I know I should be using RAW etc etc but I just prefer the simplicity of getting a VERY acceptable image out of the camera (especially on my new Canon 60D). I've read the article by Joseph on the differences between 10,11 and 12 quality but that's from a TIFF original.
When I try to export at 100% but at a range of qualities I see NO discernible difference between the output files even at 200% when re-imported in A3 but I do see a massive difference in file size. I've viewed them on my 27” iMac, iPad 1 and my 46” LCD tv via Apple TV 2 and they all look the same. I did zoom in to 1,000% and I did notice some very slight colour changes on the odd pixel but nothing of any significance - certainly no creation of artifacts.
To all intents and purposes the files are identical.
My question is - am I seeing this because I am starting from a JPG image or am I missing something?
Robin,
To truly see the difference, you need to layer them in Photoshop and run the difference algorithm. However if you’re perfectly happy with the JPG coming out of the camera, and you are perfectly happy with the JPG exported from Aperture, then that’s fine. Where you will see a big difference is if you get into editing the photo—especially when you need to fix an over or underexposed photo. However if your objective is to simply shoot and export, then you’re fine shooting JPEG.
We put a lot of emphasis on RAW around here, but at the end of the day, all that matters is that you’re happy.
You pointed out the size difference in the JPG files from camera vs from Aperture, but you didn’t say which was bigger or smaller? Is that part of your question?
@PhotoJoseph
— Have you signed up for the mailing list?
The reference to the sizes was that exporting an image from A3 at say 11, 9 and 7 quality gave for example 6meg, 5meg and 4meg jpg sizes - not surprising i.e. less quality = less size. The problem I had was that I really couldn’t see any difference between these three JPGs. I was really wondering if there really was a difference - I can’t see any point in uploading 6meg images to Zenfolio when the 4meg will do.
To some extent it doesn’t matter - as time moves on and cameras get better and better and displays get clearer and clearer whose going to want to look at these older images anyway!
I predict a day when you don’t have to think how many pixels a camera has or how many a tablet/screen has - every picture will be just perfect. Why should we have to consider the technology behind the concept.
Robin,
it’s really a question of understanding how compression works. JPEG is what is known as a ‘lossy’ compression. This means what it sounds like. It throws away data in order to save space. Its usefulness compared to ‘lossless’ compression, is that lossy can save much more space than lossless.
Very simply, JPEG works by looking for groups of pixels that are similar. The ‘lower quality’ compression you choose, the more casual JPEG is in deciding whether pixels are similar or not. So, for example, in deep shadow areas of an image, JPEG might decide that most of the pixels are similar enough to be considered the same and so instead of having to write a piece of code, describing the precise colour and position of each individual pixel, it just says … ‘this colour, multiplied by 300’ or something like that. Obviously in reality it’s more precise/complex than this … but you get the idea.
My understanding of what happens when you ‘JPEG a JPEG’ and a crucial piece of the puzzle for understanding how best to incorporate JPEG into your workflow, is that JPEG has no knowledge at all, of any previous compression applied to an image. It treats all images the same. So if you JPEG an existing JPEG, it will apply the quality setting that you choose, as if it were the first time the image was being saved. This means that if you chose a higher quality, it will look for smaller groups of pixels to compress and so the file size will increase, even though there are already many many groups of absolutely identical pixels.
As for the advantages of JPEG. Those are purely file size. For viewing on a monitor where you can see the entire image, this is absolutely fine, up to the point at which the compression (or re-compression) starts to create visible artefacts. Some images will show these more readily than others. JPEG is a poor compression format for anything with hard-edges or straight lines. So if landscape and portraits are your thing, you can take JPEG a long way.
The great disadvantages of JPEG are for printing of any sort and for any kind of extensive manipulation or compositing, where JPEG is totally unsuitable and actually, for even modest curves adjustments, where JPEGs can quickly fall apart.
If you are saving out of your camera at maximum quality JPEG and decide on rare occasions to push an image a little further than usual and quality becomes a concern, then convert it to Tiff first, then work on the Tiff, then convert back to JPEG once you’re happy. A max quality JPEG converted to Tiff, will be very close in quality, bearing in mind, that it depends to a fair extent on the subject matter.