GroovyBee, on Mon Aug 6, 2012 12:28 PM, said:
What you also have to consider about that myth is that not all programmers are created equal. What is "hard" to one programmer can be considered as a walk in the park to somebody else. Its all down to a persons ability, knowledge and experience. For example, from the current jag programmers I've talked to none of them consider the jag to be hard to program for (myself included).
The people who perpetuate this myth seem to belong to one of three catagories.
1. They are not themselves programmers.
2. They have never actually tried to program on the Jaguar.
3. They lacked the ability to stay focused long enough to complete a project.
There is nothing inherantly difficult about programming this system that a basic acceptance of your own skill level and experience cannot overcome. Bad workmen will always blame their tools.
As far as 3D rendering goes, however, the way I think about it is this.
I have Griff/Inner Circle [ST demo crew] 3D render source code, along with several other famous ST crews render engine source codes from the ST days. A direct one to one port of this code to the Jaguar, 68000-68000 would give the Jaguar a 2x speed boost over the ST.
Factor in that the ST planar bitmap mode, when converted to Jaguar chunky would give another speed boost in the rendering section of code, and that the Jaguar's CRY mode would make doing the lighting calculations trivial, and you get yet more speed.
Factor in the blitter for line filling, instead of software rendering, you get even more speed.
Factor in a non direct port to the GPU - which would be again fairly trivial given the architectural similarities in the instruction sets of the 68000 and the JRISCs and you get an even bigger performance increase.
Given that the basic ST version of the engine could deliver a frame rate and experience above and beyond most of the existing Jaguar 3D polygon games, it's not beyond reason to expect super slick smooth graphics from the system. Check out 'No Second Prize' on the ST for an example - 8Mhz, pure 68000, silky smooth goodness. All it will take is someone willing to put in the time and effort to do the hard work. A lot more of which is required than is needed for the ability to whine about the state of affairs and bemoan the lack of pre-built libraries that don't exist.
So, Am I going to be the one to port this code? No. I have absolutely no interest in 3D games. I like my retro 2D, thank you very much. That is my personal preference. I grew up with classic arcade games, that's what I like playing, that's what I enjoy making. That's what the rest of the people in the team I work with enjoy. We do this for fun. For ourselves. Because we enjoy it. However, it's all out there on the net, waiting. It won't take much from someone with the dedication, experience and motivation to do it.
So, pretty much what I'm saying is: Those who can (and have the motivation, skillset and time) do, those who can't sit about on forums complaining that they've been waiting for someone to write code for them for years so they can realise the dreams they can't make happen themselves.