Shop OBEX P1 Docs P2 Docs Learn Events
What to do while waiting to buy a Propeller Chip? — Parallax Forums

What to do while waiting to buy a Propeller Chip?

LoopyBytelooseLoopyByteloose Posts: 12,537
edited 2006-03-31 08:31 in General Discussion
I have become hopelessly compulsive the past few days [noparse][[/noparse]since the news of the new Propeller chip].

Seems like I cannot stay away from the Forum.· I realize that this product is at least a month away from release.· Any suggestions on coping?

Seems like reference books to Multi ·tasking· processing, State Machines, and other such topics might be appropriate.
I have only a vague understanding of semaphores.

Of course, I could just go to Starbucks and study Chinese all day, but that wouldn't be as pro-active.
·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

Post Edited (Kramer) : 3/1/2006 4:46:39 AM GMT

Comments

  • John R.John R. Posts: 1,376
    edited 2006-02-27 11:50
    Kramer;

    You haven't been paying attention. It's not multi-tasking. It's multi-processing!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.

    8 + 8 = 10
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-02-27 14:57
    Fair enough. I corrected the original -- I do understand it is Multi processing, just a Freudian slip, I guess.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • Paul BakerPaul Baker Posts: 6,351
    edited 2006-02-27 18:30
    Well you could go find some textbooks to understand multi-processing. But you'll have to be prepared to forget some of the things you read about. The reason is the propeller was designed from the ground up, it does not specifically inherit anything from anything else that has ever been done.

    Certainly you'll find similarities with aspects of it to other exsiting systems, but there are quite a few novel methods of doing various things, and sometimes preconceptions can be your undoing. If you read about how the industry handles a specific issue and you find no analogy on the propeller, then you could set youself back by thinking too much in the box.

    For instance there is no debugger for the chip, thats right no debugger, the task of debugging a multi-processor in a deterministic way (what's done during debugging vs. whats done when running it) is a difficult thing to accomplish. You instead have an object you can load in to send data to a television, for a Basic Stamp person, this will be much like the debug terminal, but for an SX person use to having an on the chip debugger where you can step by step travel through the program, this will seem like a step backwards. But once you contemplate step by step debugging a single processor while the other 7 are running full speed, and you have inter cog communications occuring, you can see this becoming a nightmare.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·1+1=10
  • John R.John R. Posts: 1,376
    edited 2006-02-27 23:48
    With someting this revolutionary, you might be better off going in blind and having some fun at first.

    Then after you have some type of "feel" for the processor, some background may be in order.

    I think back on bringing some members of the IT staff up to speed on object and message based programming. The "kids" who didn't have much experience caught on quickly. The "experienced" programmers struggled, because they were trying to compare the "new stuff" to the procedural stuff they were used to.

    Sometimes "preconceived" ideas just get in the way.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.

    8 + 8 = 10
  • BeanBean Posts: 8,129
    edited 2006-02-28 02:43
    John,
    I agree. When I was trying to learn OOP I had a heck of time. Your right I kept trying to figure out "How would I do this without objects". Then I swear in one day a light went on in my head (or maybe a light went out) and I understood it in about an hour.

    Bean.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "SX-Video·Module" Now available from Parallax for only $28.95

    http://www.parallax.com/detail.asp?product_id=30012

    "SX-Video OSD module" Now available from Parallax for only·$49.95
    http://www.parallax.com/detail.asp?product_id=30015

    Product web site: www.sxvm.com

    "Never under estimate the importance of being nice when you don't have to be."
    ·
  • Mike GreenMike Green Posts: 23,101
    edited 2006-02-28 20:17
    One advantage of a relatively small codespace (500 instructions) is that you are forced (hopefully) into thinking in terms of functional objects (like OOP). For a short time (or indefinitely) one cog becomes a specific device or function that another cog can make use of by leaving parameters in some global area and/or setting some I/O bits or whatever. Much like OOP, the trick is in partitioning a system into relatively small, mostly independent pieces. These pieces can be tested individually with a simulator or instrumented "testbed" until they mostly work, then integrated. Nice.
  • John R.John R. Posts: 1,376
    edited 2006-02-28 23:12
    Gee, in some ways·this sounds strangely like "structured programming".· I find it interesting how some key concepts evolve and change names and even some meanings, but as fundamental concepts (encapsulation, division of labor, task specific, do one thing well, etc.) still remain relevant.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.

    8 + 8 = 10
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-01 04:57
    Well it is 512x32bits - not that small.
    But Leonardo Da Vinci said, 'Small spaces discipline the mind'.

    The whole beauty of the BasicStamps, the SXes, and this are that they are small enough that you can see the system's attributes. Today's desktops are so vast that you might just spend years on one aspect [noparse][[/noparse]or data entry].

    I do appreciate the comments regarding Objects, OOP and so forth. There seems to be an obvious paradigm shift from the the SXes focus on interrupts and threaded multi-tasking is quite useful, but now I see that it isn't the only way to juggle multiple tasks.

