Hardware-specific questions - hardware

I'm good at programming yet I feel like I don't know enough about the architecture of the hardware I'm working on.
What does the Northbridge on the mainboard do?
What does the L2 cache of my processor do?
Can Windows XP use multiple processors? Not in terms of concrete multitasking in all programs but using the capacity of all cores if needed instead of always only one core.
How can my processor/mainboard interact with multiple kinds of graphics/sound cards?

North bridge control Memory usually, http://en.wikipedia.org/wiki/Northbridge_(computing)
L2 cache info http://www.tomshardware.com/reviews/cache-size-matter,1709.html
etc etc
These answers can be resolved/answered via google and 15min :)

Related

how can build single board computer like Raspberry Pi for run OS?

my question is : how can build single board computer like Raspberry Pi for run OS ?
user ARM micro processor and debian arm os , can use USB and etc.
like raspberry pi and other single board computer
i search but find nothing for help me !!! :(
The reason you can find nothing is probably because it is a specialist task undertaken by companies with appropriate resources in terms of expertise, equipment, tools and money.
High-end microprocessors capable of running an OS such as Linux use high-pin-density surface mount packages such as BGA or TQFP, these (especially BGA) require specialist equipment to manufacture and cannot reliably or realistically be assembled by hand. The pin count and density necessitates the use of multi-layer boards, these again require specialist manufacture.
What you would have to do if you wanted your own board, is to design your board, source the components, and then have it manufactured by a contract electronics assembly house. Short runs and one-off's will cost you may times that of just buying a COTS development or application board. It is only cost-effective if you are ultimately manufacturing a product that will sell in high volumes. It is only these volumes that make the RPi so inexpensive (and until recently Chinese manufacture).
Even if you designed and had your own board built, that in itself requires specialist knowledge and skill. The bus speeds on such processors require very specific layout to maintain signal integrity and timing and to avoid EMC problems. The cost of suitable schematic capture and board layout software might also be prohibitive, no doubt there are some reasonably capable open source tools - but you will have to find one that generates output your manufacturer can use to set-up their machinery.
Some lower-end 8 bit microcontrollers with low pin count are suitable for hand soldering or even DIP socketing, using a bread-board or prototyping board, but that is not what you are after.
[Further thoughts added 14 Sep 2012]
This is probably only worth doing if one or more of the following are true:
Your aim is to gain experience in board design, manufacture and bring-up as an academic or career development exercise and you have the necessary financial resources.
You envisage high production volumes where the economies of scale make it less expensive than a COTS board.
You have product requirements for specific features or form-factor not supported by COTS boards.
You have restricted product requirements where a custom board tailored to those and having no redundant features might, in sufficient volumes be cost-effective.
Note that COTS boards come in two types: Application modules intended for integration in a larger system or product, and development boards that tend to have a wide range of peripherals, switches, indicators and connectivity options and often a prototyping area for your own use.
I know this is an old question, but I've been looking into the same thing, possibly for different reasons, and it now comes up at the top of a google search providing more reasons not to ask or even look into it than it provides answers.
For an overview of what it takes to build a linux running board from scratch this link is incredibly useful:
http://hforsten.com/making-embedded-linux-computer.html
It details:
The bare minimum you need in terms of hardware ( ARM processor, NAND flash etc )
The complexities of getting a board designed
The process of programming the new chip on the board to include bootloaders and then pointing them to a linux kernel for the chip to boot.
Whether the OP wishes to pursue every or just some of these challenges, it is useful to know what the challenges are.
And these won't be all of them, adding displays, graphics and other hardware and interfaces is not covered, but this is a start.
Single board computers(SBC) are expected to take more load than normal hobby board and so it has slightly complicated structure in terms of PCB and components. You should be ready to work with BGA packages. Almost all of processors in SBCs are BGA (no DIP/QAFP). Here is the best blogpost that I recently came across. Its very nicely designed and fabricated board running Linux on ARM processor. Author has really done a great job at designing as well as documenting the process. I hope it helps you to understand both hardware and software side of SBCs.
A lot of answers are discouraging. But, I would say you can do it, as I have done it already with imx233. Its not easy, its not a weekend project. My project link is MyIMX233.
It took me about 4-5months
It didn't cost me much, a small fine tip soldering iron is what I used.
The hard part is learning to design PCB.
Next task would be to find a PCB manufacturer with good enough precision, and prototyping price.
Next task would be to source components.
You may not get it right, I got the PCB right by my 3rd iteration. After that I was able to repeatedly produce 3 more boards all of which worked fine.
PCB Design - I used opensource KiCAD. You need to take care in doing impedance matching between RAM and processor buses, and some other high speed buses. I managed to do it in 2 layer board with 5mil/5mil trace space.
Component Sourcing - I got imx233 LQFP once via mouser, and once via element14.
RAM - 64MB tssop.
Soldering - I can say its easy to mess up here, but key is patience. And one caution don't use frying pan and solder past to do reflow soldering. I literally fried my first 2 processors like this. Even hot air soldering by a mobile repair shop was also not good enough.
Boot loading image - I didn't take much chance here, just went with Archlinux image by olimex.
If you want to skip the trouble of circuit designing between RAM & processor, skip imx233 and go for Allwinner V3S. In 2017/2018 this would be easiest approach.
Bottom line is I am a software engineer by profession, and if I can do it, then you can do it.
Why not using an FPGA board?
Something with Zynq like the Zybo board or from Altera like the DE0-Nano SoCKit.
There you already have the ARM core, memory, etc... plus the possibility to add the logic you miss.

