site banner
ArjinFerman  Tinfoil Gigachad  6d ago (text post)   380 thread views

Tinker Tuesday for September 2, 2025

This thread is for anyone working on personal projects to share their progress, and hold themselves somewhat accountable to a group of peers.

Post your project, your progress from last week, and what you hope to accomplish this week.

If you want to be pinged with a reminder asking about your project, let me know, and I'll harass you each week until you cancel the service

2
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

EDIT: I realize that I am not quite following the instructions for this forum. I thought of this is as more of a technical discussion forum and did not want to post this in the CW thread.

Coding agents.

I use Gemini and Claude. I pay $20 for Anthropic Pro, so I am not using API tokens for Claude. With Gemini, you can do a lot with just a google account, or at least you definitely could on day 1.

I am hating Gemini lately. If I load the Pro model (default), I get mostly nothing but 503 errors. This was not the case roughly a month ago, when Google was promising a Pro response every 60 seconds and "you won't run out of tokens" or whatever. Fine, it's free, I get it. So now I use gemini-2.5-flash mostly. But this motherfucker is constantly undermining me. If I can get flash into a groove, we do fine, but I fight this guy a lot. I also get a vibe of "petulant laborer". It kind of cracks me up.

Claude is great, super friendly and helpful. I almost exclusively use Sonnet, and Sonnet has a case of the dumbs. It's extremely manageable, and I get more work done with Claude than Gemini. I can switch to Opus, and I will run out of my Pro credits and then open a Gemini session.

When I get stuck, I go to chatgpt.com and copy/paste. This usually unsticks me or gives me a new direction. I see this for 3 reasons:

  1. a new model
  2. a web prompt is a more direct interface to the model than the coding agent
  3. OpenAI sauce

I guess I'll use the OpenAI cursor thing sooner or later. I am CLI-only, headless linux for dev. I have a graphical environment for browsing, etc.

I rarely have the agent make commits, and I have regretted every session where the phrase "vibe coding" popped into my head. I try to discuss more with the agents than make edits, but I do have them make a lot of edits. They are more edit-happy than I would like. The main value-add for the agent, for me, is just the read-access to my project. That they can edit files directly is nice, but also kinda scary and has gone wrong for me. I'm much more comfortable with code generation than code editing. I love these guys for code generation, but I always edit their code, which usually works but ugly. I ask for review more than generation more than edits.

I have read that Claude and/or Claude Code has gotten dumber lately, due to "quantization", which I think makes a lower resolution model that is cheaper to operate. So that may be why I feel like Sonnet has a case of the dumbs. I rarely interact with Opus or Gemini Pro these days, on the CLI.

Okay long summer outdoors stuff is winding down. Back to Tron bike lighting!

Seems like I fried my ESP32-C3 by over-tightening the adjustment screw on the 12v step-down converter. Apparently if you break the screw it just kind of wobbles and the "5v" you're reading right now might change in an instant.

Then I broke a second buck converter doing this same oven-tightening before realizing the ESP32 was already fried anyway.

Electricity is kind of unforgiving :/

I did one of these successfully for the proof of concept but I guess forgot or got sloppy in the intervening months. On the bright side, everything involved is only $2-4 each!

The adjustment screws on those controllers are almost always simple multi-turn potentiometers, so it's a little weird to see voltage wobble even if one breaks off entirely. Lower-end buck converters can sometimes act up if you try to calibrate them without a load. Putting a 2.5-5k ohm resistor (for 12v, this should keep under 1/8th watt) on the output side may be worth a shot.

I got interested in the ESP32 stuff just from learning a little about the language, Espressif? It might have been something else, and you can see I never got into it. But one day! I will be a hardware guy doing a lot with cheap chips.

I'm onto my fifth try at a litRPG. It's progressing at a crawl, maybe 50 words per day on average as I squeeze in time after my job.

I have so many fun ideas for big fights, dramatic character developments, etc. but the pacing is really suffering as I flit straight from one big thing to the next. There needs to be downtime, I just don't find it interesting to write or think the result is very good.

Didn't get a lot done regarding Substack integration, but addressed a few bugs that were bothering me:

  • Apparently Twitter avatar URLs expire after some time. Either that, or some people I follow changed them (though I haven't noticed any difference), so I had to update them if any difference is detected.
  • "Starring" / adding a Tweet to favorites also cascaded to all it's responses, kinda defeating the purpose. Fixed, but introduced a new bug where "starred" quotes and retweets aren't rendered properly.
  • Paragraphs / endlines failed to render properly when viewing Twitter directly over the API.

How have you been doing @Southkraut?

