macOS is deprecating openGL

news
software

(BeerBaron) #239

The complete opposite is true. @skw gave a good list of counter-examples, but it’s far from exhaustive. Remember when Java was the next big thing that all Mac OS applications were going to run on?

Oh, okay, we’re arguing about what “big” and “small” is. If one thing is six times bigger than the other thing, that other thing is small to me. But that’s not important, the point is, is it big enough?

Not at all. Like I said, a huge amount of software doesn’t ship on Mac, but also a lot of software is Mac-only. Supporting multiple platforms has to pay off. The actual market share of Mac users for your niche may be 95% or 0%, depending on the kind of software you offer.

I wholeheartedly disagree. In an abstraction layer, when you add some feature and it requires something your layer doesn’t cover, you need to touch N+1 places. Sometimes one feature isn’t available on all platforms, leaving you with the lowest common denominator, or a leaky abstraction.

You also have the mental overhead of how your abstraction layer translates into actual platform code for N platforms. Refactorings become more expensive, the architecture will be more rigid and less likely to improve.

Last but not least, this is the kind of work that absolutely nobody likes to do, which is bad news for a project like Blender, which is supported by volunteer and below-market-rate work.

No developer has stated anything to that effect, as far as I’m aware. You should read the whole thread. Ton has signaled interest in MoltenVK, but he isn’t keen on acting right now.


(m9105826) #240

The only reason that abstraction layer exists is because Unreal and Unity need to run on non-standard hardware that may or may not have access to standard APIs. Blender (nor any other content creation suite) does not suffer from this limitation. Indeed, until just recently, OpenGL was the “abstraction layer” that allowed it to run wherever it wanted.


(m9105826) #241

Blockquote
The share of Linux desktops is a small (< 2%), yet Blender supports Linux.

The obvious difference being that it costs no extra development time to support Linux, because Linux doesn’t demand proprietary nonsense.


(apoclypse) #242

[quote=“BeerBaron, post:239, topic:1113803, full:true”]

13% is huge if you factor the users as opposed to numbers.

DCC is a very niche market, tiny in that huge number you like to quote for the Windows platform. DCC users are a much bigger ratio on the Mac. The mac has an almost disproportionate amount of user who are graphic artists, or work in media, web developers, etc. I mean by your logic Adobe should abandon the platform altogether. They haven’t because Apple user make up a significant amount of their user base and revenue. They don’t care that Windows has 95% market share because in their piece of the pie Windows market share alone isn’t where the money is at and never will be a long as Apple is still around.

Linux in their minuscule <2% share still command a significant portion of cg houses which is why Autodesk happily accommodates that market. If we look at pure numbers Maya shouldn’t even exist on Linux. Pure marketshare numbers tell you nothing about how marketable or how successful your software will be.


(Ace Dragon) #243

If we go back even further though, that logic would suggest that Microsoft should’ve instantly failed as a company because Windows wouldn’t have any software (Macs. made up the vast majority of all desktops in the 1980’s and early 1990’s).

I do agree.


(m9105826) #244

Apple is big for DCC in terms of design, but in all of the CG houses I’ve worked at and visited over the years, I could probably count all of the Macs I encountered on my fingers and toes (excluding comp departments). And since the legitimate Mac Pro (non-trashcan) was discontinued a few years back, I don’t think I’ve seen a single new Mac in a modeling/sculpting/rigging/animating/rendering environment. Apple hasn’t even made a cursory attempt to keep current with hardware advances, and there isn’t a single production house in the world that could justify the impact of the “Apple tax” on their bottom line when looking at new equipment. Dell, HP, and sometimes boutique niche builders like BOXX have long dominated the 3D/CG space, and for good reason.


(apoclypse) #245

True and I wasn’t saying Apple is good choice versus other platforms. I was saying that looking at marketshare alone is not a good indicator of what platform you should be developing for.

That being said, while Apple doesn’t have a foothold in the professional hardcore 3D CG market (FX houses, AAA games, etc). Apple does well with motion graphics, smaller scale 3d production. Game development (indie) for iOS is still popular to do on macOS, etc.


(BeerBaron) #246

If you read my entire post, you’ll see that I’m making that point myself. Some software is exclusive to the Mac, because it makes sense in certain niches. The point is, cross-platform development costs extra money (and nerves!), so it needs to pay off. Mac OS is not a platform you need to support.

For Blender, the share of Mac users is about 10% going by the download statistics.

Most of their software doesn’t run on Linux, but some of it was originally written for SGI (Unix) workstations and those were replaced by Linux. Yes, some CG houses run Linux, but that’s also more of an SGI legacy thing.


(apoclypse) #247

Blender definitely doesn’t need to support macs, but in a market where revenue is slim and cost of development is high even a percentage of a 13% is (especially on a usually affluent willing to spend crowd like mac users) it would be silly to disregard that market and say you don’t need it. Especially since there are a lot of iOS game developers or indie developers out there. For Blender not being a paid product this could be a non issue. They may not need to care. I don’t think it will help their brand tho.

