I like the idea of the Prop and the Prop II, however I see some real issues with it ever becoming anything more than an enthusiasts niche. Unless that is in line with corporate expectations. This is not meant to be trollish, and I am not going to suggest someone use a PIC or AVR.
As someone who uses Pic and AVR's, I've been interested in the Prop for quite some time. I know the story that the fab process Parallax is using is inimicable in some way to flash. Why exactly is it not possible to move to a process that will allow flash? No one else in the business has similar technical constraints, so this has to come down to a cost issue.
It would appear to me that going with no on-board flash has saved some $$ at the expense of serious uptake by industry or the even the enthusiast crowd, (Arduino, etc). Its hard enough as it is for a newcomer to break in to the uC market, I fail to see how/why anyone in the industry would ever pick the Prop versus a PIC or AVR as the part count for Prop means automatically adding the EEPROM to the BOM. Additional part count, board space, and design work, just to get your controller to boot? Granted, it is not rocket science, but I think it is enough to make a fair number of people think 'If this is what I have to do just to get it to boot, what other quirks are there going to be, besides this 'no interrupts' thing?".
The Prop is a very cool idea, and kudos for someone actually bringing something like this to fruition. Seriously, its like a dream come true with the masive amount of power available. However, after 4 years, I'm not sure I've seen much really happen to expand the Prop's uptake outside of the enthusiast niche. Thats not a bad thing, however reality is what it is, and having 8 'cores' doesn't seem to have piqued much real interest, outside of the 'wow, cool!'.
And, I know what everyone is going to say. But it has basically out of the box keyboard, mouse and VGA.
Well, since 2006(?), this has probably been the most powerful Uc you can buy, with more tricks up its sleeve than Houdini. And yet it still seems to be about as popular as the proverbial red-headed step child.
Should this matter?
Well, only if one realizes that uptake and gross sales are what keeps Parallax afloat. Looks like Prop II has been in development since '08, and no end is in sight. I think this would be markedly different if there were real volume sales, which would translate into profits and R&D development and newer models coming online like the Prop II in much shorter timeframes. Like what is the industry norm.
I can't believe I am saying this, but perhaps this is one instance where there is too much power? For most Uc projects, the Prop is simply over-provisioned, and over-powered. This is almost more of a multi-core processor than a uC.
For the average uC project, where part count is a major consideration, there is little to justify the Prop if you only need 10-20 Mhz with an 8/16-bit'er.
If you need more, you can pick up single chip solutions which are a couple of dollars, and less powerful than the Prop, without the additional BOM, and limited memory and other Prop quirks.
This sort of leaves the Prop with a much more narrow application range into which it can market itself from a price/performance POV.
Could you build a mini-Predator drone from a Prop, from what I've seen thats a solid yes. From a PIC or AVR, well not with just one, maybe a handful.
But, how many common uC projects require up to 8 threads on a multi-core, instead of 1 or maybe 2 threads? Much smaller subset I believe, hence harder to justify.
A lot of interesting ideas in the previous pages, but ultimately not a lot that I saw that would improve desireability among those not already Propheads.
If Parallax is making decent income from the current Prop, it'd be cool to see a number of those suggestions acted upon. Really cool to see them so open also.
However if it takes 4-5 years to get too that point because there is such limited adoption, then I would rather see something more akin to a review of the successes/failures of the current Prop, with an eye specifically cast on why it did not take the uC world by storm.
Because it could have, and being years ahead of the competition is a fair rationale, but not the whole story.
At this point in time, I don't think we are still at a point where most people could use, or really need 8 cores, so not much sense in going higher. Seems thats been identified already.
I would really stop and take hard look at whatever equations are going into the decision to not have flash, and not being a more industry standard one chip solution. Pardon if this has in fact been changed.
I would also consider adding some real RAM to the device. Its great that you can get 80 Mhz from each Cog, but if you are going to starve it of RAM, then whats the point? Heck, I'd consider knocking a couple of cores off the die for the RAM space if necessary to get there. Who knows, if you did that, or changed fab process you could probably hit even higher speeds.
Just my $0.02 worth, so please consider my comments on their merits ( if any), and not as a troll for PIC, AVR, XMOS, Freescale, etc, etc.
I'll probably order my Prop dev kit in the next month or so, and it still seems like its years ahead of everyone else in many respects.
I think you've missed a lot of what distinguishes the Propeller from pretty much every other uC available with the possible exception of the XMOS uC favored by Leon. If you look at most uCs on the market, what you'll find is a general purpose processor core of some type with several K of EEPROM for program storage, some SRAM, possibly some data EEPROM with a higher write durability surrounded by a variety of fixed special purpose I/O controllers. There are usually many many models of these uCs with various memory sizes and all sorts of combinations of possible I/O controllers.
What's different about the Propeller is that it has a set of essentially identical high speed processors, each with their own memory, multipurpose counters, and video generator, all around a shared memory and some common control logic. Each of these processors is capable of acting as a video display generator, multi-port UART, floating point processor, multiport PWM controller, dual ADC, etc. For high bandwidth tasks, several processors can be synchronized to share the work. The processors can be dynamically reconfigured as needed.
Regarding the lack of on-chip flash and the "limited" SRAM, the chip is already quite large with the 8 processors and their local memory. Adding more SRAM is problematic with the feature sizes available when the chip was designed. Note that the Prop-II will have at least 8x the on-chip SRAM. Also note that the Prop-II will have much much higher quiescent current demands because of the same smaller features and thinner oxide needed. The Prop-II will not replace the Prop-I, partly because it will be more expensive and partly because it won't achieve the same low power consumption as the Prop-I.
You raise a good point about the fact that other uC manufacturers seem able to include on-chip flash memory. In addition to the chip real estate costs, Parallax has made it clear that the chip fab processes available do not allow for on-chip flash. There may be IP issues involved and licensing costs that may add too much to the cost of manufacture. From a product manufacturing standpoint, adding an external EEPROM is a small incremental cost above that of the Propeller chip itself that needs to be figured into any evaluation of the Prop for a product.
Unlike most of the other uC manufacturers, Parallax is in this business for the long term. Some of their earliest products are still available and their expectation of the lifetime of the Propeller is very long term. The Prop-I has just begun to appear in large numbers of Parallax products and this will only expand as time goes on.
Badpacket does have a couple points. The prop is the coolest chip ever, but i don't think it will ever catch on in the commercial market. Most people and companies see the prop and immediately brand it a hobbyist toy, and not as the powerhouse it actually is. I also agree about the ram. I would sacrifice speed in an instant if it meant more ram/ program space.....Coming from someone who has tried a bunch of mircos(Arduino, Picaxe, basic stamp, Basic Micro Nano), let me tell you: the propeller is a great chip, you will like what you find.
XMOS has the same problem with on-chip flash memory. They have said that they could incorporate it if there was sufficient demand, but they would have to go to a 125 nm process, slowing the chip down a lot. Most FPGAs don't have on-chip flash, either, for the same reasons.
XMOS has good solutions to the security problem - a secure boot module consisting of 8k of OTP memory for each core, and cryptographic functions in the instruction set.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Interesting comments. Other chips sporting large flash do not have much sram (except for just recently). This is a problem when you are trying to load other programs as you do not have enough sram to run the program. It is fine if you have a fixed program that fits into the flash. Here is where the Prop shines. Development is fast because we do not have to program flash each time, and there is no wear issues either.
The other are where the prop shines is it's multipurpose I/O blocks - I say blocks because they are configurable. Why do commercial companies use FPGA's? Because they are trying to cut inventory. It is much better to hold 1 part for multiple designs. Most other uCs require different chips depending on the UARTs etc in the chip, and they are pin dependant. The prop doesn't care which pins are which and moreover, we design the IO block with a cog.
However, I am disappointed that there has not been more commercial takeup of the prop. IMHO PropII release will broaden the market for both the PropII & PropI. I think it will legitimise the Prop further because of a second chip in the series. Of course, it will be so much more powerful.
I too would like to see more cog ram. At least wit PropII we can bring in overlays or run LMM much faster. The architecture of the instruction set makes this a problem. However, I think banked sram could have worked (say switching 256 long banks into the first or last 1/2 of cog ram). I do understand about the die size of sram, but AFAIK we will have an extra 512 longs for each cog in the form of fifo memory. How this can be accessed remains to be seen, but it just may have some additional uses to make us happy.
We are just starting to see lots of things happen with the PropI. They said the prop could never do many things, including be a USB Host. Micah and others has proved them wrong, and we now have USB Bluetooth with a $2 USB part. What next???
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
Newer 16-bit PICs have assignable pins for peripheral functions like UARTs and ADC inputs. It's a clever idea, and Microchip holds a patent on it. It isn't needed if you don't have peripherals on chip, of course.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
badpacket said...
I can't believe I am saying this, but perhaps this is one instance where there is too much power? For most Uc projects, the Prop is simply over-provisioned, and over-powered. This is almost more of a multi-core processor than a uC.
For the average uC project, where part count is a major consideration, there is little to justify the Prop if you only need 10-20 Mhz with an 8/16-bit'er.
If you need more, you can pick up single chip solutions which are a couple of dollars, and less powerful than the Prop, without the additional BOM, and limited memory and other Prop quirks.
This sort of leaves the Prop with a much more narrow application range into which it can market itself from a price/performance POV.
A computer company encouraged me to come investigate the prop for one of their products because it is low cost and it can do a number of things.· It is also very encouraging that the Prop II will be turbine propelled because if I connect it to a computer then I want a Ferrari rather than a 20 MHZ chip connected to it.
The low cost AVR chips for Atmel may be good enough for automotives but they are too slow for some POV displays.
I can't get a lot of responses from Basic Micro but one of them told me to come over here because there are better forums.· A retired contractor in the microcontroller business told me to get involved over here.
I called up Microchip and talked to someone who couldn't really give me advice or answer a lot of my questions.· I went to the Arduino forums and while they answered my questions, the left a few of my questions unanswered.· The users on the main Arduino site told me to buy before I try and wouldn't tell me the difference between two units and someone on the Society of Robots said the Arduino isn't made for beginners when it comes to Robotics and I assume they know more about Robots than art students.· They say·the·Arduino websites·have tutorials but then I can name at least two Arduino websites that don't sell crystals or resonators because they won't teach you to connect them and that is because there·are noticeable gaps in what they are teaching you.· And while I like people and I don't have problem with a lot of college art students answering my questions but·I also went to college and tutored people in Data Processing classes where I had to write programs and a lot of college students also counsel people the wrong way because they are inexperienced and I've seen a lot of wrong advice being given.· There is a difference between someone who makes it their hobby and someone who makes it their profession.· There is a learning curve and sometimes other forums get me frustrated because the people don't want to learn anything new like the Prop and the Arduino sites frustrate me because they don't answer all of my questions or there are noticeable·gaps in their tutorials.
I looked at several sources on the web about the Prop II and I think that Parallax could almost fire their salespeople because it will sell itself once people see the advantages of the Prop II.
I think that Parallax is a one stop shop.· I don't have to go to Digikey to get a crystal because Parallax sells them but I would have to go to Digikey to buy a crystal for the Arduino because two or more Arduino sites don't sell them.· And when you have to buy your parts from several places, you lose money on shipping.
I ordered last Thursday's web special on the Basic Stamp Activity kit because it includes "What is a Microcontroller?"· All of the activities are found in one book instead of having to download several tutorials from the web.· Having directions on an open book means less eye strain and it means that I can bookmark an open book and turn to the right page whenever I want.· I can find all of the tutorials in "What is a Microcontroller?" because Parallax is a one stop shop.
I know people who are making an emulator on FPGA and people who have nothing to do with it are offering their disrespect for their efforts.· I've seen a youtube video of a working game running on it so they have a product and it is up and running.· I think that if you wait a few weeks or months, Parallax may release more innovative information about the Prop II which you may find exciting if you are willing to wait for those details.
Back in the 80's the Apple, Commodore and IBM users had arguments as to which computer was best.· I'll mention an interest in building a computer today and they aren't interested in building anything but in times past they were seen as using a computer for bragging rights.
"Back in the 80's the Apple, Commodore and IBM users had arguments as to which computer was best. I'll mention an interest in building a computer today and they aren't interested in building anything but in times past they were seen as using a computer for bragging rights."
Yes! Some of them are. I think Andre' has the right idea on linking retro style gaming to this discipline.
Your comments on community, and one stop shopping have a lot of merit. IMHO, a big part of this is branding, and another part of it is inertia. On one hand, the friendly, accessible approach brings people like me here, wanting to learn, looking to career build over time, or perhaps just getting a few things done. On the other, that's seen as small potatoes, with all the negative that goes along with that.
The most interesting comments I've gotten, when I countered the "it's a toy" with some solid information, is "too good to be true", or "flash in the pan", etc... I've come to realize that for a fairly large number of people, both change, and the idea that some things can be easier, or done with fewer components, is just seen as bad.