Thanks as always for asking.

I finished my refactoring, ironed out some early runtime errors, and finally got to see things run and...not work, at all. Unreal really doesn't make code-based procedural generation easy. My meshes aren't visible, the Actor hierarchy completely lacks all subcomponents that I generated, and so obviously nothing can happen.

I need to do some research or go back to the Discord to ask some questions for this. Most likely I'm either doing some small thing wrong, or else I'm doing it all entirely wrong and this kind of runtime proc-gen just isn't meant to happen in Unreal.

Feels like one of the biggest problems Unreal has for me, bigger yet than it using C++ for scripting or its monstrous size, is the fact that so much of it is built around blueprints.

Starting to put together a log with my last ten years of endurance training. It's been a... frustrating experience. My record keeping was not good for about 60% of this period (I have no records of most of 2018, which is quite frustrating because that was a particularly good year for me, especially with summer training). Also frustrating to see how much of my long term potential was squandered on impatience which led to burnout/injury/illness.

Can I ask the purpose of building that dataset?

Since 2018 have you been using an app like strava?

To try and find patterns in my training and have a record of past training all in one place.

The problem is I deleted my strava account in 2023. I made a new account in 2024, but everything from 2015-2023 I have to try and find on old logs or on places like Garmin connect.

Work continues on the "polish" draft of my NaNoWriMo project. I'm indebted to @jake for providing feedback on the second draft, both in terms of the big-picture stuff (character motivations, stress-testing the plausibility of the basic premise and subsequent plot developments) and incredibly attentive copy-editing on the level of individual words and sentences.

So I think I last posted about my chairs in depth 2 months ago?

The last 2 months were the most tedious, least exciting part of the project. Sanding. Also a heat wave to kept me out of my shop for a nearly a month. And then my wife wanted me to make some shelves since I was taking a break from my chairs anyways. I had 68 individual parts to sand though a progression of three different grits, many of which were small or narrow enough that I didn't trust my orbital sander to not just round the whole damn thing over and ruin the edge profiles. So lots and lots of hand sanding.

I'm actually still in the middle of shellacing, waxing and assembling them. But I rushed ahead on my wife's advice to get one chair done to keep me motivated. The dry fit went fantastic, so onto the clamps. So many clamps. I didn't exactly trust my joinery to be perfect without some assistance, so I put my 24 kg kettlebell on top to motivate the legs to make full contact with the floor while the glue was drying. It's been a trick that has worked well in the past with shop furniture for my heavier tools. A piece of plywood to protect the finished seat seemed prudent though.

24 hours later, out of the clamps, and I attach the seat more permanently with some hardware that slots into a seam I cut out of the front and back skirt pieces. This allows for some wood movement in the large seat panels. All in all, I couldn't be happier with the final result, although the clamps did slightly mar the finish.

I had adjusted the dimensions from the initial plans I found to more closely match other chairs I had in my house already. This chair making guide helped with the dimensions a lot too. The original plans had it several inches larger in almost every dimension. It was just excessive, and intuitively seemed uncomfortable as hell. This is a perfect fit, and in fact is far more comfortable at the table I built than the folding chairs I was using. My elbows feel like they are at a more natural height, and the back being straighter makes it easier to sit up straight and reach across the table, versus the slouch other chairs have given me at it. A very happy accident, as I was I afraid I hadn't given the chairs enough backward tilt to be comfortable.

I have 27 more pieces to go with finishing, and 3 more chairs to glue up. It's a risky trade off finishing the pieces before assembly. But I've found it to be worth it, since my glue ups are always a messy disaster, and trying to finish the work afterwards always shows huge blotches in the finish where the glue got away from me and refused to sand out or wipe off cleanly. In fact, on the chair I did glue up, I used way too much accidentally and it squelched out of all the mortises and dripped absolutely fucking everywhere. Thankfully because the work was already completely waxed, it wiped off easily, and what didn't wipe off flaked off when it dried. That said, I think on the next chairs I won't leave it in clamps as long, and hopefully that spares the finish some of the damage it took. This chair was actually made using all the jankiest most warped pieces, and it needed to stay in clamps for the full 24 hours so that it didn't pry itself apart. All the other chairs fit together much more naturally so I can probably take them out of clamps after only an hour.

Also, my daughter keeps talking about how I'm the best carpenter she's ever met. My pointing out that I'm the only carpenter she's ever met doesn't dent her enthusiasm one iota. It's very cute. She likes to show off to her friends everything around the house I've made.

So now that I'm a chairmaker, if anyone has any political enemies they need taking care of, I suppose I'm open for business. You supply the next of kin though.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats