serisman wrote on 2022-11-25, 19:47:

Thanks for checking it over for me. Yeah, your version is definitely easier to manufacture and more premium looking. For my se […]
Show full quote

rasteri wrote on 2022-11-25, 12:30:

Looks really great! I was thinking of doing something similar originally but I wanted the PCBs to be manufacturable without manually soldering on wires so I decided on the DIN ports instead.

I haven't tested this pinout fully yet - if you can wait a couple of days I should have it tested by the end of the weekend.

EDIT : what are you planning to do for an enclosure? 3D printed?

Thanks for checking it over for me. Yeah, your version is definitely easier to manufacture and more premium looking. For my setup, though, I would rather not have such a large external box with long cables.

I went ahead and added it to my PCB order. I can always spin a new version in the future if there are important changes that need to be incorporated. It looks pretty safe to me though. The serial pins can't really change if we want to continue to use hardware serial, and the PS/2 pins look like just standard GPIOs so are probably less important where they map to.

I did make a few tweaks to my (mini) PCB. Primarily switching to SMD pull-up resistors instead of the SIP resistor networks. I realized I don't have the correct SIP RNs on hand anyway, and the individual resistors gives more flexibility in tweaking values (or leaving unconnected if needed). Speaking of which... your schematic doesn't specify values for any of the pull-ups. I assume the PS/2 ones are 4.7k to 10k and the LED ones are 330 to 1k? I added optional pull-ups for the serial lines as well, not sure if they are useful or not?

Yes, I plan on designing a 3D printed enclosure, or maybe even just use some clear heatshrink tubing.

Oh, I also designed a small PCB for the serial extension that can be embedded directly inside a DB9 shell.

Both designs are available on OSHPARK, although keep in mind these are not yet fully tested designs. I'll add them to a github repo once confirmed working.
HIDman (mini): https://oshpark.com/shared_projects/GPxhWAbz
Serial DB9 to TTL: https://oshpark.com/shared_projects/djSCnWyn

I like the Serial DB9 (Female) to TTL if only mouse is needed, I guess it would need external power to be able to power the mouse?

tabm0de wrote on 2022-12-06, 17:10:

serisman wrote on 2022-11-25, 19:47:

Serial DB9 to TTL: https://oshpark.com/shared_projects/djSCnWyn

I like the Serial DB9 (Female) to TTL if only mouse is needed, I guess it would need external power to be able to power the mouse?

Yes, if emulating only a Serial mouse, you would need an external 5V power input. If also emulating (and connected to) an AT or PS/2 keyboard port, it can pull 5V power from the keyboard port.

@rasteri I've started experimenting with the CH559 chip and your code - am I right that the current KiCad schematics and the main.c and ps2.c files don't match up with regards to the PS/2 ports...?
I'm seeing P0.5 for KBCLK, P0.3 for KBDATA, P3.7 for MSCLK and P4.3 for MSDATA in the schematics, but P3.4, P3.5, P2.0 and P2.1 in code? Unless I'm missing something (and I often do)! 😀

Eivind wrote on 2022-12-09, 22:37:

@rasteri I've started experimenting with the CH559 chip and your code - am I right that the current KiCad schematics and the main.c and ps2.c files don't match up with regards to the PS/2 ports...?
I'm seeing P0.5 for KBCLK, P0.3 for KBDATA, P3.7 for MSCLK and P4.3 for MSDATA in the schematics, but P3.4, P3.5, P2.0 and P2.1 in code? Unless I'm missing something (and I often do)! 😀

Oh I hadn't pushed the last commit - try now 😀

rasteri wrote on 2022-12-10, 12:26:

Oh I hadn't pushed the last commit - try now 😀

Thanks, that did the trick! I noticed my USB mice weren't recognized if they were already connected at power-up, so I bumped the mDelaymS(5) to mDelaymS(100) in Task.c initSystem.
Keyboard seems to be working perfectly, though I couldn't get the typematic rate up from its default in DOS (using mode con rate).
Mouse however was a bit strange, ctmouse recognizes the ps/2 mouse, but whenever I try a program that uses a mouse, the cursor immediately jumps to the lower right corner (ie, max value) when I move it. Have you experienced anything similar?

