I'm still bothered by the title of this thread. Why must it be "versus"?
I agree! They are so apples-and-oranges. I've used ARM processors both professionally and recreationally; same as the Prop. In fact I've used the Prop to debug, test, and qualify ARM-based designs. The Prop seems singularly well-suited for that sort of use.
Every time a new task comes along, it is almost reflexive as to which would be the better choice; it hardly requires conscious thought. I'm very happy that we have both tool types.
I think it's silly comparing a pricey eval board with a low cost full blown computer than can smoke it in terms of sheer cpu power, why not gen up a low cost interface board to support the Raspberry? Say something under $40 and ready to go out of the box.
This is what I'm working towards. Prop will talk to sensors and actuators, RPi will talk to user/internet and provide storage/communication/crunching/gui. The firmware for the prop, the low lever infrastructure, is mostly complete. Cost should be Rpi + 1 prop chip + 3 resistors and 2 capacitors. The prop firmware will be completed and made to run with a standard linux PC before its moved to the RPi.
Anybody interested in working the linux side please pipe in!
This is what I'm working towards. Prop will talk to sensors and actuators, RPi will talk to user/internet and provide storage/communication/crunching/gui. The firmware for the prop, the low lever infrastructure, is mostly complete. Cost should be Rpi + 1 prop chip + 3 resistors and 2 capacitors. The prop firmware will be completed and made to run with a standard linux PC before its moved to the RPi.
That's exactly the set up I have been dreaming of and proposing here for a long while now. Except of course I would do the Prop development on the Pi itself. The whole self hosted development system would fit inside your robot or whatever project it is built into.
Now that we have propgcc we can develop C code for the Prop on the Pi. Already we can develop Spin/PASM on the Pi using HomeSpun. Works on my expensive ARM board here so why not the Pi.
Then there is the new open source Spin compiler from Parallax coming along nicely that could also be run on ARM.
Propeller-Pi anyone?
Sadly I'm very pressed fro time to think about yet another project.
Except of course I would do the Prop development on the Pi itself.
Now that we have propgcc we can develop C code for the Prop on the Pi. Already we can develop Spin/PASM on the Pi using HomeSpun. Works on my expensive ARM board here so why not the Pi.
Propeller-Pi anyone?
Very close, except we will do development for the prop on the prop itself, and/or via the pi, or remote workstation over internet. Should be transparent and equivalent for any method.
To you, it should mean that you don't have to bother with a large chunk of new work, you just "hook in" from your app on what ever platform you are working:
If one is working on the Pi, the prop has some "standard" firmware that handle the sensor.actuators; if one is working from the prop (like me) I would just use whatever somebody develops as the linux app that needs my prop's sensors.
Thing we're doing is a "standard plumbing" scheme, so we can collaborate easier, without having to be expert on every single module of the complex apps. This isn't guaranteed to work, but it might be attractive for certain applications. It not ready now, I will shout and dance around (even more) when their is something to play with.
"We now have in excess of 100,000 confirmed orders for the Raspberry Pi globally and can confirm that everyone who ordered before 18th April (i.e. today!) will definitely receive their Raspberry Pi before the end of June 2012, whatever your existing order confirmation says! Those placing new orders from today can expect a July delivery."
I have several projects, actually one big project, with Linux box + prop box. I'm now planning to buy a Pi, hopefully in the second run of 100K units, if this flash-in-the-pan-longshot survives that long. Anybody else (aside from heater) thinking about a multi-architecture (is that the right word?) system?
if this flash-in-the-pan-longshot survives that long.
That's funny prof, my guess is Parallax would love that kind of order volume for Prop II launch.
Oh and the 100,000 orders, that's just Farnell, I think RS will have had a similar number too, this will be a success simply because of the price point and the form factor.
Now what is interesting is the GPIO being available to tinker with that makes it favourable to communities like ours and Arduino.
I expect there to be many many add-ons available once people start receiving their orders, no reason why they couldn't be Prop based!
Maybe it would be worth a community effort to come up with such a board especially with over 200,000 potential customers.....
