Slashdot Mirror
OLPC Halves Power Consumption For XO 1.75
angry tapir writes "One Laptop Per Child (OLPC) has reduced the price of the next version of its notebook to US$165 and power consumption has been slashed by half compared to the previous version. The XO-1.75, with its 8.9-inch touchscreen, will start shipping in the second quarter of this year."
I'm convinced, after looking at the proposed specs and the mock-ups for the that the $165 price-point is a pie-in-the-sky fantasy.
But... If it does miraculously appear for under $200 and can run Android or some other Linux variant, I'll buy 3 or 4...
If you want news from today, you have to come back tomorrow.
surprised this wasn't in the summary but "The XO-1.75 is the first OLPC laptop to use chips based on processor technology from Arm Holdings, which has been a huge factor in reducing power on the laptop, " Good stuff, and it seems as if the mythical $100 price is within shooting distance
They should have started with ARM to begin with. Had they done that then they wouldn't have had the issues with Intel back stabbing them nor Microsoft wasting their time. Better late than never I guess.
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
But didn't we learn from the promise and price fiasco with the 1.0 and beta of this hardware? I personally remember being very excited in the mid 2000's about this hardware, how it was going to save the world, bring tech to the masses and the promise of a reasonable computer bringing network computing to the masses around the world (including the US where there are 30-40 million people without modern computers). The lack of bandwidth (price and locale) as well as the internal issues in OLPC made this a forgettable device much like that of the folks at CrunchTech with the CrunchPad. Lessons learned, first in wins the game, its time for tech companies to stop playing catchup and do something real and inventive that will change the face of our user experience. Cheap hardware and software is a nice idea, I'm just not holding my breath on this evolutionary release (soft and hard).
How many laptops does a kid need?
Do you or your partner snore? - Visit www.snoring.com.au
On the one hand, this article makes a clear case that there will be children in Chad mindlessly turning a crank for one hour and 47 minutes in order to do their homework for the night.
Yet on the other hand, these kids have orders of magnitude more computing horsepower than I did as a Reagan-era high school kid in an upper middle class community. Hard to know who should envy who.
I'm generally "Interesting," "Insightful," and even "Funny" here. What the hell happens to me at parties?
Does it run Linux or is it a case of getting them poor peasants hooked on Windows so that Microsoft can find a way of taxing them at some point in the future?
McNierney recharged the XO-1.75 with a hand crank. It takes 1 hour and 47 minutes to fully recharge the battery by hand, he said.
I wish they also said the average life of one charge to put the quote into perspective.
But close to a couple hours to charge the battery makes me think they should come up with some more passive solutions.
Great! Can't wait to buy one.
Note that I said one.
Advice: on VPS providers
It always was a way to get R&D paid for under the guise of charity. They could have easily made a computer for under a hundred dollars, but that wasn't what they were trying to do.
I linked to an article about OLPC 1.5. However, this more recent article states that they are continuing to use a Fedora-based OS.
http://www.rootstrikers.org/
Flash is important for the web and Java is a MAJOR standard in learning programming and developing web and applications on Linux.
http://saveie6.com/
The main problem with the OLPC, the one thing that made the project open to subversion by companies like Intel and Microsoft, is its centralized model of development. You get the laptops or tablets from one source, say, the central government of the country that buys into the idea or some buy-one/donate-the-other scheme. I understand that it's supposed to be more of an educational than a computing project. But this set-up generates dependency. What happens when the machines are damaged? More importantly, what happens to the next batch of children without laptops? Since the machines are manufactured in the usual Asian places (hint: two countries claiming the same name), this will likely result in a foreign exchange outflow from a country that can least afford it, as certain essential non-technological items (e.g. food and basic medicine) may need to take priority.
What the OLPC should have set out to develop is a RepRap-like infrastructure that will allow the adults (or even older children) of the community that takes part in the project to manufacture the laptops by themselves from cheap, readily available components. If this isn't 100% possible, then give them at least enough transfer of technology to allow them to build the least technological parts, like the case or the keyboard. Think of a laptop case made out of recycled plastic or hard laminated cardboard. Then again, how far off is the day when we can run a desktop OS on an Arduino board?
Don't just give them fish. Teach them how to fish.
Computers made using such technology might appear crude at first, but not much cruder than the devices that ushered in the PC revolution.
then why did they switch to tablets?
i thought these machines were meant to assist in education, not for them to waste time doing worthless crap? can any of YOU imagine using a touchscreen tablet for school work? because i sure can't, not without a proper keyboard.
it seems to me that the money spent on the touchscreen would have been better used in putting in a better processor or more memory. or just lower the price.
