ARM Designer Steve Furber On Energy-Efficient Computing

ChelleChelle writes "By now, it has become evident that we are facing an energy problem — while our primary sources of energy are running out, the demand for energy is greatly increasing. In the face of this issue, energy-efficient computing has become a hot topic. For those looking for lessons, who better to ask then Steve Furber, the principal designer of the ARM (Acorn RISC Machine), a prime example of a chip that is simple, low power, and low cost. In this interview, conducted by David Brown of Sun's Solaris Engineering Group, Furber shares some of the lessons and tips on energy-efficient computing that he has learned through working on this and subsequent projects."

Comments

so why can't i buy a !@##$% low powered computer?

[Anonymous Coward]

That means a portable computer with an ARM processor and a reflective monochrome display big enough to hold normal text pages. In other words an Amazon Kindle DX (separate wired or bluetooth keyboard is fine), but with an open OS that lets me write and run my own programs without having to jailbreak past some DRM crap. Somebody please make something like that? Please??

Energy Efficient Tips

[Anonymous Coward]

Use floats where doubles aren't necessary
Bit shift instead of multiply by powers of two
xor r,r instead of mov r,0
Turn your computer off at night (record the uptime with a pen+paper if you want to keep a running total)
Compile to a 32bit target for apps that don't need 64bit addressing
etc.

Re:Energy Efficient Tips

[Darkness404]

(record the uptime with a pen+paper if you want to keep a running total)

...That isn't true uptime. The point of uptime being a bragging right isn't that you have an APC, but rather your computer is configured correctly so it doesn't randomly run out of memory and crash, your hardware doesn't overheat after a week of constant usage, etc. Almost -any- computer can work 12 hours and not fail. Finding a computer that will last a year or more without rebooting is hard. Simply adding up the amount the computer has been on is not uptime, not in the least.

Re:Energy Efficient Tips

[WrongSizeGlass]

When did uptime become 'geek gas mileage'? I really have to read more articles on /.

Re:Energy Efficient Tips

[Lunix Nutcase]

For a couple of decades?

Re:Energy Efficient Tips

[Lennie]

"Finding a computer that will last a year or more without rebooting is hard."

Really ?

In our serverroom we have a few which don't get as many security updates (internal systems), they are doing fine, thank you.

Re:Energy Efficient Tips

[Lennie]

If we had made use of the services of ksplice it would have been even easier to find them.

Re:Energy Efficient Tips

[Darkness404]

Well, by computer, I meant not a server and one that has constant general use. Of course its easy to find a server that will run for years. Its a lot harder to find a computer being constantly used, updated and on the internet that hasn't been rebooted in a year.

Re:Energy Efficient Tips

[h4rr4r]

Use ksplice. done.

Re:Energy Efficient Tips

[Anonymous Coward]

your hardware doesn't overheat after a week of constant usage, etc.

I loved my Acorn: the first computer I ever had and the only one that could get away with nary a heatsink nor fan to be seen.

If not for the lack of software it would have pissed all over the IBM clones too. Still I had Mad Professor Moriati and Lemings (sic), so I was one happy kid.

Re:Energy Efficient Tips

[Verdatum]

http://www.catb.org/~esr/jargon/html/U/uptime.html

Re:Energy Efficient Tips

[Pentium100]

For me it is hard to find a year that does not have a long (at least 35 minutes, the time it takes my UPS to discharge) power outage.

Re:Energy Efficient Tips

[mirix]

Bit shift instead of multiply by powers of two

I'd think a decent compiler should do that automagically, no?

Re:Energy Efficient Tips

[dfsmith]

Use floats where doubles aren't necessary

That assumes that floats are more energy efficient than doubles. Not necessarily true on architectures that have hardware floating point and save energy by omitting the float portion. They convert the float to double, do the math, then convert back. Check your architecture!

On my Intel Core 2, the execution speed is different by about 0.06% between float and double (edge to double). Can't speak to power consumption though. On my Intel Atom 230, the floats are faster by 0.4%. That's almost certainly saving power, but not a lot.

Re:Energy Efficient Tips

[FrangoAssado]

Maybe he means you'll save the energy the compiler would spend to optimize a multiply into a bit shift :-)

Re:Energy Efficient Tips

[ColdWetDog]

When did uptime become 'geek gas mileage'? I really have to read more articles on /.

It would be understandable if we were talking about penile uptime, but computer uptime just doesn't do much for me.

Re:Energy Efficient Tips

[TheRaven64]

Actually, you're only half right. On ARM, there is typically no double hardware, so you get a very slow path for 64-bit floating point arithmetic. On your Core 2, it's more complicated. The x87 unit only supports 80-bit floating point values. This means that any float or double will be sign-extended when it is loaded into a register. You gain a bit better cache usage from using 32-bit floats, but that's it.

No both, however, if you compile with SSE then you will be using a vector for all floating point operations. With floats, the compiler can pack concurrent operations on four of them into a single instruction, with doubles it can only pack two. I'm not sure about the Atom, but I vaguely remember that it splits SSE ops in half, so you really do two 64-bit operations. Either way, you can do twice as many float operations in the same power envelope, as long as your code is suited to vectorisation.

Modern compilers prefer to target SSE instead of x87, because register allocation with x87 is painful. Most operations only work between the top two registers in the 'stack' so you need a lot of register-register copies in a typical bit of x87 code (which burns i-cache too). This is one of the main reasons why you see a performance improvement in x86-64; if you have a 64-bit chip you can guarantee the presence of SSE, so the compiler will always use SSE instead of x87 when compiling 64-bit code. If you're someone like Apple and don't support pre-SSE chips, you can also do this and get the same benefit in 32-bit mode.

Re:Energy Efficient Tips

[colonelquesadilla]

unfortunately penile uptime doesn't do much for the average slashdotter.

Re:Energy Efficient Tips

[Anonymous Coward]

In *OUR* server room we have machines labelled "re-boot every Olympics". These are Arcsoft machines, running live and on display in a Computer Museum.

Re:Energy Efficient Tips

[h4rr4r]

Every two years is not that long really.

Re:Energy Efficient Tips

[Anonymous Coward]

Eh, our VMS hasn't been down since 1984.

Re:Energy Efficient Tips

[i8degrees]

Well, by computer, I meant not a server and one that has constant general use. Of course its easy to find a server that will run for years. Its a lot harder to find a computer being constantly used, updated and on the internet that hasn't been rebooted in a year.

Huh? You just defined the very such piece of computer technology otherwise known as a "server". I don't post here often, but this comment just makes my jaw drop unusually low for some reason, so I apologize ahead of time if I sound offensive...

1) A server, as described in a TCP/IP client & server topology, is a computer that is "always on" the internet, or so you very well hope so, especially if it is being used for "business" oriented tasks. Those packets get served to your "clients" whom serve you back, and this process cycles over until satisfied. Does this sound like another real world scenario to you?

2) A server is often updated at least once a week, sometimes more, depending on the sort of update this is. You could even argue such a hardware piece is often the most updated piece of equipment, generally speaking. (Some development systems could certainly be an exception, for instance!)

3) A server, by definition, is tasked with the role of serving, or "hosting" information back to said requesting clients, whom could and often are making all sorts of requests all at the same time (concurrently), hitting up your resources significantly as the volume of concurrency rises. This often taxes the hardware in several different key areas, sometimes very much like a video game (on steroids, even!) especially if you do not have a network topology in which allows better distribution of said serving tasks across.

In summary, have you ever tried keeping one, much less several parallel video surveillance systems up for months upon years at a time, with minimum downtime as any loss of said outage could leave you liable to your real-life business clients? Sometimes you don't have the luxury of redundancy to save your ass. When incoming bandwidth is not your bottleneck, you've gotta keep your disks seeking every bit as fast as they can keep up to, along with fast concurrency capabilities, along with allowing proper time for distributed mass A/V encoding of said daily video surveying operations to occur before considering that a solid "tape" and pushing those off to another system for backup, which mind you, is getting hit by many other requests in parallel to this first example.

Okay, so I did a terrible "summary", especially now as I am continuing onward with this new paragraph, but in other words ... real life servers do not often have it easy in their life! Their CPU, memory and disk I/O are often pushing near or at maximum utilization if you are trying to get every dollar out of your time & money spent. I push my personal workstations plenty hard, but not anywhere close to what some of my servers have gone through on a daily basis!

P.S. My examples given are of only my own personal real life business experiences in the "small time" playing league -- this represents 0.01% of what big name brand servers running many of our web sites we most all depend on every day, free or paid. Yes, my "servers" were indeed a mixture between the real stuff and the commodity desktop class hardware. Whereas I cannot prove this, I've got both win32 and personal *nix workstations that have indeed been alive w/o reset for a year. When your hardware does not fail, satisfied setups and plenty of other systems running to do various testing on, with the right style of administrated setup you can easily scale to the years. ksplice can be used on personal desktops / workstations, too.

You only *simply* have to devote a lifetime to computer science to learn the trade down to a true art.! :-)

Re:Energy Efficient Tips

[Anonymous Coward]

Nice overview, but I do not believe it to be completely correct.

Although it is technical possible to execute 4 float or 2 double operations in a single SSE operation, the user usually has to explicitly code for those operations to take full advantage. Vectorization (combining multiple float operations into a single instruction) is often difficult because it is only possible with the same operation on "adjecent" values; It is not possible to combine arbitrary operations. As a result, making efficient use of this capability often requires a rewrite of the algorithm, and compilers are usually not very good at it.

SSE1 only operated on float values, but SSE2 added support for doubles as well, and the instructions set specifically makes it possible to also operate on single float or double values. This is indeed due to the clumsiness of the x87 FP stack.

However, the difference in performance between the FP stack and single SSE2 instructions is not always as huge as could be expected. Underneath the FP stack based ISA, modern X86 processors implement a register based architecture. And while the stack only allows to operate on the stack top, the ISA offers an exchange operation to swap arbitrary stack elements. Because the underlying architecture is register based, such an exchange just reassociates a different register with a different stack location, and no data is moved. As a result, this operation is extremely efficient (practically free, except for opcode).

There are, however, other advantages to using SSE instead of FP. For example, rounding a float to an integer is extremely expensive on the FP unit. This is because a global mode switch has to be performed to switch the rounding mode to truncation, and back to rounding around such a conversion. Switching this mode basically stalls the FP pipeline. In SSE on the other hand, a dedicate instruction was added to truncate to an integer for exactly this reason.

