Are you sure that's a parallel port connector? Given gender it looks like a serial DB25 to me. Not totally sure though because of the power connector.

Do you have model number for the thing? Box of its bottom?

dionb wrote on 2024-11-05, 18:15:

Are you sure that's a parallel port connector? Given gender it looks like a serial DB25 to me. Not totally sure though because of the power connector.

Do you have model number for the thing? Box of its bottom?

I guess the next model, that look exactly like this, but no additional PSU jack on the paralel port is called WITTY TH406

The one with the power jack is also stamped Midasonic .

I just found the driver and the user manual. It is 25 pin serial cable

https://archive.org/details/ms-106-midasonic_20220726

Still no info about the additional PSU jack on it.

pichaga933 wrote on 2024-11-05, 17:38:

-I don't have the PSU for the connector. Any idea what supply voltage should I provide it with ?

Yes, I have one. It might require 5v power, tapped from keyboard connector.
The barell connector often was used for devices using such power adaptors.

Drawing tablets for serial port also used power from keyboard connector.
They shipped with a DIN pass-through adapter with a cable and a barell connector.

The question of porlarity does remain, though.
If it has a common ground, measuring the shield of the DB25 connector and the inner/outer part of the barell connector for continuity may help.

Again, it's just an idea. Speaking under correction here.

pichaga933 wrote on 2024-11-05, 19:43:

I just found the driver and the user manual. It is 25 pin serial cable

https://archive.org/details/ms-106-midasonic_20220726

Thought so. There are lot of hits on "Witty Mouse parallel" but they're all clueless sellers on auction sites, the same kind who call any 72p SIMMs "EDO".

Still no info about the additional PSU jack on it.

Given there's no mention of it in the manual, I'd start off by ignoring it and trying it without anything connected.

What I would do is grab my multimeter and measure which pins- if any - on the serial connector give connectivity with the power connector. As an otherwise pretty standard serial mouse can work fine without external power, I suspect it isn't designed to power the mouse, but it's designed to supply power from the serial port to some other peripheral you plug in. If that's the case, it would probably connect center or edge to RTS/DTR - pins 4 and 20 - (from which you can scavenge +12V) and the other one to GND - pin 7.

See here for more info about power from serial port.
https://www.epanorama.net/circuits/rspower.html

Alternately, it is actually to power something in the mouse. In that case it's probably 5V in and there won't be continuity with any of the serial pins.

dionb wrote on 2024-11-05, 23:50:

What I would do is grab my multimeter and measure which pins- if any - on the serial connector give connectivity with the power connector. As an otherwise pretty standard serial mouse can work fine without external power, I suspect it isn't designed to power the mouse, but it's designed to supply power from the serial port to some other peripheral you plug in. If that's the case, it would probably connect center or edge to RTS/DTR - pins 4 and 20 - (from which you can scavenge +12V) and the other one to GND - pin 7.

That suspection is correct, but not all devices were able to leech power from the data lines.
Depending on how power hungry the devices is it needs a dedicated power source.
Since this is 80s era technology, it can be either case.

The Datafab CF card readers for parallel port did require an external 5v power source, too, for example.
And they're from the 90s.

dionb wrote on 2024-11-05, 23:50:

but they're all clueless sellers on auction sites, the same kind who call any 72p SIMMs "EDO".

Heh, wait until you're trying to find memory for a PS/2 and discover how many ppl still use "PS/2" as generic term for 72 pin RAM also.

BitWrangler wrote on 2024-11-06, 00:58:

dionb wrote on 2024-11-05, 23:50:

but they're all clueless sellers on auction sites, the same kind who call any 72p SIMMs "EDO".

Heh, wait until you're trying to find memory for a PS/2 and discover how many ppl still use "PS/2" as generic term for 72 pin RAM also.

I do that, too! Since the 90s! 😅

The alternate description was "the big SIMMs" or "32-Bit SIMMs".
As opposed to "8-Bit SIMMs" (traditional 30pin SIMMs).

To be honest, EDO or FPM / Fast-Page Mode memory were terms I haven't really used back then.

I mean, they were in the big computer repair&upgrade books we had at home, but..
In daily life, the terms EDO and FPM didn't play a big role.

Not because of incompetence or ignorance, but because the memory technology of the day seemed so short- lived, anyway.

Technology in the 90s was evolving so fast, that it seemed unnecessary to remember all the little details of what seemed to be only temporary solutions.

SDRAM was probably also such a candidate here?
There were different types, I suppose, but in practice there was just SDRAM.

Edit: 486 cache types, write-back and write-through also weren't really a thing in daily life.
Users were happy if their 486 proccesor had a reasonable sized on-chip cache, at all!
I mean, there had been so many 486 chips made by a dozen manufacturers.
It wasn't just intel and AMD. There were NexGen, Cyrix, IBM, Texas Instruments, UMC, National Semiconductor etc.

Edit: Oh the nostalgia! The 90s were such a great time for PC sellers!
You could make so much money in ripping off customers! 🤗
No need for lying, even! Laymen were almost begging to take their money!

Edit: Ok, that may sound very mean, but some customers/users were a bit stubborn back then.
They insisted on certain things, such as big HDDs - but didn't want to take advice or hear about details.
So they ended up with big, but slow or loud HDDs. Or HDDs that used an inferior interface.
So even with good intentions, helping them wasn't easy.
Also in parts due to communication problems between techies and end users.

