jpg compression, saving and resaving?
Discussion
im still shooting .jpg. So shoot me 
Question is... if you take a shot, save.
Then load, use 2 tiny miniscule heal brush strokes then save again on the same 12-quality .jpg setting,
and repeat the above, load, apply minor heal brush strokes, and save on the same 12-quality .jpg setting a second time...
are you loosing much, or anything at all, between the second and third save, using all the same save settings?
Are you loosing anything in other areas of the image, other than the changes applied to the heal-brush area?
On another note, I had a few images with that casio calculator logo visible. Someone with a lot more PS skill than I loaded the .jpg image into cs, saved on 12-quality compression and sent back to me. the file size was increased by 100k IIRC. I use elements 3 (still
) Does PS, and PS-CS elements have/use different algorythms for processing data, meaning a load/save in one then the other will mean dataloss?
>> Edited by UKBob on Wednesday 10th May 01:43
Question is... if you take a shot, save.
Then load, use 2 tiny miniscule heal brush strokes then save again on the same 12-quality .jpg setting,
and repeat the above, load, apply minor heal brush strokes, and save on the same 12-quality .jpg setting a second time...
are you loosing much, or anything at all, between the second and third save, using all the same save settings?
Are you loosing anything in other areas of the image, other than the changes applied to the heal-brush area?
On another note, I had a few images with that casio calculator logo visible. Someone with a lot more PS skill than I loaded the .jpg image into cs, saved on 12-quality compression and sent back to me. the file size was increased by 100k IIRC. I use elements 3 (still
) Does PS, and PS-CS elements have/use different algorythms for processing data, meaning a load/save in one then the other will mean dataloss? >> Edited by UKBob on Wednesday 10th May 01:43
UKBob said:
im still shooting .jpg. So shoot me
Not a lot wrong with maximum quality jpegs, you just lose post processing abilities. For us recovering darkroom users ? That's a huge deal. For many shots that will be printed quickly and/or only published on the web ? Not so much.
UKBob said:
are you loosing much, or anything at all? Other than the changes applied to the heal-brush area?
Yes, you're adding compression artefacts each time. I'm pretty much at a loss as to why you would want to do this - as soon as you've got the images off the camera why not use .psd files ?
J111 said:Lots of reasons, none of them excusable.
are you loosing much, or anything at alleach time you resave over a .jpg?
Yes, you're adding compression artefacts each time. I'm pretty much at a loss as to why you would want to do this - as soon as you've got the images off the camera why not use .psd files ?
Dealing with 2+ images of the same file on the same PC
Having to back up 2+ images of the same file on the same PC
trying to remember which of the 2+ same files has had which changes...
inserting stock library keywords on the same 2+ files...
Terrible, I know
But why would you be creating NEW artifacts, when saving a massive image, and only changing a small portion of it, saving on the same setting? Surely Jpeg is intelligent enough not to loose **more** data than on level 12, **from** level 12, if level 12 is used? Lots to learn, i know
edited to add... ie, if I heal-spotted and changed a very small area in the top right hand corner 100 times, and saved on .jpg level 12 each time, surely the rest of the file would not be affected if the "level 12" remained the same each time?
>> Edited by UKBob on Wednesday 10th May 01:58
I do take your point about the irritation of dealing with multiple files. I would maunder on about the importance of good workflow but a) it's boring and b) I'd be a total hypocrite.
Because you're compressing it to level 12, then compressing it to level 12, and so on. Every change throws away a little more data from the whole image, there's no intelligence to know that you've only altered a little bit and to leave the rest at the same level of compression. Say you're losing 1% on each save (optimistic), by the time you've made 10 changes, you've lost 10% of your original data.
UKBob said:
But why would you be creating NEW artifacts, when saving a massive image, and only changing a small portion of it, saving on the same setting? Surely Jpeg is intelligent enough not to loose **more** data than on level 12, **from** level 12, if level 12 is used? Lots to learn, i know
Because you're compressing it to level 12, then compressing it to level 12, and so on. Every change throws away a little more data from the whole image, there's no intelligence to know that you've only altered a little bit and to leave the rest at the same level of compression. Say you're losing 1% on each save (optimistic), by the time you've made 10 changes, you've lost 10% of your original data.