Coley,Yep, "Propeller Pi". An excellent idea that may of is here seem to have. Perhaps a board in a Pi format with suitable interface and many free pins.With all possible Propeller dev tools packaged up for easy use on the Pi.
Very close, except we will do development for the prop on the prop itself, and/or via the pi, or remote workstation over internet. Should be transparent and equivalent for any method.
That's too close for me. I know you are into developing on the Prop with forth but I'd be happy with having the Prop and Pi in a box as a self contained dev system with all my familiar editors, compilers and tools running in a familiar Linux environment. Then if need be the Pi provides networking to the world for your robot/embedded gadget.
On the other hand Ebden, the guy who conceived the Raspberry Pi, has in mind it's use in stimulating and educating youngsters like the C64's and such back in the day. A machine they can "own" and tinker with rather than borrowing dads PC. So perhaps making Forth on the Prop accessible from the Pi fits right in with that philosophy.
Well, we shall see. Actually, I don't mind where the development gets done or even if it uses forth, per se; as long as my remaining hairs don't turn gray, or at least don't fall out right away.
If it costs less than a full blown PC, and is easy to use, or at least not needlessly complicated, I can be happy.
But I suggest the title of this thread is not the right question anymore. I'm thinking Pi + N Props is the thing to consider now that the CSP channels are working.
I agree the title of this thread is missing the point. Pi and Prop are not the same class of thing and should not be compared.
( I could be wicked and say that the Pi is an ultra cheap general purpose computer whereas the C3 is an over price micro-controller board)
It totally misses the point that Pi and Prop could be used together profitably each doing what the other cannot.
And yes why stop at one Prop hooked up to your Pi?
be wicked and say ... the C3 is an over price micro-controller board...
I wouldn't put it like that, I'd say it had a large variety of expensive hardware built on stock, and not every application needs such a kitchen sink approach. I bought one but have not found an app to fit it. The design is sort of a step up from the hydra deluxe retro game platform, and the extra stuff ends up not fitting my projects so far.
That's funny prof, my guess is Parallax would love that kind of order volume
Silly joke, yes; but on sources, I still see folks calling RaspberryPi "vaporware" etc. This isn't like the ten years plus development for "Duke Nukem Forever". The Prop 2 is not vaporware, it has a very well planned and thorough development process. The RP is not vaporware, the design is completed, tested, refined, and in production. The first cycle is a little rough as the players gear up, but its still amazingly big for any company to pull off, let alone a small group that does it as a side project.
But, just as with life on other planets, we cannot claim it exists until we can actually see it and touch it. My bet is I'll order my RP in August, and it will be delivered in two weeks, like any other stocked part. My Propeller-Pi might take a little longer than that, however.
"Let's be honest. The Raspberry Pi is just a gimmick and is getting free publicity from the BBC. Manufacturing and production of the Pi has been a shambles. It will be a novelty for a few lessons and then pupils will get bored. Fact."
If one really wants to used, learn and exploit Linux - any PC or laptop is more appropriate. And I am not particularly impressed with the need to crowd whole computers into a 'credit card sized' format. The Beagle Board is a better format and value. And sadly the best value Panda Board is nearly as impossible to buy as the Raspberry Pi.
I was excited about the C3 when it was supposed to intially sell for $80, but decided that I didn't really need it at $120 actual price. After all, I have a whole pile of Propellers here that I have yet to use and Propeller Proto Boards are quite reasonable. I can clone a C3 if need be.
In sum, just because it is supposed to be rock bottom cheap doesn't mean you will find it intellectually rewarding. And it doesn't have to be the smallest of small to be something that you can carry about and learn from.
Still I do like several features of the C3, including the ADC chip and SDcard interface.
Well Loopy, the first poster's statement is pretty much history today and for the later part of it could be applied to any geek device including the d Prop. The fact is those who treat computers as consumer entertain devices are not the Raspberry's market nor are those who buy the Beagle or Panda boards. And yes a lot of students from the short attention span school of thought will get bored. Who cares, the cull rate in teaching kids and young adults electronics and CS is dreadfully high anyway. You teach who you can.
