What to do 1st with a jpeg file

[R Schade]

I have a question, but the more post processing I do, the more I feel I'm not getting it, so here goes.
I want to process a photo specifically a jpeg, I know, not the most wonderful thing to do, but this is my starting point. So I have my jpeg and it's right from the camera to PS, now the image resolution from the camera is 72dpi, is it better to process the file at this resolution or to up the resolution and then post process this jpeg, I am not sure whether this is an issue or not. I am basically de-noising it cropping, changing the resolution to 300 dpi, sharpening and posting and on occasion printing it in a SRGB. Or should I crop de-noise and sharpen.
Thanks, Russ

[J Rabin]

Always best to look at the image, decide what it needs, and work in order:

1. Noise reduction 1st, before any other editing (especially color correction).
2. Fix bigger problems before minor image edits (color correction, WB, etc).
3. Fix "Global" image weakness before localized editing (tone & contrast before shadow & highlight, etc.)
4. Sharpening later.

dpi/ppi resolution means nothing viewing on screen. Only important before sharpening and output.

[snokid]

only other thing is save as tif or psd.

Bob

[René Damkot]

...and be sure to keep a backup of the original...

[J Rabin]

only other thing is save as tif or psd.
PSD is a just a proprietary Adobe TIF format. Same-same, it's a user preference choice depending on what you're purposing file to.
This fellow is starting with JPEGs It's kind of silly to make TIF, no quality gain. At end of work, just Flatten and Save As JPEG with highest quality/no compression.

[René Damkot]

no quality gain. At end of work, just Flatten and Save As JPEG with highest quality/no compression.

Well, saving as max quality jpg is fine, unless you're going to open and edit, and save and close and open and edit and save and close and open and edit, and save and close again...

In that case, a tiff or.psd file does have advantages.
Off course, a tif or psd also allows you to save (adjustment)layers....

[PhotoJourno]

As mentioned above, the first thing I do with an image, is SAVE AS, so that I can not only convert it to PS format or TIF, but also so that the original remains untouched. Early on my Digital photo career (just after jumping from film) I must have modified 50% of my files, without ever saving a copy of the original (Which is handy when the end results do not satisfy you).
Other than that, I think the list above pretty much sums it up.
Cheers,

[agaupt]

now the image resolution from the camera is 72dpi

I could be totally wrong but I don't think the image resolution from the camera is 72dpi.

Computer screens maximum dpi is 72 dpi, I think photoshop defaults to that setting. A picture with say 500 dpi will look the same as 72 dpi on the computer, the file size will just be much bigger and take longer to load. But if you printed those same pictures the there would be a major difference between the two.

dpi is only really important for printing.

[gjl711]

DPI has nothing to do with file size, or the way it looks on a monitor. Monitors do not have DPI, they have PPI. It is a completely arbitrary number until printing time. Ignore the value for now. Do all you post processing and get the pic the way you want it. When it comes time to print, set you dpi appropriately to the device your printing on.

[ebann]

I could be totally wrong but I don't think the image resolution from the camera is 72dpi.

Computer screens maximum dpi is 72 dpi, I think photoshop defaults to that setting. A picture with say 500 dpi will look the same as 72 dpi on the computer, the file size will just be much bigger and take longer to load. But if you printed those same pictures the there would be a major difference between the two.

dpi is only really important for printing.

Your camera has a fixed maximum resolution, e.g. 1600x1200 for a 2MP camera. Now the difference is how that image is displayed. Monitors follow the 72dpi rule which means that it only takes 72 dots to cover one inch on the monitor screen. That's why a low-res image is usually good enough for websites. Your printer is capable of much higher (or finer) image quality and people usually follow the 300dpi rule. Why? Because 300dpi is commonly accepted as being "photographic" image quality, i.e. nearly indistinguishable from a film printed photo. So, what does this all mean? Practically, to display a 4"x6" image on a computer screen, all you need is a 288 pixels by 432 pixels image which any sub-megapixel camera can provide. But to *print* a 4"x6" on photo paper, you would need a 1200 pixels by 1800 pixels image which a 2MP camera can produce without any problems. If you try to view the 1200x1800 image on your computer screen, you get that common effect of having scrollbars on the bottom and right side of your Photoshop window because the image is simply too large at 72dpi (in effect a 16.6"x25" "monitor" quality image). Similarly, the 288x432 image on photo paper will only be roughly 1"x1.5" photo at "photographic" quality.

In other words, if you never do 8"x10" prints and only do 4"x6" or even 5"x7" prints, all you really need is a 3MP camera! 5"x7" = 1500x2100 = 3,150,000 pixels = 3.1MP! (hint: that's why I still use D30).

One last tip. If you wanted to print 8"x10" from a 3MP camera, you would need 2400x3000 pixels = 7.2MP in order to print at photographic quality (300dpi). Is it possible to do that with a 3MP camera? Yes! How? Using bi-cubic interpolation process available in Photoshop. That will allow you to "increase" resolution but not necessarily increase detail. The result is acceptable but not perfect.

The JPEG from your 20D should already have done some picture processing like sharpening, de-noise, color/tone adjustments, etc. RAW format has none of that and would require later de-noise, sharpening, color correction, etc. But with JPEG, I suggest you to apply whatever you like and only then resize if you are going to use it on a website.

Another thing. JPEG is a lossy format meaning some detail may be lost as well as it may have compression artifacts. If you edit the image, save the result as a TIFF so that it maintains the exact alterations. If you save it as JPEG once again, you are *again* re-introducing more detail loss as well as more compression artifacts. RAW is lossless format, usually converted to TIFF after adjustments are made with RAW converters.

I hope I answered your question somehow.

[R Schade]

Well That is quite a bit of information and I do appreciate it. I was most interested in that there was going to be a problem principally with the resolution and the work flow.
That said I see ther really isn't any point in changing the resolution beyond changing it unless I decide to print something. It appaers that it is a moot point when deciding when it would needed to be changed when I post it to the net and that saves me a few steps. I have gone back through most of the old post and was able to find the old 72dpi resolution links that further supports your statements.
Yes I was using the jpeg as only a starting point. I do esspecially with Basketball and events like relay for life thus using jpeg for speed and storage resources and then there are a few I keep as a hard copy and then I do save them as a TIFF and process them. Even so, when I shoot jpegs, I capture them at the lowest compression/fine so if I do process them I don't loose very much. That point was not lost on me.
So my work will go something like this
1capture,2de-noise,3minor level or color operations(only if it needs it), 4small dose of sharpening and 5save to theweb at size for output 800x533.
Does that sound about right?
Thanks again,
Russ

[ebann]

Your workflow could be something like this:

1. Download image from your camera into your preferred image organizer, e.g. Photoshop Album, iPhoto, Portfolio, etc.
2. Delete images that are aweful and keep the good and excellent.
3. Load up your denoiser program, e.g. Noise Ninja, and remove noise, save result as TIFF.
4. Load up Photoshop and perform level, color, sharpness, etc. corrections. Save result as TIFF. This TIFF will be your corrected master image. The original JPEG will be the "raw" master in case you decide to process the image in a different way. Keep both. Some people will use the TIFF and then chuck it away to save harddrive space.
5. Resize the TIFF and save as JPEG with another name for website use. If someone wanted a copy of that image, you would send them the TIFF file.

note: 20D JPEG's are of excellent quality, hardly indistinguishable from RAW in terms of detail and artifacts. Not a bad idea to shoot JPEG. Only drawback is that JPEG is 8-bit image file while RAW is 12-bit. What does that mean? It means that RAW has more dynamic range meaning more detail in the dark areas and light areas which can be "extracted" by doing EV corrections in Photoshop. If you tried this on the JPEG, you'll get blown-out highlights and shadows.

Hope it helps.

[René Damkot]

AFAIK, a RAW doesn't have more DR, merely more 'leeway'...
Only difference in DR would be caused by the curve used to process the jpg.

[ebann]

AFAIK, a RAW doesn't have more DR, merely more 'leeway'...
Only difference in DR would be caused by the curve used to process the jpg.

Hello René! You raised a doubt in my head so I did some googling... here's everything you need to know about this stuff:

http://www.normankoren.com/digital_tonality.html

In essence here's a few quotes that summarizes pretty well what I thought when I wrote my post:

"When you store images in RAW format you can also take advantage [of] your camera's hidden dynamic range; you can access tones that may be obscured or lost when images are stored in a standard file format such as a JPEG-- the default for most digital cameras."

" JPEG is a lossy compressed format that sacrifices a small amount of information to achieve a large savings in storage. But it is not this information loss that concerns us here; it is the loss of tonal levels that takes place when the image sensor's digital output is converted to any standard 24-bit color file format; JPEG is merely the most common. Image sensors in high quality digital cameras have a bit depth of at least 12; they have 2^12 = 4096 discrete tonal levels. Standard 24-bit color files have a bit depth of only 8; they have only 2^8 = 256 discrete tonal levels. Tonal levels are lost in the conversion. The techniques presented here minimize this loss-- they maximize tonal detail throughout the RAW conversion and image editing process."

RAW has more dynamic tonal range than JPEG.

So, this "leeway" you mention is merely the possibility of salvaging an underexposed/overexposed image due to the fact that the detail/tonal information is still there in the RAW file and doing any EV correction won't bloom the light or dark areas.

I might even add that the human eye can see a good amount of tonal levels, the computer monitor less so, and the printed photo paper even less so. So the end goal is to get a RAW file and "extract" the nice details from the shadows and "extract" the nice details from the highlights and "cram" them into the final image so that when printed, one can see the "wide" dynamic tonal range. If not done this way, the printer will not be able to reproduce the entire dynamic range of the RAW file and would lose details in the shadows and in the highlights. This process is called tonal-mapping.

The OP should take a look at the picture in the main page of http://www.hdrsoft.com/ too see what I mean. Just pretend that the 3 images used to create the final image was "extracted" from a single RAW file. The idea is the same but technically extracting from a single RAW is not truly HDRI... oh well... perhaps the OP will enjoy this aspect of photography... I surely did.

There... I hope I didn't bore you to death.