Olympus-OM
[Top] [All Lists]

[OM] Re: Digital vs film resolution

Subject: [OM] Re: Digital vs film resolution
From: "Matthew Bristol" <bristolm@xxxxxxxxxxxxx>
Date: Thu, 9 Mar 2006 08:44:31 -0600
I think C.H. Ling was right.

Not to put words in his mouth, but I'm guessing James' logic was:
1 binary film grain = 1 bit
8 binary film grains = 8 bits
2^8 = 256

But, there are many combinations of film grain values that give the same
level value, so you don't get 256 discrete levels.

e.g.
film grain      1       2       3       4       5       6       7
8       level
grain value             1       0       1       1       0       1
0       1       5
grain value             0       1       1       0       1       0
1       1       5
grain value     1       0       1       1       1       0       1
0       5

The total number of discrete values that could be represented by 8
grains would be 8.  So, to allow a "clump" to have a total of 256
different available values, one would need 256 grains.

The clumps are not discrete entities on the film though.  One could
choose any 256 adjacent film grains, so a film grain could be present in
many different clumps, each clump giving different level values.  You
couldn't just say:

Number of film grains/256 = megapixel equivalent

The "megapixel" equivalent for film should then be the number of
different clumps one could choose, which should be very close to the #
of film grains.  The argument against this though, is that each "clump"
is not independent from overlapping clumps, and would give a "blending"
effect by this analysis, possibly accounting for the apparent lower
sharpness in film.

-Matt



-----Original Message-----
From: olympus-owner@xxxxxxxxxx [mailto:olympus-owner@xxxxxxxxxx] On
Behalf Of C.H.Ling
Sent: Friday, March 10, 2006 12:16 AM
To: olympus@xxxxxxxxxx
Subject: [OM] Re: Digital vs film resolution


Your calculation on the film grains seems not right, if all grain has
the 
same size you need 256 (16x16) grains to represent 256 levels. But film 
grains has variable size so you never know.

C.H.Ling


----- Original Message ----- 
From: "james king"

>
> I am having real problems with this argument about sensor spacing vs
> grain size.
> firstly since digital sensors see in black and white and require a
> coloured filter to see in colour you need 3 sensors to get colour not
> one. Secondly you need an antialias filter. thirdly there is space
> between each sensor which can therefore record nothing at all.
> *assuming* film grain is binary then you only need 8 grains to
simulate
> 256 combinations and 24 grains to get 24 bit rgb. However film is
> layered (unlike digital sensors) therefore you only need 8 grains in
> three layers ontop of each other to get 24 bit rgb whereas digital
> requires three sensors.
> my 2pence worth
> James


==============================================
List usage info:     http://www.zuikoholic.com
List nannies:        olympusadmin@xxxxxxxxxx
==============================================
This message (and any associated files) is intended only for the use of the 
individual or entity to which it is addressed, and may contain information that 
is confidential, subject to copyright or constitutes a trade secret. If you are 
not the intended recipient, you are hereby notified that any dissemination, 
copying or distribution of this message or files associated with the message 
are strictly prohibited. If you have received this message in error, kindly 
notify us immediately by replying to the message and deleting it from your 
computer
==============================================
List usage info:     http://www.zuikoholic.com
List nannies:        olympusadmin@xxxxxxxxxx
==============================================

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz