User avatar
meltwater
Posts: 1015
Joined: Tue Oct 18, 2011 11:38 am

Re: Raspberry Pi Vs. Alternatives

Mon May 13, 2013 11:30 am

gritz wrote:ARM get more widely adopted by more mainstream hardware and software developers...
You may be shocked if you counted the number of ARM devices which you encounter during your daily life. I would wager that ARM devices would probably out-number any other processor you have contact with (and you can combine Intel and AMD), the difference is they often do their job without a sticker on the box.

I do agree about the penny pinching mentality to some level, but cheap hardware does still open up new applications where expensive items just wouldn't be placed (price is always a key factor). There simply is no magic combination of kit, as every need is unique, but for learning there doesn't always need to be a perfect fit, as long as the value can be gained through the learning possible with it.

Granted, the out of the box experience toll is always a high price to pay, and this is something which is always being worked on to improve, perhaps TI will help this too by raising the bar.

You can learn a lot more by using an LED and a resistor with the Raspberry Pi than you would doing a 100 rounds of CandyCrush on an iPad, even if it is easier. So ultimately it really does come down to what you do with it anyway.
______________
http://www.themagpi.com/
A Magazine for Raspberry Pi Users
Read Online or Download for Free.

My new book: goo.gl/dmVtsc

Meltwater's Pi Hardware - pihardware.com

Like the MagPi? @TheMagP1 @TheMagPiTeam

Rene_is_I
Posts: 172
Joined: Tue Dec 25, 2012 12:52 pm

Re: Raspberry Pi Vs. Alternatives

Mon May 13, 2013 4:35 pm

meltwater wrote:
You can learn a lot more by using an LED and a resistor
Especially if it's a RGB LED, or even better, five of them ;)

meltwater wrote:
You can learn a lot more by using an LED and a resistor with the Raspberry Pi than you would doing a 100 rounds of CandyCrush on an iPad, even if it is easier. So ultimately it really does come down to what you do with it anyway.
Spot on and exactly why a good support structure is so important.
It's hard enough to try and get people to change from software users to software creators, therefore the easier the process of learning is made, the more people will ultimately try it and might even enjoy it.

gritz
Posts: 449
Joined: Sat Jan 28, 2012 2:33 am

Re: Raspberry Pi Vs. Alternatives

Mon May 13, 2013 5:18 pm

meltwater wrote:
gritz wrote:ARM get more widely adopted by more mainstream hardware and software developers...
You may be shocked if you counted the number of ARM devices which you encounter during your daily life. I would wager that ARM devices would probably out-number any other processor you have contact with (and you can combine Intel and AMD), the difference is they often do their job without a sticker on the box.
I was really thinking about the o.p. and with that in mind I referred specifically to "general purpose computers". Yes, there are a lot of ARM chips out there (more than people), but most of 'em are in smartphones, embedded in media players, routers or other consumer items, or wasting their lives in disposable Android tablets, rather than being at the heart of a user - configurable general purpose computer.
meltwater wrote:I do agree about the penny pinching mentality to some level, but cheap hardware does still open up new applications where expensive items just wouldn't be placed (price is always a key factor). There simply is no magic combination of kit, as every need is unique, but for learning there doesn't always need to be a perfect fit, as long as the value can be gained through the learning possible with it.

Granted, the out of the box experience toll is always a high price to pay, and this is something which is always being worked on to improve, perhaps TI will help this too by raising the bar.

You can learn a lot more by using an LED and a resistor with the Raspberry Pi than you would doing a 100 rounds of CandyCrush on an iPad, even if it is easier. So ultimately it really does come down to what you do with it anyway.
Agreed, but sometimes it's not about learning per se, but about getting the job done. Creative people aren't necessarily geeks (even creative people who know one end of a soldering iron from another). I dimly remember the First Great Personal Computer Revolution and the suckiness of much of the available kit put a lot of smart people off of computers for life, as considerations like not getting bitten by a wobbly ram pack and such meant that the creative part became secondary to the whole nerdy / obsessive problem of getting the thing to work in the first place.

I'm minded of the endless discussions on here regarding the ninutiae of Basic v. object oriented whatever from chaps of my vintage - and while that's fine and necessary I'm not sure that the next generation of creative people who utilise computers to do interesting stuff should be eclusively geeks, because we risk getting stuck into a loop of exclusivity for another generation. Junk hardware and clunky, unsupported software is the enemy of us all imo.

It is clearly possible to make a functional board for $50 (or even less). Making a functional package that doesn't put off the averagely bright potential user or those who might consider developing hardware and software for it takes effort and skill.

Rene_is_I
Posts: 172
Joined: Tue Dec 25, 2012 12:52 pm

Re: Raspberry Pi Vs. Alternatives

Mon May 13, 2013 5:25 pm

@gritz

Well said.
bth_Penguin-clapping-animation.gif
bth_Penguin-clapping-animation.gif (41.3 KiB) Viewed 3803 times

User avatar
meltwater
Posts: 1015
Joined: Tue Oct 18, 2011 11:38 am

Re: Raspberry Pi Vs. Alternatives

Mon May 13, 2013 6:43 pm

I guess we are looking at the same goal/issue from different directions (I hope both aiming for interesting discussion and debate on it).

I see highly adaptable, highly programmable, complex low cost devices, gradually becoming more accessible (obviously there is still a way to go here). Just as the ARM is making it way up the ladder into more desktop and main-stream applications, so will the tools and applications will follow. You are right, the user friendly polish costs a lot and is often forgotten or skimped on to meet the bottom-line, sometimes though it just takes time too.

Even in the days of vintage computers, being able to play around with hardware wasn't really that accessible, oh how I wanted to do so but couldn't, it was just really the software which was. But there was ways to push the limits, put in some control and you did have to learn something in order to do it. As always though, as you tread off the beaten path (where the smooth polish of money isn't used to ease the experience apple style) things do get harder and more complex (such is nature of it). The good thing is the more people who experience it, use it and hopefully smooth the process as they go the better (something which having cheap hardware may well achieve).

The beagle board black, may well help with this, but I think the foundation has already made good in-roads to it too (the original beagle boards have been around in some form for a good while now). Now, is there a market for a higher cost, smoother and easier to use system? I guess there could be, and probably is.

There might be, although I have a feeling educational drawers are stacked with such things, most of which are too costly to send home with a particularly talented kid to experiment further on. Somewhere between you have the range of toys, such as lego mindstorms, and Robosapien, which are great systems designed for learning (but again may be out of reach for many in that application). For this purpose, I think we are still reaching the middle ground.

As for high end plug and play systems to get the job done (as in real jobs), they typically exist too, but they will represent 1000s of man-hours of development and testing and are priced at that level. TI do a whole range of items to achieve this and a suite of software to support it, to produce very complex control, test and application specific systems using Lab-View for example. You'll find even a basic module costing a few $100 and a full set-up running into the $1000s (and way up).

It is an interesting balance point though, would having more expensive board pay for the development work to make a full plug and play education system more effectively than having a low cost board and trying to do it on the cheap? How many of the more expensive boards would need to be sold to still cover the costs of developing all the supporting material? A large user base encourages things to continue to develop too.

I guess the truth is that the hardware isn't really where the true costs sit, it is the skilled development which is piled in long after the actual unit has been built. Would doubling/quadrupling the cost of the board cover this? Would the value of the board be doubled/quadruped by it? Would someone else take all that and make a ton of money by selling a cheaper one...
______________
http://www.themagpi.com/
A Magazine for Raspberry Pi Users
Read Online or Download for Free.

My new book: goo.gl/dmVtsc

Meltwater's Pi Hardware - pihardware.com

Like the MagPi? @TheMagP1 @TheMagPiTeam

gritz
Posts: 449
Joined: Sat Jan 28, 2012 2:33 am

Re: Raspberry Pi Vs. Alternatives

Mon May 13, 2013 9:09 pm

An interesting and insightful post, meltwater - and I agree 100% with all of it.

It's still early days yet, regarding ARM general purpose computing. Although we have various generations of the ARM architecture there is no wider physical standard that is the modern equivalent of the IBM clone - devices are more or less proprietary. In a way this makes sense as ARM hardware tends to be focussed, small footprint, low power applications, so simply reinventing the IBM clone with available ARM hardware and currently available operating systems (which are almost exclusively variations on the Unux theme, including Google's Android shop front) isn't yet practical, intuitive or perhaps even desirable. As much as I often find standards to be a fudge, proprietary technologies can be an awful ballache too - until a particular proprietary technology gains such a foothold that it becomes the standard (and hopefully the right one wins out!)

