Galads Game Dev Thread

[TheOldMan]

Although I might use assembly for angled projectiles...

What do you mean by "angled projectiles"?
Things that move at an angle, or things that will 'hit' only on an angle?

If that makes sense...

[Galad]

What do you mean by "angled projectiles"?
Things that move at an angle, or things that will 'hit' only on an angle?

If that makes sense...

Yeah that makes sense and I should have clarified,Things that move at an angle.

[Arkhan]

lol doing everything in assembly is faster, so just skip C and go straight for what you're already comfortable with.

[TheOldMan]

Things that move at an angle.

45 degrees is easy. Just update X and Y.
Other angles....

if you limit the sprite position to 256 (ie, don't track sprite in world coordinates),
you can use the low byte of an int for a fractional amount. It should roll over into high byte when needed, and you can shift it into position for the sprite coord.

for world coords, use an int/char pair. use char to count steps, and bump the int when the steps are up.
It's easier to do than explain, so i suspect you'll have no problem


...and yeah, it's easier in asm.

[Arkhan]

It's called fixed-point, and galad probably already understands how to do that, lol.   If not, it's not horrible to deal with. 


That's what you end up using for floaty physics (like Atlantean)

[Punch]

45 degrees is easy. Just update X and Y.

Yes, just add 0.707 to X and Y.

Brought to you by the Citizens Against Faster Diagonal Movement Association of America

[Arkhan]

Brought to you by the Citizens Against Faster Diagonal Movement Association of America

This made me lol

[ParanoiaDragon]

lol doing everything in assembly is faster, so just skip C and go straight for what you're already comfortable with.

Yeah, I was thinking(as a non coder) ASM is the best way to go, though, I thought it was supposed to be more complicated maybe compared to C?  ASM is more stable & powerful or something, or am I just talkin' out my botox?  :-P

[elmer]

I gave you a working version of Squirrel that had been modified for the new HuC nearly a year ago.

Yea, kinda.   You changed register names, and I *think*(?) maybe made it so only vsync based timers worked with Squirrel.    It's been awhile since I looked since I don't use your HuC.   Maybe this isn't the case.

Anyway, you're the one that broke the semantic versioning concept with your changes.   Should've called it HuC 4, lol. 

Don't put the onus on us to update something that you rendered not working.   Go ahead and give out/post your patched up version.   Just make sure you explain the who/what/when/where/why of it.

Actually, Uli Hecht was the guy that changed the zero-page parameter names for the assembly variables, in order to stop them from conflicting with regular C variable names.

It was one of his many, many improvements to HuC.

I just changed his variable names again to make them consistent with the standard C naming convention for internal variables.


You are right, it would have been rude for me to have expected you to make lots of changes to Squirrel in order to have it work with the new HuC.

So, I didn't.

Instead, I took the time to look at Squirrel, and made a bunch of small modifications to it so that the same code would work with both the old and new versions of HuC.

Then I gave the changes to you, and asked you to make the single-character change to your proprietary, closed-source Squirrel MML converter tool, that would make the whole package releasable for folks that wanted to use the same version of Squirrel interchangeably with both the old and new versions of HuC.


Since, a year later on, you have indicated that you don't wish to do that, and have now given me permission to openly distribute the modified version of Squirrel, then I can just write a small filter program to change the output of the Squirrel MML converter so that it will work.

If someone really wants to use Squirrel with the new version of HuC, then I will have a reason to do that.

That's where you're likely going to get into the whole "open source/licensing/omfg what now" hoojoo, but it's like, this is shit about 6.5 people are using in the grand scheme of things.

Huh??? Where did this come from???

We settled the whole Squirrel/Open-Source question a couple of years back, and I totally respect your wish to keep your code private and closed-source.

I don't particularly like it, but I do totally respect it ... it's your code to do with as you wish.


But it's that decision, plus the appearance of all of michirin9801's wonderful PCE DefleMask tunes, that spurred the creation of Huzak.

As you and others repeatedly point out, MML is so 1980s, nobody is going to use it, and Deflemask+your new sound engine will stop judgment day, so what's the point anyways?   

DefleMask+Huzak definitely won't "stop judgment day", but they will certainly give PCE musicians and developers an alternative to Squirrel.

Are you trying to discourage developers/musicians from using Squirrel, if they wish to use the new version of HuC?

