Paul Otellini + Classmate PC
Who knows Intel's Bill Of Material for the Classmate PC? I am Charbax and I'd say it's probably more than $400 a laptop. That is based on the fact that the Classmate is nothing less than a stripped down normal business laptop with optional flash memory instead of a hard disk.
Also part of Intel's strategy is to assemble the Classmate locally instead of mass producing in China, both factors, local assembly infrastructure and low volume production both push the price way up much higher than the $250 that Craig Barrett is claiming on 60 Minutes.
The cost estimations and the specifications of every single component of the XO-1 computer are available at the official wiki. $175 was Nicholas Negroponte's latest estimation as of last months industry analysts meeting.
Did you know that Intel's annual revenue is over $35 billion dollars and that over $5 billion dollars of that each year is spent on R&D? Compare that with the OLPC Foundation reportedly using a $30 million dollars on R&D over the last two years. $20 million dollars that also includes:
- the cost of the thousands of pre-produced XO-1 beta models,
- the expensive testing, development, engineering,
- the partnerships that needed to be found to manufacture each component,
- the 4x better battery life than a conventional laptop at heavy use with full backlight settings on
- more than 10 times better battery in light use and black and white reflective screen mode
- the revolutionary wi-fi mesh network
- the DCON functionality that keeps the main processor turned off most of the time in light usage scenarios of a computer.
One laptop is a dog
How could Intel choose not to develop a "Do no evil" plan-B for this obvious situation arising in this competitive market, where someone in the computer industry are now delivering cheap and simple computers instead of powerful and over-bloated ones. How can Intel not come with anything more innovative than a stripped down normal business laptop?
The Classmate PC has a UMPC-type ULV processor requiring cooling, only having caved in on not including the HDD and instead using a 2GB Flash memory which still only saves about $30 on the BOM and only improves slightly Intels battery life. Why didn't Intel invest anything in Low-Cost Computer technology yet?
How come the cheapest Intel computers available today in the big retail stores aren't any cheaper than those that were 5 years ago? We were running approximately the same Windows XP OS with approximately the same user experience as we have today. Yet 95% of computer usage five years ago and today is opening a browser, playing a few small multimedia files and eiting documents.
I don't need a quad-core 45nm high-k metal gate processor running Windows Vista to do that.
Does the Classmate's processor have a fan? That would be really bad for long term durability
5 Billion R&D and this is what Intel wants to give us -- wow
Just to be fair: the 5 billions in R&D are not entirely dedicated to the Classmate. You should really compare apples with apples. Developing nanofabs and new technologies is costly and to this date only few companies can afford such investments. So, of those 5 billions, the great part of it goes to technologies other than the Classmate.
I am not defending Intel. I think that instead their R&D investment for the classmate was an afterthought, they basically assigned some bits (probably less than what allocated for the OLPC project) to build something usable. Their fault is not in wasting money in R&D in such a product, but to investing too little resources on it. From any way you look at it, you see that the device is basically a collection of existing items with nothing innovative. Call it a low-cost patchwork if you want.
In simple words: they got late to the game (which you can see in the poor power optimization) and now with little resources they delivered such product, just to compete with the XO. Do you think they care if it is a good tool for education and it will make a difference?
Indeed, Intel did not use $5B a year on the Classmate.
But they DID reserve $1B over 5 year for the Classmate.
http://www.nature.com/nature/journal/v447/n7140/full/447006a.html
Then $500M in hte first year sounds reasonable. And for that money the Classmate is really pathetic. It suggests that this money is not meant to make the Classmate a better educational laptop. So where DO they spend this money on? (subsidies?)
Winter
Winter: The direct quote from your Nature article is as follows:
"Over the next 5 years [Intel] is investing $1 billion to promote access to information technology in developing countries."
This suggests that the money came come from somewhere other than R&D. Not only that, but it suggests that the money does not relate specifically to hardware development costs. Your "$500m in the first year" figure sounds plucked out of the air.
It is true that we do not know exactly how Intel will spend this cash - the article refers to the next 5 years spend, not the last 5 years - but I would have thought that a chunk of it will go on marketing, distribution, training awareness and possibly even maintenance of classmate and other Intel products - there are a lot of businesses in the developing world too, you know.
