I can't wait...

*is eager*

I'm working on it.... and some other stuff, too.

No anthrax included in the box? Son, I am disappoint.

IN OTHER NEWS, GRATS ON SUCCESS.

Oh, Mose, by the way...

the keyboard port on the back of the controller is kind of awkward. But now that I think about it, I would reason that the main purpose of it is to only attach it when you want to type. So it doesn't really matter.

Also, y'all might want to have a regular controller beside you to use for regular RM stuff, because KUPID stutters occasionally when going through other options. It's not super horrible, but if you like going super-fast through the options, it might pay off.

Otherwise, it's a dream, yo.

Yeah; it is a bit awkward. But the bottom of the controller is the only part that is flat - all the other available surfaces are rounded. On a rounded surface, it would be impossible to mount the PS/2 jack.

For the future models I think that the easiest way would be to add an extension cable that comes out of the side of the controller, and connects to the keyboard. That way it would be impossible for the jack to fall in to the controller, it would be easier for me to drill the plastic, and it would just be better overall.

The controller did that even before I modded it. Kind of weird, but it is a cheap third-party controller.

^_^

I haven't forgotten about Kupid. I'm pretty busy with a lot of stuff, so it might be delayed a bit. But, I am working on something....



More to come later.

*Gasp* What could be this new "Project Board" that Thutmose is working on?

Looks to me like a device that now a days, can be hard to find. In fact, I can only find less than 15 readily available at any given time. Probably a good idea. Of course, so is the RM1 Sprite Generator *hint-hint*

If you check the file name, you'd be able to see what the picture is of.

Sadly, not a DexDrive replacement just yet.

I've made a major breakthrough in this special project.

Once I do some more testing/programming, I'll try to get a video up, more photos, and make an official announcement.

I am officially announcing that Kupid 2.0 is in development. I have been working on it for months, but have been holding back releasing news until I was more confident that my design theories would work in practice (it is far, far, more technically complicated than 1.0). I am now sufficiently confident that my design will work - below are some details.

Kupid 2.0 is a complete redesign of the original Kupid, from the ground up. Rather than going the long way around and indirectly simulating button presses, piggy-backing off a PS1 controller PCB (which is a slow process both in the actual simulation and in the amount and difficulty of soldering tiny wires) and waiting for each character to be completely entered into RPGM before the next keyboard key can be pressed, Kupid 2.0 will use a much more powerful and faster microcontroller which will emulate an entire PlayStation controller on a single chip.

Further, 2.0 will include a built-in keystroke buffer allowing you to type as quickly as you can while not having to worry about Kupid catching up - it will remember every keystroke you make and execute them automatically when it is ready.


Kupid 2.0 is estimated to be approximately twice as fast as 1.0 (estimated as inputting an average 3 to 6 characters per second (i.e. average between 36 and 72 words per minute)), but will still cost about $35 when available for sale.

As development continues, more photos will be posted, as well as videos of Kupid 2.0 in action, circuit schematics, logic graphs, and source code (written in assembly - the current code which allows the MCU to communicate with the PlayStation is about 110 lines, but will likely grow to at least 1,500 by the end).

Below is a photograph of the very first 2.0 prototype (click to open larger image) and a short section of the source code.





Questions, comments, suggestions?

Questions:

The only real question that I would have is if there is any kind of ETA on it. Knowing what you are dealing with at the moment, I'm pretty much assuming it'll be done 'when it's done.' Also, will this model also need to be reset and manually set to capital A before it'll work properly?

Comments:



Suggestions:

Keyboard shortcuts, perhaps?

Pretty much. I'd like to get it (mostly) done by the end of May. But I am in the process of getting a second job, and I still have college classes to attend, among other things.

But I can't imagine that it will take me until the end of June to complete the design. So probably around the end of May/beginning of June.

It still has that same shortcoming of not truly knowing where the on-screen cursor is. So there will be times when it needs to be reset (if it gets out of sync, the cursor gets moved, etc.). But, the 2.0 model will have a reset hot-key on the keyboard that you press to internally reset the system, so you won't have to disconnect/re-connect the unit or do any other funny things to get it to work. Just press the reset key and manually move the cursor back to "A".

^_^
Shortcuts such as? 2.0 ought to be able to do just about anything.