Wealth is the gift that keeps on giving.
I never thought I'd be a beneficiary from the OLPC project. I'd never be able to use an OLPC for anything I do. But I love how the project has put a bent in the technical landscape of portable devices industry. It was a failure as an education project perhaps, but it succeeded in more than one way as a laptop research project.
When OLPC came out in 2007, the laptops were on a lap-melting, back-breaking rush towards bigger & faster. Nearly everything came in with a Core2 or a Core2 Duo, with lots of RAM (yeah, guess what you can't save power on, RAM needs a strobe whether it has data or not). The fact that OLPC came out in 2007, sort of forced the geeks to look at weight as a valid concern for a consumer device. Not to mention questions about why a 1995 top-end laptop ran for 4 hours on batteries, when a 2005 one won't do the same at the same weight.
Less than a year after OLPC came the rush of netbooks. Finally machines that people can afford to buy (like here in India) and carry around without being tied to a wall plug. Scroll paste a few years, it is not only consumers, using them. I see Rasmus post PHP benchmarks off his netbook, I see entire teams (like Inkscape) suddenly sit up and re-work their UI workflows/dialog-space for it. I see the Notion Ink use OLPC Pixel Qi tech in the new tablet.
Socially speaking, the project has been a great failure. But technologically, it has left a huge impact on portable devices everywhere. As for the former, the project probably forgot that "Charity begins at Home". Refusing to sell full-price to americans wanting them shows a complete lack of understanding of how economies of scale & price segmentation would've worked out. I'm not going to mourn the failure of Negroponte, but I'll just give the technical folks at OLPC a big thumbs-up.
I'll happily pay 200$ for an arm netbook'ish if they'll sell me one in India. Hell, I'll even fix all the things that don't work for me - for FREE. Not all of us are poor & in need of a hand-out. Heck, I'm at the verge of putting in a pre-order for a Notion Ink Adam, for double the price, if the hype pans out.
Quidquid latine dictum sit, altum videtur
With the previous OLPC laptops, a lot of effort (IIRC) went into finding hardware chips from manufacturers willing to provide open drivers.
Will the same thing apply to this new OLPC? Or are they abandoning the idea of openness?
Why are they developing it at all when I can now buy an atom based linux netbook or android tablet for under £100? Am I missing something?
I'm not sure I would credit OLPC with that. I would think that Asus and their marketing of their EEE has more to do with the trend of netbooks than OLPC.
Socially speaking, the project has been a great failure. But technologically, it has left a huge impact on portable devices everywhere. As for the former, the project probably forgot that "Charity begins at Home". Refusing to sell full-price to americans wanting them shows a complete lack of understanding of how economies of scale & price segmentation would've worked out. I'm not going to mourn the failure of Negroponte, but I'll just give the technical folks at OLPC a big thumbs-up.
I think it is unfair to declare this project a great failure, even in the social area. The project hasn't ended yet, and considering the highly ambitious goal they have set themselves, it is no wonder it takes them time to reach it. But I agree that the technology spin-offs of the project are remarkable too, although I think you should give Negroponte some credit for that part too.
I'm not so sure mass-selling the original OLPC on other markets would have been a good idea. The support organization to do that properly would have been costly, and a distraction from the real goals of the project. Perhaps they could have created or partnered with a commercial enterprise that would license the design to sell to consumers, but getting the legalities of such a deal right would have been tricky. And the first OLPC design was really just a beta anyway, if that.
I've tracked the OLPC project and have worked in educational technology for years, and arguments over processors and power consumption are bullshit. The same goes for the philosophies of education behind educational technology. At the moment, the biggest issue is teacher training. Simply put, most teachers don't know how to use computers in the context of classroom teaching. That's even true when it comes down to the basics. Sticking an ARM processor into the case isn't going to solve that. Getting the computer to run on 2 Watts isn't going to solve that. Praying that the child is smarter than the teacher when it comes down to adopting new technologies for learning isn't going to solve that. Indeed, this emphasis upon technology over learning and these idle hopes that children are better at using technology for learning have left educational technology in the same cesspool that it was in 30 years ago: teachers, the people who are responsible for guiding children through the process of learning, are almost as ignorant about how to use it today as they were way back then.