LabVIEW + National Instruments hardware or ???

I'm in the processes of buying a new data acquisition system for my company to use for various projects. At first, it's primary purpose will be to monitor up to 20 thermocouples and control the temperature of a composites oven. However, I also plan on using it to monitor accelerometers, strain gauges, and to act as a signal generator.
I probably won't be the only one to use it, but I have a good bit of programming experience with Atmel microcontrollers (C). I've used LabVIEW before, but ~5 years ago. LabVIEW would be good because it is easy to pick up on for both me and my coworkers. On the flip side, it's expensive. Right now I have a NI CompactDAQ system with 2 voltage and one thermocouple cards + LabVIEW speced out and it's going to cost $5779!
I'm going to try to get the same I/O capabilities with different NI hardware for less $ + LabVIEW to see if I can get it for less $. I'd like to see if anyone has any suggestions other than LabVIEW for me.
Thanks in advance!
Welcome to test and measurement. It's pretty expensive for pre-built stuff, but you trade money for time.
You might check out the somewhat less expensive Agilent 34970A (and associated cards). It's a great workhorse for different kinds of sensing, and, if I recall correctly, it comes with some basic software.
For simple temperature control, you might consider a PID controller (Watlow or Omega used to be the brands, but it's been a few years since I've looked at them).
You also might look into the low-cost usb solutions from NI. The channel count is lower, but they're fairly inexpensive. They do still require software of some type, though.
There are also a fair number of good smaller companies (like Hytek Automation) that produce some types of measurement and control devices or sub-assemblies, but YMMV.
There's a lot of misconception about what will and will not work with LabView and what you do and do not need to build a decent system with it.
First off, as others have said, test and measurement is expensive. Regardless of what you end up doing, the system you describe IS going to cost thousands to build.
Second, you don't NEED to use NI hardware with LabView. For thermocouples your best bet is to look into multichannel or multiple single-channel thermocouple units - something that reads from a thermocouple and outputs to something like RS-232, etc. The OMEGABUS Digital Transmitters are an example, but many others exist.
In this way, you need only a breakout card with lots of RS-232 ports and you can grow your system as it needs it. You can still use labview to acquire the data via RS-232 and then display, log, process, etc, it however you like.
Third party signal generators would also work, for example. You can pick up good ones (with GPIB connection) reasonably cheaply and with a GPIB board can integrate it into LabView as well. This if you want something like a function generator, of course (duty cycled pulses, standard sine/triangle/ramp functions, etc). If you're talking about arbitrary signal generation then this remains a reasonably expensive thing to do (if $5000 is our goalpost for "expensive").
This also hinges on what you're needing the signal generation for - if you're thinking for control signals then, again, there may be cheaper and more robust opitons available. For temperature control, for example, separate hardware PID controllers are probably the best bet. This also takes care of your thermocouple problem since PID controllers will typically accept thermocouple inputs as well. In this way you only need one interface (RS-232, for example) to the external PID controller and you have total access in LabView to temperature readings as well as the ability to control setpoints and PID parameters in one unit.
Perhaps if you could elaborate on not just the system components as you've planned them at present, but the ultimaty system functionality, it may be easier to suggest alternatives - not simply alternative hardware, but alternative system design altogether.
edit :
Have a look at Omega CNi8C22-C24 and CNiS8C24-C24 units -> these are temperature and strain DIN PID units which will take inputs from your thermocouples and strain gauges, process the inputs into proper measurements, and communicate with LabView (or anything else) via RS-232.
