Shop OBEX P1 Docs P2 Docs Learn Events
Who do you think will buy the Prop2 ? - Page 7 — Parallax Forums

Who do you think will buy the Prop2 ?

145791016

Comments

  • Tubular wrote: »
    That's really interesting, Zane. I have to agree with the hidden FPGA costs, and have always regarded the P1 is solving the kind of issues you'd normally turn to an FPGA for, but with just the right things locked down (for speedy development).

    Recently I've been noticing several gadgets that really need "the propeller treatment". One is my 2011 car radio which struggles to do any two things too closely, eg changing radio station and then volume, because its busy animating the display in between. So keypresses are 'lost'

    The other one in past couple of weeks is the thermal camera scene. Keep finding artificial limits such as not being able to record movies longer than 5 minutes unless you have a PC connected, and the Flir I bought has an MSX feature which overlays visual edges with the lower res thermal image. There's an offset between the two sensors which the software should be able to shift/compensate for, but it just doesn't.

    These things and more would be trivial for P2 to sort out.

    The commercial side will go with what is cheapest, if it means using a $3.00 M4 ARM over a $8,00 P2 then it's not even a contest.

    P2 Sales will probably be in the lower volume higher margin markets where the cost of the micro is irrelevant. Like CNC, telemetry/data acquisition, PLC, robotics sectors.
  • Cluso99Cluso99 Posts: 18,066
    rod1963 wrote: »
    Tubular wrote: »
    That's really interesting, Zane. I have to agree with the hidden FPGA costs, and have always regarded the P1 is solving the kind of issues you'd normally turn to an FPGA for, but with just the right things locked down (for speedy development).

    Recently I've been noticing several gadgets that really need "the propeller treatment". One is my 2011 car radio which struggles to do any two things too closely, eg changing radio station and then volume, because its busy animating the display in between. So keypresses are 'lost'

    The other one in past couple of weeks is the thermal camera scene. Keep finding artificial limits such as not being able to record movies longer than 5 minutes unless you have a PC connected, and the Flir I bought has an MSX feature which overlays visual edges with the lower res thermal image. There's an offset between the two sensors which the software should be able to shift/compensate for, but it just doesn't.

    These things and more would be trivial for P2 to sort out.

    The commercial side will go with what is cheapest, if it means using a $3.00 M4 ARM over a $8,00 P2 then it's not even a contest.

    P2 Sales will probably be in the lower volume higher margin markets where the cost of the micro is irrelevant. Like CNC, telemetry/data acquisition, PLC, robotics sectors.
    Or lower volume markets where development time costs are considered. It's much easier to have 16 processes running separately than one big process where timing interactions and interrupts mean a way more complex solution.
  • The landscape of micros and embedded systems has changed vastly since the SX and Propeller 1 days. This is what I would think needs to happen to give the Propeller 2 a chance.

    * Put together a message of why this microcontroller needs to exist. The propeller is a minority in this market, you need to send a clear message of why it is useful compared to an ARM/PIC/AVR. This would be done with example software and a youtube videos. Documents help as well, but a video is required to get peoples attention. If you cannot present a clear/fun message of why people/companies should use this microcontroller, they wont.
    * C needs to be supported in Propeller tool. The only worse uphill battle for a minority microcontroller is having a minority language to learn. People learned SPIN because at the time of the Propeller 1 because it was one of the most powerful micros, so people grit their teeth and learned it. With C being the dominant language of every other micro, no one is going to take the time to learn SPIN. Make C a priority and not some 3rd party unsupported feature.
    * Find something different and exciting that can be done with Propeller 2, create it, and show it off.
    * Develop educational packages around the Propeller 2. Create lots of course material that center around the Propeller 2 that can fill 1-2 semesters, also create material that will cover middle school, high school, and college. If this were up to me, I would create a Prop 2 package that had a breadboard, VGA/HDMI out (or option for an LCD display), an additional micro to handle USB for keyboard (unless USB keyboards can be used on the Prop2), and create a program where people can write code right on the device, run programs from that code, and use the breadboard to drive electronics tutorials like blinking LEDs, buttons, sound, etc.
    * Finish it. This thing has been lingering for years, wrap it up and finish it. Everything else is a moot point if it is in eternal production.

