Camera RAW have I got it right? Draft of why we use RAW.

When one tries to get ones head around a concept it is often easier to write down what you think you have learnt to help learn about it. I have tried to do this with what I have learnt about camera RAW. So next is to ask have I got it right or wrong? So I am inviting members to rip this apart in the hope that once this is done I can re-write in a way where others can read and understand. So here it is:-

What is RAW, in the main it is the 12 bit information captured by the camera before its internal computer decides what bits to use for the Jpeg image. In the camera we can set Jpeg record Pixels, Quality, Tone, Saturation, Sharpness, Contrast, White Balance, Colour Space, with my Pentax K10D one can after taking the picture convert it from RAW to Jpeg with a limited user set functions size (10m, 6m or 2m), quality (Best, Better, Good), colour (around 11 options), ISO (-2 to +2), Image tone (Natural or Bright), Saturation (-3 to +3), Sharpness (-3 to +3), Contrast (-3 to +3).

The conversion can also be done out of the camera. The whole idea is we just don’t have the time to set all the parameters in the camera before every shot and we rely with Jpeg on the camera getting it right. Using RAW we take control of the conversion. There are many converters we can use out of the camera. Often the manufacturer will bundle in so basic program with the camera. Programs like Photoshop full, or elements, Light room and others can convert RAW images often with add on programs. Others use RAW direct like most of the HRD programs which can take 2 or more RAW images and combine to get 32 bit images.

Most offer more control than found in camera. Some programs like Gimp will have a few RAW converter options with in the case of Gimp UFraw and RawTherapee. However some manufacturers change their RAW files so with for example the Nikon D7000 Adobe Camera RAW 6.5 is the lowest version that will handle the files. Adobe have offered a free converter to adobe’s own RAW format (DNG) or you can use one of the free programs like RawTherapee.

Each conversion program offers different features Lightroom for example offers a history feature. RawTherapee offers Tone Mapping, and Photoshop has extra features with the full program to that offered with elements. Most seem to offer some lens correction some have spot removal, adjustment brushes, and graduated filter. As a result how much is done in the conversion will vary program to program. In some cases it may be required to make multiple Jpeg images each capturing different aspects then combine them in another program. One could even convert to a 16 bit Tiff file and do all the adjustments in the main program.

So methods will vary but the result is the same one is able to select manually which bits to keep and which to dispose of post exposure. Light levels must be the main advantage but colour temperature must come first as it effects light levels. However using features like the graduated filter where a tint can be added often it is a case of revisiting each option a few times.

The converting RAW to Jpeg can take a lot of processor time and many of the programs allow you to set what you want to do with a batch of files then set the computer to convert and you can walk away and leave the computer to do each file in turn. This in turn means the use of some file handling program as well as the RAW program so with for example Adobe Photoshop one can end up using three programs together. Bridge allows you to select and open the RAW files, Camera RAW has a row of icons down the left showing each file opened and once settings are made you can auto open in the main Photoshop program. However it also has the option to save directly as Jpeg files without opening the main program. Much will depend on the size of the computer used but clearly opening 100 images in Photoshop could be a problem so saving as Jpeg to be opened latter is often the better option.
The information as to what you have done can be either added to the RAW files this is done with DNG RAW files or saved as a sidecar which is another file normally with same main name as original but different type designation. For example .pp3 (RawTherapee) or .xmp (Adobe - Extensible Metadata Platform) however depending on operating system these files are sometimes hidden so one can move a RAW files and lose the information which goes with it.

There are programs which give Instant Jpeg from RAW these were in the main marketed because there was a lack of Codex programs available to allow one to see thumbnails of the RAW files. Typically these programs also reduce file size so 10Mb RAW files will be converted into a 1.25Mb Jpeg. Although handy to organise files they are not really much good for anything else.

Most manufactures release Codex programs to allow one to see thumbnails of the RAW files. But some times these do not work with older operating systems like Windows XP. However RawTherapee and Bridge can be used to organise files and these will show thumbnails. Bridge has an option to vary the size of the thumbnail however it uses more cache when the size is changed.

There are other programs like Rawstudio, and Darktable but the question is how much to do with each program, and this is so dependent on what other software is available there is no real answer. Even staying with just Adobe products with CS5 with Camera RAW 6.5 converting to Black and White and then returning the colour to selected areas would require two images and layers. Likely it would be better to use Photoshop main to convert, and then use history brush. But with Light Room the selective return of colour is included with the RAW converter. This is repeated again and again when comparing programs.

