Smaller Prop

paulmac · 2007-02-02 23:13

With all the talk in "What would you want more of, cogs or RAM?" It occurs to me that a samller Propeller might be usefull.
Has anyone considered the current Prop with a smaller number of cogs? Perhaps the extra silicon could be used for more RAM...

I often find myself thinking of an application, wanting to use the Prop and realising that it is overkill. The problem then being that I would have to learn another MCU so I wouldn't be "wasting" a Prop. That or trying to shoe-horn more features in to use the extra cost.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
I stand on the shoulders of giants

Paul Baker · 2007-02-02 23:16

We've kicked the idea around a bit, but the cost of getting a new chip to market is quite substantial and because of that it wouldn't be much cheaper than the current chip.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer

Parallax, Inc.

bassmaster · 2007-02-02 23:59

No kidding 12 bucks, and a eeprom, some wire, and resistors and your set! I love it, I find more usages for the cogs the more I play with it. sure beats my gilbarco pump simulator I wrote on the stamp with 32 bytes of ram. I had to use nibbles and bits and move the active pump status byte around in data, whew. lots of work!, the Language is easy to learn and you can do more with spin than pbasic. FYI My laptop is overkill, I never play games on it but it has 256mb video ram, never use the bluetooth, or the wired lan, or the modem, or the svideo, etc....

Post Edited (bassmaster) : 2/3/2007 12:07:18 AM GMT

rokicki · 2007-02-03 00:07

Right, I think the prop (especially with the new protoboards) finds a very sweet spot between the ARM processors, the PIC family,
the Stamps, and FPGA solutions. The language may be a tad odd, but I'll take that over the huge unwieldy and inaccessible
toolchains for some of the other solutions. It's definitely a "clean slate" solution and is very elegant for that. Easy things should
be easy, and on the Propeller, they really are.

bassmaster · 2007-02-03 12:48

Paulmac,

Enjoy the challenge, you will at the least be able to put, "Multi processing microcontroller experience" on your resume!! That alone is worth < 20 hours of your life and a few bucks for a prop, depending on your career.

Example:

I make very good income in DFW, never graduated High school, grew up, got my GED, went to college late in life, and trained myself, Ground crankshafts for 8 bucks an hour while studying, Now I have been a Sr. Software/Systems/Applicatons Engineer for the last 10 years, was a Sr. Systems Engineer at NEC, left NEC on my own for more money, they even offered my 20k more per year to stay! Electronics are where its at, Outsourcing is bad but usually happens to web and gui developers more than people like me. The more experience with hardware interfacing and languages, the more valuable you will be. I put pbasic on my resume, and It got mentioned in 2 interviews! One with Raytheon, who tried to hire me for a 6 figure job, but I did not want to relocate. Microcontroller experience is always a plus, even if your just writing code for a pos scanner.

Trust me, of all the things I have "had" to learn, Spin is a walk in the park.

It's as easy to learn as pbasic, my 12 year old son learned the basics of spin in a few hours, My 7 year old daughter is now playing with a bs2 homework board, she loves to show me flashing led's and pot's to change the brightness. I am delving into the prop ASM now, its a little more challenging than spin, but no worse then mplab asm. I'ts just a little different. If you have any C or ASM experience like me the asm and spin is easier.

Keep in mind, you do not "HAVE" to use the prop asm to make some powerfull applications.

Post Edited (bassmaster) : 2/3/2007 5:09:32 PM GMT

Dennis Ferron · 2007-02-03 18:45

Speaking of it being clean-slate, I had a heck of a time explaining what the Propeller was to someone at work the other day. He's another intern, but from a different college (Perdue).

Me: "It's a chip with an all-new architecture invented by Parallax, totally clean slate."
Him: "That's ridiculous, Parallax doesn't make chips."
Me: "That's why it was such a big deal - we all thought that too, until they unveiled it."
Him: "Oh I didn't know that; at Perdue, we use 68HC11 derivatives. We kind of look down our nose at Parallax stuff."
Me: "Maybe you should give them another look."

