What size to down-size to?

drcarldrcarl Registered Users Posts: 104 Major grins
edited September 7, 2007 in Finishing School
Hello Magical Forum - Answer Portal to All Questions!


I am making a 16x20 (or maybe even 20x24) collage of some party shots.

I'll be playing with perhaps 24 (or more) images which are now some 2-3 MB each in size.

Since my RAM is limited, I wonder how small I might make these files (so that I can move them around, arrange, and add drop shadows to them before flattening) and still be very "safe" as to the final image quality.

I bet I don't have to use the original file size...I just don't know how small I can go, or how to properly get there...

Can anyone tell me the proper steps to make a file smaller without sacrificing too much quality? I'm not real sure about the effects of checking or unchecking "resampling", or what the right order is when I "save as"

TIA

dr carl

Comments

  • Art ScottArt Scott Registered Users Posts: 8,959 Major grins
    edited September 7, 2007
    When reducing the size of the pic (not pixel dimension) I never check resample or scale style....never never and this was advice given to me by a photoshop professor....only tic the constrain proportions.......then I usually just change my dpi to get desired pic size...ie 300 dpi for approx 8 x 10.....600dpi for approx 4 x 5.........I only have 512 ram(and a 300gb scratch disk) myself and on large projects I just it bogg down my machine...it is all I have to work with and rather than sacrifice project quality I live with the slowness (for now)......
    "Genuine Fractals was, is and will always be the best solution for enlarging digital photos." ....Vincent Versace ... ... COPYRIGHT YOUR WORK ONLINE ... ... My Website

  • BinaryFxBinaryFx Registered Users Posts: 707 Major grins
    edited September 7, 2007
    If you have page layout or illustration software, one could place/link to the separate images and combine them (depending on the montage). This puts all the memory intensive tasks at output, rather than in composition.

    If doing this in Photoshop, one may start at the final size of 16x20 (with or without bleed?) - or decide to work in two or four sections rather than at the final size and later combine these tiles into the full size document. The flattened tiles should be smaller than the layered working files.

    One also has to settle on a resolution for the canvas, before one brings in the original images that will be montaged. The output device, image content and viewing conditions will decide this point. For a 150lpi halftone press output, the final size image is often 300ppi, or x2 the output screen ruling. It is also commonly noted that this may be overkill (depending on image content) and that one may be able to use a quality factor of only x1.5 or 225ppi for a 150lpi halftone output. For Epson inkjets, 360ppi is often quoted as an 'ideal' input resolution at 100% size (due to these printers using native output resolutions of 360, 720, 1440 dpi etc). It is also noted by many that using 240ppi instead of 360ppi (2/3) is acceptable. So, one may decide to work at a lower resolution than is ideal, if memory overheads are a concern. So your blank canvas may be 16x20 @ 360ppi, or 16x20 @ 240ppi, before you place the separate images as smart objects, resize them, mask them etc.

    If using Photoshop CS2 or higher, it would perhaps be wise to place the separate files as smart objects (from an image quality and freedom of manipulation point of view).


    Hope this helps,

    Stephen Marsh.
    http://members.ozemail.com.au/~binaryfx/
Sign In or Register to comment.