This isn't necessarily a software answer, but if you want low cost data aquisition, you might want to look at the labjack. It's basically a microcontroller & usb interface wrapped in a nice box (like an arduino (Atmel AVR + USB-Serial converter) but closed source) with a lot of drivers and functions for various languages, including labview.
Reading a thermocouple can be tough because microvolts are significant, so you either need a high resolution A/D or an amplifier on the input. I think NI may sell a specialized digitizer for thermocouple readings, but again you'll pay.
As far as the software answer, labview will work nicely with almost any hardware you choose -- e.g. I built my own temperature controller based on an arduino (with an AD7780) wrote a little interface using serial commands and then talked with it using labview. But if you're willing to pay a premium for a guaranteed to work out of the box solution, you can't go wrong with labview and an NI part.
LabWindows CVI is NI's C IDE, with good integration with their instrument libraries and drivers. If you're willing to write C code, maybe you could get by with the base version of LabWindows CVI, versus having to buy a higher-end LabView version that has the functionality you need. LabWindows CVI and LabView are priced identically for the base versions, so
that may not be much of an advantage.
Given the range of measurement types you plan to make and the fact that you want colleagues to be able to use this, I would suggest LabVIEW is a good choice - it will support everything you want to do and make it straightforward to put a decent GUI on it. Assuming you're on Windows then the base package should be adequate and if you want to build stand-alone applications, either to deploy on other PCs or to make a particular setup as simple as possible for your colleagues, you can buy the application builder separately later.
As for the DAQ hardware, you can certainly save money - e.g. Measurement Computing have a low cost 8-channel USB thermocouple input device - but that may cost you in setup time or be less robust to repeated changes in your hardware configuration for different tests.
I've got a bit of experience with LabView stuff, and if you can afford it, it's awesome (and useful for a lot of different applications).
However, if your applications are simple you might actually be able to hack together something with one or two arduino's here, it's OSS, and has some good cheap hardware boards.
LabView really comes into its own with real time applications or RAD (because GUI dev is super easy), so if all you're doing is running a couple of thermopiles I'd find something cheaper.
A few thousand dollars is not a lot of money for process monitoring and control systems. If you do a cost/benefit analysis, you will very quickly recover your development costs if the scope of the system is right and if it does the job it is intended to do.
Another tool to consider is National Instruments measurement studio with VB .NET. This way you can still use the NI hardware if you want and can still build nice gui's quickly.
Alternatively, as others have said, it is perfectly viable to get industrial serial based instruments and talk to them with LabVIEW, VB .NET, c# or whatever you like.
If you go down the route of serial instruments, another piece of hardware that might be useful is a serial terminal (example). These allow you to connect arbitrary numbers of devices to your network. You computers can then use them as though they were physical COM ports.
Have you looked at MATLAB. They have a toolbox called Data Acquisition. compactDAQ is a supported hardware.
LabVIEW is a great visual programming environment. In terms if we want to drag,drop and visualize our system. NI Hardware also comes with the NIDAQmx Library which can be accessed through our code. Probably a feasible solution for you would be to import the libraries into another programming language and write code for all the activities which otherwise you were going to perform using LabVIEW. Though other overheads like code optimization would be the users responsibility, you are free to tweak the normal method flow, by introducing your own improvements at suitable junctures in the DAQ process.

