Ok - I'll probably regret this, but here goes. I have let most of your recent posts pass since like most people I figure you have a right to draw comparisons between the Prop and other chips. I don't really think the Prop suffers from such comparisons, and sometimes you even make valid and interesting points.
But you just never know when to stop. Your recent posts have nothing to do with the thread topic. It's an advertisment for XMOS - pure and simple. You would have been much better advised to just leave things alone, since you had some people in these forums on your side - either willing to support you, or at least willing to ignore your agitating for XMOS at every opportunity.
I don't think you will have quite so many on your side after these recent posts.
Ross.
EDITED: since Leon continues to post, I edited this to refer to his "recent" posts, not his "last" post.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
Leon, my point wasn't that the Atom was gunning for XMOS' market -- actually, if you read what I wrote you'd see I am arguing just the opposite. It's that XMOS is targeting low-end PC level processing power, which means lots of power consumption, heat, and electrical noise. Not to mention the most unfriendly for hobbyists chip form factor ever invented, BGA. So you'd better get used to using something like Prop Protoboards because if you don't have some serious soldering equipment you aren't gonna be working with the actual chips.
Xmos is targeting a very different market than Parallax. Their customers are going to be mostly corporate, but looking for solutions that fit in the embedded space for relatively high-end functionality. There is nothing hobbyist or education about any product in BGA form factor.
Parallax is trying to bring as much of their functionality as possible to people who started out on Basic Stamps. These are individuals who are going to haul out a soldering iron which they will hold in their hand. Those people have no more chance of using an Xmos processor than I do of winning the Tour de France. Xmos advertises to those people (I see their ads all the time on hackaday and slashdot) but unless those people work for a company that has serious fab capability they're never gonna work with those chips.
Meanwhile you can buy a DIP40 propeller. And even the P2 will be in a package that can at least be theoretically hand soldered. And if you're working mostly with text and simple data rather than video, that will give you a whole world you never imagined could be powered by two AA batteries.
I'm guessing that part of the problem is that this thread is about a 2 cog prop, taking away most of what makes this chip so unique and leaving it open to comparison by other processors. The only thing I haven't really noticed yet, however, is a comparison of ease to use with other chips vs. the prop. From what I can see, this is the main reason for Peter to bring this up, because he doesn't want to have to deal with all the extra difficulties to make the "better" pic/xmos processors work. The prop is simple, and that's why most of the people on these forums are here. Even if the 2 cog prop would be more expensive per feature than other pic processors, it wouldn't matter to many here because just making it work wouldn't be worth lower cost.
As I said, it's only the larger chips that are in BGA. The XS1-L1-64 and -128 are in QFP, which can be hand-soldered easily. I could even solder the two-core XS1-L2 chip at home (it's a sort of two-row QFN) as I have a hot-air system. I can get BGAs assembled quite cheaply, anyway. Several low-cost ready-made boards are available from XMOS and Sparkfun, which are very popular with hobbyists and students. The $99 XC-1A has a four-core BGA device and can be used simply by plugging it into a USB port as it has the JTAG interface on-board. I've interfaced the earlier XC-1 board to a Propeller chip, it was quite trivial.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Do notice I brought up the Pic more than the XMOS as it was a fairer comparison to the 2 cog Prop. It seems people on this thread have strayed quite far from the topic...
I think there's a basic thing that Leon ( and maybe a few others ) might be missing. Chip in particular and the other fine people at Parallax have brought something back to computing that has been missing for most of us for a very long time: Excitement, Joy, Wonder. In short, Fun.
The very idea that some "INDIVIDUAL", some crazy techno-nerd can "Lock himself in the basement" for a few years and emerge with a miracle to rival the mighty teams of Intel and the others is pretty fantastic. The last time I know of someone doing something similar was Linus Torvalds with the creation of Linux. Linus LOVED the technology to the point that he wasn't concerned with the money. He was just as poor as the rest of us even AFTER he became famous for his operating system. It was the USERS of his system that got together and donated the money to pay off the last couple of years on the loan he took out to buy his computer.
I think there is something similar going on here with many of us. I KNOW that there are other chips out there that can do most of the things that I use the propeller for. I KNOW that there are processors with features that I WISH the propeller had, but I LOVE the "David and Goliath" nature of what Parallax represents and I will go so far as to "Hammer" a Propeller into a design that might have been somewhat better suited for another processor.
This thread is an affirmation that many of us would easily design with another Parallax processor if it met the one thing that we often can't design around: Price. Another thing that SOME of us want to make the best of is our knowledge base. A 2-4 core Propeller would let us use most of the knowledge that we have gained over the last few years without having to resort to new tools, new languages, new support, etc.
I would MUCH prefer to see the Prop II sitting on my desk in the near future over any cut-down version of the chip we already know and love, but if it WAS as simple as handing the project to someone else and saying: Cut it in half! I think many of us would "Vote with our dollars" for a Parallax solution whenever we could. This is as much about "Community" and Ethics, and triumph of the "Human Spirit" as it is anything technical.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
" Anything worth doing... is worth overdoing. "
KenBash: Well said! The discussion on this thread WAS about if it were possible to have someone else within Parallax, other than Chip, to take the "blocks" of the Prop 1 and re-arrange them, what would be required and useful. Would a 2 cog prop work. I don't think 2 cogs would but maybe 3 or 4 cogs would. I put in the ring some various cut and pasted blocks as ideas and what could be in them (or removed from the Prop 1) and put it up for discussion.
You will note I also posted a bigger version that may be a Prop 1.5+ (64 I/O + 64KB RAM & 64KB ROM) - I think this may be a better investment for Parallax than the current Prop 1.5 or smaller Prop, providing it could be done by someone other than Chip.
OT Leon has successfully hijacked another thread. His recent post was a pure and simple blatant post for Xmos. One must ask, how much does Xmos pay him, or how much free hardware does he get, or does he have some other vested interest? BTW don't program one of those things in ASM - the instruction set is terrible! And certainly it could not compare to a cut down Prop.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
A 2 or even 4 cog Prop doesn't make sense period. The strength of the Prop is its abundance of Cogs not its lack thereof. Cut it back to 2 and there's no reason to use a Prop. Other 32bitters can do just as good or better at that point, doubly so if you don't need video.
Then there's price. Unless you can undercut and outperform the various ARM offerings or PICS you aren't going to get noticed by business ergo no design wins. And I don't see anyone trying to show the world how well the current Prop performs against other processors except bad mouth a certain poster who dared bring up a competitor and who caused some regulars here to have a case of the vapors.
Proof is in the pudding gentlemen, not hyperbole and wishful thinking.
No, Chip is better off focusing on the Prop II and not wasting time and money on some crippled variant to please hobbyists when Parallax still doesn't have a serious design win for the Prop as it is. And consider this: In two or so years when the Prop II is released, the other vendors are going to have their own new generations of processors out as well, so his new design is going to have to really stand out.
Lastly, if you want to see what a two Cog Prop in a 28 pin package can do. Simulate it with a real Prop. Then compare it with a PIC32, AVR32, ARM, etc.
Having met Leon at Maker Faire (in his XMOS T-Shirt) I can say he has a definate leaning toward XMOS as his favourite.
I do believe when he mentions other processors, be it XMOS or PIC etc he is merely stating that there are othe processors out there more suitable for the task and there is nothing wrong with that.
Remember that most of us using the Props are entusiastic about it and take it personally when others say it isn't as good as we think it is.
I have taken offence to Leon promoting other chips in the past but now I just don't take any notice in fact I expect it of him.
I'm just about to deliver my first commercial project with the Propeller.
I could have use a PIC because in all honesty it doesn't need lots of speed, however, it wasn't price sensitive and I prototyped the whole thing with a Prop so it made sense for me to stick with it for the production model.
Why did I do this? Well, simply because I like the architecture, it's easy and friendly and that counts a lot for me.
Everyone has their favourites, Leon has XMOS and I have Propeller, different folks, different strokes!
I have an XMOS board, I got one when the first dev board came out as did some others who frequent this forum, I tried it and decided it wasn't for me. I haven't looked at it since.