As for the performance benefit with X86-64, I believe this to be largely caused by the larger register set. in IA32, there are only 7 registers which are sort-of general purpose. The hardware has to spend quite some effort in tracking loads and stores, to figure out it can actually keep such values in registers too (underneath, a modern X86 processor has lots of registers). X86-64 makes it a lot easier for the compiler to express what it wants.

Re:Energy Efficient Tips

[jonaskoelker]

This is one of the main reasons why you see a performance improvement in x86-64

Isn't that just a question of setting the right -march=foobar --omg-optimizations flags in /etc/emerge.make.conf.zomfg.fast?

Netbooks will make the ARM viable.

[Anonymous Coward]

Thank goodness for netbooks. They will finally make the ARM a viable CPU for use in a wide range of higher-end PCs. We just need to see Windows support for ARM, and then we'll be well on our way towards it being a widely available option.

Frankly, the ARM is a much nicer architecture to target when writing compiler back-ends and when writing high-performance assembly code by hand. It just isn't riddled with the archaic crud that the x86-32 and x86-64 architectures are littered with.

Re:Netbooks will make the ARM viable.

[Lunix Nutcase]

We just need to see Windows support for ARM, and then we'll be well on our way towards it being a widely available option.

And you also mean the porting of thousands and thousands of x86 apps as well? If the ARM version of Windows can't run the apps people want, they aren't going to by an ARM netbook.

Re:Netbooks will make the ARM viable.

[Anonymous Coward]

So because netbooks with less powerful CPUs are selling, this is going to translate into a demand for less powerful CPUs in high-end PCs? Why? And even if Microsoft were to back ARM that doesn't really help. The Linux world is used to having every architecture supported because it is all done in one place - the maintainer of the distribution. The Windows world is very fragmented, but more importantly it is (for the most part) BINARY ONLY. If some exotic Linux app isn't distributed in binary form for your distribution/archtecture, you more than likely can get the source and build it yourself (possibly with some tweaking required). On Windows, if Adobe doesn't offer Photoshop for ARM then you're shit out of luck. This is a much bigger hurdle by far than whether or not Microsoft is going to support ARM. The Windows world is still in the relative infancy of supporting the x64 platform.

Re:Netbooks will make the ARM viable.

[__aasqbs9791]

Just out of curiosity, while I've done some programming professionally, I haven't touched C or Assembly in well over 15 years, since I needed immediate results and haven't had bosses that allowed anything less, how much work is it to convert something from X86 to ARM? Assuming you didn't write it with the intention of every needing to do so, vs having planned for such a possibility.

Re:Netbooks will make the ARM viable.

[DanTheStone]

You're forgetting that the future of low-powered home machines is just in accessing the internet. Everything will just be done in the browser, so the computers only need to run the browsers.

This post can be read as serious or sarcasm; I leave it up to the reader.

Re:Netbooks will make the ARM viable.

[ChatHuant]

And you also mean the porting of thousands and thousands of x86 apps as well?

That's what emulation is for; if MS decides to port Windows to ARM, they'll probably provide an emulator and a shim layer for native calls, and at least support a number of major Windows apps (even though they'll probably run S.L.O.W.L.Y. to begin with). Silverlight and .NET apps should also run more or less out of the box.

Re:Netbooks will make the ARM viable.

[ogdenk]

Windows CE has been available on ARM for over a decade.

Besides, I want an ARM netbook because it's NOT a PC and DOES NOT run Windows.

Re:Netbooks will make the ARM viable.

[del_diablo]

What conversion? If you write an app in C++, it RUNS on anything with a compiler. Now, if you use bad compilers such as MS one there will even be trouble with compiling it on other platforms. I.E: gcc and llvm compitable means it will compile for anything those 2 can compile for(litteraly about anything). You never plan for it, instead you do the code properly the first time. What to not do: overdone amount of assembly(needs to be changed), compiled against binary blobs(needs to be ported), using weird libaries, ineffective coding, using a weird compiler, etc.

Re:Netbooks will make the ARM viable.

[emt377]

Just out of curiosity, while I've done some programming professionally, I haven't touched C or Assembly in well over 15 years, since I needed immediate results and haven't had bosses that allowed anything less, how much work is it to convert something from X86 to ARM? Assuming you didn't write it with the intention of every needing to do so, vs having planned for such a possibility.

Depends on the ARM CPU. ARM7/ARM9 are alignment sensitive. ARM Cortex has a bus/cache interface that allows arbitrary alignment. Porting to the former may be difficult depending on the software, or may simply be tedious, the latter is usually as easy as a recompile if the platform and toolchain is similar (e.g. Linux+gcc).

Re:Netbooks will make the ARM viable.

[Tumbleweed]

And you also mean the porting of thousands and thousands of x86 apps as well?

But most people don't USE thousands and thousands of apps. Or even hundreds of apps. Most people don't even use dozens of apps. You got your web browser and your IM client (maybe), (maybe) an email client, a DVD player, and I bet that's it for the vast majority of computer users outside of work. IM & email clients can be handled by the browser. Seriously - Chrome OS or Android running on a nice 1-2gHz dual-core ARM beast with hardware accelerated HD video w/ 4GB of memory is probably all that most people need at home, as long as it's got a decent screen, keyboard and mouse, they're set.

Download your Android apps, of which there *are* thousands (though many different versions of a much smaller number of TYPES of applications) for expanding into more obscure things. Most of which would be games, of course.

I'd like to see Chrome OS & Android merge sooner rather than later. Absolutely no point in having these be separate projects - with the rise of 'superphones,' there's not that much difference in horsepower anymore, and one platform target is better than two from a developer standpoint (usually).

IMO, anyway. :)

I'm hoping the day when I can take my dual-core 1+ gHz superphone out of my pocket, put it in a dock at home to get a big screen, real keyboard and mouse and true broadband home connection. Shouldn't be more than a year away. C'mon, Sprint, with the Supersonic! I'll fire up the beastly machine when I need to run Photoshop or whatever, but most of the time at home, my computer use is watching TV/movies/websurfing/email.

Re:Netbooks will make the ARM viable.

[Dog-Cow]

You haven't actually ported any non-trivial app, have you?

Re:Netbooks will make the ARM viable.

[mzs]

Does Cortex still behave like this?

You have a struct { int8_t a; int16_t b; int16_t c; }

It will get layed-out on PPC, SPARC, and X86 like so:

a0bbcc

Will it get laid-out in ARM like so?

a0bbcc00

That's how things were often configured in ARM toolchains because in this way the same sequence of instructions can be used to access an array of such structures, not a different sequence for the ones that are aligned on 32-bits and those that are not. Do you loose the useful rotations with misaligned loads with Cortex then?

The strict alignment is already there in portable code since on PPC floats and doubles (and vector types) need to be aligned properly and on SPARC every type needs to be. But what I described above is a different wrinkle about how things get padded that can and does trip-up coders since they are new to it.

Re:Netbooks will make the ARM viable.

[AmberBlackCat]

I wonder how much energy we could save if Windows and Linux would only load what's necessary for you to load an application, rather than load a million "services/processes" on startup. I think I'd be fine if my computer loaded just the mouse, keyboard, and display support and a background picture. Then if the computer needed to generate a sound or play an mp3, it could load the audio support. And if I need the network connection, it can load the network support. And if I plug in my iPod, it can load iTunes and all of the baggage that comes with that. There are times I use my computer without any of those things, yet they all load anyway. Aside from saving energy and making startup faster, I bet the hard disk would last longer with less reads. Why can't they go back to only loading anything when it's needed, rather than assuming you'll need them?

Re:Netbooks will make the ARM viable.

[AmberBlackCat]

They'd have to make an app store that works exactly like the iTunes store. That is, give people a way to sell apps for a low cost and make it really difficult for most people to get them without paying. It worked for the iPod. I'm thinking it didn't work for Android cause they left out the DRM part, which is good for my brother but bad for people trying to sell the apps.

Re:Netbooks will make the ARM viable.

[hedwards]

You can do that, but you've just given up most of the advantage to ARM. In general we should be moving away from the x86 architecture, it's kind of kludgy and includes workarounds for things which haven't really been necessary for over a decade. The only reason why there's a debate is that there's a huge code base that requires emulation when moved to ARM.

Re:Netbooks will make the ARM viable.

[h4rr4r]

So disable them. Not hard.

No one wants to wait. Why in the heck would I want to modprobe in the stuff for my network card rather than wait the tenth of a second it takes to do it on boot?

Re:Netbooks will make the ARM viable.

[Darinbob]

The ARM has its own cruft though. Thumb, Thumb 2, Jazelle... The first ARM was a nice clean architecture, something worth being studied in school. After that though you have a lot of history and tweaking and ugly necessities of being backwards compatible with older designs.

In some sense that's not a bad thing. ARM was never designed to be in the niches that it finds itself in really. It is low power because it is simple, and it is simple because it started life as a very small RISC architecture. The fine article is discussing the design of the first Acorn RISC Machine, which is not at all the same thing you're going to find on a laptop or netbook. When you get to the beefier and faster ARM varieties they are considered low power but only because it is being compared to the ridiculous competition from Intel.

The main niche it is in now is in embedded systems; the designs can be licensed to others so you have a wide variety of manufacturers to choose, lots of system-on-chip options, and you can even embed it in your own ASIC designs. Thumb mode makes it fit nicely between tiny/low-power 8-bit designs and high performance/big-bus 32-bit options.

Netbooks is a new thing (and vaguely defined). But it presumably wants the biggest small CPU you can get, and that's where the ARM fits.

Re:Netbooks will make the ARM viable.

[mzs]

So I looked a bit and I don't see anything new in cortex that allows to remove this traditional ARM padding.

Re:Netbooks will make the ARM viable.

[marcansoft]

Run PowerTOP on Linux (and use a tickless kernel, of course). There are some offenders, but most of those background services aren't using any power. As long as the processes are sleeping most of the time and don't wake up often (once every few seconds at most), they aren't going to increase your power usage.

There are a few slightly annoying ones (ntp tends to wake up once per second, and I think mysql wakes up twice per second), but most of the crap comes from poorly-written GUI apps that poll for stuff or feel the need to wake up tens or hundreds of times per second. Bad user preferences also don't help (hint: anything that's moving on the screen at any sort of framerate while the computer is otherwise idle is going to massively increase power usage over a truly quiescent CPU).

