Just posted my 1st monthly #devlog summarizing decisions/progress/plan making an MMO without funding, team, engine, or GenAI.
decision tldr: canned chat, sacrificing readability, unencrypted protocol, isometric 2d pixelart, action combat
qvixote.com/don/p/177349...
#indiedev #solodev
Did a 3k word draft today for the monthly #devlog (will edit and post it the 14th). Then I did my 2nd #pixelart sprite.
Yesterday's felt creative, this one felt like a puzzle game. I am not happy with the results, but after 2 hours working on it I'll just take the L.
#indiedev #gamedev
I know I said I would do the monthly devlog thing today...
But so far I'm 3k words in and it's not an entertaining read.
I have often suggested to students doing game jams to practice any of the individual skills, so I wouldn't be surprised if doing 90~180 minutes per day of work on a small solo project was the best way of practicing all of them.
A lot of good gamedevs I know didn't really start doing any gamedev stuff until their 30s, so don't worry too much because if you truly want to make games, you're already at a huge advantage for having discovered this so early and being able to do some work towards it.
It's just hard. Took me 9y of 8+h/day to make one I was proud of in a team, 14y to the first solo/compo one. To put this into perspective, with 20% of that effort (~3y) I was already proud of a couple 'novel' data structures (one of them turned out to already exist and is the HAMT)
That being said I don't know if I'll be able to work on the game tomorrow because I want to do a full review of the first month of work (30.5 days), a "what went right, wrong, what was done, and what I want to do this month"
Went with 24x24 after measuring a random 1080p screenshot of PoE2, scaled to 540p. idk if it's too small for what I am looking to do I suck too much to know myself.
Next day I may try 48x48 if no one gives me a better idea.
Today I spent most of the day researching and then a couple of hours drawing for the first time in decades. (Maybe even for the first time without anyone forcing me to do so). Kind of fun, kind of hard, like playing your first soulslike.
#solodev #gamedev #indiedev #mmo #devlog #pixelart
I am still working on the whole text parsing for the JIT.
The IR is still provisional, I am happy with some of the things I tried but without a specific editor stuff like the line absolute/relative implicit registers just don't work well
Nothing to show today, I have been working mostly on a parser for the small JIT I made days 3-4 of development, I also implemented the SSRA allocator by @mattkeeter.com
(which if I understand it properly was actually first implemented by Mike Pall for LuaJIT).
#solodev #gamedev #indiedev #devlog
Would it be a very bad idea to use oklch during both authoring and rendering? Only having the last postprocess step convert it to sRGB?
Also: Yesterday I implemented postprocessing into my engine.
#enginedev #gamedev #indiedev
This is the way, just slap a simple broadphase like spatial hashing on top and it goes brrr (fast enough).
Because it would be very easy to just skip the learning, generate some images and slop the word 'art' on them
I decided to make a tool within my tiny engine so that each of my future pixel art sprites has a persistent, complete editing history
#indiedev #gamedev #dev #gameengine #devlog #pixelart
However the big demoralizer to me is AI.
AI being so 'ahead' of what I'll be able to do for so long, has increased any resistance I felt two orders of magnitude.
So, starting the 11th, I will be drawing a sprite each day.
I may do the ( #pixel_dailies ) that I think I could use for the game.
Almost never in my life I have been in a room where a programmer who was more experienced than me wasn't better.
So, of course I'll suck at art: I never even tried to draw.
It has been decades since I was forced to draw anything.
I have been avoiding doing any art for almost a month.
Would be ez to say "I have 0 talent", or "I have aphantasia"
But I do not believe in talent.
A lot of people have called me lucky to be so incredibly talented at coding. Many would be better than me if they had spent close to 100k hours coding
This one looks really nice to me, but it feels like it uses too much screen space, maybe I'll try tuning the size of this a bit more.
This one feels dogshit because my brain immediately thinks of these as different options I could pick.
This one feels okaish? I liked more the big box one, but this is less intrusive.
I've been prototyping / playing around with a few ideas for questing text. I really like this color theme for plain text (the theme is github.com/mubin-thinks...), but I do not like any of the layouts I've tried so far.
#indiedev #gamedev #mmodev
Debug image of the game engine with metrics about cpu update times, draw command build time, gpu time (including CPU<>GPU communication time), and debug information about the world, the camera, entities under the mouse...
I didn't need font rendering, but couldn't resist.
I am using using nanoprintf and (a modified) libschrift.
To avoid changing the vulkan code I atlas using each color channel as a different texture, using dot product to pick.
I'm happy with the whole engine being <250KiB.
#gamedev #indiedev
Also, somewhat random, but I found my old anime op/ed playlist and my productivity went through the roof, it felt like being a young undergrad again.
Unsure if it's a good idea to use it, since I stopped listening to it shortly after retiring because it gave me PTSD-esque nightmares of crunch.
The harder part will be to add again the ability to create new sprites from screen data, but once I manage I will be able to remove cmake and raylib from the project.
I now can describe characters in screen-space (128x128, with a 96x32 capsule), made the editor able to move around objects (in xz, no height yet) & started porting my painting tool (its still missing the harder features)
#gamedev #indiedev #solodev #mmodev
Showing green and red guidelines on 540p and 360p, the red ones correspond to the most popular screen resolutions according to steam hardware stats
Have been working on stuff I needed for editor tooling, and just added guidelines I can toggle. I need those because, instead of 640x360 I am targeting "the smallest pixel perfect integer scale of 960x540 that would fill the screen", and cropping to the center (like old console overscan)
#gamedev
Yes, that's why I like activitypub more, and still do the longer posts there.
But I would have preferred if the focus was on allowing everyone to self-host their own identity/corner (pds/blog/localstorage/app) and request to be linked with any instance/homeserver they think they might fit in.
I will also stop doing hourly progress. I like the accountability, but they break my flow, and make me pick the low hanging fruits, making it harder to pick the important ones. I will instead post whenever I get something new working, which will be still a lot since I have a lot of TODOs
If you code for 14 hours, and end up with less code and a 7x smaller executable. Did you work at all? No, I didn't finish the netcode. Yes, I went back to the foundations and finally finished (the first implementation) of my first vulkan renderer.
#indiedev #gameengine
I am hoping to have a first online playable at some point later today, but it's a bit tight.
For a fully working proto-mmo engine I am still missing a cooperative scheduler, a cryptographic primitive, 2d and 3d dual-grid implementations, a debug ui such as imgui, some reliability protocol, a simple audio system.
Long term I'm also missing scripting, some shared memory, and hotreloading.
So far I've got: player<>3d world physics, a raylib and (incomplete) vulkan renderer, a sprite editor, logging, two allocators, four kinds of timing tools, serialization, synchronization, and atomic primitives, 3 randomness primitives, a whole bunch of design docs and mostly functional core loops.
Today it's been one month since I decided to un-retire and start making my MMO. Excluding raylib, stb, doctest I coded 6981 lines of C++ code. ~3500 without throwaway prototypes, benchmarks, and tests.
Given that I spent 308h working on the project, this is ~11 lines/hour
#indiedev
