[Mblackwell]

Is it important that it be per-frame and not per-tic? Either way you could fudge it by having a variable counted in EVENT_DISPLAYREST (which is updated per-frame) and one inside APLAYER (which is updated per-tic, also 30 tics = 1 second).

[Helixhorned]

 James, on 29 May 2012 - 12:38 PM, said:

Thanks for fixing the glitch with showview Helix, at last this means I can use skyboxes now with relative impunity

And thanks for writing stuff that manages to break EDuke32! By the way, it's ambigious how a showview should behave when the screen is tilted in the first place: both possibilities (sub-viewport has view tilted or not) can be useful in different circumstances, like when using it to draw a skybox to the full screen vs. a HUD element that shows the view from a camera.

 Fox, on 03 June 2012 - 04:42 PM, said:

Is it possible to add in Eduke32 a variable that holds the value of "how many frames already passed in the current tic"? For example, let's say the current framerate is 400, when half of a tic has passed this value would equal 200. This would be usefull so that I could create an interpolation code of my own.

To be strict, there's no such thing as the "current frame rate" defined (unlike "current velocity" of a physical object), since the events are discrete in time. What is defined though, as you suggest, is the number of frames drawn since a certain reference time. However, it's not clear to me how you'd get interpolation code out of that. For example, say F frames were drawn during a time corresponding to game tics N-1 and N, and the current time is somewhere in between. Then you have no way of knowing (only guessing) how many frames will be drawn until the time curresponding to tic N.
What's done in the engine instead is to interpolate the positions based on the 4-ticks-per-game-tic (120 Hz unless changed by inittimer) timer "totalclock". I would advise you against replicating it though, because you might run into clashes with hard-coded behavior. What kind of interpolation do you find lacking in EDuke32? The "jerky movement" of some actors might not necessarily fall into that bag, because the root cause of the jerkiness appears to be how the actor itself is moved.

[Fox]

I don't know for sure. Right now there is nothing specifically I want, but I remember I needed something back then. And I didn't knew that the actor interpolation was made based on 1/4 tic, I tought it was based on the framerate.

[Plagman]

As MB said, you just need to count one display event and one game event to achieve that.

[Jblade]

Music doesn't seem to play properly with the newest snapshot - any music tracks assigned to Episode 5 and onwards don't play. You can replicate this by playing either the multiplayer or the classic collection map in AMC TC (or any other Duke mod with more than 4 episodes defined)

[TerminX]

Blame Hendricks266 and his doubling of the max number of levels for that one... stuff like this is what happens when you blindly increase limits without thoroughly examining all related code just because somebody asked for it.

[Jblade]

 TerminX, on 04 June 2012 - 01:54 PM, said:

Blame Hendricks266 and his doubling of the max number of levels for that one... stuff like this is what happens when you blindly increase limits without thoroughly examining all related code just because somebody asked for it.

....would this be a bad time to mention it was me who asked for it?

[Paul B]

 James, on 04 June 2012 - 02:12 PM, said:

....would this be a bad time to mention it was me who asked for it?

One point for honesty! =)

[Danukem]

I have to say, one thing that I am enjoying about my break from modding is that I'm not the one encountering new bugs, reporting them, trying to isolate them and find workarounds. I feel that I more than paid my dues doing that shit and it was one of the reasons I needed a break. It can be very aggravating when you are working on a TC, you already have a ton of work to do and a limited amount of free time to do it in, and then you get held up by a breakage that isn't even your fault.

[Jblade]

It's fixed now, thanks for the speedy work once again TerminX!

[TerminX]

 Trooper Dan, on 04 June 2012 - 10:23 PM, said:

I have to say, one thing that I am enjoying about my break from modding is that I'm not the one encountering new bugs, reporting them, trying to isolate them and find workarounds. I feel that I more than paid my dues doing that shit and it was one of the reasons I needed a break. It can be very aggravating when you are working on a TC, you already have a ton of work to do and a limited amount of free time to do it in, and then you get held up by a breakage that isn't even your fault.

That's kind of the nature of the beast when you're working with development builds that are automatically generated after each commit. It might be irritating sometimes but it definitely allows for faster development and implementation of new features, and it improves the quality of EDuke32 as a whole.

[Danukem]

 TerminX, on 05 June 2012 - 11:15 AM, said:

That's kind of the nature of the beast when you're working with development builds that are automatically generated after each commit. It might be irritating sometimes but it definitely allows for faster development and implementation of new features, and it improves the quality of EDuke32 as a whole.

To clarify: I expect new features to have breakages, and it doesn't bother me very much. But when older features that were already working fine get broken, then that's aggravating, nature of the beast or not.

[Jblade]

Just a quick question/query - I just got reminded of that sprite tagging feature which is amazingly useful, but my only 'issue' is that I can't use it for the new sprites I coded for AMC TC effects and stuff...is there already a way I can add my sprites to the list of things you can give names to or would it require a new def line to be coded in or something?

Also speaking of effectors and stuff, I took a look at the M32 script section and couldn't see anything but is there also a way to add my new enemy/effect sprites to the 'hide sprites' command in Mapster? I took a look to see if I could just do it manually but didn't turn up much look.

[Helixhorned]

James, on 07 June 2012 - 12:54 AM, said:

Just a quick question/query - I just got reminded of that sprite tagging feature which is amazingly useful, but my only 'issue' is that I can't use it for the new sprites I coded for AMC TC effects and stuff...is there already a way I can add my sprites to the list of things you can give names to or would it require a new def line to be coded in or something?