Of course, Intel could just be freeing up $1bn for tax relief... (I don't know whether such a thing is actually possible, but I thought I'd better get a dig at Intel in there somewhere since it seems to be mandatory in the comments here nowadays).
David,
Winter is not here to engage in intelligent conversation. His goal, apparently, is to distort or fabricate the facts in order to sustain his arguments, no matter how fatfetched they might be.
The lastest example is his claim that the Intel machines' real cost is $200 over the selling price. Wayan asked him to provide the source for this information; he promptly ignored the question.
I have very little respect for Winter, who has proven, time and again, to be of very questionable intellectual integrity.
Author wrote:
"Innovations including
- the 4x better battery life than a conventional laptop at heavy use with full backlight settings on
- more than 10 times better battery in light use and black and white reflective screen mode"
Well, it seems things MAY not that rosy!. Look at the surprising results obtained by Luis Ramirez, an XO advocate:
http://luisramirez.cl/blog/?p=575#more-575
All my respect goes to Luis for having the honesty to report things the way he sees them, in spite of his preference for the XO machine. Thank you, Luis!
Troy wrote:
"The lastest example is his claim that the Intel machines' real cost is $200 over the selling price. Wayan asked him to provide the source for this information; he promptly ignored the question."
The answer is hanging in Wayne's moderation. Somehow, I couldn't post directly in that thread. Off line life prevented me from answering earlier.
But the short reply is, I got the number here:
http://www.olpcnews.com/people/negroponte/olpc_pot_intel_black.html
You seem to have missed that post, as neither you nor anyone else complained about the estimate then.
Winter
Winter wrote:
"But the short reply is, I got the number here:
http://www.olpcnews.com/people/negroponte/olpc_pot_intel_black.html"
Typical Winter m.o: his "facts" are based on silly rumors!
Did he ever even bother to check the legitimacy of those rumors before passing them off as facts? Of course not...
Why bother with the truth when you can pull facts out of thin air?
"Well, it seems things MAY not that rosy!. Look at the surprising results obtained by Luis Ramirez, an XO advocate:
http://luisramirez.cl/blog/?p=575#more-575"
Indeed. The B2 had severe power problems, as was readily acknowledged by all involved. The B3 is rumoured to do better ;-)
However, all the projections of the XOs battery life are based on the Linux kernel people getting the power management code right. And they DO take that seriously. Greg Kroah-Hartman and Marcello Tossati are only two of the top kernel developers involved in the OLPC kernel (not all on power management though).
From:
http://laptop.media.mit.edu/laptopnews.nsf/2e76a5a80bc36cbf85256cd700545fa5/6709ac0696003ee3852572e00083567f?OpenDocument
>>>>>
8. Power management: This week we passed a major milestone: working ebook mode using the read activity as vehicle. Suspend and resume are working on both GX and LX--although we have a few bugs left on the LX. Walter Bender demonstrated the fruits of many people's labor in Argentina, Peru, and Uruguay. In one test--with the caveat that the the WiFi was off, an XO ran for 23 hours on a 92% battery charge. Even with the backlight on, we have seen ebook mode run for more than 13 hours. Thanks go to Don Hopkins, Chris Ball, Mitch Bradley, and Andres Salomon.
The last major functional piece of resume is working; Marcelo Tosatti reports that it appears that basic functionality (detection of device insertion) is now working, even though he has no idea why (perhaps some change in mainline). All we need to do is to power up the USB ports after resume. Once this is done, we should have completed basic suspend and resume work, and move on to performance (both speed of resume, and power management in general. This will also allow us to have the mesh alive in ebook mode as soon as we have completely autonomous mesh firmware.
<<<<<
Winter
(who seems to instill severe doubts about his intellectual integrety)
Luis Ramirez's findings regarding battery life closely match the experience of another reviewer:
Quote:
**************************
Negroponte has a very optimistic target for the XO's power consumption: two watts. Compared to around 10 watts for even the thriftiest laptops, that would be an excellent performance, giving the machine of runtime of more than 10 hours despite the small battery capacity. The prototypes are nowhere near this performance because power management does not yet work. At present, the laptop runs for 2.5 hours with and 3.5 hours without display illumination, consuming 9.1 and 6.5 watts, respectively.
**************************
Source:
http://www.heise.de/mobil/artikel/88916/3
Troy wrote:
"Did he ever even bother to check the legitimacy of those rumors before passing them off as facts? Of course not...