For me, it's nice to have a variety of low cost, small dev boards, because it means the entry points to FPGA development to micro-controller applications available to anyone with a $100 budget and would be intellectually rewarding for most hobbyists/hackers who want to expand their horizons.
In some respects it's a replay of the 80's with all the hackable, low cost boards out there from a multitude of vendors that take the place of the Pet20, AppleII, C64, etc. All in all, I can't think of a better time to be a electronics/computer hobbyist.
As for Raspberry itself, it's future is in the hands of the early adopters that will write the future manuals and software for it. If their base gets big enough
"Let's be honest. The Raspberry Pi is just a gimmick and is getting free publicity from the BBC. Manufacturing and production of the Pi has been a shambles. It will be a novelty for a few lessons and then pupils will get bored. Fact."
How could you possibly like that first comment? It's nonsense.
He says the Pi is just a gimmick. How so? It's just another ARM board in a
growing market where we already have Pandas and IGEPS and Beagles. Only this
one is designed to hit an especially low price point, for reasons we will see
below. Where is the gimmick in that?
He says "...and is getting free publicity from the BBC". Well why not? The
Raspberry Pi is designed by a charitable organization. The Raspberry Pi
Foundation. Who have a specific educational agenda. I an glad to see the Beeb
does get behind such efforts.
He says, "Manufacturing and production of the Pi has been a shambles". Well
perhaps. On the other hand they come at it with no prior experience in such
things and have generated a demand far in excess of what they ever imagined.
Under the circumstances I think they have handled it very well.
He says, "It will be a novelty for a few lessons and then pupils will get
bored". Again perhaps. Pupils can get bored with pretty much anything you push
on them in school. Still better to have such things in schools for the few
percent who develop an interest than not.
Just to round things off the commenter says "Let's be honest." and "Fact" as if
any of the comment was either.
Now to your comment:
If one really wants to used, learn and exploit Linux - any PC or laptop is more
appropriate.
At some level I agree with you. But Ebden from Raspberry Pi has this idea that
these machines should be like other kids toy's. The kids should own them. They
should be able to do what ever they like with them. They should be able to break
them or loose them without anyone getting upset about it. All with the idea of
encouraging exploration and experiment. All barriers between a childs mind and
the machine should be eliminated. To that end these machines have to be
small and cheap. This is not quite the case with laptops and PC's yet which are
still relatively expensive and where everyone worries about breaking the
operating system.
I just don't understand why there has been so much negative reaction to the Pi
which I belive has the nobelest of intentions.
By the time you add video, keyboard, power supply and such to the Raspberry Pi - it is not really a bargain. It does take NTSC and PAL, but NOT VGA. You will have to provide a HDTV monitor for better video. Can students and schools really afford HDTV monitors for everyone?
The Propeller offers VGA as well as NTSC and PAL.
Furthermore, the C3 and all Propeller's are fully deterministic processors - while Linux is NOT and AVR chips with frequent interrupts are NOT. For true hard RTOS, the Propeller has more to offer. Parallax has always been focused on deterministic microcontrolllers, NOT multitasked OSes such as Linux. You learn different things with the Propeller and at a lower level.
Finally, the people that are selling the PandaBoard, the BeagleBoard, and the Raspberry Pi seem to be one in the same and have demonstrated over time that they are far more interested in all the media attention gained than in the actual delivery of product in a timely manner. The waiting is part of an advertising ploy. Delivery is only fulfilled when there is an additional benefit. IN the case of the PandaBoard, it is nearly impossible to locate one as it seems it wasn't cost effective.
There has been a tremendous amount of hype about cheap computers for students. The EEEpc in its original release has been one of the few to effectively deploy a complete device and that was for about $300USD. It wanted to be much cheaper, but found that a complete computer could not be made for less. Buying parts here and there is NOT going to create a lower cost student full platform and exposes the device to all sorts of damage. So the dream seems rather absurd.
BTW, I do like Linux (and strongly prefer it over Windows) and I even like AVR chips. But I strongly believe that educational hardware is highly dependent on have a good community of support. The appropriate hardware is a secondary issue. Will the Raspberry Pi or the C3 ever gain the support of educators en masse?