Start using RAW now, and download rawshooter from pixmantec.
When I started with DSLR I always shot JPG, then after the switch to RAW it was like a whole new world.
I recently had to go back to an image I had taken in JPG for a client and felt frustrated at the limitiations on PP that brought.
Recently had a conversation with anthoer pro friend who ALWAYS shot JPG, never saw the point of RAW. After demoing rawshooter to him for 5 mins he switched mode on his camera there and then and has never looked back - he too regrets not switching earlier.
When I started with DSLR I always shot JPG, then after the switch to RAW it was like a whole new world.
I recently had to go back to an image I had taken in JPG for a client and felt frustrated at the limitiations on PP that brought.
Recently had a conversation with anthoer pro friend who ALWAYS shot JPG, never saw the point of RAW. After demoing rawshooter to him for 5 mins he switched mode on his camera there and then and has never looked back - he too regrets not switching earlier.
Bee_Jay said:
Start using RAW now, and download rawshooter from pixmantec.
When I started with DSLR I always shot JPG, then after the switch to RAW it was like a whole new world.
I recently had to go back to an image I had taken in JPG for a client and felt frustrated at the limitiations on PP that brought.
Recently had a conversation with anthoer pro friend who ALWAYS shot JPG, never saw the point of RAW. After demoing rawshooter to him for 5 mins he switched mode on his camera there and then and has never looked back - he too regrets not switching earlier.
Several people have given this advice recently, and although I'm a total numpty when it comes to post-processing I have taken it and also shoot RAW all the time - even I notice the benefit.
My additional advice would be: shoot RAW and buy more and bigger CF cards
Bee_Jay said:
....switch to RAW it was like a whole new world.
....I recently had to go back to an image I had taken in JPG for a client and felt frustrated at the limitiations on PP that brought.
....regrets not switching earlier.
That's it in a nutshell. Why the reluctance to shoot raw anyway?
edit to add: had you shot in raw, you could have corrected the white balance and exposure in your set you shot the other day in about 1 minute. As it is you can't use them as a set... see the advantage?
>> Edited by _dobbo_ on Wednesday 10th May 11:54
Bee_Jay said:Is that considered the best product? I assume there are dozens of products on the market. Part of the reason I 'cant be arsed' is Im not sure how it works, or what to choose. Im sure my canon came with some kind of raw processing program. Do you have to switch between PS and pixmantec, back and forth all the time? What of tiff's - ive read that photos should also be saved or edited as tiffs, cant remember why though
Start using RAW now, and download rawshooter from pixmantec.
Bee_Jay said:I didnt think there would be any point, all my shots are average at best, and technically far from perfect with a lot of room for improvement. Sounds like Im missing out on something.
had a conversation with anthoer pro friend who ALWAYS shot JPG, never saw the point of RAW. After demoing rawshooter to him for 5 mins he switched mode on his camera
_dobbo_ said:Yes. Ok, so RAW it is.
....switch to RAW it was like a whole new world.
....I recently had to go back to an image I had taken in JPG for a client and felt frustrated at the limitiations on PP that brought.
....regrets not switching earlier.
That's it in a nutshell. Why the reluctance to shoot raw anyway?
edit to add: had you shot in raw, you could have corrected the white balance and exposure in your set you shot the other day in about 1 minute. As it is you can't use them as a set... see the advantage?
re: the reluctance - I wanted to be sure "when the time came" (which is now, I guess) that whatever raw processing program I used was the best. Ive got yearS worth of unsorted files which will take me an age to sort out, and having double the amount of files really put me off.
I'll switch to raw then. How does it work, step by step? Im sure I read somewhere that PS cant load raw files (why on earth not??
