TransKey-II in development

[+mytek]

A pinout of the TK-II would really help..

 

Thank You!

 

Douglas

 

Schematics for every Atari related gizmo I've made are on my website. Use the link in my signature to take you there, then click on the TK-II tab. Or better yet use the TransKeys link .

[rockdoc2010]

Hello 576XE

 

 

We've had something like that:

 

 

They probably didn't sell enough to redesign them for the XL and XE series.

 

Sincerely

 

Mathy

 

PS it's the CX85

 

 

But the switch matrix is mechanical and the drivers are needed to use this thru the joystick port. but you could in theory wire the switches into a 1200XL or likewise keyboard during breakout.. ie wire the CX85 swithces to your mylar..

 

Douglas

[rockdoc2010]

 

Schematics for every Atari related gizmo I've made are on my website. Use the link in my signature to take you there, then click on the TK-II tab. Or better yet use the TransKeys link .

 

 

Found it and i am in grossed in fingering it out thanks. I thought the schm files were for your mini boards but it seems that the most of your mini boards are connector space..

 

Thanks

Douglas

[rockdoc2010]

BTW,

 

What is the key capacity? can i use a 104 key? Atari only had a 54 key board!????

 

Douglas

[Sleepy]

Hi,

 

I´ve an 1088XEL with TKII on board, using german keyboard-layout.

 

This works fine, the only little thing attracted my attention is that, with customizing the "Y" and "Z" to the german layout, the graphical symbols of the keys naturally changed, too.

 

That the "Y" is linked to the "Y"-key as well that the "Z" is linked to the "Z"-key is fine, but unfortunately the graphic symbols also changed; this is a little bit irritating when using the grafic font to create a graphic.

 

 

My question: Is it possible to modify the TKII-firmware this way that "Y" and "Z" are still linked to the changed layout, but with the graphic characters remain on the original place on the keyboard?

 

So that graphic character...

is linked to the "Y"-key

 

and graphiccharacter...

is linked to the "Z"-key?

 

I try to take a look on the firmware provided on the mytek-TK-site, but the linked (free) version of flowcode 8 doesn´t open the .fcf-file (only fcfx and fcsx files are accepted). Is there an older free version avaiable which can open .fcf?

 

Sleepy

Edited March 10, 2019 by Sleepy

[+mytek]

I can fix this in the next firmware revision due to come out for 1088XLD support. I never even gave this any thought at the time when I first provided German keyboard mapping.

 

My version of FlowCode is very old (V4.54). It's an expensive program, and I never found any reason to justify the expense of upgrading. Disappointing that they've not provided a conversion or import that would work in the newest versions.

[Sleepy]

Great!

 

My version of FlowCode is very old (V4.54). It's an expensive program, and I never found any reason to justify the expense of upgrading. Disappointing that they've not provided a conversion or import that would work in the newest versions.

 

Maybe in the full version; I tested the free version. In the decription was nothing written about restricted fileformat; just limited number of components etc.:

 

Maybe compatibility for older versions is not a basic feature...

 

Sleepy

[Sleepy]

Planning the concrete layout for an individual (german) keyboard, a second question came up:

 

If the graphical character moves to the correct place as shown in post 480, does the according international character moves to the original place, too?

 

So that graphical and international character are at the original place, only "Z" and "Y" are changed?

 

Sleepy

Edited March 28, 2019 by Sleepy

[+mytek]

Since the international characters are limited as provided by Atari, I was only able to match up with certain ones, others are SOL. But anything that was possible was appropriately remapped. And although I thought I had missed this, the CTRL graphics characters were also moved along with the Z and Y keys.

[+mytek]

A newer version of the firmware has just been released, and for the first time ever there is now an on-line manual that fully describes it.

 

Main TK-II Page

On-Line Manual

 

In this version I weeded out all the quirky and unreliable aspects, including some legacy stuff that just never worked like it was suppose to. For most users you wont see a tremendous difference in usability, although there is one new feature that might prove useful, a delayed power-up reset release. This can be toggled ON or OFF by pressing CTRL+ALT+D (the present setting can be viewed by pressing ALT+D). When this feature is turned OFF, the Atari powers up with a normal length reset cycle. However when this feature is turned ON, reset is held for a couple of seconds longer. This extended reset allows the PS/2 keyboard ample time to initialize, thus enabling the Atari to see 'held' console keys such as OPTION (F7) to disable built-in Basic, or START (F5) to boot from cassette. Please note that this feature will not work in a XEGS due to it's rather unusual reset circuit that will not allow an extended reset.