The one area that the Propeller has a massive advantage over XMOS is this forum, this is where XMOS can't get anywhere close I bet Leon wishes there was the same amount of activity on the XMOS boards as there is over here.......
Back to the topic in hand.
A two or four cog Prop, yeah I like the idea but it's not practical and commercially it would be a disaster.
It couldn't compete with the plethora of other chips out there that could beat it in so many ways.
Parallax would be better making a Prop with some security and more i/o, this is what is holding it back in the commercial market.
Roll on Prop 1.5 or whatever it is called.....
Regards,
Coley
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ PropGFX - The home of the Hybrid Development System and PropGFX Lite
yeah ! Assembler rulez ! [noparse]:)[/noparse] The assembler in the xmos is great and so compact, really well thought...
But the 4 COG prop ? would be a great idea, specially if the do some multi-chip package with a big EEPROM inside, that would be something ! I'd leave 32 KB of HUB RAM though.
I often find myself wanting a simple application, such as a sine wave source, and I find myself reluctant to use a Prop. Not because it is more expensive, it just seem "just too damn good" to be given such a menial task. I want to put a little composite video, VGA and tone generator into a small box, for monitor proving, the Prop is just the beasty for that.
I never went the PIC route but I do have a drawer full of AVRs, which do have the same advantage of free development tools. They usually get the small, menial things to do.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Why did I think a new, more challenging, job was a good idea ??
Ale: I am quite happy to debate ASM on another thread. It's OT here.
I would rather see any possible resources placed into the Prop 1.5, and particularly a larger die size (increased cost) with more ROM & RAM as well as the 64 I/O, rather than a smaller Prop. The price seems to be the main driving force behind this. If we could significanlty increase the volume of Prop sales, then I would expect this would be the best way to get a price reduction.
As for a 2 die solution (prop + eeprom) I would expect this would be a much more expensive solution that a seperate Eeprom. It is all to do with volumes. If I understand correctly, because of the low volumes, the manufacturer used does not have access to all the features such as eeprom, flash etc.
Wouldn't a Prop 1.6 (64 IO + 64KB ROM + 64KB RAM), cost ~$10, with the same geometry (low power) as the Prop 1 ROCK !!! Delivery could be <6 mths ???
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
I've decided to weigh in here as a way introducing myself to the members here. I chose this topic to respond to only because I cannot resist a good controversy, and definitely not because I am an expert or anything. Far from it.
There do seem to be many special circumstance where a two-cog prop could be a big seller. I am sure that a lot of commercial products which can be easily developed on the propeller could also be squished into two cogs. A two-cog unit driving a touchscreen could be a useful product by itself. I'm so new to this device that I don't even know if that's even possible, but I this is the kind of question that can get this discussion back on track.
I do know one thing, it should be fairly straightforward to implement certain prototypes on the propeller. Having a downsized less expensive unit available could definitely shake things up a bit. It's just that nobody know exactly how much it would shake things up. I wish I had the requisite orders in hand so that I could put my money where my mouth is.
Leon said...
I'd rather use one of those for a small job than an expensive and large Propeller chip.
I think this is a perfect example of your strong bias, Leon. I think you should have said "I'd rather use one of those for a small job than an expensive and large Propeller or XMOS chip." At least, considering they're the about the same size and cost.
magdrop said...
I've decided to weigh in here as a way introducing myself to the members here. I chose this topic to respond to only because I cannot resist a good controversy, and definitely not because I am an expert or anything. Far from it.
If threatening bans·ends·in ONE MORE member that gets out of lurking, I propose to·schedule a·regular fighting session·every friday night!
wjsteele said...
... I think you should have said "I'd rather use one of those for a small job than an expensive and large Propeller or XMOS chip." ...
But I didn't think you guys wanted him to mention that other chip.
Actually, I wouldn't even consider the unmentionable chip for a quick hardware hack as it would be severe overkill with 400 MIPS - it would be a Propeller if I needed it's special features like video or TV.
Post Edited By Moderator (Dave Andreae (Parallax)) : 7/12/2010 7:39:39 PM GMT
Prop1.5 = current prop with PORT B implemented .... no other changes
any changes would delay it significantly, and distract from Prop2
Cluso99 said...
