Author Topic: GH60 prototype betatesting [Call for layouts, GUI testing] (Read 184926 times)

JPG · « **Reply #250 on:** Mon, 25 February 2013, 12:25:46 »

Sometimes you need to play with the bios to make certain key "wake" the computer.

IvanIvanovich · « **Reply #251 on:** Mon, 25 February 2013, 12:30:25 »

But I never need to do that on any other keyboard. Any key will wake the pc using any of my Poker, Cherry G80, etc. Only the GH60 does not wake it. So while I am not sure why, I view it as a 'bug' as it is a basic function in my opinion and certainly a nuisance for me that it is not working.

alaricljs · « **Reply #252 on:** Mon, 25 February 2013, 12:32:35 »

It's already been covered by komar and myself. The 2 firmwares in question do not at the moment support waking up the PC from sleep/standby/hibernate.

komar007 · « **Reply #253 on:** Mon, 25 February 2013, 12:36:55 »

Quote from: IvanIvanovich on Mon, 25 February 2013, 12:30:25

But I never need to do that on any other keyboard. Any key will wake the pc using any of my Poker, Cherry G80, etc. Only the GH60 does not wake it. So while I am not sure why, I view it as a 'bug' as it is a basic function in my opinion and certainly a nuisance for me that it is not working.

It is not a bug, it's lack of feature;)

rknize · « **Reply #254 on:** Mon, 25 February 2013, 12:39:11 »

It's been a while since I debugged USB protocol stuff, but for the remote resume-from-suspend to work the host sends a special command to the device before suspending it. IIRC, the device is supposed to go into a low current mode (2.5mA or something). The USB protocol allows such a peripheral to cause the USB host controller to wake the system via an interrupt. I'm not sure how the relevant libraries for this controller will handle all of that.

IvanIvanovich · « **Reply #255 on:** Mon, 25 February 2013, 12:45:12 »

I see. So does all the custom DIY boards not resume from sleep?

alaricljs · « **Reply #256 on:** Mon, 25 February 2013, 12:46:38 »

Even though I've had a hefty number of Korean boards pass through my hands that is something I have never tested for.

hasu · « **Reply #257 on:** Mon, 25 February 2013, 17:55:31 »

Quote from: IvanIvanovich on Mon, 25 February 2013, 11:25:08

I'm used to be able to press any key at all on any other keyboard to wake the pc don't need a special key, and I have it disable on my mouse to wake the pc because it would wake it randomly.

^This. I didn't want to let pc wake randomly with accidental key press

I don't use this remote wakeup feature myself.
At this time PJRC stack only supports USB remote wakeup and keybind is not flexible. You have to build with PJRC stack and press LShift+RShift+PrintScreen to wake your pc up.

Feature request and bug report also should go with github issue tracker, it is useful to remind me of what to do.
https://github.com/tmk/tmk_keyboard/issues/16

IvanIvanovich · « **Reply #258 on:** Mon, 25 February 2013, 19:21:48 »

Not the friendliest keycombo since my PRT key in a Fn layer and my locking shift should make it really fun, but I might give it a go later. For now I guess it is not that big of a deal to reach over and give a quick tap to the power button to wake the pc.

alaricljs · « **Reply #259 on:** Mon, 25 February 2013, 19:30:54 »

The combo is configurable, config.h has the lshift/rshift combo (IS_COMMAND) and all the different key presses that go with it are in common/command.c.

IvanIvanovich · « **Reply #260 on:** Mon, 25 February 2013, 19:38:04 »

It would be nice if I can use space. Since it is the largest key it is easy to locate before my espresso after rolling out of bed in the morning

)

alaricljs · « **Reply #261 on:** Mon, 25 February 2013, 19:49:47 »

https://github.com/tmk/tmk_keyboard/blob/master/common/command.c line 246 chance KC_PSCREEN to KC_SPACE

This at least gets you shift-shift-space to wake.

komar007 · « **Reply #262 on:** Tue, 26 February 2013, 17:43:42 »

I've just published a very preliminary version of the Python GUI and a new firmware version which supports it. Everything is in the first post.
Please help find bugs and get it to work properly.
I'm sorry for the terrible quality of some parts of the code, but this is a work in progress. Because of that, please don't make changes yet which you want to be included in upstream later, since some parts of the code will change greatly.

IvanIvanovich · « **Reply #263 on:** Wed, 27 February 2013, 19:39:29 »

I download the gui files, but I am not sure what is supposed to do with it. I download python files but I could only open it in edit mode. It have been at least a decade since I looked at anything python related so maybe some basic instruction would be good.

hasu · « **Reply #264 on:** Thu, 28 February 2013, 05:34:25 »

komar, really really great work!

