aki-nyan //nullptr::live
LightNews LightNews
banner
aki.lethalbit.net
aki-nyan //nullptr::live
@aki.lethalbit.net
1K followers 410 following 4K posts
abyssal witch | deranged catgirl hardware/software engineer + vtuber | that crazy SCSI girl | she/her | 🔞

banner: @rally-nine.bsky.social

www: https://lethalbit.net/
twitch: http://twitch.tv/akinyanya
mastodon: https://vt.social/@lethalbit
Posts Replies Media Videos
aki.lethalbit.net
Sure, why not, might as well,
White center-justified text on a black background that reads:k post this on your profile and let your followers associate you with a fictional character that you remind them of
November 10, 2025 at 11:10 PM
aki.lethalbit.net
Yeah, same BLAS, same,
The FORTRAN 77 code: RETURN C C THE PROBLEM CANNOT BE SOLVED C 30 IERR = 1 RETURN END
November 8, 2025 at 6:40 PM
aki.lethalbit.net
I still need to find a fun use for the two HALON ABORT buttons and enclosures I have,
A photo of a large red button with a shroud and a kinda grimy faceplate that reads "HALON" on the top "ABORT" on the bottom, "PUSH" on the left and "HOLD" on the right,
November 6, 2025 at 10:55 AM
aki.lethalbit.net
Annie laying on the top of my PC with her paws crossed, she's looking down to the left
November 4, 2025 at 4:49 AM
aki.lethalbit.net
For context, before/after:
A screenshot of the KiCad bus unfolding menu showing multiple bus vectors, test.meow[0..7]+ test.meow[0..7]- test.purr[0..7]+ test.purr[0..7]- test.tail test.ears[0..1] Only the `test.ears[0..1]` menu item has sub-menus to allow unfolding `test.ears0` and `test.ears1` A screenshot of the KiCad bus unfolding menu showing multiple bus vectors, test.meow[0..7] test.purr[0..7] test.tail test.ears[0..1] The `test.meow[0..7]` menu is expanded showing entries for `test.meow0` to `test.meow7`, and `test.meow2` is expanded showing `test.meow2+` and `test.emow2-`
November 3, 2025 at 11:52 PM
aki.lethalbit.net
Annie laying on top of my PC with her eyes close and a paw curled up under her head.
October 28, 2025 at 5:15 AM
aki.lethalbit.net
Ya know, as someone who is struggling to properly pick up drawing, I sure have a lot of sketch books...

Granted these have been collecting for the last like, 10 years or so, as i've always wanted to try to get into it, but gosh, it feels pretty bad being unable to clear the hurdle plaguing me rn.
A stack of 6 sketch books sitting on top of a PC with a PCBite plate and wires on top of that.
October 25, 2025 at 9:22 AM
aki.lethalbit.net
lurking
A photo of Annie crouched down and looking directly at the camera from a gap between the desk and the bottom of a monitor. There is some blurry out of focus electronics in front of her, include three black magic probes, a tigard, a icebreaker bitsy, and some other things.
October 24, 2025 at 2:33 AM
aki.lethalbit.net
For some more context, I'm trying to un-fuck this and my brain is too smol and full of parser-combinator to figure it out T~T
The Rust code: ast::RootExpr::Module(module) => { let mod_name = module.0.name().clone(); let modules = ast.modules(); if modules.contains_key(mod_name) { emitter.emit(Rich::custom( module.1, format!("Module '{mod_name}' already defined"), )); } else { let _ = modules.insert(mod_name, module); } },
October 22, 2025 at 8:11 AM
aki.lethalbit.net
1) Not a super helpful error message
2) Ohno, why is there like 100 of the same error on the same line
The following text repeated over and over and over again: Error: found ''' expected something else, or '"' ╭─[ ./tests/test.v:614:8 ] │ 614 │ .RST(1'b0), │ ┬ │ ╰── found ''' expected something else, or '"' ─────╯
October 22, 2025 at 5:24 AM
aki.lethalbit.net
Annie, laying on top of my PC with one paw over the edge looking into the camera
October 16, 2025 at 10:37 AM
aki.lethalbit.net
Thanks past Aki, really helpful stash message there
A screenshot of a git stash `stash@{0}` whos description is `On main: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa` from 2025-8-13
October 15, 2025 at 8:43 AM
aki.lethalbit.net
I *love* the Python 3.14 release logo *so* much

www.python.org/downloads/re...
The Python 3.14 logo, it's two snakes enjoying a pie with 3.14 on the top and π crimping one snake is blue one is yellow, the yellow snake has its tongue sticking out and it's eyes closed,
October 13, 2025 at 11:36 AM
aki.lethalbit.net
I'm sure that's totally fine:tm: bweeeeh.

