New Ogre project on OS X (macOS)

[JayMcBee]

Hi there -

so I'm about so start a new OS X - or macOS - app (and extend another existing one..) that both need awesome 3D graphics


2-3 years ago I did a prototype with the 1.1 SDK and back then already decided to never go back to low-level OpenGL programming if at all possible.

However, the Ogre '[version chooser' (What version to choose?) suggests that in order to use the latest 2.1 SDK "You don’t care about OS X & mobile support for now and can wait for it.". Guess I do, though.

Seems like an awkward time to start a new, long-term project using Ogre.. I would really prefer running the latest version to take advantage of the latest features, in particular performance and cool graphics.

Could anybody give some broad timeline regarding the availability of 2.1 on OS X/macOS?
Does it make sense to hold our breaths (the current plan is to release Q3/Q4 2016)..?

I've also noticed there's a (pretty active!) 2.1 metal branch - maybe that'll be the best option for the coming months?

Any feedback much appreciated!


Cheers,
Jay

[EricB]

Your best bet is to use v1.10 if you want to release a product sometime in the next ~3 years.
I believe dark_sylinc is currently working on the metal branch based on some commercial investment. The answers you seek are right there in the 2.1 sub-forum: http://www.ogre3d.org/forums/viewtopic.php?f=25&t=85890

Please note that v2.0 is a tech demo and probably shouldn't be used in a commercial product. 2.1 Is still highly in development. So expect instability and API changes right under you. If you don't mind constantly rewriting the game and fighting the engine often due to bugs or changes then 2.1 is for you. Otherwise stick to the stable old 1.x branch.

My personal opinion is that 2.0 should not have been released (and recommendations to use on the site does many folks a disservice.) And 2.1 should be labeled as '2.0 Beta', which should be used by those experimenting, learning, writing foss code. Not for anyone wanting stable code bases.

[JayMcBee]

Your best bet is to use v1.10 if you want to release a product sometime in the next ~3 years.
..
My personal opinion is that 2.0 should not have been released (and recommendations to use on the site does many folks a disservice.) And 2.1 should be labeled as '2.0 Beta', which should be used by those experimenting, learning, writing foss code. Not for anyone wanting stable code bases.

Thanks for the feedback!
Wow, sounds rather like we should target a different graphics engine entirely more actively under development..?!

Seriously pondering now..

Cheers,
Jay

[frostbyte]

what do you mean "more actively under development..?!"
BronzeHead mentioned "2.1 Is still highly in development"....thats where the focus is....
dark_sylinc mentioned a rough estimate of 3-4 month for the 2.1 metal branch

i think BronzeHead did'nt meant ogre-OSX-metal branch would be ready within 3 years
but that you would spend a lot of time dealing with bugs and updates since ogre2.1 was meant to be a "Community preview" and is not officialy released yet
you would have faster a dev-cycle using 1.10( which btw have a backport of PBS and hlms ) and porting to 2.1 when time comes...

sorry to tell you but OSX is not really a target platform for gamers/3d since apple decided for this platform to be focused on business sector
http://store.steampowered.com/hwsurvey

so due to OSX small market share and lack of good OSX graphics driver support, ogre-OSX gets much less attention, this may change with METAL

last thing: in terms of features and stability i must say that Ogre is still the best FOSS RenderEngine i know of...
sure it can be buggy and difficult to handle, but situation on many other engines( including commercials ) may not be much better...
https://answers.unrealengine.com/spaces ... ashes.html

[xrgo]

I am in love with 2.1, performance is great, no stability problems, in the last year has been like 2 disruptive changes and took me like 5 mins to solve, so... its a wip, but its veeeery usable (we are using it for commercial products with success). If you can afford wait for metal I can definitely recommend it.
About using other engines, its not fair to compare, Ogre is a rendering only engine, which gives me control of everything... I use the libs that I like, with only the components that I like, etc... but obviously Unreal or Unity are good choices for "just start making games right away!"

but situation on many other engines( including commercials ) may not be much better...
https://answers.unrealengine.com/spaces ... ashes.html

And I can contribute with this: https://www.reddit.com/r/gamedev/commen ... not_unity/ (quite old though)
newer https://www.reddit.com/r/gamedev/commen ... not_unity/

[EricB]

Wow, sounds rather like we should target a different graphics engine entirely more actively under development..?!

Ogre3D has a lot of active development. In fact you could probably say it's the most actively developed FOSS engine out there.

i think BronzeHead did'nt meant ogre-OSX-metal branch would be ready within 3 years

I wasn't referring to the Metal renderer specifically. It will be useable in short order. Specially since it's commercially funded.

The 3 years refers to a stable period of finished features, stable API, and stabilized bug free (as much as possible) code. My personal opinion is you don't go chasing active development branches when making commercial products. Where a 1% difference in Steam reviews (79% vs 80%) results in 20% revenue difference. Every little negative crash review adds up.

Personally I would look into 2.1 when development on 2.2 is started. (Currently I'm on 1.7.4 for Windows/Linux and 1.8.1 on OSX until my project is completely finished, then I'll bump up to 1.10 right before going into the maintenance period before EOL)

so due to OSX small market share and lack of good OSX graphics driver support, ogre-OSX gets much less attention, this may change with METAL

Our game, running on 1.8.1 on OSX works perfectly fine. One could say even better than our Linux port. (Short of full screen on some machines. It's an issue with OS X's Window Manager... Picky little thing.)

I am in love with 2.1, performance is great, no stability problems, in the last year has been like 2 disruptive changes and took me like 5 mins to solve, so... its a wip, but its veeeery usable (we are using it for commercial products with success). If you can afford wait for metal I can definitely recommend it.

Not saying you can't use 2.1 for a commercial product. I know of 2? games on Steam that are doing it now. However, there is no guarantee that some massive API upheaval may happen in the future, or a bug that completely screws up your development for weeks pops up. If you have money and/or time to throw at it, congrats! I operate on a shoe string budget (as probably most folks who use Ogre) so I'd rather have a little bit more guaranteed stability over brand-new under-tested features.

[JayMcBee]

Thanks for all the feedback - and sorry if the original question offended anybody, I probably got the original answer wrong or maybe the general ideas behind the various branches, i.e. how/where they should be used.

I'm now using 1.9 for my (OS X, iOS) projects and with the backports mentioned I guess that's a great basis to cover the app features planned for the near future.
FWIW my background is not games but simulations and 3D editors in productivity apps so no need to squeeze every last frame out of the machine. Pretty keen on experimenting with large scenes on current iPad hardware, though.. smooth animations would be crucial for a decent user adoption.

Anywho - thanks for clarifying!

Cheers,
J