A note on the USB-to-PS/2 mouse adapter that came with Microsoft mouse devices | Hacker News Reader

A note on the USB-to-PS/2 mouse adapter that came with Microsoft mouse devices

devblogs.microsoft.com

384 pointsluu142 comments

Oh wow, what a blast from the past. Around the year 2000 or 2001 I got an optical mouse as a present (either birthday or new year, can't remember). I really wanted it for a long time because playing StarCraft with a mechanical one was rather painful.

Needless to say it was a USB mouse and I only had ps/2. I purchased an adapter but it never worked despite spending a lot of time trying and testing it with different PCs and mice.

Then I took the adapter to an electronics repair shop. They opened it up and explained to me it was fake: it just connected the wires without any circuitry for the real USB support. They had no idea why anyone would manufacturer such a thing.

Well, now I know why. I spent quite a long time thinking I was sold a counterfeit adapter.

After reading the article I wondered if the adapter would work with other mice and if not wouldn't that cause confusion. So I guess we got an answer to those questions.

Chances are there are other mice that support both protocols on the controller, but use a different USB-PS/2 pin mapping.

My memory suggests that Logitech mice at the time had a PS/2 connector "natively" on the cable that plugs into an adapter for USB. Surely they did the same, skip on any active controller in the adapter and do it all on the controller they already have - those adapters where everywhere, and included with every mouse for a long time, which surely would not have happened if they incurred any meaningful cost pressure.

Now did they use the same mapping? (assuming my memories are not completely wrong anyways) With one being USB-PS/2 and the other PS/2-USB, a test would be stacking both into a noop-adapter USB-USB or PS/2-PS/2 and see if it works.

PS:

Sorry I think I was wrong, had an edit open but failed to submit: I believe that I was thinking of keyboards that did the same (in particular of Cherry keyboards), not Logi mice. The same, and for longer I think, because PS/2 held out so much longer in its keyboard variant. At first because of unreliable (or missing?) USB implementation on BIOS level, then because of lingering distrust.

Where PS/2 continued to exist surprisingly long (perhaps still?): the internal connection from laptop mainboard to touchpad. Not the DIN connector, but the protocol. And as a consequence, the cheapest external touchpads (those riding the efficiency of scale of the much higher volume internal market) had been PS/2 for a surprisingly long time.

I tried with Logitech mice (MX510, in pearly red) and it didn’t work.

The same happened to me when ordering a DIN-to-PS2 keyboard adapter from Aliexpress. But that one doesn't really need any extra circuit IIRC so it was even weirder they made a fake one.

But making a counterfeit and going to through the trouble to pointlessly connect incompatible signals makes even less sense, doesn't it? Unless the idea is that it should "look legit" by having something on the inside, I guess. But then some 1-cent IC would have made it more believable so it still doesn't quite make sense. Perhaps (perhaps!) I'm now overthinking this. Sorry.

I've always wondered that about these passive USB to FireWire adapters: https://www.amazon.com/s?k=firewire+to+usb+adapter

Is there something out there doing the same sort of trick as the mouse? I can't find any reference to one so as far as I can tell these things are entirely useless.

What in god's name?

There's a photo of the inside of one which just shows glue and wires, which backs up parent poster's conspiracy theory thinking about companies just making nonsense cables:

https://m.media-amazon.com/images/I/618Us-meOFL.jpg

There are dozens of these things. All the positive reviews look fake, too.

I can't find any reference to any legitimate product ever that had some sort of passive USB/Firewire adapter.

One review points out that Firewire uses 12V power, and a few others get power warnings from their devices potentially indicating that the pins are just being shorted somewhere.

Electronics shop assumed it was an active converter, rather than passive.

Well, shit! This explains away the tons of frustration I had as a 10 year old battling with seemingly random optical mice not working with the PS/2 adapter.

"Where were you then, Raymond Chen? I could have really used your advice 25 years ago!"

Once had a very awkward conversation with a sales person in PC World.

I had been given a old retired server from work to take home and learn with... only problem was it had no USB.

I was asking for a PS2 keyboard, he said they don't sell them but I could use any USB keyboard, I told him I don't have any USB ports and I needed a PS2 keyboard.

