I always shoot in the highest possible settings and in RAW+JPEG mode.

9 times out of 10 I only keep the jpg, but sometimes the raw file is invaluable, especially if you're shooting in strange light conditions when the jpeg engine in the camera can get it horribly wrong.

As said above, you can take detail away but you can't add it afterwards. Shooting in RAW+JPEG means I use 10-15MB per shot. I only have 1GB cards at the moment, and I don't find the size limiting.

As I've said, it's not so much 'quality' that makes me shoot large pics, but the fact that it means you have the biggest, most detailed, original possible which gives you a lot more scope to play with and edit/crop etc. If you don't really manipulate the images after you take them (or expect to print them large) then smaller images mean a saving on storage space and may seem a better idea.

[/color]

Agreed. Also, by shooting in RAW mode you have so much more information to play with after shooting.

Of course you need to do lots of mucking about with the photo to make it presentable, and you need extra software to do that with (the stuff which comes free is pretty hopeless - I use Bibble). But often you can correct quite severe mistakes of exposure and the like to salvage good shots from otherwise unusable ones.

Case in point (hugely overexposed original):

No problem, it's easier to think of it as a 'quality' setting. 'Fine' is better quality (but take up more space on the card) but the quality is independent of the size of the image (megapixels).

As I've said, it's not so much 'quality' that makes me shoot large pics, but the fact that it means you have the biggest, most detailed, original possible which gives you a lot more scope to play with and edit/crop etc. If you don't really manipulate the images after you take them (or expect to print them large) then smaller images mean a saving on storage space and may seem a better idea.

Yeah, I understand that quality and size are separate, hence why i chose to reduce size over quality. At the mo I dont manipulate them to a large degree, I may do when I get further into my course though

I didn't realise HOW the camera produced the smaller image though, I didn't know it did it by using less of the sensor, but I get that now lol

I didn't realise HOW the camera produced the smaller image though, I didn't know it did it by using less of the sensor, but I get that now lol

I'd be interested to know how it really does it, does it simply use less pixels on the CCD and adjust it that way (hardware) or does it shoot full size and reduce to the set size? (software).

As if it's the 2nd, surely theres a processor/battery overhead to shooting smaller than native size? And if the 1st would that mean some kind of interpolation algorithm which would limit the quality of the reduced pic over one resized manually?

Might need to Google that when I've got spare time.

*oops, double post

Logically, if it uses a small part of the sensor then the focal length should be affected - smaller sensors produce narrower (more zoomed) images.

Right, I've just tested it (took two photos of a metre rule). There is no change in zoom, therefore the size of the image is a software function.

That actually explains why you can't use the size function in RAW mode. The camera is capturing the image, then resizing and compressing the JPEG.

Logically, if it uses a small part of the sensor then the focal length should be affected - smaller sensors produce narrower (more zoomed) images.

Right, I've just tested it (took two photos of a metre rule). There is no change in zoom, therefore the size of the image is a software function.

I assumed it can't be physically using a smaller area as the image would be cropped. But is it a software resize (after the full image is captured) or a hardware resize (using less pixels of the CCD as it is captured)? The only way I can think of checking is to see if there's a noticeable difference in write time?

If it's the former resize then I'd expect batteries to wear faster when taking a large amount of shots as there's more intensive image manipulation going on inside the camera. It could also affect multiple shot rates and such? It would also mean the quality of the image is dependent on the algorithm used to reduce the image in camera (bicubic resampling etc.) ie you could be losing quality as well as size by shooting at a lower resolution on the camera rather than resizing manually.

Kinda pointless really, but has made me wonder.

Using the old 'simplest answer is usually the right one' technique I'd go for the former also - i.e. fullsize RAW image is captured, then JPEG'd and resized. As I don't use these modes I wouldn't like to guess at the performance hit or quality of the output, but I know these camera engines are getting extremely quick - also resizing down is a relatively straightforward process where quality isn't much of an issue - you don't get artifacts downsizing unlike enlarging.

I'd still rather do all that stuff on the PC though.

Yes, using common sense tells me that as you can shoot RAW+JPEG it must capure RAW and compress for the JPEG. So highly likely it also resizes from a RAW image.