Pages: [1]   Go Down

Author Topic: Jpeg work flow  (Read 6799 times)

DoDa!

  • Newbie
  • *
  • Offline Offline
  • Posts: 2
Jpeg work flow
« on: September 21, 2007, 05:34:36 pm »

Two questions

1)  I know the first words out of some mouths when I say I only save in JPEG will be ... You should be shooting RAW!, but for my purposes at this time I don’t need to....  What I do need is a better work flow! And how should I save/backup?   After I download ( Canon 30D, Large Jpeg/ no compression) should I convert to a tiff and backup creating a non-lousy file for future or should I just make a duplicate JPEG to work with and backup what I’ve downloaded ( I also burn to disc as well / no compression)

2)  I’m I right in thinking that every time I open and close an jpeg that it losses quality when re-compressed.  So if it started at  X number of pixels and compresses to say 80% the next time it repeats the 80% ( of the 80%) thus it is 64% of original so that opening and closing the file a few times make it just a fuzzy dot?
Logged

Morgan_Moore

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2356
    • sammorganmoore.com
Jpeg work flow
« Reply #1 on: September 21, 2007, 05:56:59 pm »

Quote
two Qs
[a href=\"index.php?act=findpost&pid=141076\"][{POST_SNAPBACK}][/a]

Saving jpgs is fine - just always work on a copy - you wont gain what isnt there by saving as tiff - just fill your HD/cd/dvds up

You could even change the 'properties' of the originals to 'read only' to stop acidental mungling

Degradation only happens when you modify a file

so open, modify, save, open, modify save ruins it slowly but open close open close doesnt

(I think!)

and yes you should be shooting RAW!

or you could shoot RAW and Jpeg and just archive the raws for when you decide it was a good idea after all



SMM
Logged
Sam Morgan Moore Bristol UK

Jonathan Wienke

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5829
    • http://visual-vacations.com/
Jpeg work flow
« Reply #2 on: September 21, 2007, 06:02:46 pm »

Quote
You should be shooting RAW!, but for my purposes at this time I don’t need to....  What I do need is a better work flow!

You should be shooting RAW, for many reasons, workflow being a very significant one. With RAW, you can buy a Color Checker and color calibrate ACR to your camera, which will save you a huge amount of time fixing colors the camera didn't get right. You also have much greater control over the look of your images, and can copy the processing settings from one image to many others at once, something you can't do with JPEGs. And then there's the general image quality improvement you get by converting RAWs to 16-bit and editing and saving them as 16-bit PSD or TIFF. Only save a copy of an image as JPEG, so that you have a higher quality version available for future editing if necessary.
Logged

PeterLange

  • Guest
Jpeg work flow
« Reply #3 on: September 23, 2007, 05:25:36 am »

Quote
... or should I just make a duplicate JPEG to work with and backup what I’ve downloaded ( I also burn to disc as well / no compression)
Makes sense for me.

/> Duplicate the image in Photoshop
/> Change the copy to 16 bit/ch for further editing.
/> Careful noise reduction and re-sharpening might be among the first to do.
/> Brightness, Contrast, Saturation, etc... may follow. Working with Actions saves a lot of time.
/> If you have adjustment layers it can make sense to save a psd master, but if you’re somewhat sure that you like the results, merge the layers, goto 8 bit again and save as high quality JPEG.
/> So everything stays with just two nice, small files: the original JPEG and the edited version. For minor belated changes, the edited version might be good enough and double JPG saving won’t tear the pixel apart (at least not with regard to final print). For a major rework, start with the original again.


Raw of course has it’s advantages, no doubt about this, but IF you strive for ACR calibration you should better know what you are doing. Popular scripts i.e. the Fors script are conceptually flawed which makes it a gambling with colors. As simple as that. We could discuss down to the level of sensors’ spectral response if needed.

Also, as far as JPEG quality is concerned I would not listen to people who bought their last camera years ago. Here’s a more balanced view, very much in line with my measurements, particularly referring to Canon's digic III processor:

>> I usually only shoot raw, but a half hour spent with the latest version of Canon's DPP software when I first got the camera showed me that I was not going to be happy working with DPP. Image quality from DPP isn't the issue. It's fine. Indeed it may be as good as it gets. But the workflow is primitive compared to Camera Raw, Lightroom or Capture One. So, I decided to shoot combined raw + Fine Large JPG, just in case.

Interestingly, though I left the in-camera JPGs at the default settings I found that they produced highly usable files. Tonal rendition, saturation and sharpening were all pretty much dead on and very close to what I could produce with DPP unless the image was technically problematic. I was fortunate to have a pre-release of one of my preferred raw processing programs (coming soon) that has support for the 40D. I used these to process my raw files. <<

Michael Reichmann on the 40D
http://www.luminous-landscape.com/reviews/...D-handson.shtml


Good luck
& best regards, Peter

--
Logged

juicy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 254
Jpeg work flow
« Reply #4 on: September 23, 2007, 05:49:00 am »

Hi!

Quote
Popular scripts i.e. the Fors script are conceptually flawed which makes it a gambling with colors. As simple as that. We could discuss down to the level of sensors’ spectral response if needed.

Has this already been discussed somewhere to the deeper level and if so, could someone provide a link/thread etc?  If not, then I would encourage doing it here in LL, maybe in another thread.

Cheers,
J
Logged

PeterLange

  • Guest
Jpeg work flow
« Reply #5 on: September 23, 2007, 12:27:44 pm »

Do a forum search or start here with bjanes’ posts #32 and 36:
http://luminous-landscape.com/forum/index....pic=17064&st=31
Logged

juicy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 254
Jpeg work flow
« Reply #6 on: September 26, 2007, 07:49:02 am »

Hi!

Quote
Do a forum search or start here with bjanes’ posts #32 and 36:
http://luminous-landscape.com/forum/index....pic=17064&st=31
[a href=\"index.php?act=findpost&pid=141416\"][{POST_SNAPBACK}][/a]


Thank you very much, I had forgotten that thread and was too stupid not to search before asking...  

Cheers,
J
Logged

Jonathan Wienke

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5829
    • http://visual-vacations.com/
Jpeg work flow
« Reply #7 on: September 26, 2007, 10:34:43 am »

Quote
IF you strive for ACR calibration you should better know what you are doing. Popular scripts i.e. the Fors script are conceptually flawed which makes it a gambling with colors.

That's quite a bit of an overstatement. The Fors sript does an excellent job of finding the best possible calibration values for ACR if you give it a properly-shot RAW to work with. I've calibrated 4 cameras with the Fors script, (Olympus SP-350 and Canon 1Ds, 1D-MkII, and 10D) and I get consistent and true-to-life color from all of them. I've shot jobs with 3 cameras, and not being able to tell which image was captured with which camera from color differences is a nice thing. ACR's calibration system is more limited than full-on profiling, but for most users it will significantly reduce or eliminate the need for doing remedial color correction in Photoshop if WB is set right in ACR. It also is more useful in a variety of lighting conditions; full-on ICC profiling can result in greater color accuracy, but the profile is only valid for the lighting conditions under which it was made, and switching to a different lighting will deliver worse results than ACR's calibration method.
Logged
Pages: [1]   Go Up