Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.compression > #2131 > unrolled thread

let me make it simple

Back to article view | Back to comp.compression

Contents

  let me make it simple Michael Harrington <michaelharrington4rep@gmail.com> - 2013-11-14 07:20 -0800
    Re: let me make it simple Robert Wessel <robertwessel2@yahoo.com> - 2013-11-14 15:26 -0600
      Re: let me make it simple Michael Harrington <michaelharrington4rep@gmail.com> - 2013-11-14 18:54 -0800
        Re: let me make it simple pfraser <pete_fraser@comcast.net> - 2013-11-14 19:54 -0800
          Re: let me make it simple Michael Harrington <michaelharrington4rep@gmail.com> - 2013-11-14 20:44 -0800
          Re: let me make it simple Michael Harrington <michaelharrington4rep@gmail.com> - 2013-11-14 20:58 -0800
        Re: let me make it simple Robert Wessel <robertwessel2@yahoo.com> - 2013-11-15 01:09 -0600
      Re: let me make it simple Michael Harrington <michaelharrington4rep@gmail.com> - 2013-11-14 18:56 -0800
        Re: let me make it simple Thomas Richter <thor@math.tu-berlin.de> - 2013-11-15 08:59 +0100
          Re: let me make it simple Michael Harrington <michaelharrington4rep@gmail.com> - 2013-11-15 00:41 -0800
            Re: let me make it simple pfraser <pete_fraser@comcast.net> - 2013-11-15 07:28 -0800
    Re: let me make it simple Thomas Richter <thor@math.tu-berlin.de> - 2013-11-15 08:51 +0100
    Re: let me make it simple Fibonacci Code <anglikai@gmail.com> - 2013-11-16 22:31 -0800

#2131 — let me make it simple

Inch2 to micron2

If color spots are 100 microns big and encode 23 bits data then this is 108 terabytes in data hard-encoded.


Since the current tech is I think 1.2 terabyte per square inch.... does this mean I win a free peer review?

[toc] | [next] | [standalone]

#2133

On Thu, 14 Nov 2013 07:20:36 -0800 (PST), Michael Harrington
<michaelharrington4rep@gmail.com> wrote:

>Inch2 to micron2
>
>If color spots are 100 microns big and encode 23 bits data then this is 108 terabytes in data hard-encoded.
>
>
>Since the current tech is I think 1.2 terabyte per square inch.... does this mean I win a free peer review?


Assuming you mean spots of 100 square microns, you'd have some 6.5
million spots per square inch.  At 23 bits per spot, you'd get about
18.5 *mega*bytes per square inch.  So your density number appears to
be off by a factor of about six million.

[toc] | [prev] | [next] | [standalone]

#2134

Hmmm, odd

https://www.google.com/search?hl=en-US&ie=UTF-8&source=boatbrowser&q=1+square+inch+to+square+micron

So there is 645 million microns2 to a inch2. According to google and other sources. Thus total microns is quite a large number.




Now 10 million colors to be answered:

http://ux.stackexchange.com/questions/30127/monitors-display-more-colors-than-human-eye-can-distinguish



A kilobyte is 1024 bytes, a mega is 1024kb, a giga is 1024mb, a tb is 1024gb

So if color is 24 bits per "spot" and if each spot is 100x100 microns our math is as follows

645160000*645160000 = total microns
100*100 = color spot
23 bits per color spot

Here is all the math except the answer. How many bits or bytes do I have the potential for?

[toc] | [prev] | [next] | [standalone]

#2136

Michael Harrington wrote:
> Hmmm, odd
>
> https://www.google.com/search?hl=en-US&ie=UTF-8&source=boatbrowser&q=1+square+inch+to+square+micron
>
> So there is 645 million microns2 to a inch2. According to google and other sources. Thus total microns is quite a large number.

Seems about right.

> So if color is 24 bits per "spot" and if each spot is 100x100 microns our math is as follows
>
> 645160000*645160000 = total microns

You have squared it already -- you can't square it again.

> 100*100 = color spot
> 23 bits per color spot
>
> Here is all the math except the answer. How many bits or bytes do I have the potential for?
>

You have 6.45 * 10^8 u^2 / inch^2

Your spot is 100 x 100 u = 10^4 u^2

So you get 6.45 * 10^4 spots / in^2

Each spot is 24 bits or 3 Bytes.
So you get just under 200 kBytes / in^2
That would be an appallingly low density, even if you could do it.
The problems involved in reading the color quickly, cheaply, and 
consistently would not be trivial.

[toc] | [prev] | [next] | [standalone]

#2137

Your right, i squared a square. 

3d to 2d mistake.

But let me draw up full extent math and see where it goes.

Sigh, lack of sleep ftl

[toc] | [prev] | [next] | [standalone]

#2138

Your right, i squared a square. 

