Nostr notes by gloriouscow

So @npub1g8x…7wul linked me to this cool PDF that has a lot of ...

2026-10-24T18:30:36Z

So Alice Averlong🏳️‍⚧️ (npub1g8x…7wul) linked me to this cool PDF that has a lot of inside photos and discussion about the design of the 286.

https://www.intel.com/content/dam/www/central-libraries/us/en/documents/2025-05/history-1984-annual-report.pdf

Inside I couldn't help but notice there's a photo of a CAD workstation, ostensibly showing someone working on the 80286 design.

...there's no way we could actually find the spot where he was working on the actual die, is there?

#reverseengineering #retrocomputing #vintagecomputing

A beautiful shot of the Intel/Harris 80C286 CPU by ...

2026-10-19T16:57:40Z

A beautiful shot of the Intel/Harris 80C286 CPU by DJ🌞[arbejdsløs] :donor: (npub1cwd…4d06), pins labelled by me.

You can see that VCC (+5V) and VSS (Ground) have two bond wires each, to handle greater current than a single bond wire would allow.

#retrocomputing #vintagecomputing #reverseengineering

I find this very interesting, that the 386 microcode is just sort ...

2026-10-17T16:30:13Z

In reply to nevent1q…82vk
_________________________

I find this very interesting, that the 386 microcode is just sort of a wider 286. GIven how the 286 was sort of a janky piece of shit and the 386 was Intel's killer chip that basically cemented the PC as the dominant home computer, I think that's pretty surprising.

So if we have the same source selector size, what are the chances ...

2026-10-17T16:25:27Z

In reply to nevent1q…63m9
_________________________

So if we have the same source selector size, what are the chances that the source values are mapped the same?

I'm betting dollars to donuts that the values 0-7 are AX,CX,DX,BX,SP,BP,SI,DI, just like intel always orders things.

But does this make sense? It kinda does. The 386 has more ...

2026-10-17T16:21:02Z

In reply to nevent1q…202y
_________________________

But does this make sense?

It kinda does. The 386 has more destinations - it added a bunch of debug registers, the FS and GS segments, etc. So it would need a larger destination selector.

As for bus control, the picture is similar. The 286 doesn't need the 386's DWORD-sized transfer codes, nor SPCR or LPCR. That leaves us with 32 values for bus control. Oh hey, that fits in 5 bits.

If that's true, that would give us this table.<code>| ...

2026-10-17T16:17:14Z

In reply to nevent1q…ujut
_________________________

If that's true, that would give us this table.<code>| Field | 386 | 286 | | ---------------------------- | --: | --: | | Source selector | 6 | 6 | | Destination selector | 7 | 6 | | ALU source / branch offset | 6 | 6 | | ALU operation / branch cond. | 7 | 7 | | Opcode | 3 | 3 | | Subcode | 2 | 2 | | Bus control | 6 | 5 | | Total | 37 | 35 | </code>

here's the other half. it kinda looks like one output goes ...

2026-10-17T16:14:57Z

In reply to nevent1q…mz00
_________________________

here's the other half. it kinda looks like one output goes off on its own, but i think it's just doing a sneak and it's joining up with the 4 lines above.

This makes the 286 have the same microcode field layout, except the Destination and Bus control fields are one bit smaller.

It's somewhat easy to see the field delineations on the die, ...

2026-10-17T16:10:44Z

In reply to nevent1q…lrlc
_________________________

It's somewhat easy to see the field delineations on the die, using DJ🌞[arbejdsløs] :donor: (npub1cwd…4d06) 's beautiful die images of the NMOS 80286 that, unfortunately, revealed an unreadable implant ROM.

Fields can be ascertained by the way the ROM outputs run off in groups.

Here we can see the first microcode fields are 6, 6 and 6 bits.

omg mark of the beast.

Here's the 386's word format.<code>| Field | 386 ...

