Steve Jobs talks MacBook Air, China Mobile, Amazon Kindle, more

1235

Comments

  • Reply 81 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by G-News View Post


    You're not thinking far enough.

    Think along the whole way:

    Your computer, your modem, the next telephone node, the next central telephone node, the ISPs serverroom, another ISPs serverroom, googles mainframe, and then the whole way back. Not included are things like DNS resolution etc etc.

    That estimate was the result of a study, so it's certainly not off by orders of magnitude.

    The only thing you could argue is that a lot of power is consumed regardless of whether you actually perform the search or not. But the infrastructure to enable the search in the first place is consuming crazy amounts of power.

    I'll see if I can find a link.



    It still requires almost no power per individual search. Most of the power used in a computer are for fixed operations. The power supply, the mobo, the networking, the cpu's use a variable amount, but testing has shown that it's rarely more than 40% of the total. If a given server used 500 watts, how much is needed just for turn-on? About 250 watts. The rest is consumed by the cpu's, and memory at a variable rate. how many requests pass through that server at any given second? Can be thousands to tens of thousands, depending on how busy the server is.



    Larger servers may use four times as much power, but can serve more than ten times as many requests per second, hundreds of thousands.



    When you consider that a server can serve a million requests an hour (or ten million), even a thousand watts over than time results is very little power per request.



    As most of the areas you mentioned have fixed power requirements, even if nothing is being passed, the amount per transaction is still very low. Even if you add the fixed power uses, the number is still very low.



    For example, as you mentioned it, my internet gateway (DSL modem plus router) uses 10 watts. I use it all day, and even into the night. There are three of us here. That comes out to about 0.1¢ per hour, using the costs I gave earlier.



    There are thousands of requests I make a day, between having pages being served up, Google searches, e-mail, newsgroups, long downloads.



    Divide that 10 watts up between all of that (remembering that most of the power is being used for the router for our networking), and it's a miniscule amount of power. If it were jut a DSL modem, and not a networking device, it would use perhaps 2 watts. that would cost someone about 0.02¢ per hour to keep turned on and working.



    You have to figure out the percentage of time any of these devices are being used for the purposes you are talking about, and other purposes to get actual usage data.



    Otherwise, it wouldn't be economical.
  • Reply 82 of 108
    jeffdmjeffdm Posts: 12,951member
    Quote:
    Originally Posted by G-News View Post


    Some Links on the topic (German):

    http://www.sueddeutsche.de/computer/artikel/680/131447/

    they mention about 70W for a search



    http://www.innovativ-in.de/blog/2007...fur-das-klima/

    they mention 4Wh power consumption per search in the Google mainframe alone.



    http://perpenduum.com/2007/10/a-sing...s-in-one-hour/

    They mention 11Wh per search, so maybe I remembered it wrong, but it's still a hell lot of power, considering how many searches there are each day. and now if you consider how much power the entire internet uses...



    And here's another study: http://enterprise.amd.com/Downloads/...pletefinal.pdf



    And another article: http://blogs.sun.com/rolfk/entry/you...int_when_using



    Sure, maybe they are all not totally accurate, but when you start considering how much technological infrastructure is behind everything somehow related to the internet, then you just have to realize how much energy is being consumed.

    And sadly, a lot of that energy is wasted into heat too.



    It seems like a lot, but without figures on what it takes to print, bind and ship a book, it's just a one-sided non-comparison.
  • Reply 83 of 108
    g-newsg-news Posts: 1,107member
    Let me fill that gap, then:



    http://www.penguin.co.uk/static/cs/u...ndotohelp.html



    2.5kg of CO2 carbon footprint for a 500page paperback.

    That equals 368 google searches, assuming 6.8g per search.

    Assuming 0.2seconds avg per Google search, that 73.5 seconds.

    Assuming 400KB/sec download rate, that amounts to 29411KB.

    So, roughly speaking, downloading just 28.7MB of data has about the same CO2 footprint as a 500 page paperback.



    If that is anywhere near the actual figures, moving from books to ebooks is not going to benefit the environment anything at all. Because we haven't included the CO2 footprint of the eBook reader yet, nor have we included that devices power draw.

    q.e.d.
  • Reply 84 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by G-News View Post


    Some Links on the topic (German):

    http://www.sueddeutsche.de/computer/artikel/680/131447/

    they mention about 70W for a search



    http://www.innovativ-in.de/blog/2007...fur-das-klima/

    they mention 4Wh power consumption per search in the Google mainframe alone.



    http://perpenduum.com/2007/10/a-sing...s-in-one-hour/

    They mention 11Wh per search, so maybe I remembered it wrong, but it's still a hell lot of power, considering how many searches there are each day. and now if you consider how much power the entire internet uses...



