How Best to enlarge a photo Huge!
Nov 1, 2013 03:30:33 #
bull drink water wrote:
to those who know more than me. would shooting in raw and converting it to tiff make a difference?
Yes. So would shooting it with a medium format digital or a D800. The more pixels in the original the less enlargement needed to get to where you want to be.
Nov 1, 2013 07:46:00 #
RichieC
Loc: Adirondacks
bull drink water wrote:
to those who know more than me. would shooting in raw and converting it to tiff make a difference?
Raw is 12 or 14 bit... every pixels' color is described with 12-14 decimal places... tif or jpeg use just 8. That means even though the pixels look the same color, there may be tiny differences. Raw is a logarithmic scale- 4 times the resolution of 8 bit... This is per channel- so final figure multiplied again by three ( RGB Channels). - the difference in image information is profound. _ well depends if you captured it all... 14 bit photo of black will return a black photo. LOL
When enlarging, you'll use every "bit" of this info- pardon the pun. It will (should) return a better result. It will return the best result possible.. lets say it that way.
I am interested in seeing this wall!
Nov 1, 2013 08:01:57 #
rebride wrote:
To judge these don't all need to be cropped the same?
The sections are shown at 100% to show difference. So I think that makes for a fair comparison.
To answer another question, using the RAW would result in just a little improvement. For the sake of my time, I just used the JPEG SOOC to illustrate the differences in magnification and software.
Nov 1, 2013 10:03:42 #
FYI 14 bits is equivalent to 2 to the 13th power or 8192. This is equivalent to only 8 decimal places, not 14.
1/8192=0.00012207
1/8191=0.00012209
1/8192=0.00012207
1/8191=0.00012209
Nov 1, 2013 10:42:49 #
Nov 1, 2013 12:21:58 #
RichieC
Loc: Adirondacks
steveg48 wrote:
FYI 14 bits is equivalent to 2 to the 13th power or 8192. This is equivalent to only 8 decimal places, not 14.
1/8192=0.00012207
1/8191=0.00012209
1/8192=0.00012207
1/8191=0.00012209
You are very technically correct... I forgot the binary system of 1- 20. However the premise is the same.
8 bit returns 256 possible tones- per channel ( This is JPEG )
12 bit returns 4,000 possible tones per channel
14 bit returns 16,0000 possible tones per channel
An 8-bit jpeg can represent around 16 million colors, whereas a high bit-depth file can represent over 28 billion
However,,, lol thank you for reminding me... studied it 30 years ago when I took Fortran and Pascal... . I do strive to be accurate.
Nov 1, 2013 12:55:04 #
Talk to billboard/sign companies. They do this all the time and may have some suggestions for sources.
Naturally, their work is not for close up viewing, but somewhere between 8X10 inches, and a 20 X40 foot bill board is a wall size solution.
After all, you can buy wall size murals all over the internet that have dimensions for walls,,, perhaps a printer who specializes in such things?
Naturally, their work is not for close up viewing, but somewhere between 8X10 inches, and a 20 X40 foot bill board is a wall size solution.
After all, you can buy wall size murals all over the internet that have dimensions for walls,,, perhaps a printer who specializes in such things?
Nov 1, 2013 12:57:07 #
steveg48 wrote:
FYI 14 bits is equivalent to 2 to the 13th power or 8192. This is equivalent to only 8 decimal places, not 14.
1/8192=0.00012207
1/8191=0.00012209
1/8192=0.00012207
1/8191=0.00012209
Are you allowing for the Giffengander and Hoofenschnoffel differentials? What about the Ghergenschlam hypothesis?
Sheldon C
Head Honcho... BBT
Nov 1, 2013 12:58:29 #
RichieC wrote:
Figure out what the magnification will be to achie... (show quote)
Perhaps the North wall of an East/West Gymnasium....
It's been said on this forum and many others, proper viewing distance is the diagonal dimension of the image. Pixel Peekers perpetually deny this "suggestion".
:thumbup:
Nov 1, 2013 13:17:32 #
RichieC
Loc: Adirondacks
Kuzano wrote:
Are you allowing for the Giffengander and Hoofenschnoffel differentials? What about the Ghergenschlam hypothesis?
Sheldon C
Head Honcho... BBT
Sheldon C
Head Honcho... BBT
AssH_le..... ROFL!
Nov 1, 2013 20:26:18 #
Nov 2, 2013 08:04:06 #
I used to use Genuine Fractals, many years ago that worked well. I think it is now Perfect Resize !
Nov 2, 2013 08:31:20 #
Kuzano wrote:
Are you allowing for the Giffengander and Hoofenschnoffel differentials? What about the hypothesis?
Sheldon C
Head Honcho... BBT
Sheldon C
Head Honcho... BBT
:cry: Her Voltstickum disproved the Ghergenschlam Hypothesis last year. Chees it really sucks when people cannot stay up on their math. Next thing you know people will think there is free health insurance
:thumbup: :roll:
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.