@slaapliedje
but then why was that article talking about "blitting code" Am I wrong in thinking most of the modern display drivers and such wouldn't need that low level of coding, and indeed maybe they're talking about a port to OS3.9?
The thing is: the PC version uses hw accelerated rendering in a way that doesn't work on most NG Amiga systems. Among other things it uses a giant 8192 x 8192 texture atlas internally, which is a size that's not even remotely possible on many of our systems. Besides that the high amount of texture data itself already exceeds the VRAM of many NG Amiga systems.
Therefore I had no choice but to write a software renderer as a replacement. Under the hood that one consists out of lots of blitting functions, each one specialized and highly optimized for the respective task, depending on drawing mode.
All in all I'm happy to say that this software renderer is actually much faster than the hw-accelerated rendering would have been (I also have a win32 build of the game that uses an Intel version of that renderer - and it runs faster with that than with the original rendering
).
I know it's wishful thinking, from what I've seen of the particles and such on screen in their initial demos, unless they cut all of that out, an older 68k system wouldn't be able to handle it.
It's impossible. I'm not saying that it is impossible to write such a game for 68k, of course it is possible (well, of course not nearly in that quality, but you know what I mean, sth. like a Chaos Engine enhanced).
But this game here has not been designed with low-end systems in mind. The unmodified original game logic ran with less than 9 fps on a sam460. It has been designed with current PCs as target systems in mind, highly trading performance for engine flexibility. Had to do lots of modifications to the original code until it ran decently with around 30 fps on a sam460 (and with some additional tweaks also on a sam440) - and I'm not done here. Once the PC version is complete I'll give it yet another big optimization run. Right now I'm pausing to let the Pixwerks guys finish their job first though. I will have to dig really deep and mess around with the source to a much higher degree compared to what I did until now and that would conflict too much with the Pixwerk's guys job to actually finish the game itself. And of course it's best to do such fat optimizations once everything is done.
@klx300r
actually it's the opposite for backers because now we're getting online multiplayer which wasn't on the cards after the kickstarter since the goal wasn't reached at that time
Unfortunately online multiplayer on NG-Amigas won't happen, because a "black-boxed" commercial network gaming library has been chosen for that task.
In theory I could write a networking system myself, but sorry, no. This would also require a server to be set up, of course it would be incompatible to the other versions of the game, etc. And I really already spent much much much more time (and still have to) with that port than anticipated, due to the facts mentioned above.
For those who don't follow the progress reports on facebook, here's my youtube channel where I post progress videos from time to time:
https://www.youtube.com/user/Daytona675x