Most users will already have a suitable TV, many families will have several. A keyboard, PS and mouse will cost $20 or so, and most users will already have them.
Very few OSs are deterministic, and there are billions of embedded systems running on "non-deterministic" processors without any problems.
Farnell and RS are manufacturing and selling the Pi, and they have nothing to do with the PandaBoard and BeagleBoard. They have all been designed by different teams.
What is the significance of the AVR? The Pi and those other systems use ARM processors. ARM claims that some of their current chips are deterministic, BTW.
Lots of people are familiar with Linux and ARM chips. Why should there be a problem with support?
By the time you add video, keyboard, power supply and such to the Raspberry Pi - it is not really a bargain.
....
Will the Raspberry Pi or the C3 ever gain the support of educators en masse?
Very few OSs are deterministic, and there are billions of embedded systems running on "non-deterministic" processors without any problems.
Some applications require cycle accuracy. The Prop is great for that. Most applications have a less stringent timing requirement, and work pefectly well with an RTOS. I think the Prop can compete well in the low end of this segment, but there are many other solutions with interrupt-driven processors that probably work better in the mid and high range of this segment.
Applications that require full OS support are probably better running Linux. However, I suspect that there are many applications that try to use Linux in a real-time manner. Linux provides lots of real-time support, but it can be quite complicated, and it is difficult to ensure that a critical thread always gets it's time-slice when it needs it.
It is very difficult for me to see either the C3 or the Raspberry Pi as something of a first platform for new learners - especially the young. Thus, I feel that there is hype involved that started with the 'One laptop per child' culture.
With Linux, you need the ATmega (or AVR, or whatever) as a rather extensive file system is necessary just to get the OS up and running. Nonetheless, it is of educational value because Unix offers a great deal of clarity of how an OS functions. Personally, I have preferred to hack and load Linux in an ASUS router as a small Linux OS system; with two USB ports, two RS232 ports, wifi, and LAN ports, it has some very attractive features (including a power supply and complete enclosure)
With the C3 and SPIN, you don't really need a file system at all. The use of FAT16 or FAT32 is rather minimalist. But the C3 offers many more learning opportunities for one to be creative with video creation and with use of file systems. A learner is offered more pratical participation at lower levels.
Hard RTOs remain an important part of eduction. Low level programing remains an important part of eduction. Creative parallel processing remains an important part of eductions. But all these items force the student to step out a comfort zone of C programing and vast libraries of prefab software and #pragma dependency.
Still, I remain wary of those out to buy the cheapest or smallest computer. It seems they want bragging rights rather the more appropriate acquired knowledge and skill. Such items may be a lot of fun for those with established expertise, but the smallest of small creates interface problems of its own. Of course, all the buzz plays into the hands of ATmega getting a lot of publicity and advertizing men establishing a career. Either way, not much is getting done besides the buzz.
Admittedly, I am not interesting in buying anything that has an indefinite backorder.
RS and Farnell have always manufactured some of their own products. For instance, Farnell used to make power supplies and test equipment. Of course, they don't actually make the Raspberry Pi boards themselves, they just get the PCBs made and assembled in China like most manufacturers do, these days.
What on earth does the ATmega have to do with the Raspberry Pi? It's an AVR device made by Atmel.
Comments
I agree! They are so apples-and-oranges. I've used ARM processors both professionally and recreationally; same as the Prop. In fact I've used the Prop to debug, test, and qualify ARM-based designs. The Prop seems singularly well-suited for that sort of use.
Every time a new task comes along, it is almost reflexive as to which would be the better choice; it hardly requires conscious thought. I'm very happy that we have both tool types.
This is what I'm working towards. Prop will talk to sensors and actuators, RPi will talk to user/internet and provide storage/communication/crunching/gui. The firmware for the prop, the low lever infrastructure, is mostly complete. Cost should be Rpi + 1 prop chip + 3 resistors and 2 capacitors. The prop firmware will be completed and made to run with a standard linux PC before its moved to the RPi.
Anybody interested in working the linux side please pipe in!