Ale: I am quite happy to debate ASM on another thread. It's OT here.
I would rather see any possible resources placed into the Prop 1.5, and particularly a larger die size (increased cost) with more ROM & RAM as well as the 64 I/O, rather than a smaller Prop. The price seems to be the main driving force behind this. If we could significanlty increase the volume of Prop sales, then I would expect this would be the best way to get a price reduction.
As for a 2 die solution (prop + eeprom) I would expect this would be a much more expensive solution that a seperate Eeprom. It is all to do with volumes. If I understand correctly, because of the low volumes, the manufacturer used does not have access to all the features such as eeprom, flash etc.
Wouldn't a Prop 1.6 (64 IO + 64KB ROM + 64KB RAM), cost ~$10, with the same geometry (low power) as the Prop 1 ROCK !!! Delivery could be <6 mths ???
I think Holly has made this point earlier in this thread, but it is worth making again: the Propeller uses cogs to achieve what most other chips do using interrupts. This means that a two cog prop would often need to dedicate one of the cogs to doing only "housekeeping" type functions, or running a device driver - leaving only one cog available to do any application processing.
While there are certainly applications where one cog would be enough, I doubt they would be sufficently numerous to justify the development cost of a two cog-prop. Especially as the market is already full of more suitable (and far cheaper!) alternatives.
The Propeller has been designed to be a good fit for applications that benefit from parallel processing. It is always going to be overkill for applications that don't need this particular feature.
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
In an assembler environment I think you are not correct in your assessment as to how many cogs it takes to do things....... you CAN use lots of cogs, but you don't neccessarily NEED to.
My experience shows that a single cog can run multiple assembler threads simultaneously..... this has been stated and proven numerous times including my demo at UPEW. In fact to this point I have yet to need a second cog, all wile simultaneously running full duplex 115,200 serial, I2C at 100 KHz, LCD Display, button input, and diving steppers at 5 Khz. And I'm positive that adding an encoder at 100 Khz would be a cinch.
The caveat is that one needs to run a compact and efficient scheduler and not run video in that cog. Once you add video, you will need a dedicated cog for that until a work-around for the WAITVID is uncovered, and I think there is an answer for that. But presently I have no need for video, so that is not being pursued just now.
Most folks are not adequately aware of the power of each cog in a Prop, and readily squander its cycles with those nasty WAITCNT/PE/PNE instructions...... there is a MUCH better way! That said, if you have cogs left over, then there is not much point in saving them. But if you only had 2 to start with, frugality would be required.
In a two cog prop, I could pretty much do what I could with an SX..... some things easier, and some things harder, but I believe it could be an inexpensive, comfortable, flexible and familiar solution to the SX EOL.
Large volume production favors the ARM over the current Propeller. But in low quantities, where price breaks don't factor in, and where the development costs are amortized across fewer widgets, I dispute the claim of the ARM being cheaper. At work we've just spent nearly $10,000 on a few copies of IAR Systems Embedded Workbench, and many hundreds of dollars on prototype boards. For a $2 chip, it's amazing what it takes to write code for an ARM!
In contrast, I spent less than $15, total, to develop an application on the Propeller. That included the microcontroller, the programming cable, the IDE, and everything else necessary.
BTW, I'd love a 2-cog Propeller Jr. It would be fun to see just what could be crowded onto it. A dedicated polling cog could handle a whole lot of I/O.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The early bird gets the worm but the second mouse gets the cheese.
Very impressive, and a credit to both your ingenuity and to the capabiliites of the Propeller - I'll have to have a look at your code. Can you point me to it?
However, I don't think it answers the question as to WHY I would want to do this, when by using an existing chip (costing a fraction of the price) I can do exactly the same thing - without having to do a triple backflip (with pike) to try and squeeze it into a cog!
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
Comments
I didn't initiate the discussion of XMOS in this thread, incorrect statements keep being made about their devices which I have tried to refute.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 7/6/2010 12:57:55 AM GMT
Ok - I'll probably regret this, but here goes. I have let most of your recent posts pass since like most people I figure you have a right to draw comparisons between the Prop and other chips. I don't really think the Prop suffers from such comparisons, and sometimes you even make valid and interesting points.