After awhile of him going back and forth with his manager, only to return to tell me i can use USB, It turned out he thought I was trying to buy a Play Station 2 keyboard and he'd never heard of PS/2!

I wear a pretty neat shirt which shows PS/2 mouse and keyboard ports and a text saying "I'm this old" at the front. As I will turn 40 next year, it felt adequate.

Most annoying tech support call ever:

Customer: Our computer won't boot. Come over and fix it.

Me: Please tell me what message you see on the screen which describes why the computer will not boot.

Customer: No, just come fix it.

Me: Okay, but we're charging for a service call even if I don't touch your computer.

Customer: Just fix it.

Me: (drive over, walk up to the office where the computer is at, pause at door and look in to read the screen across the room (younger me had much better vision) announce)

You have the mouse plugged into the keyboard port, and the keyboard plugged into the mouse port. Power the computer down and swap the connections and it will then boot.

I did tech support for a startup a long time ago. I mostly won't do any phone support for friends and family because they have the nasty habit of doing something like this:

Me: What do you see?

Them: <click> <click> <click> Nothing.

Me: What are you doing?

Them: <click><click><click> Nothing.

So I've been doing it in person or don't do it at all. Too many hours trying to visualize what in the world might be on the other person's screen.

The reason I'm moved to comment is two things: one is that my daughter in college had me help her file her taxes by starting with FaceTime on her MacBook, then sharing the screen so that I could see exactly what she was doing. She's not technically inclined and it wasn't something that I suggested, she figured out how to do it herself.

The other thing is that my mother-in-law is 90 and also not a techie. But I've been able to help her with 1 or 2 issues remotely because she's able to take a photo of the situation with her cell phone and text it to me.

Things have progressed to the point where hopefully your customer could have been convinced to send you a picture and saved you a trip.

Yeah.

Counterpoint to the above was the customer who selected OS/2 for their computer --- we installed it and delivered, but the next day they called asking for help with installing a printer --- despite my only OS/2 experience being doing the install the previous day, the UI was consistent enough that I was able to talk them through this on the phone w/o a reference system to use in parallel (we only ever sold the one OS/2 computer and never did get a second copy in).

This would make more sense with a parallel port and/or the DIN connector used by keyboards before PS/2, both of which have been extinct for much longer than PS/2, which is still being added to some main boards.

I would have included an RS232 D-Sub serial connector but those are still used if not for mice.

I mean… Yes, I agree. But I'm not, like, THAT old. o.O

Also, from a visual standpoint, it’s quite appealing the way it is: https://www.qwertee.com/shop/tees/i-m-this-old-468

I'm 5 years older than you and that t-shirt (very nice anyway!) still makes me think of late 1990s/ early -00s so it's still not THAT old. I owned an actual IBM PS/2 XT as a kid and they were not color-coded [1]

[1] http://john.ccac.rwth-aachen.de:8000/alf/ps2_76/ps2_76_2_ful...

Yep, the colors come from the PC System Design Guide [1] developed by Microsoft and Intel in the late 90s after IBM completely lost control of the PC market.

[1]: https://en.wikipedia.org/wiki/PC_System_Design_Guide

I'm that old! Now I want a similar shirt featuring the back side of an IBM 5170

That's cool, I love the colors

Or a gamepad port on a Soundblaster card.

It felt like back then everything was a bodge and computers were somehow made without any thought to gaming or mice even after they became common. Wasn't the game port really a MIDI port and gamepads pretended to be MIDI devices?

No, on the original Sound Blaster there is a jumper that switches the port between a UART for MIDI and a quad timer for IBM compatible joystick.

I believe so, yes.

I remember being blown away when that changed. "I can just use a USB port and not have to use the gamepad port?!"

I'd really like a modern version of the SB AWE32... I do wonder sometimes why front panel audio connectors never changed to a USB motherboard connector, I know it would be a few dollars more. I used to use a cheap USB dac on the front for wired headphones as the FP audio just never sounded that good..

Don't forget the Centronics 36-pin parallel connector!

PS/2 is like the 90s. If it didn't have screws for positive affixation it doesn't count.