I can't really move the FPGA up any more, or the SODIMM down, uuuuuuuugh,
a 3D view of a heatsink clipping into a SODIMM receptacle
October 12, 2025 at 9:47 AM
aki.lethalbit.net
So, it's not a *whole* like weaving now a-days, BUT there *WAS* a PCB technology that was a whole lot like it! It was called "multiwire" where all the traces were insulated enamel wire and woven throughout the PCB matrix!
A slide from an IEEE Workshop on June 9-12th in 2013, showing the Multi-wire-board technology
October 12, 2025 at 7:07 AM
aki.lethalbit.net
The FPGA is slowly spreading its tendrils
A screenshot of the KiCad pcbnew editor showing many multi-colored nets branching out from the big FPGA in the middle
October 12, 2025 at 6:33 AM
aki.lethalbit.net
*squints*
A screenshot of an ngscopeclient eye diagram of a 140 MHz PRBS31 stream with an 8V span. The eye is wide open, if not a little bit jittery, The same 140 MHz PRBS eye but through a 25cm dupont cable showing the eye squinting and degraded heavily.
October 9, 2025 at 9:24 AM
aki.lethalbit.net
Annie laying on her side with her face smushed up against the top of my PC with her eyes closed.
October 9, 2025 at 5:30 AM
aki.lethalbit.net
A photo of Annie looking off in the distance with a pile of wires in front of her
October 9, 2025 at 5:14 AM
aki.lethalbit.net
Please, my power budget, she is very ill
A screenshot showing KiCads schematic editor, it is showing the following table: Rail | I_{QUIESCENT} | I_{STARTUP} | I_{ACTIVE} | I_{TOTAL} | P_{QUIESCENT} | P_{STARTUP} | P_{MAX} | P_{AVAILABLE} 0v60 | ~5mA | ~20mA | ~500mA | ~525mA | ~3mW | ~12mW | ~330mW | 510mW 0v85 | ~1.275A | ~1.18A | ~8.5A | ~11A | ~1.08375W | ~1.003W | ~9.35W | 13.6W 0v90 | ~70mA | ~300mA | ~2.2A | ~2.57A | ~63mW | ~270mW | ~2.313W | 4.5W 1v2 | ~20mA | ~60mA | ~7.6A | ~7.68A | ~24mW | ~72mW | ~9.216W | 14.4W 1v8 | ~260mA | ~560mA | ~1.2A | ~2.02A | ~468mW | ~1.008W | ~3.78W | 4.5W 2v5 | ~5mA | ~50mA | ~120mA | ~175mA | ~12.5mW | ~125mW | ~450mW | 510mW 3v3 | ~520mA | ~270mA | ~1.13A | ~1.92A | ~1.716W | ~891mW | ~6.336W | 8.25W Above the table there are two notes: * I_{IN}: 4A @ 12V * P_{MAX}: 45.5W And below the table there is the following note: NOTE: 1.4W of 3v3 is for 2v5 and 0v6 Rails (Includes Margin)
October 7, 2025 at 2:07 AM
aki.lethalbit.net
Ah, yes, the ideal material thickness, infinity microns
The rust code: thickness: Length::new::<micron>(f64::INFINITY),
October 5, 2025 at 12:49 PM
aki.lethalbit.net
That kinda morning I guess
A screenshot of the output of `rustc` with the diagnostic note: `= note: the full name for the type has been written to '`
October 5, 2025 at 12:20 PM
aki.lethalbit.net
For context, this is what i'm talking about, observe the little blips in the bottom via pair.

The top has a small rules area in between the vias to keep out the pour from there.
A screenshot of KiCads pcbnew showing a purple copper pour with two pairs of Vias in the middle. The top set has the normal via clearance and a small rules area between them, which is keeping out the copper pour from encroaching, the bottom pair does not, which leaves little lumps between the vias.
October 4, 2025 at 12:31 PM
aki.lethalbit.net
"How many reference clocks does your instrument have?"

Yes many
A rough diagram showing a basic overview of a clock distribution network. On the far left there are two boxes, one labeled "EXT REFCLK" which is blue and one labeled "INT REFCLK" which is red. There is a white arrow coming out of "EXT REFCLK" and "INT REFCLK" going into a purple box called "SAMP CLKBUF" which has 5 red arrows coming out of it. In addition, "INT REFCLK" is going to another purple box labeled "SYS CLKBUF" which also has 5 red arrows coming out of it.
October 2, 2025 at 5:06 AM
aki.lethalbit.net
The symmetry pleases me
a screenshot of KiCad's PCBNew showing 2 USB-C connectors labeled J2 and J3, J2 is DUT and J3 is DBG They layout is mostly symmetrical along the horizontal axis, there are some ground pour slivers that still need post-cleanup
September 30, 2025 at 11:11 PM