EVENT_LINKTAGS is your friend.

Quote

Also speaking of effectors and stuff, I took a look at the M32 script section and couldn't see anything but is there also a way to add my new enemy/effect sprites to the 'hide sprites' command in Mapster? I took a look to see if I could just do it manually but didn't turn up much look.

I think there's no way currently, but that should be doable in a similar fashion to the above.

[Jblade]

Cool, thanks Just another question you might have the answer to - I'm currently rewriting some GUI CON code stuff for the AMC TC, and I know multiplayer is broken at the minute but I don't want to write stuff that will screw things up when it's working again. I take it I still have to seperate gameplay stuff from screen code, right? So in other words if I'm say working on a vending machine that changes some player variables when you click a button, I still have to code an actor to do that? or is it ok to do things like that if it's only changing the player's variables?

[Helixhorned]

Yeah, it's best and cleanest if no information leaks between the game state and whatever happens locally.

[Jblade]

Helixhorned, on 07 June 2012 - 03:54 AM, said:

Yeah, it's best and cleanest if no information leaks between the game state and whatever happens locally.

Alright, that's good to know, thanks.

[Fox]

I wonder, I wish Eduke had support to fog and colorized sectors outside of palletes. If I, for example, want a nightvision google that replaces shading with green tinting, I must define a pelette for each shade... while I wish I could write the shade in a variable and set the tinting based on that variable... perhaps that's a little beyond Eduke's range.

[Helixhorned]

Fox, on 08 June 2012 - 04:30 PM, said:

I wonder, I wish Eduke had support to fog and colorized sectors outside of palletes. If I, for example, want a nightvision google that replaces shading with green tinting, I must define a pelette for each shade... while I wish I could write the shade in a variable and set the tinting based on that variable... perhaps that's a little beyond Eduke's range.

"A palette for each shade" is pretty much exactly how the classic renderer works, if you take "palette" to mean "256-index lookup". IIRC your mod was primarily aimed at classic, right? I still think that modifying the base palette (the mapping from 256 indices to R,G,B triples) is the way to go there. You'd "simply" make every color fade to a green instead of black. But as I said earlier, EDuke32 currently has no convenient support for that sort of thing, so if necessary I can bump that up in my priority list.

[Mblackwell]

Well, it would be nice to have some way of redefining tints and fogpals in CON on the fly. It would make a few weird hacks/ideas I've had floating around no longer needed.

[Fox]

Yeah, that's what I meant. I want a way to define palette-like features on the fly.

[MusicallyInspired]

Haha! Nice!

[Fox]

Is it possible to add a command to Eduke32 which returns the x or y dimensions or offset of a tile? There are plenty of uses for that, especially for drawing sprites on screen.

[Fox]

More requests... I just hope TerminX is still ready to use.

I had like to have an actorless command like EVENT_LOADACTOR and another events that works just like useractor (without an actor). There are plenty of effects that are achieved by running a code when the level is load or contantly through the game, but they depends on pre-existing actors. So I had like to have two events that 1. run independent of actors; 2. run their codes before all actors; 3. gamevar THISACTOR is set to -1;

[Danukem]

Fox, on 17 June 2012 - 05:26 AM, said:

More requests... I just hope TerminX is still ready to use.

I had like to have an actorless command like EVENT_LOADACTOR and another events that works just like useractor (without an actor). There are plenty of effects that are achieved by running a code when the level is load or contantly through the game, but they depends on pre-existing actors. So I had like to have two events that 1. run independent of actors; 2. run their codes before all actors; 3. gamevar THISACTOR is set to -1;

I don't see why that is necessary. I presume you have at least one sprite per map, so EVENT_LOADACTOR will execute. And if you want some code to execute exactly once, just define a state that gets called from the first and only the first sprite that executes LOADACTOR (set up a flag for that purpose).

[Mblackwell]

I tend to just run all code like that from APLAYER since it's always going to be around. Also if you want code to run before anything else (visibly) runs you can always use EVENT_DISPLAYROOMS.

[TerminX]

I've been spending my limited free time on Max Payne 3 lately (I haven't had time to actually play a game in ages). Making the tile sizes available is a really simple change though... it's literally 2 lines copied from Mapster32 into EDuke32.

[KillerBudgie]

Is Max Payne 3 any good? I'd like to hear from if it's any good so I may decide to go out and buy it for myself.

[Mblackwell]

I PMed Hendricks about this forever ago but it was never resolved:

Revision 2726 broke the ability to not use any grp at all (and may break the ability to use an entirely new grp file although that I have no way of testing that). This means most of my projects can't use anything past r2725.

Edit:

Going way back (not sure how I didn't catch this sooner, I must not have updated EDuke32 in the mods that used this):

r2569 | helixhorned | 2012-03-29 14:17:03 -0700 (Thu, 29 Mar 2012) | 7 lines

Make makepalookup() accept NULL for 'remapbuf', meaning "use identity mapping".

Also,
 - use this in game.c and astub.c palookup loading code
 - when makepalookup() is passed a 0 palnum, return early.  This means that
   'fogpal' will silently fail when attempting to change pal 0.
 - in 'makepalookup' DEF command, error out if passed a pal of 0

Unfortunately failing on fogpal 0 means that my mods which have a differently colored default visibility fog than black (for various reasons) are broken.

[Mblackwell]

Just saw a commit to fix problem 2. Thanks Helix!