. We're a relatively small indie studio whom primarily makes freemium mobile games. We primarily focus on the IOS platform as our products tend to earn roughly 80% of our income from advertising and roughly 20% from IAP. Whilst we have a much larger user-base on Android, advertising partners will generally pay a great deal more for an IOS user, as there is believed to be a correlation between the use of an Apple device, and the amount of disposable income a user possesses. 

Over the past two or so years we've been heavily considering creating a PC and Mac collection of our mobile games with a refined meta loop, not only to try and move players away from using hacked android APKs on their desktops but also allowing for influencers to easily capture gameplay footage, as a step towards furthering our relationship with Apple. As an example, we've created Apple TV ports of several of our titles, all of which have not earned nearly enough money to break even on the technical investment of the port, but as a way to elevate our position when the business comes to feature consideration for IOS.

Whilst we primarily work with Unity, and are therefore not directly affected by the deprecation of OpenGL, the reality is that many local developers of our size are. And, when talking about mobile or desktop builds of a product, it is important to realise that almost every decision that a developer makes is down to opportunity cost. Ensuring that your title can work across multiple APIs ads a great deal of additional technical risk and overhead to a product, and unless a developer is working with an engine like Unity, it is very likely that developers will have to think much harder about how and if it is worth their while to release across both platforms instead of focusing on just one. In the mobile space, if this decision for the Mac OS is a precursor to Apple locking down IOS to Metal, I dare say many developers would likely end up with a much larger gap between Android and IOS releases, should they chose to release on both platforms at all. In terms of desktop development, as the marketplace is currently so tiny on the Mac, I dare say you will see far fewer developers taking the time and energy to even dip their toes in that pool. 

Sure, a company like 2K, Valve or Bethesda will obviously be able to provide the resources to ensuring that their title can perform on metal, but the reality is that smaller AA or Indie studios simply don't have the resources to make that happen without trading off somewhere else in the production (marketing, content, features etc.). Its all about opportunity cost, and the last thing that the Mac OS needs is another barrier for entry for developers, particularly when it isn't really necessary. Sure, Metal might allow for the same frame to render faster than Vulkan, but again, that should be the devs decision. And from a user's perspective, I'm sure they would rather have a selection of games that look okay on their system, rather than a few hit titles that look spectacular with nothing else available. 

Sorry for rant, hopefully this was in some way informative.

[normality]

14 hours ago, DrMacintosh said:

Developers need to figure out how to tell Apple that Vulkan is important. Vulkan has basically no adoption in any category, and Vulkan, while seemingly being a fantastic API, has not proven itself to be the best choice for a company like Apple which has spent millions developing their own API to meet the demands of their OS. 

 

If Devs want Vulkan then Devs need to prove that Vulkan has a future. 

Going back a few pages with this quote but I think it is relevant.

Apple is obviously a very ... *interesting* partner to work with. We work with Apple and Google frequently for software development, however Apple doesn't publicly acknowledge that these services exist, nor are they at all transparent when informing us of changes to the platform or the direction of our pre-release titles. Google however is much more receptive and friendly, and will do a great deal more to give a developer confidence in their platform.

Apple might be very strong headed and secretive when it comes to their customer facing side, but I can tell you that they are just as difficult if not more so as a developer. Its not uncommon to get short, vague emails from Apple, and we literally have company wide meetings where we all attempt to dissect and interpret the coded information as best we can. If we return an email with questions, generally we will just be advised to re-read the previous email, and phone calls will often go unanswered if they believe that you will ask questions around sensitive topics. 

Trying to push Apple to make changes to their platform is unthinkable, as you risk loosing support from the company, as well as visibility on their platform, something you absolutely need as an Apple developer. As such, it is incredibly uncommon for a developer to have enough clout and resilience to push Apple on any issue, as they must be prepared for the fallout. 

As such, I'm not really sure how you can turn the ship around. I think this one just has to play itself out, and perhaps a year or so after the changes come into effect (again assuming Vulkan is also dismissed) and content on the platform has dried up, Apple might then realise the impact of their decision. But honestly, I just don't think there are enough stable game developers with relationships with Apple to make any impact on this decision at this stage. 

[Thaldor]

51 minutes ago, normality said:

Trying to push Apple to make changes to their platform is unthinkable, as you risk loosing support from the company, as well as visibility on their platform, something you absolutely need as an Apple developer. As such, it is incredibly uncommon for a developer to have enough clout and resilience to push Apple on any issue, as they must be prepared for the fallout.

This ^

 

Probably only AAA game studios and bigger publishers could afford loosing support from Apple and manage to still break even with their game. But these players are as interested on Mac OS as platform as something that isn't interested at all. Smaller companies cannot afford to loose the support and featured status from Apple and I can very well see that happening if they start to push Apple not to do something they have decided to do. And even if thousand developers or even ten thousand developers voice their opinion, I don't see Apple care about it since they still have hundreds of thousands developers who mainly do fart apps and they can just say "it's only a marginal amount of developers from our developers". Probably only developer that could have some voice would be Feral, but seeing them voicing against Apple is something that probably never happen.