A partial option is that any controller with analog triggers should work, especially as some driving games use them for acceleration (though I would assume remapping would work too). EDIT: One of the wikis says you may need more steps for analog triggers to work (on windows)

PCSX2 also now has a ‘pressure modifier’ key+setting (default is 50%) that may work in some cases (EDIT: specifically for digital buttons). At least if you have a key for it (like the OG Steam Controller has extra, like the grip buttons… change grip to keyboard key in sc-controller then bind that to the pressure modifier)

I know it, but I don’t know or use it, honestly

I forgot to say: some of how I figure stuff out is just by autocomplete (in my case, an LSP plugin), and if that isn’t obvious I look at the docs (Raylib has this).

I like to start with pure code first (no framework), at least where it makes sense (like my sweeper example, I focused on generating the initial maps first) and making that usable via import. The framework/engine-specific file is where I put all the code that is more closely tied to input, game state, and rendering. Doing it this way also means that making a different implementation is a bit easier.

‘pure code first’ doesn’t really apply to more advanced(/realtime) stuff like a physics game or platformer though, as you’d be more closely using an engine’s functions and types.

but what quick projects? What can I do? I have no idea

I know exactly what you mean, I am the same way. The most common sort-of-thing I’ve done typically is generator-y involving loading stuff from text files (adventure book reader (gtk4 framework), polygons (raylib)) as it’s an easy start though I didn’t get far with those for one reason or another (I don’t write+font didn’t scale as desired, polygons were too technical of a thing for me to do a complete implementation viable to use).

Took a while with thinking about structure before I attempted it (I didn’t follow a guide or anything), but the last thing I’ve done is a minesweeper clone with the map generated as a few sequence-of-sequences (in Nim-lang). I finished that with Godot for the GUI+tile rendering as it seems easier for me (I lean in on a lot of the editor features, too), though projects like this could certainly be done with RayGUI (+tilemap editors maybe?). Stalled when it comes to sharing though (for a few different issues).

So yeah, my method is to find that intersection of something understandable and at least somewhat interesting to you. Something you can actually use. Which sure, may be easier said than done. If you do follow guides, it might make sense if you go in with a plan for adaption after completion (like dodge-the-creeps turned into a roguelike shooter).

Interesting. I guess I’m not that far along (sort of stalled now), and quite possibly may never really need that.

Though for this one:

Some examples I usually have in my computers are ‘newFile’, ‘newTextFile.txt’ (just blank text files)

creating a blank file and renaming to .txt before editing seems good enough for me.

I do use music and video (Flash animations in video too), though yeah they have been moved to slower drives (because data, easier migration). I use XFCE but don’t use desktop icons.

At least projects is an understandable purpose, I don’t use the Templates, Desktop and Public folders at all, and aside from desktop (which I know is a workflow thing that I don’t even use) I would need someone to explain them to me. I’m guessing public would be for a multi-user system, templates maybe for printing stuff (I do not).

I’m of the mind that it’s probably better to start out with engines/frameworks to get your footing before you move on to lower-level stuff. Then again, I’m not very far either (and sort of stalled for a few reasons).

Godot might be a good avenue with GDextension (GDnative for 3.X), particularly with C/C++. Nim bindings exist (and this is my language of choice) but if you like C/C++ it will certainly have a lot less friction and with those you probably can get far with just the official docs. For C++, you could probably edit the engine itself especially for older versions.

Raylib is another option (and you will likely do a lot more boilerplate stuff) for pretty much any language. And again, with C/C++ it will likely be even more straightforward.

EDIT: Sorry, I see that you know about Raylib already. To say it more directly, I am saying it is likely better to find quick projects that you want to make on the device you’re already coding on (likely a computer) rather than trying to immediately support a specific device or dream projects. Though other low-cost devices (single-board-computers, microcontrollers) if you can get them (at/below MSRP) might be an easier way to fill that niche as well.

I mean consistent sound is fully in-line with what I’m saying, I am fine with robotic sound though the issue I have is that it can be grating for newer. Which I just assumed was something about how samples are used (compared to older speech synthesis). Is the sound actually part of the design to allow such high-speed?

Even if it were, older-style synthesis could likely have that as a parameter or option (or just… a dedicated voice).

I’ve seen some videos on screen-readers with a somewhat fast voice (not quite as fast as your link) that does sound better, similar voices to DECtalk Paul. They don’t seem to always give the voice name but I’ve seen some mention of IBMTTS so it might be related (though current results give AI service stuff that I’m not sure would trace back to those old videos (2016) but either way it might be some Paul derivative). EDIT: It might be ETI Eloquence?

It seems ETI Eloquence is both beloved in the blind community as well as something that has had support issues (proprietary abandonware). And I’ve seen one person on the subject:

It’s frustrating to say the least. Eloquence haters are like, what’s the big deal, but I’m like, show me a voice that is fast and responsive, and doesn’t make me wanna claw my eyes out like eSpeak does. I don’t like concatenative voices because you hear where the splices take place and it’s just weird and off-putting. They are also not as snappy.