Another element in larger scale commercial adoption is the relationships, business to business, that tend to lock in components for a mutual gain. When that is occurring, neither party has a lot of interest in seeking new devices. Highlighting how they may better compete doing so actually will pack a punch, but then they may well just revisit the relationship in order to sweeten the deal, having some additional leverage... If that occurs, it's a low cost deal, and that's quick, consistent, and low risk.
Along those lines, managing a vendor relationship comes with a cost, that plays into what I just wrote above. It's in the interests of the more established players to maintain their lock-in, and prevent new relationships from forming too. Why deal with a single chip supplier, when an entire catalog can be sourced for about the same costs? I wonder about that one often, when the discussion circles around this way.
Well, without writing a book, I remain convinced that barriers to adoption are not technical. For the tasks that lie within a reasonable scope of what the Prop is good at, the chip dominates, and there are a lot of those tasks! This will remain true for Prop II, meaning I believe some greater effort, aimed long the lines in this post will be required to see greater success.
I have not done commercial designs for about 10 years. However, I always factored costs into a project for the number of components. The more the components the more it really costs, for both assembly and fault-finding. There is also a benefit in using common components amongst designed which translates into inventory costs. The prop shines here.
There is also the software costs and here the prop is not seen in a good light, and yet in fact, it really shines. Why? Well, the 8 cores allow objects to be reused simply. There are no interrupts to confuse things. You can use spin where speed is unimportant (yes C is also available) and pasm where you require the speed. PASM is quite simple because of the reduced instruction set.
The major problem for the Prop I see is that it is not widely publicised. I found it by accident, just as I was about to embark on some PIC playing (yes it is a hobby for me now). Needless to say, my PIC stuff is unopened and gathering dust. I have others note they found the prop by accident too.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
The Prop is in a strange situation. Given that you want to build a micro-controller of some kind, Chip obviously had that itch, and given that you would like it to make some money, what are you going to do? Design another PIC, AVR, whatever style device? No there is no point, that market is awash with such devices. Going up scale, build another ARM, MIPS whatever style device? Again no, this market is also crowded. These markets are also fiercely competitive.
Solution: come up with something "totally out of the box" and perhaps tailored to a niche market that no one else is addressing. Parallax has done a brilliant job in that respect with the Propeller, the Prop Tool, the OBEX and their customer base in the education and hobbyist market.
Unfortunately this leads to the following phenomena. Those who are familiar with micro's, micro-controllers, embedded software and electronics etc in a professional, commercial environment can easily skip over the Prop when they see it. Dismissing it as "a toy", "a hobbyist gadget", "to weird", "fringe", "form a tiny irrelevant company". What can you do with only 512 LONGs of RAM in a 32 bit CPU? And why do I want an interpreter to slow me down? And oh, by the way, there is no native C compiler for it. The list goes on and on.
I'm guilty of this myself, there I was trying to get to grips with PICs and AVR's after years of working on bigger things like 68Ks, ARMS, PowerPC's and Linux. Seeing the Propeller in a catalog I just turned the page not even recognizing it as useful innovative device from the brief description there. And who is Parallax anyway?
Not until I got thoroughly frustrated with trying to get PICs and AVRs to do relatively simple things did I pick up a Prop on a whim and start to discover what a wonderful thing it is.
Now marketing is a strange thing. You might have the best product in it's field but selling it into a huge, entrenched, competitive market place might just be impossible. See how many CPU architectures have fallen by the way side in the evolution of the PC for example. Even Intel discovered it cannot compete even against it's own products for example they failed with the Itanium, previously with the 860 and in the depths of time with the 432.
I suspect it would be suicide for Parallax to expend to many resources is trying to do this.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Thanks, and I agree there is a lot I don't know about the Prop that cut too the core of its essence.
However, I am still kind of stuck on why a chip that obviously has far more power than anything else in its class, would be dismissed by an engineer.
I don't quite think its just that unfamiliarity or resistance to change. Is it possible that the Prop is considered more of a hobbyist chip?
Or, is it more that having soft peripherals instead of actual silicone for bog standard ones is considered too much flexibility or requires a reliance on 'software' that is just out of the comfort zone?
I think its probably some parts of each, along with a limited reputation in the Industrial space perhaps. Also, having what is essentially 8 cores running your peripherals along with your main code is quite the challenge, even for mainstream software developers who've had years to work on dual, quad, and octo cores in the PC arena.
Add to this the lack of interrupts, which are pretty much the base of embedded, it seems like a significant road to hoe.
The more I think about it, the more sense this makes to me at least.
I am not sure how a Prop II, with more of almost everything is going to effectively reduce a number of these barriers. Maybe it is just a matter of time, and there is no expedient way to hasten the pace.
I'm still of the mind that having a more basic set of hardware peripherals like PWM, timers, ADC, UART, JTAG, etc would allow more people to look at the Prop as an actual uC with 8 cores, versus a piece of silicon with 8 cores which can each be a peripheral, though some might require 2 or more cogs. And, not having something industry standard like C is a decent hit not to overlook.
I found this post on a robotics forum, and it seems to echo what I've heard: http://webcache.googleusercontent.com/search?q=cache:AcQQ8GgSoswJ:www.chibots.org/?q=node/463+parallax+propeller+adc&cd=4&hl=en&ct=clnk&gl=us&client=firefox-a
Again, for all its wonder, I'm not sure if uber-customization/configurability have not actually been more detrimental to serious interest in the Prop once a real engineer sits down and looks at what he will actually have left to work with once he gets his basic peripherals assigned to cogs.
I am not a real engineer, or even much of a hobbyist compared to most here. These are jusome thoughts.
Chuck Z, there are numerous Arduino/AVR sites that should be able to answer any question you could ever have. I don't know the exact rules for this forum, but if you Google AVR and Freaks, I doubt theres a question you can ask that won't get an answer. I hear your complaint about some forums, however you have to realize the Arduino is an Artsy thing for many, and your technical questions are probably over their head. If you want a more technical Q&A experience, you need to find a forum that has people who are more technical versus Artsy/Hobbyest types. Reference my suggestion above.
Also, I don't see any problem with your ordering issues. If you need to order an AVR/Pic crystal, well you have to do the same with the EEPROM for the Prop. You can order an AVR from a million shops, and most of them will have the exact cystal and caps you need.