Re:Netbooks will make the ARM viable.

[ratboy666]

I've converted code. NT to NT, x86 to MIPs and Alpha.

Converted a scripting language tool; just about 5,000 places in the code needed correction. Note that "endian", OS and alignment issues didn't come into it at all. Still, 5,000 changes...

I imagine that other Windows/x86 specific applications may be just as nasty.

I imagine an x86 translator would be a better idea if attempting "Windows on ARM".

Re:Netbooks will make the ARM viable.

[pslam]

Depends on the ARM CPU. ARM7/ARM9 are alignment sensitive. ARM Cortex has a bus/cache interface that allows arbitrary alignment. Porting to the former may be difficult depending on the software, or may simply be tedious, the latter is usually as easy as a recompile if the platform and toolchain is similar (e.g. Linux+gcc).

C code which accesses data unaligned is illegal anyway. Correctly written C and C++ programs which do not assume structure packing, size or alignment will compile and run without changes.

Sadly, there are a lot of programs written assuming otherwise. Usually stuff written for an x86/Windows environment. You'd be surprised just how much GNU/Linux source compiles and runs without a single change: basically the entire of Ubuntu. Hell, I used to run a Debian ARM system back in 1999. This isn't news to me, at least.

Re:Netbooks will make the ARM viable.

[lenski]

I am porting an IEEE 802.15.4 (physical layer compatibility only, proprietary protocol) system from an old school 8/16 bit CPU to a cortex-M3 (Atmel AT91SAM3U4) right now. Having been coded on the byte-wide machine, all sorts of protocol structures have odd alignments.

I have verified that the new CPU is tolerant of unaligned accesses, both of shorts on odd boundaries and longs on both odd byte and odd-halfword boundaries. Forcing GCC to compile the "unnatural" packing requires the use of an attribute:

typedef struct
{
        unsigned char uc1;
        unsigned int ui1 __attribute__((packed));
        unsigned short us1 __attribute__((packed));
} Pstruct;

(I am typing this from memory, it may not be just right but should communicate the basic idea.)

Section 13.10.5 (page 99) of the SAM3U programmer's guide states that unaligned accesses are supported only by the load/store instructions LDRx and STRx. Various system management accesses (exception vectors, for instance), including the thumb-2 instructions must be properly aligned. The LDRx and STRx instructions are the standard memory access instructions and therefore support unaligned access, possibly with a performance impact.

Don't try to use the load multiple, store multiple or exclusive access instructions, or try to run the stack on unaligned boundaries.

Re:Netbooks will make the ARM viable.

[lenski]

Not to sound disagreeable, I had exactly the attitude you describe. But having been convinced (OK, forced) to develop for a Cortex-M3 (thumb-2 ONLY) part, I've become convinced that the thumb-2 instruction set is actually very good. It offers excellent code density, while still allowing me to code "just the way I want". I have checked several basic routines that we depend on for our work, and as far as I am concerned, the ARM guys have hit the sweet spot for the work we're doing in our embedded project.

I checked a fair amount of bit-twiddling functionality, including CRC-16 (big and little-endian), a few basic crypto functions, and general structure access.

Though not specifically related to thumb-2 "cruft", I believe another significant improvement in the ARM approach is that they have built an excellent stack/exception entry and return model that does a surprisingly good job of reducing our dependence on assembly language escapes or functions. Almost the entire processor, peripheral blocks, system control initialization is done in C with the contribution of a few ASM escapes that are coded to behave as ordinary functions. In the cortex world, ARM refers to this as "CMSIS", the "cortex Microcontroller Software Interface Standard".

Re:Netbooks will make the ARM viable.

[mzs]

Could you do this for me:

psize.c:
typedef struct {
                unsigned char a;
                unsigned short b;
                unsigned short c;
} pstruct;

int psize = sizeof (pstruct);

$ cc -c foo.o

And then dump out psize to see what it is. My guess is it will be 8 instead of 6, but that was since my tool chain did not assume ldhw. If I put the packed attribute on member c, then it would be 6, but then the code would have many ldb instructions.

So maybe I am dating myself, is it now common for tool chains to simply assume ldhw is there and no longer align to 32-bits in such cases like above? (ie there would be only one byte of padding after member a.) Also I used to get a useful rotation if I had the bits set to little endian and not to trap on misaligned access. Is that now no longer the case?

For example if I had this at address zero (for simplicity sake):

01234567 89abcdef

and I loaded this:

long v = *((long *)2);

Then v would be 0x23016745 if memory serves. Now that would be 0xab896745?

Re:Netbooks will make the ARM viable.

[imgod2u]

And you also mean the porting of thousands and thousands of x86 apps as well? If the ARM version of Windows can't run the apps people want, they aren't going to by an ARM netbook.

I think Apple has pretty much proved that the actual number of applications people want to be able to run isn't anywhere near in the thousands and thousands. The vast majority of people basically want a browser, document editor, spreadsheet editor, presentation editor -- basically Microsoft Office and IE/Firefox/Chrome.

A media player, a media store, and various games. That's about it. In that list, the only category in which individual efforts must be made to port the program are games. But as we've seen with the iPhone, if you build the platform, they will come.

I think netbooks/tablets will be the point where traditional PC apps stop being desirable. With the form factor and the desired battery life, people will not want to run the same programs on a netbook as on a laptop/desktop. A 10" screen will be eaten up by Office ribbon and there's no way to play WoW on such a thing.

The question isn't whether Windows 7 will be ported to ARM. The question is whether Windows Phone 7 (which primarily runs on ARM) will soon be preferred over Windows 7 for everything up to laptops.

Re:Netbooks will make the ARM viable.

[lenski]

Not at work yet and too lazy to set up the VPN connection...

Initial response is that the compiler's default is natural alignment, which needs to be disabled by use of the explicit __attribute__((packed)).

I just finished sweeping our snowy driveway, and am headed into work soon. I'll let you know what I learn.

You refer to "cc" as if this were compiling natively. FWIW, my use case is cross-compilation of thumb-2 code on an X86_64 host.

Re:Netbooks will make the ARM viable.

[noidentity]

Heh, I had to deal with this structure padding issue in some of my "portable" libraries. For convenience, I define a struct that matches a file header format. It's all char/unsigned char/arrays of such, so most compilers shouldn't pad it. But apparently on ARM it was being padded at the end to a multiple of 4 bytes. I solved this by not relying on sizeof(Mystruct).

In all versions, I have assertions to be sure the compiler isn't inserting any padding. That's how I originally found out about the ARM issue, due to one of these failing. So you can chastise me for relying on lack of padding between members, but not for assuming it, because I don't. In my current code, I assert that offsetof(Mystruct,last_member) is correct, BTW.

Re:Netbooks will make the ARM viable.

[ncc74656]

And you also mean the porting of thousands and thousands of x86 apps as well? If the ARM version of Windows can't run the apps people want, they aren't going to by an ARM netbook.

Haven't we been down this road before? Twice, in fact?

Re:Netbooks will make the ARM viable.

[mzs]

Yeah I used gcc cross compiler, it was installed as ccarm when I did ARM stuff, I just used cc as a generic term for "C Compiler."

I'm waiting eagerly since I had SO MUCH network code that did not work right due to the extra padding of shorts in structures to make such structures always 32-bit aligned. I would send and receive using sizeof and so many things would end-up having extra 2 bytes of padding here and there being not at all what I expected at first and of course not working between ASM, X86, and SPARC versions.

Back then I think I could only apply the packed attribute to an entire structure not individual members and these were header files I had not written. I ended-up putting the troublesome functions in seperate C files and compiled them with -fpack-struct or some such option and I put explicit int8_t xyz_padN (N = 0, 1, 2, ...) members in them so things would line-up correctly for ARM packed structs and other archs that were compiled without the option. It was a true PITA with lots of use of cscope.

Re:Netbooks will make the ARM viable.

[lenski]

typedef struct {
                unsigned char a;
                unsigned short b;
                unsigned short c;
} pstruct1;

typedef struct {
                unsigned char a;
                unsigned short b;
                unsigned short c;
} __attribute__((packed)) pstruct2; ..... in "main"....

        printf("sizeof(pstruct1)=%d, sizeof(pstruct2)=%d\r\n", sizeof(pstruct1), sizeof(pstruct2));

...and the output:

sizeof(pstruct1)=6, sizeof(pstruct2)=5

I'll check back occasionally to see if you have further questions; since I'm in the early stages of this porting process, I have plenty of questions and welcome additional thoughts.

Sorry about the late response, I've been tracking down endianness, packing and crypto issues in the comms layer of this porting job.

Re:Netbooks will make the ARM viable.

[mzs]

Thank you very much. It seems that your toolchain aligns structs in the natural way that X86, SPARC, and PPC do. That's great! I guess it assumes you have load and store word that work right. What toolchain and version are you using?

I had to port old code on another occasion that assumed everything other than char and unsigned char was aligned on 2-bytes, that was annoying as well. That's probably what you have on your hands as well. I feel your pain.

Again what I did is I compiled (this time everything since the PPC was SO much faster than the 8088) with that -fpack-struct option and added the xyz_padN members where needed. But first I needed to change all the int, unsigned, unsigned int, to int16_t and uint16_t, and similarly for other types. And then I had all the byte order assumptions to deal with, joy.

Again thanks for the info about natural alignment becoming the norm on newer ARM cores.

Re:Netbooks will make the ARM viable.

[lenski]

We're using the downloadable version of the Code Sourcery toolchain, to which we were directed by Atmel. FWIW, (IMHO) Code Sourcery has done a good job of collecting the often disparate components that make up a good toolchain.

I am operating under a basic assumption that converting from an old-school 8-bitter to the CortexM3 is guaranteed to take some careful thinking, planning and verification using various tools (particularly including objdump!)...