You'll be thrilled to hear what they're doing with higher wattage USB-C cables, then! https://www.startech.com/en-eu/cables/s2cepr2m-usbsl-cable

The AT connector predated that and is essentially a jumbo PS/2 connector. Mice were DB9 serial, though.

Its funny because I'm younger than you by a couple of years and I can remember the big parallel ports on keyboards before PS/2 became popular! PS/2 still feels modern to me in comparison

Big parallel ports? There was some proprietary DB-something connectors (for example, the Olivetti M24/AT&T 6300 had one), but if you're younger than 40 you likely have been using an AT connector (basically PS/2, but physically larger) for the keyboard, and a DB-9 or DB-25 for the serial mouse - and that is easy to mistake for a parallel port.

PS/2 and AT refer to the IBM computer models where the connectors were introduced, of course.

Since this is an age competition, I can remember the annoyance of XT and AT keyboards and motherboards having the same connector but being incompatible. Some keyboards had a switch to change modes. I don't know what advantage the AT keyboard provided but it must have been huge to be worth breaking compatibility in both directions.

How about some of the PC “clones” that used older dumb terminal components and ports, probably to ease upgrades on accessories for some corporations.

Our computer lab in college had these NEC/AT&T 386 desktops that had RJ-11 keyboard ports that didn’t work anywhere else.

> Its funny because I'm younger than you by a couple of years

I'm 33 and also grew up with computers that only had PS/2 ports! Granted, USB existed, but the PCs I had as a youngster were old PCs.

I'm quite sure I literally used the adapter mentioned in the submission, as one of the first mouses I bought myself was a Microsoft mouse that came with a PS/2<>USB adapter!

For a few years during the Playstation 2's reign, I used a PS/2 -> USB adapter on my PS2 for my keyboard so I could text-chat while playing Tony Hawk games online. Always saw some humor in that.

and OS/2 meant half operating system. :)

I think I have some old mice soemwhere that came with a ps2 to usb adapter. (I should go pull them out, I don't recall f they were PS/2 mice with ps/2 to usb adapters or USB mice with USB to ps/2 adapters)

And yet... On almost every new ATX motherboard, there are PS2 inputs. And barely any options at all even available for purchase for individuals with, say, more than one RJ45 port, regardless of price. I don't get just how stagnant that industry is.

A friend who works in banking said their use of PS/2 inputs are a security measure.

They can disable USB support, but still have keyboards and mice.

I don’t think that’s the case anymore. I just popped on to newegg and scrolled through their “best seller” motherboards, none of them have PS/2 ports. They’ve started to die off on enthusiast motherboards for the past few years.

Some more details (and interesting historical perspective) can be found here: https://superuser.com/questions/704920/how-to-determine-if-u...

Apparently around 2014 the vast majority of mouses and keyboards were dual-protocol, but in 2019 the first reports of them being USB-only started appearing. I suspect a lot of the cheapest mouse SoCs are still dual-protocol as they worked when first designed, and they just kept producing them without any changes.

Pinout and explanation of how they detect protocol here: https://retrocomputing.stackexchange.com/questions/28288/how...

I have an old Roland S760 sampler from the early 90s. One of the innovations Roland made in this (lower cost) space was to add the ability to hook up a monitor and a mouse, to make chopping samples and programming patches much easier than peering at a tiny screen like on an Akai S1000 of the time (which I also have), or even worse on the S760's tiny two-line display

The mouse connector is MSX, which are very difficult to find these days, so mine came with a DIY MSX->PS2 connector. Sadly the PS2 mouse it came with was on it's last legs, so I wanted to replace it with something more modern. All I could find at the time were MS mice that came with a USB-PS2 connector. When I plugged it in, it barely worked and I could never understand why.

I did finally manage to track down a NOS PS2 mouse and it worked perfectly again. Until this post, I never understood just WHY it didn't work, so thank you!

The S760 is a fantastic sampler. It's kinda like a JD990 but as a sampler and without the full synth engine complexity. Made famous by Daft Punk and a lot of others as the S760 digital multimode filter became one of their signature sounds.

Interesting, I have a Roland S-330 from the late 1980s and my main concern getting it was the floppy drive viability. It didn’t occur to me that the mouse would be difficult.