That's exactly the set up I have been dreaming of and proposing here for a long while now. Except of course I would do the Prop development on the Pi itself. The whole self hosted development system would fit inside your robot or whatever project it is built into.
Now that we have propgcc we can develop C code for the Prop on the Pi. Already we can develop Spin/PASM on the Pi using HomeSpun. Works on my expensive ARM board here so why not the Pi.
Then there is the new open source Spin compiler from Parallax coming along nicely that could also be run on ARM.
Propeller-Pi anyone?
Sadly I'm very pressed fro time to think about yet another project.
It's already on my list of things to do with the RPi
Very close, except we will do development for the prop on the prop itself, and/or via the pi, or remote workstation over internet. Should be transparent and equivalent for any method.
To you, it should mean that you don't have to bother with a large chunk of new work, you just "hook in" from your app on what ever platform you are working:
If one is working on the Pi, the prop has some "standard" firmware that handle the sensor.actuators; if one is working from the prop (like me) I would just use whatever somebody develops as the linux app that needs my prop's sensors.
Thing we're doing is a "standard plumbing" scheme, so we can collaborate easier, without having to be expert on every single module of the complex apps. This isn't guaranteed to work, but it might be attractive for certain applications. It not ready now, I will shout and dance around (even more) when their is something to play with.
"We now have in excess of 100,000 confirmed orders for the Raspberry Pi globally and can confirm that everyone who ordered before 18th April (i.e. today!) will definitely receive their Raspberry Pi before the end of June 2012, whatever your existing order confirmation says! Those placing new orders from today can expect a July delivery."
I have several projects, actually one big project, with Linux box + prop box. I'm now planning to buy a Pi, hopefully in the second run of 100K units, if this flash-in-the-pan-longshot survives that long. Anybody else (aside from heater) thinking about a multi-architecture (is that the right word?) system?
That's funny prof, my guess is Parallax would love that kind of order volume for Prop II launch.
Oh and the 100,000 orders, that's just Farnell, I think RS will have had a similar number too, this will be a success simply because of the price point and the form factor.
Now what is interesting is the GPIO being available to tinker with that makes it favourable to communities like ours and Arduino.
I expect there to be many many add-ons available once people start receiving their orders, no reason why they couldn't be Prop based!
Maybe it would be worth a community effort to come up with such a board especially with over 200,000 potential customers.....
Coley
That's too close for me. I know you are into developing on the Prop with forth but I'd be happy with having the Prop and Pi in a box as a self contained dev system with all my familiar editors, compilers and tools running in a familiar Linux environment. Then if need be the Pi provides networking to the world for your robot/embedded gadget.
On the other hand Ebden, the guy who conceived the Raspberry Pi, has in mind it's use in stimulating and educating youngsters like the C64's and such back in the day. A machine they can "own" and tinker with rather than borrowing dads PC. So perhaps making Forth on the Prop accessible from the Pi fits right in with that philosophy.
If it costs less than a full blown PC, and is easy to use, or at least not needlessly complicated, I can be happy.
But I suggest the title of this thread is not the right question anymore. I'm thinking Pi + N Props is the thing to consider now that the CSP channels are working.
I agree the title of this thread is missing the point. Pi and Prop are not the same class of thing and should not be compared.
( I could be wicked and say that the Pi is an ultra cheap general purpose computer whereas the C3 is an over price micro-controller board)
It totally misses the point that Pi and Prop could be used together profitably each doing what the other cannot.
And yes why stop at one Prop hooked up to your Pi?
I wouldn't put it like that, I'd say it had a large variety of expensive hardware built on stock, and not every application needs such a kitchen sink approach. I bought one but have not found an app to fit it. The design is sort of a step up from the hydra deluxe retro game platform, and the extra stuff ends up not fitting my projects so far.
But since you said it first, I won't disagree.
http://www.bbc.co.uk/news/technology-17857189
It should show that it doesn't have much in common with the C3.
I've just had an update from RS: they are about to receive their second batch of boards, 7,000 this time. I might get mine next month.