Until I started to write this report I had never considered multi conversions inside the camera but clearly using ±2 ISO one could make two Jpeg images from one RAW images load into layers then create a composite photo.

RAW + Jpeg is of very little use, as one can produce Jpeg from the RAW in camera when ever required, and set to do this automatically will increase save time, so can stall the camera when taking multi images as with for example panorama or HDR. Card space and band width will mean there are times when Jpeg is required, but this is really selected as a special rather than a norm.

Personally I use Adobe Camera RAW 6.5, and by time I have finished very little needs doing in CS5, which, as a result, becomes more of a graphic artist’s package, than a photography package. I find the Spot removal, Adjustment Brush, Graduated filter and Post Crop Vignette very useful as well as the Basic controls. Most the others are seldom used. However likely I should, Targeted adjustment tool I tried to use for first time today. Like many more tools this allows you to do the same as one has always done, but in a more intuitive way.

The main point is to use all the tools to bring out features which once converted to 8 bit will be partly lost.

Thanks for reading I wait to see the comments. All best Eric

Hi Eric,

I will try and read this in more depth when I have a bit more time.

One thing where my work flow would differ from yours is that I would not routinely create TIFs or JPGs unless I was going to work in PS or wanted to create a jpg for web or AV purposes. So I have little need to batch process. I would always print straight out of LR. Occasionally there are several images I might want to use in one session in which case the export function in LR will handle multiple files very easily.

Thank you David I realise Light Room is better at handling RAW from Kevin's demo last Monday.
If I was asked Photoshop Elements or Light Room the latter would be my selection. However I have tried except for final personal note to keep it general.
My problem is many of the converters have much more than just conversion. The idea of converting all to 16 bit tiff files did cross my mind once as 16 bit tiff you can use curves to reduce to 8 bit however when I tried it I found Camera RAW 6.5 saved as 8 bit tiff so rather useless and although RawTherapee would save as 16 bit tiff when I came to convert I was unable to get conversion up with CS5 as it did with CS4 so with CS5 you have to go to HDR toning to convert and with CS4 you go to Mode 8 bits channel to get the same conversion window up. Likely a completely different way to do same thing with Light Room?

As a result instruction sets of how to convert a 16 bit tiff to 8 bit becomes very messy if one is going to retain the information held in the 12 or 14 bit RAW file. Hence not talking about using 16 bit tiff. There is also the 58.08 Mb size of a 16 bit tiff compared with 8.93 MB of the original RAW file which is going to be a problem with older computers.

I tried to get some photos printed on the high street and it would seem although you may get the odd one who can handle tiff files in the main you have no option but use a Jpeg.

Although I would agree when sending to print houses you have more options and buying ones own high quality printer is likely the best option to serious photographers this was all started with the beginners group so final picture really does need to be Jpeg.

I have checked with both Pentax K10D and Nikon D7000 and in both cases you can convert from RAW to Jpeg in camera with user controls giving + or - 2EV stops so one could make 2 or more Jpegs to be latter combined once out of the camera. What would be interesting is this also a feature offered with other makes and model of camera?

This also raises the often asked question of setting camera to RAW + Jpeg since a Jpeg can be produced from the RAW latter except for special situations like taking card directly out of camera to be printed in high street shop why would one want to auto produce RAW + Jpeg?

As before all thoughts please if you think I am wrong please say so.

I have been experimenting to see if one could use Gimp to process RAW files and capture the whole dynamic range of the RAW file in a Jpeg file.

Gimp will not handle RAW images, so first bit is in the camera, and we produce two Jpeg images one +2 EV and one -2 EV. Then these are loaded into Gimp, do not use “Open” but use “Open as Layers” this will then give us two layers. With lighter layer on the top, add a layer mask selecting “Greyscale copy of layer” and tick “Invert mask” this will merge the two images together grabbing the whole range of the two exposures. As with all true HDR (rather than tone mapping) the picture will become rather wishy washy and likely one will want to adjust the Opacity to get the effect wanted.

Once found it was so easy but I have been most of the night trying to work out how to combine the two layers. Tried every mode without any look and also hunted the internet. This to me is the whole problem with Gimp finding out how.

I have been directed to this website http://www.digitaltrends.com/computing/how-to-get-photoshop-for-free/ and it would seem you can get CS2 for free which must be better than Gimp so giving up finding out how Gimp works.