[9fans] I appear to have stumped the brains at chacha.com

[9fans] I appear to have stumped the brains at chacha.com

[ron minnich]

a simple question: how many microrprocessors are there in the world?

Let's restrict to 32 bits +mmu

Anybody want to guess? I can guess, but I am sure I am wrong.

Don't forget to count all the camcorders.

ron

Re: [9fans] I appear to have stumped the brains at chacha.com

[andrey mirtchovski]

is this a google interview question? i love those........ NOT!

first thing to do is to limit the number from above, so at least we
know how large the playing field is:

assume a microprocessor is 1 cm³ on average. the slabs of silicone
used to make such processors are 1 m³  in volume, so one of those can
make 1 000 000 (10⁶) microprocessors. to make a single slab of
silicone 453kg of raw manufacturing-grade gravel (sand) is used,
together with 250kg of coal and other carbons.
(http://www.answers.com/topic/silicon?cat=health).

next we need to find how much sand there is in the world, and knowing
that we have certainly not exhausted all of it to make silicone (there
are still sandy beaches in some places of the world) we can calculate
an upper limit for the number of microprocessors in existence:

sand grains are 0.1mm across (on average), that means there are 1
trillion (10¹²) grains of sand in a cubic meter. the number of grains
of sand on all earth's beaches has been estimated at 10²⁴ that 10 to
the power of 24, (source: http://www.astro.utu.fi/~cflynn/sand.html),
so about 10¹² m³.

sand weighs at 1 500-1 800kg per m³
(http://www.answers.com/topic/sand?cat=health), which gives us about
1.65*10¹⁴kilograms of sand. that makes (combined with coal) roughly
about 4*10¹¹ slabs, which, at 10⁶ microprocessors per slab, gives us a
maximum number of microprocessors that can exist on an Earth with
observable sandy beaches to be 4*10¹⁷

although i'm much more inclined to just start counting from 1 when i'm
asked this type of questions.

ps: limiting that number from below is left as an exercise to the reader :)

Re: [9fans] I appear to have stumped the brains at chacha.com

[Francisco J Ballesteros]

Are you sure you do not reach a limit before when you calculate how many
cores are needed to cover the earth surface? (on average)

Or... when you calculate how much power is required to power all of them at
the same time?



On 10/16/07, andrey mirtchovski <mirtchovski@gmail.com> wrote:
> is this a google interview question? i love those........ NOT!
>
> first thing to do is to limit the number from above, so at least we
> know how large the playing field is:
>
> assume a microprocessor is 1 cm³ on average. the slabs of silicone
> used to make such processors are 1 m³  in volume, so one of those can
> make 1 000 000 (10⁶) microprocessors. to make a single slab of
> silicone 453kg of raw manufacturing-grade gravel (sand) is used,
> together with 250kg of coal and other carbons.
> (http://www.answers.com/topic/silicon?cat=health).
>
> next we need to find how much sand there is in the world, and knowing
> that we have certainly not exhausted all of it to make silicone (there
> are still sandy beaches in some places of the world) we can calculate
> an upper limit for the number of microprocessors in existence:
>
> sand grains are 0.1mm across (on average), that means there are 1
> trillion (10¹²) grains of sand in a cubic meter. the number of grains
> of sand on all earth's beaches has been estimated at 10²⁴ that 10 to
> the power of 24, (source: http://www.astro.utu.fi/~cflynn/sand.html),
> so about 10¹² m³.
>
> sand weighs at 1 500-1 800kg per m³
> (http://www.answers.com/topic/sand?cat=health), which gives us about
> 1.65*10¹⁴kilograms of sand. that makes (combined with coal) roughly
> about 4*10¹¹ slabs, which, at 10⁶ microprocessors per slab, gives us a
> maximum number of microprocessors that can exist on an Earth with
> observable sandy beaches to be 4*10¹⁷
>
> although i'm much more inclined to just start counting from 1 when i'm
> asked this type of questions.
>
> ps: limiting that number from below is left as an exercise to the reader :)
>

Re: [9fans] I appear to have stumped the brains at chacha.com

[Skip Tavakkolian]

> is this a google interview question? i love those........ NOT!
>
> assume a microprocessor is 1 cm³ on average. the slabs of silicone
> used to make such processors are 1 m³  in volume, so one of those can
> make 1 000 000 (10⁶) microprocessors. to make a single slab of
> silicone 453kg of raw manufacturing-grade gravel (sand) is used,
> together with 250kg of coal and other carbons.

is that why silicon wafers are shaped like manhole covers?

Re: [9fans] I appear to have stumped the brains at chacha.com

[erik quanstrom]

>> is this a google interview question? i love those........ NOT!
>>
>> assume a microprocessor is 1 cm³ on average. the slabs of silicone
>> used to make such processors are 1 m³  in volume, so one of those can
>> make 1 000 000 (10⁶) microprocessors. to make a single slab of
>> silicone 453kg of raw manufacturing-grade gravel (sand) is used,
>> together with 250kg of coal and other carbons.
> 
> is that why silicon wafers are shaped like manhole covers?

stolen from russ' links:  scroll past the german:

http://www.hebig.org/blogs/archives/main/000962.php

Re: [9fans] I appear to have stumped the brains at chacha.com

[ron minnich]

You got further than chacha:

Status: Connecting ...
Status: Looking for a guide ...
Looking: ...
Status: Jacqueline has connected to help you with your search on how
many microprocessors are there in the world. Please wait while your
guide searches for your results.
Jacqueline: Welcome to ChaCha!
You: thanks
Jacqueline: Hello! Thank you for coming to ChaCha
Jacqueline: One moment please
You: fyi there is an ad in top right of this web page but in firefox
it is cut off, shows bumper sticker "My town's news is better than
yours", but you can't see the name of the newspaper :-)
You: you might want to let the ad guys know :-)
Jacqueline: sorry?
Jacqueline: Are you trying to find an exact figure?
You: something close. Is it more like 10 billion or 1 billion or
Jacqueline: I am not finding an exact number
You: I am no longer sure.
Jacqueline: I am going to transfer you
Jacqueline: One moment please
You: I think an exact number is totally impossible.
Transfer: You are being transferred to another guide who can help you
search even better!
Status: Looking for a guide ...
Status: Cecil has connected to help you with your search on how many
microprocessors are there in the world. Please wait while your guide
searches for your results.
Cecil: Hello!
Cecil: I'm searching for your results.
You: thanks
You: This is actually a hard one.
You: To get the right number we'd have to count watches, cars,
microwaves, DVD players, ipods, it just goes on forever.
Cecil: Thanks for being patient!
You: I know I can't figure it out :-)
Cecil: Rest assured we will find your results!
Cecil: I will need to transfer you.
You: ok
Cecil: Thank you for using ChaCha!
Status: Your ChaCha guide was Cecil ID: 82277
Status:

That past tense on the status line? they dropped me. So much for
guided search. Google might not need to give up just yet.

Anyways, I am gonna guess 1B 32-bit processors in current use, which
is pretty conservative: iPods alone are 110M units. It still makes for
nice graph, starting at 1955 (or so) with 1.

ron

Re: [9fans] I appear to have stumped the brains at chacha.com

[erik quanstrom]

> ps: limiting that number from below is left as an exercise to the reader :)

my feynmann, we can further limit this number from above by noting
that we likely do not have 10^8 processors/person.  100 seems like
a reasonable upper bound.  that would leave us with a "mere" 10^11
processors tops.

:-)

- erik

Re: [9fans] I appear to have stumped the brains at chacha.com

[ron minnich]

btw, speaking of orders of magnitude:

big cluster in 1990: 4 nodes
big cluster in 2005: 4096 nodes

big "supercomputer" in 1992: 256 CPUs
big "supercomputer" in 2008: 256K CPUs

so, 10 bits in about 15 years in each case. You can fill in the data
points, and project and look at what people are doing and, guess what,
no sign it will stop. There's a cheat here: that big "supercomputer"
is multicore, but still ...

These aren't Google systems. These are systems that are "tightly
coupled", i.e. run one app and only run as fast as the slowest
computer and the longest network latency will allow. Network latency
is measured in units of hundreds of nanoseconds. At any given time,
there's only one  or two of these being built (that we know about).

ron

Re: [9fans] I appear to have stumped the brains at chacha.com

[Skip Tavakkolian]

> a simple question: how many microrprocessors are there in the world?
> 
> Let's restrict to 32 bits +mmu

assuming you mean real not virtual processors.

World population is 6.5B.  Optimistically 25% can afford any
electronics with 32bit processors (i.e.  percentage not preoccupied
solely with survival and bare necessities).  Roughly 1.6B. Some have
many (pc, laptop, pda, camera, mobiles); if on the average each person
has 2 - including those collecting dust in the closet - that's 3.2B.
It jibes with the estimated total number of mobile subscribers
(about 2.6B) worldwide.

wouldn't manufacturers keep track of unit production?

Re: [9fans] I appear to have stumped the brains at chacha.com

[Bakul Shah]

> a simple question: how many microrprocessors are there in the world?

Dead or alive?  In use or ever made?

Why do you want to know?

According to ITU the world had 640M PCs in 2003 and 772M PCs
n 2004.  But I think their numbers are conservative.  India
for instance grew from 9.3M to 13M in a year and they both
seem rather low.  China grew from 50M to 53M and the growth
rate seems very low.  Elsewhere I read China's computer
numbers double every two years.  And I am sure so does
India's.

http://www.itu.int/ITU-D/ict/statistics/at_glance/internet04.pdf

Though overall 1 computer for every 8 or 10 people "feels"
about right.

> Let's restrict to 32 bits +mmu

> Anybody want to guess? I can guess, but I am sure I am wrong.
> 
> Don't forget to count all the camcorders.

Camcorders have MMUs these days?

Probably there are more cpus in cellphones than anything else
but no mmu for most of them.

Perhaps chacha is going senile....

Re: [9fans] I appear to have stumped the brains at chacha.com

[Jack Johnson]

On 10/16/07, Bakul Shah <bakul+plan9@bitblocks.com> wrote:
> > a simple question: how many microrprocessors are there in the world?
>
> Dead or alive?  In use or ever made?

If I were counting, I'd be thinking still in existence, but not
necessarily in use. So, that magic line from 1955 to the present, some
reasonable portion under the curve.

I think the MMU is the interesting limiter.  Normally I would have
done something like:

% of cell phones + % of cell phone users with a DVD player + % of cell
phone users with a computer + % of cell phone users with a digital
cable or satellite receiver + most of the trash + some fudge factor

Though, if you could calculate the average number of microprocessors
per person by nation (or by their nation's GDP, or some other easy
extrapolation).

-Jack