Why bother with the truth when you can pull facts out of thin air?"
All deals from Intel are secret. So if people go on unsubstatiated rumours, it is their own fault. They can lso easily tell the world when it is wrong. They did watch the 60 minutes video I assume.
Intel themselves have stated that the true cost of the Classmate are around $350. Others have made credible arguments that it is closer (or over) $400.
Charbax has posted around the world that Intel sells the Classmate at $180. We could ask him where he got that number. Hey Troy, you could do it yourself!
http://charbax.com/2007/05/22/illegal-dumping-of-low-cost-computer-intel-classmate-does-not-cost-180/
In the 60 minutes program, Negroponte essentially said the same (without the exact price):
>>>>>>>>>
Leslie Stahl: Intel gave every student in this class in Mexico a 'Classmate' which Negroponte believes is part of an effort to kill him off.
Nicholas Negroponte: It's predatory.
Leslie Stahl: At a recent lecture at MIT he accused Intel of dumping, of going to the same governments he's trying to sell to and offering the 'Classmate' below cost.
Nicholas Negroponte: Intel should be ashamed of themselves. It's just shameless.
Leslie Stahl: Craig Barrett is Intel's Chairman of the Board.
Leslie Stahl: Negroponte believes that you're trying to drive him out of business.
Craig Barrett: We're not trying to drive him out of business, we're trying to bring capabilities to young people and it's more than just Intel. It's going to take the whole industry to do this.
<<<<<<<<<
The nice part is that Barrett doesn't even rejects the dumping charges (and you CAN be sure he knew of them).
Need more evidence? I am always willing to help you use Google.
Winter
So, in all that rambling, where is the evidence that the Intel mnachines cost $400?
It doesn't take a genius to see that the whole thing is pure nonsense. But then again, here is your "reasoning" (I quote you):
*****************************
All deals from Intel are secret. So if people go on unsubstatiated rumours, it is their own fault.
****************************
Nice and honest logic, isn't it?
Following your "logic", since you keep your real name private while defending the XO from legitimate questions, it is "logical" to conclude, BEYOND ANY DOUBT, that you are Nicholas Negroponte!!!!
And, if that conclusion is proven false, we can always blame you for keeping your name private!
Wnter wrote:
**********************
However, all the projections of the XOs battery life are based on the Linux kernel people getting the power management code right. And they DO take that seriously. Greg Kroah-Hartman and Marcello Tossati are only two of the top kernel developers involved in the OLPC kernel (not all on power management though).
From:
http://laptop.media.mit.edu/laptopnews.nsf/2e76a5a80bc36cbf85256cd700545fa5/6709ac0696003ee3852572e00083567f?OpenDocument
********************************
You need to learn to "read between lines", Winter. Be smart.
Let me give you an example:
On that same page, look at item #7 - it says:
7. Human power: Pedal Power Haiti wants to try our laptops with the
pedal power system they are using from Dissigno (a San Francisco-based
human-power startup). This system is big, but folds up, and is in test
already in Nepal and Haiti. The output is between 12–14.6 volts at
50–70 Watts. The system was designed to charge a car battery, and can
work with our gang charger systems as well.
****************************
Let's do some "reading between lines":
1. Ok, Dissigno is the creator of the Pedal power System
2. Dissigno is "a San Francisco-based
human-power startup"
How true is that?
#2 FALSE - Dissigno is NOT "a San Francisco-based human-power startup" - They are a marketing company. Here's their website and mission statement:
http://www.dissigno.com/about/index.html
#1 FALSE - The creator of the "pedal power system" is NOT Dissigno - it is this guy:
David Sowerwine - http://www.ecosystemsnepal.com/founder.php
and this is his creation:
http://www.ecosystemsnepal.com/research_development.php
^^^^^^^^^^^^^^^^^^^^^^^^^^^
See, Winter, how much info you can get with due diligence, by doing legitimate reasearch and a bit of intelligence and fairness?
Will,
"Does the Classmate's processor have a fan? That would be really bad for long term durability"
It looks that way:
"One unexpected thing in Classmate (which actually was new for me) was the fact that it has a cooling fan. It was awkward because the room was a bit cold and the fan went on and off at least 5 times during the test. It was quite annoying because it is slightly louder than a normal laptop"
( http://luisramirez.cl/blog/?p=575#more-575 )
Troy wrote:
"So, in all that rambling, where is the evidence that the Intel mnachines cost $400?"