But you just never know when to stop. Your recent posts have nothing to do with the thread topic. It's an advertisment for XMOS - pure and simple. You would have been much better advised to just leave things alone, since you had some people in these forums on your side - either willing to support you, or at least willing to ignore your agitating for XMOS at every opportunity.
I don't think you will have quite so many on your side after these recent posts.
Ross.
EDITED: since Leon continues to post, I edited this to refer to his "recent" posts, not his "last" post.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
Post Edited (RossH) : 7/6/2010 1:06:19 AM GMT
Xmos is targeting a very different market than Parallax. Their customers are going to be mostly corporate, but looking for solutions that fit in the embedded space for relatively high-end functionality. There is nothing hobbyist or education about any product in BGA form factor.
Parallax is trying to bring as much of their functionality as possible to people who started out on Basic Stamps. These are individuals who are going to haul out a soldering iron which they will hold in their hand. Those people have no more chance of using an Xmos processor than I do of winning the Tour de France. Xmos advertises to those people (I see their ads all the time on hackaday and slashdot) but unless those people work for a company that has serious fab capability they're never gonna work with those chips.
Meanwhile you can buy a DIP40 propeller. And even the P2 will be in a package that can at least be theoretically hand soldered. And if you're working mostly with text and simple data rather than video, that will give you a whole world you never imagined could be powered by two AA batteries.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
90 * 2 = Pi
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
90 * 2 = Pi
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
90 * 2 = Pi
The very idea that some "INDIVIDUAL", some crazy techno-nerd can "Lock himself in the basement" for a few years and emerge with a miracle to rival the mighty teams of Intel and the others is pretty fantastic. The last time I know of someone doing something similar was Linus Torvalds with the creation of Linux. Linus LOVED the technology to the point that he wasn't concerned with the money. He was just as poor as the rest of us even AFTER he became famous for his operating system. It was the USERS of his system that got together and donated the money to pay off the last couple of years on the loan he took out to buy his computer.
I think there is something similar going on here with many of us. I KNOW that there are other chips out there that can do most of the things that I use the propeller for. I KNOW that there are processors with features that I WISH the propeller had, but I LOVE the "David and Goliath" nature of what Parallax represents and I will go so far as to "Hammer" a Propeller into a design that might have been somewhat better suited for another processor.
This thread is an affirmation that many of us would easily design with another Parallax processor if it met the one thing that we often can't design around: Price. Another thing that SOME of us want to make the best of is our knowledge base. A 2-4 core Propeller would let us use most of the knowledge that we have gained over the last few years without having to resort to new tools, new languages, new support, etc.
I would MUCH prefer to see the Prop II sitting on my desk in the near future over any cut-down version of the chip we already know and love, but if it WAS as simple as handing the project to someone else and saying: Cut it in half! I think many of us would "Vote with our dollars" for a Parallax solution whenever we could. This is as much about "Community" and Ethics, and triumph of the "Human Spirit" as it is anything technical.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
" Anything worth doing... is worth overdoing. "
··············································· ( R.A.H. )
····································
You will note I also posted a bigger version that may be a Prop 1.5+ (64 I/O + 64KB RAM & 64KB ROM) - I think this may be a better investment for Parallax than the current Prop 1.5 or smaller Prop, providing it could be done by someone other than Chip.
OT Leon has successfully hijacked another thread. His recent post was a pure and simple blatant post for Xmos. One must ask, how much does Xmos pay him, or how much free hardware does he get, or does he have some other vested interest? BTW don't program one of those things in ASM - the instruction set is terrible! And certainly it could not compare to a cut down Prop.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
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
I personally would not be here, if it were not for somebody mentioning "that other cool chip". For how many of us is that true?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
Then there's price. Unless you can undercut and outperform the various ARM offerings or PICS you aren't going to get noticed by business ergo no design wins. And I don't see anyone trying to show the world how well the current Prop performs against other processors except bad mouth a certain poster who dared bring up a competitor and who caused some regulars here to have a case of the vapors.