Lawson · 2007-02-03 19:03

I assume "Perdue" is Purdue? in Indiana? I started college there, I've since transfered to another school. They did seem to be a bit behind the times there :P

Oh yea and on the "Issue" of wasting a Prop on an application. I'm all for overkill. I made a weapon controller for my largest Battle Bot that used a basic stamp 2 and quickly ran into a performance wall. I wasted a lot of coding time working arround the performance limit that I could've spent adding more functionality and squashing bugs. After all that extra time I'm still going to have to upgrade from the stamp 2 to get the controller to do everything it needs to do. (BS2P or spin-stamp ideally)

Paul Baker · 2007-02-04 04:22

Another note about overkill, anyone with experience in embedded systems design (especially those who create systems for people who don't understand electronics themselves) know all about project creep. This is where a framework for the project is established between the designer and customer and somewhere during the design the customer comes back with "couldn't we have it do X too?". Frequently the designer has to explain to the customer that there aren't enough resources remaining to do the additional functionality the customer desires. Now in a perfect world this doesn't happen, the scope of the project is set at the outset and doesn't change. But frequently the customer doesn't fully understand exactly what they want until they start seeing the system come together. So using a system with spare capacity for a project means it can handle most project creep situations.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer

Parallax, Inc.

Brian_B · 2007-02-04 04:30

I think it's a mindset ,I have a pile of propellers here (I think they were 13.00 a piece) and I feel bad using such a awsome chip to do such little things

Brian

Bill Henning · 2007-02-04 04:53

Agreed [noparse]:)[/noparse]

Why do you think I just got a pile of SX28 chips, and five SX48 proto boards?

I find they make great I/O co-processors for the propeller! And I can add almost four SX-28's for the cost of adding a second prop - and get more than twice the I/O pins - or almost four SX48's for a TON of I/O but I CANNOT imagine hand soldering that teeny surface mount chip :-(

Ofcourse when the new propeller shows up, hopefully the current one will drop in price - and then it will be an AMAZING copro for the new prop!

Brian Beckius said...
I think it's a mindset ,I have a pile of propellers here (I think they were 13.00 a piece) and I feel bad using such a awsome chip to do such little things

Brian

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.mikronauts.com - a new blog about microcontrollers

paulmac · 2007-02-04 08:55

@bassmaster
My Prop is in the mail. I'm looking forward to the challenge. I have fully committed my resources to using the Propeller over any other MCU. No more plans of scavenging MCUs from keyboards and such (well not on a large scale anyway [noparse]:)[/noparse]. The Prop's flexibility and potential...
This will be fun; I'm looking forward to having my cars climate control MCU being able to do most of the stuff the Hydra can.

Thanks for the ecouragement and words of wisdom.

@Bill Henning
Yeah, I've been eyeing up the SX chips.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
I stand on the shoulders of giants

Post Edited (paulmac) : 2/4/2007 9:00:50 AM GMT

bassmaster · 2007-02-04 13:06

Yea! another convert!

Welcome brother..., hmm... maybe I should change my name to Father Sean.

Dennis Ferron · 2007-02-04 18:04

About the overkill thing being sort of a mindset - I'm currently trying to learn older 8 and 16 bit microprocessors like the 8080, 8086, and 6502, etc., because I believe that to truly understand where you are, you need to know where you've been. I could use the Propeller to make a whiz-bang video processor so that I could make TV game consoles, one based on each of these micros, or with a plug-in to accept a different legacy microprocessor at different times. But I keep thinking, "The Propeller is so powerful compared to these old uProcessors, I could easily offload all of the work onto it. At what point does the Propeller make the game console's legacy processor superfluous?"

Brian_B · 2007-02-04 19:51

Dennis I agree wiyh you 100% , I really think the best micro to learn is the 8052. it is the closest thing to the propeller I've seen. Another fun one is the 6809 , I have a book to learn how to program graphics in assembly on the 6809 ,really a good primer to the propeller.

Brian

Smaller Prop

Comments