Let's Make Robots!

Time to move on?

The Arduino Duemilanove was my first microcontroller that I purchased, and has been my main micro on all of my projects, robotic or otherwise. At one point, I felt that it was time to move on from the Arduino, and search for a more practical solution, so I purchased a PicKit v2 and some PICs. I hoped that with the PicKit I would be able to program individual microchips and use them in my projects and leave them there only to program with ISCP (In Circuit Serial Programming) That was fail. I thought the Arduino was tedious, well, let's just say I went back to appreciate the "ease of use" of my arduino. 

 Now, I feel like it's time to move on again, so I looked into the Picaxe. A ton of members on this site use it, so I figured why not start there? (Compared to Society of Robots where everyone is either using the Axon, or some type of AVR.) I'm thinking about buying the Picaxe 28x Starter Kit (which runs about $50.00 USD, as much as both the Arduino and PicKit cost me.) in hopes of finding a new, better, easier, more adaptable micro that I can be proud of. 

 Any thoughts? 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
I agree on the cross-platform IDE.  I've heard that it works okay on OS X, but it's definitely not as easy as a straight install. I got it running in Ubuntu though wine and could generate binaries, but I couldn't get it to write to the EEPROM.

I also would prefer built-in flash, but a 32kb eeprom is pretty cheap.  The EEPROM adds to the parts count, but the stuff you can do on-chip  (It runs at 160 MIPS) means you usually have a lower total parts count than using a less powerful ucontroller.  Speech Synthesis, TV out, etc. 

The Propeller does cost more ($8 for the IC, and a $1.50 for an EEPROM), but you are getting a WAY more powerful chip.  So, it's either $4 for an AVR or $9.5 for a Prop. 

Documentation is okay, though not as good as Arduino or others.  You can program it in C or ASM, or use their own language, spin.  I think it's not as popular is because (1) when it initially came out, there was no C compiler, (2) it has a pretty unconventional multi-core architecture, (3) OS X support is wonky, (4) it's a lot newer than PIC or AVR. 

Of course, just because it works for me doesn't mean it works for everyone.  But I've had a lot of fun with it.  

Am I the only guy that uses a Propeller?  I think the PICaxe's are great, especially for the price, but I'm a big fan of Propellers - they're extremely fast (80Mhz) and all you need is the USB protoboard to start programming.  The built-in video circuitry is also helpful when you're doing debugging.  

 

I have a propeller, an older Serial PropStick, but I've not done much with it  yet. There are a few things that stop me from choosing it when starting a project, not the least of which is the lack of a vendor-supported IDE for not-windows. Yes, there are other options, and they work fairly well, but they are all lacking something or other for the newbie Prop programmer (aka Me).

Another thing I don't like is the lack of built-in flash memory. A seperate (usually i2c) eeprom is required for keeping a bot's programming through a power cycle. It's not much, but it's another required part.

Finally, the cost that turns me off too. I can leave an "arduino" in a robot for about four bucks. With a Propeller it's ~$10 for the chip, plus a couple-ish for the eeprom, works out to about three times the cost. Yeah, it's only $8 difference, but that's a pack of smokes these days ;)

All that said, the Propeller is a pretty snazzy item. I really do owe mine a chance. All I need is some inspiration, and a budget... Haha.

The Prop looks good, and probably can be very useful. I'd like to try it sometime. But have other bits to work with too. I'd like to put my Arduino into a bot of some sort. And work with other AVR projects. I've got a fairly fast device in the Isopod. I guess it would be good to see an app for it that might apply well. Right now, regular micros cover a number of needs for robots. If I want speech recognition or vision systems, I'd add a PC or laptop to the robot already controlled by smaller micros. There might be a niche where the Prop controls a motherload of actuators (for facial expressions) or reads a lot of of sensors (finger tip touch to recognise surfaces) but not sure yet.

I think the lack of popularity with Propellers has made the amount of documentation fall to the way side compared to that of PICs or AVRs, therefore, there is a small amount of self-taught hobbyists willing to venture into using Propellers. 

I'm not sure if there is a limited amount of Propeller documentation or not, but it's sole lack of market popularity leads me to believe that there is less user-generated documentation as compared to the more well-known AVR or PIC.  

Well that's why I bought a Pickit, to do ICSP so I could just have the chip in the bot, and not have to worry about it, also so I could program in PIC ASM, it was just so difficult to implement. 

 Also, I decided to move to PICs from the Arduino because it felt fake. Being able to program a PIC in it's assembly language, or C language feels more legitimate. (I don't know if there's any other way to explain it.) 

 I guess I was looking at the Picaxe as a better solution for rapid prototyping and proof of concepts. I heard it was easier to program than the Ardunio.

PICAXE vs Arduino, really it just depends on if you find basic easier than C. Why did the Arduino feel fake? It doesn`t have to be programmed in the IDE if you don`t like feeling coddled. You can use straight up C or even assembly and upload it to the board. Also with this hack, your arduino can act as an ICSP. The instructions are a little confusing, and you will have to make up a cable, but I use it all the time for various AVRs and it works perfectly. I had to resort to this when I upgraded my PC to one without any legacy ports.

What was so tedious trying to program the PIC?

Thanks, ezekiel, for linking to that. I'd seen it before but not for a while. Looks like it's been updated since the last time I read it.

Well I just thought that the arduino felt fake because I wasn't really understanding how the microchip was being programmed, where as with ASM, (which I didn't know could be used to program the Arduino) you have to pay attention to every detail, and you know step by step what's happening.

 The PIC was tedious because of the process needed and no so much the coding used. The setup needed for MPLAB was just very perticilar with the assembly and linking process, it was just a lot to handle with just online instructions and coming straight from the simplicity of the Arduino. 

Remember, too, that you can program a standalone AVR (some of em anyway) just like an arduino. Instead of leaving a $35-50 arduino in your bot, you can leave a $5 chip in it (an ATMega168 is like $4.11 in singles, cheaper in quantities). Depending on what hardware you have, or are willing to pick up, you can just have a six pin ICSP header on your bot, or a variation of a 3 to 6 pin UART/serial/whatever header for quickie USB programming. You can even use your Arduino and a few jumper wires to program a chip that's left in the bot.

You don't have to use Arduino to program the chip (although the bootloader makes things really simple sometimes). The AVR likes "regular" C/C++ just fine too.

Unless your goal is just to get away from Arduino/AVR entirely, in which case I can't help you. Picaxe may definitely be easier (basic vs C), but I (personally) wouldn't call it "more adaptable" than moving to a standalone AVR. "Better" is a relative term that is certain to incite a riot :)