However, in the bigger picture, outside of the hobby arena your complaints regarding ordering one-offs are not that big a deal. If Prop were to become more mainstream, then one would expect it to be carried by Digi-Key and Mouser if that worked with Parallax's sales/distribution process.
badpacket said...
If Prop were to become more mainstream, then one would expect it to be carried by Digi-Key and Mouser if that worked with Parallax's sales/distribution process.
Mouser has 247 DIP 40 Propellers in stock. Digi-Key carries them too at a better price in quantity than Parallax [noparse]:)[/noparse]
As far as hobbyist chip goes ... maybe so. Parallax is certainly considered a hobbyist centric company.
I had the "So what about the Propeller?" conversation with a professor last week. He likes the chip, but his conclusion was that it's ideas are not mainstream enough relative to other micros to be considered for his class. He's not much of an adventurer though.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
May the road rise to meet you; may the sun shine on your back.
May you create something useful, even if it's just a hack.
The main problem with the Propeller for many commercial applications is simply the price - it's far too expensive! Features like 160 MIPS, the on-chip VGA and the nice counter/timers aren't useful for most applications, where a $1 - $2 PIC or AVR will do the job. Higher performance systems tend to need more memory, where devices like the ARM are much more suitable, and are available for the same price as the 8-bit chips.
Hobbyists do appreciate having all that stuff on one chip, even if it costs more, as they only have to learn one device, and the price simply doesn't matter.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
badpacket: "However, I am still kind of stuck on why a chip that obviously has far more power than anything else in its class, would be dismissed by an engineer."
Well it's not just that an engineer might dismiss it as being out of his field of vision. You have to consider whole companies, design teams, purchasing departments, managers, chummy relation ships between vendors and consumer companies, industries, image, reputation, expectations, politics. The inertia of all of that. Good engineers are curious open minded types who often love to chase new ideas, solutions and gadgets. That's why they are engineers.
A couple cases in point:
Years ago I worked for a huge company building radars for ships, radars for airports, early warning systems, missiles etc etc. They had for years built their own embedded computer to manage these things. They had their own 16 bit CPU design in TTL chips and later bit-slice technology. At some point management decreed that there will be no more development of future versions of this computer and they they would use off the shelf kit from outside. Eventually it transpired the off the shelf kit was too expensive, unavailable and/or not up to the job. Luckily the engineers had continued work on their baby as a "skunk works" project for a few years. Only when the crunch came could they pull it out of the hat as the solution to the growing problems.
Last week I was in a meeting between my company and a large electronics/embedded design house discussing possibilities for the future development of our products. I started to draw on the white board a little solution to one issue and by way of example put a Propeller and/or XMOS device in the middle both of which have the required features and capabilities. I was immediately told that I should not be thinking about the internal guts of the proposed solution and the conversation moved on. Now, I know these guys a bit and can see that their solution would include an ARM and I/O expanders on SPI etc etc. They do a lot of that. I was a little bit cheered to see the chief hardware engineer in that meeting writing "Parallax" and "XMOS" in his note book though. He had never heard of them.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
badpacket said...
Chuck Z, there are numerous Arduino/AVR sites that should be able to answer any question you could ever have. I don't know the exact rules for this forum, but if you Google AVR and Freaks, I doubt theres a question you can ask that won't get an answer. I hear your complaint about some forums, however you have to realize the Arduino is an Artsy thing for many, and your technical questions are probably over their head. If you want a more technical Q&A experience, you need to find a forum that has people who are more technical versus Artsy/Hobbyest types. Reference my suggestion above.
Badpacket,
AVR Freaks isn't accepting new registrations and I don't go to Arduino sites to promote other chips.
I told one hobbyist about AVRs and the Prop and he basically said he has settled on two Microchip pics and that is all he wants to do.
I can get to talk to several engineers on this site.· I don't recognize many engineers on other forums at all.· When a representative answers the phone at Microchip, it could be customer service which isn't an Engineer and when I specifically called about a technical question, I felt like I was talking to customer service.
You can say all these other sites have this and that but I experience frustration on the other sites.· You can say that they can answer my questions but I had a teacher in High School who got through life quoting her textbooks without really understanding the content.· I would ask either / or questions and her answer was "yes" because she was good at telling me information without listening, understanding or explaining.· That is why an engineer here is more suited to answering questions.· They not only know the subject matter but they can take it apart, re-arrange it, and build it again; not quite so on the Arduino sites because you're told to buy a development board or a bootloader so when you want something more technical, you discover the gaps in their learning process because it wasn't covered.· Making it easy isn't so easy when you take out the steps to learning because you find out that people really didn't learn.· Many people mastered Basic on a Commodore 64 but fewer people mastered 65XX opcodes because everything isn't easy.· I can't just give me son the answers to math questions because I have to let him figure some of it out if he wants to learn.
So lets get back to talking about Prop II before someone·says something·about the hijacking of this thread.
jazzed, thanks for reminding me. Its been quite a while but you're right they do carry it. So, the complaints about having to buy an IC here or there, and then have to buy supporting discretes somewhere else and the couple of bucks extra shipping is trivial and non-existent.
heater, thanks, I did realize that however I'm new and went with a somewhat rhetorical query rather than come across as 'slamming' the Prop. I would love to see the Prop become more mainstream, because it is sort of a hobbyists wet-dream.
I'll admit to a little confusion reading some of the threads where it seemed like some people want the Prop to really become a player in the uC arena, while others were asking for everything including the kitchen sink, which seems to preclude it ever growing from the hobby/educational market. I'd actually like to hear something from Chip as to what the vision is for the Prop. That would probably shed more light than anything else.
Chuck Z, I take it you've got a problem with my reply?
Well, let see. As to promoting another forum, I thought I made it clear I wasn't aware of specific rules, and did at least not post the site name, URL, just a simple hint. That other forum already has someone who spams it with PIC's in mot every post, I'm not interested in being such a troll here.
As for hijacking the thread, I do believe -you- were the one who brought up the Arduino, and spent a paragraph or two bemoaning it for failures to meet your expectations. All I did was try to advise you that it is a product that is primarily for non-engineering types, and most forums are populated by such. You had questions and complaints that were technically above the level of expertise normally encountered, and in reality, you should have been asking those questions in a forum where there are numerous professional developers, and in many cases probably only answerable by devs who work on the actual AVR versus the abstracted level known as the Arduino.
If this is how you normally reply to those who try to give you some help for your questions, then I'm not surprised at the response or lack of you seem to have gotten.
