Scorpion Engine

[earok]

Originally Posted by Cobe

Well its not like Mixel is coding the game itself without game maker engine and someone's gonna tell how couldn't you make it run on stock A1200, Jim Power runs on A500..........
If Commodore lived a year or 2 more, fast ram for A1200 imho would become mandatory like 512k for a500 and it not only adds memory but doubles the speed.
And most of us "active" cd32 users have at least tf328. Look at the lists on Amibay..

Actually.. yeah I think.. you've just got to make the game for whatever setup you feel like. If you aim at an OCS Amiga, people will ask for AGA enhancements, and if you aim at an AGA Amiga, people will ask why it doesn't run on OCS or why it isn't enhanced for Vampire. Can't keep everyone happy. I vaguely even recall someone complaining about a well optimized 512K OCS game just because it wasn't also NTSC compatible.


For what it's worth, I really want Scorpion's performance to be comparable to native games. To the point where it really should be able to make a Jim Power like-game for the same requirements even if it doesn't quite push the hardware the way only a custom one-off engine could.

[malko]

Thank you for this vision @earok

[Tigerskunk]

Originally Posted by earok

If you aim at an OCS Amiga, people will ask for AGA enhancements, and if you aim at an AGA Amiga, people will ask why it doesn't run on OCS or why it isn't enhanced for Vampire.

Exactly.

Originally Posted by earok

Can't keep everyone happy. I vaguely even recall someone complaining about a well optimized 512K OCS game just because it wasn't also NTSC compatible.

and this.

If I have learned one thing, you will never make everybody happy in the Amiga community, it's just too diverse in its hardware setups and preferences.

Just do what YOU feel like is your vision for an Amiga game, whether that's a OCS Amiga1000 game without any additional RAM or a game for Vampire accelerators or whatever.
YOU need to be happy with it and feel it's fun for you to work on it..

And then people can follow, and get your game, or they don't.
Just don't give a fuck, and it will work out anyway.

[nikosidis]

This is why C64 is much more popular than Amiga. Even if there is C128 it is almost no games made for that config and that is only more RAM! Think about Amiga with all the different hardware! It is crazy and actually not a good thing regarding interest in developing for the platform. A lot have found out that A500 with 1MB is what most have and most aim for. More than A1200 with standard 2MB chip is at least asking for almost no sales and very little interest.

What can be done and done before is to make some extras if extra ram is detected.

[Tigerskunk]

Originally Posted by nikosidis

This is why C64 is much more popular than Amiga. Even if there is C128 it is almost no games made for that config and that is only more RAM! Think about Amiga with all the different hardware! It is crazy and actually not a good thing regarding interest in developing for the platform. A lot have found out that A500 with 1MB is what most have and most aim for. More than A1200 with standard 2MB chip is at least asking for almost no sales and very little interest.

What can be done and done before is to make some extras if extra ram is detected.

The only problem is really that you need a strong idea of what you want to create, and then just follow through with it.

I have heard sooooo many things and lamentations over the years. But in the end, if people would really want to play your game they'd find ways to.

"Could your game please be..."
* NTSC
* AGA
* WhdLoad
* hard disk installable
* for 68060 CPUs
* for RTG cards
* with more dithering
* with less dithering
* more like game XYZ (most often "Turrican", regardless of if your game is a SHMUP, Run&Gun, or a text adventure)
* less like game XYZ
* faster
* slower
* harder
* less hard, need a health bar or I cannot play it (which comes back to "can your game please be more like Turrican")
* with music that's less shit
* 14 bit music streaming ("Oh, you release a game on floppy? That's shit, please release on CD, go with the times on this 36 years old retro platform. So it's less like 1992, and more like 1994. So modern!")
* with some special enhancements that targets something my exotic setup and nobody else has, but will cost you a full two weeks to develop
* fill in your own complaint here

In the end the answer is:

no

[Mixel]

You guys are the best btw I sometimes need to hear this stuff apparently, haha.

Is anyone solely making a living from indie retro dev anyway? I'd have thought it's all just labours of love regardless of platform?

The diversity of Amiga is a boon to me.. If everyone targeted the low end we'd be seriously hobbling the potential for original games (not just ports from other platforms). When I originally "left" the Amiga platform I had an A1200/040, if you'd told me, in the late 90s, that I'd be making Amiga games in 2021(!) but for some reason I'd be targeting the 1MB A500 I'd abandoned years ago by that point (of all things) I'd think you were crazy, haha. My dream project is not a game for an unexpanded A500+, and for me to keep momentum I have to love what I'm doing... Market size isn't even part of the equation. If I wanted maximum reach I wouldn't be making it for *any* retro platform, I'd be bludgeoning my head against a wall trying to make it work in Multimedia Fusion or something. (and abandoning it early in the process most likely)

IMO Scorpion has the potential to be a way bigger deal than folk realise. From my POV it's becoming the best tool of it's type for making certain styles of game *for any platform* (for a certain subset of creators) and by some miracle it just happens to export Amiga games. If it outputted games for other systems I'd still rather use it than any other system I've tried, and I've tried a lot.