    Now if you look at this list and see that you will struggle with one or more of these items, then it may be time to consider the idea of sun setting this project. Even with X amount of years sunk into R&D, will it ultimately be worth the additional cost to bring it to market if it cannot make the impact it needs? Today this is a much more competitive space than it was 10 years ago.

    Anyway this is my 2 cents.
  • Heater.Heater. Posts: 21,230
    Most of your bullet points I can agree with. They about the marketing and support of the finished product.

    The second point I agree with. C is vital for a wider audience. Except I would not include the Propeller Tool in the picture, The Propeller tool is what it is. A beautifully simple IDE for Spin and the Propeller. No need to mess it up with complications of multiple language support.

    Arguably those that want C/C++ rather than yet another language would also prefer to use a familiar IDE, not yet another IDE. Developing an IDE is a large non-trivial task, there are many already, why reinvent that wheel? There is already SimpleIDE for C on the Propeller. Many products have adopted the familiar Arduino IDE to attract users. Not my favorite idea. Personally I think building Propeller 2 programming support into Microsoft's Visual Studio Code would be a great way to go. Open Source and Cross platform, all the heavy lifting is already done.

  • RaymanRayman Posts: 13,797
    I would love to have Visual Studio support... Would be very nice to do C++ for P2 inside Visual Studio.
  • evanhevanh Posts: 15,126
    We already know why the Propellers exist. The primary market is dependable educational packages for schools. JT, none of that is any different to when Prop1 launched. C hasn't just sprung into existence in the last decade. C isn't as friendly for teaching. That makes a revised Spin the logical choice for first deployment.

    Heater,
    Are you saying I can download an RPM or DEB of Visual Studio?
  • Heater.Heater. Posts: 21,230
    Ah, yeah, I did not mean Visual Studio. I meant Visual Studio Code:

    https://code.visualstudio.com/

    Microsoft's greatest wonder. Open Source and Free and cross-platform. Based on the Electron platform by Github. In turn based on Google's Chrome engine and node.js.

  • evanhevanh Posts: 15,126
    Ugh! Five tracking scripts detected with scripting disabled. That goes up to ten tracking scripts with just the host domain enabled. Display formatting is all messed up ...
  • Heater.Heater. Posts: 21,230
    Well, it is Microsoft we are talking about :)

    When I disable JS then https://code.visualstudio.com/ looks just fine. Except the big screen shot image is missing.
  • evanhevanh Posts: 15,126
    Ah, I also have across the board third-party domain blocking on by default as well. :D
  • evanhevanh Posts: 15,126
    Just saw a new term named - "HIVE is not von Neumann because of the sparseness of its data and its ability to simultaneously perform different processes on different areas of memory simultaneously," Trung said. "This non-von-Neumann approach allows one big map that can be accessed by many processors at the same time, each using its own local scratch-pad memory while simultaneously performing scatter-and-gather operations across global memory."

    That sounds like a Prop2 to me. :)
  • evanhevanh Posts: 15,126
    Except for the missing terabytes of MRAM.
  • evanh wrote: »
    We already know why the Propellers exist. The primary market is dependable educational packages for schools. JT, none of that is any different to when Prop1 launched. C hasn't just sprung into existence in the last decade. C isn't as friendly for teaching. That makes a revised Spin the logical choice for first deployment.

    That would send the wrong message that Parallax and the P2 are for hobbyists and schools and telling industry to drop dead.

    You want to show that Parallax is serious about getting sizeable commercial wins - that means a C/C++ development package ready to go when they roll out the P2.

    Really imagine having a both at Design West and only having Spin and Pasm. No one will take the P2 seriously if they do that.

    Here's the thing Parallax has to make back the 5 million plus they've invested in the P2 by the time we get silicon. That means getting design wins on the commercial side from multiple industry sectors. That means having industry standard tools to be used with the P2 that engineers can show their PHB/management that the P2 is worth looking at. That means C/C++ not Spin.

    Schools can't deliver that sort of volume unless you're Apple.

