But there are times when you want to keep the source code not available, unless to developers. The pinball for example has a game contest built in and would not want the methods to win the contest to go public. an extra charge SDK would include the reference and the entire source code, the reference being the library routines and how to develop for it. Unless only the base driving routines remain open source and available?

_________________
"My biggest dream in life? Building black plywood Habitrails"

Agreed - full open source isn't always what you want. I just wanted to clarify that it isn't incompatible with for-profit.

As you say, if you have "secret sauce" then you need at least part of your project closed-source.

You could use someone else's LGPL-licensed CPU core or library in such a project, but not a GPL-licensed one. As a user of IP, you'd have no problem with MIT or BSD licensed components.

(As a producer of IP, those simpler licenses allow your work to be improved without the benefits coming to all your users, which might be undesirable.)

Cheers
Ed

The secret sauce is what I would want to keep secret, the BSD licensed I would have no problem in those on the redistribute or letting it be known on a title screen that those are being used and sources to obtain them. Same with improvements. The thing is also, for users who do get the source, I want them to be able to sell their work for profit as well.

EXAMPLE: I sell my videogame system and a develoepr wants to use the hardware to make their own game utilizing the secret sauce code form me and the public code. I have no problem with them selling their work for profit. The secret sauce is the extra cost SDK kit from me which gives them the right to use my secret routines in their work and sell for profit while keeping those secret as well.

License fun bizarre...

Tony is confusing tonight.

_________________
"My biggest dream in life? Building black plywood Habitrails"

@Nightmaretony

Wouldn't an acceptable solution to the "secret sauce" problem be to put things like contest rules in config files? That way all the code can be open and you get to keep things like number of points required to win secret (of course, if your game rules are complicated enough to require a scripting language you're back in trouble). IIRC that's how iD Software opens their games - they release all the source code, but game assets like the graphics and sound are "closed".

I could definitely do that for other projects. For the pinball, I am using libraries and the contest occurs only in one game mode, so that game mode can be kept proprietary. It checks for secret scoring combinations and does mathematical calculations to create a checksum for the player to write down as proof that they achieved the contest. (for example, if the score ends in 70 and the secret combination was achieved in a 3,5,2,6,7,1,4 order, then the checksum would put up a message of NITE KREW as the message to send as proof of the achievement. Simply looking up a data table and sending me THE BIG NUTTED SQUIRREL wouldn't win, for example.

But yup, I can keep certain small portions proprietary. got no issue there and leave the rest as open and available.

Those who purchase the develo0pment kit would have full access to all code and assets, of course.


No wonder the licenses be confusing, got my own confusing needs here

_________________
"My biggest dream in life? Building black plywood Habitrails"

What should I do? I don't know anything about VDHL but I know how to program and I'm willing to learn. Should I get the Gameduino with the GPL license so I can basically have a video chip or should I get the bare FPGA and create my own from scratch? I would basically be trying to create something better than someone who programmed for expert companies and some of the code for the Gameduino may be written in Forth.

What I want it for is a GPU for a hobby computer. There are seven days left to decide.

(warning, long post. For your question about using Gameduino as a GPU for a hobby machine, I think it looks good, and straightforward - see the hardware page.)


Hmm, your earlier message tells us that some of the samples have some GPL content:

but I see the kickstarter page (where you can more or less buy one for $53) says that all the code is GPL. I suppose that would mean the FORTH runtime, the game library, and the samples. (Edit: James clarified that there is no runtime. So, if you use the game library, you're in GPL territory, similarly if you mutate one of the samples for your project. But if you work it as a graphics adaptor there's no GPL aspect.)

My conclusion would still be that this is fine for all purposes: if you had secret sauce you'd need to put it in configuration files which you license separately from your game source files.

Getting any kind of FPGA board seems like a good idea to me, if you think you'll put in the effort to learn how to drive it. Certainly easier than making your own board: that can come later. You'll have a choice of schematics, verilog or VHDL, and you can mix and match, and there's quite a lot of examples out there to make use of and learn from.

As far as I can see, the Gameduino hardware includes the FPGA and has a JTAG port so you can re-program it - all you'd need is a cable, and confidence that you can re-load the original design if you need to.

Working with the Gameduino in the intended way, as a graphics coprocessor, means writing values to its memory over the SPI bus. You don't have to work in FORTH on the (very interesting looking, minimal, 16-bit) J1 CPU and indeed you don't even have to read/write values directly if you use the GD games dev library on an Arduino.

So, it's an Arduino addon, it's an FPGA dev board, it's a graphics subsystem and it's a FORTH CPU playground. And because it's open source, you're invited to make your own.