Well, the PS/2 connector can be a combined mouse/keyboard connector. That might leave enough space for a DB9 serial connector.

Eivind wrote on 2022-12-11, 15:10:

Mouse however was a bit strange, ctmouse recognizes the ps/2 mouse, but whenever I try a program that uses a mouse, the cursor immediately jumps to the lower right corner (ie, max value) when I move it. Have you experienced anything similar?

Maybe its report format is a bit weird. Can you gather some log data for me?

I've just pushed an update to HIDMAN that will re-enable the debugging menu facility.

Once you have it flashed, fire up a text editor (MSDOS edit, windows notepad etc) on the PC that HIDMAN is attached to and momentarily hit the reflashing button (that's wired to 4.6). A menu will be typed up by HIDMAN.

Hit "2" for Log HID Data, and then unplug/replug your mouse to get the report descriptor(s). Wiggle it around a bit to gather some report data then save the text file and post it here (or DM it to me).

(Long term I may add a "safe mode" that forces keyboards/mice to BOOT MODE to account for weirdness like this)

EDIT : Oh, and "mode con rate=blah delay=blah" works for me. I'll add some more debugging options eventually.

rasteri wrote on 2022-12-11, 23:04:

Maybe its report format is a bit weird. Can you gather some log data for me? […]
Show full quote

Eivind wrote on 2022-12-11, 15:10:

Mouse however was a bit strange, ctmouse recognizes the ps/2 mouse, but whenever I try a program that uses a mouse, the cursor immediately jumps to the lower right corner (ie, max value) when I move it. Have you experienced anything similar?

Maybe its report format is a bit weird. Can you gather some log data for me?

I've just pushed an update to HIDMAN that will re-enable the debugging menu facility.

Once you have it flashed, fire up a text editor (MSDOS edit, windows notepad etc) on the PC that HIDMAN is attached to and momentarily hit the reflashing button (that's wired to 4.6). A menu will be typed up by HIDMAN.

Hit "2" for Log HID Data, and then unplug/replug your mouse to get the report descriptor(s). Wiggle it around a bit to gather some report data then save the text file and post it here (or DM it to me).

(Long term I may add a "safe mode" that forces keyboards/mice to BOOT MODE to account for weirdness like this)

EDIT : Oh, and "mode con rate=blah delay=blah" works for me. I'll add some more debugging options eventually.

Nifty! 😀

Here's a dump of me removing and reinserting a USB mouse:

1HIDMAN v0.1 Main Menu
2
31. Configure game controller mappings
42. Log HID Data
5
6ESC to exit menu
7
8Logging HID Data. Press ESC to stop...
9I0 L8- 0 0 0 0 0 0 0 0 
10
11
12Interface 0 Report Descriptor - 
13
1405 01 09 06 A1 01 05 07 19 E0 29 E7 15 00 25 01 
1575 01 95 08 81 02 95 01 75 08 81 01 95 03 75 01 
1605 08 19 01 29 03 91 02 95 05 75 01 91 01 95 06 
1775 08 26 FF 00 05 07 19 00 29 91 81 00 C0 
18
19
20
21Interface 1 Report Descriptor - 
22
2305 0C 09 01 A1 01 85 01 15 01 26 5E 00 75 08 95 
2402 0A 25 02 09 B6 09 B5 09 B7 09 CD 09 EA 09 E9 
2509 E2 0A 92 01 0A 8A 01 0A 21 02 0A 23 02 0A 2A 
2602 0A 24 02 0A 83 01 1B 01 00 09 00 2B 4C 00 09 
2700 05 0C 0A 27 02 0A 26 02 09 B8 81 60 C0 05 01 
2809 80 A1 01 85 02 15 01 25 03 75 02 95 01 09 82 
2909 81 09 83 81 60 75 06 81 03 C0 
30
31
32
33Interface 0 Report Descriptor - 
34
3505 01 09 02 A1 01 85 01 09 01 A1 00 05 09 19 01 
3629 05 15 00 25 01 95 05 75 01 81 02 95 03 75 01 
3781 03 05 01 09 30 09 31 16 00 80 26 FF 7F 75 10 
3895 02 81 06 09 38 15 81 25 7F 75 08 95 01 81 06 
3905 0C 0A 38 02 81 06 C0 C0 05 0C 09 01 A1 01 85 
4002 19 00 2A 3C 02 15 00 26 3C 02 95 01 75 10 81 
4100 C0 05 0A 09 00 A1 01 85 03 19 00 29 00 15 00 
4225 00 75 08 95 04 81 01 C0 05 0B 09 00 A1 01 85 
4304 19 00 29 00 15 00 25 00 75 08 95 02 B1 01 85 
4405 95 02 B1 01 85 06 95 0F B1 01 85 07 95 16 B1 
4501 85 09 95 05 B1 01 85 0A 95 02 B1 01 C0 
46
47
48
49Interface 1 Report Descriptor - 
50
5105 01 09 06 A1 01 05 07 19 E0 29 E7 15 00 25 01 
5275 01 95 08 81 02 95 01 75 08 81 01 95 05 75 01 
5305 08 19 01 29 05 91 02 95 01 75 03 91 01 95 06 
5475 08 15 00 25 65 05 07 19 00 29 65 81 00 C0 
55
56I0 L8- 0 0 29 0 0 0 0 0 
57
58
59HIDMAN v0.1 Main Menu
60