    And here's another study: http://enterprise.amd.com/Downloads/...pletefinal.pdf



    And another article: http://blogs.sun.com/rolfk/entry/you...int_when_using



    Sure, maybe they are all not totally accurate, but when you start considering how much technological infrastructure is behind everything somehow related to the internet, then you just have to realize how much energy is being consumed.

    And sadly, a lot of that energy is wasted into heat too.



    Those kinds of numbers have been debunked over the years though.



    I remember studies (I would have to look for them, if I have time) that said that if those numbers were correct, even the lowest, the computer networks on the planet would be using more power than every other use put together. That obviously isn't happening. Even without the studies, you can look to see how many page requests are made around the world a day to see the numbers are far off, even using you own numbers for power.



    There is well over a trillion searches, network passes, etc every day. If each used 70 watts, that would be 70 trillion watts. That's a joke number.
  • Reply 85 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by G-News View Post


    Let me fill that gap, then:



    http://www.penguin.co.uk/static/cs/u...ndotohelp.html



    2.5kg of CO2 carbon footprint for a 500page paperback.

    That equals 368 google searches, assuming 6.8g per search.

    Assuming 0.2seconds avg per Google search, that 73.5 seconds.

    Assuming 400KB/sec download rate, that amounts to 29411KB.

    So, roughly speaking, downloading just 28.7MB of data has about the same CO2 footprint as a 500 page paperback.



    If that is anywhere near the actual figures, moving from books to ebooks is not going to benefit the environment anything at all. Because we haven't included the CO2 footprint of the eBook reader yet, nor have we included that devices power draw.

    q.e.d.



    That's not even close to being correct.
  • Reply 86 of 108
    g-newsg-news Posts: 1,107member
    why not? I used the numbers that are available in the links I provided.

    Assuming that they are even half way close, that estimate isn't so far off.



    Rethorically speaking "That's not even close to being correct." is a so called killer phrase.

    Your trying to kill the dialog without providing any information to back it up. If you don't like the idea of us all wasting tons of energy every year, that's fine with me, but that doesn't make my assumption any less valid.
  • Reply 87 of 108
    jeffdmjeffdm Posts: 12,951member
    Quote:
    Originally Posted by G-News View Post


    Let me fill that gap, then:



    http://www.penguin.co.uk/static/cs/u...ndotohelp.html



    2.5kg of CO2 carbon footprint for a 500page paperback.

    That equals 368 google searches, assuming 6.8g per search.

    Assuming 0.2seconds avg per Google search, that 73.5 seconds.

    Assuming 400KB/sec download rate, that amounts to 29411KB.

    So, roughly speaking, downloading just 28.7MB of data has about the same CO2 footprint as a 500 page paperback.



    If that is anywhere near the actual figures, moving from books to ebooks is not going to benefit the environment anything at all. Because we haven't included the CO2 footprint of the eBook reader yet, nor have we included that devices power draw.

    q.e.d.



    Even if that's true, I think most novels take less than 1MB of space. Once downloaded, it doesn't need to be re-downloaded.
  • Reply 88 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by G-News View Post


    why not? I used the numbers that are available in the links I provided.

    Assuming that they are even half way close, that estimate isn't so far off.



    Rethorically speaking "That's not even close to being correct." is a so called killer phrase.

    Your trying to kill the dialog without providing any information to back it up. If you don't like the idea of us all wasting tons of energy every year, that's fine with me, but that doesn't make my assumption any less valid.



    I gave you nformation. If you don't believe the daily internet traffic numbers, you can go look them up for yourself.



    It's a pretty simple step from there to multiply them by the 70 watt number you gave to get a total.



    It's had to find all the numbers sometimes, but this chart will show the numbers of peope nline daily across the world. Notice that the figures are outdated. The worldwide number is for 2005, and is considerably larger at the end of 2007.



    It doesn't give a number for actual page requests though. That's harder to find, though I have read it several times. but, if you can figure out some approximate number, as an average, we can make a guess.



    This doesn't tell how many are networked together, and passing this data, and that would add a great deal to the numbers.



    I'll see what else I can find.



    http://www.infoplease.com/ipa/A0933606.html
  • Reply 89 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by JeffDM View Post


    Even if that's true, I think most novels take less than 1MB of space. Once downloaded, it doesn't need to be re-downloaded.