Silly joke, yes; but on sources, I still see folks calling RaspberryPi "vaporware" etc. This isn't like the ten years plus development for "Duke Nukem Forever". The Prop 2 is not vaporware, it has a very well planned and thorough development process. The RP is not vaporware, the design is completed, tested, refined, and in production. The first cycle is a little rough as the players gear up, but its still amazingly big for any company to pull off, let alone a small group that does it as a side project.
But, just as with life on other planets, we cannot claim it exists until we can actually see it and touch it. My bet is I'll order my RP in August, and it will be delivered in two weeks, like any other stocked part. My Propeller-Pi might take a little longer than that, however.
Why did they remove all but 14 of the 364 comments?
"Let's be honest. The Raspberry Pi is just a gimmick and is getting free publicity from the BBC. Manufacturing and production of the Pi has been a shambles. It will be a novelty for a few lessons and then pupils will get bored. Fact."
If one really wants to used, learn and exploit Linux - any PC or laptop is more appropriate. And I am not particularly impressed with the need to crowd whole computers into a 'credit card sized' format. The Beagle Board is a better format and value. And sadly the best value Panda Board is nearly as impossible to buy as the Raspberry Pi.
I was excited about the C3 when it was supposed to intially sell for $80, but decided that I didn't really need it at $120 actual price. After all, I have a whole pile of Propellers here that I have yet to use and Propeller Proto Boards are quite reasonable. I can clone a C3 if need be.
In sum, just because it is supposed to be rock bottom cheap doesn't mean you will find it intellectually rewarding. And it doesn't have to be the smallest of small to be something that you can carry about and learn from.
Still I do like several features of the C3, including the ADC chip and SDcard interface.
For me, it's nice to have a variety of low cost, small dev boards, because it means the entry points to FPGA development to micro-controller applications available to anyone with a $100 budget and would be intellectually rewarding for most hobbyists/hackers who want to expand their horizons.
In some respects it's a replay of the 80's with all the hackable, low cost boards out there from a multitude of vendors that take the place of the Pet20, AppleII, C64, etc. All in all, I can't think of a better time to be a electronics/computer hobbyist.
As for Raspberry itself, it's future is in the hands of the early adopters that will write the future manuals and software for it. If their base gets big enough
If that is a fact, where is the evidence?
How could you possibly like that first comment? It's nonsense.
He says the Pi is just a gimmick. How so? It's just another ARM board in a
growing market where we already have Pandas and IGEPS and Beagles. Only this
one is designed to hit an especially low price point, for reasons we will see
below. Where is the gimmick in that?
He says "...and is getting free publicity from the BBC". Well why not? The
Raspberry Pi is designed by a charitable organization. The Raspberry Pi
Foundation. Who have a specific educational agenda. I an glad to see the Beeb
does get behind such efforts.
He says, "Manufacturing and production of the Pi has been a shambles". Well
perhaps. On the other hand they come at it with no prior experience in such
things and have generated a demand far in excess of what they ever imagined.
Under the circumstances I think they have handled it very well.
He says, "It will be a novelty for a few lessons and then pupils will get
bored". Again perhaps. Pupils can get bored with pretty much anything you push
on them in school. Still better to have such things in schools for the few
percent who develop an interest than not.
Just to round things off the commenter says "Let's be honest." and "Fact" as if
any of the comment was either.
Now to your comment:
At some level I agree with you. But Ebden from Raspberry Pi has this idea that
these machines should be like other kids toy's. The kids should own them. They
should be able to do what ever they like with them. They should be able to break
them or loose them without anyone getting upset about it. All with the idea of
encouraging exploration and experiment. All barriers between a childs mind and
the machine should be eliminated. To that end these machines have to be
small and cheap. This is not quite the case with laptops and PC's yet which are
still relatively expensive and where everyone worries about breaking the
operating system.
I just don't understand why there has been so much negative reaction to the Pi
which I belive has the nobelest of intentions.
Have a listen to Ebden talk about the Pi here to see what his plan is all about.
http://www.youtube.com/watch?feature=player_embedded&v=6xFzVuxldqs#!
http://www.raspberrypi.org/
The OP compared it to the C3. How about comparing the Model A at $25 to the QuickStart, which also costs $25.