(For what it's worth, I think that there is some value in the 'student is smarter than the teacher' mentality when it comes down to educational technology. Yet that only works for a subset of children, since it involves a lot of self-motivation.)
"Socially speaking, the project has been a great failure."
By what standard has it been a failure?
The project is helping millions of 3rd world kids wrt education. Not as many as Negroponte would have liked to, but how's that a great failure?
Are you aware that many of he great tech corporations want this not-for-profit paradigm changing project to be a failure?
My laptop only cost $30 plus shipping from ebay. Yeah it's slow (700 megahertz) but it has enough RAM (1/3 gigabyte) to run Firefox and get online. As you said, for most people that's all they care about. A cheap computer that can give them net access... and that's the point of the OLPC project.
RQ:
This OLPC claims it has minimal power consumption. How do customers know, before they buy, how much their new laptop or desktop will use for consumption? Mine desktop uses 110 watts which is a heck of a lot, and I'd love to downgrade to, say, 25 watts.
"I disapprove of what you say, but I will defend to the death your right to say it." - historian Evelyn Beatrice Hall
Actually, I still use my original XO-1 for one thing that no apparently other netbook or laptop can do: it is the only netbook I know of that can be read in full sunlight. When I go to the beach, I can sit 5 hours in the sun while working - in my case, using the bash shell, which is perfect for my particular project. To my knowledge, no other netbook or laptop can do that. If there is one, let me know so I can prepare for next summer... I still hate the OLPC keyboard.
(Actually I don't like the beach that much but go to keep the GF happy. I've given up on arguments about dangers of too much sun. I sit under an umbrella with SPF 30 sunblock while she perfects her tan.)
AFAIK RAM uses only a little amount of power, even when you have lots and lots of it. Since having more memory is bound to shave of some power consumption in other areas, it's probably the dumbest area to cut down in order to save power.
Switch back to Slashdot's D1 system.
When OLPC came out in 2007, the laptops were on a lap-melting, back-breaking rush towards bigger & faster. Nearly everything came in with a Core2 or a Core2 Duo, with lots of RAM (yeah, guess what you can't save power on, RAM needs a strobe whether it has data or not). The fact that OLPC came out in 2007, sort of forced the geeks to look at weight as a valid concern for a consumer device. Not to mention questions about why a 1995 top-end laptop ran for 4 hours on batteries, when a 2005 one won't do the same at the same weight.
In my experience, smaller and lighter laptops (subnotebooks) have been around for a long time, but before the OLPC, they were expensive premium items. The OLPC made it clear that you could sacrifice some of the premium features, and make it both small and cheap.
While the advances in technology were probably playing a part, there was also a social acceptance factor. Surprisingly, people were content with machines that looked and felt cheap. You could see the same thing with regular laptops, which some years ago were mainly used by executives. Now you see them at homes and colleges everywhere, because they are simply convenient for those uses.
Having used laptops as my primary computers since 1997, I feel the same general idea about them as I do with cell phones (been using since 1995): they are more "personal" computers. Cell phones make social sense, because you usually want to call a person instead of a place. In the same way, your computer belongs to you, instead of some fixed place.
There is another interesting factor with social acceptance. Sitting at a desktop computer isolates you from the people nearby, but a laptop is less intrusive. For example, sitting on a couch with the laptop on your lap, there is nothing in the way of eye contact with others. Netbooks, tablets and cell phones are even less intrusive, and thus more socially accepted.
Escher was the first MC and Giger invented the HR department.
Nice to see UK ARM making more roads into computing. Us old Acornphiles knew that the humble ARM chip would take over the World 20 years ago!
Why do these kids need a full powered laptop and can't just have a Kindle with (or without) unlimited 3G.
- The battery life is phenomenal
- You can drop it
- Low learning curve for teachers
- Extremely light weight = cheap to ship oversees
- Project Gutenburg - 33,000 free books
- You can read on it outside (why not make the entire back of it a solar panel?)
Why are they using cranks when there are solar powered backpacks? (if the amazon price is $75, then the seller is just making $45...)
http://www.amazon.com/Solar-Pow-Backpack-Navy-ET0016N/dp/B004B1A9IW/ref=sr_1_13?s=electronics&ie=UTF8&qid=1294496630&sr=1-13
or http://www.youtube.com/watch?v=1ly3vzxztxA
hate to break it to you, but
number of linux laptops shipped by olpc: 2,000,000+
number of windows laptops shipped by olpc: 0
if deployment schools want windows, they have to work that out with microsoft. olpc's only involvement was patching the firmware to make it possible, and microsoft paid for that. they've never spent a dime on windows shit or shipped a windows laptop.
