bgfx module/target

MikeHart

Administrator
CX Code Contributor
3rd Party Module Dev
3rd Party Target Dev
3rd Party Tool Dev
Joined
Jun 19, 2017
Messages
3,454
Yes, I noticed that it is kinda slower than native mojo1. But you did a great job so far. I am looking for a new render backend for CX and BFGX might be a good choice. Would you mind if we integrate your work somehow into CX?
 

Ferdi

New member
3rd Party Target Dev
Joined
Jun 21, 2017
Messages
87
You can. But, you sure you want to do it now in the current form? My biggest worry, currently Mojo is using dynamic vertex buffer and dynamic index buffer, after doing examples, I am pretty sure this should be transient vertex buffer and transient index buffer. I am not sure whether dynamic will cause problem, like memory leak. Transient buffers are freed after each frame by bgfx.

EDIT: sorry Mike, I am out of the loop, did an important target stop working?
 
Last edited:

MikeHart

Administrator
CX Code Contributor
3rd Party Module Dev
3rd Party Target Dev
3rd Party Tool Dev
Joined
Jun 19, 2017
Messages
3,454
No all targets are working atm. But the Apple ones eill stop soon i think. So switching to a new render backend is definitely on the horizont.
I don't plan on just slapping it in. But if bgfx is up to the task, then your work is definitely a head start. Compatibilty with mojo1 and 2 is a must for sure.
 

Ferdi

New member
3rd Party Target Dev
Joined
Jun 21, 2017
Messages
87
Hi Mike, I can't do Mojo 2, I could not get it working. I forgot exactly why I can't get it working, but it is in this thread. I can get some of examples going but not all. If you want something like Mojo 2, it is best to create new API. And if it is new API, I rather the community just tell me what the API are and see if I can get it working. If I make my own API, it will be weird and no one will use it.

Another option is to extend Mojo 1.
 

MikeHart

Administrator
CX Code Contributor
3rd Party Module Dev
3rd Party Target Dev
3rd Party Tool Dev
Joined
Jun 19, 2017
Messages
3,454
Would you did with the mojo1bgfx import was basically a mojo2 kind of wrapper, at least I from what I saw. A complete new API is the most easy path which I would love to take. And yes, I could work with you together on such an API. Supporting mojo1 through that wrapper would be awesome still. Let me think a little about it.
 
Last edited:
Top Bottom