|
Post by paranoiadragon on Jul 3, 2022 8:58:08 GMT
I was recently reading an old thread from 2007 on Sega16 that started out being about why the Mega Drive in Japan failed. Joe Redifer, Black Tiger, and Nat were amongst it. It eventually changed into "blast processing" and Sega enthusiasts saying Sonic 1 couldn't be done on the Turbo. It'd maybe have a few more colors, but no parallax and be slow even though the main CPU is just a hair slower than the Genesis.
I was working on a Sonic project at one point, maybe with Touku? It's been quite a number of years when we were originally doing that. The only thing I got to do was the music for the title screen and Green Hill zone. Though I do wish I could go back and adjust the title screen, as there was an instrument or two that I'd like to make a little bit louder. And my title screen I think I would like to change the main instrument for the melody.
At the time, whoever was I was working with, was using assets from the Master system version. I'm actually not entirely sure why. Surely, the graphics as is on the Genesis could have been replicated even with no parallax whatsoever. Although I do know he wanted to do parallax, there was a bit in his demo. There might even be still a video of it up on YouTube but I haven't been able to find it.
One of things I recall him saying was that the Turbo can't do tile or Sprite flipping which adds to the resources that are needed to make a Sonic game. So I can only presume that it would take extra frames of animation to flip things.
Another one of my understanding is that the turbo is quite a bit faster when it comes to how many millions of instructions per second it can do or MIPS. I did a little bit of googling, but couldn't find exact numbers for the Genesis with what little bit I did. But it seemed like the turbo could do maybe twice as many mips. With my limited understanding, I thought that the turbo could zip sprites and or tiles very quickly across the screen due to the MIPS, like what is seen in a multitude of Shmups on the turbo.
I would think that the turbo CD along with arcade card could easily replicate the base game without the parallax, but probably also provide enough resources for any tricks needed to include parallax of any sort as well as Sprite and tile animation. I would maybe think that the hills in the background could be static. Though I'm not sure if overlapping of the hills and trees and such would require all of them to be sprites. I know Ninja gaiden is not a good example because it does have jerkiness, but it does have buildings overlapping the background. Then there's I believe the clouds and then of course the parallax of the water, I think is called h-int if I'm remembering right.
Am I right, that the base game could easily be done with no problems of any slowdown or anything on the turbo in general? Not only that, but be more detailed at least with the amount of colors on screen?
I bring this up here because I really do think it would be great to provide at least a proof of concept, a Green Hill zone demo that shows the game running just fine on the turbo along with this expanded colors. Maybe the rings scattering all over the place could be a problem but then again maybe not if they're quickly alternating between disappearing and reappearing?
Would anyone be willing to work on such a demo? I give them my already finished music of course. In a perfect world I'd have no problem in finishing the entire soundtrack but real life has been busier these days and I have some problems with my midi keyboard that have kind of held me back from making more music lately. I'm still not sure if I just need to get it repaired by someone or if I can maybe pop it open and take out perhaps a lithium battery that would end up resetting the entire machine beyond what the factory reset can do. In a way that would be nice if that's all it took instead of me having to bring it to someone somewhere.
I might mention that it was stated that the turbo cannot do Castle of illusion or some other game that starts with an o that I can't remember right now. It's not a game I'd ever heard of before, it was something like Ozzimatli or Osomaru or something. Personally, I think it will have absolutely no problem pulling out Castle of illusion. Obviously there's the always the problem of figuring out the right tricks for parallax, but otherwise I would think it would be easily doable and end up looking better. Especially seeing as how that's in the earlier game for the Genesis I don't recall anything earth-shattering about it. No real tricks involved in that game that I can recall.
|
|
|
Post by gredler on Jul 3, 2022 17:49:23 GMT
Would anyone be willing to work on such a demo? I give them my already finished music of course. Interesting proposition, esp given the recent modern platform ports. What are you using to make the music? Is it for CD only or one of the HuC packages (MML/squirrel, or deflemask)?