2026-10-17T16:04:54Z

In reply to nevent1q…dy74
_________________________

Here's the 386's word format.<code>| Field | 386 bits | | ---------------------------- | -------: | | Source selector | 6 | | Destination selector | 7 | | ALU source / branch offset | 6 | | ALU operation / branch cond. | 7 | | Opcode | 3 | | Subcode | 2 | | Bus control | 6 | | Total | 37 | </code>

Preparations are starting for decoding the Intel 80286 microcode, ...

2026-10-17T16:03:59Z

Preparations are starting for decoding the Intel 80286 microcode, which we will hopefully have images of soon.

We can actually start figuring a few things out already, even without the bit extraction.

The 80286 has a 35-bit microcode word. This is straightforward to determine - there are 35 outputs coming out of the microcode ROM.

35 bits is a lot larger than the microcode word of the 286's little brother, the 186. The 80186 has a 21-bit microcode word, which is no bigger than the 8088.

It's much closer to the 80386, which has a 37-bit microcode word. Suspiciously close, in fact.

#retrocomputing #vintagecomputing #emulation

You can leak all three internal temporary registers on the V20, ...

2026-10-10T15:40:53Z

In reply to nevent1q…y6jt
_________________________

You can leak all three internal temporary registers on the V20, which would be extremely fun to try to emulate without a microcode-based emulator.

The eventual idea is to have microcode-based cores for 8088,8086,V20,V30, and 80286 in MartyPC.

I'm not sure if a microcode-based 386 is practical in software. I think MartyPC might just stay 16-bit forever.

I've now got a microcode-based V20 emulator that passes all ...

2026-10-10T15:21:43Z

In reply to nevent1q…6tng
_________________________

I've now got a microcode-based V20 emulator that passes all my V20 CPU test suite.

It's not cycle-exact yet for every instruction, but its a start.

I've pretty much completed analysis of the NEC V20/V30 ...

2026-10-09T03:43:59Z

I've pretty much completed analysis of the NEC V20/V30 microcode

https://github.com/dbalsom/x86_microcode/blob/main/v20/README.md

There's a few remaining unknowns, but for the most part, it's done. It's enough to program a microcode-based, cycle-accurate V20 or V30 emulator.

Someone should do that.

#emulation #retrocomputing

It's tantalizing because you can essentially see the bits. ...

2026-09-25T03:56:08Z

In reply to nevent1q…ktwa
_________________________

It's tantalizing because you can essentially see the bits. The column-selection circuitry is at the top, the row multiplexers and ROM output lines on the right.

Just saw a tantalizing image of the 80c286 microcode ROM, via ...

2026-09-25T03:49:03Z

Just saw a tantalizing image of the 80c286 microcode ROM, via sqpat (npub1mag…w696)

It's not an implant ROM like we found on the NMOS 286. That's surprising, since it means the ROM mask was redone for the 80c286, something that I assumed was unlikely after talking to a few folks.

The image is not quite clear enough to attempt a bit extraction. But it means that it's possible.

#retrocomputing #vintagecomputing

Did you know this rectangular cutout* in the spindle was called ...

2026-09-17T18:05:27Z

In reply to nevent1q…82nj
_________________________

Did you know this rectangular cutout* in the spindle was called the 'chucking hole'? Now you do.

*EDIT: It might be the center hole. Damn sloppy books.

Inside was the original sales receipt, which I have censored a ...

2026-09-17T18:01:58Z

In reply to nevent1q…vtmp
_________________________

Inside was the original sales receipt, which I have censored a bit although I doubt that credit card number from 1988 still works.

This book was purchased on 6/11/1988 for 47.87.

Nothing like relaxing with a good book #retrocomputing ...

2026-09-17T17:57:37Z

Nothing like relaxing with a good book

#retrocomputing #vintagecomputing

i'm proud of the 64x64 floppies. getting the arrow legible at ...

2026-09-16T16:54:09Z

In reply to nevent1q…vya2
_________________________

i'm proud of the 64x64 floppies. getting the arrow legible at this resolution was fun

Still grinding on floppy disk pixel art. ...

2026-09-16T16:45:20Z

Still grinding on floppy disk pixel art.

Come check our talk at Saturday 10:30 AM at VCF Midwest! ...