Proof is in the pudding gentlemen, not hyperbole and wishful thinking.
No, Chip is better off focusing on the Prop II and not wasting time and money on some crippled variant to please hobbyists when Parallax still doesn't have a serious design win for the Prop as it is. And consider this: In two or so years when the Prop II is released, the other vendors are going to have their own new generations of processors out as well, so his new design is going to have to really stand out.
Lastly, if you want to see what a two Cog Prop in a 28 pin package can do. Simulate it with a real Prop. Then compare it with a PIC32, AVR32, ARM, etc.
In Australia we have a saying: "As silly as a two-bob watch". Which can be explained to non-Aussies as meaning "as silly as a two-cog Prop".
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
Post Edited (RossH) : 7/6/2010 5:37:11 AM GMT
I do believe when he mentions other processors, be it XMOS or PIC etc he is merely stating that there are othe processors out there more suitable for the task and there is nothing wrong with that.
Remember that most of us using the Props are entusiastic about it and take it personally when others say it isn't as good as we think it is.
I have taken offence to Leon promoting other chips in the past but now I just don't take any notice in fact I expect it of him.
I'm just about to deliver my first commercial project with the Propeller.
I could have use a PIC because in all honesty it doesn't need lots of speed, however, it wasn't price sensitive and I prototyped the whole thing with a Prop so it made sense for me to stick with it for the production model.
Why did I do this? Well, simply because I like the architecture, it's easy and friendly and that counts a lot for me.
Everyone has their favourites, Leon has XMOS and I have Propeller, different folks, different strokes!
I have an XMOS board, I got one when the first dev board came out as did some others who frequent this forum, I tried it and decided it wasn't for me. I haven't looked at it since.
The one area that the Propeller has a massive advantage over XMOS is this forum, this is where XMOS can't get anywhere close I bet Leon wishes there was the same amount of activity on the XMOS boards as there is over here.......
Back to the topic in hand.
A two or four cog Prop, yeah I like the idea but it's not practical and commercially it would be a disaster.
It couldn't compete with the plethora of other chips out there that could beat it in so many ways.
Parallax would be better making a Prop with some security and more i/o, this is what is holding it back in the commercial market.
Roll on Prop 1.5 or whatever it is called.....
Regards,
Coley
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PropGFX - The home of the Hybrid Development System and PropGFX Lite
But the 4 COG prop ? would be a great idea, specially if the do some multi-chip package with a big EEPROM inside, that would be something ! I'd leave 32 KB of HUB RAM though.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Visit some of my articles at Propeller Wiki:
MATH on the propeller propeller.wikispaces.com/MATH
pPropQL: propeller.wikispaces.com/pPropQL
pPropQL020: propeller.wikispaces.com/pPropQL020
OMU for the pPropQL/020 propeller.wikispaces.com/OMU
pPropellerSim - A propeller simulator for ASM development sourceforge.net/projects/ppropellersim
I never went the PIC route but I do have a drawer full of AVRs, which do have the same advantage of free development tools. They usually get the small, menial things to do.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Why did I think a new, more challenging, job was a good idea ??
I would rather see any possible resources placed into the Prop 1.5, and particularly a larger die size (increased cost) with more ROM & RAM as well as the 64 I/O, rather than a smaller Prop. The price seems to be the main driving force behind this. If we could significanlty increase the volume of Prop sales, then I would expect this would be the best way to get a price reduction.
As for a 2 die solution (prop + eeprom) I would expect this would be a much more expensive solution that a seperate Eeprom. It is all to do with volumes. If I understand correctly, because of the low volumes, the manufacturer used does not have access to all the features such as eeprom, flash etc.
Wouldn't a Prop 1.6 (64 IO + 64KB ROM + 64KB RAM), cost ~$10, with the same geometry (low power) as the Prop 1 ROCK !!! Delivery could be <6 mths ???
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
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
There do seem to be many special circumstance where a two-cog prop could be a big seller. I am sure that a lot of commercial products which can be easily developed on the propeller could also be squished into two cogs. A two-cog unit driving a touchscreen could be a useful product by itself. I'm so new to this device that I don't even know if that's even possible, but I this is the kind of question that can get this discussion back on track.