embedded application

In the last two months I've worked as a simple application using a computer vision library(OpenCV).
I wish to run that application directly from the webcam without the need of an OS. I'm curious to know if that my application can be burned into a chip in order to not have the OS to run it.
Ofcorse the process can be expensive, but I'm just curious. Do you have any links about that?
ps: the application is written in C.
I'd use something bigger than a PIC, for example a small 32 bit ARM processor.
Yes. It is theoretically possible to port your app to PIC chips.
But...
There are C compilers for the PIC chip, however, due to the limitations of a microcontroller, you might find that the compiler, and the microcontroller itself is far too limited for computer vision work, especially if your initial implementation of the app was done on a full-blown PC:
You'll only have integer math available to you, in most cases, if not all (can't quote me on that, but our devs at work don't have floating point math for their PIC apps and it causes many foul words to emanate from their cubes). Either that, or you'll need to hook to an external math coprocessor.
You'll have to figure out how to get the PIC chip to talk USB to the camera. I know this is possible, but it will require additional hardware, and R&D time.
If you need strict timing control,
you might even have to program the
app in assembler.
You'd have to port portions of OpenCV to the PIC chip, if it hasn't been already. My guess is not.
If your'e not already familiar with microcontroller programming, you'll need some time to get up to speed on the differences between desktop PC programming and microcontroller programming, and you'll have to gain some experience in that. This may not be an issue for you.
Basically, it would probably be best to re-write the whole program from scratch given a PIC chip constraint. Good thing is though, you've done a lot of design work already. It would mainly be hardware/porting work.
OR...
You could try using a small embedded x86 single-board PC, perhaps in the PC/104 form factor, with your OS/app on a CF card. It's a real bone fide PC, you just add your software. Good thing is, you probably wouldn't have to re-write your app, unless it had ridiculous memory footprint. Embedded PC vendors are starting to ship boards based on 1 GHz Intel Atoms, and if you needed more help you could perhaps hook a daughterboard onto the PC-104 bus. You'll work around all of the limitations listed above, as your using an equivalent platform to the PC you developed your app on. And it has USB ports! If you do a thorough cost analysis and if your'e cool with a larger form factor, you might find it to be cheaper/quicker to use a system based on a SBC than rolling a solution using PIC chips/microcontrollers.
A quick search of PC-104 on Google would reveal many vendors of SBCs.
OR...
And this would be really cheap - just get a off-the-shelf cheap Netbook, overwrite the OEM OS, and run the code on there. Hackish, but cheap, and really easy - your hardware issues would be resolved within a week.
Just some ideas.
I think you'll find this might grow into pretty large project.
It's obviously possible to implement a stand-alone hardware solution to do something like this. Off the top of my head, Rabbit's solutions might get you to the finish-line faster. But you might be able to find some home-grown Beagle Board or Gumstix projects as well.
Two Google links I wanted to emphasize:
Rabbit: "Camera Interface Application Kit"
Gumstix: "Connecting a CMOS camera to a Gumstix Connex motherboard"
I would second Nate's recommendation to take a look at Rabbit's core modules.
Also, GHIElectronics has a product called the Embedded Master that runs .Net MicroFramework and has USB host/device capabilities built-in as well as a rich library that is a subset of the .Net framework. It runs on an Arm processor and is fairly inexpensive (> $85). Though not nearly as cheap as a single PIC chip it does come with a lot of glue logic pre-built onto the module.
CMUCam
I think you should have a look at the CMUcam project, which offers affordable hardware and an image processing library which runs on their hardware.

