Show newer

Infodump on my Display Kit re-use project

Also contains lots of loose info on Type-C P/D alt-mode (DisplayPort) negotiation and implementation options that might be interesting (or totally wrong, this is all pretty new to me)

community.frame.work/t/custom-

In general I’m at a bit of a crossroads in the project now, unsure what option to design around since none seem particularly enticing…

fucking hell I've done a lot of research/writing today.. first part up community.frame.work/t/custom-

almost done with second half, but dinner first

oh.. even after consolidation and deduplication I still have 96 open tabs for this project

oh shit it's finally bandcamp friday!! time to buy all the things

oh sick! the fitbit actually recognized and tracked it this time. did 1250 meters apparently

Show thread

re: USB-C P/D hardware design question 

So far: there's a single standard (TCPC) for port controllers, so you can use a chip from TI/NXP/Infineon that handles the PD protocol, and talk I2C to it from your microcontroller to configure modes etc. Means you have free choice in microcontroller but the only software stack is an adapation of Google's chromebook firmware, and all the others are purely focused on power, not alt modes.

Texas Instruments chips do offer a more all-in-one solution, but require some Windows application to flash configuration values into them

Infineon CCG3 has the "EZ-PD" stack, but also seems to have a specific version that comes pre-programmed for DP dongle applications, but isn't available through JLC SMT nor LCSC, and I'd especially like something JLC SMT has so I don't have to solder 40-QFN etc..

STM32 has 'Cube' software blocks for either doing the entire PD negotiation on chip, or talking the I2C TCPC protocol. Also seems like JLC SMT stocks a bunch of those again.

Show thread

> Although USB Type-C can appear to be very
complicated at first, we believe that the initial learning curve is worth all of the advantages

I hope so too

The IMANU DeadRocks set is so good musically but it's just so sad cause the crowd is terrible, especially when he starts explaining something about his newest track and just gives up...

So. We all know what the spoons theory (#spoonie) is. You might even have heard of the derivative #fork theory as it pertains to mental health.

But Pina (a.k.a. ADHD Alien) has given people with #ADHD the perfect metaphor for our struggle with executive dysfunction: #Buttons.

adhd-alien.com/2019/07/16/why-

Now for there to be a button emoji that isn't a generic circle or circle-in-a-circle. 🥺

re: personal development, in software development 

sometimes I think the GoToSocial interface is still quite clunky but I had to use online banking today and with that comparison i'm doing just incredible lmao

Show thread

personal development, in software development 

These last few months I've gotten quite a bit more comfortable using Profilers (FireFox' overall, and React devtools), and code internals like custom React hooks and Redux-related stuff.

And user interface design stuff overall :)

The asymmetry between developers using powerful new machines and not noticing performance is terrible for the average user is often mentioned, but I've noticed something else with React:

A React bundle in development, especially with StrictMode, has *way* worse performance, because it's doing a lot of extra things to make the development experience nicer, stuff like rendering *every change* twice, to check for consistency in pure components.

This makes it a lot easier to notice (and debug) performance issues that might not even be noticeable in production builds, or only on way lower specced hardware

fuckin hyped to do some pcb design again, it's been ages

Show older
Pixietown

Small server part of the pixie.town infrastructure. Registration is closed.