Kupid 2.0 is going to be based around an 8 megahertz Atmel AVR ATtiny2313 microcontroller, rather than an 8 MHz PICAXE MCU (which is what 1.0 runs on). The PICAXE chip, though running on the same clock speed, executes the program more slowly than the AVR since it is compiled from a program written in a higher-level language (BASIC), while I am programming the AVR chip in assembly code - meaning that each instruction I write is executed in a single clock cycle (in this case, 8 million instructions per second).

Programming an AVR in assembly gives me full control of the chip, every feature, and the execution of the program. This allows me to optimize the program so that it runs very quickly and efficiently, much more than what would be possible on the PICAXE chip.

Atmel AVR MCUs are significantly cheaper than PICAXE chips ($8 versus $3 for a more powerful chip). And by redesigning Kupid with the AVR as the brains, I have eliminated the need for many of the wires as well as eliminating the darlington array, which cuts the cost even further.

While Kupid 2.0 is a significantly more technically complicated design, it is much more efficient, powerful, and flexible than 1.0.

Once the design is complete, I will release all technical details.

Yes. Kupid 1.0 is canceled, unless 2.0 doesn't work out. Though I am confident that 2.0 will be a success.

I wasn't happy with 1.0 for a wide number of reasons, not the least of which was the absence of a character buffer. It would also take a lot of time to make each unit, and the cost to me would be high. 2.0 is an all around better design both for me and those who will purchase it when it's available.

Well congrats to Ix being the only person in the world to ever own a commercial version of the first Kupid. ^_^

Anyways... some keyboard shortcuts I can think of off the top of my head would be a Ctrl + C and Ctrl + V for the obvious reasons. Though that doesn't change the fact that you can't really select a series of characters without using that dumb "how many characters do you want to select" function within RM1. And maybe get the Home and End keys to work. That would be kind of useful if someone was typing out a rather long dialog box. And maybe even get Shift to work like a shift key and not a caps lock.

Ultimately, these aren't all that important as long as I can just type.

Well, I can use that old scheme I thought of where you can press a hot-key to tell the MCU to store the characters you type afterward into memory, press the hot-key again to stop storing the characters, and another hot-key to paste it in.

Home & End function differently in different programs on a PC - what exactly would you want those keys to do in RPGM?

When you press shift down the keyboard sends a byte to the host, and sends another byte when you release the key.

The PICAXE had a built in keyboard reading function with limited capabilities - it could only read that initial byte, and not the second one, making it impossible for shift to work like shift does on a computer.

However, the AVR chip has no such function. So I have to program one entirely from scratch in assembly. But, that means that I will be able to read every byte sent by the keyboard, so it should be possible to make shift work properly.

While you are typing in a dialog box, it would be easy to just strike the Home key to automatically go to the top of the dialog box for proof reading. Or maybe the same thing with Page Up. And then just the opposite with End and Page Down. Something like that, I suppose.

Home Key, Page Up, and Page Down shouldn't be a problem - but since Kupid has no way of knowing how many pages of text there is, it wouldn't be possible to just press the End Key and go to the end of the text.

Well, I just found out that I got that second job. I start June 1st. I have to give a presentation at the company on the 26th of May; need to get my materials together, and finish creating the mock-up of the data system they want.

And I start classes at my new college the beginning of June too. I switched majors to electrical engineering.

Going to be very busy. Hopefully this won't put too much of a delay on the development of Kupid. I'll keep posting updates as I progress.

I haven't made any progress in 2.0 for a while, I've been rather busy with work.

But I am preparing to release the source code to Kupid 1.0 shortly. I just need to finish going through it, adding more comments, and deleting vestigial code. I'll be releasing it and all my Kupid 1.0 designs under the Open Software License ("OSL") v. 3.0, so people will be free to use, modify, and extend my code and designs for use in their own projects (within the limitations set forth by the license).

I imagine that I'll release Kupid 2.0, and DDR, under the same license when the time comes.

My seldom-updated blog is okay for what it is, but I am also developing a simple project site that will host everything for all my RPGM projects that's a bit more concise and easier to navigate. When it's done, that's when and where the source code will be released.

Thutmose's Workshop 2.0 is online (a redesign, and some added content), along with the complete source code and schematics for Kupid 1.0. Just go to "Project Pages" in the menu an look for "Source Code, Schematics, and Photos" - there's other info about Kupid 1.0 there as well (a lot of it taken from what I wrote in the Mag Wiki).

So if you're curious about what actually makes Kupid 1.0 tick, download the source files and check it out! ^_^

http://thutmosesworkshop.com