> /home/lcs/armdev_2009q3/bin/arm-none-eabi-gcc -v
Using built-in specs.
Target: arm-none-eabi
Configured with: /scratch/julian/2009q3-respin-eabi-lite/src/gcc-4.4/configure --build=i686-pc-linux-gnu --host=i686-pc-linux-gnu --target=arm-none-eabi --enable-threads --disable-libmudflap --disable-libssp --disable-libstdcxx-pch --enable-extra-sgxxlite-multilibs --with-gnu-as --with-gnu-ld --with-specs='%{O2:%{!fno-remove-local-statics: -fremove-local-statics}} %{O*:%{O|O0|O1|O2|Os:;:%{!fno-remove-local-statics: -fremove-local-statics}}}' --enable-languages=c,c++ --disable-shared --disable-lto --with-newlib --with-pkgversion='Sourcery G++ Lite 2009q3-68' --with-bugurl=https://support.codesourcery.com/GNUToolchain/ --disable-nls --prefix=/opt/codesourcery --with-headers=yes --with-sysroot=/opt/codesourcery/arm-none-eabi --with-build-sysroot=/scratch/julian/2009q3-respin-eabi-lite/install/arm-none-eabi --with-gmp=/scratch/julian/2009q3-respin-eabi-lite/obj/host-libs-2009q3-68-arm-none-eabi-i686-pc-linux-gnu/usr --with-mpfr=/scratch/julian/2009q3-respin-eabi-lite/obj/host-libs-2009q3-68-arm-none-eabi-i686-pc-linux-gnu/usr --with-ppl=/scratch/julian/2009q3-respin-eabi-lite/obj/host-libs-2009q3-68-arm-none-eabi-i686-pc-linux-gnu/usr --with-host-libstdcxx='-static-libgcc -Wl,-Bstatic,-lstdc++,-Bdynamic -lm' --with-cloog=/scratch/julian/2009q3-respin-eabi-lite/obj/host-libs-2009q3-68-arm-none-eabi-i686-pc-linux-gnu/usr --disable-libgomp --enable-poison-system-directories --with-build-time-tools=/scratch/julian/2009q3-respin-eabi-lite/install/arm-none-eabi/bin --with-build-time-tools=/scratch/julian/2009q3-respin-eabi-lite/install/arm-none-eabi/bin
Thread model: single
gcc version 4.4.1 (Sourcery G++ Lite 2009q3-68)

Re:Netbooks will make the ARM viable.

[badkarmadayaccount]

Separate x86 transcoder chip. Not that much powersaving, but better than any software.

No always has been

[Darkness404]

In the face of this issue energy-efficient computing has become a hot topic

No. That has, and never was the case. The problem is A) Programs now take a lot more CPU usage (compare CPU usage for the same task in Office 2007 and Office 97) B) CPUs are designed primarily to be faster (needed because of point A) C) Battery technology isn't improving as rapidly as the rest of the components.

Look at the Poqet PC of the '80s, it had very aggressive power management which wouldn't work today. Computers have -always- tried to be energy-efficient in the portable sector. And quite honestly, its about the only sector that needs work on energy-efficiency to gain any benefit.

Re:No always has been

[gbjbaanb]

isn't that the point - energy efficient computing means software as well as hardware features.

So Office 2007 takes way more CPU and memory to run; all the intensive cloud and web based server apps take a lot of power to cope with peak demand, and so we all run out of energy and end up with brown outs and much more expensive electricity.

When that happens, maybe people will start to take notice and efficient, not bloated, software will become more fashionable. Microsoft will no doubt bring out D# and sell us all more servers and development tools all over again - see everyone wins :)

Re:No always has been

[DdJ]

This line of reasoning makes one wonder how much energy it'll take to edit documents with the new iPad versions of iWork, on an iPad with a keyboard attached. Honestly it's something I'm actually considering.

The best subnotebook-like device I ever used in terms of power management was my eMate. I would take that thing to a conference, and in one case actually a business trip to Europe, and not have to charge it until I got home (ie. I got days of use out of it before I had to recharge it). But nobody builds stuff like that anymore.

Re:No always has been

[farble1670]

Computers have -always- tried to be energy-efficient in the portable sector. And quite honestly, its about the only sector that needs work on energy-efficiency to gain any benefit.

that couldn't be further from the truth. energy costs are just going up. for households it's mildly important as they can usually sleep their computers when not in use. for businesses, energy efficiency ranges from very important to critical when they have massive server rooms full of tens of thousands of CPUs powered and busy 24x7.

moreover, for developing countries, it's again critical. while the L in OLPC stands for laptop and therefore technically qualifies as mobile, it's more about having a battery to deal with locales where the electric grid is often shut down either on purpose to save energy or inadvertently because of a poor / out of date infrastructure.

Re:No always has been

[Cajun Hell]

And quite honestly, its about the only sector that needs work on energy-efficiency to gain any benefit.

Google disagrees with you, in a really big way.

Also, anyone who has hooked up a Kill-A-Watt to their computer, and then calculated how much money per year they're spending on it, disagrees with you.

This one asshole spent an estimated half a million dollars (of someone else's money) on electricity (which is probably the main reason he really got in trouble), not counting the harder-to-measure increased electric bill for the air conditioning (he was doing this in Arizona).

Energy costs money. People care about money.

Re:No always has been

[evilviper]

Also, anyone who has hooked up a Kill-A-Watt to their computer, and then calculated how much money per year they're spending on it, disagrees with you.

I've had a Kill-A-Watt for many years. Time and time again I've thought about getting a new system, a smaller monitor, etc.

Every time I do, the payback spans out into 5+ years. Even here in California, a 100W system is using 3 cents of electricity if it's powered-on 24/7. If it idles lower it may be less than half that. If you put it into standby/suspend or power it off, expect it to be less than 1/10th of that...

Let's go crazy, though, and consider a monster 300W system, running 24/7 for no good reason... You're paying well under $33/year for electricity. How much is it going to cost to replace this system? If you could replace it with something that used NO ELECTRICITY AT ALL for $300USD, it would still take 10 years before you come out ahead. Those aren't good numbers.

As always, the most environmentally friendly thing you can do is use an existing device as long as possible. And when that's no longer tenable, buying the absolute cheapest computer/car/house you can get your hands on is the best way to go, even if it's not incredibly efficient...

Re:No always has been

[Darkness404]

When that happens,

-if- that happens. Look, we have a lot of oil, a lot of coal and a lot of uranium plus wind energy that can be utilized, more efficient hydro-electric designs and now we are -finally- building more nuclear power plants. Mix this in with the decreasing cost of solar cells and we are going to have -very- cheap energy soon.

Yes, coal and oil aren't endless, but they have a pretty big supply. Nuclear power is the most promising, and if Obama doesn't screw up the free market any more (not that we still have it in the US anymore...) we will end up with even cheaper energy than we have right now.

Re:No always has been

[gbjbaanb]

not so much, sure I think our stocks of fossil fuels will last for ages, but it'll get increasingly difficult to get it out of the ground and process it. Add a bit of profiteering from the energy companies (as if they'd do that!) and energy prices will continue to rise. Some people are talking of peak opil currently, some people think peak production occured a few years ago.

Uranium... well, most of it is bought and used abroad. In 2001 the EU determined that known Uranium supplies would last 42 years. That was before globla warming was so popular and Nuclear energy back on the table. A lot of the uranium resources are found in countries with a less-than-stable political situation. (Australia and Canada produce most though).

So, although renewable energy sources are becoming much more popular (possibly as governments look at the numbers for traditional production and think "OMG") they are still inefficient compared to the traditional methods.

I'm happy that electricity production will be fine, but there will be a bit of trouble as production fails to keep up with the ever-increasing demand, and prices will continue to rise.

Just a marketing prank!

[Anonymous Coward]

They just want to capitalize on this whole "green" movement, especially considering their product line coincides with it. If they waited any longer, they might not be able to pull this one off.
 
Market on, but don't make yourself foolish.

Re:where do you get your facts?

[WrongSizeGlass]

ARM = Anonymous Rude Moron ... at least your case ;-)

Noooo!!! Tell me it isn't so!

[Colin Smith]

The Singularity must come!

Real physical limits. Energy production. Getting rid of heat.

Course we could always go nuclear and resume the exponential increase in energy available per individual which has been driving progress for the last 100,000 years, and which stopped in the 1970s (it explains the no flying cars thing, and lack of moon habitats).

 

Re:Noooo!!! Tell me it isn't so!

[lobiusmoop]

There are 2 ways to increase energy available per individual... Personally I'm hoping we get the whole overpopulation issue resolved first (perferably the humane way), we're kinda screwed overall if we don't.

Re:Noooo!!! Tell me it isn't so!

[srussia]

There are 2 ways to increase energy available per individual...

Global warming kills 2 birds with 1 stone. More heat=more energy, catastrophic change=fewer individuals.

Ah, the ways of the FSM are mysterious indeed!

Re:Noooo!!! Tell me it isn't so!

[Yvan256]

FSM? Heretic! The IPU will get you and you won't even see her coming!

Re:Noooo!!! Tell me it isn't so!

[Anonymous Coward]

Flying cars would be noisy. But you have a point in the fact that most people still can't afford to own a decent private airplane or even to own a part in one.

The energy costs of launching a rocket into translunar orbit (or any orbit) are neglible compared to the labor costs of employing thousands of engineers, scientists and bean counters to be able to get the thing on the launchpad.

Now, if we could just invent an AI that could replace all of those. Or at least one smart enough to replace the bean counters...

We can't do that with today's understanding, even given an infinite array of CPUs and infinite energy to run them at full speed.

Re:so why can't i buy a !@##$% low powered compute

[fuzzyfuzzyfungus]

http://openinkpot.org/wiki/Hardware

There are a bunch of cheap 'n cheerful Kindle clones that are supported by the openinkpot firmware, if e-ink is your style.

If you need the refresh rates of an LCD, I think you may have a problem. There is the OpenMoko Wikireader which is super low power and pretty cheap; but the screen(although touch sensitive) is a little smaller than what you want and the CPU is more of a microcontroller than what you probably want.

You might also consider going the OLPC route. That gets you a rather nice monochrome screen(with backlight color optional) and, while x86, it is a ~1watt embedded Geode, not too power hungry. That one is also by far the most "normal computer" like of the bunch(i.e. no having to code around super slow refresh rates, and running basically standard linux, rather than building binaries for some microcontroller).

Nothing exactly like what you want(that I know of); but there are some approaches...

RISC = Good

[WrongSizeGlass]

I've always been a big fan of RISC. I even have a copy of ARM System Architecture (c) 1996 by one Mr Steve Furber I pulled out of the basement of Strand Books quite a while ago.