I confirmed your app on a tad old xubuntu, I need to install 'python3-minimal' and 'python3-tk' packages for this test.

Writing layout binary from GUI worked flawlessly and quickly. Editing layout also worked well except for one problem.

My problem is that I couldn't define 'layer action without scancode', I just want to change layer without sending any key. I know your gh60.lay has such key on key 57 but I couldn't change existent key(key 61) into such normal layer change key.

komar007 · « **Reply #265 on:** Thu, 28 February 2013, 08:51:06 »

Quote from: hasu on Thu, 28 February 2013, 05:34:25

komar, really really great work!

I confirmed your app on a tad old xubuntu, I need to install 'python3-minimal' and 'python3-tk' packages for this test.

Writing layout binary from GUI worked flawlessly and quickly. Editing layout also worked well except for one problem.

My problem is that I couldn't define 'layer action without scancode', I just want to change layer without sending any key. I know your gh60.lay has such key on key 57 but I couldn't change existent key(key 61) into such normal layer change key.

Thanks, hasu!
I fixed that today, hoping I had introduced the bug just recently, but as it appears it's in 0.1 as well.

My code's in a mess now, so I'll publish the fix later. For now, a workaround is to write 0 as the scancode.

Acanthophis · « **Reply #266 on:** Sun, 03 March 2013, 19:36:32 »

Will this layout be possible?

Link

WhiteFireDragon · « **Reply #267 on:** Mon, 04 March 2013, 01:47:22 »

No. PCB does not support holes where those switches are. Why would you want something like that anyways? It's too much of a mix between 1x, 1.25, and 1.5x sizes on the bottom row.

Acanthophis · « **Reply #268 on:** Mon, 04 March 2013, 07:32:18 »

It's just a individual thing I guess. Suits my typing style and my hand/finger size/reach pretty good, at least I would like to think so.
Well, I guess normal Pure layout will suffice.
Thanks for an answer, though.

Acanthophis · « **Reply #269 on:** Mon, 04 March 2013, 09:41:50 »

Any thoughts on the FN layer?
Is it feasable to program it this way?

Link

kmiller8 · « **Reply #270 on:** Mon, 04 March 2013, 10:20:05 »

Ok so after a few weeks of testing my board. I have two things.

1. Would there be a way to make the Fn button act as a windows key on a short-press, but FN for holding? I have it ingrained in my head to use the left windows key for Win+R and many other windows functions.

2. \\ not sure if that's the cap or the FW \\ It's been like that since day 1 \\ these were brand new blue switches \\ so I don't think there's anything wrong with them, but I could be mistaken since I seem to be the only one with the \\ problem. \\\\\\\\\\\\\\\\ granted, now i can type \\ twice as fast

jdcarpe · « **Reply #271 on:** Mon, 04 March 2013, 10:25:08 »

Quote from: Acanthophis on Mon, 04 March 2013, 09:41:50

Any thoughts on the FN layer?
Is it feasable to program it this way?

Link

Yes, that layout is totally possible. hasu's firmware includes all of those media keys, etc., you just have to change your keymap.c to reflect the layout of your Fn layer.

mkawa · « **Reply #272 on:** Mon, 04 March 2013, 10:28:28 »

another call for someone who needs a board to help with software dev (especially with konar's keymapping front-end)

komar007 · « **Reply #273 on:** Mon, 04 March 2013, 10:33:09 »

Quote from: kmiller8 on Mon, 04 March 2013, 10:20:05

Ok so after a few weeks of testing my board. I have two things.

1. Would there be a way to make the Fn button act as a windows key on a short-press, but FN for holding? I have it ingrained in my head to use the left windows key for Win+R and many other windows functions.

We'd need an idea how to implement it... Since you only know if the press is long or short after you've released the button:D
The best I can think of is to delay the keypress until either the threshold has been reached or the key was released before it, in which case press and release events have to be sent instantly.

Quote from: kmiller8 on Mon, 04 March 2013, 10:20:05

2. \\ not sure if that's the cap or the FW \\ It's been like that since day 1 \\ these were brand new blue switches \\ so I don't think there's anything wrong with them, but I could be mistaken since I seem to be the only one with the \\ problem. \\\\\\\\\\\\\\\\ granted, now i can type \\ twice as fast

Which FW and OS are you using?
There's a bug in the linux hid support which causes something like that (2 codes mapped to this key). Not sure this is the case here, since this rather caused immediate key release behaviour...