Multi core programming

I want to get into multi core programming (not language specific) and wondered what hardware could be recommended for exploring this field.
My aim is to upgrade my existing desktop.
If at all possible, I would suggest getting a dual-socket machine, preferably with quad-core chips. You can certainly get a single-socket machine, but dual-socket would let you start seeing some of the effects of NUMA memory that are going to be exacerbated as the core counts get higher and higher.
Why do you care? There are two huge problems facing multi-core developers right now:
The programming model Parallel programming is hard, and there is (currently) no getting around this. A quad-core system will let you start playing around with real concurrency and all of the popular paradigms (threads, UPC, MPI, OpenMP, etc).
Memory Whenever you start having multiple threads, there is going to be contention for resources, and the memory wall is growing larger and larger. A recent article at arstechnica outlines some (very preliminary) research at Sandia that shows just how bad this might become if current trends continue. Multicore machines are going to have to keep everything fed, and this will require that people be intimately familiar with their memory system. Dual-socket adds NUMA to the mix (at least on AMD machines), which should get you started down this difficult road.
If you're interested in more info on performance inconsistencies with multi-socket machines, you might also check out this technical report on the subject.
Also, others have suggested getting a system with a CUDA-capable GPU, which I think is also a great way to get into multithreaded programming. It's lower level than the stuff I mentioned above, but throw one of those on your machine if you can. The new Portland Group compilers have provisional support for optimizing loops with CUDA, so you could play around with your GPU even if you don't want to learn CUDA yourself.
Quad-core, because it'll permit you to do problems where the number of concurrent processes is > 2, which often non-trivializes problems.
I would also, for sheer geek squee, pick up a nice NVidia card and use the CUDA API. If you have the bucks, there's a stand-alone CUDA workstation that plugs into your main computer via a cable and an expansion slot.
It depends what you want to do.
If you want to learn the basics of multithreaded programming, then you can do that on your existing single-core PC. (If you have 2 threads, then the OS will switch between them on a single-core PC. Then when you move to a dual-core PC they should automatically run in parallel on separate cores, for a 2x speedup). This has the advantage of being free! The disadvantages are that you won't see a speedup (in fact a parallel implementation is probably slightly slower due to overheads), and that buggy code has a slightly higher chance of working.
However, although you can learn multithreaded programming on a single-core box, a dual-core (or even HyperThreading) CPU would be a great help.
If you want to really stress-test the code you're writing, then as "blue tuxedo" says, you should go for as many cores as you can easily afford, and if possible get hyperthreading too.
If you want to learn about algorithms for running on graphics cards - which is a very different area to x86 multicore - then get CUDA and buy a normal nVidia graphics card that supports it.
I'd recommend at least a quad-core processor.
You could try tinkering with CUDA. It's free, not that hard to use and will run on any recent NVIDIA card.
Alternatively, you could get a PlayStation 3 and the Linux SDK and work out how to program a Cell processor. Note that the next cheapest option for Cell BE development is an order of magnitude more expensive than a PS3.
Finally, any modern motherboard that will take a Core Quad or quad-core Opteron (get a good one from Asus or some other reputable manufacturer) will let you experiment with a multi-core PC system for a reasonable sum of money.
The difficult thing with multithreaded/core programming is that it opens a whole new can of worms. The bugs you'll be faced with are usually not the one you're used to. Race conditions can remain dormant for ages until they bite and your mainstream language compiler won't assist you in any way. You'll get random data and/or crashes that only happen once a day/week/month/year, usually under the most mysterious conditions...
One things remains true fortunately : the higher the concurrency exhibited by a computer, the more race conditions you'll unveil.
So if you're serious about multithreaded/core programming, then go for as many cpu cores as possible. Keep in mind that neither hyperthreading nor SMT allow for the level of concurrency that multiple cores provide.
I would agree that, depending on what you ultimately want to do, you can probably get by with just your current single-core system. Multi-core programming is basically multi-threaded programming, and you can certainly do that on a single-core chip.
When I was a student, one of our projects was to build a thread-safe implementation the malloc library for C. Even on a single core processor, that was more than enough to cure me of my desire to get into multi-threaded programming. I would try something small like that before you start thinking about spending lots of money.
I agree with the others where I would upgrade to a quad-core processor. I am also a BIG FAN of ASUS Motherboards (the P5Q Pro is excellent for Core2Quad and Core2Duo processors)!
The draw for multi-core programming is that you have more resources to get things done faster. If you are serious about multi-core programming, then I would absolutely get a quad-core processor. I don't believe that you should get the new i7 architecture from Intel to take advantage of multi-core processing because anything written to take advantage of the Core2Duo or Core2Quad will just run better on the newer architecture.
If you are going to dabble in multi-core programming, then I would get a good Core2Duo processor. Remember, it's not just how many cores you have, but also how FAST the cores are to process the jobs. My Core2Duo running at 4GHz routinely completes jobs faster than my Core2Quad running at 2.4GHz even with a multi-core program.
Let me know if this helps!
JFV

Best platform for learning embedded programming? [closed]