    Most novels, I can say from experience, take less than 300 KB. They take so little time to download, I can't measure it (at 6.144 Kb/s).
  • Reply 90 of 108
    physguyphysguy Posts: 920member
    Quote:
    Originally Posted by G-News View Post


    Some Links on the topic (German):

    http://www.sueddeutsche.de/computer/artikel/680/131447/

    they mention about 70W for a search



    http://www.innovativ-in.de/blog/2007...fur-das-klima/

    they mention 4Wh power consumption per search in the Google mainframe alone.



    http://perpenduum.com/2007/10/a-sing...s-in-one-hour/

    They mention 11Wh per search, so maybe I remembered it wrong, but it's still a hell lot of power, considering how many searches there are each day. and now if you consider how much power the entire internet uses...



    And here's another study: http://enterprise.amd.com/Downloads/...pletefinal.pdf



    And another article: http://blogs.sun.com/rolfk/entry/you...int_when_using



    Sure, maybe they are all not totally accurate, but when you start considering how much technological infrastructure is behind everything somehow related to the internet, then you just have to realize how much energy is being consumed.

    And sadly, a lot of that energy is wasted into heat too.



    Practically all energy used to do work is 'wasted into heat' at the end. It did work getting there, moved an electron, flipped a bit, moved a car forward, moved your finger up and down. It's called the Second Law of Thermodynamics. What's your point? (BTW 70W and is not an amount of energy but power. IF 70W is correct and it only take 1 sec to do a search then the search only cost 0.02 WHr. IF a kWHr cost $0.25 (which I think is high) then the search only cost $0.000005)
  • Reply 91 of 108
    g-newsg-news Posts: 1,107member
    Where not talking about costs, but energy consumption.

    One of the articles I linked cites Google having a 1bio$ power bill each year.

    That's a lot of power.



    Also, this link might be of interest to you.

    http://hardware.slashdot.org/article.../09/27/2157230
  • Reply 92 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by G-News View Post


    Where not talking about costs, but energy consumption.

    One of the articles I linked cites Google having a 1bio$ power bill each year.

    That's a lot of power.



    Also, this link might be of interest to you.

    http://hardware.slashdot.org/article.../09/27/2157230



    The problem with all of these metrics is that they are so inaccurate as to be meaningless.



    Did you actually read what was said?



    Quote:

    The total includes the energy used by desktop computers and monitors (which makes up two-thirds of the total), plus other energy sinks including modems, routers, data processing equipment and cooling equipment."



    Do you realize that almost all of that equipment is primarily used for OTHER uses, and that being on the internet is just a matter of the fact that people are using those computers for work (one of the biggest percentages of where users go on the internet), and for other purposes at home, where the computer just happens to be on the internet because so many people now have broadband, so that from the first second they turn the machine on, it's on the internet, even if their browser is turned off?



    Data processing centers may have an internet connection, but the vast amount of power is not used for anything related to internet use.



    Going by that, I would say, that at the most, half of the power he quoted could be ascribed to internet use. I doubt if it's even that high, because a lot of the other equipment used is not for that purpose either, and he doesn't bother to break it out.



    That's why I can't stand these very unscientific "studies". there is NO standard. Everyone uses their own ideas as to what should be measured, how it should be measured, and even their own guesses as to how much power each piece of equipment is using.



    Where did he even get that number from?
  • Reply 93 of 108
    jeffdmjeffdm Posts: 12,951member
    Quote:
    Originally Posted by physguy View Post


    Practically all energy used to do work is 'wasted into heat' at the end. It did work getting there, moved an electron, flipped a bit, moved a car forward, moved your finger up and down. It's called the Second Law of Thermodynamics. What's your point? (BTW 70W and is not an amount of energy but power. IF 70W is correct and it only take 1 sec to do a search then the search only cost 0.02 WHr. IF a kWHr cost $0.25 (which I think is high) then the search only cost $0.000005)



    I think G-News' point was to try to show that it takes more power to get an eBook than it does to cut trees, make paper, print, bind and ship a paper book. I missed the link that gave figures on how much power paper books take.
  • Reply 94 of 108
    g-newsg-news Posts: 1,107member
    My original point was that the infrastructure that would be required to replace worldwide paperbook production with ebooks, servers serving the data and readers displaying them would likely be less environmentally friendly, than the paper books are now. That's my whole point and I think I have shown enough clues (I'm not calling it evidence), that this assumption is not totally taken out of thin air.
  • Reply 95 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by G-News View Post


    My original point was that the infrastructure that would be required to replace worldwide paperbook production with ebooks, servers serving the data and readers displaying them would likely be less environmentally friendly, than the paper books are now. That's my whole point and I think I have shown enough clues (I'm not calling it evidence), that this assumption is not totally taken out of thin air.



    no, it's not out of thin air, but the numbers are so skewed.