If you want to test yourself out on verilog or VHDL you can download the free 'ISE webpack' software from Xilinx and go through an exercise of designing something, and simulating it. (Or, if you're happy with command line working, use Icarus verilog which is open source and a great deal smaller.)

A great advantage is also that the FPGA design time board layout/debugging is already done.

_________________
"My biggest dream in life? Building black plywood Habitrails"

I posted two articles in the programmable logic section so it will be a learning experience for me too.

There is also a board you can build to interface the VGA to the FPGA board on a breadboard:

http://devbisme.webfactional.com/blogs/ ... fpga-board

The Gameduino is available again and it looks like it is here to stay!

http://www.seeedstudio.com/depot/gamedu ... p-860.html

For those of you who didn't get in on the Gameduino's first product run, it is now available at Seeedstudio.com

The only thing you will need is a joystick shield kit from Sparkfun (it is assumed you have an Arduino):

http://www.sparkfun.com/products/9760

i would have to say that is quite a nice addition to a 6502 and with a very impressive performance at that.

Try as it may, the newer chips are looking to get a lot more performance that you would really have to think of upgrading to get the newer set. A very simple 6502 and another processor to boot for simple performance values like usb would really feel outdated.

It hasn't been adapted yet for the 6502 or other 65xx processors. One of the users here is making a board with a compatible shield (daughterboard) for the Arduino and if I remember correctly it is FPGA so it would probably be faster than a stock 6502.

I bought some books to re-learn the 6502 but I'm not going to pay $200 for a kit to get started when I can get a Maple Leaf for less than $50 and have a faster processor and more memory without all the problems of finding help and products that work with old hardware. I would have liked some extra help with the 65XX line but the Maple Leaf line will have to do; I live in a world where newer is better and there is more help for the newer.

My Gameduino came, I soldered the Sparkfun Joystick, I bought an Arduino and an LCD that I have to solder to some pin headers. I'm thinking about learning to make the LCD work and my next step is to build my own shield (daughter board) and wire an IBM XT keyboard to a shield where I can work with the Arduino and LCD in making it more computer like. Then I will start work on making that work with the Maple Leaf and I'll go from there. I imagine that I'll have a couple of prototype shields as I learn about different configurations.

I'm learning about SMT and Microchip and I've been acquiring wires and parts to work with. I may go off in different directions but I'll have fun doing it and I encourage others to look into Microchip because they have a 16 bit chip coming out that only costs $2.00 dollars and runs at 60 Mips.

And you will always find Moore's Law coming into play for increasing computer power at cost.

You can also look into the Propeller for serious multitasking fun.

If you must go Upwinger in that regard, do take care and enjoy the journey. We all find our own places and cliques to work with.

_________________
"My biggest dream in life? Building black plywood Habitrails"

I play with Microchip controllers in my spare time. My current project is a parallel chess engine using eight 40 MIP 16-bit devices as slaves to another USB enabled master device that runs the GUI.

Keeps me quiet. Shout if you need any help.

_________________
Andrew Jacobs
6502 & PIC Stuff - http://www.obelisk.me.uk/
Cross-Platform 6502/65C02/65816 Macro Assembler - http://www.obelisk.me.uk/dev65/
Open Source Projects - https://github.com/andrew-jacobs

I had time to just skim over some of their 16 and 32-bit PIC controllers and noticed they had JTAG connections. Is their software free? easy to use? to program them.

_________________
65Org16:https://github.com/ElEctric-EyE/verilog-6502

The Windows based MPLAB IDE and the Windows/Linus MPLAB X software are both free. You can get free C compilers for all the devices but they come with some code size limitations. The easiest way to program them is with a PICKit 3 (£43 with a demo board) which connects to your PC via USB and provides in circuit programming and debugging.

To get a 24/30/33F device working yourself all you need is 6 pin 0.1" header to attach the debugger, a 10K pull up on /MCLR, a 10uF(?) tant cap. on between GND and the Vcore pin and 100nF caps on every pair of GND/Vcc/AVcc pins.

You can have an external oscillator if you need one (the internal FRC 7.38Mhz oscillator can be cranked up to almost 40 MIPs on a 33F). A simple USB interface needs another cap and a resistor on the sense pin.

_________________
Andrew Jacobs
6502 & PIC Stuff - http://www.obelisk.me.uk/
Cross-Platform 6502/65C02/65816 Macro Assembler - http://www.obelisk.me.uk/dev65/
Open Source Projects - https://github.com/andrew-jacobs