Perhaps there isn't that much call for general purpose computers as we know them any more and the future is an extension of what we are beginning to see now - a range of boxes each built for a particular task and a few boards with more flexibility and functionality, but requiring a pretty deep skillset. The majority of consumer / hobbyist tasks will be covered, while those with more esoteric needs will continue to plump for x86. Or not...
I guess the truth is that the hardware isn't really where the true costs sit, it is the skilled development which is piled in long after the actual unit has been built. Would doubling/quadrupling the cost of the board cover this? Would the value of the board be doubled/quadruped by it? Would someone else take all that and make a ton of money by selling a cheaper one...
There's a possibility of "after you" reticence here. Very often the pioneers of new technology aren't the ones who profit from it - rather their best ideas are taken by next generation companies with more development budget (but possibly a more generic or less imspired approach) But at least then the stuff I can only do currently on x88 Win / Mac might migrate to ARM. Just so long as we're not stuck with dumb terminals and a leaky, corrupt cloud! Maybe in five or ten years we'll all know how it shakes out :)

Anyway, hats off to The Pi Foundation for (somewhat unwittingly) stepping up to the plate and revitalising hobbyist general purpose computing. It's not just about the hardware, no matter what the nerds may say! I would also love to see the RISC operating system brought up to date as I believe that the future of ARM as a platform for people who want to get work done requires an OS free from terminal faff, possible reliance on third party packages of dubious quality, and even thinly disguised spyware if it's to be taken seriously by third party developers, which kind of brings me back to my original argument of being prepared to pay a reasonable amount of money for stuff that works as it should! {/soapbox]

I realise that the last bit is pure wishful thinking though. Convincing consumers to pay for software for a $50 board will be a task and no-one would want to sink a huge amount of development cash into building an operating system for a platform that is currently a bit of a sprawl. And any operating system is nothing without third party software and hardware support, so it's the usual chicken and egg thing.

Rene_is_I
Posts: 172
Joined: Tue Dec 25, 2012 12:52 pm

Re: Raspberry Pi Vs. Alternatives

Tue May 14, 2013 6:56 am

Another thing to watch will be how much of the ARM market Intel is able to get with its range of Atom Z CPUs which I'm assuming is x86 based.
Already Motorola is using them with Android on their RAZRi smartphones.
No doubt there will be others and perhaps even Pi like boards based on them.

User avatar
meltwater
Posts: 1015
Joined: Tue Oct 18, 2011 11:38 am

Re: Raspberry Pi Vs. Alternatives

Tue May 14, 2013 9:45 am

The Atom is an interesting one since it in aiming for the market which the ARM is coming from, while ARM is reaching to the Intel market. Which architecture will suit best in which market remains to be seen (on the surface it may not make a big difference either way). ARM has done this to some extent with netbooks quite well, and even run windows 7, within those products (as far as I know, not used them really to be sure) the internal architecture isn't an issue. It isn't really ARM/Intel which makes it hard to program or use, just the stuff which sits on top. Intel have just had Microsoft working on their products for decades to smooth the process out (and they still have trouble doing so).

The direction of where computing is going is a billion pound question really. Are we going to continue with the trend of mobile devices or return to the desktop power houses, or something else.
We have Microsoft, attempting at a gamble to tie the two together (coming late to the phone party, trying to be early for the next). Google insist they aren't planning on this yet with ChromeOS and Apple made some noises but I guess not likely yet. There has also been rumbling of tablets being a short term thing (although I feel this is more wishful thinking than anything else, as the margins appear to be very low for tablets at present and laptops are having to cut prices to compare).

Apple clearly have sown up the market in the past, with Android now making up serious ground only recently. I certainly now feel Android has reached the point which Windows Mobile reached before it was axed - however it was just a lumbering zombie by that time with patch upon patch, killing it was the kind thing to do (lots of legacy).

I can understand totally about the lack of nice user experience, I wonder if it would have helped if Ubuntu had been the distro of choice (if the Raspberry Pi had an A8 Cortex or they hadn't dropped ARM11 support) since a key focus there was to close the User experience gap on Microsoft Windows (and Apple of course). In some ways they were getting there, with many novice users happily using it over windows not just for price.
For that matter if Microsoft Windows did run on the Raspberry Pi (even if it was something you paid for) that may have eased the Linux shock I guess. Even then programming on a windows environment can be just as much a nightmare as Linux sometimes, so there would still be the pain somewhere, even if it isn't on boot.
The whole route needs to be cleared really from OS, to programming methods and hardware interfaces, but these aren't traditionally publicly areas so most stages will have hidden pit-falls somewhere until they get smoothed out (be it by paid development staff or 1001 random helpful people).

