Indeed. Many ports are the ST version verbatim with only changes to the graphics due to the different bitplane organization and different sounds (Amiga uses samples, ST uses synthesis).
The existence of the ST definitely had a negative impact on Amiga software development, especially games.
Jack Tramiel unfortunately managed to convince about everyone that the ST was selling way more than the Amiga which led developers to make that machine the lead one for the 1985-1988 period when the ST was actually not selling well at all: it seems that the Amiga did in fact outsell it fairly early after the A500 model was introduced but alas the damage was made and developers blissfully unaware of the reality of the market continued to develop for the ST first with the Amiga version as an afterthought.
As Kamalh noted, replacing CPU based sprite routines with blitter ones would not take more than a few days of work for any competent developer so the fact that this was rarely done serves as a great illustration of the sorry state of game development of that era with publishers cutting costs as much as possible without regard for the platforms capabilities. There were plenty of talented coders but they rarely were given a chance to shine and a great example is the port of Final Fight for the Amiga where a single coder (Richard Aplin) was tasked with both ST and Amiga conversions under six months of work without access to any original Capcom material safe for an Arcade cabinet of which he ended up dumping and disassembling the ROMs of his own initiative. Even if I consider the port to be pretty crap, it is nevertheless quite a technical exploit under these conditions.
Combine this with Commodore absolute ineptitude in packaging and marketing the machine and you had a recipe for long term failure, a shame given how ahead of its time and cheaper the machine was.
There were initially some license fees to become a registered developer and have access to the documentation but these were relaxed after a few years and anyone could purchase the hardware documentation, which incidentally I did as soon as it became available in the Amiga shop in my home town.
Interesting! I would say that this supports my hypothesis.
The randomness of the result of the modulation is why I was mentioning tools: with proper tools, one would expect to be able to sample an actual instrument, isolate a clean loop of the waveform, then use some analysis software to spit out the best approximate modulation parameters. Even during these days this would have been a fairly simple program to write.
I mean, most keyboard synths of that era used the same hardware as the console and arcade machines and they managed to produce relatively good sounding instruments so the capability definitely existed.
The reason I was mentioning cars is that any vertex not projected for the cars can be reused for something else so it is a worthwhile tradeoff. It is true that they do not usually suffer from perspective correction issues.
Also, the cars may be small but they require more details than the road or side objects because they need to be beautiful and realistic so they have a higher vertices/polygons density than the rest of game 3D elements. I do not remember the exact numbers but I think the cars in VRally 2 have more than 140 polygons at the highest LOD (of which there were two), so reducing this to two-thirds via pre-computing of polygon lists would allow an equivalent number of vertices to be added to the close parts of the road strip. Moreover cars are fairly boxy overall so they are well suited for angle based polygon elimination.
I only have Lotus II and Super Hang On on my MegaDrive (Genesis to be precise) and both have the same issue (Super Hang On being much worse but it is an early game) but I have been indeed wondering about that. I see no reason why a smooth 50/60Hz raster road cannot be easily achieved with tiled graphics + row-scroll but I have yet to see one on the MD.


Reply With Quote