3d to 2d mistake.

But let me draw up full extent math and see where it goes.

Sigh, lack of sleep ftl

[toc] | [prev] | [next] | [standalone]

#2139

On Thu, 14 Nov 2013 18:54:06 -0800 (PST), Michael Harrington
<michaelharrington4rep@gmail.com> wrote:

>Hmmm, odd
>
>https://www.google.com/search?hl=en-US&ie=UTF-8&source=boatbrowser&q=1+square+inch+to+square+micron
>
>So there is 645 million microns2 to a inch2. According to google and other sources. Thus total microns is quite a large number.
>
>
>
>
>Now 10 million colors to be answered:
>
>http://ux.stackexchange.com/questions/30127/monitors-display-more-colors-than-human-eye-can-distinguish
>
>
>
>A kilobyte is 1024 bytes, a mega is 1024kb, a giga is 1024mb, a tb is 1024gb
>
>So if color is 24 bits per "spot" and if each spot is 100x100 microns our math is as follows
>
>645160000*645160000 = total microns
>100*100 = color spot
>23 bits per color spot
>
>Here is all the math except the answer. How many bits or bytes do I have the potential for?


I originally had a sentence mentioning that if your spots were, in
fact, 100 microns square (rather than 100 square microns), you needed
to reduce the density number by an additional factor of 100, but I
assumed you would not have made that basic an error.

In any event, 100 microns square leaves you with 185KB, which is the
same number pfraser mentioned.

In any event, for persistent external storage the convention is to use
the decimal values of the SI prefixes (IOW, a 100GB disk drive stores
100,000,000,000 bytes).  Not that the resulting 11% error makes any
real difference here.

[toc] | [prev] | [next] | [standalone]

#2135

So this shows where you made your error. you are off by a measure of 6 million.



Can I has a peer review yet?

I mean seriously.... just email me your phone, we talk for 15 minutes, and then you do your omg and we get past this stuff.

Michaelhh@gmail.com

[toc] | [prev] | [next] | [standalone]

#2141

Am 15.11.2013 03:56, schrieb Michael Harrington:

>
> I mean seriously.... just email me your phone, we talk for 15 minutes, and then you do your omg and we get past this stuff.

Are you serious? Why don't you just take the 15 minutes to check your 
math and get some *real* numbers, and why should I waste such 15 minutes?

Be serious. What you describe here is essentially a photographic sensor 
you find in every digital camera. 8 bits per sample, 24 bits per cell. 
You get realistic sizes for sensors and realistic resolutions for 
sensors by googling a bit. We are probably around 16MPixel (just to give 
you some magnitude) right now, sometimes more, sometimes less. 16 
megapixel makes 16 megabyte (due to the Bayer-pattern, there is only one 
color per pixel, not three). That is *not* an impressive density for 
storage. Dynamic RAM has much more than that. It stores 8GByte on a 
smaller area.

Just to get your numbers straight. It's nothing you cannot research by 
five minutes on the internet using google or wikipedia.

[toc] | [prev] | [next] | [standalone]

#2143

I will do a comprehensive write up on my 3 tier system. including math proofs and walkthroughs. give me a few days.

[toc] | [prev] | [next] | [standalone]

#2144

Michael Harrington wrote:
> I will do a comprehensive write up on my 3 tier system. including math proofs and walkthroughs. give me a few days.
>
We just destroyed your math proof.
You admitted as much yourself.

Why continue?

[toc] | [prev] | [next] | [standalone]

#2140

Am 14.11.2013 16:20, schrieb Michael Harrington:
> Inch2 to micron2
>
> If color spots are 100 microns big and encode 23 bits data then this is 108 terabytes in data hard-encoded.


One inch is 2.54 cm (why can't you just use the metric system?). One 
micron is 1/1000th of a millimeter, 100 micron thus 0.1mm, which gives 
you 254 dots per micron or approximately 2^16 dots on a square inch. For 
easier calculation, just say 32 bits per dot, thus 32 * 2^16 bits a 
square inch, thus 2^21 bits = 2^18 bytes = 2^8 KByte = 256 KByte a 
square inch. I don't call this impressive.

> Since the current tech is I think 1.2 terabyte per square inch.... does this mean I win a free peer review?

No, it means you win a free course in basic algebra. Seriously, if you 
don't take five minutes time to check your calculation, why should I?

[toc] | [prev] | [next] | [standalone]

#2147

Do you see, it doesn't matter it is colour or black and white, colour is just waves, you just "name" it colour that is all. Long long ago, people already store audios and visual in analogue way, that is waves. Before digital technology comes and replace them. Now you say you want to go back to "colour". The bit density of current HDD is in constant improvement. I don't think your "colour" can keep up to that.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.compression

csiph-web