LOL that sounds embarassingly Scorpion evangelical, but I can't think of any other system that gives this balance of ease of use/results/performance. Even if Scorpion stayed at its current optimisation level and people used it to make Vampire/Pi/Mister/UAE games in the same way people make Pico8 games it's a stupidly nice tool already.

Originally Posted by Steril707

The only problem is really that you need a strong idea of what you want to create, and then just follow through with it.

This is great advice TY. Sharpening resolve.

I've already had almost half of those suggestions(usually more suggestions than complaints, though I've had a couple of those too) and people have seen less than 5% of the game.

[nikosidis]

Mixel: It is fantastic to hear the way you think about Scorpion Engine. I also have the feeling it is something very special and it is great it is just for our platform
Most of us been there with 040, 060 or NG Amiga systems. It is just not where our system does anything special. The special about Amiga, C64, Spectrum is in its limitations. That is what make it original compared to modern stuff. The low res. where you can see the pixels. The limitations in colours, sound.
Market size is not huge but for sure growing. Sam's journey on C64 sold around 3000 copies and the price is 20 Euro for digital download, boxed versions more. Most developers do retro development for fun but most like to target specs. people have. The more feedback and money there is in something the more quality we will see.

[Tigerskunk]

Originally Posted by Mixel

You guys are the best btw I sometimes need to hear this stuff apparently, haha.

Is anyone solely making a living from indie retro dev anyway? I'd have thought it's all just labours of love regardless of platform?

It is a labour of love. Even more reason to just do what you want to do regardless of the lamenters.

Originally Posted by Mixel

The diversity of Amiga is a boon to me.. If everyone targeted the low end we'd be seriously hobbling the potential for original games (not just ports from other platforms). When I originally "left" the Amiga platform I had an A1200/040, if you'd told me, in the late 90s, that I'd be making Amiga games in 2021(!) but for some reason I'd be targeting the 1MB A500 I'd abandoned years ago by that point (of all things) I'd think you were crazy, haha.

Well, if that's your point of view of why YOU target that specific range, then that's good for you.
I'd just add, that the reason why people are doing game dev on these old machines now doesn't need to be the reason why people did it back then.

Originally Posted by Mixel

My dream project is not a game for an unexpanded A500+, and for me to keep momentum I have to love what I'm doing...

Yep

Originally Posted by Mixel

Market size isn't even part of the equation. If I wanted maximum reach I wouldn't be making it for *any* retro platform.

Yep, it's a labour of love.

Originally Posted by Mixel

IMO Scorpion has the potential to be a way bigger deal than folk realise.

Thinking the same here.
If this follows through with what is already possible on it, I see most Amiga game dev happening on it over the next years.

[earok]

Thanks for the kind words @mixel @nikosidis @malko @steril707, I really hope that Scorpion lives up to the promise you guys see in it!

And yeah, @steril707, really sorry you've had to face that too :/

@Mixel Interesting you brought up multimedia fusion! The Klik N Play series is one of the main inspirations for Scorpion, I never really got into Multimedia Fusion but KNP and TGF were what I swore by before I learned to code.

----

Anyway, experimental update for the day!

The "parallax distance" property has been replaced with "parallax scroll rate". This is a floating point value so you've got much more control over it than before.

- A value of 0 won't scroll the background at all, it'll stay static.
- A value above 0 will scroll relative to the camera movement speed (it's possible to scroll faster than the camera, which may be useful for foreground parallax effects)
- A value below 0 will scroll backwards (See: Jim Power SNES). Maybe not a great thing to use but it's there.

A good default value is probably the player x speed divided by 2.

Note that the one caveat is, the scroll rate is fixed and it scrolls whenever the camera moves left or right in a frame. So it might look a little weird if you've got a character that accelerates - I'll need to do some testing/experimentation there.

Over the weekend I think I'll look at adding a couple more features to parallax, and then releasing an official update. I don't think I can include the DaemonClaw demo but I can certainly add parallax to Amigo the Fox so there's at least one working parallax demo in the pack.

[Mixel]

Originally Posted by Steril707

Well, if that's your point of view of why YOU target that specific range, then that's good for you.
I'd just add, that the reason why people are doing game dev on these old machines now doesn't need to be the reason why people did it back then.

Oh completely, my reasons are really changeable anyway. Haha. “90s me” would’ve been wrong to think someone developing A500 games in the distant future is crazy btw I just think it’s funny in retrospect. Alll the respect to people targeting the low end! Particularly some of these crazy arcade ports we’ve been getting lately!

Thinking the same here.
If this follows through with what is already possible on it, I see most Amiga game dev happening on it over the next years.

I think it’s already beginning, but obviously there are a lot of genres it can’t really do at all.. Something really interesting I see it could lead to; an uptick in new devs who have very different ideas re: target config than us Amiga old timers would think sensible.. like WAY beyond my “might potentially need a bit of fast ram?” and that will risk attracting the ire of a bunch of classic fans. If someone wants to make eg: a Caveman Ninja alike, or Castlevania SOTN scope game, or something Metal Slugish (eg games that wouldn’t have happened well without a PS1 or NeoGeo etc), but it requires an 030+ and 4MB of ram I really don’t think anyone should be discouraging them. That wouldn’t be a failure on the dev’s part, or scorpion’s. The games would still have all that awesome Paula and OCS/AGA goodness and feel like Amiga games, just delivering on the somewhat unfulfilled potential for next gen Amiga games.