) which is why a raw file program is needed. Right? So, import into rawshooter and do what, exactly? Everything, or just some things? When is PS needed? Can rawshooter do all your healing and cloning work, like PS?(must install PS, still using elements
) >> Edited by UKBob on Wednesday 10th May 12:10
OK,
You can open RAW up directly in photoshop (If you find you can't, download the component from Adobe) - but only one at a time unless you use CS2 (I think).
However, if you want to process a whole shoot in one go, with the same white balance settings and exposure comp applied to all simultaneously, the odd crop here and there etc then Rawshooter is way better from a speed and workflow perspective - plus it is free. The paid for version gives you more options (like cropping) and I am lead to believe that basic healing is on it's way too.
Typically after a shoot using Rawshooter we will quickly flick through the whole shoot, prioritise photos (ctrl-1 etc) and then make all of the major adjustments in RS (exposure comp, white balance, saturation, vibrance, shadow and highlight contrast etc.) either quickly individually or using cut and paste from one image to all with minor tweaks. Cropping is done in RS too (we use the Pro version). Once all is set, we kick off a convert to JPEG and sit back. Then if any need work in PS (usually cloning or healing) then we will do those individually, but we find that most of our stuff (80-90%) is good to go to the printers straight out of Rawshooter - it even does sharpening. Using this method Rob and I get through over 600 images in less than an hour from start to finish including conversion.
>> Edited by Bee_Jay on Wednesday 10th May 12:35
You can open RAW up directly in photoshop (If you find you can't, download the component from Adobe) - but only one at a time unless you use CS2 (I think).
However, if you want to process a whole shoot in one go, with the same white balance settings and exposure comp applied to all simultaneously, the odd crop here and there etc then Rawshooter is way better from a speed and workflow perspective - plus it is free. The paid for version gives you more options (like cropping) and I am lead to believe that basic healing is on it's way too.
Typically after a shoot using Rawshooter we will quickly flick through the whole shoot, prioritise photos (ctrl-1 etc) and then make all of the major adjustments in RS (exposure comp, white balance, saturation, vibrance, shadow and highlight contrast etc.) either quickly individually or using cut and paste from one image to all with minor tweaks. Cropping is done in RS too (we use the Pro version). Once all is set, we kick off a convert to JPEG and sit back. Then if any need work in PS (usually cloning or healing) then we will do those individually, but we find that most of our stuff (80-90%) is good to go to the printers straight out of Rawshooter - it even does sharpening. Using this method Rob and I get through over 600 images in less than an hour from start to finish including conversion.
>> Edited by Bee_Jay on Wednesday 10th May 12:35
We use JPEGs instead of TIFFs because most of our stuff is printed at 20"x30" max and it saves on files sizes and upload times to the lab.
If however we are doing anything large-scale or canvases etc then we try to use TIFFs.
However the images below on the walls were JPGs from an 8MP Canon 20D !!!
If however we are doing anything large-scale or canvases etc then we try to use TIFFs.
However the images below on the walls were JPGs from an 8MP Canon 20D !!!
Yep, didn't think they did, but JPG's loose data pretty badly quite quickly doing that.
The best current compression in my opinion is PNG. It's a much more advanced version of the GIF type format of compression. Just done a test and my PNG is perfectly accurate to my BMP source, and it's also 8x smaller than my 100% JPEG with altered pixels.
JPG is ok for source > final version, but always keep the source and edit that before finalising to JPG, because editing a JPG to re-save after it's added artifacting is not really ideal unless your working method doesn't have source/PDF backup.
Dave
>> Edited by Mr Whippy on Wednesday 10th May 15:28
The best current compression in my opinion is PNG. It's a much more advanced version of the GIF type format of compression. Just done a test and my PNG is perfectly accurate to my BMP source, and it's also 8x smaller than my 100% JPEG with altered pixels.
JPG is ok for source > final version, but always keep the source and edit that before finalising to JPG, because editing a JPG to re-save after it's added artifacting is not really ideal unless your working method doesn't have source/PDF backup.
Dave
>> Edited by Mr Whippy on Wednesday 10th May 15:28
Gassing Station | Photography & Video | Top of Page | What's New | My Stuff