    Making a book is a very environmentally negative process. The making of paper is VERY pollution heavy. Some of the biggest polluters are paper manufacturers. If you put that into your equation it tips the balance way over. It's also, other than the effects on our water supply, a very energy intensive process. Logging and transporting those many thousands of tons of wood to the paper maker uses a vast amount of energy. Most of the logs are wasted for paper, and so must then be converted to other purposes, or destroyed.



    Printing presses can be very expensive for large publishing houses. They also use a large amount of electricity.



    Packaging those books again uses a lot of paper. Transporting those books uses a lot of oil. Then the books that don't sell are shipped back to the publisher, so they can destroy them. More energy used.



    This is a VERY energy intensive business.



    Whereas, the incremental energy use from downloading books is miniscule.
  • Reply 96 of 108
    Quote:
    Originally Posted by melgross View Post


    no, it's not out of thin air, but the numbers are so skewed.



    Making a book is a very environmentally negative process. The making of paper is VERY pollution heavy. Some of the biggest polluters are paper manufacturers. If you put that into your equation it tips the balance way over. It's also, other than the effects on our water supply, a very energy intensive process. Logging and transporting those many thousands of tons of wood to the paper maker uses a vast amount of energy. Most of the logs are wasted for paper, and so must then be converted to other purposes, or destroyed.



    Printing presses can be very expensive for large publishing houses. They also use a large amount of electricity.



    Packaging those books again uses a lot of paper. Transporting those books uses a lot of oil. Then the books that don't sell are shipped back to the publisher, so they can destroy them. More energy used.



    This is a VERY energy intensive business.



    Whereas, the incremental energy use from downloading books is miniscule.



    Convert to HEMP over Wood chip pulp production and recycle all Pulp into Composite wood boards.
  • Reply 97 of 108
    melgrossmelgross Posts: 33,510member
    Quote:
    Originally Posted by mdriftmeyer View Post


    Convert to HEMP over Wood chip pulp production and recycle all Pulp into Composite wood boards.



    HEMP!



    Sure, that'll be the day.



    But even in the unlikely case they did, there would still be problems. besides, hemp wouldn't make very good paper or cardboard . It doesn't have the proper cellular structure.



    They want to make it into clothing. That would work, if someone didn't smoke it first.
  • Reply 98 of 108
    g-newsg-news Posts: 1,107member
    I guess that estimate of 2.5kg of CO2 per 500page copy includes logging, paper, printing and distribution. I can only assume that, but 2.5kg is a rather high mark already. Needless to say, the CO2 footprint of a kindle is probably closer to 500kg. (as are, unfortunately, most electronic devices).
  • Reply 99 of 108
    physguyphysguy Posts: 920member
    Quote:
    Originally Posted by G-News View Post


    I guess that estimate of 2.5kg of CO2 per 500page copy includes logging, paper, printing and distribution. I can only assume that, but 2.5kg is a rather high mark already. Needless to say, the CO2 footprint of a kindle is probably closer to 500kg. (as are, unfortunately, most electronic devices).



    Even if both numbers are simply accepted it is still no contest. Something like the Kindle still wins hands down because it is reusable, unlike the book - recycling still uses energy. Don't forget a device like the kindle not only replaces many books (200 gets you to parity) but you have used fewer newspapers, magazines, etc. etc. A kindle device (or a computer) is a much more efficient distribution system for information.



    These arguments are akin to those that argue that burning ethanol is 'carbon neutral'. Nothing could be more incorrect. The 'carbon neutral' argument is that since the plant (lets assume corn for convenience but the plant is immaterial) that produced the ethanol captured the carbon from the atmosphere then putting it back is neutral. This is an artificial linking of the two processes. There are two paths.



    1)

    -grow corn - capture carbon

    - burn corn - release carbon



    2)

    -grow corn - capture carbon

    - leave corn on/in the ground - soil/fertilizers - carbon still captured

    - burn oil - release carbon



    Same net results (approximately) for the two paths.



    Ethanol is important as a renewable source, not as a carbon-free source.



    There are only three sources of energy on the earth period.



    1) Solar

    - Oil captured it millions of years ago

    - Ethanol, food, etc. - captured it weeks ago

    - 'Solar' power captured it moments ago as did wind, etc.

    2) Nuclear

    - Fission - captured in some supernova eons ago

    - fusion - captured in the Big-Bang

    3) Gravity

    - Tidal

    - Geo-thermal



    Cheers
  • Reply 100 of 108
    Jobs is extremely right on Kindle. Kindle is too ugly to be called a success.



    Sachin Dhall
Sign In or Register to comment.