Originally Posted by earok

@Mixel Interesting you brought up multimedia fusion! The Klik N Play series is one of the main inspirations for Scorpion, I never really got into Multimedia Fusion but KNP and TGF were what I swore by before I learned to code.

I like MMF but it never meshed with how my mind works like scorpion or PGF, it’s sooo modular and weird. It’s still a possible candidate when it comes to porting CMO away to Win/Mac but a tiny part of me hopes it takes me so damned long to finish Amiga CMOHN that by the time I’m done Scorpion might compile to windows, lol. (No pressure, haha) I’ve also tinkered with Godot, Game Maker, App Game Studio, Virtools, Pixel Game Maker MV, Construct, and never got anywhere significant with ANY of them. ? (and that’s probably why I come across as some sort of Scorpion cultist now, lol) - I mean Scorpion doesn’t even have any documentation - or a big community of people making tutorials etc - if I, a total programming dunce can work this much of it out anyone probably can, and Earok, Tsak, anyone else involved in it should be incredibly proud of what they’re doing.

- A value below 0 will scroll backwards (See: Jim Power SNES). Maybe not a great thing to use but it's there.

Hahaha you actually added it! Thanks! Going to use this on some bosses and particularly cursed areas, it’s so disorienting I love it, lol.

[earok]

I haven't done much on MMF myself, I think maybe the reason I couldn't get into it was because it seemed a lot more complicated than KNP and TGF (which were specifically designed for making games as opposed to general apps).

Anyway, so.. I was planning on looking at releasing the next update over easter since I haven't had any major bug reports in a few days, but I've kind of made a complicated mess of things.



Line by line parallax is working and IMO it's pretty neat, but it's far from finished. The copperlist has way too many unnecessary instructions, foreground is a bit broken and a bigger problem is adding some sort of editor for it. I guess I could do something extremely rudimentary for this release and work on a tidier visual editor for a future release.

[Mixel]

“IMO it’s pretty neat” hahah understatement much? Rudimentary editing is great too!

Totally unsolicited idea and you probably know exactly how it should work already, but.. If making a visual editor would be complicated could you load a separate png of the same height as the bg image with just pure black with white pixels to denote where slices would start/stop? That’d be a nice easy way to set up and test the slices in a graphics app instead of scorpion itself? (Genuinely may be a really dumb idea, I don’t know!) then it would just display the number of speed selector boxes as read from the slices file? I have no idea how the foreground ones would work though.

Going to look incredible in practice, so exciting!

[earok]

Fixed up the foreground scrolling and added the ability to have non-repeating images (eg, the moon). This is all with two sprites, so there's still six available for other things.

Every time I work on parallax I get a new idea for improving parallax more. So I'm still not done, but I'm getting closer. I've got one more big idea for really making it shine, and then I think that's about as far as I can push it. Maybe another week before all of the parallax stuff gets rolled into the official release.



For configuring it, I think it's just going to have a CSV "spreadsheet" with values that you can tweak. Since there's a fair few things you may want to tweak (how fast it scrolls left, how fast it scrolls right, if it scrolls automatically even if you're not moving, if it's background, if it's foreground, if it only uses part of the screen like the moon and if so, where on the screen it is etc) and those values might be different from one line to the next.

[earok]

I think this is going to be my last experiment on parallax for awhile.

You can see there's now multiple overlapping layers of mountains and spikes. But the addition of this comes at an additional RAM cost, so I'm not totally happy about it.

The technical explanation is this - before, 8 copies of the background were generated to create the background parallax, in addition to using SPRxCTL to nudge by 1 pixel (see: Codetapper's analysis of Risky Woods).

What I've done here is disabled the SPRxCTL trick and created 16 copies of the background, each of which can be manually edited. (eg, the back layer of mountains is manually painted to be in the same place on all 16, whereas the front layer of spikes is manually painted to scroll 2x pixels across each time).

In theory this could be exploited to do rain, snow and other weather effects, or maybe animated parallax backgrounds like the gears on that mechanical level in Apidya, or maybe waves that bob up and down.

But anyway, that's enough for now. Going to take a clean break for a bit before I tidy it up for experimental branch release.

[viddi]

Wow, simply stunning.

[nikosidis]

Hehe, sick that what is possible with this program now!! Whow.

[spoUP]

Very very cool Earok!
Progress is very fast!

[Tigerskunk]

Everybody, get into Amiga game dev asap!

[malko]

^ cool and impressive, yes
One tiny question @earok : Since the initial post with the moon, the moon is aligned with the hero : centred on the screen).
Can it be anywhere on the x axis ?

[earok]

Cheers everyone

I need to do a little more work on it, but it should be possible both to position the moon anywhere or make it scroll/wrap around the sides of the screen.