    The Propeller really will add depth to the educational value via filling out the array of products.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • MikeKMikeK Posts: 118
    edited 2006-03-01 23:30
    Paul Baker: Been there, done that. On IBM 3033, 3081 and 3090 series mainframes in the 1980's. The stuff we were debugging ran on the bare hardware (our stuff was a mini-OS), so we were limited to using the hardware's debugging tools. Writing to any sort of hardcopy output device was not available. We made extensive use of trace logs in memory that we could dump to tape (once, after the failure) to see what was going on (off line).

    In an MP environment, you've got to be aware that any debugging code (any code, actually) you put in can alter the inter-processor timing enough to cause a problem to disappear, appear less often, appear in another form, etc.

    The only time I ever worked 24 hours straight (7am-7am) was to debug a problem on a multi-processor. It wasn't a pleasant experience. :-(

    I'm not familiar with the Propeller's architecture, but if one cog can alter another cog's memory (or shared memory), and that memory can contain executable instructions, the observed behavior can be interesting, to say the least.

    Mike
  • Paul BakerPaul Baker Posts: 6,351
    edited 2006-03-01 23:43
    Mike, yeah thats why I said to some people it would appear a step backwards. As per the propeller, memory in each cog is not accessible by other cogs. The shared memory is accesable by all cogs so the possibility of thrashing memory exists. The thing I think will mess people up the most is people using a pointer when they meant to use an immediate value, but since this is mainly an issue in the source and not destination of the argument, this should lead to unexpected data rather than thrashing an inappropriate memory location.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·1+1=10
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-29 06:34
    My guess is that this coming Saturday, April 1st will be the official 'on-sale' day for the Propeller. Just a hunch.

    The waiting has made me consider immediately buying the Hydra as well.
    It appears that I would be significantly missing out in terms of Video support without it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2006-03-29 16:27
    Not so -- even if April 1st was not on a Saturday we would not launch a new product on that day. Funny, I got a really stern note from a Nuts & Volts reader claiming that the Propeller is an elaborate April Fool's day joke. Yeah, you know us Parallax folks: nothing to do but trick people....

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-30 07:45
    The next logical choice is April 17th as people will be finished with their Income Taxes [noparse][[/noparse]ugh!] and looking for refund checks.

    Frankyly, I am a bit impatient. Thankfully, the Forum has provided enough information to keep me studying and to consider just exactly what i would like to do with it.

    Somehow, I sensed that someone wouldn't believe it. After all, April 1st is for fools, isn't it? Or is the day meant to be just plain, good fun for everyone?

    Maybe I will try to place half my order with Mouser [noparse][[/noparse]as they already listed the Propeller] and half with Parallax.
    Then I can be surer that I will get an early delivery.

    This will be quite an investment for me, but since I want to become proficent in both Assember and Object Oriented languges, what platform could be better? And who could better teach me? After all, I don't have anyone around that teaches Electronics in English here.

    Maybe I will program my BasicStamp to provide a count-down to April 17th.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • JavalinJavalin Posts: 892
    edited 2006-03-30 12:22
    Of course if you just wanted to learn ASM and OO - then you could get an SX and Javalin!

    smile.gif
  • Paul Sr.Paul Sr. Posts: 435
    edited 2006-03-30 15:18
    Kramer said...
    The next logical choice is April 17th as people will be finished with their Income Taxes [noparse][[/noparse]ugh!] and looking for refund checks.

    The flaw in your logic is that, for the most part, only those of us who OWE wait to file until the last minute - everyone I know who was due $$ back has already received and spent it!

    I have a feeling that it doesn't matter one iota when the Propeller is released - those of us who "gotta have one" will buy one regardless of taxes....or much of anything else!!
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-30 15:59
    I suppose your right. I am looking to dip into 'vacation savings' to spend a bit.
    It was either that or go to Bali alone. Not much fun to sit on a beach without someone.

    I have the SX, don't have the Javilin as I have never been quite sure about Java. Of course the Propeller really moves the whole paradigm of hobby microcontroller into a new and uncharted directions. I just want to enjoy the adventure.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • Paul Sr.Paul Sr. Posts: 435
    edited 2006-03-30 16:27
    Kramer said...
    Not much fun to sit on a beach without someone.

    Laptop, Propeller, Hydra.... Sounds like GREAT beach toys to me..... Don't need no stinkin' wife and kids.....

    Now, if my wife gets her hands on this post, I will have to kill you....
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-30 17:58
    I already have a first project for Hydra and it is a quite ambitious video game.
    It would use color [noparse][[/noparse]at least four] and require rotation of a duodecahedron in any and all directions.


    Why go the the beach when you need an extension cord?
    I am already in the tropics and it is hot outside.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2006-03-31 08:31
    Looks like Monday April 3rd is the day we can Pre-Order.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
Sign In or Register to comment.