Am I destroying my image quality by doing this?

[Dylan Johnson]

I had a shoot with a friend the other day, and I did my usual editing and then I started thing about how I was exporting my files. These were the steps I took after the shoot:

1. Import my RAW files into LR4

2. Did my usual editing

3. Exporting in JPG format

4. Imported those JPGs into Photoshop CS4

5. Did my retouching on the images (No extra layers or masks were created)

6. Exported those JPGs in JPG format again at the maximum quality.

Now, my question is am I destroying the image quality that way? Should I be exporting everything as a TIF until my final save of the image after all editing? Should I go back and re-edit all the images or do you think the quality of those files are okay to give to a client (or friend in this case).

Thanks.

 

[MrScary]

Export as a Tiff or a RAW then save As Tiff. Work on the Tiff. When finished you can either save As a Tiff or JPG. If JPG, make sure you do NOT work on it again.

I'm sure someone will pop in and tell you something different, just to confuse you.

 

[Dylan Johnson]

MrScary wrote:

Export as a Tiff or a RAW then save As Tiff. Work on the Tiff. When finished you can either save As a Tiff or JPG. If JPG, make sure you do NOT work on it again.

I'm sure someone will pop in and tell you something different, just to confuse you.

 

[Sailor Blue]

Truthfully, unless you have to do a lot of editing then you are probably not having any real problems editing a JPG, but you can be sure that you aren't creating any problems if you work on a TIFF as described below.

Check to see that you have set Lightroom Edit/Preferences/External Editing to TIFF, ProPhoto RGB, and 16 bits/component.

Now when you use the Lightroom Ctrl + E (Cmnd +E on a Mac) the file will be automatically exported to your copy of Photoshop in the above format.

Do all your editing on the 16-bit ProPhoto TIFF file then use the Ctrl +S keys to save the file. The TIFF file will be saved and a thumbnail will show up in Lightroom. Now use the Photoshop Edit/Convert to Profile/sRGB IEC61966-2.1 command to convert the image from ProPhoto to sRGB. Doing this profile conversion is necessary if you want your final JPG image to display the correct colors.

Now use the Image/Image Size/Bicubic Sharper (best for reduction) command to reduce the image to the desired size and then use the File/Save As/JPEG command to save the JPG image for your friends.

If I remember correctly Photoshop CS4 also has a File/Save for Web and Devices command. It that is true then you can use this command to shortcut the above procedure. After saving the TIFF file just use the Image/Image Size/Bicubic Sharper (best for reduction) command to reduce the image to the desired size and then use the File/Save for Web and Devices command. This command will automatically convert the saved file to sRGB when it saves it as a JPG.

Here is a great free video tutorial on how to work mostly non-destructively in Photoshop. It is for CS6 so expect fewer commands to work non-destructively with CS4.

B&H - Tim Grey - Photoshop CS6 for the Photographer - YouTube

--
Living and loving it in Bangkok, Thailand. Canon 7D - See the gear list for the rest.

 

[vander]

.jpg is a "lossy" format, whereas ".tif", for example is "lossless".

To understand how it works, picture a .jpg this way. Each image is made up of pixels, thousands and thousands of small boxes of different RGB colours.

A .jpg, when saved compresses the image by tossing away closely related colours in areas of high concentration. This is done and generally cannot be picked up by the eye.

However if you open that image which was previously saved as a .jpg and work on it and save it again, it reduces even more similar pixels in colour, resulting in an even more compressed image.

Likely you could do this dozens of times on a large file and would likely not see any difference, but if you keep doing it you'll definitely notice over time.

Here is a good video and example. http://petapixel.com/2010/02/04/saving-jpeg-photos-hundreds-of-times/

 

[PenguinPhotoCo]

should be fine.

I shoot raw, LR outs as jpg then portraiutre softens the skin (which creates a second gen JPG). I then take these into PS for any editing (cloning, retouching, etc) and output in the crops for the lab. If I'm printing them I'll save a PSD.
On all my albuims the LR-out jpgs go into fotofusion, from that to PS then are composited and exported from FF - 24" wide album spreads look great.

I don't see any issues even on large prints (rarely sell larger than 20x24 but have made promo banners 60") -but I"m shooting people where I don't perhaps need the detail that a cityscape would need.

 

[JJW1956]

I process in ACR (not Lightroom) and then export to CS6 for any fine tuning, then I save as a tiff, then I open the tiff in CS6 and save a jpeg, then I have to open the jpeg again in CS6 to sharpen that jpeg. Seems to work ok for me.

 

[BAK]

Have you looked at one of your photos?

Was it destroyed?

If you give a TIFF to your friend, what is he going to do with it?

Not much, probably.

BAK

 

[DuaneV]

I convert my RAWs to tiff and then do all the work on them and print from them. If I upload to my website or facebook page, then I add my watermark, resize and save as a jpeg and upload that.

 

[Dylan Johnson]

BAK wrote:

Have you looked at one of your photos?

Was it destroyed?

If you give a TIFF to your friend, what is he going to do with it?

Not much, probably.

BAK

I had a hard time telling which one was which, but its not so much what format to give to clients, its what format I should be doing all my editing in before the final save as JPEG,

Thanks for the reply though.

 

[Doug Haag]

I am an amateur photographer, so I may not know what the hell I am talking about. But when I first heard warnings about jpg being a "lossy" format, I undertook an experiment that may be similar to the video you posted, vander. Of course, my technique may be nonsense and some of you will surely explain why that is the case.

Anyway, the professional level software I use for editing (not PhotoShop or any Adobe product) has a feature that may or may not be available in PS. If you overlay two original images, this feature of my editing software will produce a third image that highlights any individual pixel in the two overlayed originals that is not identical between them.

I saved a jpg over and over again (each time with a new file name) without making any intervening editing changes. The only difference between the first image and the last was a series of "lossy" saves. I then overlayed the original image with one of the later saves in the series. It seemed to me that this was a more accurate method to detect whether there were any changes. Rather than relying on my ability to detect subtle changes by "eyeballing" side-by-side images, this presumably was a very accurate computer-based measurement of pixel by pixel differences.

With a half dozen or so successive saves, the program detected no differences. However, there were subtle differences after ten or a dozen successive saves. So I concluded that a handfull of successive jpg saves was nothing to be concerned about in terms of image degradation.

At the time, I didn't think to compare an original Tiff to the first save of that image as a jpg. That would certainly reveal whether the initial jpg saving step caused any degradation of what was there in the Tiff.

If this is a valid analytical process, I don't know whether you can duplicate it because, not being a PhotoShop user, I don't know if PS contains this feature of my software.

 

[Robert J. Gonzalez]

Starting now. Tiffs are useless, they just take up space on your harddrive.

In preferences in LR choose PSD, 16 bit, and prophoto as your parameters under external editing.

Right click an image in lightroom and select edit in photoshop. Do you editing and save and go back to lightroom and continue.

MrScary wrote:

Export as a Tiff or a RAW then save As Tiff. Work on the Tiff. When finished you can either save As a Tiff or JPG. If JPG, make sure you do NOT work on it again.

I'm sure someone will pop in and tell you something different, just to confuse you.

--
MrScary (DennisR)
Swansea, Wales. UK
http://russ4tography.com
http://www.flickr.com/photos/scarecrowdr/

 

[proseak]

Robert J. Gonzalez wrote "Tiffs are useless"

Why?

 

[Victor Engel]

JPEG should be considered as a final format, not a working format. Loss of quality resulting from saving to jpeg is normally satisfactory when it's used this way. The color model is changed, and textures of hue and lightness are represented as parameters to equations rather than color data. You can read up on the jpeg format for more details if you like.

But basically, it divides a picture into square blocks of 8 by 8 pixels. Depending on the compression level you select when saving, you will see varying degrees of blockiness in your picture, aligned with this 8-pixel grid. Sometimes you will see a sort of checkerboard-like pattern within each 8x8 block, too.

The compromises chosen by the designers of the jpeg format were chosen to take advantage of the blind spots in human vision. For example, we are more sensitive to high frequency in the lightness channel than in the hue channel (or similar items in other color models). Jpeg uses YCbCr, which is lightness and two chrominance (color) channels. The chrominance channels are compressed more highly than the lightness channel, normally.

The real problem comes in when you edit a file saved as jpeg where your edit exposes these compromises. Things like curves adjustments, hue and saturation adjustments can move boundary conditions from positions that are imperceptible to conditions where they show up clearly. A common example is deep blue sky.

 

[ShotFromTheSidewalk]

the first time i adjust the saturation in a sky on a jpg, i saw banding. this tells me the jpg is really not as good as it looks. now, depending on your methods, none of this may matter. or one day it might. do some testing with test images with fine details (u know, those lines that get closer and closer. find some colours that gradually change and see what happens when you make your usual adjustments.).

working all raw and tiff takes disk space galore and a pretty good computer. maybe don't do it all right now. just do the important photos that way. you could aim at that workflow sometime in the future.

 

[Victor Engel]

This is why Lightroom is good. It uses the raw files and never wastes space with tiff files unless you ask it to generate them. If it can't be done on Lightroom, you can use the Edit in Photoshop menu item, and Photoshop will open up the raw file using the ACR parameters Lightroom used. Do the work in Photoshop, then save the finished product as a jpeg. No Tiff file is ever needed. If you want to save your Photoshop work, though, you'd want to save as PSD or Tiff.

 

[Dylan Johnson]

I always save the raw files. I have 2TB storage and a Samsunng Ultrabook
--
Canon 50d gripped, 70-200 f2.8L IS USM, 50mm f1.8, 18-55 f3.5-5.6, 430ex II.

 

[Robert J. Gonzalez]

proseak wrote:

Robert J. Gonzalez wrote "Tiffs are useless"

Why?

The take up space on your hardrive as I already said so why did you ask why?

Plus they offer no image quality advantage over PSDs.

 

[Victor Engel]

Robert J. Gonzalez wrote:

proseak wrote:

Robert J. Gonzalez wrote "Tiffs are useless"

Why?

The take up space on your hardrive as I already said so why did you ask why?

Plus they offer no image quality advantage over PSDs.

They offer a wider compatibility, though.

 

[DotCom Editor]

CaptiVision Photography wrote:

I had a shoot with a friend the other day, and I did my usual editing and then I started thing about how I was exporting my files. These were the steps I took after the shoot:

1. Import my RAW files into LR4

2. Did my usual editing

3. Exporting in JPG format

You just threw away 50% of your data by saving to an 8-bit format. Further, you just went from 65,000 shades of R, G, and B to a mere 256 shades of each.

4. Imported those JPGs into Photoshop CS4

If you're moving images from LR into PS, there are better ways to do it than going to 8-but and a smaller color space.

5. Did my retouching on the images (No extra layers or masks were created)

6. Exported those JPGs in JPG format again at the maximum quality.

You just further compressed something you previously compressed, throwing away even more data from your images. Are bands starting to show up in the blue sky? Now you know why.

Now, my question is am I destroying the image quality that way? Should I be exporting everything as a TIF until my final save of the image after all editing? Should I go back and re-edit all the images or do you think the quality of those files are okay to give to a client (or friend in this case).

JPG is and END format, never, ever, ever and intermediary format. Use JPG only for posting images on websites, sending to a commercial lab, or embedding in an e-mail message to grandma. I must say that 99% of my work never exists in JPG at any time!