|
|
|
Post by spenoza on Jul 3, 2022 17:55:39 GMT
The multi-scrolling background layers would be a challenge, certainly, and not being able to mirror tiles or sprites vertically does increase data sizes, but there would be no issues with processing speed or sprite handling. It might be easier to work from Master System level designs simply because the PCE can’t layer sprites all the way across the screen in 320 horizontal resolution. Switching to the Supergrafx would certainly ameliorate a lot of the issues.
|
|
touko
Punkic Cyborg
Posts: 106
|
Post by touko on Jul 4, 2022 7:46:09 GMT
paranoiadragon ,no it was not from me but from Xav, another french guy who left the dev scene some years ago now .
|
|
|
Post by paranoiadragon on Jul 4, 2022 8:22:29 GMT
I do Redbook audio. I haven't learned how to do chip tunes yet. Maybe someday. I know Fragmare, who I usually think of for graphics, has done some great chip tunes.
If we remove the vertical parallax (I think some have been calling it Ant farm parallax), would that help? If I remember right, vertical parallax is more difficult to pull off. Though, tons of Shmups pull it off, but I believe forced scrolling makes that easier. And you don't think the ring sprites would be a problem?
The main issue was that the regular PCE couldn't handle the game. Of coarse, if someone wanted to make a demo that's bi-compatible, that'd be kinda rad, but obviously extra work. Also, I was just remembering, isn't parallax easier to fake when overlapped by squared off sprites or tiles, as opposed to being rounded off? I recall something about that.
Of coarse, the other thing that puzzled me was the claim of Castle of illusion not being possible. Once again, parallax comes into play. Would need tricks for any possible parallax, but otherwise I didn't see anything that the TG can't pull off without problems. Heck, I would imagine the choppiness of when Mickey swings on a vine could easily be fixed, don't know what happened with that in the original. The parallax itself is usually only 1 layer so it might be just a matter of figuring out a way to maybe simplify the bg, less unique dynamic tiles, no vertical parallax, etc. if possible.
Oh ok, Xav, I couldn't remember. I knew who I worked with was in France. It was maybe 10 plus years ago that he was working on that demo. I wonder why Xav left? That's a shame.
|
|
touko
Punkic Cyborg
Posts: 106
|
Post by touko on Jul 4, 2022 9:18:52 GMT
Unfortunately, i don't know the reason,may be health problems or something similar .
|
|
|
Post by turboxray on Jul 4, 2022 17:57:23 GMT
I looked into this quite a few times over the years. Specifically Green Hill Zone, because the grass is not a flat plane.. but small variances/dips/etc. And the grass itself 'spikes' up over the far BG layer. That presents a real challenge for PCE's parallax tricks/methods. I found the best approach is dynamic tiles for the far layer, and normal tiles for the front layer.. augmented with sprites for the edge transitions (and things like rocks, trees, flowers, see-through waterfalls, etc). The dynamic tiles needs to be really well thought out, and needs to be line precision relocatable (this is not an issue, but you'll have to do multi-pass method combined with VDC write inc of 32, and stagger your tiles to work with that). The dynamic tiles need to be ST1/ST2 opcodes, so it's fastest and it can take advantage of just using ST2 (skipping ST1) for even faster speeds. Areas like the water would really benefit from this, and any open solid color areas. You rebuild the dynamic tiles EVERY frame... just to give an idea of how complex and fast this needs to be. The dynamic tiles are done in "groups", which means it's not a simple pattern across the screen which is nice, but you also have to update the tile map for placement. I wouldn't start with the master system one. I would take the entire stage layout of the Genesis version, and shrink it to 80% horizontally to correct for PCE's 256px PAR (pixel aspect ratio). This will give you