2026-09-08T18:26:52Z

Come check our talk at Saturday 10:30 AM at VCF Midwest!

Matthew | McJonesTech (npub1r3p…yrar) and I will be discussing the Seequa Chameleon, a quirky PC luggable from the early 80's that had an 8088 and Z80 CPU.

Come see our exhibit, too!

https://geerlingguy.github.io/vcfmw21-map/#079

I'll have a collection of light pens for various systems on display, as well as a working light pen you can play with on an IBM 5150.

And I'll have floppy stickers!

#retrocomputing #vintagecomputing #vcfmw2026

Read the story about how the NEC V20 microcode was decoded. ...

2026-09-03T20:37:28Z

Read the story about how the NEC V20 microcode was decoded.

https://martypc.blogspot.com/2026/09/decoding-nec-v20-microcode.html

#retrocomputing #vintagecomputing #reverseengineering #necv20 #x86

Well you kind of just did lol. MartyPC isn't 'new', ...

2026-09-03T16:09:56Z

In reply to nevent1q…6c7j
_________________________

Well you kind of just did lol.

MartyPC isn't 'new', though. I have been working on it for over four years now. It was running the PC demo Area 5150 back when it was still impossible to vibe-code an entire emulator.

If you've only just recently heard about it it's because I got laid off recently and I've been able to work on it more or less full time. Many of the new features I've been announcing have been in some state of in-progress for years, getting moldy in local branches. I'm just finally dragging them across the finish line.

I've been able to see what AI does with MartyPC when people fork it and try to add features - someone forked MartyPC and had Fable try to add a 286 and it did, by my estimation, a shit job of it.

Also the way that I validate my emulation is by comparing to a real CPU using hardware interfaces I build myself or bus traces I take off a real system with a logic analyzer, and I'd like to see Claude do that.

I do need to write up a contribution policy, not that I've had a ton of contributors (sqpat (npub1mag…w696) did kindly contribute an advanced 'Fantasy EMS' card, but I know he didn't use AI)

This is a fascinating PCB design ...

2026-09-03T00:09:07Z

In reply to nevent1q…d6r2
_________________________

This is a fascinating PCB design

the curvy traces are really helping contribute to the "mutant ...

2026-09-03T00:03:27Z

In reply to nevent1q…ypaa
_________________________

the curvy traces are really helping contribute to the "mutant swamp-thing PCB" vibe i'm getting here

this is weird lookin'. is this some conformal coating or is ...

2026-09-03T00:01:55Z

In reply to nevent1q…5hrf
_________________________

this is weird lookin'. is this some conformal coating or is it crystalized Jolt Cola? who knows. It looks wet but it's just solid

Grody to the max :( ...

2026-09-02T23:59:51Z

In reply to nevent1q…ecja
_________________________

Grody to the max :(

it's our old friend moisture ingress that might explain the ...

2026-09-02T23:55:27Z

In reply to nevent1q…lsd4
_________________________

it's our old friend moisture ingress

that might explain the buzzing

UH OH ...

2026-09-02T23:53:53Z

In reply to nevent1q…df5n
_________________________

UH OH

IBM's guide to operations says to turn the volume to ...

2026-09-02T23:35:51Z

In reply to nevent1q…l7ew
_________________________

IBM's guide to operations says to turn the volume to "high".

Well, this is what stuff looks like with the volume knob cranked. Notice it doesn't actually get any louder, it just clips massively.

I wonder if I have the right caps on hand to recap this thing...

...wouldn't hurt to take a peek inside?

Top is the source wav, bottom is recording back off the deck. ...

2026-09-02T23:32:15Z

In reply to nevent1q…rdhx
_________________________

Top is the source wav, bottom is recording back off the deck.

It's surprisingly good.

There's definitely an issue with the volume control on the deck, though. When i turn the knob while recording the envelope barely moves.

The volume knob is currently more like an "add noise" knob.

Unfortunately, I'm not getting anything but Device timeout. ...

2026-09-02T23:10:40Z

In reply to nevent1q…zckh
_________________________

Unfortunately, I'm not getting anything but Device timeout.