To everyone else, seems like a great forum and a really unique uC from a friendly company. I'll be ordering a dev kit this week, and look forward to getting in to the West coast Expo and chatting with you.
badpacket,
Chip has really said quite a lot about the development process and his philosophy about the Prop I and Prop II. It's scattered through some humongous threads and several Webinars and you just have to go through it if you want to get the flavor of it. There's also a lot of the whole philosophy of Parallax and some of their experience that "informs" how they do things in threads about IP protection and other topics.
Mike
If this is how you normally reply to those who try to give you some help for your questions, then I'm not surprised at the response or lack of you seem to have gotten.
My response is a response to the lack of help and not the cause of why I'm not answered on the Arduino forum.· I suppose if they read my response here, they would say I should have spoken up over there but if I speak up too loudly over there then I'm "complaining".· If that is how beginners are treated then that is bad teaching because my professors at Temple stopped teaching with one method if someone didn't learn and they would start teaching with a different method.· The reason things aren't mentioned is because it isn't worth the effort to the teacher because it is supposed to be general information to the informed and work for them to expain the obvious.
I would love to open a web store and sell microcontrollers but some of the teaching hasn't caught up to the learning curve which is one of the reasons I haven't.· I'm sure that it is easy to teach but it is like my instructor in accounting said in college; you have to learn it now in college because when you're on the job, no one is going to tell you how to do it.· It carries over into personal life and the forums.· It is also why my relatives aren't interested in microcontrollers; they say it is difficult and they say,'you go learn it first' before you try to get us involved because they see it as difficult.· The reason why Arduino has a bootloader is because real microcontrollers are difficult for beginners.· Maybe the reason why they aren't telling is because they don't want to scare people half to death how hard and involved some things can be because it is about being easy instead of being involved.· The other reason may be due to the fact they have more users they can handle than to give everyone·competent time.
I'm glad you like microcontrollers and this is just my explanation and I don't have a problem with you but every side has a different point of view and my point is just because it is easy doesn't mean that people understand it.· The reason why it is easy is because people don't understand it.
Are the so-called Ajax techniques rich enough and fast enough that a dev system could be made browser-centric? Or is that just wishful thinking?
Very much so. I'm using WaveMaker, a browser-based IDE for graphical development of enterprise Java applications. It makes doing database front-ends a breeze. It's quite easy to use, and one quickly forgets that it runs in the browser.
It's open source, so you can download it and try it out just to get a feel. There are plenty of on-line video tutorials; making a user interface to say a simple web service can be done in a couple of minutes, start-to-finish. By finish I mean you get an archive file that can be deployed to a java application server.
WaveMaker's IDE eats its own dogfood -- the same WaveMaker libraries that end up powering your web application, so in principle the code could be reused for a Parallax IDE.
Another, much simpler web-based enterprise development platform, also open source but Python based, is web2py.
Comments
I like the idea of the Prop and the Prop II, however I see some real issues with it ever becoming anything more than an enthusiasts niche. Unless that is in line with corporate expectations. This is not meant to be trollish, and I am not going to suggest someone use a PIC or AVR.
As someone who uses Pic and AVR's, I've been interested in the Prop for quite some time. I know the story that the fab process Parallax is using is inimicable in some way to flash. Why exactly is it not possible to move to a process that will allow flash? No one else in the business has similar technical constraints, so this has to come down to a cost issue.
It would appear to me that going with no on-board flash has saved some $$ at the expense of serious uptake by industry or the even the enthusiast crowd, (Arduino, etc). Its hard enough as it is for a newcomer to break in to the uC market, I fail to see how/why anyone in the industry would ever pick the Prop versus a PIC or AVR as the part count for Prop means automatically adding the EEPROM to the BOM. Additional part count, board space, and design work, just to get your controller to boot? Granted, it is not rocket science, but I think it is enough to make a fair number of people think 'If this is what I have to do just to get it to boot, what other quirks are there going to be, besides this 'no interrupts' thing?".
The Prop is a very cool idea, and kudos for someone actually bringing something like this to fruition. Seriously, its like a dream come true with the masive amount of power available. However, after 4 years, I'm not sure I've seen much really happen to expand the Prop's uptake outside of the enthusiast niche. Thats not a bad thing, however reality is what it is, and having 8 'cores' doesn't seem to have piqued much real interest, outside of the 'wow, cool!'.
And, I know what everyone is going to say. But it has basically out of the box keyboard, mouse and VGA.
Well, since 2006(?), this has probably been the most powerful Uc you can buy, with more tricks up its sleeve than Houdini. And yet it still seems to be about as popular as the proverbial red-headed step child.
Should this matter?
Well, only if one realizes that uptake and gross sales are what keeps Parallax afloat. Looks like Prop II has been in development since '08, and no end is in sight. I think this would be markedly different if there were real volume sales, which would translate into profits and R&D development and newer models coming online like the Prop II in much shorter timeframes. Like what is the industry norm.
I can't believe I am saying this, but perhaps this is one instance where there is too much power? For most Uc projects, the Prop is simply over-provisioned, and over-powered. This is almost more of a multi-core processor than a uC.
For the average uC project, where part count is a major consideration, there is little to justify the Prop if you only need 10-20 Mhz with an 8/16-bit'er.
If you need more, you can pick up single chip solutions which are a couple of dollars, and less powerful than the Prop, without the additional BOM, and limited memory and other Prop quirks.
This sort of leaves the Prop with a much more narrow application range into which it can market itself from a price/performance POV.
Could you build a mini-Predator drone from a Prop, from what I've seen thats a solid yes. From a PIC or AVR, well not with just one, maybe a handful.
But, how many common uC projects require up to 8 threads on a multi-core, instead of 1 or maybe 2 threads? Much smaller subset I believe, hence harder to justify.
A lot of interesting ideas in the previous pages, but ultimately not a lot that I saw that would improve desireability among those not already Propheads.
If Parallax is making decent income from the current Prop, it'd be cool to see a number of those suggestions acted upon. Really cool to see them so open also.
However if it takes 4-5 years to get too that point because there is such limited adoption, then I would rather see something more akin to a review of the successes/failures of the current Prop, with an eye specifically cast on why it did not take the uC world by storm.
Because it could have, and being years ahead of the competition is a fair rationale, but not the whole story.
At this point in time, I don't think we are still at a point where most people could use, or really need 8 cores, so not much sense in going higher. Seems thats been identified already.
I would really stop and take hard look at whatever equations are going into the decision to not have flash, and not being a more industry standard one chip solution. Pardon if this has in fact been changed.