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
I'm looking to learn about embedded programming (in C mainly, but I hope to brush up on my ASM as well) and I was wondering what the best platform would be. I have some experience in using Atmel AVR's and programming them with the stk500 and found that to be relatively easy. I especially like AVR Studio and the debugger that lets you view that state of registers.
However, If I was to take the time to learn, I would rather learn about something that is prevalent in industry. I am thinking ARM, that is unless someone has a better suggestion.
I would also be looking for some reference material, I have found the books section on the ARM website and if one is a technically better book than another I would appreciate a heads up.
The last thing I would be looking for is a prototyping/programming board like the STK500 that has some buttons and so forth.
Thanks =]
"embedded programming" is a very broad term. AVR is pretty well in that category, but it's a step below ARM, in that it's both simpler to use, as well as less powerful.
If you just want to play around with ARM, buy a Nintendo DS or a Gameboy Advance. These are very cheap compared to the hardware inside (wonders of mass production), and they both have free development toolchains based off of gcc which can compile to them.
If you want to play around with embedded linux, BeagleBoard is looking to be a good option, only $150 and it has a ton of features.
Personally I think AVR is best for the smaller-sized 8-bit platforms, and ARM is best for the larger, more powerful 32-bit based platforms. Like many AVR fans, I don't like PIC. It just seems worse in pretty much every way. Also avoid anything that requires you to write any type of BASIC.
If you just want to play around with it, I'd suggest the Arduino platform (http://www.arduino.cc). It's based on the ATmega168 or ATmega8, depending on the version. It uses a C-like language and has its own IDE.
Myself I've worked in embedded programming for 9 years now and have experience on TI MSP430, Atmel AVR (a couple of flavours) and will be using an ARM soon.
My suggestion is to pickup something that has some extra features in the processor like ethernet controller and CAN controller, even get two or three if you can. Embedded devices are nice to work with, but once they can talk to other similar devices via CAN or get onto a network, they can become much more fun to play with.
ADI's Blackfin is another option since it's quite a straight forward architecture to program, yet can also do some fairly hefty DSP stuff should you choose to go down that route. It helps that the assembly language is quite sane too.
The Blackfin STAMP boards are an inexpensive (~$100 last I checked) way in, and they support the free GCC tools and uClinux.
Whatever architecture you choose I'd definitely recommend first downloading the toolchain\SDK and looking through the sample projects and tutorials - generally having a bit of a play about. You can often get quite acquainted with the architecture through simulation without even touching any hardware.
ARM has the nicest instruction set of the widely used embedded platforms, leaving you free to pick up the general principles of writing software for embedded platforms without getting bogged down in weird details like non-orthogonal registers or branch delay slots. There are plenty of emulators - ARM's own, while not free, is cycle-accurate; and a huge variety of programmable ARM-based hardware is cheap and easy to come by as well.
The TI MSP430 is a great platform for learning how to program microcontrollers. TI has a variety of FREE Tools and some cheap evaluation boards (starting at $20). Plus, it's a low-power, modern microcontroller.
A nice choice would be PIC18 by Microchip
It has quite alot of material, documentation, tutorials and projects on the internet
Free IDE and compiler.
you can pull your own flash writer in a few minutes.
(Although for a debugger to work you'll need to work harder)
If you're a student (or has a student email address) Microchip will send you free sample chips. So basically you can have a full development environment for close to nothing.
PICs are quite prevalent in the industry. Specifically as controllers for robots for some reason although they can do so much more.
Arduino seems to be the platform of choice these days for beginners although there are lots of others. I like the Olimex boards personally but they are not really for beginners.
Microchip's PIC range of CPUs are also excellent for beginners, especially if you want to program in assembler.
BTW, Assembler is not used as much as it used to. The general rule with embedded is if you've got 4k of memory or more, use C. You get portability and you can develop code faster.
I suppose it depends on your skill level and what you want to do with the chip. I usually choose which embedded chip to use by the available peripherals. If you want a USB port, find one with USB built in, if you want analogue-to-digital, find one with an ADC etc. If you've got a simple application, use an 8-bit but if you need serious number crunching, go 32 bits.
I'd like to suggest the beagleboard from TI. It has a Omap3 on it. That's a Cortex-A8 ARM11 CPU, a C64x+ DSP and a video accelerator as well.
The board does not need an expensive jtag device. A serial cable an an SD-Card is all you need to get started. Board costs only $150 and there is a very active community.
www.beagleboard.org
Your question sort of has been answered in this question.