 

ALT+Q will now toggle a 'Quick Cursor' movement when using an XL or XE OS. This setting is temporary and will not be retained when the system is powered down, but will stay in effect once activated, or until ALT+Q is pressed again.

 

A couple of hard coded macros were eliminated due to their somewhat dangerous side effect of erasing a program in memory in certain situations. These were the former ALT+D = DOS (now reassigned), and the former ALT+N = NEW (now reassigned) macros.

 

And if you have a 1088XLD, ALT+N = Next Disk which is the command to activate the Swap Disk Latch when booting from a multi-disk set, and ALT+M will toggle through Mouse Port Selection that determines which joystick port is assigned to the PS/2 mouse (this is a non-volatile setting retained even when the system is powered down).

 

Edit: I forgot to mention that the default Num Lock state upon powering up can now be configured. To do so, simply press the CTRL+ALT+NumLock keys to toggle the setting (ALT+NumLock will show its present configuration).

 

I hope you find this update useful.

Michael

 

[ebiguy]

That’s a good news and thank you for this release.

 

I am wondering how ALT-M coexists with the mouse selection daughter board in the 1088XEL?

Do the bicolor led reflects the new status when ALT-M is pressed for example ?

Both keep last status when powered down so who wins on power on ?

[+mytek]

5 hours ago, ebiguy said:

That’s a good news and thank you for this release.

 

I am wondering how ALT-M coexists with the mouse selection daughter board in the 1088XEL?

Do the bicolor led reflects the new status when ALT-M is pressed for example ?

Both keep last status when powered down so who wins on power on ?

Only a 1088XLD will respond to either the ALT+M and ALT+N functions. The XEL will act like any other Atari with a TK-II board installed and nothing will occur by pressing those key combinations. As to why this is, you'll have to read the manual and specifically refer to the last section on how things electrically are connected, and detected.

 

[+mytek]

I had to do some rewrites on the TK-II manual, especially concerning the following...

 

 

I think that says it much better then what I had before, but still keeps it simple and to the point.

 

And for a more lengthy explanation...

Quote

Delayed Reset Release Feature
If set to ON (toggled by CTRL+ALT+D), the reset is held for a extended period (measured in seconds) when the unit is first powered up, giving enough time for the PS/2 keyboard to fully initialize, and thereby have it's key presses registered by the Atari OS following the release of the extended reset. A useful aspect of this in action, is to disable internal Basic on an XL/XE system when powering up by holding Option (F7).

This feature is not required by either the 1088XEL or 1088XLD, and not even available from the later.

This is the nice thing about having the manual be an on-line version, making it much easier to keep it up to date.

 

[+mytek]

Just to add a bit of trivia...

 

When I first conducted tests with an extended reset, there was a visual side effect that was quite disconcerting. Antic while waiting to get reset, would sometimes power up with a very scrambled set of parameters that produced some random and not very pleasant display artifacts. It would get sorted out once reset was released, but until then it was quite a mess.

 

The solution was to do a two step process. So when initially powering up, reset would be quickly asserted and almost immediately released, and then quickly reasserted once again for the longer delayed reset pulse. This allowed Antic to come up properly with an entirely black screen display defined, and still allowed for an extended reset to occur. The double reset pulse happens so fast, that it goes by without any noticeable visual effect.

 

Sometimes you have to resort to trickery to get what you want .

 

[+mytek]

I don't know if anyone noticed, but Lotharek now sells his own PS/2 keyboard adapter called the AKI which is a remake of a very old project (predates the TK-II). It's pretty cheap at 17 USD (of course shipping & handling still applies). I don't know much about it other than some very old info from the original AKI German site. Apparently this board's firmware is based on the Dark AKI updated code, so not sure if there have been any functional changes. Unfortunately no manual seems to exist on Lotharek's website, so I guess it's assumed that users will learn through pushing keys to see what happens  . I did try to do some Google searches under "Dark AKI" but that didn't turn up anything useful.

 

Here's a comparison photo, with Lotharek's keyboard product on the right, and one of mine on the left.