A *desktop* from that era, late 98se/early windows XP is fetching a hundred dollars, with 256 megs ram,like pc100 or 133, let alone anything newer, and any similar era laptop is $200 and up. (I don't do ebay or paypal so that is out, I don't like their ethics for one and some other reasons, and I like to actually see and checkout hardware I am buying...) Believe me, if I could have gotten a half way decent laptop good enough for net surfing for 50 bucks near me in the last two years I would have jumped on it. 200 "firm" is the cheapest I have seen at any CL listing or local mom and pop computer store. I don't want a "netbook", but a real laptop, but will just hold out until I can buy a new one online with a warranty from some big name brand place rather than pay outrageous "outside the major urban areas" prices on old junk. As it is, I am milking out my ancient built from parts desktop as it is. Seems like it is no use anymore, unless you upgrade constantly, you can't reuse any of your old stuff anymore, the ram changes, the socket changes, the interfaces change, your cards IO changes, etc, so in essence you need to buy brand new about every two years or so. I *used* to incrementally upgrade, but you really can't do that much today, the "older" stuff that might still work for you is now more expensive than brand new, so there's no point in doing it..I am still running DDR, like pc2100, an AGP vid card, and PATA/eide hard drives and optical drive. Can't upgrade any of that stuff without going to brand new everything anymore, a new machine, so you might as well just buy a new one with a warranty (as much as that is useful..). And forget $15 dollar a month DSL, you can't get dsl or cable here, I had to settle for an (relatively) expensive cellular data plan to have anything better than dialup.
And this isn't way out in south moose droppings alaska or anything, just normal far enough out to be some farms mixed in with subdivisions. We are NEVER going to get the wired internet providers to provide service beyond where they were already at years and years ago, unless mandated by law, like telephones and electric service was in the past. They are more interested in taking their billions and buying up media companies rather than actually upgrading their pipes in areas that are stuck on copper that was run when tesla and westinghouse and edison were still alive.
Don't get me wrong, I'd love to have a 50 buck laptop and 15 dollar dsl to use with it, but don't assume everyone across the country has those options. There's a gigantic diff betwen heavy right and left coasts and a few major urban areas and *everywhere else* in the US now and internet is just one of the schisms that is growing daily.
We are really a totally different USA and it feels like we are occupied territory and have to pay for everyone else's cheap stuff, then get put down all the time by those same areas when all we really want is a fair 21st century shake at things.
I have a G1G1 OLPC 1.0 that we bought in a fit of enthusiasm back when they first came out. These days, it sits untouched.
The greatest highlight on the device was the great screen. I certainly hope Pixel Qi finally starts shipping in volume at a reasonable price. The eReader mode looks great, and I'm sure they've improved in the intervening time. The tough, splashproof hardware was nice too.
Unfortunately, lowlights abounded: It was incredibly slow. It took forever to boot and applications starting took way too long to start. The keyboard was atrocious (even kids are better with a full-sized keyboard). The touchpad was wonky. The software only made sense if you were with a whole classroom full of other OLPCs (at the university I work at, we had a couple of get-togethers, and we managed to do a little bit of the "sharing" stuff with it). It was cool when it worked, but hard to get to work. Documentation wasn't great and even with a collaborative effort by 10 or so IT folks, we couldn't get things to work consistently.
Even with the enthusiasm most of us had for the idea, it was hard to get enthusiastic about developing for it as you couldn't do much out of the box and those of us with the expertise to program found it uncomfortable to work on. Yeah, you could plug in a keyboard an mouse, but still, it wasn't a great experience.
I also tried a few of the alternate faster-booting linux variants on it, and those improved the software, but the keyboard/touchpad still made it unpleasant to work on.
I know, I know. This thing is way beyond what a kid in abject poverty in a developing nation would have otherwise, but if people with expertise in computers can't get it to work and find it uncomfortable to work on, how would the expect to get developers to work on it?
Then the later issues with Classmate competition, XP, management, delivery, etc all compounded apon the initial hardware design issues....
I thought it was a great project, but not so much that I actually wanted to use the device after I got one. I tried for a long while...