the correct aspect ratio, because when the real PAR for low res 256px kicks in, it will be correct to the Genesis version (for NTSC, not PAL). From there, using it as a base, start reworking the stage "tiles" as chunks, in the layout. Start with simply blocks, and then refine. TiagoSC, who did the SNES port, should have taken this route. Instead he just used all the assets as-is of the Genesis port. Dealing with the grass issue... I would divide the sections into 32px wide like so: With the blue being tiles that you manually mask against the tiles used in dynamic tiles, and the pink area is sprites. You can't use all sprites here because Sonic's overlapping sprite would cause drop out. Using sprites for half of it means you would cut down the time it would take the CPU to manually AND/OR the few lines of all those tiles. Otherwise, a compromise would be to have grass tops that are flat edges and not spikey. They could still have dips here and there (dips would be the sprite areas). Lastly, I would clip the horizontal display to 240px just in case there are areas that need full edge to edge scrolling (later levels, not green hill zone, that have flat areas). There are a number of SNES games that clip from 256px to 240px. It's not a big deal since that was typically in overscan area anyways on a consumer set. Of course things like HUD and such will need to be put into a solid HUD area. I think that's worth it for impressive parallax. It's an ambitious project, and needs lots of highly optimized assembly, special tools, etc. But I've crunched the numbers and did proof of concepts for the ideas mentioned, and it's doable. It's also a lot of work. I think the question needs to be asked though: Do you really want to see a Sonic port/demo of Green Hill Zone on the PCE, or would you rather have a new Bonk demo that does all of this kind of stuff??? Because after a Green Hill Zone demo, which would be definitely nice, then what? The whole game? People are going to eventually forget about it (the novelty wears off), and/or demand you finish making the game/port. I mean, the game already exists on the Genesis. Outside a nice tech demo, I rather see something like above methods applied to a new Bonk style game (or something original). And this is why I personally have had to stop myself from doing a Sonic demo with all the above mentioned stuffs as much as I would love too (given I have severely limited free time). I learned from the NES2PCE stuff, which took up a lot of time, is in the end just novelty.
|
|
|
Post by DarkKobold on Jul 4, 2022 22:51:09 GMT
I think the question needs to be asked though: Do you really want to see a Sonic port/demo of Green Hill Zone on the PCE, or would you rather have a new Bonk demo that does all of this kind of stuff??? Because after a Green Hill Zone demo, which would be definitely nice, then what? The whole game? People are going to eventually forget about it (the novelty wears off), and/or demand you finish making the game/port. I mean, the game already exists on the Genesis. Outside a nice tech demo, I rather see something like above methods applied to a new Bonk style game (or something original). And this is why I personally have had to stop myself from doing a Sonic demo with all the above mentioned stuffs as much as I would love too (given I have severely limited free time). I learned from the NES2PCE stuff, which took up a lot of time, is in the end just novelty. I'll be honest, there's a ton more important stuff that really could use your skill set. This seems like a massive distraction from all of the big plans, like HuTrack, TGDK, and mapper support in HuC/TGDK. This seems like a total waste of time, just to prove the PCE could do it. And why? To play Sonic on a 2 button controller instead of 3 buttons? It wouldn't actually use any of the advantage of the TG16, because you aren't going to redraw the art to use more colors. You've already demonstrated an engine that uses parallax, but no one can use it because it isn't released.