EDIT: on the other hand the solution to this problem is to send both codes at once, which may, in turn, cause windows to freak out and see the keypress twice... I don't know.

jdcarpe · « **Reply #274 on:** Mon, 04 March 2013, 10:33:16 »

komar, does your GUI absolutely require Python 3.0? I need to have Python 2.7 installed on my machine, so I didn't know if your code would run under 2.7.

komar007 · « **Reply #275 on:** Mon, 04 March 2013, 10:36:05 »

I'll see how compatible it is...
BTW, is there a problem with having more pythons? I have 6 on my machine...

mkawa · « **Reply #276 on:** Mon, 04 March 2013, 10:40:27 »

UGH PYTHON

kmiller8 · « **Reply #277 on:** Mon, 04 March 2013, 10:50:31 »

Quote from: komar007 on Mon, 04 March 2013, 10:33:09

We'd need an idea how to implement it... Since you only know if the press is long or short after you've released the button:D
The best I can think of is to delay the keypress until either the threshold has been reached or the key was released before it, in which case press and release events have to be sent instantly.

Which FW and OS are you using?
There's a bug in the linux hid support which causes something like that (2 codes mapped to this key). Not sure this is the case here, since this rather caused immediate key release behaviour...

EDIT: on the other hand the solution to this problem is to send both codes at once, which may, in turn, cause windows to freak out and see the keypress twice... I don't know.

haha, 'twas a silly thought, I'll just have to adjust to windows on the right side.

FW, I have no idea, whatever WFD sent, and Windows 32-bit Pro.

I'll try to update the FW to something newer and see if that fixes it

kmiller8 · « **Reply #278 on:** Mon, 04 March 2013, 11:22:05 »

Edit from a phone: it's broken. What did I do wrong ;_;

EDIT 2: reflashed the pre-release FW \ \ \\ \ \ \ \ \ \ \ \ \ it's better now i guess

)

komar007 · « **Reply #279 on:** Mon, 04 March 2013, 11:59:11 »

Unfortunately, I've just found out it doesn't work on win7 and 8:/
I have no idea about windows, so I ask for help.

The best we can do is to provide a windows version of the hid communication layer using winapi.
The worse solution is to make libusb and pyusb work correctly in windows 7 and 8. This has to be doable either in the code or by changing something in the way the driver is installed, since libusb and pyusb are supposed to work at least on windows 7... But again, I need someone who can do this, or I'll be searching blindly.

kmiller8 · « **Reply #280 on:** Mon, 04 March 2013, 14:28:57 »

Quote from: komar007 on Mon, 04 March 2013, 11:59:11

Unfortunately, I've just found out it doesn't work on win7 and 8:/
I have no idea about windows, so I ask for help.

haha ok. well I guess I'm stuck on this pre-release \\ \\ \\

komar007 · « **Reply #281 on:** Mon, 04 March 2013, 15:16:02 »

The FW works all right, you just cannot program the layout.
If you have a winxp somewhere, you can use it to program the keyboard, and then use it in 7.
Though this FW is buggy in general;)

IvanIvanovich · « **Reply #282 on:** Mon, 04 March 2013, 15:24:16 »

You can still use the method from hasu to reprogram for now on Windows, though you really need to read, read, re-read, and read again the instructions. I used that with the winavr and flip on win7 x64 and it work fine.

I'm not a programmer and don't really understand much of those things, but would it be possible to just make a front end gui to work on winavr for this? Seems it would be a good idea to take advantage of that since it is working reliably and just make it more user friendly.

Acanthophis · « **Reply #283 on:** Tue, 05 March 2013, 01:18:20 »

Quote from: jdcarpe on Mon, 04 March 2013, 10:25:08

Quote from: Acanthophis on Mon, 04 March 2013, 09:41:50
Any thoughts on the FN layer?
Is it feasable to program it this way?

Link

Yes, that layout is totally possible. hasu's firmware includes all of those media keys, etc., you just have to change your keymap.c to reflect the layout of your Fn layer.

Thanks, I'll look into hasu's FW.

nano · « **Reply #284 on:** Tue, 05 March 2013, 06:00:55 »

I'm very new to GH and this is my first Post

I just wanted to let you know that your code on github is awesome. I've studied it for the last two days and all I can say is wow.
consistent style
well documented
great use of modern c

Continue like this!

John Fourth · « **Reply #285 on:** Tue, 05 March 2013, 21:55:50 »

Any boards left to help with development? I have C/C++ and Linux experience.