You probably already know this stuff or have decided against it for authenticity or whatever, but just in case, you can use some of the Gotek USB flash drive floppy disk emulators:

https://github.com/keirf/FlashFloppy/wiki/Host-Platforms#rol...

And this may or may not be out of date, but may still be useful:

https://llamamusic.com/s50s550/FlashFloppy_GOTEK.html

You might have luck finding connectors on Buyee.jp. MSX was huge in Japan, and Roland is a Japanese corporation

I never thought to check the Japanese auction houses - I have a workmate there who can purchase locally as well. There is one for sale, but it's a bit expensive - I'll keep my eyes open thanks

An active converter is needed to use modern USB on retro computers. I made one a few years ago for many different computers.

https://github.com/dekuNukem/USB4VC

There's also a more recent project called HIDman, but only for PCs.

https://github.com/rasteri/HIDman

I once didn't want to wait to get a ps/2 keyboard in the mail to set up an older PC so I made a quick and dirty adapter out of an arduino.

Another one is ps2x2pico. I've been using it in my retro machines and it works great. For my Logitech G305 mouse (1000hz), I have ps2x2pico configured to force a 100hz mouse polling rate so the cursor is super smooth in Windows 98 without needing to use a program like PS2Rate. For my Keychron S1 keyboard, N-key rollover works flawlessly.

https://github.com/No0ne/ps2x2pico

Just reading this title and with only microsoft.com as the URL, I knew with 99% certainty this was going to be oldnewthing (Raymond Chen)

Good point. I see his current domain is "devblogs.microsoft.com". A few times I have seen @dang (or a teammate?) upgrade the HN software to show more URL details on the homepage. It would be nice to do the same for this one. A good example is fortune.com. Their blogs are utter rubbish, and their URLs now appear on the front page as something like "fortune.com/blogs" instead of just "fortune.com" (reserved for actual articles from the magazine).

I do enjoy his quirky little comments. Nothing ground breaking, but pretty much guaranteed to make me go :Huh! Never knew that"

Definitely one of the top blogs on that website. I wish win32 function articles would link references in blog posts, his explanations and examples helped me out a lot. Larry Osterman was another good blogger when he was still active.

I have a keyboard from my first 386/33 computer. It can be connected to my current Mac through a sequence of adapters: DIN5 -> PS/2 -> USB-A -> USB-C.

I am looking forward to extending my sequence of adapters in the future.

Windows 11 will happily load sermouse.sys if you hook a serial mouse to a USB serial adapter and plug it in.

Even on arm64!

Now that’s an old-new-thing post I’d like to read!

I assume because serial mouse support ca. 1996 is baked deeply into the USB standard, so you can’t pass connect-a-thon/compliance tests without it?

It might also just be a situation of "it's there, doesn't hurt anyone, doesn't cause much work, removing it could hurt someone". Windows needs (or at least used to need) keyboard support earlier in the boot sequence than mouse support. It's plausible that they load the entire USB stack early on to get keyboards etc working, then sometime later some ancient code iterates all serial ports to check for serial mice. To that driver, USB-Serial bridges and "native" serial ports should look the same through the hardware abstraction layer.

The leaked Windows XP code contains a lot of ancient code, some apparently taken from MSDOS. They later removed some of that, but apparently not this part.

> It's plausible that they load the entire USB stack early on to get keyboards etc working, then sometime later some ancient code iterates all serial ports to check for serial mice. To that driver, USB-Serial bridges and "native" serial ports should look the same through the hardware abstraction layer.

That's exactly it. When a serial port is initialized on Windows, it looks for data coming in, and if the right 'garbage' at the right time is picked up, Windows will then hook the mouse driver.

https://roborooter.com/post/serial-mice/ discusses the fairly simple protocol at hand.

This can lead into the unintentional effect of "plug in <serial-based device, like a GPS>, mouse suddenly goes wild". In my case... works as designed - serial mouse on USB-C dongle just works ;)

https://www.taltech.com/support/windows_2000_nt_serial_mice_...

It isn't. Serial mouse protocols are nothing like USB HID protocol.