    If they can't make the money back, there will be no P3.

    Spin can always be rolled out later.


  • evanhevanh Posts: 15,126
    It is only what I said, schools is number one, nothing more. Do you really think Parallax solutions have ever competed head on with other processors? Yet here they are, doing yet another chip.

    I'm sure there will be effort put into rebuilding the tool set after the Prop2 is done - When it's actually a stable product. There's no reason why it can't eventually have C support, but industrial uses can't be the top priority.
  • Heater.Heater. Posts: 21,230
    evanh,
    Just saw a new term named - "HIVE is not von Neumann because of the sparseness of its data and its ability to simultaneously perform different processes on different areas of memory simultaneously," Trung said. "This non-von-Neumann approach allows one big map that can be accessed by many processors at the same time, each using its own local scratch-pad memory while simultaneously performing scatter-and-gather operations across global memory."
    It does not sound like the P2 to me.

    So called non Von Neumann architectures have been around for ever. Normally such parallel processing architectures are all about getting more computational speed.

    They are all limited by Amdahl's law:
    https://en.wikipedia.org/wiki/Amdahl's_law

    If I wanted maximal computational speed I would not choose a P1 or P2. That is not what it's about.

    No, to my mind the Propeller is about the total independence of processes, like code that runs in COG. Such that it has totally deterministic timing and is not messed up by whatever other processes you happen to want to run.

    It's about the tight coupling between processing and IO.


  • evanhevanh Posts: 15,126
    I was joking a little because the description was basically a multi-point switching mechanism. The Prop2 HubRAM is, of course, built around exactly that (including the "local scratch-pad memory" - CogRAM).

    Here's the link - http://www.eetimes.com/document.asp?doc_id=1331871& You'll see there they are not aiming for fast parallel computation but rather massive data processing instead, ie: make deeper/wider analysis of the data collected from tracking scripts. It's presumably got a lot of physical similarities in hardware though.
  • Heater.Heater. Posts: 21,230
    Thanks for the link.

    There is a lot of hype in there.

    When it comes to the tight linkage between processing and IO I guess it has no similarities to the concept of the P1 or P2.

  • evanhevanh Posts: 15,126
    Right, nothing tight at all. Many hierarchical levels, potentially including equivalent network structure for multi-cabinet switching, ie: a lot of fibre cables. I/O to them would ultimately be live feeds from data collection systems like Stingrays.
  • drewldrewl Posts: 1
    edited 2017-06-10 09:48
    I guess i'll give my 2cents seeing better informed or interested parties have already pointed at their issues. Personally I came to parallax products through the purchase of a pimoroni propeller hat for the raspberrypi, Great combination for makers coz the prop adds so many of the extra gpio in a configurable manner that the pi lacks off the bat. Anyway back to the question of a prop2.. I would buy it because it sounds cool, and has significant capabilities over the original prop. When it comes to robotics or timing i believe that a prop2 should have a significant advantage over the competition that use ARM (beaglebone, et al) if like other posts have said C,C++, and a form of assembly/bytecode programming is provided. About the IDE, well the parallax IDEs haven't been to shabby in the past although Eclipse integration would be nice seeing it is becoming a standard platform to build a programming environment around, although a user friendly interpreted language a forgiving structure can't hurt (python seems to have traction in sci/edu fields). The joy of the prop1 is knowing how easy things can be to implement in certain scenarios. For the maker i would still recommend they look at a prop1 although those with a wallet or love will buy a super edition prop2. I think that the prop community could gain by strengthening relationships with educators like raspberry pi or beagle bone and making easy to use hats/capes and software system for extra functionality and control. Many young folk rely of a pi3 these days as their development platform!!. Glad i still got an old phenom in the corner and a fair laptop.
  • First off, let me just say that I love the folks at Parallax, the products that they provide, the efforts that they promote through education, and this wonderful forum community, and what I have to say, I say out of love, so please allow me to be brutally honest.

    As a customer, which buys Parallax products from time to time, there have been several serious disappointments, which I have experienced and which I would like to share, just to give Parallax one customer's viewpoint, so that when they roll out the P2, there will not be other customers with a similar viewpoint.