[Arkhan]

Actually, Uli Hecht was the guy that changed the zero-page parameter names for the assembly variables, in order to stop them from conflicting with regular C variable names.

It was one of his many, many improvements to HuC.

I just changed his variable names again to make them consistent with the standard C naming convention for internal variables.

Ah, well, this is where it should have been changed to HuC 4, as it breaks compatibilty with anything prior.

Fortunately, as mentioned, we have about 6.5 (6.75?) people doing devwork, so it's not going to cause some great uproar or confusion really. 

You are right, it would have been rude for me to have expected you to make lots of changes to Squirrel in order to have it work with the new HuC.

So, I didn't.

Instead, I took the time to look at Squirrel, and made a bunch of small modifications to it so that the same code would work with both the old and new versions of HuC.

Then I gave the changes to you, and asked you to make the single-character change to your proprietary, closed-source Squirrel MML converter tool, that would make the whole package releasable for folks that wanted to use the same version of Squirrel interchangeably with both the old and new versions of HuC.

Correct us if we're wrong, mainly OldMan as he looked at it last, doesn't it (Squirrel) not work with anything but vsync timers using the new HuC?  If that's the case, it severely limits it's functionality to the point where there isn't really a point in using it.   The vsync timer sucks.

As for a single character change to the conversion utility, either there was a miscommunication (likely) or there is more to it.  I don't recall much of that so that might have been between yourself and OldMan.

That was all happening while I was finishing Inferno, so I wasn't doing anything PCE related except posting and occasionally reading stuff.   

I try to work on one thing at a time so they get finished.  I don't like bouncing around.

Since, a year later on, you have indicated that you don't wish to do that, and have now given me permission to openly distribute the modified version of Squirrel, then I can just write a small filter program to change the output of the Squirrel MML converter so that it will work.

If someone really wants to use Squirrel with the new version of HuC, then I will have a reason to do that.

It's not like you needed permission, since you just admitted that you'd been handing it out anyways, lol.

Huh??? Where did this come from???

I don't know.  You just have a tendency to get overly pedantic at the strangest of times.

But it's that decision, plus the appearance of all of michirin9801's wonderful PCE DefleMask tunes, that spurred the creation of Huzak.

I really doubt the non-open nature of an MML to PCE conversion utility is what spurred the creation of Huzak, lol.   I'm certain it's just the appearance of covers of chiptunes on PCE with no means to play them on real hardware.   If the Deflemask stuff wasn't being pumped out for PCE, I really doubt you'd have bothered simply because of Squirrel/MML.   I figure you would've just made fun of MML, and done nothing else really.

DefleMask+Huzak definitely won't "stop judgment day", but they will certainly give PCE musicians and developers an alternative to Squirrel.

Definitely.  It might be the first time tracker-related chiptuning on the PCE is released in any functional capacity.

Are you trying to discourage developers/musicians from using Squirrel, if they wish to use the new version of HuC?

No, lol. 

MML discourages people in the first place.  Not sure if you noticed, nobody really uses it anyways because everyone's basically telephone gamed the woes and fears of the dreaded MML to the point where nobody can be bothered to even actually try it.

I was just operating under the assumption it doesn't really work right so no point bothering. 

If it's something that is easy to sort out, now that I'm not doing an MSX project outside of gluing things together in a poorly ventilated room, I could f*ck around with it.

[nodtveidt]

MML isn't hard unless you're afraid of big scary letters.

[Gredler]

I've tried MML  a bunch, and made a "song " and sound effects with it, but i am not a sound guy let alone a musician... yet


That being said, I don't care if it's huzak or aquirrel, as long as it works and the avg joe musician can crank out some tunes with it.

[Arkhan]

MML isn't hard unless you're afraid of big scary letters.

TRIGGGERRRRRRRED

lol

[Galad]

TheOldMan,thank you and I appreciate the info but I'm going to go for fixed point implementation in asm as that is easier for me.Your comment wasn't a waste though as I did learn something from you.

[Galad]

It's called fixed-point, and galad probably already understands how to do that, lol.   

Indeed I do

lol doing everything in assembly is faster, so just skip C and go straight for what you're already comfortable with.

Arkhan I don't know why I didn't listen to you as that's the second time you've suggested that,If I would have just listened I could have saved myself hours,instead of screwing around with what I was doing.I lost development time over this.And yes my code  is working 100% now
in assembly,Thanks for the help gentlemen.