I have something similar at my desk, an Apple Extended Keyboard II. It is connected to an ADB to USB adapter (which needs its coin cell battery replaced every 5-7 years) and then that is connected to a USB-A to USB-C adapter to connect to my MacBook. Despite being about 35 years old, it still functions beautifully and I use it every day. I really hope that I don’t have to add another adapter to my chain in the future but you never know…

I would think the PS/2 -> USB-A adapter in this chain is a smart one.

Yes, of course.

Um. Yes. It must be, or it would not work.

I still had my DEC LK201 hooked up on my main home linux desktop until a few years ago. Haha. I don't know why, but I liked the key layout.

Can't you get a DIN5 -> USB-A adaptor? Not sure if that would work, might be something designed for audio devices.

AT keyboard (DIN5) and PS/2 keyboards are in terms of electrical signals and the low-level framing protocol equivalent. As for the higher protocol level PS/2 is technically an superset of AT keyboard, but when used on PC platforms (or with Linux) the difference does not matter. And in reality most of the cheap keyboards you can buy only implement small subset of the protocol that is used by BIOS and Windows, such keyboards will not work with things like DEC Alpha workstations. (this also leads to somewhat surprising observation that under windows some keyboards work when plugged into mouse port, some do not)

But where's the fun in that?

> I am looking forward to extending my sequence of adapters in the future.

Keep us updated about how long it gets. :D

Pics or it didn't happen. (Seriously, I'd love to see that unholy train.)

Keyboards from the late 90s and early 2000s often did something similar. The controller supported both PS/2 and USB and the keyboard would come with a PS/2 plug and a passive adapter. That's why if you're trying to use a PS/2 keyboard designed prior to the introduction of USB (like a Model M) on a modern computer, you need to make sure to get an active PS/2 to USB adapter that actually translates the protocol.

Yeah, I discovered this the hard way when I tried to plug a somewhat modern keyboard to a pc through ps/2 (for the theoretical interrupt-driven speed) and didn't work.

A lot of mechanical gaming keyboards still support ps/2 though for the aforementioned interrupts.

I guess this has been long solved, but I remember around 2010 or so it was generally still recommended to use keyboards with PS/2 because of lower latency and support for N-key rollover.

I would expect that PS/2 to USB could be handled on the computer.

I have little experience, but from debugging some early 3D printers on Linux, I've learned that the U in USB really stands for universal: that it'll just pass on serial stuff to/from the OS, drivers or software on the computer to handle.

In addition to electronic singalling being completely different, the serial interface used by many 3d printers is not the same as low level serial interface sent over the wire. There is a low level usb protocol for stuff like data packets ,checksums , metadata negotiation, multiple endpoints in single device, basic transfer types(bulk, isochronous, control, interrupt). All common device types and standard drivers including USB to serial, HID, removable storage is built on top of this layer instead of raw serial sent over the wire. That's why you can have stuff like USB hubs which can forward data to and from multiple devices. And also why you can plug in a USB device, and the OS recognize and lookup driver without user manually configuring what device is plugged into which port.

Having done some embedded development, I am pretty sure that those 3D printers had a serial-over-USB converter like the famous FTDI family of chips. Those chips just simulate a plain old RS232 connection over USB.

There are some motherboards that can switch either PS/2 or RS232 signalling onto USB-A connectors, but this is completely non-standard hack mostly seen on some enterprise stuff few years back that seems to have mostly died away.

Not really no, the timing requirements of the USB protocol are very tight and handled by dedicated circuits. The exception is microcontrollers where you can sometimes achieve the timing in software.

This is similar to how a (cheap) DisplayPort-to-HDMI adapter worked, the adapter itself is completely passive.

Many devices with DisplayPort output support multimode, which allows the device to switch from outputting DP packets to outputting TMDS-based HDMI or DVI signals. You'll sometimes see a ++DP logo on these ports.

This is also why you don't see a cheap HDMI(source)-to-DisplayPort(sink) adapter, the smarts just aren't built into the source device.

> the adapter itself is completely passive

I think it was not completely passive, but simple level conversion; DisplayPort is AC-coupled signal, HDMI is DC-coupled.

There is a difference between DP+ and DP++. DP+ uses external level shifter, DP++ uses passive adapter and the card switches different transmitter into the path internally.