I'm going to need to record the audio from the tape back to my computer and see if it resembles anything close to the original signal. I'm gonna bet decent money that that buzz is also coming through the headphone jack (EAR) and that's gonna be a problem.

43 years of never being used but she still clicks. I'm a ...

2026-09-02T23:01:10Z

In reply to nevent1q…lehh
_________________________

43 years of never being used but she still clicks. I'm a little disappointed in how quiet it is. There's almost no point in recording it for MartyPC. I don't think you'd hear it with the case on.

Protip: AUX is an input, it's not bidirectional.

2026-09-02T22:58:39Z

In reply to nevent1q…uvww
_________________________

Protip: AUX is an input, it's not bidirectional.

Okay, according to the eBay listing, "Data In" is ...

2026-09-02T22:51:02Z

In reply to nevent1q…8v6d
_________________________

Okay, according to the eBay listing, "Data In" is cassette output.

Here's our cable. I'm not sure if "Data Out" is ...

2026-09-02T22:39:41Z

In reply to nevent1q…9p35
_________________________

Here's our cable. I'm not sure if "Data Out" is from the computers perspective or the tape deck's. I'm sure we'll figure it out.

The remote plug is smaller.. 2.5mm.

Let's just plug in the remote and see if the relay works.

The cassette remote control is controlled by a relay. Here you ...

2026-09-02T22:24:17Z

In reply to nevent1q…c0m2
_________________________

The cassette remote control is controlled by a relay. Here you can see it on the motherboard.

There's a very good chance this relay has never been used since maybe it was tested in the factory.

The cassette port on the IBM 5150 is next to the keyboard port. ...

2026-09-02T22:22:03Z

In reply to nevent1q…fcuy
_________________________

The cassette port on the IBM 5150 is next to the keyboard port. Don't mix them up.

We have successfully(?) transferred our WAV to tape. Heres what a ...

2026-09-02T22:10:59Z

In reply to nevent1q…dg5q
_________________________

We have successfully(?) transferred our WAV to tape.

Heres what a data tape sounds like:

So it's not the motor that is loud. The motor is fine. ...

2026-09-02T22:07:28Z

In reply to nevent1q…sqa4
_________________________

So it's not the motor that is loud. The motor is fine. There's just a very concerning buzz when you have the volume above 0. I'm guessing this deck could stand to be recapped.

My main concern is whether this buzz will carry over to the output signal. If it does that might cut short our little tape party.

Huh, the record button won't press down. Whatever could be ...

2026-09-02T21:51:48Z

In reply to nevent1q…uzng
_________________________

Huh, the record button won't press down. Whatever could be the problem?

Well, you see, my dear friends broke off the write protect tab on this cassette, because of course they did.

how dare anyone ever want to record over a 60 minute long advertisement against marijuana use.

Nothing a little tape can't handle

admit it this is the kind of quality content you follow me for

2026-09-02T21:42:48Z

In reply to nevent1q…v7p6
_________________________

admit it this is the kind of quality content you follow me for

I can't find my AUX cable. That's a shame. But who needs ...

2026-09-02T21:40:48Z

In reply to nevent1q…a8pf
_________________________

I can't find my AUX cable. That's a shame. But who needs one when you can just do this

So the first thing to do will be to record a WAV file of a ...

2026-09-02T21:28:56Z

In reply to nevent1q…uv09
_________________________

So the first thing to do will be to record a WAV file of a cassette onto an actual tape. I will need an AUX cable for this.

Pardon me while I find mine.

Oh dear, there's audio on Side B. Oh well. I'm not losing ...

2026-09-02T21:25:44Z

In reply to nevent1q…9e6s
_________________________

Oh dear, there's audio on Side B. Oh well.

I'm not losing a priceless memento here. I mean this is really shitty. Just imagine three stoned teenagers who have no idea how to play a guitar just jamming for 60 straight minutes making up lyrics on the fly about wrestling memes from the mid-90's and you will have the general impression

The motor is concerningly loud. But it's rewinding! I'm ...

2026-09-02T21:20:52Z

In reply to nevent1q…sezd
_________________________

The motor is concerningly loud. But it's rewinding!

