[Spectere]

ajs, on Jun 10 2008, 11:08 AM, said:

(BTW if you get a chance pls do test the 2.82 [RC] PSP build http://mzx.devzero.c.../mzx282-psp.zip )

Awesome. I'll give it a go when I get home from work tonight.

Is there anything in particular that you want us to look for?

This post has been edited by Spectere: 10 June 2008 - 06:52 PM

[weasel]

Alright, I'm sorry, I tried, but I couldn't download the RC build. Not at a reasonable pace, anyway...and it seems this computer hates USB storage devices, so I couldn't get it on here anyway. It'll have to wait until I get home, unfortunately...

[ajs]

If it starts, that'd be a pretty good piece of news. If it loads a game, well..

--ajs.

[Terryn]

Before I forget, AGAIN.

I'd like a way for the editor-less builds to still use cheat functions - not for the sake of cheating, but for the sake of getting through the worst-case scenario in which a bug in a game would otherwise force someone to replay good portions of a game.

[Terryn]

Other things.

1) board_editor_hide_help hides too much. It should probably only hide the hotkey help, but it also hides very helpful information, such as what item is in the current buffer and whether the editor is in block selection/overlay/text placement/etc mode. I want it to display only that information when the help is hidden.

2) Already discussed this, but I'm setting it down in public. I want the ability to revert a Robot to the state it was before it is entered via the Robotic Editor. Currently, there's not much of an easy way to escape the Robot if something like a bad replace all goes awry and one wants to just load a recent backup.

[Terryn]

Again!

I'd like the octave range in sound commands to be raised to the 8th octave (as opposed to the current 6th). It's less idiotic of a request than it seems, considering MZX handles WAVs and OGGs now, and 44.1KHz/48KHz just barely makes it into the end of the 6-octave range, making speedup sound manipulation of OGG/WAVs somewhat limited. (Mathematically, extending the octave range by two will allow 192KHz wavs/oggs to be played at/near native speeds with octave settings, but that's a benefit that I don't see being used for pretty obvious reasons. 96KHz is comparatively feasible, sort of. The extension to eight octaves is simply because it's more "natural" than seven.)

[smilymzx]

I wanted a Energizer (or at least as a new Robotic command) without MZX's default Player animation. So the Player can have customizable animations. ''Adlo (the MZX game)'' uses the robot for Energizers, but does not work right!

This post has been edited by smilymzx: 05 July 2008 - 08:42 PM

[nooodl]

Nobody uses normal energizers. Use a robot.

[Lachesis]

ajs, on Aug 2 2008, 03:52 PM, said:

At least BASIC could always be interrupted with Ctrl-C or similar. MZX has no such feature!

Yo, maybe a key combo that drops you out of gameplay, to the title screen or editor, which is usable busyloops occur. I'd rather not have to open up the Task Manager every time something unexpected happens in my code, resulting in a busyloop.

[LogiCow]

A more simple solution would be to make escape work during busy loops.

[Revvy]

That wouldn't help with endless textboxes.

[Pyro1588]

i'd like to throw my support in for the ctrl+c or whatever key combination to kill the gameplay, at least when you're testing.

[smilymzx]

a list of nifty features that aren't supported yet:

A choice of character Height: 8x8, 8x12, or 8x16, as well with 8x14. or maybe let a user choose a height, (8 through 16)

Extend how much characters in a charset. Only choices I would go is 512 or 256(default)

Extend how much Colors in a Palette: 16(Default), 32, 64, 128, or 256(SMZX)

Optional: A choice of character Legnth: Only 8 or 9 is most possible

This is also optional: Extend and Make, and Default your Resolustion as long as your screen does not reach over 800x600 pixles.

These features are really needed to make my Mario clone, At least for the release version to be done

Some might be already asked for, But I need those someday. (Think of it as a Wishlist)

This post has been edited by smilymzx: 14 August 2008 - 05:34 AM

[Seventh Shade]

Can we get the ability to play sound effects and individual volumes. Like say:

sam 420 "~50~boom.wav"

would play boom.wav at 50% volume. This would let us do neat things like have thing get louder as the player gets closer to them.

[LogiCow]

Support for the mousewheel, mouse forward and mouse back buttons.
"buttons" is not intuitive. A better way would be to add "MouseN" counters.

[ajs]

With SDL just returning a bitmask here, it would be possible to support buttons 4 (WHEELUP) and 5 (WHEELDOWN) as extensions of the already available BUTTONS bitmask. As for this MouseN thing, I'm not really sure why you want that. You've got expressions.

--ajs.

[Terryn]

Well, the deal with BUTTONS is that the more buttons are supported, the more unintuitive the combinatorics become. If one wants to eventually support a mouse with a clickwheel and two side buttons, that ends up being seven actions (leftclick, rightclick, clickwheelclick, scrollup, scrolldown, leftsideclick, rightsideclick) which would need to give a specific number to each feasible combination due to BUTTONS being tied to a single counter. Further complicating things is that left and right clicks together returns 3, throwing a wrench into making an expanded numbering system sane.

[RoSS]

Terryn, on Aug 18 2008, 08:57 PM, said:

Well, the deal with BUTTONS is that the more buttons are supported, the more unintuitive the combinatorics become. If one wants to eventually support a mouse with a clickwheel and two side buttons, that ends up being seven actions (leftclick, rightclick, clickwheelclick, scrollup, scrolldown, leftsideclick, rightsideclick) which would need to give a specific number to each feasible combination due to BUTTONS being tied to a single counter. Further complicating things is that left and right clicks together returns 3, throwing a wrench into making an expanded numbering system sane.