> I think it was not completely passive, but simple level conversion; DisplayPort is AC-coupled signal, HDMI is DC-coupled.

Wait, I don't think that's accurate. I've worked with HDMI/DVI and both are AC-coupled, that is, you put a 10 or 20 nF cap in series on the TMDS lines if you're driving directly from an FPGA's digital IO pins.

Signal level and coupling are separate things.

> Signal level and coupling are separate things.

sure, but the part is called a level shifter. MAX9406 DisplayPort to DVI™/HDMI Level Shifter

https://www.analog.com/en/products/max9406.html

Cheap-ish, luckily. I was quite relieved to find a reasonable ($50ish) active HDMI-DP converter. I have an old Dell monitor that only does HDMI 1.3 so to drive its full resolution you need to use its Displayport (or dual link DVI) input. And modern devices generally don't have either. The gadget solves the problem and gets me my 2560x1440 resolution.

if it uses 'power' (aside wire resistance) is not passive, e.g. I'd consider resistors, capacitors, coils - passive, even I'd agree about diodes, even though they'd add latency. DP->HDMI requires some electronics (a level shifter) as bare min.

Sligthly (un)related: The other day the builtin keyboard on my rather new Thinkpad X1C 12th stopped working. I connected a USB keyboard and tried to figure out whether there was an error in Linux kernel ring buffer or anything else explaining the problem.

To my surprise the internal keyboard is not even USB, but i8042. Not a hardware guy, but I had silently assumed everything is USB these days. I guess i8042 is the PS/2 protocol TFA is talking about?

(No error found, but after a reboot it has been working again.)

PS/2 keyboard/mouse are interrupt driven which saves power on laptops. While USB ones are polling based which require controllers running all the time.

Good point. On a video recording system we had problems with dropped frames from the USB camera. When an external USB keyboard was connected the problems became rarer. Obviously the keyboard kept some component to enter deeper energy states, and when everything staid up all the time frames were no longer lost as commonly.

Surprising to me as a layperson in this field is that camera with 30 fps would allow entering a deeper state than a keyboards with a data rate orders of magnitude smaller.

https://wiki.osdev.org/%228042%22_PS/2_Controller

To add to this: the keyboard matrix scan controller on laptops is usually located inside the Super I/O chip and connected to the CPU/southbridge via LPC or eSPI. Since LPC/eSPI peripherals are mapped to I/O ports, implementing an 8042 controller means that the keyboard can work even without setting up a much more complicated USB host stack.

On laptops it's usually not SuperIO, but EC (Embedded Controller), the same chip that handles battery, power, etc. and is usually connected over LPC to "present" the keyboard as PS/2 compatible input (some new ones are pure USB though, like my current laptop)

Some devices have SPI interfaces for HID though

That's a great site. Finally I think I understand WTF Lenovo mean by "IPMI over Keyboard Controller Style (KCS) Access"

Wow, thank you to share this website. It is so cool. I never saw it before. New Wiki rabbit hole to explore!

I don't recall seeing a USB keyboard in a laptop (admittedly I don't do that for a living at any rate). Even backlights tend to be separate pwm lines. It also make sense as USB would require a controller on the keyboard side.

In laptops, the EC is what usually connects to the keyboard and trackpad/pointing stick, and it communicates with the chipset via LPC (a variant of the old ISA bus). Many of them have a built-in 8042-compatible IP block for this. It looks exactly like a PS/2 keyboard and mouse to software, but the physical layer may no longer be present (some old laptops that support external PS/2 keyboard+mouse may have an actual PS/2 interface.)

If the keyboard stopped working (completely), I'd suspect a bug or random glitch in the EC firmware.

Until 5 or so years ago many laptop trackpads were simply PS2 mice (even with drivers installed).

Looks like it's again time for ThinkPad Linux user PSA:

  echo 255 > /sys/devices/platform/i8042/serio1/serio2/sensitivity

May laptop's mouspad is connected over i2c. I wonder what the reason for that is besides saving a few bucks on cables.

I suspected something list this, so these always scared me.