Re:RISC = Good

[Michael Kristopeit]

i like MIPS... in college, we had one semester of hands on lab work learning everything about a specific MIPS implementation, then another semester writing a compiler capable of compiling itself for the architecture.

once you can grasp the simplicity and understand exactly whats going on in the chip, features like HyperThreading seem almost stupid because of how much complexity and exceptions they add to the system.

Re:RISC = Good

[WrongSizeGlass]

I spent years working with PowerPC ... and then took it in the backside when Apple switched camps to Intel x86. With the resurgence of RISC in the mobile market I may be able to use some of my RISC experience.

I still think RISC is the better all around architecture approach. We used to play around with 'chip simulators' by using the plentiful registers in the RISC chips to hold the registers of the smaller RISC chips we were simulating. Great for VM's, too. Ahh, those were the days.

Re:RISC = Good

[Yvan256]

I'm a big fan of arm. I have two of those!

Re:RISC = Good

[emt377]

i like MIPS... in college, we had one semester of hands on lab work learning everything about a specific MIPS implementation, then another semester writing a compiler capable of compiling itself for the architecture.

once you can grasp the simplicity and understand exactly whats going on in the chip, features like HyperThreading seem almost stupid because of how much complexity and exceptions they add to the system.

Of course modern MIPS processors are threaded as well...

Re:RISC = Good

[Michael Kristopeit]

true... modern cars run on electric batteries... it doesn't mean diesel engines aren't the best choice for moving a semi trailer of goods cross country.

my main problem with the threaded approach is the overhead to allow real-time applications such as audio recording to run in a pool of other processes using the same core. a lot of predictability is lost when the pipeline flow is variable... the overhead of regaining a similar level of predictability wipes out most of the perceived gains, and lowers maximum achievable quality.

it's all application specific obviously, but i believe multiple MIPS cores can always do any job in any environment whereas a single threaded MIPS core is sometimes not capable of doing a specific job...

Re:RISC = Good

[imgod2u]

SMT is probably one of the best architectural features you can get in terms of complexity of implementation vs speedup. It isn't trivial by any means, but far less hazardous (get it?) than an OoO back-end or speculative execution.

Re:so why can't i buy a !@##$% low powered compute

[Darkness404]

Because there is almost no market for it. Lets see here

A) Monochrome displays (other than E-ink) are generally considered to be low-tech, cheap, pieces of junk. Look at how well monochrome TVs are selling... Yes, it does ease strain on eyes and increase readability, but has the other side of making most of... well anything unpleasant to look at other than text.

B) It will be expensive. Amazon can afford to produce the Kindle at-cost or even with a bit of a loss because they will gain sales in e-books and such. Even if they sell it at a profit they still can buy parts in bulk and make them cheaper than a product with a run of only ~1,000 units or less.

C) There aren't enough apps. What apps would make sense to port to this device? Lynx? Most other things would need more CPU power (making it non-energy efficient) or a colour screen.

Really, other than you, this wouldn't appeal to a large enough audience of people. Best thing to do would be to jailbreak an existing e-reader, its simply a bad business decision to make a product with almost no market.

The funny part is newer computers are more

[NotSoHeavyD3]

efficient. I mean if you consider any unit of computation vs energy expended. I bet my current desktop computer would compare from a computation point of view to a super computer from the late 80's. (GFLOP to GFLOP) However my current computer pulls about 300W, I'm pretty sure that's alot better than any super computer from the 80's that would compare to it.

Re:The funny part is newer computers are more

[Michael Kristopeit]

my mac mini only pulls 14W and isn't far behind your current desktop in performance. my fit-pc2 only pulls 6W. relative to power consumption, both of those machines are easily besting super computers from the 80s as well as your current desktop.

Re:The funny part is newer computers are more

[WrongSizeGlass]

efficient. I mean if you consider any unit of computation vs energy expended. I bet my current desktop computer would compare from a computation point of view to a super computer from the late 80's. (GFLOP to GFLOP) However my current computer pulls about 300W, I'm pretty sure that's alot better than any super computer from the 80's that would compare to it.

300W? I'm pretty sure a coffee maker from the 80's was even that efficient.

Re:The funny part is newer computers are more

[BikeHelmet]

My Athlon II X2 + 8800GS pulled 95w from the wall, according to my Kill-A-Watt.

Now I have a Phenom II X4 and GTS 250. It pulls about 107w from the wall when idle, and as much as 160w when gaming. (~175w when encoding video)

The funny thing is, I've seen old P4 computers drawing 250w+.

Re:The funny part is newer computers are more

[mirix]

And late model PIII's, (which were smoking the early P4's) only sucked 30-40W at the die. Netburst was just a big fail in general. A slow flamethrower.

Re:The funny part is newer computers are more

[Anonymous Coward]

now let's see stuffing up those energy-efficient PCs into racks which supercomputers were in '80s.

Re:The funny part is newer computers are more

[Anonymous Coward]

Coffee machines have to heat water up, an operation that actually uses power. I'd be surprised if any coffee machine could work with 300 watts.

Re:The funny part is newer computers are more

[ncc74656]

The funny thing is, I've seen old P4 computers drawing 250w+.

They probably had less-efficient power supplies than your newer machines. Power-supply efficiency only seems to have become a selling point in the past couple or three years.

Indeed

[Colin Smith]

The phone market has always been looking for ways of extending battery life. I have a phone which is basically a computer with an antenna. It plays videos, music has wordprocessor, gps, maps etc and the battery still lasts up to 3 days.

 

Re:Indeed

[colonelquesadilla]

The phone market has always been looking for ways of extending battery life. I have a phone which is basically a computer with an antenna. It plays videos, music has wordprocessor, gps, maps etc and the battery still lasts up to 3 days.

Sure, but not while it's doing those things continuously. Also people underestimate the effect of the antenna. A cell phone radio puts out a fair amount of power, and due to design limitations, the antenna can't be exactly impedence matched. We had a guy from TI in talking about RF switching and phase shifting, and the cell phone manufactureres are looking into new technology in order to increase the efficiency of the radio to increase battery life.

Begging the Question

[bill_mcgonigle]

while our primary sources of energy are running out

Just cleaning up our light-water reactor waste (which we cannot leave around for 300,000 years) can power the Earth's advancing societies for a century.

There are much better reasons to go for low-power computing, portability and economics chief among them.

Re:where do you get your facts?

[bill_mcgonigle]

Your credibility as a reference was lost when you fail to expand the ARM chip acronym correctly.

Hey, on Slashdot old school is acceptable.

Evident?

[fm6]

By now, it has become evident that we are facing an energy problem -- while our primary sources of energy are running out, the demand for energy is greatly increasing.

Evident to you and me, maybe. But there are lots of folks who insist that all these issues can be explained away. A lot of them follow Slashdot, and I'm a little surprised they haven't already chimed in.

(Forgive the double post. Should have previewed.)

Re:so why can't i buy a !@##$% low powered compute

[biryokumaru]

The only OLPC laptops I can find for sale now are on eBay, from people who would willingly profit from a charity. Is there a better source?

You can. But apparently are unable to use Google

[Colin Smith]

http://en.wikipedia.org/wiki/Comparison_of_e-book_readers
 

Connectivity and standards are the solution

[jhfry]

If Internet connectivity were ubiquitous and cheap and proper standards were developed and encouraged, we would see a tremendous improvement in efficiency.
1. It takes a very minimal amount of power to use hosted applications, so the end users devices would be low power.
2. Data centers have serious incentives to be efficient, when your annual electric bill is in the $100,000+ range, even a 1% improvement is worth considering, when was the last time you cared about saving 1% on your electric bill.

The WWW is getting us on the right track, but what we really need is to develop a new Internet protocol for hosted applications. I see little reason that we need to continue to try and add complexity to the WWW, HTML was never really intended for Web 2.0+ apps. If this new protocol were properly designed, and very open, and had strictly enforced standards, hardware could be made to accelerate its more power hunger aspects (sound, video, 3d, etc.). This would result in very low powered components that do one thing very well, coupled with a very low powered cpu you could have a full featured machine that consumes minimal power.

Re:Connectivity and standards are the solution

[badkarmadayaccount]

WebGL, audio and video tags. What the hell is wrong with HTML? Perfectly sound way to describe UIs.

Bull...

[Wyatt Earp]

"By now, it has become evident that we are facing an energy problem — while our primary sources of energy are running out,"

No, a primary easy source is running out, rock oil, we have like 40-80 years of that left.
http://en.wikipedia.org/wiki/Petroleum#Future_of_petroleum_production

Shale oil, well theres a ton of that out there, if the world wanted to, they'd be able to access that. The US has about 1,750,000,000,000 barrels
http://en.wikipedia.org/wiki/Petroleum#Consumption_statistics
http://en.wikipedia.org/wiki/Oil_shale_reserves#Definition_of_reserves

And there is a ton of natural gas, and shale gas and coal, not to mention fission, solar, and/or wind.

So, no we are not running out of our primary sources of energy.

Re:Bull...

[Colin Smith]

Look up "Energy Return on Energy Invested".

Saudi oil has been 100:1.

Shale... 5:1 maybe, 3:1.
http://en.wikipedia.org/wiki/Oil_shale

When they really start using shale, you know the shit is really hitting the fan.

And no matter how much is left (quoted in the reserve figures as recoverable), could be a trillion trillion barrels, nobody is going to bother trying to get it out when it takes a unit of energy in to get a unit of energy out.

Re:Bull...

[Wyatt Earp]

5:1 - 3:1 is still a net profit.

Re:Bull...

[maeka]

5:1 - 3:1 is still a net profit.

Only if you're assuming energy cost is the primary cost.

Re:Bull...

[micahfk]

Wanna make a $10,000 bet on the price?

http://en.wikipedia.org/wiki/Wager_between_Julian_Simon_and_Paul_Ehrlich
http://www.nytimes.com/2005/08/23/opinion/23tierney.html?_r=1&n=Top%2FOpinion%2FEditorials%20and%20Op-Ed%2FOp-Ed%2FColumnists%2FJohn%20Tierney&oref=login

In case you want an excuse to back out.

I doubt it

[Colin Smith]

If the unit of computation is to put a single character on the screen for example. Today, it requires several supercomputer class processors to do the same job as one 286 during the 80s.
 

