Originally Posted by Harry
Loading a big part of the game into memory was an advantage for those games which used in worlds (parts of) the same graphics over and over again. For this reason many track formats even had a rudimentary custom file structure on them. These parts could then be copied to expansion memory and loaded from there. I made once a version of Projectyle for me which loaded the half of the disk which contained the arenas, score boards etc. multimedia data in a 512 k expansion memory, and it showed the goal faces with 2 instead of 30 seconds delay. (A pity that the original company hadnt done this).
Otherwise, the used memory detection routines were mostly broken, until the fact that Pang detected the Amiga-XT-Brigdeboard-Card-State-Transfer area as ram, crashing the Game, because the XT-Side modified the data. The ONLY correct way to detect a ram area is to allocate it with the OS, eg allocate from $80000-$d0000 in chipmem with AllocAbs and later assume that that are 512 k. Same for Non-Chip-Ram: Allocate the largest free "Fastram" block, make sure it is larger than 256 k (to exclude these small expansions), and clip its boundary to a 512 k border; in the case of old customized handmade hd installs you may also leave the odd start if the game runs with it and it is large enough.
Yup. Of course you can buffer disk tracks (or files) in any type of memory. The thing is, that is optional - i.e. doesn't up the hardware requirements - in other words a great option
Using fastmem for real-time buffering (which is obviously what I meant above) is very limited - mostly you can only use it to save loading (And this is also how it has been used since slowmem expansions became commonplace. Before that, assuming 'everyone' had it limited sales.)
Loading on any computer has throughout time been used as just a way of extending RAM to a slower memory. That's how it was used to make more impressive games than just 512K allowed, and that's how modern games fit GB of textures on a 512MB graphics card today.
(You know all this of course, but if everyone on the forum spoke to experts only it would be a forum that excluded people.)
What I think is important isn't whether some old game crashes on some Amiga hardware combo today, but whether there will ever be a substantial amount of excellent and absorbing new Amiga games in the future.
I've had my company since forever and many times I've planned to give it a try, only to go "oh, who are kidding, who's going to care." I registered amigafuture.com at one point to set up a kickstarter.com-like community where anyone could post a project plan (or at the very least a luxurious showcase of projects for people to gather round), but again the doubt. (And the life(tm) and time and new interests factors.) I'd leave my job and work full-time as Amiga game coder for my company in an instant. I have a simple almost-finished game and an Exile/Gravity-Force game that is playable but only one level on my harddisk, but how do I make a living? It's hopeless. I'm sure I'm not alone in this feeling.
Elite: Dangerous got funding on kickstarter.com on the second try just now, and that's good. (And the money is insane!) But it's not an Amiga game. Who cares.
Well, I do. And others do. But more people caring with their time and money, that's a challenge. It's obviously not impossible, just - a - bit - hard.
OK, I better stop 'opening up' now. Before I give you all chronic depression. Haha, good night