I would also consider adding some real RAM to the device. Its great that you can get 80 Mhz from each Cog, but if you are going to starve it of RAM, then whats the point? Heck, I'd consider knocking a couple of cores off the die for the RAM space if necessary to get there. Who knows, if you did that, or changed fab process you could probably hit even higher speeds.
Just my $0.02 worth, so please consider my comments on their merits ( if any), and not as a troll for PIC, AVR, XMOS, Freescale, etc, etc.
I'll probably order my Prop dev kit in the next month or so, and it still seems like its years ahead of everyone else in many respects.
What's different about the Propeller is that it has a set of essentially identical high speed processors, each with their own memory, multipurpose counters, and video generator, all around a shared memory and some common control logic. Each of these processors is capable of acting as a video display generator, multi-port UART, floating point processor, multiport PWM controller, dual ADC, etc. For high bandwidth tasks, several processors can be synchronized to share the work. The processors can be dynamically reconfigured as needed.
Regarding the lack of on-chip flash and the "limited" SRAM, the chip is already quite large with the 8 processors and their local memory. Adding more SRAM is problematic with the feature sizes available when the chip was designed. Note that the Prop-II will have at least 8x the on-chip SRAM. Also note that the Prop-II will have much much higher quiescent current demands because of the same smaller features and thinner oxide needed. The Prop-II will not replace the Prop-I, partly because it will be more expensive and partly because it won't achieve the same low power consumption as the Prop-I.
You raise a good point about the fact that other uC manufacturers seem able to include on-chip flash memory. In addition to the chip real estate costs, Parallax has made it clear that the chip fab processes available do not allow for on-chip flash. There may be IP issues involved and licensing costs that may add too much to the cost of manufacture. From a product manufacturing standpoint, adding an external EEPROM is a small incremental cost above that of the Propeller chip itself that needs to be figured into any evaluation of the Prop for a product.
Unlike most of the other uC manufacturers, Parallax is in this business for the long term. Some of their earliest products are still available and their expectation of the lifetime of the Propeller is very long term. The Prop-I has just begun to appear in large numbers of Parallax products and this will only expand as time goes on.
XMOS has good solutions to the security problem - a secure boot module consisting of 8k of OTP memory for each core, and cryptographic functions in the instruction set.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 5/17/2010 2:08:05 AM GMT
The other are where the prop shines is it's multipurpose I/O blocks - I say blocks because they are configurable. Why do commercial companies use FPGA's? Because they are trying to cut inventory. It is much better to hold 1 part for multiple designs. Most other uCs require different chips depending on the UARTs etc in the chip, and they are pin dependant. The prop doesn't care which pins are which and moreover, we design the IO block with a cog.
However, I am disappointed that there has not been more commercial takeup of the prop. IMHO PropII release will broaden the market for both the PropII & PropI. I think it will legitimise the Prop further because of a second chip in the series. Of course, it will be so much more powerful.
I too would like to see more cog ram. At least wit PropII we can bring in overlays or run LMM much faster. The architecture of the instruction set makes this a problem. However, I think banked sram could have worked (say switching 256 long banks into the first or last 1/2 of cog ram). I do understand about the die size of sram, but AFAIK we will have an extra 512 longs for each cog in the form of fifo memory. How this can be accessed remains to be seen, but it just may have some additional uses to make us happy.
We are just starting to see lots of things happen with the PropI. They said the prop could never do many things, including be a USB Host. Micah and others has proved them wrong, and we now have USB Bluetooth with a $2 USB part. What next???
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
The low cost AVR chips for Atmel may be good enough for automotives but they are too slow for some POV displays.
I can't get a lot of responses from Basic Micro but one of them told me to come over here because there are better forums.· A retired contractor in the microcontroller business told me to get involved over here.
I called up Microchip and talked to someone who couldn't really give me advice or answer a lot of my questions.· I went to the Arduino forums and while they answered my questions, the left a few of my questions unanswered.· The users on the main Arduino site told me to buy before I try and wouldn't tell me the difference between two units and someone on the Society of Robots said the Arduino isn't made for beginners when it comes to Robotics and I assume they know more about Robots than art students.· They say·the·Arduino websites·have tutorials but then I can name at least two Arduino websites that don't sell crystals or resonators because they won't teach you to connect them and that is because there·are noticeable gaps in what they are teaching you.· And while I like people and I don't have problem with a lot of college art students answering my questions but·I also went to college and tutored people in Data Processing classes where I had to write programs and a lot of college students also counsel people the wrong way because they are inexperienced and I've seen a lot of wrong advice being given.· There is a difference between someone who makes it their hobby and someone who makes it their profession.· There is a learning curve and sometimes other forums get me frustrated because the people don't want to learn anything new like the Prop and the Arduino sites frustrate me because they don't answer all of my questions or there are noticeable·gaps in their tutorials.
I looked at several sources on the web about the Prop II and I think that Parallax could almost fire their salespeople because it will sell itself once people see the advantages of the Prop II.
I think that Parallax is a one stop shop.· I don't have to go to Digikey to get a crystal because Parallax sells them but I would have to go to Digikey to buy a crystal for the Arduino because two or more Arduino sites don't sell them.· And when you have to buy your parts from several places, you lose money on shipping.
I ordered last Thursday's web special on the Basic Stamp Activity kit because it includes "What is a Microcontroller?"· All of the activities are found in one book instead of having to download several tutorials from the web.· Having directions on an open book means less eye strain and it means that I can bookmark an open book and turn to the right page whenever I want.· I can find all of the tutorials in "What is a Microcontroller?" because Parallax is a one stop shop.
I know people who are making an emulator on FPGA and people who have nothing to do with it are offering their disrespect for their efforts.· I've seen a youtube video of a working game running on it so they have a product and it is up and running.· I think that if you wait a few weeks or months, Parallax may release more innovative information about the Prop II which you may find exciting if you are willing to wait for those details.
Back in the 80's the Apple, Commodore and IBM users had arguments as to which computer was best.· I'll mention an interest in building a computer today and they aren't interested in building anything but in times past they were seen as using a computer for bragging rights.
Yes! Some of them are. I think Andre' has the right idea on linking retro style gaming to this discipline.
Your comments on community, and one stop shopping have a lot of merit. IMHO, a big part of this is branding, and another part of it is inertia. On one hand, the friendly, accessible approach brings people like me here, wanting to learn, looking to career build over time, or perhaps just getting a few things done. On the other, that's seen as small potatoes, with all the negative that goes along with that.