 

 

TK-II On-Line Manual

TK-II Versions

Lotharek AKI Product Page

 

[+CharlieChaplin]

Well,

 

the AKI was originally available from two german Abbuc members, the webpage was here: defunct AKI link but the link to the AKI project no longer exists. Attached are some old schematics, GIFs and JPGs I have about AKI.

 

Afaik, Dark AKI was a newer project (or updated/upgraded AKI) from Poland. Details were once available at www.atarionline.pl  alas, I can no longer find them there. Maybe you have more luck or @Kaz atarionline.pl can provide us with more info or a link...?!?

 

 

AKI.ZIP

[+ZuluGula]

http://www.atari.org.pl/forum/viewtopic.php?pid=206227#p206227

You will need transaction. 

[+CharlieChaplin]

Thanks for the link! To download something from atari-area you must be logged in, to log in you must register - not everyone does this (or is able to do it, due to language barrier), so I downloaded and zipped the stuff I found there...  (if I missed something, let me know).

 

 

 

Atari AKI Full.zip xaki103 plus coldboot.rar Dark_AKI.zip

[+mytek]

CharlieChaplin thanks for the files ? .

 

I spent some time on the DARKAKI manual today, and translated it from Polish to English: DarkAKI Functions v.1.1.en.pdf

 

It looks pretty good, and virtually everything came across in the translation except for: dot.ąęćłńóśŜź , which I haven't the faintest idea what that means .

 

 

Edit: So DarkAKI definitely had a lot of new features added to it from what I remember of the original AKI. And unlike the TK-II (and the 1990's Transkey) they've opted to use function keys F1-F4 like the ATARI800 emulator (and similar to Altirra), reassigning the 1200XL F1-F4 keys to F9-F12. So this seems to make the TK-II out of step with the emulators in the use of the first 4 function keys.

 

Well back in 1990 it just seemed logical to do it the way I did, especially considering that those emulators hadn't even been created yet, so they obviously couldn't be followed as a standard at that time. If anything Transkey should have set the standard for the emulators, since it predated them concerning the use of a PC keyboard for Atari purposes. And being in the USA in the 90's there were a lot of 1200XL's around which had the 4 function keys labeled F1-F4, and of course that lived on inside the OS in the later XL and XE systems. So my first thought when I was designing the Transkey, was that those 1200XL functions should be assigned to the same function keys on a PC keyboard as they were on the 1200XL... and that's what I did.

 

When I developed the TK-II, I hadn't used any of the new emulators, having just come back to the Atari's after a 25 year absence. And I didn't even know about the AKI project. Oh well it is what it is, and I have no plans to change it.

 

So this brings me to the point where I have to ask, why did the emulators chose to put the console key functions on F1-F4 and push the 1200XL F1-F4 functions up to F9-F12? I really can't see the logic in doing it that way, but I am curious as to the reason why .

 

[flashjazzcat]

2 hours ago, mytek said:

why did the emulators chose to put the console key functions on F1-F4 and push the 1200XL F1-F4 functions up to F9-F12?

Think about how much software (aside from the OS screen editor) actually makes use of the 1200XL function keys (which aren't found on any other machine) to this day, and you might have the answer. START, SELECT, OPTION and RESET are frequently used, so it makes sense to put them on the most conspicuous and easily accessible function keys.

[+ZuluGula]

4 hours ago, mytek said:

It looks pretty good, and virtually everything came across in the translation except for: dot.ąęćłńóśŜź , which I haven't the faintest idea what that means .

Basically, those letters have the same meaning as: a,e,c,l,n*s,z, with exception of ó which is the same like u. 

https://en.m.wikibooks.org/wiki/Polish/Polish_pronunciation

[+mytek]

16 hours ago, flashjazzcat said:

Think about how much software (aside from the OS screen editor) actually makes use of the 1200XL function keys (which aren't found on any other machine) to this day, and you might have the answer. START, SELECT, OPTION and RESET are frequently used, so it makes sense to put them on the most conspicuous and easily accessible function keys.

Yes that makes sense to some degree. But with the PS/2's Function keys normally grouped in 4's, it really isn't very difficult to still see the pattern of the console keys when placed in the middle group as they are in the Transkey or TK-II as seen below. And unless someone has very short arms, it would seem that the middle row would be just as easily accessible, maybe even easier since it is centered and closer to the user .

 

 

And as was done in the original AKI and Steve Carden's remake the KRH.