Jo22 wrote on 2024-11-06, 01:21:

[...]

So even with good intentions, helping them wasn't easy.
Also in parts due to communication problems between techies and end users.

If even techies were saying things like: "it seemed unnecessary to remember all the little details of what seemed to be only temporary solutions" and non-techies actively rejected being even confronted with that level of terminology, no surprise...

It wasn't just in the 1990s - the number of topics here now on EDO vs FP, SDRAM densities (not being bothered to look up the exact figures and instead using relative terms like "high density" despite what qualifies as "high" could change on a month-by-month basis), types of So370 socket etc.

But also in the wider world. ISPs insist on calling fiber devices that neither modulate nor demodulate "modems" because customers supposedly are familiar with the term (they aren't, in fact the same providers periodically resort to calling the devices "box" or "hub" or whatever instead for that very reason) achieving nothing more than confusing the issue with the few people involved who actually do understand the terminology. I've seen a presentation that literally said you needed to hook up the modem to the modem that you then connected to the modem (they meant: hook up a WiFi Access Point to a router which was then connected to the fiber network termination unit).

It's intellectual laziness that always leads to more confusion, not less.

🤷‍♂️

RAM surely was important, but the exact kind of internal operation mode.. Not that much, I suppose.
What matters was if it had parity, how short access time was (ns), which voltage it had used, if it was doubled-sided or not (later types).

I assume, in practice, things like FPM or EDO were rather silently assumed depending on RAM type or form factor.
As a rule of thumb, 30pin SIMMs were FPM by default, 72pin PS/2 SIMMs were EDO by default.
Especially since systems using SIMMs were 286/386/486 PCs, which all supported FPM but not necessarily EDO.
Later 486 PC mainboards had EDO support, but by that time the PS/2 SIMMs were the new star.

I think that's comparable to EPROMs of the time.
If someone was going to program an 2764 EPROM, it was important if it was CMOS or not (27C64 or 2764), how high the programming voltage was (ca. 12v, 21v, ca. 24v), how high access time was (ns).

However, who really checked in the datasheet which programming sequence was needed for each EPROM?
In practice, most users never cared. They've used that 100µs pulse standard sequence from the 1970s (!) rather than the recommended sequence by the manufacturer.
Probably also because the old C64 software they've used for the programmer didn't feature anything else.

That being said, my father and me never ripped of customers per se.
The worst we did was to convince them to have more RAM, which payed off when new Windows versions were released.
It was common practice in PC selling business in the 90s to ripp of users/customers, though.

The story with the fake caches comes to mind, for example. There's an article about it.
http://redhill.net.au/b/b-bad.php#fakecache

To make matters worse, only a few customers really had a precise idea of what they wanted, but were rather vague about it.
Also probably because everything was changing so fast at the time.
If a family member or friend giving advices for buying a new computer was a C64 user with knowledge from 5-7 years prior, it wasn't really helpful, either.
The confusion that stemt from it wasn't to be underestimated.

Edit: Then there were certain users who cared about the Megahertz and the CPU, simply to be able to brag about it.
- They wanted, say, a 33, 50 or 100 MHz computer, no matter the underlying architecture (386, 486, 586).
- Or they demanded for, say, a 40 MHz 386 PC or 486 PC, but didn't care if it was SX or DX or if it had any motherboard cache or not.
It was simply about the numbers. They wanted to be able to brag about it in the public.

Edit: My apologies for the bad wording here. I'm typing this on a tablet computer right now.
What I meant to say was that certain technical details went into background in daily life for various reasons.
But sure, computer enthusiasts did care about them. Just like they care if a game is running at 15 fps or 21 fps.
Ordinary users.. not so much. They rather cared if the overall experience was smooth or not, if the PC worked trouble-free or not.

Thanks for the help. I guess there are no real parallel port mice.

pichaga933 wrote on 2024-11-06, 23:31:

Thanks for the help. I guess there are no real parallel port mice.

Unlikely. Bus mice and serial mice were common on PCs of the 80s.

PS: I've found a picture of another Witty Mouse.
It says "no extra power supply required", so maybe some models had needed it at one point ?

I think the closest thing to parallel port mice I have seen would be "user port" mice for some 8 bits. Though you could probably hack it up for intuition legal stuff on an Amiga by availing yourself of parallel port joystick and joymouse hacks.

BitWrangler wrote on 2024-11-07, 02:05:

I think the closest thing to parallel port mice I have seen would be "user port" mice for some 8 bits. Though you could probably hack it up for intuition legal stuff on an Amiga by availing yourself of parallel port joystick and joymouse hacks.

Good point! The C64 had two types, essentially. Analogue via SID chip (paddle inputs) and digital (joystick emulation).

The digital joystick mice can be made working on a parallel port, maybe.
Using an Atari Joystick to PC adapter (LPT port) and with an appropiate driver.

On PC side, there are already DOS mouse drivers that support analogue joysticks on gameport.

The Amstrad PCW/Schneider Joyce word proccessor had an odd mouse, too. The AMX mice?

Unfortunately on the PC there is no Game Port , Just 1 Com port and 1 Parallel Port. That is it. And I am using serial to WIFI modem on the Com port. So my hope was kept in the Parallel port for a pointing device... Do not have the options for expansion cards, the serial controller onboard is just for 1 com port. It would need a bit of Hardware development of a special add on, so I could add second com port, but this will be in another topic.