I'm using side B of a tape from a shitty band some friends of mine had in high school, so I have to rewind it.

New tapes are like eight bucks! That's silly.

I'm guessing if I thought about this for more than ten ...

2026-09-02T21:16:41Z

In reply to nevent1q…pd6g
_________________________

I'm guessing if I thought about this for more than ten seconds I wouldn't be nervous because you can typically plug AC plugs either way

Well you see that little notch on the plug? There's a little ...

2026-09-02T21:14:36Z

In reply to nevent1q…rtgj
_________________________

Well you see that little notch on the plug? There's a little spring loaded retracting thingy that latches onto that thing. So i suppose that's the way to plug it in.

Here goes nothing.

The AC cord is keyed, but the receptacle is not. Which means I ...

2026-09-02T21:10:29Z

In reply to nevent1q…lw83
_________________________

The AC cord is keyed, but the receptacle is not. Which means I can plug it on either way. I'm guessing I shouldn't.

But I think we shan't use a PCjr. Tis a silly machine.

2026-09-02T21:05:09Z

In reply to nevent1q…nadz
_________________________

But I think we shan't use a PCjr. Tis a silly machine.

We need the appropriate breakout cable. If we wanted to use an ...

2026-09-02T21:04:07Z

In reply to nevent1q…m8lx
_________________________

We need the appropriate breakout cable. If we wanted to use an IBM PCjr, we could use this cable, which thanks to Computer Reset, is still shrink wrapped.

The IBM PC is designed for decks with a 'remote' line, as ...

2026-09-02T21:00:57Z

In reply to nevent1q…usg0
_________________________

The IBM PC is designed for decks with a 'remote' line, as seen here.

The remote signal allows the Play/Record buttons to be depressed and the computer to control the motor. So it can stop the cassette after loading software. Handy.

Here's my cassette deck. It's a Radio Shack CCR-81, ...

2026-09-02T20:58:37Z

In reply to nevent1q…t6xa
_________________________

Here's my cassette deck. It's a Radio Shack CCR-81, designed for the TRS-80.

It's time for another retro adventure with cow! Let's ...

2026-09-02T20:55:34Z

It's time for another retro adventure with cow!

Let's (try) to use a cassette deck with an IBM PC. I'm going in blind. I have all the stuff but I have not tested ANYTHING. Should be fun.

#retrocomputing #vintagecomputing #ibmpc

Some of the project members have been using Claude for ...

2026-09-01T20:13:17Z

In reply to nevent1q…v7ax
_________________________

Some of the project members have been using Claude for development. It's an umbrella project with dozens of people involved and dozens of platforms with their own hardware interfaces, and it's not really clear who's doing what.

I get that people don't want to support AI in any way, but I don't know the fact that there's a vibe-coded weather app for your Intellivision warrants throwing the entire project and significant amounts of engineering effort under the bus

so i guess #fujinet is very controversial but I assume you can ...

2026-09-01T20:04:28Z

In reply to nevent1q…fj8v
_________________________

so i guess #fujinet is very controversial but I assume you can make a decision for yourself whether mounting floppy disk images over your wifi constitutes becoming a accomplice to fascism.

Now, I emulate an 8250 UART, which caps out at 38400 baud, so ...

2026-09-01T19:38:02Z

In reply to nevent1q…t5n5
_________________________

Now, I emulate an 8250 UART, which caps out at 38400 baud, so this is not very fast. It's like PCjr levels of slow.

Actually PrintMaster is mad the disk is write-protected. and well I can't mount disk images R/W over the internet. Print Shop won't run from D: because it expects to be in drive A:

Someone should really QA this stuff

Here's I've loaded Printmaster.img. Hm. I wonder what ...

2026-09-01T19:34:46Z

In reply to nevent1q…c6g9
_________________________

Here's I've loaded Printmaster.img. Hm. I wonder what that could be.

With a Fujinet-RS232 we can select disk images from the internet ...

2026-09-01T19:32:31Z

In reply to nevent1q…qp6l
_________________________

With a Fujinet-RS232 we can select disk images from the internet using a web browser from various servers.

