Doesn't happen on my cpc664! Tested and confirmed.
I've seen it on cpc464 with newer keyboard and cpc6128, and even plus. But not my cpc664!
What do you mean by keyboard clash??
/Ygdrazil
Quote from: arnoldemu on 09:09, 24 October 11
Doesn't happen on my cpc664! Tested and confirmed.
I've seen it on cpc464 with newer keyboard and cpc6128, and even plus. But not my cpc664!
The circuits are identical, so what makes the 664 different?
Bryce.
Maybe this http://www.cpcwiki.eu/index.php/Programming:Keyboard_scanning#Keyboard_clash (http://www.cpcwiki.eu/forum/../index.php/Programming:Keyboard_scanning#Keyboard_clash)
/ygdrazil
Quote from: Ygdrazil on 09:41, 24 October 11
What do you mean by keyboard clash??
/Ygdrazil
Quote from: Bryce on 09:42, 24 October 11
The circuits are identical, so what makes the 664 different?
Bryce.
I don't know, but there is something strange going on when I press multiple keys. Perhaps clash should happen, but it doesn't.
Pressing I,O,P together doesn't produce another key. If I try and hold down more keys, it then automatically releases some other keys which are already pressed.
It's not possible to hold down all keys like it is on the other keyboards.
Perhaps the keyboard is broken in some way, but every key does work on the keyboard... I tested that.
Maybe they added some sort of "intelligent filtering" in the Firmware?
Bryce.
Perhaps the 664 keyboard was faulty for a reason.
I mean, those are notoriously faulty yet clashless.
Quote from: Bryce on 12:48, 24 October 11
Maybe they added some sort of "intelligent filtering" in the Firmware?
Bryce.
The firmware can't overcome the faults of the hardware. But maybe the keyboard itself is modified.
I have 6128 without keyboard clash - due to a Siemens keyboard (which looks obviously different).
Quote from: MacDeath on 06:58, 25 October 11
Perhaps the 664 keyboard was faulty for a reason.
I mean, those are notoriously faulty yet clashless.
Where shall be the gain in being faulty? Arnoldemus keyboard is a "gain of function" mutation, not a "loss of function" ;-)
I have a nice tool under FutureOS (Show Keys), that will show you the keyboard matrix semi-graphically. Use this to verify your keyboards (ESC ends!). Link:
http://www.colorado-boys-muenchen.de/users/futureos/files/FutureOS_Program_Sources.zip (http://www.colorado-boys-muenchen.de/users/futureos/files/FutureOS_Program_Sources.zip)
Quote from: TFM/FS on 20:14, 25 October 11
Where shall be the gain in being faulty? Arnoldemus keyboard is a "gain of function" mutation, not a "loss of function" ;-)
I have a nice tool under FutureOS (Show Keys), that will show you the keyboard matrix semi-graphically. Use this to verify your keyboards (ESC ends!). Link:
http://www.colorado-boys-muenchen.de/users/futureos/files/FutureOS_Program_Sources.zip (http://www.colorado-boys-muenchen.de/users/futureos/files/FutureOS_Program_Sources.zip)
I used my own program that performs a similar function.
And this is how I verified my CPC664 keyboard is unique.
Quote from: TFM/FS on 20:11, 25 October 11
The firmware can't overcome the faults of the hardware. But maybe the keyboard itself is modified.
I have 6128 without keyboard clash - due to a Siemens keyboard (which looks obviously different).
The siemens keyboard is a special adapted keyboard? or it's the same look as a normal 6128 but it has a different manufacturer?
Quote from: arnoldemu on 20:57, 25 October 11
The siemens keyboard is a special adapted keyboard? or it's the same look as a normal 6128 but it has a different manufacturer?
It's completely different. I menetioned to point to the keyboard of the CPC, and to "unblame" the non-keyboard-part of the CPC :)
Quote from: TFM/FS on 22:51, 25 October 11
It's completely different. I menetioned to point to the keyboard of the CPC, and to "unblame" the non-keyboard-part of the CPC :)
As I thought... I am sure it makes coding much more enjoyable.
It would be possible to correct the clash in Firmware by filtering the results: ie: If keybaord input > 4 and = (the four keys that happen in a clash) then ignore the fourth character.
If the Siemens keyboard uses a completely different matrix layout, then it must have different firmware too? Or is it a PS/2 keyboard connected to a standard 6128 via a controller?
Bryce.
Quote from: Bryce on 08:49, 26 October 11
It would be possible to correct the clash in Firmware by filtering the results: ie: If keybaord input > 4 and = (the four keys that happen in a clash) then ignore the fourth character.
Haha, so which one of the four keys is the wrong one? Imagine first you "see" two pressed keys then it's four. All you can do it guess! And I rather take a Guinnes than to guess.
Quote from: Bryce on 08:49, 26 October 11
If the Siemens keyboard uses a completely different matrix layout, then it must have different firmware too? Or is it a PS/2 keyboard connected to a standard 6128 via a controller?
Bryce.
A bit a rewiring and all is fine :-)
Got to remember the original series of 664 had faulty keyboards... so many got changed, perhaps with better 3rd party companies ones, hence superior and clashless..?
The quality of the keyboard doesn't effect whether the clash happens or not. Only the scanning method and keyboard architecture influences whether there will/can be clashes. If they didn't change the scanning or architecture, then even the best keyboard will still produce clashes. I can see no electronic differences between the 464 and the 664, so they have obviously changed something in how they scan the keys.
Bryce.
Na die werden hald eine Hand voll Dioden eingesträut haben.
English: May they have added some diodes.
They could have, but they didn't, otherwise they would be in the schematic.
Bryce.
Quote from: Bryce on 20:17, 27 October 11
They could have, but they didn't, otherwise they would be in the schematic.
Bryce.
Let me know if I'm wrong. There is one schematic, and there are two kinds of keyboards? :-X
Quote from: TFM/FS on 00:12, 28 October 11
Let me know if I'm wrong. There is one schematic, and there are two kinds of keyboards? :-X
I think it's the actual keyboard itself. But I will not open it to find out more because I would not be able to put it back together again and I don't want a broken CPC664.
The other possibility is that the keyboard is starting to fail (although all the keys work on their own)... because if I press lots of keys together the keyboard decides it doesn't want some of the keys I already pressed to be down and releases them for me (in terms of what the cpc sees).
Really, others who have cpc664 should verify my findings.
I tested so far:
CPC464 old keyboard, CPC6128, CPC664, CPC464 new keyboard, KC Compact. All except my CPC664 show keyboard clash.
Well according to the schematics, and mainboard pictures on the wiki (I don't have a real 664 to look at), the 664 is identical to the 464 and uses exactly the same components. If they did use diodes, then they must be mounted under the keyboard, which I very much doubt.
what I suspect they did was to block ALL multiple presses in the firmware, by only accepting one key input per scan, ie: the first key detected is the only one that gets processed. This would stop clashes, but also have some negative effects when you are beating on multiple keys while playing your favourite game.
Bryce.
Well, as Arnoldemu suggested, tests on 664 CPCs are needed, anything else is just grey theory.
Volunteers please ;-)