- AlanH
The current OLPC XO-OS now dual boots to either the Sugar learning environment (for kids) or the more traditional GNOME GUI for a more windows-like experience. This is a really big improvement IMHO for "more traditional" and older users as it made a lot of Linux apps work out of the box on the XO. With Sugar, you need to "Sugarize" apps first to be able to run with the GUI if you didn't want to launch them via command line.
http://www.object404.com
Bah, I doubt OLPC had anything to do with it at all. The thing that launched the whole market in 2008 was because Intel suddenly had a dedicated low-power design in the Atom not just LV/ULV versions that were custom versions/bins of existing designs with large dies. The extremely small die size compared to all the other processors gave Intel a reason to flood the market with cheap computers and still make a very, very solid margin.
New processor designs take 3-4 years from start to completion which is why Intel has 4 years betweem every tock, AMD has said the same about GPU designs. That puts the Atom back in 2004-2005, while whatever the OLPC project did in 2006-2007 which was their main years is completely irrelevant to that. It was an existing project at Intel, of course they didn't mind if the OLPC paved some ground but it was going to happen regardless.
Maybe it got more attention in India but I can promise around here nobody had heard about the OLPC project, nobody has seen an OLPC and yet netbooks and minipcs with Atom were a big hit. Negroponte doesn't exactly have a monopoly on predicting that computers will be smaller, faster and cheaper. And near as I can tell, netbooks ended up running mostly Windows with about zero of any OLPC innovations. Sure he achieved a few things, but this is mostly just wishful thinking.
Live today, because you never know what tomorrow brings
See here for video card power consumption: http://forums.atomicmpc.com.au/index.php?showtopic=264
Laptop GPUs use somewhat less but the numbers can be hard to find.
CPU power consumption specs can be seen on wikipedia (or on Newegg)
110W is not actually that much these days. I don`t think you are going to get down to 25W unless you get a small form factor motherboard that uses mobile chips.
You're entitled to your perceptions, but the XO is far from the CrunchPad. Since Negroponte's overpromising, OLPC has shipped two generations of hardware, several software releases for the laptops (and a school server), and developers have created a few hundred activities for the UI. The hardware and software was "real and inventive", now they're actually executing and delivering it while the technorati drool over new shiny.
From http://wiki.laptop.org/go/Deployments , "There are over 1.85 million XOs in the field as of August, 2010." 500,000 kids in Peru and 400,000 in Uruguay have the laptops, tens of thousands in other countries.
=S
You're insane if you think that a developing company can assemble laptops for even 4x what it costs Quanta to crank them out by the hundred thousands. Getting to one laptop per child in a school is expensive enough already.
Transfer of technology is important (though less important than educating kids), and the OLPC open source software facilitates that; anyone can participate in development and I think Uruguay is off developing its own software. But turning a mass-produced rugged device into a locally-made expensive cardboard joke is over-reaching.
=S
MS FUD scared educational ministries into asking for Windows, but I don't think any deployment is actually using Windows on the XO. Every single one shipped running a version of Fedora Linux with the "Sugar" UI, and recent software releases ( http://wiki.laptop.org/go/Release_notes/10.1.2 ) offer a Gnome desktop as an alternative.
=S
Bah, I doubt OLPC had anything to do with it at all. The thing that launched the whole market in 2008 was because Intel suddenly had a dedicated low-power design in the Atom not just LV/ULV versions that were custom versions/bins of existing designs with large dies. The extremely small die size compared to all the other processors gave Intel a reason to flood the market with cheap computers and still make a very, very solid margin.
Intel pushed Celeron-M / i915 platform for "Classmate" as a competitor to OLPC. Asus partnered with Intel for the original EeePC: http://www.pcmag.com/article2/0,2817,2141742,00.asp What does it contain? Celeron-M, not Atom. In fact the Celeron-M can't even make use of the power saving SpeedStep on identical Pentium-Ms. It seemed clear these Celeron-M machines (either in Classmate, or in EeePC) was a stop-gap until Atom could be released.
Atom was originally designed for other embedded applications, and not mini-PCs.
It is unconscionable to construct a summary specifically about power consumption without actually stating what the effing power consumption is. We have units for these things, people. Why always use the relative ones?
There is another interesting factor with social acceptance. Sitting at a desktop computer isolates you from the people nearby, but a laptop is less intrusive. For example, sitting on a couch with the laptop on your lap, there is nothing in the way of eye contact with others. Netbooks, tablets and cell phones are even less intrusive, and thus more socially accepted.
A key but very undersold benefit of the XO and it's Sugar UI is their focus on collaboration, by design. The activities (apps) let kids share and learn together, so they are encouraged to interact.
OLPC Australia