It is a little like Python in some ways, it tries to be friendly by not worrying you about data types (making things easier for the beginner), but if you use it beyond the basics you soon find you have to deal with them in the end and by that stage it probably would have been easier to have just had them there in the first place. As with the discussions about the "best" programming language, there are millions of attempts to "make it easy" but usually when you try to implement something in the way you really want to do it, you will often find you are a prisoner of the "easy" language having to jump through hoops to achieve that result (which the designer of the language or programming method hadn't quite thought of).

Apple have been experts at producing a magic button device where all the options and settings are boiled off to the minimum. The trouble is they (and the developers too of course) are the ones who then decide how you do things (which I guess is fine for most people) but if your needs are different then you have no choice but to wait for them to create it for you. Behind the shiny gloss, iOS and android are just as difficult to program as the Raspberry Pi (in fact harder since you still have to use the tools and methods which they provide, so doing something different is all the more difficult).

The other side of the coin is that we a ton of choice, and a ton of information available, so it is possible to do it independently, although the choice comes at a cost (as you rightly point out), there are so many "standards" and ways to do things it is hard to find the best or easiest way to achieve something without following down several dead-ends first.

This is partly why things like the MagPi hopefully is a good thing (and I hope the stuff I produce), by documenting doing something from A to B with the steps to get there.

It might not match your needs fully, but by not being a shiny button to do something for you, you can then take that process (with all the quirks and pit-falls on the route) and adapt it. We can decide if we want to go to the restaurant and enjoy a good meal or we can buy our own ingredients and cook up something ourselves, maybe follow a recipe or two, perhaps put in a little extra salt if we like.

Now, as you say, we just need to get to the stage where switching on the Cooker doesn't need a book to do it before we can do anything really useful. I think we are getting there slowly (but much faster with the help of the Raspberry Pi community).

I've also thought the fact the foundation aren't sitting there raking in piles of cash, means people can give their time to improve things without the feeling their work is going into their pockets. As for TI, well hopefully they will be using their own resources to do similar good work, I guess if they do make a ton of money it will be down to how well they smooth the process for the general market anyway. Who knows that polished product could also emerge from somewhere too, having the cheap hardware means that developing something is cheaper too, and more value when they sell it. A growing market and competition in that market usually ends up being great for the consumer.

PS - thanks for the discussion, great stuff to think about and see other perspectives on things.

At the end of the day though, you are right. People want to be able to do stuff, and most probably don't really care that much how, as long as it works.
______________
http://www.themagpi.com/
A Magazine for Raspberry Pi Users
Read Online or Download for Free.

My new book: goo.gl/dmVtsc

Meltwater's Pi Hardware - pihardware.com

Like the MagPi? @TheMagP1 @TheMagPiTeam

Rene_is_I
Posts: 172
Joined: Tue Dec 25, 2012 12:52 pm

Re: Raspberry Pi Vs. Alternatives

Wed May 15, 2013 6:14 pm

As regards x86 based mini computers, I found this:
http://www.gizmosphere.org/

Although it's small (10X10cm), it consumes a whack of power, it's expensive at $200 and the support seems lacking.
In short, I wonder how many people would buy something like this when for a little more, one could get a mini industrial type mini-atx board and add some serial and parallel pci cards and experiment till the cows come home and since they are "standard", there is a huge choice of Distros available or even...dare I say it.....run Windows on it (at your peril of course).
Come to think of it, why don't manufactures make a multi-purpose I/O cards some thing like this from years ago:
http://www.hardwarebook.info/GeekPort

These days it would cost next to nothing to produce a PCI or PCIe card with lots of GPIOs on it and judging by the amount of "tinkering" being discussed in these forums, I doubt there would be a shortage of buyers especially if both Linux and Windows drivers were made available.
These boards could then be plugged into pretty much any old pc lying about.
Of course it would hardly be portable, but other then that, it would be a great teaching and experimenting tool.

EDIT:

A nice thing about industrial type mini-atx boards is that they still tend to have ISA in one form or another.
ISA is a piece of cake to interface to (in relation to PCI) and although slow by today's standards, it's plenty fast enough for general use and experimenting.

User avatar
Zalamel
Posts: 23
Joined: Wed May 08, 2013 11:25 am

Re: Raspberry Pi Vs. Alternatives

Wed May 15, 2013 8:45 pm

Image

I've received my Raspberry Pi!

And color me impressed! Even though I'd seen pictures, and a couple videos of it, I'm still amazed at how truly tiny the Pi really is. So tiny, in fact, that I'm inspired to construct something utilizing it.

I plan on building a small, fully functional, handheld computer. Somewhat like what this guy has done: https://www.youtube.com/watch?v=E89s2h9swIc Except, well...Better. ;)

I'll be making it out of a custom wooden case, probably in the shape of the original Gameboy, but bigger. It will house a 5" screen, dismantled USB Controller, and a micro-keyboard + trackpad combo (The same one from the video, in fact).

So not only will this work as just a general purpose, wi-fi capable computer, it'll also be the ultimate handheld emulator, thanks to it's built in, proper controller. And with the right battery, it'll hopefully boast something near to a 5 hour battery life! It should also be incredibly reliable, compared to a smart phone or tablet, due to the Pi's (From what I can see) quality construction.

It's also the closest thing I can get a to a Cyberspace Deck, like something you'd read in an 80's Cyberpunk novel.

Kind of boggles my mind that none of this would have been reasonably possible for me to do, just a couple of years ago...

As for the original purpose of making a super durable, archival survival computer...I'll just get another Pi for that. In fact, I see myself owning quite a few of these little micro computers in the future.. :lol:
meltwater wrote: PS - thanks for the discussion, great stuff to think about and see other perspectives on things.
I second that. That was quite an enjoying, and intriguing read.
Rene_is_I wrote:As regards x86 based mini computers, I found this:
http://www.gizmosphere.org/
That's actually a very neat little board! However, for that price, I'm not sure what I'd do with it, that the Pi couldn't do, albeit slower.

Rene_is_I
Posts: 172
Joined: Tue Dec 25, 2012 12:52 pm

Re: Raspberry Pi Vs. Alternatives

Wed May 15, 2013 9:08 pm

Well done, enjoy it but don't forget to add the camera module for a truly super-duper handheld computer.
http://www.raspberrypi.org/archives/3890

Since one of your applications is to use the Pi's as a computer in survival mode, have you looked at less conventional types/methods of generating electrical power?
http://en.wikipedia.org/wiki/Lemon_battery
http://en.wikipedia.org/wiki/Baghdad_Battery

Don't know if either process would yield enough current to power the Pi, but interesting none the less.

User avatar
clive
Forum Moderator
Forum Moderator
Posts: 1013
Joined: Tue Feb 07, 2012 8:19 pm

Re: Raspberry Pi Vs. Alternatives

Wed May 15, 2013 10:22 pm

Zalamel wrote:Image

I've received my Raspberry Pi!

And color me impressed! Even though I'd seen pictures, and a couple videos of it, I'm still amazed at how truly tiny the Pi really is. So tiny, in fact, that I'm inspired to construct something utilizing it.
Hi Zalamel. Love the picture; love the ideas; love the attitude (love the avatar -- Amiga FTW ;)).

Posts like yours remind me of why we make the Raspberry Pi -- good luck with the project(s)! :D

User avatar
Zalamel
Posts: 23
Joined: Wed May 08, 2013 11:25 am

Re: Raspberry Pi Vs. Alternatives

Thu May 16, 2013 5:05 am

Rene_is_I wrote:Well done, enjoy it but don't forget to add the camera module for a truly super-duper handheld computer.
http://www.raspberrypi.org/archives/3890

Since one of your applications is to use the Pi's as a computer in survival mode, have you looked at less conventional types/methods of generating electrical power?
http://en.wikipedia.org/wiki/Lemon_battery
http://en.wikipedia.org/wiki/Baghdad_Battery

Don't know if either process would yield enough current to power the Pi, but interesting none the less.
O_O'...I...I actually did completely forget about the camera board...That really would just absolutely improve the usefulness of it, as I don't actually have a camera that can record 1080p...Yet ;)

Thanks for reminding me! :D

As for those alternative methods of power, I think the Baghdad batteries might just be able to power a Pi, if you had enough of them, but I suspect the current would fluctuate too much for the Pi to reliably run off it. They might also be able to trickle charge another, modern battery, however I'm not sure how long they last, or how practical that would be. I may look into it.

I'm pretty sure the lemon batteries would be a bit too impractical, I think it takes 4 lemons to just about power 1 LED.

What I had in mind was a small water turbine connected to an alternator from a car, which could be placed in a stream or river. Then of course there's solar or wind, which are fairly reliable means of power.

I appreciate the suggestions though! :)
clive wrote: Hi Zalamel. Love the picture; love the ideas; love the attitude (love the avatar -- Amiga FTW ;)).

Posts like yours remind me of why we make the Raspberry Pi -- good luck with the project(s)! :D
Thanks Clive! I've been wanting to do something like this ever since I was a kid, so I'm pretty excited about it. The Pi, and boards like it, are finally making it possible for a relative layman like me, to build one!

Image

Return to “Off topic discussion”