Re:I doubt it

[Amouth]

but now we have cuved edges and smooth colors - and wait.. still using a square dot to show it.. crap

Re:so why can't i buy a !@##$% low powered compute

[Arthur Grumbine]

Notion Ink has you covered if you can wait until June for their Adam. It's basically everything you just described, times 9000. And pricing in the bargain laptop range ($350 - $800). I'm not kidding, check it out.

The Kindle's not so bad for this, actually

[weston]

It's already running Linux out of the box, and the hacks getting everything a unix geek might want on it really don't sound like they're about bypassing DRM so much as they are getting tools onto the system that Amazon just left out.

Static link cross compile a telnetd and toolchain and get 'em both on there and you're set to go.

The only reason I haven't bought in on that action yet is that as far as I know there's no decent third party full size portable keyboard. If it did bluetooth, I'd be totally sold. As it is, I'm almost sold.

Re:The Kindle's not so bad for this, actually

[mirix]

I assume the ARM on the kindle has a serial port, or at the very least some GPIOs (so you can bitbang a serial port, or I2C). add a $2 MCU to convert a PS/2 keyboard to serial or I2C, and you're in business.

You could have the MCU in the keyboard, and use a 2.5mm headphone jack or so to plug the keyboard into the kindle.

Also, there are very small uart to bluetooth boards in the $15-20 range.. a pair of them work like a bridge... so:
RS232 - bluetooth - RS232; bidir.

Slap one of those in your kindle, another in your keyboard w/mcu, and you're in business wirelessly.

maybe I'm missing something though?

Re:so why can't i buy a !@##$% low powered compute

[niko9]

After reading Joseph Jenkin's excellent book, The Humanure Handbook (http://www.jenkinspublishing.com/humanure.html and available for free here: http://humanurehandbook.com/downloads/Humanure_Handbook_all.pdf), I've been doing a lot of thinking recently about not just human waste but the waste that comes from everyday computing.

I think what you and I want is the combination of a Beagle Board and a Pixel Qi display (http://www.pixelqi.com/). That display has been mentioned on Slashdot before. You could also start developing super efficient programs on something like a PC Engines ALIX board (http://www.pcengines.ch/alix3d3.htm) or a Technologic Systems ARM based board: http://www.embeddedarm.com/products/arm-sbc.php#ts-7800-series The TX boards come with Debian pre-installed and boot up in less that2-3 seconds. They seem to be very hacker friendly.

Of course, the other side of the equation is the WWW that we connect to. Is there anyone giving any consideration to efficient computing on the server side? Web forums vs. mailing lists; fancy web sites vs. lean, mean, and candid web design? Are there resources online where I can learn how to design a lean, mean, yet somewhat modern looking web site and run it off something like the aforementioned ALIX (500 Mhz AMD Geode) board? Does HTML help at all in this regard?

I even miss the old but lean Slashdot of long ago.

ACORN Risc Machine

[Anonymous Coward]

I want NOTHING to do with a processor that fraudulently got Obama elected! We haven't even seen his BIRTH CERTIFICATE! I CAN'T BELIEV....

Wait .. they have nothing to do with each other?

Nevermind, carry on.

Most People Don't Care About Most Apps

[weston]

And you also mean the porting of thousands and thousands of x86 apps as well?

I suppose there's probably a market for that; there's certainly some subset of the population that's attached to specific desktop applications.

But I'd bet a larger subset of the market just wants to write documents, send/receive email, and browse the web. Ubuntu or some equivalently friendly Linux distribution will do the job nicely there.

Make Boots & Scans Faster

[CohibaVancouver]

One of the reasons my PC is a power hog is I leave it on all the time. Why?

1) Because boots take a long time. If boots were near-instant like they were on my TRS-80 in 1979 I'd turn it off. (Yeah, yeah, Macs boot fast, yeah yeah, I can 'suspend' but none of that junk ever works properly on WinTel.)

2) Because backups, patches and scans run at night time. If I didn't need those, I'd turn it off.

Figure out a solution there and I'd turn my box off.

Re:Make Boots & Scans Faster

[farble1670]

I can 'suspend' but none of that junk ever works properly on WinTel

every computer i've owned or used in the last 10 years has been able to hibernate or sleep. that includes macs, linux and win98 to win 7. if you buy a computer that can't reliably sleep, you should return it and get your $ back.

Re:Make Boots & Scans Faster

[h4rr4r]

Hard to run cron jobs while the machine is sleeping.

Re:Make Boots & Scans Faster

[Blakey Rat]

yeah yeah, I can 'suspend' but none of that junk ever works properly on WinTel.

Yah it does. Unless your computer's defective. If it's defective, take it back and get one that works.

Seriously, suspend/hibernate is a solved problem for the last decade. When's the last time you tried it? Windows 98?

Re:Make Boots & Scans Faster

[CohibaVancouver]

When's the last time you tried it?

Last week.

My (c) 2005 gateway (XP) has no audio when it comes back from sleep. My Dell Inspiron PIII has a Linksys WiFi card in the PCMCIA slot. The card dies when it goes to sleep and only comes back with a reboot.

Re:Make Boots & Scans Faster

[mirix]

How many bytes was the OS on a TRS-80?

You can't really compare that...

Re:Make Boots & Scans Faster

[CohibaVancouver]

How many bytes was the OS on a TRS-80?

The computer 'booted' to a BASIC interpreter at the command line:

READY
>_

Re:Make Boots & Scans Faster

[Anonymous Coward]

It takes an average BIOS engineer few tries to get the power management stuff right these days, apparently. Perhaps a BIOS update would solve the issue. Making sure to install few quite important patches for older Windows systems would also significantly lessen the possibility of losing data while going to sleep.

Re:Make Boots & Scans Faster

[mirix]

And the whole thing probably fit in an 8k ROM or something?

that explains the speed :)
Now... you need a big enough ROM to stuff a modern OS in!

Energy running out

[coldmist]

while our primary sources of energy are running out

And in the 1920s, they claimed we were running out of oil. In the 1970s, they claimed we were running out of oil. Just last year they found a new oilfield off of Brazil bigger than anything found yet. Last year. After everyone said no new large fields would ever be found.

Coal? Clinton locked up the Grand Staircase in Utah, the largest clean coal deposit, with 62 Billion tons of coal.

I don't know. I hate scare-mongering that has been going on already for 100 years, and shown wrong for 100 years, and the next generation doesn't see how poorly it looks.

Re:Energy running out

[Eharley]

We're not likely to run out soon, it's just going to get expensive.

Re:Energy running out

[h4rr4r]

On the first I agree, but there is no such thing as clean coal. That is just a marketing stunt.

Re:Energy running out

[emt377]

Just last year they found a new oilfield off of Brazil bigger than anything found yet. Last year. After everyone said no new large fields would ever be found.

The Tupi field is estimated to hold 8b barrels of oil. Given our current global consumption that's a three month extension. It's the biggest field discovered in 30 years - which is pretty telling. Find ten of these and we've got a few extra years. Find only another one or two and it makes no difference. Meanwhile, when the global business cycle points up again our oil consumption is going to follow likewise - again. Prices will rocket, and economic growth will be choked. Oil is really a limited resource and the way we've built our entire economy around it is going to limit our capacity for global growth.

Re:Energy running out

[Anonymous Coward]

We still boomed at Oil @ $200 per barrel tho (2007)... I suspect that in the end, wishy washy will win it.

Re:Energy running out

[Anonymous Coward]

thing is we wouldn't have to be worried about oil now because we had an incredibly powerful energy source, possibly the most concentrated amount of energy and it's processes are so simple school children know how it works, fission. but the flower children and the neocons screwed the pooch. Now we are having to deal with a screwed up and ancient power infrastructure and why is that? because it was much more profitible to rape the natural landscape and destabilize the heart of western civilization then to use modern technology. Wankers.

Re:Energy running out

[h4rr4r]

Fun fact for you mr retard, those "dot heads" are Caucasian just like you.

Re:Energy running out

[colonelquesadilla]

dude... seriously... wtf. But even if we all decided to be hateful racist self entitled assholes there are huge holes in the plan. How do you propose to murder most of the population of the planet? Nukes? Think it's still gonna be nice to live here afterwards? So say you can do it with conventional weapons. Returning everywhere but western europe and north america to "natural habitats" wouldn't solve jack shit, you just end up with fewer resources and fewer people. doesn't much matter though, I doubt there is any point in arguing about this with you since you clearly stated your goal was not to solve some sort of coming resource crisis but to kill all the s and are just using this as a convenient opening for it.

Re:Energy running out

[Anonymous Coward]

No shit. We are the saudi arabia of coal (we even have a muslim HNIC). World War II was won thanks to diesel and gasoline that came from coal (coal liquefaction). Of course, with current technology, we're the saudi arabia of oil, too.

Re:Energy running out

[stonewallred]

Fact for you,they are light skinned niggers. They do not look like me or my people. They do not talk like me or my people. And therefore they are animals to be killed or domesticated as I see fit. Of course you have been conditioned to hate yourself and your white skin, taught by a evil education system and ignorant tools to believe that to seek domination over the monkeys is somehow wrong or morally evil. And the lands could be turned back to a mostly natural state, free from the hordes of sub-humans masquerading as real humans. The impact from drilling and mining would be minimal compared to the damage the apes are doing today by just being there. I say kill them all because the vast majority are too stupid to be good slaves. And bio-weapons, a few nukes against major structural targets like damns, cities and power plants, and then mopping up for a few years as the apes quickly devolved back to their natural state devoid of the artificial veneer of civilization we have given them. Within ten years of the start, we would able to book safaris over there and hunt the few remnants as sport.And we, as true humans, would be set for the advancement of our race without the baggage of stupid animals holding us back by false and evil ideas of equality.

Re:Energy running out

[bendodge]

Even if it does, we don't need the government to save us preemptively from expense (be it healthcare or energy). In both cases reducing red tape is 80% of the solution. I'm tired of being rescued. Just get out of the way.

(Please realize that this post is just a personal rant and has little do do with the parent. He just offered a nice jumping point.)

Re:Energy running out

[h4rr4r]

Either that is the best racist troll yet, or Crazytown has a new mayor.

Re:Energy running out

[k8to]

True, but this reflects lower availability, which for a commodity like energy is a pretty important thing!

Re:so why can't i buy a !@##$% low powered compute

[Skal Tura]

Kindle might have very low computational power, but actually there are some quite powerfull, and can do more than just lynx.