The problem I have with Dectalk is that it slurs like a drunk as you speed it up.

I just listened to the samples and it seems a bit hit-or-miss. Some of them still stumble over words, have stilted pacing, or just sound off in some other way (raspy-ness, speed). It seems to vary more voice-to-voice than by the quality setting.

I mean I’m sure some of these voices are fine and probably better than other AI models in terms of performance… though they are a bit uncanny valley and I still think a voice meant to sound robotic (while still having personality) is probably an easier target. I didn’t notice anything like that in the samples, though I did see a couple of YT videos with a GlaDOS voice (sounding fairly accurate) that mention Piper (though I know such a thing likely wouldn’t be front-and-center due to licensing).

Honestly, a lot of newer TTS is worse than the 80s/90s stuff like DECtalk or PlainTalk (/MacinTalk). Both of which, while not exactly human-sounding, actually sounded better (at least in a sort of aesthetic way). For an example, Microsoft Sam (and whatever the voice is default for espeak) is such a downgrade IMO.

I’m not sure how heavy Piper models are (data or running), but I’m sure TTS could be better without neural anything.

Yeah, definitely bots. A different one (an account called ‘nobody’) on that instance posted one to a community called Religious fruitcake.

You’re right, I wouldn’t. None of what I said is specific to 3.1, and I do use Blender for ancient techniques (visible vertex color) rather than old software (even though I would like something simpler).

As to why anyone would do it? Familiarity and a “workstation” feel I’d guess, especially the more in-era stuff is added (software workflow, CRTs, scanners/printers). Maybe it’s just another way to avoid the modern mess of ads, AI, frequent updates/changes etc.

For someone producing a retro project, I could also see using an older OS as something akin to method acting (similar-to but-not-quite ‘dogfooding’). Stew in the exact design language and technical sensibilities you’re trying to replicate, rather than reading about it or looking at screenshots.

Bryce 3D was also just an example (and LGR’s video being the exact sort of energy I’m talking about), though I’m not sure how the newest version would compare on workflow and aesthetic. Seems like it’s more focused on realistic landscapes.

Sure, but when you’re doing a 5-hour render and it’s using something like 1/64th of your CPU (for LGR’s Threadripper example) that is a lot of room for improvement even if you don’t care about ‘maximum’. Especially for rendering multiple frames.

I mean a slight upside is that it is a lighter workload and it shouldn’t impact system usability.

I’m guessing the big issue here will be software lacking support for multi-threading. At least that’s the impression I got watching the LGR video on Bryce 3D.

Even some light threading or ability to user newer instruction sets/math etc might help but probably no easy way to do that unless some really smart people make patches or something. And even then, I might see someone so inclined to do so via WINE or even just that they’d use Blender instead.

I’m going to go with the multiple causes for slowness (and that’s interpreted languages in general). In some cases, things might be usable if I weren’t on Zen+ still (newer stuff has better IPC among other things).

Things like JIT or no-GIL might reduce that, but I’m not sure that it’s that easy to fix (not being default (plus multiple options) seems to complicate bindings even).

Politician leans too close into mic: “WORKS AS INTENDED”

Reporter: “but this is a clear-”

Politician: “lmao wontfix”

I did have this music bookmarked (6 1/2 hours of Commodore 64 SID Music) despite no direct connection/nostalgia to the C64. It just has something better than other retro music.

So how do I make my own SID tunes?

In most cases, you will need access to either a real C64 or a C64 emulator such as CCS64 or VICE. Ideally it is preferred that you actually compose on the real thing

I’d prefer pure midi (for use in a game engine like Godot, midi players are a fun idea too), anything where the soundfonts are free and not too big. Far down on the list for me though, and I’d probably need to lean on public domain midi files at first anyway.

I’m most interested in a polygon editor first (webGL animation/Wick Editor export) and a non-destructive workflow with Godot (SDF textures, Blender as an intermediate, maybe even exporting .scn files for polygons) second.

Though for this (and also the similar, separate effort of PixiEditor) I fear that it will just try to do everything else instead.

For this specific, I see people complain that it’s web/electron-based.

Hey, I’m not sure if it fixes what you were worried about, but there was a new release that fixed more than a few issues.

Furthermore the creator of a 3.X gdnative project (Enu) is involved so that may be interesting longer-term.

Linux specific: XFCE has window manager tweak settings (in accessibility tab) that makes any maximized window effectively borderless (combined with auto-hide panel, of course).

Games are weird though so the ones that are fixed at native-res when windowed (NO maximize) don’t benefit from that unfortunately. I personally have an ultra-minimal window theme that reduces that issue (it’s only 12px off the top if you don’t move it up). EDIT: I think Smallscreen is the smallest default window theme, though the sides and bottom of the window are non-0 so that adds difficulty to placement.