611. Configure game controller mappings
622. Log HID Data
63
64ESC to exit menu
65
66Goodbye

...however, after those report descriptors, if I move the mouse at all it spits out a few lines and then seemingly hangs (so I can't save the dump), but here's a couple of screenshots:

Eivind wrote on 2022-12-12, 00:20:

Here's a dump of me removing and reinserting a USB mouse:

Aha - your mouse uses 16bit X/Y reports. I have this working in another branch but it's not 100% there yet.

I just pushed a commit that forces Boot Mode for keyboard/mouse devices. This will reduce mouse resolution and stop NKRO but should be compatible with all devices.

Let me know if that works.

rasteri wrote on 2022-12-12, 01:12:

Aha - your mouse uses 16bit X/Y reports. I have this working in another branch but it's not 100% there yet.

I just pushed a commit that forces Boot Mode for keyboard/mouse devices. This will reduce mouse resolution and stop NKRO but should be compatible with all devices.

Let me know if that works.

After your last update (commit #9591d39), here's where I'm at:

Mouse no.1 (Razer Deathadder 2013):
Works perfectly AFAIK, the only gripe I have is that it's a bit too sensitive, even if I use ctmouse /r1. Like 1cm movement between edges of the screen.

Mouse no.2 (Roccat Kova [+]):
Left movement of the mouse causes the cursor to go up, and right goes down. Up and down movements are ignored. Any mouse buttons held down causes the cursor to move diagonally up or down to the right if I move the mouse left or right at the same time! 😀 It's also too sensitive (for my taste at least).

The HIDman debug mode still hangs after x lines of output while logging and moving the mouse around.

I've attached two files which are outputs of lsusb -v for each of the mice in a Linux VM, if that is of any help.
Thanks for your awesome work, we're almost there! 😁

Eivind wrote on 2022-12-12, 10:13:

I've attached two files which are outputs of lsusb -v for each of the mice in a Linux VM, if that is of any help.
Thanks for your awesome work, we're almost there! 😁

Could you get me the other mouse's HID report descriptor from the debug menu? I don't need you to wiggle it this time.

rasteri wrote on 2022-12-12, 18:37:

Could you get me the other mouse's HID report descriptor from the debug menu? I don't need you to wiggle it this time.

I've attached the Roccat Kova [+] mouse's report descriptor.
For the Razer Deathadder 2013, when starting logging and attaching the mouse the input hangs after spitting out the report descriptor, and it looks like the mouse isn't properly "started" (enumerated?). It doesn't light up its internal LEDs at least.
I obviously couldn't save it, but I took a photo of the report descriptor.

serisman wrote on 2022-11-25, 19:47:

Both designs are available on OSHPARK, although keep in mind these are not yet fully tested designs. I'll add them to a github repo once confirmed working.
HIDman (mini): https://oshpark.com/shared_projects/GPxhWAbz
Serial DB9 to TTL: https://oshpark.com/shared_projects/djSCnWyn

I got my PCBs back today and whipped one up (single USB port with Logitech Wireless adapter to PS/2 KB/Mouse). I did some light testing in DOS and Windows 3.11 and everything works as expected!