    It seems to me that over the years, there has been many times that Parallax started off in a very good direction, but failed to maintain traction and momentum, such as with plain old documentation, the OBEX, the Gold Standard object, and the SimpleIDE. I cannot imagine a large business or corporation paying an engineer to seek out vital information in an online forum, when key aspects of a product and examples of use should be readily available, for rapid development of ideas and applications. Thorough documentation and a vast supply of good examples should be ready upon P2 deployment, so that any engineer can quickly utilize the P2, without seeking answers from forum members. Additionally, commercial customer's should be able to reasonably expect a well developed IDE for building and maintaining code, to go along with their new found "wonder" chip. When SimpleIDE first came into existence, I thought to myself, "WOW this is great, I hope that Parallax maintains the momentum on this software push", but then, everything basically stopped, and the same holds true for the Propeller Tool. The P2 really should have a well developed IDE, with all the bells and whistles, that developers love to use. Before commercial sales will commence, electronic engineers must be impressed with the abilities and capabilities of the P2, but in my opinion, the programmers responsible for creating, writing, and deploying P2 code, will have to be equally impressed with the IDE.

    I suppose that most large businesses and corporations will seek out the path of least resistance, in getting their product to market, and I believe that Parallax should put a huge amount of emphasis, on providing their potential commercial customers, with the proverbial "EASY" button.

    Admittedly, I am not that knowledgeable about the P2, or it's future capabilities, but I would hope that it will have enough memory to support a full blown G-Code interpreter, as well as enough memory to run a CNC machine, with many bells and whistles. I still vote that Parallax should pursue the CNC market, and with that in mind, I would suggest that Parallax consult with Mickster about the possibilities, because I am now of the belief, that he truly seeks out cutting edge technology for machinery and interfacing, and I believe he would be a very valuable asset, as a consultant, should Parallax choose to pursue this avenue.

    Providing that the P2 has significantly grown in memory, I cannot think of a better application for the Propeller, other than CNC and automation.

  • I would like to suggest the retrocomputing / retrogames community as a possible target for the P2. As far as I can tell the P2 may have the features and power to fully emulate old home computers and maybe also some old arcade games. Imagine a C64 or ZX Spectrum fully emulated with the P2, or Pacman, Frogger, Galaga, etc. I would like to see if a MAME-like emulation could be done with the P2.

    About the language support, I strongly suggest that C / C++ will be available from day 1. PropellerGCC has already some support for P2 (I believe it partially supports what was called P2-HOT) so I think that an initial support could be done very quickly: update the instruction set and build programs that runs with the most easy execution profile (hub-exec ?).

  • jmgjmg Posts: 15,140
    macca wrote: »
    About the language support, I strongly suggest that C / C++ will be available from day 1. PropellerGCC has already some support for P2 (I believe it partially supports what was called P2-HOT) so I think that an initial support could be done very quickly: update the instruction set and build programs that runs with the most easy execution profile (hub-exec ?).
    Yes, there are already alpha-quality pathways for P2 with C, and those will further rapidly improve.
    There is more time than many seem to think, before production volume devices hit the distributors, and C is closer than many also think.

    Short answer: P2 C is not going to be a problem.


  • jmgjmg Posts: 15,140
    idbruce wrote: »
    Admittedly, I am not that knowledgeable about the P2, or it's future capabilities, but I would hope that it will have enough memory to support a full blown G-Code interpreter, as well as enough memory to run a CNC machine, with many bells and whistles....

    Providing that the P2 has significantly grown in memory, I cannot think of a better application for the Propeller, other than CNC and automation.
    Present target is 512k, so yes, "the P2 has significantly grown in memory", and it should easily support a full blown G-Code interpreter, as well as enough memory to run a CNC machine, with many bells and whistles.
    The AVR is one reference for G-Code, (one that has already hit design saturation), but that is an 8-bit MCU, with 32k code, so is a fraction of P2's capability.
    Porting saturated AVR code is not likely to be fun, but there may be C-code on thinks like RaspPi that manage the higher movement calcs that can more easily be ported.
    P2's COGs + smart pins are far enough from AVRs Timers/interrupts, that a clean slate will be better for lowest level code.
  • jmgjmg Posts: 15,140
    idbruce wrote: »
    ... The P2 really should have a well developed IDE, with all the bells and whistles, that developers love to use. Before commercial sales will commence, electronic engineers must be impressed with the abilities and capabilities of the P2, but in my opinion, the programmers responsible for creating, writing, and deploying P2 code, will have to be equally impressed with the IDE.
    Ah yes, but 'IDE' means many things to many people, which is part of the problem with 'IDE debates'