Also 10% of downloads is nothing to sniff at. Windows may have many more downloads but that doesn’t tell you anything about the kinds of users who are downloading the app. It could be people who have no intention of ever producing anything with the software and would like to just tinker around. We need more data imo to really gauge if dropping the Mac platform just based on pure download percentage is a good idea. I just don’t think its as black and white as that.


(BeerBaron) #248

I can’t make sense of this. Yes, cost of development is high, cost of supporting another proprietary graphics API is especially high, so those 13% of users have to translate into a disproportionate amount of revenue.

Development costs are notoriously hard to estimate and most developers don’t exactly have a good business sense, so when left to their own devices, they go out and write abstraction layers and support Mac OS and Linux on principle. My whole point is merely that you can forego all that and just ship on Windows. Maybe you’ll miss out on profits, maybe you won’t, but if your business fails, it won’t be because you didn’t support Mac OS or Linux.

This is what the majority of software companies (including game companies) do, so it should be obvious.


(John Forde) #249

Proprietary or not doesn’t make any difference. Supporting Linux requires extra development time.


(John Forde) #250

It very well might fail because you don’t support MacOS or Linux. The broad coverage of platforms, not the individual market shares, is the point here. Which is why the major game engines all support MacOS and Metal. Why should I choose to use your software if it doesn’t support all the platforms I am targeting?


(John Forde) #251

This is absolutely incorrect. Macs have NEVER had a majority. Their biggest percentage of market share is now.


(Jason van Gumster) #252

Not if a large proportion of your developers are already running Linux.


(apoclypse) #253

Game companies don’t make games for Apple’s because of a number of factors, market share may be one of them but it’s probably not the primary factor. Like I said Mac users in general are more willing to spend on software than on other platforms. It’s the user base that’s the issue there. Like I said the majority of users on the platform are professionals of some sort who may not be interested in games at all. You are also talking about a platform with higher cost of entry and generally lower specs in-terms of gpu performance than you average windows machine.

Let’s look at it like this. The iOS Appstore has a huge amount of games, probably more than most other types of apps, iOS is not the platform with the most marketshare. Android is, by a significant factor. Yet iOS still gets a huge amount of development. Why? Because iOS users are more willing to spend on their apps than Android users, Apple has all the revenue by a significant margin.

It makes no sense to me to determine what platform you will develop for on marketshare alone. You should base that on where your users are and who is most willing to pay for the software. On top of that you should be determining if entry into an already crowded market is a good idea because your ROI may only be weeks, versus months on a platform with less competition, you could potentially make more money selling to an under appreciated market versus just catering to the larger market where there may be five apps that do the same thing you do.

Its not as obvious as you say. You state most software developers but give no examples. I can provide examples where developers have actively cater to Mac user because that’s where their specific users are (DAWs for example. FLStudio just received a mac version). You aren’t going to release a DAW today and not have a mac version. That’s where most of the audience is and they will buy your software quite happily.


(BeerBaron) #254

Game engines are a special case, they are already designed with multiple renderer backends for different platforms in mind, because you need one for every console. At that point, adding another backend becomes cheaper. Also, iOS is an important platform for Metal.

On the other hand, Unreal Engine Editor didn’t support Mac OS for the longest time, Unity Editor supported it because it started out on Mac, neither officially support Linux (though it works). None of the console SDKs support Mac OS or Linux.

Yet, not all games that could ship on Mac OS or Linux (because they use Unity/Unreal) actually do.

Nobody is arguing that.

Nobody disagrees with that. I made that point myself. You’re talking to a straw man. Yes, DAW is strong on Mac OS. DAW also doesn’t need 3D graphics.

The whole argument isn’t “supporting Mac OS is pointless”, it’s “supporting Mac OS is optional unless your niche is totally Mac-centric”.


(Gandalf) #255

well if you code for iOs you need to code in mac or not code for iOs


(apoclypse) #256

Right and this is the point I’m disagreeing with, because it’s not just mac centric niches we are talking about here. You point is that it’s optional based on market share. Your niche doesn’t have to be mac centric, it just has to be small enough that alienating potential customers makes no good business sense.


(BeerBaron) #257

That’s just not true. Supporting a platform has a fixed cost no matter how many “potential customers” you have. If that fixed cost isn’t carried by the amount of actual customers, you have a net loss. If the niche is smaller, it’s actually more likely that the fixed cost isn’t carried.

Autodesk’s Alias is a good example of a niche product. Autodesks knows how many Mac customers they have, they have an idea what it would cost to support Metal and they came to the conclusion that it would be a net loss. Hence, they kill the Mac version. Of course all of Autodesk easily has the resources to port Alias to Metal, it’s just not a winning proposition.

The whole equation was different when you would get Windows, Mac OS and Linux in one package by choosing OpenGL over Direct3D. Of course in practice OpenGL on Mac OS was inferior, but that’s something developers had to find out the hard way.


(apoclypse) #258

Alias is not a good indicator of anything. It was project that was starved for resources (even before metal) because Autodesk already has a cross-section of other software that does the same thing (just like they dropped Softimage for the same reason). If anything I’m surprised it was still around and wouldn’t be surprised if Windows support dies at some point too.