Apps? Just look at where you got lynx from ... And you won't have a shortage

More than just lynx etc. very low computationally intensive? Just look at the chinese netbooks ...

And there you got your actual products aswell

Re:so why can't i buy a !@##$% low powered compute

[fuzzyfuzzyfungus]

I don't know, though I wouldn't necessarily be too worried about the ebay units. There might be some international cartel snapping them up by the truckload from poor kids, swapping whatever oddball keyboard the locale required for a custom manufactured EN_US one, and selling them on Ebay; but that seems like a lot of trouble for a not-necessarily-all-that-popular product.

I'd tend to suspect, especially if you buy from a seller with a history of other geek junk buying and selling but no major quantity of OLPCs, that you would end up buying an original G1G1 unit from some geek early adopter who has since upgraded to a newer netbook. My casual ebay search suggests that such people are getting anything from ~$100 to just over $200 for their units. Not exactly impressive cash for a device that sold for $200+ a $200 donation originally. It's just geeks selling off toys they've gotten tired of, probably so they can buy new ones.

It's a limited supply, so you had better not build any long term plans around it; but I don't see any ethical objections...

Great job marketing

[Anonymous Coward]

that is simple, low power, and low cost

ARM may have a lot of coolness going for them right now since they are taking on big bad boy Intel. However ARM is certainly not low cost (ask anyone who has bought a source license or that pays royalties).

Re:Great job marketing

[mirix]

And how much do you have to pay to make a modern x86 chip under licence?

The physical chips are definitely cheaper than anything x86 with similar speed.

Re:Bull...you are not even counting coal

[codepunk]

We have over 250 years worth of easily available coal here in the US.

bar

[Anonymous Coward]

Steve Foobar? Maybe its time to hit the bed.

Re:so why can't i buy a !@##$% low powered compute

[maxume]

Everyone running a data center is giving thought to efficiency. They pay for electricity and cooling.

Re:so why can't i buy a !@##$% low powered compute

[grumpyman]

C) There aren't enough apps. What apps would make sense to port to this device? Lynx? Most other things would need more CPU power (making it non-energy efficient) or a colour screen.

It's way more than Lynx. There's elm, gopher, nn and with a built-in modem, you should be able to connect to BBS too.

Re:where do you get your facts?

[BestNicksRTaken]

It was Acorn RISC Machine way before it became Advanced RISC Machines Ltd; by almost a decade, when Furber ran the show.

Re:Bull...you are not even counting coal

[h4rr4r]

Nasty, dirty shitty coal. Coal power should just be illegal already.

Nuke, wind, solar, natural gas all are alternatives with far less pollution and co2 release.

ARM is viable right now, except for some drivers

[Anonymous Coward]

Fuck Windows. All we need are graphics drivers that we have the source code for. We have the source code to everything we run these days, and most of it has been made portable and compiled for and tested on ARM. Except graphics drivers. That one thing is holding us back.

Re:where do you get your facts?

[TheRaven64]

Probably from a reliable source. The chip that he designed was the Acorn RISC Machine. When ARM was spun out as a joint venture with Apple, it was renamed. Advanced RISC Machines is a backronym intended to keep the same initials but remove the Acorn branding (which Apple didn't want).

Re:so why can't i buy a !@##$% low powered compute

[DarkOx]

I not so sure, it was not all that long ago most of us who even had a computer were working on an 80x86 or 80x88 with a 512k of memory, connected to a 80x25 character monocrome display, with no network interface, and primary storage consisting of a box next to the unit filled with 360k floppy disks.

You know what we were like pigs in s**t happy too. I have cell phone, not a facy smart phone. I just use it I don't think about it. I charge it Sunday night before I go to bed and I talk and text on it as much as I want to all week long. The batter never goes flat over that time frame.

I had a smart phone for a little while. It could not even stand by all day. I had to go back because I was afraid I might miss SMS alerts from critial systems because the things battery went flat.

I would be very happy with a device that was inexpensive enough to leave on the back seat of my car for weeks on end that would be just there if I needed it and ready to go. Something that could connect to the internet wireless-ly and run lynx and those kinda of apps would be just perfect, even if the screen refresh was slow an e-ink based. Things like e-mail would be very doable as well. We don't live in the same world that 80x86 lived in. There is all kind of infrastructure around, let the POP or IMAP server do the thinking, just add a command like TXTPLZ to the protocols that would instruct the server to render messages sent in other formats out as plain characters. Get send html no problem
  becomes 13 and 10. Hell use libcacca to render images as ascii art.

The software you need to write and run on the device could do allot by simply offloading the thinking to the oh lord dare I say it? cloud...

Wierd priorities

[Areyoukiddingme]

My computer? Really? THAT'S the power hungry device I should be worried about? What about, oooh, I don't know, how about my electric clothes dryer with the 240V plug and the massive double circuit breaker in my breaker box?!

Priorities people, priorities.

Re:Wierd priorities

[BeaverCleaver]

How many hours a day is your computer switched on? How often do you use your dryer? This IS slashdot.

Re:so why can't i buy a !@##$% low powered compute

[commodore64_love]

How about just giving me Windows 95, which could run on only 16 megabytes RAM and a 33 megahertz PC. On a modern Atom-equipped PC such an efficient/sleek OS would use so little power your battery could go for days-and-days.

Try a "graphing" calculator

[irtza]

Here are some cheap platforms that are robust, programmable and have a fairly substantial geek appeal that may meet your very needs: TI-89, TI-92, TI-86 HP48gx I haven't used my TI-89 or TI-85 for over five years, but I turned on the calc about a month or so ago and it runs beutifully. I heard casio makes some as well, but never used one of those. PS: I love putting things in "quotes"

Fearmongering.

[Hurricane78]

Set up a couple of solar thermal power plants (e.g. in Arizona), lay a couple of high-voltage DC lines, or convert it to hydrogen, or any form or battery, and be good. We will have left this planet, long before we use more energy than the sun can deliver (especially when you add space-based power plants). And the technology is cheap, simple (a poor African nation could do it without having to having to take a loan), recyclable, and there is a lot of really dead and hot land out there (certainly deader than what our power plants stand on nowadays).

They just don’t want to set them up, so they can keep their power.

Re:Fearmongering.

[Pinky's Brain]

For less than the cost of the financial stimulus package, or the Iraq war for that matter, the US could produce almost all it's electricity with solar thermal plants with present level technology (and the cost for plants would probably be quartered by the time you are done because of economies of scale, so it would cost far less). Hugely reducing electricity costs in the US would probably do more for the economy than just about anything else the money could be spend on.

As a European I'm envious ... the US really has it all, virtual dead deserts with round the year sunlight, a reserve currency which gives you nearly limitless free money to spend on these kinds of projects, and hell quite a nice supply of oil reserves as well ... it's frankly a miracle how your politicians manage to fuck that kind of potential up.

Re:Fearmongering.

[Anonymous Coward]

> it's frankly a miracle how your politicians manage to fuck that kind of potential up.

The US has the best government money can buy.

Re:Fearmongering.

[Neoprofin]

It's a government by the people...

Har har.

Re:Fearmongering.

[Anonymous Coward]

Except the USA consumption is a bad model, it far outstrips supply & the dollar is dying as the reserve currency.

Re:Fearmongering.

[stonewallred]

We could just round up all the sub-human niggers and slants, along with the sand niggers and niggers with dots on their heads and put them in huge treadmills to produce our electricity. Get rid of all the diesel powered tractors and big trucks, and hook them monkeys into harnesses and have them pull the plows and trailers. Plus keep them in pens, they don't need lights or TVs and radios. And we'd save tons on prisons. Plus we would get rid of welfare. Just feed them, breed them and when they get old, sick or hurt, shoot them. Make the world a better place, enslave a negro today.

Re:Fearmongering.

[Anonymous Coward]

You still need to cover peak load in the evenings when the sun is down and the winds might not be blowing. Having enough nuclear capacity to run them for the daily peaks might be very expensive.

Some method of storing the energy is needed.

If we had efficient electrolysis to produce H2, we might be able to cover peak demand and also develop a distribution infrastructure for H2 for vehicles.

Another option is fuel cells to store the energy while the sun shines and the wind blows - I've read about vanadium fuel cells that should be quite safe, if you could build them big enough and cheaply enough.

Re:Fearmongering.

[Anonymous Coward]

It isn't a miracle; it's an everyday(my whole lifelong) event. If they didn't fuck it up they would not know what to do next.

Re:so why can't i buy a !@##$% low powered compute

[hedwards]

I'm confused, you want to buy one, but the seller shouldn't profit. Last I checked the OLPC laptops were pretty much only being sold to schools or donated. For a while you could buy two and donate one if you wanted to have one. But right now you pretty much have to buy one via ebay if you want to have one for yourself.

It's just marketing

[Sycraft-fu]

Aka lying.

While it is true that according to best estimates oil is going to become rather scarce in the not too distant future, oil isn't a major provider of power. It is used for power generation, but not near as much as others. Coal and nuclear are the heavy hitters in power generation and both are still extremely abundant. So while we need to consider what will fuel our cars, we are not approaching a situation where the lights go out.

He's just trying to jump on the green bandwagon to hype up his product. ARM has made a name for themselves providing low power chips. That's why they are so popular with cell phones and the like. He's just playing on that and the big carbon hype going on to try and promote his products.

As a practical matter, while we do want to find ways to be more efficient always, we still want to use energy. Energy consumption allows for the things that make modern life so much better, allow us to live longer, healthier, etc. What we really want to do is increase efficiency AND find ways to generate even more power in a sustainable manner, be it fusion, solar, geothermal, etc.

Re:It's just marketing

[k8to]

Uh, right.

Because, as abundant chemical energy to power transportation becomes scarce, there will be no pressure to use electricity for transportation needs. No, none at all......

But back in the real world, these two forms of energy can be converted one to the other, and are, and a scarcity of energy of one type does indicate an overall energy scarcity, especially when you look at the energy budget of transportation as compared to the current power grid.

Cloud computing is energy efficient

[cryfreedomlove]

Isn't cloud computing a better angle on energy efficient computing than trying to optimize the power usage of any one computer? Most computers are very under utilized and suck up power doing nothing. Cloud based computers get allocated to the next user when they fall idle so you get more useful computation per watt of energy burned. That seems smarter than going after energy efficiency for any one computer and then letting it burn that smaller amount of energy while sitting, for the most part, idle.