    Some think a Syntax capable editor, and a batch-build are a fine Instant Development Environment.
    Almost any Programmer's Editor delivers that.

    Others want their IDEs to be able to 'find declaration' and offer parameter hints when typing. ie be more Language focused (Usually C)

    Still others expect their IDE to include a Debugger & Simulator, and Debug is one area Parallax are very poor on right now.


  • RaymanRayman Posts: 13,797
    I'd love to replace my MAME-PC with a P2... I've got an actual arcade machine that I'm running with a PC:

    http://www.rayslogic.com/hardware/donkeykong2mame/donkeykong2mame.htm

    Spent a whole day recently upgrading the thing and had a heck of a time finding a sound card that is compatible with modern PCs and works in DOS mode.

    Anyway if GCC could compile MAME somehow and we could make a front end launcher for it, that would be great. You could probably fit my whole game library (now on 3 DVDs) on one SD card these days...
  • jmg wrote:
    Still others expect their IDE to include a Debugger & Simulator, and Debug is one area Parallax are very poor on right now.
    I don't see that as a problem. A software debugger is nothing more than a crutch, and a poor one at that. My debugger is an oscilloscope, which gives me real-time data about how my program is running.

    -Phil
  • Some could argue that an oscilloscope is a crutch, and it would be difficult to track down a bad pointer with an oscilloscope. Personally, I prefer adding prints when debugging code. Each programmer has his own preference on how they debug code. I've watched people that really know how to use a debugger, and they do can do some amazing things with it. I've never been able to figure out how to analyze a stack dump using GDB, but I've seen other people that could track down the source of a problem using that technique.
  • Dave Hein wrote: »
    Some could argue that an oscilloscope is a crutch, and it would be difficult to track down a bad pointer with an oscilloscope. Personally, I prefer adding prints when debugging code. Each programmer has his own preference on how they debug code. I've watched people that really know how to use a debugger, and they do can do some amazing things with it. I've never been able to figure out how to analyze a stack dump using GDB, but I've seen other people that could track down the source of a problem using that technique.
    Even though I've written several debuggers I have to admit that I also usually just insert printfs. One of the first things I do with a new piece of hardware is setup a UART for debug output. Partly that is because I got used to that approach when dealing with hardware that didn't support a debugger.

  • I have found the P2 the easiest microcontroller to debug.
    With a streamer that can capture pin states(and smart pin status) to hub as a built
    in "logic analyzer" to its debug interrupt that allow single step.
    Throw a diagnostic cog into the mix and I have been able to solve ALL software issues
    and even some FPGA ones too.
    I have even used P2 to debug/test "other" microcontrollers.
    P2 PASM is a joy to code.

    When we get P2+SPIN2 the world will be a better place. :)


  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2017-06-11 04:00
    I can't understand why Spin is being dismissed so lightly sometimes and it complements the Prop so nicely too and is very easy to learn and very easy to get started even if you have little programming knowledge. It's a pity though that the compiler couldn't be resident on the chip itself and interactive as I find having my Tachyon Forth resident on the chip a very powerful development and debugging tool/environment with immediate access to all symbols and functions by name, and that is only a small part of it all.

    Couldn't we do something similar with Spin2 in augmenting the bytecode interpreter with a compiler and an interactive console? For sure these extra functions would have to be loaded into serial Flash initially but just imagine if all the Parallax boards were "ready to Spin" from the moment you unpack it and plug it in (and powered from USB please). Even your phone or tablet could serve as a simple terminal then or over Bluetooth just as I do now with Tachyon.
Sign In or Register to comment.