Actually, magic always scares me. I was putting together an LTE internet gateway from a Mikrotik device flashed with OpenWRT and was confused about how the SIM card slot on the main board interacts with a mini PCIe LTE modem. I looked at the mini PCIe pinouts and discovered the modem is actually USB (mini PCIe has pins for this) and mini PCIe has pins that are wired to the SIM card. I later ran into an issue where the card wouldn't work, but if I tape over the USB3 pins, it would drop to USB2 (still enough bandwidth for LTE) and work.

Didn't those adapters exist in both directions? Green in one and violet in the other? I have the feeling if I started to dig in my treasures I should still find some. One way to avoid e-waste is not to throw anything away...

The purple/violet was usually the color to refer to the keyboard, whereas green was code for 'mouse'.

As I recall you had the usb to ps/2 adapters for attatching legacy peripherals, but those usually came with two female ps/2 leads, a green one and a purple one, connecting to the one male usb-A.

One could hack the Linux kernel to connect an extra keyboard in the mouse connector. :)

It's just color coding for mice (green) and keyboard (purple). The connector's the same.

These and some other colors were specified in the PC-99 standard in 1999:

https://en.m.wikipedia.org/wiki/PC_System_Design_Guide#Color...

I knew that, and know most of the colours, but it's only now that I realise plugs on VGA cables are blue to match, plugs on DVI cables are white to match, and even my cheap USB-to-serial adapter has a teal RS232 plug.

The best part about Microsoft mouse USB=PS2 adapters is Microsoft made several mice models with incompatible wired differently adapters that look the same from the outside :)

I always assumed these were active devices doing some quick translation between USB and PS/2 but after dealing with USB controllers, this would have been exceptionally difficult (at the time) to fit inside of a dongle barely larger than the connectors.

I am using an USB-to-PS/2 adapter right now.

My mouse is a Logitech MX-500, which is the last model I have with PS/2 support, the next model MX-510 was USB only.

The motherboard is an MSI B550 TOMAHAWK, the computer was assembled just a few months ago.

The mouse still works and it has proven to be more reliable than modern ones.

If I may ask, why? The motherboard has USB, the mouse is USB, why bother with PS/2?

Also, and this is just a reflection of my experience, but I haven't had any issues with USB mice for a very long time. From my Logitech MX Master to very very very cheap (even free) optical ones.

Why not?

There is no “bother” here. The mouse has a cable, the cable goes to the back of the tower with USB or PS/2.

And I have one more USB port to plug other stuff.

This has answered a 25 year old question in my mind.

The very old Microsoft IntelliMouse are some of the best designed mice in the world. They always just work as designed. (Until you found someone with a glass desk, then it was time for a mousepad.)

https://en.wikipedia.org/wiki/IntelliMouse

My intellimouse explorer lasted until someone stole it after ten years, which isn't bad. I do think since then mainstream mouses are less durable (they only last 5 years!) but they didn't have the loose click feeling that the Microsoft mouses did. So Logitech and Razer for me.

So that's why the adaptor didn't work with other mice

Interesting. I have a USB IBM TrackPoint Keyboard that I use with a USB to PS/2 mouse and keyboard adapter. I bought the cheapest adapter I could find and I was surprised it worked. Now I know why—the keyboard must have a chip that switches between the two protocols.

These days it wouldn’t be too hard to make an active adapter with a small microcontroller in it. Not as cheap as just routing some wires though.

The motherboard on my old Intel 4790K would only go into the BIOS with a keypress from a PS/2 keyboard, not from a USB one... That computer is still in service, a friend uses it for general browsing, etc. When I gave it to him, I gave him the 3 purple keyboard adapters I had and an explanation to keep track of them.

I couldn't get one anywhere locally, even at Fry's Electronics at the time before they closed (man that place was a ghost town towards the end). I don't know if the keyboard adapters were the same as the mouse adapters, but I do remember that some would work with some devices and not others.

A note about even older AT keyboard connections they should not be hot plugged! Some older motherboards had a fuse soldered in that would blow if the keyboard was hot plugged. Fixed several in the days.

My whole life has been a lie... And yes I do have a drawer with several of these alongside VGA adapters and anything else I might need during the end times.