Re:so why can't i buy a !@##$% low powered compute

[h4rr4r]

How about using a lightwieght linux distro? You can have a real OS and run it on similar hardware.

Also no, the battery would not go on for days, the cpu would need to be something more efficient than the atom and lighting the screen is probably a much bigger draw.

Re:so why can't i buy a !@##$% low powered compute

[fuzzyfuzzyfungus]

I can't tell whether the grandparent comment is referring to "Is there anyone giving any consideration to efficient computing on the server side?" in the sense of "Is anybody working on answering my HTTP requests at lower energy cost?"(in which case the answer is an unequivocal 'Fuck yeah, they're paying the power bills') or in the sense of "Is anybody working on websites/web services that will be less computationally expensive for the clients?"

If it is this latter case, the answer is "mostly no, just look at average page sizes and amount of complex flash and JS rocketing up over time"; but there are isolated efforts here and there. Opera's mini browser for seriously weak phones basically depends on having an Opera proxy pre-chew the content before shoving it down the pipe, and I'm pretty sure that RIM does similar digesting to email.

Re:so why can't i buy a !@##$% low powered compute

[maxume]

I got more of a green vibe than I got a battery life vibe.

Re:so why can't i buy a !@##$% low powered compute

[Anonymous Coward]

NVIDIA? Sorry but that breaks the deal for me. It might not matter for such a toy, but I refuse to give such a retarded company any of my money. I would rather buy completely unaccelerated video.

Re:Too conservative.

[Anonymous Coward]

You can compare your desktop against a supercomputer from 1995 and come out on top.

E.g., a Cray-3 from 1995 had four CPUs @ 500 MHz. Each CPU could execute ~ two flops/clock,
for a total of 4 GFLOPS. Power draw was about 40kW, not counting coolant pumps, MG
inefficiencies, &c. A Y-MP/16 of that era probably about doubled that (more CPUs but
slower clocks) at twice the power draw.

A top-end laptop probably has about that much horsepower now.

Re:where do you get your facts?

[SpinyNorman]

I worked at Acorn in the early 80's and knew Steve Furber, but it's amusing to think that most of the Slashdot crowd probably wasn't even born then. I assume the average age here nowadays must be college age or thereabouts.

Not surprising they don't know what ARM originally stood for.

Re:so why can't i buy a !@##$% low powered compute

[marcansoft]

Windows 95 is way, way worse than any modern OS, even Vista and Windows 7. Back in the '95 days OSes didn't put the CPU into idle mode and certainly didn't support any power management, effectively running the CPU at full throttle 100% of the time.

Sorry, you made the erroneous assumption that low system requirements equals low power. Try again.

Re:so why can't i buy a !@##$% low powered compute

[Miseph]

And you can play those new, high-tech Multi-User Dungeon games, too!

Re:so why can't i buy a !@##$% low powered compute

[bcrowell]

Amazon can afford to produce the Kindle at-cost or even with a bit of a loss because they will gain sales in e-books and such.

Amazon loses money on the e-books they sell. They pay about $14 wholesale for the digital version of a newly released hardcover, and only charge the customer about $10.

problem is

[codepunk]

None of them are as cheap as coal, if you had to rely on the alternatives you would not be able to afford to heat or cool your home.

Re:problem is

[compro01]

Depends on your precise definition of "cheap". Coal is cheap because, barring other forces like regulation, it can externalize most of the costs. No need to pay much for waste storage when you can pitch it into the air or bury it with common trash.

Snark? or do you think you're for real?

[lenski]

It's a comfortably snide comment but it does not reflect the decades-long reality of the ARM architecture.

The ARM guys started out in a performance/watt sweet spot from the beginning and they continue to get it right today.

Re:Snark? or do you think you're for real?

[Anonymous Coward]

The point was toward their (new) marketing side. Great architectures and all, but don't associate it with the BS fearmongering of late.

Re:so why can't i buy a !@##$% low powered compute

[Anonymous Coward]

Isn't the core CPU portion of the Tegra chip ARM based? I'm not sure I understand your comment.

Re:so why can't i buy a !@##$% low powered compute

[xpurple]

I want something that is low power, has a nice keyboard and runs linux. Text only mode is fine as all I really want to do is run emacs.

Re:so why can't i buy a !@##$% low powered compute

[b4upoo]

I'm not so certain that a decent,monochrome, computer running Linux would not sell well.
                  There is a universe of high quality books at Project Gutenberg that I often enjoy and it is a shame to use a high powered PC just to read and download books. And these days a decent chess game can take place on a low powered PC as well. From my point of view there are now so many uses for computers that having numerous units for various tasks seems quite reasonable and we could actually save energy as well if we use the correct unit for most jobs. Why not ,for example, have a small PC in the kitchen that keeps track of your grocery shopping list and relates those needs to menus for the week? Other small computers might be applied to individual hobbies such as keeping your golfing history and golf contacts and schedules as its only purpose.

ohai

[Anonymous Coward]

*than Steve Furber

Re:so why can't i buy a !@##$% low powered compute

[Anonymous Coward]

The only OLPC laptops I can find for sale now are on eBay, from people who would willingly profit from a charity. Is there a better source?

Become a contributor to the OLPC Project and get one today. http://blog.laptop.org/2010/02/25/xo-1-5-early-production-laptops-free-to-contributors-worldwide/

Looking in the wrong places

[pedestrian crossing]

People put too much stock in the home energy savings angle.

I hooked up a Kill-A-Watt to my whole computer/media center stack (computer/monitor/stereo/mixer/external HDDs/assorted electronics), and it was pulling about 65 watts. The only thing that pushed that number up was cranking the stereo.

It costs about 5 bucks a month to leave this stack running 24/7. 5 bucks. If I was to be looking for significant energy savings, I think I would be looking somewhere else.

Want to save energy? Insulate your house. Take shorter showers.

Want to save money? Stop drinking cappuccinos and eating fast food, but leave the fscking computer alone.

anyone who has hooked up a Kill-A-Watt to their computer, and then calculated how much money per year they're spending on it, disagrees with you.

I've done it and I disagree with -you-.

Re:Looking in the wrong places

[Anonymous Coward]

Uh, the monitor alone probably pulls 65 watts... No fair testing when your computer is idling, the monitor is sleeping, and the external drives are spun down.

Re:Looking in the wrong places

[evilviper]

I hooked up a Kill-A-Watt to my whole computer/media center stack (computer/monitor/stereo/mixer/external HDDs/assorted electronics), and it was pulling about 65 watts. The only thing that pushed that number up was cranking the stereo.

A 17" LCD will draw 35 watts. I don't believe for a second the rest of your PC is even idling at under 60 watts on its own. The only way your numbers make any sense is if your "computer" is a laptop, drawing about 30 watts all together. In which case, you are the exception, not the rule...

Re:where do you get your facts?

[Xargle]

Your credibility as a reference was lost when you fail to expand the ARM chip acronym correctly.

Hah! Fail, Mr AC Troll.

The price will fluctuate up and down

[Colin Smith]

Wanna make a $10,000 bet on the price?

I'm already making a much much bigger bet than that. On the order of several hundred k.

The oil price will go up till it kills the economy, puts millions out of work. Then it'll crash because nobody can afford it. Then as everthing recovers over the next couple of years it'll go back up again with demand. The specific price is going to depend on the level of inflation by the government.

There is a chart of whale oil prices which are a reasonable model of what's going to happen.

http://www.theoildrum.com/node/4672

Other commodities are recyclable. You can melt them down and reuse them. Oil, not. It's burned it's gone. Course we may switch to electric. Which means renewables, coal and primarily nuclear. But we have a huge oil infrastructure which would have to be replaced. Trillions of dollars. Which means lots of inflation and lots of recession.

 

Re:The price will fluctuate up and down

[compro01]

But we have a huge oil infrastructure which would have to be replaced.

Oil fields sure, but refineries and front end infrastructure would likely be re-purposed, as I doubt that liquid hydrocarbons are going away as a means of energy storage and transport anytime soon. The energy density is just so much better than other currently available means. By volume or mass, gasoline has over 40x better energy density than the best batteries we have, and hydrogen can't hold a candle to it by volume.

Re:The price will fluctuate up and down

[micahfk]

All of which avoids the point that there is no peak oil theory. Supply and demand, new technology, new exploration (mostly outside the US really) all play a part in bringing commodities downward.

Oh, and those graphs are a 15 year timescale for whale oil--we're talking a longer time period.

Re:so why can't i buy a !@##$% low powered compute

[fuzzyfuzzyfungus]

Well, that depends. If you handed Win95 a standard Atom based modern PC(assuming you could even boot it) you would get terrible results, as you note. No ACPI, no serious power management.

If, by some strange miracle, Intel decided that there was a market for modern-day PCs equivalent in performance to ~33MHz Pentiums with 16 megs of RAM, they could definitely fab something that fulfilled that performance objective while using next to no power, even if the OS did no power management at all.

Had to look

[zogger]

Wasn't sure what that was. For people who don't know, here is the write up on the emate. http://en.wikipedia.org/wiki/EMate_300

Interesting article

[noidentity]

That was well worth reading. It touches on some interesting things, like how they originally kept power usage down to allow a cheaper plastic body for the chip, and how we don't have good profilers to find power usage hotspots in code.

Re:so why can't i buy a !@##$% low powered compute

[commodore64_love]

Sorry. I guess I should have said "Windows 95 [with upgrades to make it work with a modern Atom CPU]" in my previous post. Obviously I would not use the stock Win95

Anyway...

I use Windows 98 on my laptop with only 64 megabytes. It works just fine. I'm not sure why people say these older OSes are "no good" since it runs web browsing & word processing software a-okay and without needing a lot of power to do it (4 hours on a NiMH battery). Just replace the inefficient AMD K5 with an efficient Atom.

Oh and as for the Linux suggestion, I have used things like Puppy Linux on this laptop and enjoyed the speed of its RAM-based operation, but it failed to see the modem or the sound card. I like OSes that work, and it didn't seem to function properly.

Re:so why can't i buy a !@##$% low powered compute

[Anonymous Coward]

a) The refresh rate on e-ink is about 10x too slow to make it worth it

b) Connecting to the cloud requires radio or other wireless technology. That eats up battery power.