|
|
|
Post by turboxray on Jul 5, 2022 0:12:19 GMT
I think the question needs to be asked though: Do you really want to see a Sonic port/demo of Green Hill Zone on the PCE, or would you rather have a new Bonk demo that does all of this kind of stuff??? Because after a Green Hill Zone demo, which would be definitely nice, then what? The whole game? People are going to eventually forget about it (the novelty wears off), and/or demand you finish making the game/port. I mean, the game already exists on the Genesis. Outside a nice tech demo, I rather see something like above methods applied to a new Bonk style game (or something original). And this is why I personally have had to stop myself from doing a Sonic demo with all the above mentioned stuffs as much as I would love too (given I have severely limited free time). I learned from the NES2PCE stuff, which took up a lot of time, is in the end just novelty. I'll be honest, there's a ton more important stuff that really could use your skill set. This seems like a massive distraction from all of the big plans, like HuTrack, TGDK, and mapper support in HuC/TGDK. This seems like a total waste of time, just to prove the PCE could do it. And why? To play Sonic on a 2 button controller instead of 3 buttons? It wouldn't actually use any of the advantage of the TG16, because you aren't going to redraw the art to use more colors. You've already demonstrated an engine that uses parallax, but no one can use it because it isn't released. I'm not talking about me, specifically, I'm talking in general. I'm just using myself as an example. There are other people with similar skillsets and little-to-no-time as well. It's a common trend; "time is a precious commodity". It's an honest question to ask, though, given how little is happening in the PCE DEV scene. But it seems to be all rage/trend in the retro scene lately (going way beyond Genesis/MD stuffs). People want remakes and ports. To play devils advocate, if you develop such an engine - you can use it for something else. Nothing says you can't develop something else with it. But the hardest part of a Sonic demo isn't the parallax or any of this other stuff.. it's implementing the exact controls and physics - IMO. That's the no-fun grunt work. HuC is not going to cut it. Because no one has asked about using it haha, so there's no rush to make it public. I even asked around.. people directly (privately), got an "ehh.. mayyybe". I mean, that's totally fine. I made it for my own projects first and foremost, but I mean if easy parallax for the PCE doesn't get people hyped about PCE dev.. I'm not sure what else will. Kind of got the feeling anything else related to Huc/TGDK will be the same. So it's more like, is HuC/TGDK more of a distraction to what I should be doing?
|
|
|
Post by DarkKobold on Jul 5, 2022 0:48:40 GMT
"Because no one has asked about using it haha, so there's no rush to make it public. I even asked around.. people directly (privately), got an "ehh.. mayyybe". I mean, that's totally fine. I made it for my own projects first and foremost, but I mean if easy parallax for the PCE doesn't get people hyped about PCE dev.. I'm not sure what else will. Kind of got the feeling anything else related to Huc/TGDK will be the same. So it's more like, is HuC/TGDK more of a distraction to what I should be doing?" I mean, this is getting way off topic, but I think most people who approach TG Dev and see that MML/squirrel is the only public music option scares away the general GBDK/SGDK type user. I feel like we've had a lot of people come and go lately because of that exact reason. Getting general users excited about TG development really requires finishing up HuTrack. EDIT: I can not get the formatting of the quote right. Sometimes these boards just crap the bed.
|
|
|
Post by gredler on Jul 5, 2022 1:12:29 GMT
Yeah I would agree that the current available tools are basically incomplete because of the lack of sound support outside of MML on old versions of HuC. The sound engine barrier is a major one for completing projects.
To bring this back on subject that is why I was asking OP if the completed Sonic tunes were hucard or CD, as that affects how much data can be loaded at once without a load pause which Sonic levels may be easier to create as hucard, no?
|
|
|
Post by paranoiadragon on Jul 5, 2022 6:27:15 GMT
Yeah, switching to another track like invincibility or 1-up would require chip tunes, or maybe adpcm? The entire game having to stop and load just to do the invincibility or 1-up I feel would be unacceptable. It's one thing to have a game load a boss at the end of a level, like say, Gate of Thunder. But it's another thing to have a game load in the middle of a level, and then have to load again back to the original music. I guess Strider jumps back and forth between red book audio tracks, and it's not too annoying. But I think for a game as fast paced as Sonic, it would really disrupt the gameplay.
And yeah, I'd much rather see a demo of Bonk over Sonic any day of the week. I'm honestly not a fan of the Sonic games, but I think they have decent music. The two things that irritate me most about the games is the rings scattering all over the place, I think that's where my OCD comes into play. And then also where you have to breathe air bubbles. Although probably the music that goes along with running out of air greatly frustrated me.