Yes, this is a common issue for folks who are into vintage mechanical keyboards. Be aware that you almost always need an /active/ USB to PS/2 converter. For the Model M, many keyboard generations had detachable cables that used the SDL standard and there are several options for in-cable conversions based on Blue Cube, TeensyUSB, or similar setups to allow you have a "SDL to USB" cable.

There's a purple adapter too that came with some keyboards. It looked like it came out of the same molds, so I suspect it's a similar situation.

This is really good timing because I just naively bought one of those little green adapters for test for three apparently fruitless causes -- a mini disc recorder, a vt520 and a vt525. I now have some hidmans (hidmen?) on the way with some hope that I/we someone can figure out whats missing to get it working with the VT52{0,5}.

hidsman

The Atmel AT90USB162 chip multiplexed the USB D+/- pins with an SDATA and SCK PS/2 port. It is possible to build a "serial" to PS/2 or USB bridge device with very few components. The part seems to still be available from Microchip.

Wish the article further explained how the mouse did this internally.

There is a section about it on Wikipedia PS/2 page:

https://en.wikipedia.org/wiki/PS/2_port#Conversion_between_P...

And the page it referenced:

https://www.nxp.com/docs/en/reference-manual/DRM014.pdf#page...

In Page.11, it show the algorithm to distinguish PS/2 and USB, based on USB D+ and PS/2 clock signal.

Funny enough PS/2 has found its niche and is included on most “gaming” motherboards because - less latency than USB! ps/2 has higher system priority

I wonder if you could bit-bang PS2 from the host side - though this is probably one of those things where we need to "consider if we should."

Yes, the PS/2 keyboard protocol is the AT keyboard protocol, which is pretty easy to bitbang on processors that aren't extremely slow.

https://en.wikipedia.org/wiki/PS/2_port#Conversion_between_P...

PS/2 is a flavor of line serial communication (0 and 5V): ground, power, clock, data.

"PS/2 mice send interrupts at a default rate of 100 Hz when they have data to send to the computer."

Which is how a 'dual mode controller' would drive it's IO pins in a way that work for USB or PS/2.

USB uses https://en.wikipedia.org/wiki/Differential_signalling , which Wikipedia's article explains have the advantages of greater signal to noise ratio and resistance to some common types of data-channel noise.

I don't happen to know what the _data_ voltage level of a USB link is offhand and skimming the article was insufficiently clear (Either of -5V and +5V, OR float is ~2.5V and one pulls up to Vcc the other down to Gnd make some sense.). The initial power pin level is also supposed to be ~5V.

Also, if you mean from a typical PC... A modern AMD system's block diagram shows most USB ports driven from CPU socket pins or from the primary board chipset directly (E.G. AMD's X870E or whatever). I would be almost amazed if those weren't in house / licensed ASIC blocks with ROM firmware.

> USB link is offhand

3.3V <-> GND

The data? I checked and Ground and Power pins are 0 and 5V respectively as I recalled.

This StackExchange thread has some good diagrams and useful answers:

https://electronics.stackexchange.com/questions/190592/why-d...

Lower speed links are suppsoed to pull the - data pin slightly negative and, effectively, a safely 'TTL high' (relative to 5V source) of greater than 2.8V. The power difference is 5V to better guard against too much loss over a long cable.

Higher speed links are quoted as using even lower voltages (the graph appears 0 and 0.4 v swapped as flipped).

I wonder if the pin connection is standard across manufacturers that do this. I assume anyone coming after microsoft's mouse release copied them.

It's likely there are two polarities of the adapter in the wild given the pin mapping but it's probably true that most copied Microsoft

Wild west. Even Microsoft made models not compatible with earlier ones.

I don't understand why so many modern motherboards continue to include ps/2 connections on their rear i/o panels.

I’ve worked in environments where USB is 100% disabled, including HID.

Today, I believe but can’t find an affirmative document, that some motherboards have an option to hijack USB HID mouse & keyboards and present them to the OS as something other than USB.

"Legacy USB support" is exactly this, uses SMM/ring -2 software routine to pass data from USB devices into emulated legacy 8042 keyboard & mouse.

a) gamers (supposedly lower latency and better n-key-rollover, realistically not really relevant)

b) servers for KVMs/crash carts/... that are still VGA + PS/2

Reply on news.ycombinator.com