Well, what else can we do? Well, we can connect to a #fujinet ...

2026-09-01T19:28:15Z

In reply to nevent1q…jstr
_________________________

Well, what else can we do?

Well, we can connect to a #fujinet Fujinet-RS232 without needing com0com, for starters

oh, this is Telnet-related fun. we can't see what we're ...

2026-09-01T19:18:29Z

In reply to nevent1q…x7r6
_________________________

oh, this is Telnet-related fun. we can't see what we're typing, because Telnet negotiates host/client echo settings. I assume that host echo is off here.

I assume the little superscript 2 we got at the top there is some ...

2026-09-01T19:15:27Z

In reply to nevent1q…zgru
_________________________

I assume the little superscript 2 we got at the top there is some telnet-related schmutz.

I can't actually send anything though, that's weird!

2026-09-01T19:13:33Z

In reply to nevent1q…u7yv
_________________________

I can't actually send anything though, that's weird!

If we open Procomm Plus though and hit Reconnect, voila we have ...

2026-09-01T19:11:58Z

In reply to nevent1q…mwvy
_________________________

If we open Procomm Plus though and hit Reconnect, voila

we have gone to a silly place

now an immediate feature to add comes to mind, connect_trigger or ...

2026-09-01T19:10:02Z

In reply to nevent1q…uv08
_________________________

now an immediate feature to add comes to mind, connect_trigger or such which would delay connection until you actually sent something.

let's try going ...

2026-09-01T19:07:08Z

In reply to nevent1q…jslj
_________________________

let's try going here<code>[[emulator.serial_bridge.connection]] guest_port = 0 transport = "tcp" port_name = "A silly place" address = "77.37.63.221:8888" connect_timeout_ms = 3000 auto_reconnect = true reconnect_interval_ms = 1000 </code>

Oh boy, we're connected.

And here is where I make a fateful terminology decision that will ...

2026-09-01T18:54:47Z

In reply to nevent1q…0gkv
_________________________

And here is where I make a fateful terminology decision that will confuse the shit out of people.Connection

/Disconnection are separate concepts from Attachment/Detachment.

This way if your TCP session dies , like your internet goes out, we are still Attached to the IP and port number, but we are Disconnected.

You can reconnect, but if you want to connect to something else, you have to Detach. I will probably regret this.

Unfortunately this feature will not work on the web, because a ...

2026-09-01T18:32:26Z

In reply to nevent1q…emq9
_________________________

Unfortunately this feature will not work on the web, because a web page cannot just make random outbound TCP connections. It should be fairly obvious why you're not allowed to do that, as fun is that would be

It's fairly easy to abstract a stream of some given type ...

2026-09-01T18:28:03Z

In reply to nevent1q…0aje
_________________________

It's fairly easy to abstract a stream of some given type using a trait in Rust. I'm calling this trait SerialBridgeStream. Our virtual SerialPorts will take dynamic objects that implement SerialBridgeStream so we can expand to other transports later without another major refactor.

telnet is a very thin protocol layer over TCP though, and you ...

2026-09-01T18:24:35Z

In reply to nevent1q…gftm
_________________________

telnet is a very thin protocol layer over TCP though, and you might be forgiven for not even noticing you're using a raw TCP socket anyway.

Also, fun fact, the telnetd server in Linux had like a massive CVE in it for something like twenty years that was only discovered in January of this year.

The 'transport' key is important, because eventually ...

2026-09-01T18:21:41Z

In reply to nevent1q…yq3h
_________________________

The 'transport' key is important, because eventually I'll add telnet (and maybe in some far flung future, ssh.. don't hold your breath)

With a telnet transport you could set up COM2 to be directly connected to your favorite MUD.

Although a static connection is a bit limited, so what we would eventually want is a virtual modem that can take Hayes commands and dial IP addresses. This sort of thing is not anything new in emulator-land, I just haven't stolen it yet.

The configuration for this will look something like ...