I do know one thing, it should be fairly straightforward to implement certain prototypes on the propeller. Having a downsized less expensive unit available could definitely shake things up a bit. It's just that nobody know exactly how much it would shake things up. I wish I had the requisite orders in hand so that I could put my money where my mouth is.
I think this is a perfect example of your strong bias, Leon. I think you should have said "I'd rather use one of those for a small job than an expensive and large Propeller or XMOS chip." At least, considering they're the about the same size and cost.
Bill
If threatening bans·ends·in ONE MORE member that gets out of lurking, I propose to·schedule a·regular fighting session·every friday night!
-Phil
Post Edited By Moderator (Dave Andreae (Parallax)) : 7/12/2010 7:39:39 PM GMT
Prop1.5 = current prop with PORT B implemented .... no other changes
any changes would delay it significantly, and distract from Prop2
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.mikronauts.com E-mail: mikronauts _at_ gmail _dot_ com
My products: Morpheus / Mem+ / PropCade / FlexMem / VMCOG / Propteus / Proteus / SerPlug
and 6.250MHz Crystals to run Propellers at 100MHz & 5.0" OEM TFT VGA LCD modules
Las - Large model assembler Largos - upcoming nano operating system
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
That is good, very good....
·
Welcome out of the shadows!
I think Holly has made this point earlier in this thread, but it is worth making again: the Propeller uses cogs to achieve what most other chips do using interrupts. This means that a two cog prop would often need to dedicate one of the cogs to doing only "housekeeping" type functions, or running a device driver - leaving only one cog available to do any application processing.
While there are certainly applications where one cog would be enough, I doubt they would be sufficently numerous to justify the development cost of a two cog-prop. Especially as the market is already full of more suitable (and far cheaper!) alternatives.
The Propeller has been designed to be a good fit for applications that benefit from parallel processing. It is always going to be overkill for applications that don't need this particular feature.
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
In an assembler environment I think you are not correct in your assessment as to how many cogs it takes to do things....... you CAN use lots of cogs, but you don't neccessarily NEED to.
My experience shows that a single cog can run multiple assembler threads simultaneously..... this has been stated and proven numerous times including my demo at UPEW. In fact to this point I have yet to need a second cog, all wile simultaneously running full duplex 115,200 serial, I2C at 100 KHz, LCD Display, button input, and diving steppers at 5 Khz. And I'm positive that adding an encoder at 100 Khz would be a cinch.
The caveat is that one needs to run a compact and efficient scheduler and not run video in that cog. Once you add video, you will need a dedicated cog for that until a work-around for the WAITVID is uncovered, and I think there is an answer for that. But presently I have no need for video, so that is not being pursued just now.
Most folks are not adequately aware of the power of each cog in a Prop, and readily squander its cycles with those nasty WAITCNT/PE/PNE instructions...... there is a MUCH better way! That said, if you have cogs left over, then there is not much point in saving them. But if you only had 2 to start with, frugality would be required.
In a two cog prop, I could pretty much do what I could with an SX..... some things easier, and some things harder, but I believe it could be an inexpensive, comfortable, flexible and familiar solution to the SX EOL.
Cheers,
Peter (pjv)
In contrast, I spent less than $15, total, to develop an application on the Propeller. That included the microcontroller, the programming cable, the IDE, and everything else necessary.
BTW, I'd love a 2-cog Propeller Jr. It would be fun to see just what could be crowded onto it. A dedicated polling cog could handle a whole lot of I/O.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The early bird gets the worm but the second mouse gets the cheese.
Very impressive, and a credit to both your ingenuity and to the capabiliites of the Propeller - I'll have to have a look at your code. Can you point me to it?
However, I don't think it answers the question as to WHY I would want to do this, when by using an existing chip (costing a fraction of the price) I can do exactly the same thing - without having to do a triple backflip (with pike) to try and squeeze it into a cog!
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
You don't need to buy IAR Workbench - I believe there are several free C compilers (including gcc) that support these chips.
But even if a 2-cog prop could be put on the market for $2, it would still not be competitive against the ARM.
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina