Shooting in RAW?

Snurk71

DIS Veteran
Joined
May 17, 2001
Messages
3,239
I think I kind of understand the concept of shooting in RAW - that you have the best editing capabilities. But what does that really mean between the editing you can do on a RAW picture vs. a JPG at the highest MP? What more can you do with a RAW image vs. a JPG (in a non photo expert explanation? Do you need a high powered program like the full PS to work with RAW, or would it be beneficial to shoot in RAW if using PSE as an editor?

I'm stepping up to a DSLR and curious whether I should be looking to shoot to RAW or the JPG I've always done in the past with a P&S.

Thanks
 
A RAW file contains all of the data captured by the camera, whereas a jpg file has been compressed, which necessarily causes some of the data to be lost. The data that the camera discards in the compression may or may not be important to any editing you might want to do. The thing that I hear mentioned most frequently as a benefit to shooting RAW insofar as editing capabilities are concerned is white balance--i.e., you can often make much better corrections to white balance with a RAW file. You also usually have more latitude with regard to correcting exposure with a RAW file. In short, you can often fix your mistakes better in post processing with a RAW as opposed to a JPG file. Of course, some photographers will tell you that if you just take the time to get it right in the first place, you don't need RAW.

The downsides to RAW seem to be the larger file size (though as memory gets cheaper, that is less of an issue) and the fact that the camera manufacturers RAW formats are all proprietary, so you have to have software that can handle them. PSE 4.0 has a RAW conversion tool, but I don't think the earlier versions do. I've read that the PSE version is not as "powerful" as the full PS or some other RAW converters, but I haven't used anything but PSE (and that only a few times), so I can't really comment on that aspect of it.
 
The RAW fans call it the "digital negative" because while it contains information on color settings, etc, it mainly stores the untouched original photo. The JPG is not only compressed but has the various filters already applied - for different color levels, white balance, sharpening, etc.

The other big problem with Raw is that it's also slooooooow to deal with! Forget using a normal photo viewer like Irfanview to quickly flip through your pictures like you can with JPGs. It'll work but be dramatically slower. It's also slower when you're taking the pictures, you'll be waiting longer for the camera to write the picture to the card.

I usually shoot in JPG but occasionally will do some Raw, but I generally don't do enough post-production on the photos to make any difference.
 
Read this

very very good article. I shoot almost strictly in RAW now. I enjoy doing the post process work, and it has made my PS skill increase.
 

I had always shot JPG and been happy with the results but for my last trip I switched to RAW on my Nikon D200 and I may never go back. The latitude I am provided for post processing far outweighs the cost of memory or processing. A few things of note, it does require you to change your workflow for images as you must preprocess them either through a RAW converter or some other piece of software before they are available for viewing by the masses. The lack of compression artifacts is great and changing the white balance, hue, exposure, etc. after the fact is awesome. Please don't get me wrong, making sure the settings are correct before you take the picture is always preferred but should you happen to forget, RAW gives you a mulligan to reset. If you have Windows XP, Microsoft has a Raw Image Thumbnailer and Viewer that makes the RAW images a little more integrated like other graphic formats. I have not noticed any difference in the time it takes to record an image in RAW versus JPG in the camera. But then again I use SanDisk Ultra III Compact Flash cards so I would not expect to see a delay.

Jeff
 
RAW is very misunderstood by most photographers, but it is the best image quality available from your camera, and that's why I bought a SLR and L glass in the first place.

RAW takes more write time but I have only run out of buffer once, taking photos of the tigers at AK, and it wasn't a big slowdown anyway. Newer cameras than my 10D are faster at writing to the card so this is a small issue now.

I usually shoot with a small/fine embedded JPG so I have something to email or post quickly. Some viewers, like BreezeBrowser display the embedded thumbnail so display time is very quick, even for RAW images.

The only disadvantage I can see is the memory space issue.

I took some photos on a misty foggy day on the Blue Ridge Parkway (not uncommon) and after expanding the levels the JPG had numerous very visible gaps in the values, the RAW had none. I'm sold.
 
Kind of think of it as what you see is what you get. With RAW the camera stores what it sees without changing anything and leaves it up to you.

Here is an example, You shoot in jpeg and the camera selects the white balance for you if you have it on automatic, you may select sunlight as the balance if you have it in manual. After you take the picture you are stuck with it.
With RAW you can change the white balance any way you want after you take the picture. You aren't stuck with the balance.
 
I have been shootin RAW since I first got the 10D, but recently with so many debates I tried Raw+jpeg for my entire WDW trip(AUG). On some images I was very happy with jpeg but on some of the images that demand more dynamic range RAW rules. I am now sure that I was not wasting my time, and as for workflow I find it takes just as much time to process RAW files as it does jpegs.

Look at it this way, have you ever heard a jpeg user claim that "jpeg out of camera is best"?. No they always say they that jpegs are good enough, but never the word best. Even the camera manufacturers, state that RAW yields the best results.


Rawshooter Essentials, FREE(not limited trial)

You can set defaults(contrast/saturation/etc...) and walk away while it converts your raw files to the file format you choose. If you

http://www.pixmantec.com/products/rawshooter_essentials.asp
 
The reason that you can get more dynamic range out of RAW files and correct exposure better is that the camera generally records 12 bits of data for each photo site whereas a jpg has been reduced to 8 bit color. That's a lot of extra information to play around with. If you use RAW then you have all of that extra color information to work with to make corrections before you save the file as jpg (which is 8 bit). If you just shoot jpg then you are allowing the camera to select color balance, contrast curve, etc.

I still shoot jpgs most of the time since I'm generally not that interested in making fine art photos (I just want photos of the family). When I want to take nice portraits or shoot something that I think I'll want to play with then I'll switch to RAW. Otherwise I just try to take good exposures and manage the white balance in the camera to minimize post processing.

By the way, I love Rawshooter for it's features and you should definitely try the free version. But, I always prefer the final output of Phase One's Capture One LE. It's interface is really nice though not as nice as Rawshooter. I always want to use Rawshooter because it's more fun but at the end of the day Capture One has better images (in my opinion). I hope I didn't start a fight... :confused3
 
PaulD said:
The reason that you can get more dynamic range out of RAW files and correct exposure better is that the camera generally records 12 bits of data for each photo site whereas a jpg has been reduced to 8 bit color. That's a lot of extra information to play around with.

Well that has to do with processing, and IMO not so much with bits...

If you process a raw file and then convert to jpeg(8bit), the "expanded" dynamic range is can still be present. So I can only assume it has nothing to do with bits and more to do with processing b4 the conversion.


And yes C1 yeilds great results, I reccomended RSE cuz it is free.
I mostly tend to use ACR because of it being integrated into CS2, but for higher demanding stuff I use C1 pro or RSP.
 
PaulD said:
The reason that you can get more dynamic range out of RAW files and correct exposure better is that the camera generally records 12 bits of data for each photo site whereas a jpg has been reduced to 8 bit color.

One of the common misconceptions about RAW is that there is more dynamic range. There is more range... sometimes.

The sensor and it's amplifier determine the dynamic range. Once captured, the 12 bit image can often be compressed to 8 bits while still retaining all the dynamic range. The issue is the number of levels available. Too much range does not fit well in 256 levels, there will be gaps between levels. In some places the gaps are visible, like in a clear blue sky.
4096 levels fits a wide range much better, greatly reducing the gaps between levels and providing a much smoother image.

When the range is compressed to minimize the gaps, either in the highlights or the shadows, RAW will indeed provide more dynamic range in those areas.

If we could take the time to carefully meter the scene to determine how much range is there, and if it fit into 8 bits, we could decide if we need RAW or if JPG would be enough.

Memory cards start sounding pretty inexpensive at that point! ;)
 
You know, every time I see this thread I cringe and open it just to make sure we are still talking about storing images in RAW format and that the discussion has not morphed into talking about photographers wearing no clothes while taking pictures. There are just some visuals I don't need to have.

Jeff
 
The reason that you can get more dynamic range out of RAW files and correct exposure better is that the camera generally records 12 bits of data for each photo site whereas a jpg has been reduced to 8 bit color.

Actually, both a RAW file and a JPG file are compressed. The difference that I think you are driving at is that a JPG file uses a form of "lossy" compression whereas a RAW file uses "lossless" compression. A lossy compression algorythm throws away information that it determines to be less important. A lossless compression algorythm is one in which the uncompressed data is absolutely identical to the input data.

The fact that JPG is a lossy format makes it important that you minimize the number of times you open, edit, and save a JPG file. Every time you do so, the compression process loses more data. To see what I mean, open any JPG photo, make a minor edit, save it, and close it. Repeat this process through a few iterations. Then compare the original with the multi-edited version and you'll see a significant loss of detail and higher levels of noise/artifacts.

A RAW file contains all of the data captured by the camera, whereas a jpg file has been compressed, which necessarily causes some of the data to be lost.

Sort of, but not quite. As has been said, in RAW you get 12 bits of data (4096 different possible values) for each of the three colors whereas in JPG you only get 8 bits of data (256 possible values). The JPG can still cover the same dynamic range with 8 bits that the RAW file covered with 12 bits, however, if it does so, the differences between each color value are necessarily larger.

For example, if the RAW file had every possible value of blue from none at all to the brightest possible, it would have 4096 different shades of blue represented. A JPG version could still go from no blue at all to the brightest possible blue, but it would have to do it with only 256 different shades. The picture covers the same dynamic range, but the differences between the shades of blue are larger.

So when would that matter? It very rarely does straight out of the camera. However, if you start editing the photo, doing things like changing the brightness or contrast, you may find that you need those subtle differences that were present in the RAW file but not in the JPG.

With RAW you can change the white balance any way you want after you take the picture. You aren't stuck with the balance.

You aren't really stuck with the white balance either way. In both photos, you've captured a range of color values for every picture. With the JPG file, you've also adjusted that range of colors to account for an expected color cast. If the process that converted to JPG got the white balance wrong, you can still adjust it. The problem with doing this on a JPG is that you've already lost a lot of the original information when you converted the file. It may not be possible to adjust the white balance very much without making the picture look bad. The same thing can happen with a RAW file, but because you've still got a lot more information, you've got much more room for error.
 
Anewman said:
Well that has to do with processing, and IMO not so much with bits...

If you process a raw file and then convert to jpeg(8bit), the "expanded" dynamic range is can still be present. So I can only assume it has nothing to do with bits and more to do with processing b4 the conversion.

Well, if you have a good exposure and the scene isn't really high contrast then you probably don't need to use RAW (at least not for exposure issues). It becomes more useful when you have a high contrast scene and you want to enhance the extremes of the image (i.e. the shadows and the highlights). boBQuincy makes some comments about the gaps between levels and that's really the point. If you have more information to work with then you can make adjustments to these areas without having problems with creating a stairstep effect instead of smooth transitions. You can certainly do a lot with Photoshop but it's easier to work with the original unaltered data so that you can make all the decisions about how that data is rendered into an image.

And it's not all about dynamic range and exposure. There are a lot of decisions that the camera makes for you when it renders the image. For example color issues like white balance, saturation, hue, and how each color is rendered (i.e. should green look like this or this?). Sharpening is another biggie. All of these thing can be controlled if you use RAW. That's why the big camera manufacturers provide you the option.

I'm not saying everything should be shot in RAW. I certainly don't unless I think I might need to do some editing later (and I want to spend the time doing it). RAW shooting just offers one more set of tools and techniques to use to get the most from you images. If you want the ultimate control over every aspect of your photo then you should start with a RAW file. If it sounds like more hassle than it's worth then just keep doing what you're doing and be happy. It's all about having fun, right?
 
If I'm not mistaken, whatever program you use to PP RAW files displays the file on your screen using the parameters stored in the RAW at the time the picture was taken (settings in the camera). So if they were off, like WB, it will appear that way on your screen. With that, I've read that different RAW converters can vary in their ability to PP RAW files. IOW, some do a better job at it than others.
 
MarkBarbieri said:
Actually, both a RAW file and a JPG file are compressed. The difference that I think you are driving at is that a JPG file uses a form of "lossy" compression whereas a RAW file uses "lossless" compression. A lossy compression algorythm throws away information that it determines to be less important. A lossless compression algorythm is one in which the uncompressed data is absolutely identical to the input data.
I understand the difference however that doesn't change the fact that RAW is not necessarily compressed. My camera stores them uncompressed - every picture is exactly the same size, no matter the contents. Since RAW is not a standard but proprietary, each manufacturer is free to come up with their own standard. Some do lossless compression, some don't.
 
i shoot everything but outdoor sunny team sport in RAW. i do all of my image processing in a 'non-destructive' fashion through Apple Aperture (and some Adobe Lightroom). i can easily lift and stamp to batch process 800 RAW images while highlighting, say, 25 or so that have specific needs.

my main reasons are as follows (more or less in priority order):

1) RAW gives me an unmodified and lossless representation of what the sensor captured. it's as close to a negative as i'm going to get. the moment you go jpg, the following irreversible damage is applied to the image:
- conversion to lossy compression with data loss
- maximum colour depth is reduced with a permanent WB temperature/tint applied to image
- in-camera sharpening is applied

2) precise whitebalance control. auto whitebalance is good to start, but things like mixed light sources or light sources that change temperature over time (flourescent, mercury vapour, etc.) so you could easily have 4 or 5 different WB requirement every second.

3) dynamic range - again like good film stock, gives increased ability to salvage images due to wide variety of exposure issues and allow for things like tone mapping. this also generally means much better/accurate saturation and contrast.

4) RAW files are smaller than TIF files

since daytime team shots are cropped and readied for printing (or printed onsite), .jpg works fine and the images are then archived by team, age, and season. in other words, i guess it's a nice way of saying i don't really care so much because the parents are happy :)
 
My camera stores them uncompressed

OK, I suppose it might have been just as incorrect for me to say that RAW files are compressed as it was for the earlier poster to say that they are not. Still, I'm surprised that any modern digital camera would store RAW files without compressing them. What a huge waste of space.

A typical RAW capable digital camera these days has something like an 8 megapixel resolution and a 12 bits per channel of color depth per pixel. That's 32 bits for 8 megapixels or a total of 24 megabytes per file. That's some big files.

Actually, now that I think about it, it may be only 8 megabytes for a RAW image if it's really an 8 megapixel, 12 bit depth bayer array. That's because each pixel in the original image is just red, blue, or green. It gets the other colors from the pixels around it.

From what I've read, Canon uses lossless compression on their RAW files while Nikon and Kodak use slightly lossy compression algorithms on theirs, so I guess I was even more wrong with my original statement.
 














Save Up to 30% on Rooms at Walt Disney World!

Save up to 30% on rooms at select Disney Resorts Collection hotels when you stay 5 consecutive nights or longer in late summer and early fall. Plus, enjoy other savings for shorter stays.This offer is valid for stays most nights from August 1 to October 11, 2025.
CLICK HERE













DIS Facebook DIS youtube DIS Instagram DIS Pinterest

Back
Top