While that is true somewhat, if buttons was made up of a series of bits, like (right side click) (leftside click) (scroll down) (scroll up) (wheel click) (right click) (left click), where say BUTTONS=1001001 means you're clicking right side and left button while scrolling up. While it doesn't look super intuitive, if you had all these set, seeing whether that would be
if "('buttons' a ('LEFT_BUTTON' o 'RIGHT_SIDE_BUTTON' o 'SCROLL_UP') )" = "buttons"
or
if "('buttons' x 'LEFT_BUTTON' x 'RIGHT_SIDE_BUTTON' x 'SCROLL_UP')" = 0

[Wervyn]

RoSS, on Aug 18 2008, 11:01 PM, said:

While that is true somewhat, if buttons was made up of a series of bits, like (right side click) (leftside click) (scroll down) (scroll up) (wheel click) (right click) (left click), where say BUTTONS=1001001 means you're clicking right side and left button while scrolling up. While it doesn't look super intuitive, if you had all these set, seeing whether that would be
if "('buttons' a ('LEFT_BUTTON' o 'RIGHT_SIDE_BUTTON' o 'SCROLL_UP') )" = "buttons"
or
if "('buttons' x 'LEFT_BUTTON' x 'RIGHT_SIDE_BUTTON' x 'SCROLL_UP')" = 0

I think the most intuitive way to do this would be "if "buttons" = "('LEFT_BUTTON'+'RIGHT_SIDE_BUTTON'+'SCROLL_UP')" then "label"", but that's just me. Regardless, RoSS is right, I don't think we need to pollute a bunch more counter namespace with builtins to represent single bits. As long as it's well documented, bitmasking flags is really easy. I do think it's worth adding support for more mouse buttons, though, definitely for the scroll wheel at least. Perhaps in a way that's configurable somehow. And if it is in the (game) config file, it might be possible to tie in gamepad or joystick support to the counter as well.

EDIT: I just realized that Terryn didn't realize that the buttons counter was already designed as a bitmask, and that "3 = left+right buttons" is completely sane. I had taken that for granted all these years. Um, yeah.

"Business or pleasure sir?" "ECSTACY!"

[Galladin]

I myself would be extremely happy to see a mousewheel support...it would make things a lot easier with my main project.

[ajs]

I've had a look at this in more detail. Whilst it's true that SDL exports the wheel status with the other buttons as a bitmask, I'd forgotten about the MOUSEUP/MOUSEDOWN interaction. With any button, you will get a MOUSEDOWN event and MZX will then update its internal button representation. It'll update it again on MOUSEUP.

The issue is with wheels that they work like one-shot "ticks". It's unlikely that such a "tick" will conveniently occur during a robot's poll of the BUTTONS counter, so the bits are never set. This is unlike the left/right mouse button which people will instinctively "hold" until some action occurs.

So the rest of MZX internally handles this by storing the "last button" pressed (note, this actually is not a mask, and can only represent one button). Since the user will want to detect each "tick" in turn for their game, it really makes more sense to introduce a LAST_BUTTON counter that maps this functionality through to robots. Unfortunately that'll require a world version bump, so unless this feature is very popular I don't think I'll implement it right now.

Another option would be to hack it, and have the fourth/fifth bits of BUTTONS "stick" such that on the first read, if the wheel was ever used since the last poll, the bits would be set, and after the read they would be zeroed explicitly. This would work for wheels and doesn't break compatibility, but I don't like it as much as the SDL documentation explicitly notes that buttons 4 and 5 may not NECESSARILY be mouse wheels. Doing this kind of thing for regular buttons would simply be wrong.

The third and probably best way to handle it would be to leave BUTTONS alone and add a STICKY_BUTTONS counter (or similar) which does the above for ALL buttons. This could be kind of useful for certain engines, I think. Again it would require a version bump.

--ajs.

[RoSS]

What you could do is have some sort of MOUSE_WHEEL counter that counts ticks until it's read. +5 would mean the there was five ticks upward since the last read, and -3 would mean there was three ticks down since the last read. Of course, there's the question of if you scrolled it up 5, then down 3 and read it, should it read 2 or -3. Personally, I think the latter makes more sense, but I could see a reasonable argument for the first.

EDIT: Werv, I have no idea what I was smoking when I came up with my solutions XD

[ajs]

Think I'm going to go for the "sticky button" counter since SDL can't guarantee that the mouse has a scrollwheel or (therefore) that buttons 4/5 are scroll events. As a result, this feature will be delayed to 2.83.

--ajs.

[Lachesis]

Can you make the Delete Board key be further away from the Save Game key?

or make it ALT+D or something

[ajs]

Pressing D brings up a window "Delete board:" which looks entirely different to the save dialog, and which you can leave with escape. Is this seriously a problem?

--ajs.

[paulguy]

I think even worse is the ALT+R and ALT+T being so close. I've wiped several boards because of my instant reflex reaction to hit enter at a blue dialog box...

[Revvy]

ajs, on Aug 20 2008, 08:47 AM, said:

Pressing D brings up a window "Delete board:" which looks entirely different to the save dialog, and which you can leave with escape. Is this seriously a problem?

--ajs.

I can understand why this would be a problem. I've got the S+ENTER+ENTER combo burned into a reflex now, and if I pressed D by accident then I doubt I could stop it in time.

[weasel]

Perhaps the first steps could be taken here to add alternative hotkeys that users would be more used to from other programs - for instance, perhaps Ctrl+S would save the world without the dialog box?

[astral]

Revvy, on Aug 21 2008, 02:02 PM, said:

I can understand why this would be a problem. I've got the S+ENTER+ENTER combo burned into a reflex now, and if I pressed D by accident then I doubt I could stop it in time.

S'what the built in backup feature is for.

[StarbladeEnkai]

Has someone mentioned "copying between vlayer and overlay" yet?