So what IS the real cost of the Classmate?
Cost estimates:
http://news.bbc.co.uk/2/hi/technology/6675833.stm
http://www.iht.com/articles/ap/2006/12/05/technology/LA_TEC_Brazil_Cheap_Laptops.php
http://www.engadget.com/2006/12/06/intels-classmate-pc-to-go-head-to-head-with-olpc-in-brazil/
http://news.com.com/Intels+bridge+for+the+digital+divide/2100-1005_3-6084250.html
http://mobile.vnunet.com/vnunet/news/2186476/intel-ships-first-olpc
http://www.vnunet.com/vnunet/news/2186476/intel-ships-first-olpc
http://blogs.guardian.co.uk/technology/archives/cat-103/
About the selling price by Intel:
http://arstechnica.com/news.ars/post/20060929-7864.html
http://www.bit-tech.net/news/2007/04/18/intel_classmate_pc_at_idf/
About the dumping:
http://money.cnn.com/2007/04/27/technology/fastforward_xo.fortune/index.htm
"The Classmate is essentially a standard but simplified desktop that like the XO is decorated in green plastic, but less fetchingly. Negroponte claims Intel is pricing it at $180, significantly below its cost of production, and for lots of only 10,000. In an e-mail, Intel spokesman Bill Calder responds that the Classmate's price is generally higher and he's not sure what Negroponte is referring to."
fans are bad. Mechanical parts under constant use are always the first to break. Tsk Tsk Intel. $ 1 billion couldn't get you a fanless CPU?
Hi everybody,
About the battery test I performed last night. I should insist that I was not very keen on doing a side by side comparison between a fully operational commercial product (Classmate PC) and a prototype (B2) with several pending issues and bugs. I'm confident that the super-smart team working on OLPC is going to fix ALL of them soon and I'm sure that will include battery life. So I just want to repeat here the same I said in my blog to target people working on the field (developing nations) working on the OLPC cause:
"This simple test should also help the OLPC advocates to realise that it is necessary to be extremely cautious about showing any prototype to educational authorities or technical government staff. I would not recommend giving a B1 or B2 unit for official testing purposes to any government authority at the moment. Those attempting to do such a thing can actually endanger the goals of this educational project."
Full test available here
http://luisramirez.cl/blog/?p=575
Winter wrote:
************************
About the dumping:
http://money.cnn.com/2007/04/27/technology/fast
forward_xo.fortune/index.htm
"...Negroponte claims Intel is pricing it at $180, significantly below its cost of production..."
*****************************
So, now it turns out that Negroponte's own baseless accusation is the only "evidence" you can produce!
Maybe the "evidence" is buried in one of the many meaningless links you provide. Why don't you provide ONE SINGLE LINK along with the corresponding text where this "evidence" lies?
I bet you can't do that, because there is no such thing. After all, Wayan asked the same question and all you could show is the same rumor that Charbax is spreading...
"Fans are bad. Mechanical parts under constant use are always the first to break. Tsk Tsk Intel. $ 1 billion couldn't get you a fanless CPU?"
Intel simply doesn't have the equivalent of the AMD Geode, meaning low power, low cost chips. As of now, Intel was not in the urge to make such chips, because they are not the one you can make a big profit. So making one would require significant investments, which Intel is not going to do.
Troy:
"So, now it turns out that Negroponte's own baseless accusation is the only "evidence" you can produce!"
So you know Negropontes accussations are false. Great, how do you know that? Show us where we can find evidence for THAT.
Intel never disputed this. They only said the price GENERALLY is higher.
This seems to be a clear case of Shoot the messenger. Instead of requiring evidence, for once, give us some for your own accussations. I really want to know what Intel is selling the Classmates for. What did Pakistan pay?
Winter
Winter wrote:
"So you know Negropontes accussations are false. Great, how do you know that? Show us where we can find evidence for THAT."
Don't be silly, Winter. You can't prove a negative. Everybody knows that.
If Negroponte makes the claim, HE is the one that must support it.
I'll give you an example, given that you don't "get it" that easily:
If I claim I have seen "polka-dot elephants", who has the burden of proof here?