The most interesting comments I've gotten, when I countered the "it's a toy" with some solid information, is "too good to be true", or "flash in the pan", etc... I've come to realize that for a fairly large number of people, both change, and the idea that some things can be easier, or done with fewer components, is just seen as bad.
Another element in larger scale commercial adoption is the relationships, business to business, that tend to lock in components for a mutual gain. When that is occurring, neither party has a lot of interest in seeking new devices. Highlighting how they may better compete doing so actually will pack a punch, but then they may well just revisit the relationship in order to sweeten the deal, having some additional leverage... If that occurs, it's a low cost deal, and that's quick, consistent, and low risk.
Along those lines, managing a vendor relationship comes with a cost, that plays into what I just wrote above. It's in the interests of the more established players to maintain their lock-in, and prevent new relationships from forming too. Why deal with a single chip supplier, when an entire catalog can be sourced for about the same costs? I wonder about that one often, when the discussion circles around this way.
Well, without writing a book, I remain convinced that barriers to adoption are not technical. For the tasks that lie within a reasonable scope of what the Prop is good at, the chip dominates, and there are a lot of those tasks! This will remain true for Prop II, meaning I believe some greater effort, aimed long the lines in this post will be required to see greater success.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Safety Tip: Life is as good as YOU think it is!
Post Edited (potatohead) : 5/17/2010 1:44:16 PM GMT
There is also the software costs and here the prop is not seen in a good light, and yet in fact, it really shines. Why? Well, the 8 cores allow objects to be reused simply. There are no interrupts to confuse things. You can use spin where speed is unimportant (yes C is also available) and pasm where you require the speed. PASM is quite simple because of the reduced instruction set.
The major problem for the Prop I see is that it is not widely publicised. I found it by accident, just as I was about to embark on some PIC playing (yes it is a hobby for me now). Needless to say, my PIC stuff is unopened and gathering dust. I have others note they found the prop by accident too.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
Solution: come up with something "totally out of the box" and perhaps tailored to a niche market that no one else is addressing. Parallax has done a brilliant job in that respect with the Propeller, the Prop Tool, the OBEX and their customer base in the education and hobbyist market.
Unfortunately this leads to the following phenomena. Those who are familiar with micro's, micro-controllers, embedded software and electronics etc in a professional, commercial environment can easily skip over the Prop when they see it. Dismissing it as "a toy", "a hobbyist gadget", "to weird", "fringe", "form a tiny irrelevant company". What can you do with only 512 LONGs of RAM in a 32 bit CPU? And why do I want an interpreter to slow me down? And oh, by the way, there is no native C compiler for it. The list goes on and on.
I'm guilty of this myself, there I was trying to get to grips with PICs and AVR's after years of working on bigger things like 68Ks, ARMS, PowerPC's and Linux. Seeing the Propeller in a catalog I just turned the page not even recognizing it as useful innovative device from the brief description there. And who is Parallax anyway?
Not until I got thoroughly frustrated with trying to get PICs and AVRs to do relatively simple things did I pick up a Prop on a whim and start to discover what a wonderful thing it is.
Now marketing is a strange thing. You might have the best product in it's field but selling it into a huge, entrenched, competitive market place might just be impossible. See how many CPU architectures have fallen by the way side in the evolution of the PC for example. Even Intel discovered it cannot compete even against it's own products for example they failed with the Itanium, previously with the 860 and in the depths of time with the 432.
I suspect it would be suicide for Parallax to expend to many resources is trying to do this.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Thanks, and I agree there is a lot I don't know about the Prop that cut too the core of its essence.
However, I am still kind of stuck on why a chip that obviously has far more power than anything else in its class, would be dismissed by an engineer.
I don't quite think its just that unfamiliarity or resistance to change. Is it possible that the Prop is considered more of a hobbyist chip?
Or, is it more that having soft peripherals instead of actual silicone for bog standard ones is considered too much flexibility or requires a reliance on 'software' that is just out of the comfort zone?
I think its probably some parts of each, along with a limited reputation in the Industrial space perhaps. Also, having what is essentially 8 cores running your peripherals along with your main code is quite the challenge, even for mainstream software developers who've had years to work on dual, quad, and octo cores in the PC arena.
Add to this the lack of interrupts, which are pretty much the base of embedded, it seems like a significant road to hoe.
The more I think about it, the more sense this makes to me at least.
I am not sure how a Prop II, with more of almost everything is going to effectively reduce a number of these barriers. Maybe it is just a matter of time, and there is no expedient way to hasten the pace.
I'm still of the mind that having a more basic set of hardware peripherals like PWM, timers, ADC, UART, JTAG, etc would allow more people to look at the Prop as an actual uC with 8 cores, versus a piece of silicon with 8 cores which can each be a peripheral, though some might require 2 or more cogs. And, not having something industry standard like C is a decent hit not to overlook.
I found this post on a robotics forum, and it seems to echo what I've heard: http://webcache.googleusercontent.com/search?q=cache:AcQQ8GgSoswJ:www.chibots.org/?q=node/463+parallax+propeller+adc&cd=4&hl=en&ct=clnk&gl=us&client=firefox-a
Again, for all its wonder, I'm not sure if uber-customization/configurability have not actually been more detrimental to serious interest in the Prop once a real engineer sits down and looks at what he will actually have left to work with once he gets his basic peripherals assigned to cogs.
I am not a real engineer, or even much of a hobbyist compared to most here. These are jusome thoughts.
Chuck Z, there are numerous Arduino/AVR sites that should be able to answer any question you could ever have. I don't know the exact rules for this forum, but if you Google AVR and Freaks, I doubt theres a question you can ask that won't get an answer. I hear your complaint about some forums, however you have to realize the Arduino is an Artsy thing for many, and your technical questions are probably over their head. If you want a more technical Q&A experience, you need to find a forum that has people who are more technical versus Artsy/Hobbyest types. Reference my suggestion above.
Also, I don't see any problem with your ordering issues. If you need to order an AVR/Pic crystal, well you have to do the same with the EEPROM for the Prop. You can order an AVR from a million shops, and most of them will have the exact cystal and caps you need.
However, in the bigger picture, outside of the hobby arena your complaints regarding ordering one-offs are not that big a deal. If Prop were to become more mainstream, then one would expect it to be carried by Digi-Key and Mouser if that worked with Parallax's sales/distribution process.
As far as hobbyist chip goes ... maybe so. Parallax is certainly considered a hobbyist centric company.