To add to that, the embedded processor industry is very segmented, it doesn't have a major player like Intel/x86 is for the "desktop" processor industry. The ARM processor does have a large share, so does MIPS I believe, and there are many smaller more specific microcontroller like chips available (like the MSP430 etc from TI).
As for documentation, I do embedded development for a day job, and the documentation we have access to (as software developers) is rather sparse. Your best bet is to use the documentation available on the processor manufacturers site.
Take a look at Processing and the associated Arduino and Wiring boards.
If you just want to have fun, then try the Parallax Propeller chip. The HYDRA game platform looks like a blast. There's a $100 C compiler for it now.
I started on BASIC stamps, moved up through SX chips and PICs into 8051s, then 68332s, various DSPs, FPGA soft processors, etc.
8051s are more useful in the real world... the things won't go away. There's TONS of derivatives and crazy stuff for them. (Just stay away from the DS80C400) The energy industry is absolutely full of them.
Start with something tiny. If you have external RAM and plenty of registers... what's the difference between that and a SBC?
Many moons ago I've worked with 8-bitters like 68HC05 and Z80, later AVR and MSP430 (16-bit). However most recent projects were on ARM7. Several manufacturers offer ARM controllers, in all colors and sizes (well, not really color).
ARM(7) is replacing 8-bit architecture: it's more performant (32-bit RISC at faster instruction cycles than most 8-bitters), has more memory and is available with several IO-configurations.
I worked with NXP LPC2000 controllers, which are also inexpensive (< 1 USD for a 32-bitter!).
If you're in Europe http://www.olimex.com/dev/index.html has some nice low-cost development boards. Works in the rest of the world too :-)
For a fun project to test, have a look at xgamestation
But for a more industrial used one chip solution programming, look at PIC
For my Computer Architecture course I had to work with both a PIC and an AVR; in my opinion the PIC was easier to work with, but that's maybe because that's what we worked with the most and we had the most time to get used to. We used the AVR maybe only a couple of times so I couldn't get the hang of it perfectly but it also was nothing overly complicated, or at least not more frustrating than the other.
I think you can also order microprocessor samples from Microchip's website so you could also get started with that?
Second that:
Arduino platform http://www.arduino.cc
HTH
For learning, you can't go past the AVR. The chips are cheap and they'll run with zero external components - they also supply enough current to drive an LED straight from the port.
You can start with a cheap programmer such as lady-ada's USBTinyISP (USD$22 for a kit) which can power your board with 5V from the USB port. Get the free tools WinAVR (GCC based) and AVRStudio and get a small project working in no time.
Yes the AVRs have limitations - but developing software for microcontrollers is largely about managing resources and coping with those problems. It's unlikely that you'll experience problems such as running out of stack space, RAM or ROM when you're making hobbist projects for powerful ARM platforms.
That said, ARM is also a great platform which is widely used in the industry, however, for learning I highly recommend AVRs.
I would suggest Microchip's PIC18F series. I just started developing for them with the RealICE in-circuit emulator, but the pickit2 is a decent debugger for the price. You could say this for the AVR's also, but there is a large following for the device all over the web. I was able to have a - buggy, yet functional - embedded USB device running within days due to all the PIC related chatter.
The only thing I don't like about the PICs is that a lot of the sample code is VERY entwined into the demo boards. That can make it hard to tear out sections that you need and still have an application that will build and run for your application.
Texas Instruments has released a very interesting development kit at a very low price: The eZ430-Chronos Development Tool contains an MSP430 with display and various sensors in a sports watch, including a usb debug programmer and a usb radio access point for 50$
There is also a wiki containing lots and lots of information.
I have already created a stackexchange proposal for the eZ430-Chronos Kit.
You should try and learn from developpers kits provided by Embedded Artists. After you get the kit, check their instructional videos and videos provided by NXP, which are not as detailed as they could be, but they cover a lot of things. Problems with learning ARM as your first architecture and try to do something practicall are:
You need to buy dev. kit.
You need a good book to learn ARM assembly, because sooner or later you will come across ARM startup code, which is quite a deal for a beginner. The book i mentioned allso covers some C programming.
Combine book mentioned above with a user guide for your speciffic processor like this one. Make sure you get this as studying this in combination with above book is the only way to learn your ARM proc. in detail.
If you want to make a transfer from ARM assembly to C programming you will need to read this book, which covers a different ARM processor but is easier for C beginner. The down side is that it doesn't explain any ARM assembly, but this is why you need the first book.
There is no easy way.
mikroElektronika has nice ARM boards and C, Pascal and Basic compilers that might suite your demands.