2026-09-01T18:18:57Z

In reply to nevent1q…l863
_________________________

The configuration for this will look
something like this.<code>[[emulator.serial_bridge.connection]] guest_port = 0 transport = "tcp" address = "127.0.0.1:42000" auto_reconnect = true reconnect_delay_ms = 1000 connect_timeout_ms = 3000 </code>

So, if we're giving up physical control lines anyway, ...

2026-09-01T18:16:50Z

In reply to nevent1q…xmcd
_________________________

So, if we're giving up physical control lines anyway, what's better than a stupid pseudoterminal?

I'll tell you what! A TCP socket!

Pseudoterminals also don't exist as like, permanent devices ...

2026-09-01T18:15:58Z

In reply to nevent1q…nenj
_________________________

Pseudoterminals also don't exist as like, permanent devices unless I'm missing something. MartyPC can create one, but it's not attached to anything, which means I have to do some linux-specific magic to connect it to another pseudo-terminal after the fact, and it just becomes a whole bunch of platform-specific nonsense I don't want to deal with.

That's all well and good, but Pseudoterminals don't have ...

2026-09-01T18:12:15Z

In reply to nevent1q…s6g4
_________________________

That's all well and good, but Pseudoterminals don't have physical control signals like DTR.

No serial mouse for you.

You would think Linux, being the all powerful hacker OS, would ...

2026-09-01T18:11:32Z

In reply to nevent1q…lnp5
_________________________

You would think Linux, being the all powerful hacker OS, would have something equivalent to do this, making it super easy to make virtual serial ports and connect them together.

But no. I found various tantalizing tidbits like some sketchy ass kernel modules but Linux expects you to use Pseudoterminals (PTY devices) for this sort of thing.

Here's what it looks like. ...

2026-09-01T18:09:58Z

In reply to nevent1q…2tav
_________________________

Here's what it looks like.

Windows has an ancient piece of software called com0com that lets ...

2026-09-01T18:08:47Z

In reply to nevent1q…drkq
_________________________

Windows has an ancient piece of software called com0com that lets you create virtual serial ports and bridge them together.

You know this is crusty old software because you download it from SourceForge.

MartyPC has a serial port bridge feature. This is a fun feature ...

2026-09-01T18:07:54Z

In reply to nevent1q…7nsh
_________________________

MartyPC has a serial port bridge feature. This is a fun feature where you can connect a serial port on the guest machine to a serial port on your host computer.

If you have a means to connect these ports together, Fun can be had. Like connecting your virtual IBM PC to a Raspberry Pi serial terminal and browsing the internet with Lynx.

You can also even connect a real, physical serial mouse to a serial-to-usb adapter and use that. It's a bit janky though.

Okay, one last feature before I release 0.5.0 of MartyPC. I ...

2026-09-01T18:06:10Z

Okay, one last feature before I release 0.5.0 of MartyPC. I swear.

Just one last feature, bro. I can stop whenever I want

#retrocomputing #martypc

it may very well have been baking on top of some industrial ...

2026-08-29T22:20:42Z

In reply to nevent1q…mg4a
_________________________

it may very well have been baking on top of some industrial equipment for twenty years

This is a FastPoint Technologies (Formerly FTG) PXL-2000. It is a ...

2026-08-29T22:04:28Z

This is a FastPoint Technologies (Formerly FTG) PXL-2000.

It is a *USB light pen interface.*

If you never thought *USB* and *light pens* would be mentioned in the same sentence together, well, here we are.

#retrocomputing #lightpen

so not only are the graphics ruined, but the text is ruined too. ...

2026-08-26T04:16:45Z

In reply to nevent1q…jdh9
_________________________

so not only are the graphics ruined, but the text is ruined too. not a great situation.

notice that the "D" in "dealer" has become an O ...

2026-08-26T04:14:57Z

In reply to nevent1q…lmam
_________________________

notice that the "D" in "dealer" has become an O because MRC has the "Xerox Bug"

https://www.dkriesel.com/en/blog/2013/0802_xerox-workcentres_are_switching_written_numbers_when_scanning

here's an actual example. the left is from a Google Books ...

