qiuu, on 27 September 2011 - 12:14 PM, said:
I think this is the kind of project I'd enjoy doing as well, so I can totally understand your sentiment. It allows for a lot more experimentation and creativity than something like, say, Sonic 1 and 2, which was pretty goal-oriented and I can imagine partly tedious to do. Here you got some programmer art (simple but aesthetic), spiced up with nice visual effects, and then you can get going trying out whatever you feel like. And I also enjoyed playing this a lot more than Sonic 1 and 2, which, while technically more impressive due to the sheer amount of work that had to be put into it, doesn't have much new and interesting content to offer to me.
Hi qiuu, good to hear from you. When making Sonic 1 and 2 I had an Excel spreadsheet with a list of tasks and estimated durations on it - that fact alone should give you some idea of how enjoyable it was. Having said that though, I always saw that project as a stepping stone to other things, rather than a finished product. I ran out of time to do much more with the first version so I had to release it in the rather uninspiring "two games stitched together" format. However, now that the ground work's done I can get the benefits in other hacks like Sonic VR. The objects were all ready to go, the "split screen palette" trick became the level fade in/out sequence, and the menus were deliberately basic. I just needed a small amount of work on art and backgrounds and then I could spend the rest of the time doing the fun part of designing devious levels with witty names!
qiuu, on 27 September 2011 - 12:14 PM, said:
The penultimate level turned out surprisingly simple once I started using the spindash, almost felt kinda cheap doing it.
Don't worry, that's deliberate. Like many of the levels, it's simply a puzzle that you found the solution to.
qiuu, on 27 September 2011 - 12:14 PM, said:
I think the one downside of using the Sonic engine for this that comes with everything being rather macroscopic (I.e. large on the screen, thus not much fitting within the screen) is that you don't always see where you're going, especially when going downwards, which turned the part with the springs and the spikes at the ceiling into a bit of a guessing game.
That's a really good point you make about the relative screen size in Sonic games, as it's something I had to constantly keep in mind for this hack. I'm glad somebody else has noticed it and it wasn't just me! No doubt Sega very much intended this as a way to increase the sensation of speed. However, it can limit what you can actually get away with in terms of design, particularly if you want the player to look around their surroundings and think before acting. I'm not surprised you've had this difficulty as well, since the hack you're working on is very puzzle-based. This was the first time I had made levels from scratch, and I was shocked by the amount of effort it takes to make layouts that are fair to the player, in that they can react to obstacles in time and don't have to make awkward leaps of faith. The fact that the levels in Sonic VR are deliberately challenging and one-hit-kill makes it even harder to get right.
Take the level Margin of Error for example, which is a series of halfpipes connected by corridors that have spikes on the floor and ceiling. You have to get enough speed up to cross the floor spikes and land safely in the next halfpipe, but not go so fast that you fly into the ceiling ones. It's a really simple design, but it took me ages to find the right starting point for the player, because of this issue of not being able to see very far compared to the distances Sonic can travel. The original starting point for this level was in the bottom of the first halfpipe. You'd begin by spindashing to the right and immediately see the floor spikes. But then your next reaction is to charge up a full speed spindash to clear the spikes, only to hit the ones on the ceiling that you didn't see!
A better option is to start in a corridor to the left, high enough up the wall so that dropping into the halfpipe gives you enough momentum to roll down and along the bottom, up past the floor spikes, and finally reaching the same height as the ceiling spikes on the opposite side. I've then shown the player all they need to know, so it's over to them to mess it up! However, there's no guarantee that the player won't make a blind jump out into space and completely negate this by landing on the bottom of the halfpipe, or worse, on the floor spikes. The solution I eventually came up with was to put the player in a corridor to the left, but place a Casino Night bumper at the entrance. This forces the player to drop slowly into the halfpipe and get the full tour of the area like I want, and hopefully also reminds them of the similar-looking area in Spring Yard zone so they're already expecting a halfpipe with corridors on either side.
There are a bunch of other levels where it's really important to show the player what they're up against, some of which I think I managed to get right, and others I didn't. In the last level, for the spikes on the ceiling part that you mention, I messed up at the very last minute by lowering the springs slightly. The reason was to avoid too many objects in a row because of sprite flickering, but it means you can't see the spikes any more and thus have no reference point to judge distances with. Sorry about that! Hopefully I'll make fewer mistakes next time round, as I've definitely learned a lot about level design over the past couple of months.