Thanks,
John

WhiteFireDragon · « **Reply #286 on:** Tue, 05 March 2013, 22:02:59 »

Ask mkawa. I think he offered his to anyone that has more time to help with development since he doesn't have the time himself. I have 2 with me because I did the SMD soldering, but I don't own them. Maybe I can ask the owner if anyone can borrow it, as he doesn't know too much about keyboards.

agodinhost · « **Reply #287 on:** Thu, 07 March 2013, 19:06:51 »

Guys, would not be better to create another topic to the subject "GH60 Keymapper software" ?

tricheboars · « **Reply #288 on:** Fri, 08 March 2013, 00:51:13 »

Quote from: nano on Tue, 05 March 2013, 06:00:55

I'm very new to GH and this is my first Post

I just wanted to let you know that your code on github is awesome. I've studied it for the last two days and all I can say is wow.
consistent style
well documented
great use of modern c

Continue like this!

this post makes me happy. keep it wutang mother****ers!

vorn · « **Reply #289 on:** Fri, 08 March 2013, 01:19:39 »

Sorry for off-topic... tricheboars: has there ever been a Wu keycap GB?

GordonZed · « **Reply #290 on:** Fri, 08 March 2013, 20:09:44 »

Quote from: vorn on Fri, 08 March 2013, 01:19:39

Sorry for off-topic... tricheboars: has there ever been a Wu keycap GB?

Yeah, there is right now. The ZOMG! keycaps GB, I'd ask Ragnorock if there are any extras 'cause orders juat closed last week.

komar007 · « **Reply #291 on:** Sat, 09 March 2013, 05:07:34 »

I noticed that the whole board is moved a bit to the right in the original Poker case. It seems to be about 0.5mm. Can someone confirm this? I'm wondering how much the cases differ from each other. I also know that after a ton of tinkering with the mounting holes it seems the horizontal distance between the left hole and the switches is slightly different than in the Poker PCB, so I need to correct that.

BTW: has anyone tried to put the GH60 into a different case (aluminium anyone...)? Does everything fit all right?

mkawa · « **Reply #292 on:** Sat, 09 March 2013, 11:20:17 »

i didn't notice this when i test fit it into a pure case... do you have a sample of both cases komar?

komar007 · « **Reply #293 on:** Sat, 09 March 2013, 11:29:47 »

No, unfortunately just Poker.

mkawa · « **Reply #294 on:** Sat, 09 March 2013, 11:51:38 »

ok i think i have one for you, pending danucf's approval (he had dibs)

mkawa · « **Reply #295 on:** Sat, 09 March 2013, 12:44:23 »

yep, i've got one for you. can you pm me komar?

komar007 · « **Reply #296 on:** Sat, 09 March 2013, 12:55:37 »

WhiteFireDragon · « **Reply #297 on:** Sat, 09 March 2013, 13:36:36 »

Quote from: komar007 on Sat, 09 March 2013, 05:07:34

I noticed that the whole board is moved a bit to the right in the original Poker case. It seems to be about 0.5mm. Can someone confirm this? I'm wondering how much the cases differ from each other. I also know that after a ton of tinkering with the mounting holes it seems the horizontal distance between the left hole and the switches is slightly different than in the Poker PCB, so I need to correct that.

BTW: has anyone tried to put the GH60 into a different case (aluminium anyone...)? Does everything fit all right?

I think I mentioned this earlier in the thread that they're all off by just a fraction of a millimeter, including the USB connector pads. I noticed this when I designed my aluminum case. I'll give you the coordinates of exact location of the mounting holes when I can get back to my workstation computer.

komar007 · « **Reply #298 on:** Wed, 13 March 2013, 05:24:48 »

What do you think of making the top-left hole a slot too? I know this will require some twiddling to get the position right and it may allow some movement after fixing the PCB to the case, but since it seems it fits the Pure case just fine and is off in the Poker case, maybe it makes sense...
What about the vertical position of the board? I can see it fits a little to high in the case. Is it just me?

OrangeJewce · « **Reply #299 on:** Wed, 13 March 2013, 08:33:37 »

Quote from: komar007 on Mon, 04 March 2013, 15:16:02

The FW works all right, you just cannot program the layout.
If you have a winxp somewhere, you can use it to program the keyboard, and then use it in 7.
Though this FW is buggy in general;)

Komar, do you need someone to work on a windows API to get the firmware to work on win7/8? Or is it the firmware itself that's broken on anything over XP?

Cheers,

GH60 rev. B w/ ali's case	\|	Cherry G80-3000 HFU/05	\|	IBM Model M (51G8572)

News:

Author Topic: GH60 prototype betatesting [Call for layouts, GUI testing] (Read 184926 times)