I had the "So what about the Propeller?" conversation with a professor last week. He likes the chip, but his conclusion was that it's ideas are not mainstream enough relative to other micros to be considered for his class. He's not much of an adventurer though.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
May the road rise to meet you; may the sun shine on your back.
May you create something useful, even if it's just a hack.
Hobbyists do appreciate having all that stuff on one chip, even if it costs more, as they only have to learn one device, and the price simply doesn't matter.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 5/17/2010 9:01:39 AM GMT
Well it's not just that an engineer might dismiss it as being out of his field of vision. You have to consider whole companies, design teams, purchasing departments, managers, chummy relation ships between vendors and consumer companies, industries, image, reputation, expectations, politics. The inertia of all of that. Good engineers are curious open minded types who often love to chase new ideas, solutions and gadgets. That's why they are engineers.
A couple cases in point:
Years ago I worked for a huge company building radars for ships, radars for airports, early warning systems, missiles etc etc. They had for years built their own embedded computer to manage these things. They had their own 16 bit CPU design in TTL chips and later bit-slice technology. At some point management decreed that there will be no more development of future versions of this computer and they they would use off the shelf kit from outside. Eventually it transpired the off the shelf kit was too expensive, unavailable and/or not up to the job. Luckily the engineers had continued work on their baby as a "skunk works" project for a few years. Only when the crunch came could they pull it out of the hat as the solution to the growing problems.
Last week I was in a meeting between my company and a large electronics/embedded design house discussing possibilities for the future development of our products. I started to draw on the white board a little solution to one issue and by way of example put a Propeller and/or XMOS device in the middle both of which have the required features and capabilities. I was immediately told that I should not be thinking about the internal guts of the proposed solution and the conversation moved on. Now, I know these guys a bit and can see that their solution would include an ARM and I/O expanders on SPI etc etc. They do a lot of that. I was a little bit cheered to see the chief hardware engineer in that meeting writing "Parallax" and "XMOS" in his note book though. He had never heard of them.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
AVR Freaks isn't accepting new registrations and I don't go to Arduino sites to promote other chips.
I told one hobbyist about AVRs and the Prop and he basically said he has settled on two Microchip pics and that is all he wants to do.
I can get to talk to several engineers on this site.· I don't recognize many engineers on other forums at all.· When a representative answers the phone at Microchip, it could be customer service which isn't an Engineer and when I specifically called about a technical question, I felt like I was talking to customer service.
You can say all these other sites have this and that but I experience frustration on the other sites.· You can say that they can answer my questions but I had a teacher in High School who got through life quoting her textbooks without really understanding the content.· I would ask either / or questions and her answer was "yes" because she was good at telling me information without listening, understanding or explaining.· That is why an engineer here is more suited to answering questions.· They not only know the subject matter but they can take it apart, re-arrange it, and build it again; not quite so on the Arduino sites because you're told to buy a development board or a bootloader so when you want something more technical, you discover the gaps in their learning process because it wasn't covered.· Making it easy isn't so easy when you take out the steps to learning because you find out that people really didn't learn.· Many people mastered Basic on a Commodore 64 but fewer people mastered 65XX opcodes because everything isn't easy.· I can't just give me son the answers to math questions because I have to let him figure some of it out if he wants to learn.
So lets get back to talking about Prop II before someone·says something·about the hijacking of this thread.
Chuck
heater, thanks, I did realize that however I'm new and went with a somewhat rhetorical query rather than come across as 'slamming' the Prop. I would love to see the Prop become more mainstream, because it is sort of a hobbyists wet-dream.
I'll admit to a little confusion reading some of the threads where it seemed like some people want the Prop to really become a player in the uC arena, while others were asking for everything including the kitchen sink, which seems to preclude it ever growing from the hobby/educational market. I'd actually like to hear something from Chip as to what the vision is for the Prop. That would probably shed more light than anything else.
Chuck Z, I take it you've got a problem with my reply?
Well, let see. As to promoting another forum, I thought I made it clear I wasn't aware of specific rules, and did at least not post the site name, URL, just a simple hint. That other forum already has someone who spams it with PIC's in mot every post, I'm not interested in being such a troll here.
As for hijacking the thread, I do believe -you- were the one who brought up the Arduino, and spent a paragraph or two bemoaning it for failures to meet your expectations. All I did was try to advise you that it is a product that is primarily for non-engineering types, and most forums are populated by such. You had questions and complaints that were technically above the level of expertise normally encountered, and in reality, you should have been asking those questions in a forum where there are numerous professional developers, and in many cases probably only answerable by devs who work on the actual AVR versus the abstracted level known as the Arduino.
If this is how you normally reply to those who try to give you some help for your questions, then I'm not surprised at the response or lack of you seem to have gotten.
To everyone else, seems like a great forum and a really unique uC from a friendly company. I'll be ordering a dev kit this week, and look forward to getting in to the West coast Expo and chatting with you.
Chip has really said quite a lot about the development process and his philosophy about the Prop I and Prop II. It's scattered through some humongous threads and several Webinars and you just have to go through it if you want to get the flavor of it. There's also a lot of the whole philosophy of Parallax and some of their experience that "informs" how they do things in threads about IP protection and other topics.
Mike
I would love to open a web store and sell microcontrollers but some of the teaching hasn't caught up to the learning curve which is one of the reasons I haven't.· I'm sure that it is easy to teach but it is like my instructor in accounting said in college; you have to learn it now in college because when you're on the job, no one is going to tell you how to do it.· It carries over into personal life and the forums.· It is also why my relatives aren't interested in microcontrollers; they say it is difficult and they say,'you go learn it first' before you try to get us involved because they see it as difficult.· The reason why Arduino has a bootloader is because real microcontrollers are difficult for beginners.· Maybe the reason why they aren't telling is because they don't want to scare people half to death how hard and involved some things can be because it is about being easy instead of being involved.· The other reason may be due to the fact they have more users they can handle than to give everyone·competent time.
I'm glad you like microcontrollers and this is just my explanation and I don't have a problem with you but every side has a different point of view and my point is just because it is easy doesn't mean that people understand it.· The reason why it is easy is because people don't understand it.
·
It's open source, so you can download it and try it out just to get a feel. There are plenty of on-line video tutorials; making a user interface to say a simple web service can be done in a couple of minutes, start-to-finish. By finish I mean you get an archive file that can be deployed to a java application server.
WaveMaker's IDE eats its own dogfood -- the same WaveMaker libraries that end up powering your web application, so in principle the code could be reused for a Parallax IDE.
Another, much simpler web-based enterprise development platform, also open source but Python based, is web2py.