I'd love someone to take up my original idea that me and Black Tiger were working on for a rebuilt version of Super Bonk, which I was calling Bonk to the Future, (since at the very beginning of the game there's a cinema showing him being warped into the future). It's actually something I didn't realize for years until I let the title screen sit there for a while and then it went to that little Cinema. I've always felt like that game was one of the lowest points of the series, between the music, the darker color palette, and the gameplay in general. We theorized that it would improve the game to have a faster spin. Although I really do feel like better music would make the game more enjoyable as well. I can't overstate how much audio contributes to an enjoyable game.
Black Tiger did do some Sprite and tile work for it(stored on the Frozen Utopia forums, which is gone but he probably has them backed up), with us spitballing ideas of how parallax would work, but also redrawing the buildings to look better. Basically making them completely static in the background not unlike Ninja Gaiden but smooth. Graphically that's the only level that I recall us doing any sort of ideas on.
Musically I've already made seven or eight tracks of the game which can be found on SoundCloud under Paranoia Dragon. My intention of course was to make the game as much as possible sound more like the first two games, but with an enhanced sound chip. The music in the SNES game just doesn't have the spirit for the most part of the first two games in the series. Not to mention the sound set is quite different.
Any demo, or full game for that matter using these techniques would be beautiful and ultimately I would think be helpful for the dev community, and not just bragging rights. I completely brand new game is also something I keep in my mind as high priority of what I'd like to see, along with ports of arcade games that we never got. Like the arcade Contra with the layout of the NES game. Same goes for Super c but I think some of the levels in the arcade are quite different from the NES version. And then there's the remake we were making at Frozen Utopia of Golden Axe. Another game where any assets I could point to were on the Frozen Utopia forums. The music was all finished and ready to go, most of it being a combination of the Telenet music mixed with the PS2 music, but a few songs I redid from scratch. 80% of the graphics might have been ready to go if I remember right. Once again it was Black Tiger doing any kind of recoloring and such of the sprites and tiles. It looked beautiful. I know Nod said he'd still like to work on that no matter what someday though I haven't heard anything from him in a long time.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 6, 2022 21:37:38 GMT
I'd love someone to take up my original idea that me and Black Tiger were working on for a rebuilt version of Super Bonk, which I was calling Bonk to the Future, (since at the very beginning of the game there's a cinema showing him being warped into the future). It's actually something I didn't realize for years until I let the title screen sit there for a while and then it went to that little Cinema. I've always felt like that game was one of the lowest points of the series, between the music, the darker color palette, and the gameplay in general. We theorized that it would improve the game to have a faster spin. Although I really do feel like better music would make the game more enjoyable as well. I can't overstate how much audio contributes to an enjoyable game. That would be my dream game, if I ever knew how, but I wanted to start it off in the Badlands before you got into the city. One of the things I wanted to explore was turning some of the bonus level florets into Venus bonktraps as I found there were too many of them.
|
|
|
Post by paranoiadragon on Jul 7, 2022 4:27:51 GMT
That's true, I did want to somehow adjust how many times you do the bonus stages. It kind of made the game drag on. ATM, I can't remember if I had any specific idea on how to adjust, but Venus Bonktraps is a way to go. I feel like I did have some idea, it's just not coming back to me right now. I wanted to streamline them somehow.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 7, 2022 13:08:03 GMT
I've never managed to get through Super Bonk, personally. Do you enjoy it? I like the idea of Bonk in a modern city, steel structs sticking out from the place etc. Some of the stages I thought were well thought out, but others were too convoluted for such a slow-moving characters, that I wanted to see if I could fix them up on tiled or something. (Not that I necessarily think I would be good at that, but it might be a fun challenge).
Maybe you wanted more a straight port? Just using the old games' sound affects and the turbo's colour palette would be an improvement right there.
I didn't know there was a cinema, so I will have to try leaving it on the title screen.
|
|