The Propeller offers VGA as well as NTSC and PAL.
Furthermore, the C3 and all Propeller's are fully deterministic processors - while Linux is NOT and AVR chips with frequent interrupts are NOT. For true hard RTOS, the Propeller has more to offer. Parallax has always been focused on deterministic microcontrolllers, NOT multitasked OSes such as Linux. You learn different things with the Propeller and at a lower level.
Finally, the people that are selling the PandaBoard, the BeagleBoard, and the Raspberry Pi seem to be one in the same and have demonstrated over time that they are far more interested in all the media attention gained than in the actual delivery of product in a timely manner. The waiting is part of an advertising ploy. Delivery is only fulfilled when there is an additional benefit. IN the case of the PandaBoard, it is nearly impossible to locate one as it seems it wasn't cost effective.
There has been a tremendous amount of hype about cheap computers for students. The EEEpc in its original release has been one of the few to effectively deploy a complete device and that was for about $300USD. It wanted to be much cheaper, but found that a complete computer could not be made for less. Buying parts here and there is NOT going to create a lower cost student full platform and exposes the device to all sorts of damage. So the dream seems rather absurd.
BTW, I do like Linux (and strongly prefer it over Windows) and I even like AVR chips. But I strongly believe that educational hardware is highly dependent on have a good community of support. The appropriate hardware is a secondary issue. Will the Raspberry Pi or the C3 ever gain the support of educators en masse?
Very few OSs are deterministic, and there are billions of embedded systems running on "non-deterministic" processors without any problems.
Farnell and RS are manufacturing and selling the Pi, and they have nothing to do with the PandaBoard and BeagleBoard. They have all been designed by different teams.
What is the significance of the AVR? The Pi and those other systems use ARM processors. ARM claims that some of their current chips are deterministic, BTW.
Lots of people are familiar with Linux and ARM chips. Why should there be a problem with support?
LOL! You are so funny.
http://www.raspberrypi.org/archives/1139
Consider this the start of "support of the masses"
Applications that require full OS support are probably better running Linux. However, I suspect that there are many applications that try to use Linux in a real-time manner. Linux provides lots of real-time support, but it can be quite complicated, and it is difficult to ensure that a critical thread always gets it's time-slice when it needs it.
It is very difficult for me to see either the C3 or the Raspberry Pi as something of a first platform for new learners - especially the young. Thus, I feel that there is hype involved that started with the 'One laptop per child' culture.
With Linux, you need the ATmega (or AVR, or whatever) as a rather extensive file system is necessary just to get the OS up and running. Nonetheless, it is of educational value because Unix offers a great deal of clarity of how an OS functions. Personally, I have preferred to hack and load Linux in an ASUS router as a small Linux OS system; with two USB ports, two RS232 ports, wifi, and LAN ports, it has some very attractive features (including a power supply and complete enclosure)
With the C3 and SPIN, you don't really need a file system at all. The use of FAT16 or FAT32 is rather minimalist. But the C3 offers many more learning opportunities for one to be creative with video creation and with use of file systems. A learner is offered more pratical participation at lower levels.
Hard RTOs remain an important part of eduction. Low level programing remains an important part of eduction. Creative parallel processing remains an important part of eductions. But all these items force the student to step out a comfort zone of C programing and vast libraries of prefab software and #pragma dependency.
Still, I remain wary of those out to buy the cheapest or smallest computer. It seems they want bragging rights rather the more appropriate acquired knowledge and skill. Such items may be a lot of fun for those with established expertise, but the smallest of small creates interface problems of its own. Of course, all the buzz plays into the hands of ATmega getting a lot of publicity and advertizing men establishing a career. Either way, not much is getting done besides the buzz.
Admittedly, I am not interesting in buying anything that has an indefinite backorder.
http://www.electronicsweekly.com/Articles/01/05/2012/53098/raspberry-pi-get-it-from-rs-and-farnell.htm
What on earth does the ATmega have to do with the Raspberry Pi? It's an AVR device made by Atmel.