digitise your photos

home

technical talk:

OK, so if you're not too familiar with some of the technical issues then let's see if we can cut a path through some of them without resorting to too much geek speak...

Firstly, the quality and appearance of any image that you see is determined by two main factors: the resolution and the colour depth.

The term resolution really refers to the amount of detail in the image and, since images are made up of tiny dots, or pixels, the resolution specifies the number of pixels. For example, an image like this one...

...measures 300 pixels wide and 200 pixels high, giving a total of 300 x 200 = 60,000 pixels. OK so far?

The term colour depth refers to the number of colours available to make up the image. Every pixel has a colour and each colour has a code assigned to it - different shades of red, for example, will need different codes to represent each of them... and if you want to have lots of colours then you'll need lots of codes... and the more codes you have, the larger the file size...

One byte (8 bits) can hold 256 different codes, or colours, so if that's enough for your purposes then each of our 60,000 pixels will need one byte and the file size will be about - but not exactly - 60,000 bytes. Unfortunately, for good photo-realistic images, 256 colours just doesn't do it and in practice we need many many more.

If we use two bytes for every pixel, instead of one, then we can have 256 x 256 = 65536 colours. You'd think that this was enough, right? Nah! Two bytes good but three or four bytes even better! So we generally have 24-bit (3 bytes) or 32-bit (4 bytes) colour systems - more colours than you can shake a stick at! In fact, the image above uses 24-bit colour, in other words 3 bytes for each pixel, and so our 60,000 pixels need 3 x 60,000 = 180,000 bytes

So does it make a difference having - or not having - all those colours? Well here's the same image but with the colour depth reduced to 1 byte ie 256 colours:

...and here it is reduced to 16 colours:

...and now 2 colours:

In each case, the file size is drastically reduced if we use fewer colours...

...we'll come back to file sizes later.

So, back to our 24-bit colour image which measures 300 by 200 pictures. How big does it look on your screen? (At this point we apologise again for the maths...) If your screen size is, for example 1200 by 1000 pixels then it should occupy 300/1200 = 1/4 of the screen width and 200/1000 = 1/5 of the height. If your screen resolution is less than this then the image will occupy more of the screen space because it'll be a larger fraction of the whole width and height.

It is quite possible to reduce the size of the image. There are a number of ways to do this but essentially they involve losing pixels and therefore information. It's possible, also, to increase the size of the image but if you do, you can't expect to add any extra detail - that only really happens in Hollywood!

OK, so all this may be vaguely interesting in a geeky sort of way but what implications does it have for you? What resolution should your pictures be?

As you might expect, there is no single answer to this one other than to say that it depends on what you want to do with them. Hardly very helpful! If you know that you're only ever going to view the images on a screen then your screen's resolution is the main factor and these are typically not great. To use the example above if your screen size is 1200 by 1000 then an image of 1200 pixels by 1000 pixels will competely fill it. Screen sizes may increase as time goes on, but right now there's no point in you having an image resolution that is any bigger than your screen because you'll only have to reduce its size in order to view it.

If you want to use the image on a website then they can be smaller than this because you don't generally want them to fill a screen. The problem arises when you want to print your pictures, however...

Again, we need to do some maths to illustrate the situation. Our starting point is to recognise that a good quality print requires a resolution on your printer of at least 300 dpi, or dots per inch. (We can use the terms dots per inch and pixels per inch interchangeably) For the picture above, that means that if we print it at 300 dpi then it will measure 2 inches wide and 1.5 inches high, not very big. So what size do you want your prints to be? Typical small prints measure 6 inches by 4 inches so at 300dpi you need an image that is 6 x 300 = 1800 pixels wide and 4 x 300 = 1200 pixels high.

To look at another example, a typical 4 megapixel camera will take pictures that measure 2200 pixels by 1700 pixels so (dividing by 300) you could expect to get good prints measuring approximately 7 by 5.5 inches. In practice, we've had good results at sizes up to A5. We generally scan 35mm slides and negatives at 2000dpi - although we can go up to 4000dpi if necessary - and again we've had good quality prints at sizes up to A5 at 2000dpi and up to A4 size or more at 4000dpi.

To sum up, then, you will probably want to capture your images at as high a resolution as possible - you can always reduce the size but you can't increase it. Hard disk storage is relatively inexpensive and there is always the option of archiving the images to CD or DVD if you're short of space.

One final topic to consider is the issue of file size. If you've looked at the file properties for the images above (right-click on them) you'll see that not only are they of different sizes - as you would expect - but they are of different types. So what's all this about file types? There are many different image file formats, or types and they each have advantages and disadvantages. Put simply, some file formats compress the image data so the the file takes up less room on your disk, and some do not. Clearly there are advantages in having smaller file sizes; particularly when viewing images on the Internet where download speeds are crucial.

So if some file formats result in a smaller file size, why don't we all use them? Two formats widely used are GIF and JPG. Both compress file sizes, but in different ways. GIF files are not really relevant to us because they are limited to just 256 colours; adequate for many purposes but not for photographs. The JPG file format does allow 24-bit colour - fine for photos - but when it compresses a file it actually loses some of the information - it's called a lossy file format. For most people this loss would be imperceptible and the benefits of the much reduced file size far outweigh any minute loss of quality. What isn't always appreciated is that every time you edit and then save an image in JPG format it 'loses' just a little more information. It is, then, better to make any changes using the original uncompressed file (see below) and then save it as a JPG file.

It's also worth pointing out - because many may not be aware of this - that most image editing applications will allow you to specify the level of compression when saving an image in JPG format. Generally

high compression = more data loss = lower quality

and, of course, vice versa. Typically the default value will not be set for minimum data loss - ie maximum quality - and you may need to explore the options to find the one that suits your purposes.

When we scan images for you we will supply you with the original images in JPG format which is saved at minimum compression, ie best quality. We can supply the files in TIFF format if you prefer or, indeed any of a number of other formats if you so wish. In fact, we always supply b/w images in TIFF format as, unlike JPG, it can hold 16-bit greyscale data. At this point we make no apology for referring you to our comments re the need to backup all your image files, wherever they're kept - you can read more in the care and maintenance section on the 'slides' page.

These are difficult issues and we hope that the comments here have been helpful. If you have questions or concerns than please get in touch; our contact details are here.
 

 
home ] [ prints ] [ negatives ] [ slides ] [ restoration ] [ pricing ] [ technical talk ] [ contact ] [ disclaimer ] [ online payments ]
 

 

all original material on these pages copyright © 2017 DigitiseYourPhotos and respective owners