I haven't built up the DB9 Serial adapter, or tested the Serial mouse, or tested the 2nd USB port yet, but at this point I assume they will probably work fine as well.

@rasteri - Should I submit a pull request to include these PCB designs into your GitHub repo, or would you prefer they are kept separate?

serisman wrote on 2022-12-14, 00:29:

serisman wrote on 2022-11-25, 19:47:

Both designs are available on OSHPARK, although keep in mind these are not yet fully tested designs. I'll add them to a github repo once confirmed working.
HIDman (mini): https://oshpark.com/shared_projects/GPxhWAbz
Serial DB9 to TTL: https://oshpark.com/shared_projects/djSCnWyn

I got my PCBs back today and whipped one up (single USB port with Logitech Wireless adapter to PS/2 KB/Mouse). I did some light testing in DOS and Windows 3.11 and everything works as expected!

I haven't built up the DB9 Serial adapter, or tested the Serial mouse, or tested the 2nd USB port yet, but at this point I assume they will probably work fine as well.

Well, I built up the DB9 Serial adapter and attached it to my HIDman-mini, but it doesn't appear to do anything (yet?).

@rasteri - I skimmed the code, and it doesn't look like there is Serial Mouse code implemented yet. Can you confirm or deny that before I do any further troubleshooting? Also, the RGB LED doesn't seem to do much either. Is it supposed too, or is that also for a future enhancement?

Looking good i most say, doh my self i will have longer ps/2 cables as mouse not always near keyboard on some at pentiums 😀

serisman wrote on 2022-12-14, 05:17:

Well, I built up the DB9 Serial adapter and attached it to my HIDman-mini, but it doesn't appear to do anything (yet?).

@rasteri - I skimmed the code, and it doesn't look like there is Serial Mouse code implemented yet. Can you confirm or deny that before I do any further troubleshooting? Also, the RGB LED doesn't seem to do much either. Is it supposed too, or is that also for a future enhancement?

Correct - serial mice and RGB are in another branch. There's also some new report parser code that should hopefully fix Eivind's issues too. I should have some time this weekend to merge everything.

rasteri wrote on 2022-12-14, 13:25:

serisman wrote on 2022-12-14, 05:17:

Well, I built up the DB9 Serial adapter and attached it to my HIDman-mini, but it doesn't appear to do anything (yet?).

@rasteri - I skimmed the code, and it doesn't look like there is Serial Mouse code implemented yet. Can you confirm or deny that before I do any further troubleshooting? Also, the RGB LED doesn't seem to do much either. Is it supposed too, or is that also for a future enhancement?

Correct - serial mice and RGB are in another branch. There's also some new report parser code that should hopefully fix Eivind's issues too. I should have some time this weekend to merge everything.

Thanks for confirming my suspicion! Definitely no rush on my account. I don't really need a Serial mouse at the moment, I just wanted to test it out to make sure my hardware was working properly.

Anyway, I was able to design and 3D print a cover for my HIDman-mini, and DB9-TTL adapter. Turned out pretty good. I'll clean up my files over the next couple of days and put them in a new GitHub repo in case anyone else wants to duplicate this.

tabm0de wrote on 2022-12-14, 07:17:

Looking good i most say, doh my self i will have longer ps/2 cables as mouse not always near keyboard on some at pentiums 😀

Well, yeah, the cable length can be optimized to whatever is needed. I built this prototype with a specific computer in mind, and wanted a smaller adapter. I may build a more general purpose one with slightly longer cables next.

serisman wrote on 2022-12-15, 03:55:

tabm0de wrote on 2022-12-14, 07:17:

Looking good i most say, doh my self i will have longer ps/2 cables as mouse not always near keyboard on some at pentiums 😀

Well, yeah, the cable length can be optimized to whatever is needed. I built this prototype with a specific computer in mind, and wanted a smaller adapter. I may build a more general purpose one with slightly longer cables next.

😀 as soon I know all is good I need to order few 😀

Just to say, I've ordered a load of used hi-dpi gaming mice on eBay, and I'm determined to get them all working in non-boot-mode (or at least with appropriate sensitivity ).

Thanks all for sticking with the project, we'll get it stable 😀

(For the moment though, all the cheap generic mice I've tried work great so that's one potential workaround 😜 )