Me?
or the people who say: 'Not true, Troy. there are no "polka-dot elephants"!'?
How could anyone prove that there are no "polka-dot elephants"?
get it now?
*************************
another way of answering your silly question:
I quote you, Winter:
"So you know Negropontes accussations are false. Great, how do you know that?"
Answer:
For the same reason I know your accusations are false: there no evidence you two are telling the truth.
Do you really feel that if you repeat your transparent little lies enough times, people will begin to belive them?
That's insane...
Does anyone know how many Watts per hour the Classmate uses? That would be useful to figure out how much energy you would have to draw from a solar panel or regular grid electricity to power it.
Also, it would be great to know how many Watts per hour the CPU uses.
Luis, did you ever get any info on what kind of learning apps Intel plans for the Classmate? The ones on skoool.es aren't bad but they are quite similar to a lot of material that already exists.
It would be interesting to know if Intel has something like eToys up their sleeve, though I highly doubt it.
Will,
I have done some research. Have a look at the tech specs sheet for the Intel® Mobile Processor ULV 900 MHz (90nm, code number 353) the Classmates uses:
http://download.intel.com/design/mobile/datashts/30311008.pdf
At pag. 64 you have listed the thermal design power for the all class of processors. The 353 has a TDP of 5.0W.
I admit that this is the best I got. Of course the processor Intel uses in the Classmate could be a specific version for it, with its own power optimization. I doubt though.
Nick
Winter & Troy:
While I love your energy and spirit in comments and guest posts, neither I nor other OLPC News readers want to wade through the line-by-line, almost word-by-word personal attacks
Rather than getting all up in each other's grill, please focus on one laptop per child and OLPC.
Thanks.
Wayan,
your request is fair enough. I have devoted enough energy to bringing light to many aspects of Prof. Negroponte's project. That's my contribution and hopefully others will continue contributing in their own manner.
Time to invest in other interests.
Good luck to all (yes, even to my "nemesis" Winter!).
:-)
Nick: thanks for the specs on the processor. 5W is a lot compared to .8 W for the XO. However, the XO has other processing elements such as the DCON chip, Marvell wireless controller, and others that chew up power. Still OLPC says the XO uses 2-5W per hour depending on usage mode. Can anyone independently verify this?
Also, does anyone know how much total power the Classmate uses?
I did some quick research on 6 cell Li-Ion batteries and they seem to range from 48-53 Watts per hours. Granted, this is not a very scientific method. Let's crudely estimate that since the Windows Classmate runs for 4 hours then it uses approximately 12-13 Watts per hour. That is a lot more than the 2-5 Watts OLPC reports.
Results for 6 Cell Li-Ion batteries
I know I will sound as a school teacher, but Watts are Joule per second. Watt-hours are times 3600 Joule. Watts per hour translates to an acceleration in energy ;-)
Battery capacity is given in ampere-hours (=Amp Hours=mA hours), which translates into an amount of charge (=Coulomb).
Winter
Here is the specification for the lap-top wiki for the battery capacity
Battery:
* Fully-enclosed “hard” case; user removable
* Pack type: 4 Cells, 6V series configuration
* Two chemistries:
o NIMH, with a capacity of 16.5 Watt-hours
o LiFeP, with a capacity of 22 Watt-hours
Many of the specifications for the XO are available in the OLPC wiki.
http://wiki.laptop.org/go/Hardware_specification
Here is the latest news about Intel from NY times:
SAN FRANCISCO (Reuters) - Intel Corp. (Nasdaq:INTC - news) detailed plans on Tuesday to team up with Asustek Computer Inc. (2357.TW), the world's largest maker of computer motherboards, to make a notebook PC that would cost as little as $200 aimed at the education market in developing countries.
Intel may consume a lot of power today (5W-10W). However, innovation is going to reduce the power consumption in the future. The OLPC is driving companies to compete and spurring innovations in low-power computation which is critical for developing countries. Is that not good?
Many of the comments in this forum remind me of the Apple-IBM wars of the 80s and 90s. I feel that we should discuss issues that are critical to education of children and I feel ultimately the best solution would be when all the players have interoperability with each other. Slamming any particular group (XO or Intel) is unproductive. There have been many innovations and there will be more. I do not represent either organization when I like to remind readers, many who were not born at that time, that Intel introduced one of the earliest of microprocessors in 1971.