 

 

So basically the PS/2 keyboard to Atari Interfaces were all based on the scheme I show above, with the following exceptions on the AKI/KRH:

 

F11 - INVERSE (got moved to tilde key below Escape, since this key has symbols on it not shared by Atari)

F12 - BREAK* (got moved to the actual BREAK key on a PS/2 keyboard)

 

*Note: I believe the reason that the actual BREAK key wasn't used on the PS/2 keyboard, was related to the difficulty in decoding it properly. I struggled for a long time trying to figure out how to reliably do so, but I finally did and decided it made more sense to assign it as such, instead of using F12.

 

Quote

Footnote 1) Actually, the "Pause/Break" key does not have a break code in scan code sets one and two. When this key is
pressed, its make code is sent; when it's released, it doesn't send anything. So how do you tell when this key has been
released? You can't. OP Note: but I did figure out a way to do so, although it's so strange that I forget exactly how I did it.

 

Of course one could argue that the 1200XL button layout also shows a somewhat emulator like arrangement of the console keys vs. the F1-F4 keys  .

 

 

So from a upgrade hardware standpoint. Until the DarkAKI version hit the streets, all PS/2 keyboard interfaces were pretty much based on the same Atari related assignment of the F-Keys as was used in the Transkey/AKI. When I started the TK-II project in May of 2015 (yes it's been a while), I was doing so in response to a thread that was started around that time period where several people were looking for either an original Transkey or an AKI, neither of which were being manufactured any more. So I stepped in and decided to do a re-boot of the Transkey based on a modern all-in-one MCU to satisfy this need, and logically based the F-Key assignments on those previous two keyboard interfaces.

 

Bottom line in my reasoning behind all this, was thinking that someone that is looking for one of these interfaces, has A) maybe used one before, or B) possibly seen literature that described it in detail (there was a review article about Transkey in START magazine Volume 5 Number 3 November 1990 by Matthew Ratcliff). So it seemed best to not make too many changes to what was done, unless it was something better (i.e., use actual BREAK key instead of F12), or place something in a more reasonable location (i.e., tilde as INVERSE key). I also incorporated a few subtle changes to how MACROS would be recorded and accessed based a bit more on the AKI method of doing so, but also added a small twist of my own in the process.

 

So now to address the elephant in the room...

Now that Lotharek has brought out the DarkAKI PS/2 Interface, should I rearrange the function keys on the TK-II to somewhat match?

 

It would be a lot of work, since it's just not firmware changes only, but also would create a considerable amount of document changes (i.e., TK-II Manual). And of course it introduces the possibility of creating new bugs in the code. Also keep in mind that the present code methodology was not meant to be rearranged easily, and has hooks for some of the other features associated with those keys that could get affected by the move (more potential bugs).

 

And of course anyone that has gotten used to the present layout, would have to relearn the new arrangement, although if they use emulation they would probably welcome that .

 

 

14 hours ago, ZuluGula said:

Basically, those letters have the same meaning as: a,e,c,l,n*s,z, with exception of ó which is the same like u. 

https://en.m.wikibooks.org/wiki/Polish/Polish_pronunciation

Interesting, but I fail to see how that really helps me to decode that leftover fragment from the original Polish manual.

 

Do you have some idea about what it means?

 

[flashjazzcat]

Thanks for reminding me that Help is on F10. I can never remember that.

 

[+mytek]

2 minutes ago, flashjazzcat said:

Thanks for reminding me that Help is on F10. I can never remember that.

Your welcome .

 

Speaking of HELP, on a PC F1 is normally used for that function. Perhaps I see a bit more logic behind what was done and why, since the DarkAKI also is using F1 as HELP for the Atari. And I believe this is the same on the ATARI800 emulator. Is this true for Altirra as well?

 

[flashjazzcat]

4 minutes ago, mytek said:

Is this true for Altirra as well?

Not by default: F1 is turbo mode. But you can completely remap the keyboard to suit yourself.

 

(Help is on F6, BTW, so I have unfortunately gotten into the habit of holding F6 and tapping RESET [F5] to get into the U1MB setup menu).

 

Is it not possible to completely abstract TK-II key assignments from the core code by using an indirection table (i.e. the firmware acts on tokens derived from a look up on the translation table)?

Edited December 18, 2019 by flashjazzcat