2026-08-26T04:13:03Z

In reply to nevent1q…2nzh
_________________________

here's an actual example.

the left is from a Google Books scan. The right is from IA:

imo this severely undermines IA's mission to preserve our history and there's a metric shit ton of content on IA that unfortunately probably needs to be reprocessed if not rescanned.

MRC is a compression technique ("Mixed Raster Content"). ...

2026-08-26T04:10:29Z

In reply to nevent1q…j8se
_________________________

MRC is a compression technique ("Mixed Raster Content").

It basically lifts the text off a graphic onto its own layer, stores it as 1bpp which can be compressed as such, then the underlying graphic, now complete with text-shaped holes in it, can then be compressed with jpeg or something.

the problem is whatever workflow IA uses to do this dials compression to all the way to maximum which results in 1bpp text over a smeary, unreadable mess

Amazing work but please please please stop using MRC in PDFs.

2026-08-25T21:38:33Z

In reply to nevent1q…nx0u
_________________________

Amazing work but please please please stop using MRC in PDFs.

my eternal frustration with art, is that i can envision exactly ...

2026-08-22T00:55:11Z

In reply to nevent1q…xtfe
_________________________

my eternal frustration with art, is that i can envision exactly what i want to make, but it rarely turns out how i envision.

you have to somehow be satisfied with where you end up, even if it wasn't quite what you expected

This is starting to look something like what i had in mind. ...

2026-08-22T00:49:33Z

In reply to nevent1q…nlnw
_________________________

This is starting to look something like what i had in mind.

I've always loved stars in pixel art. They're stupid ...

2026-08-22T00:23:57Z

In reply to nevent1q…k0s6
_________________________

I've always loved stars in pixel art. They're stupid simple to make, but they always look pretty

Trying to give an impression of the milky way. Does it work? ...

2026-08-22T00:17:55Z

In reply to nevent1q…050e
_________________________

Trying to give an impression of the milky way. Does it work?

some happy little trees ...

2026-08-21T23:52:38Z

In reply to nevent1q…p7hc
_________________________

some happy little trees

I'm thinking something like this. Snowy landscape, night sky ...

2026-08-21T23:45:15Z

In reply to nevent1q…xsq9
_________________________

I'm thinking something like this. Snowy landscape, night sky (aurora?), a bit of snow accumulation on his head.

I was attempting a new logo for the Macintosh emulator Snow. I ...

2026-08-21T23:33:25Z

In reply to nevent1q…dtl0
_________________________

I was attempting a new logo for the Macintosh emulator Snow. I was thinking of giving him the Jack Torrance treatment and putting icicles and shit on him, but I'm not sure.

Do we like pixel art Macs? #retrocomputing #pixelart ...

2026-08-21T23:21:32Z

Do we like pixel art Macs?

#retrocomputing #pixelart

report them for illegal kerning

2026-08-21T00:21:35Z

In reply to nevent1q…2pje
_________________________

report them for illegal kerning

Can I also just say how much I love Aseprite? I've been doing ...

2026-08-17T16:41:23Z

In reply to nevent1q…zu80
_________________________

Can I also just say how much I love Aseprite?

I've been doing pixel art for years in just Photoshop but I'm kicking myself because Aseprite is so much better.

If you do any pixel art whatsoever go buy it. It's a great bargain at twice the price
https://www.aseprite.org/

When you drag a disk image into MartyPC 0.4.2, you'll get ...

2026-08-17T16:33:29Z

In reply to nevent1q…kvxd
_________________________

When you drag a disk image into MartyPC 0.4.2, you'll get drop targets to let you choose what drive you want to mount the disk into.

I thought it would be fun to show the actual physical floppy type in pixel art in these boxes. Makes it a bit more visually clear.

Pixel art floppy disk! #retrocomputing #floppydisk #pixelart ...

2026-08-17T16:25:52Z

Pixel art floppy disk!

#retrocomputing #floppydisk #pixelart

I think maybe it might look better if horizontal deflection also ...

2026-08-16T19:48:06Z

In reply to nevent1q…qawm
_________________________

I think maybe it might look better if horizontal deflection also collapsed, but slightly slower