What am I doing wrong/right?

[DisneySuiteFreak]

Here is what happens when you save a JPEG of a fine gradient in photoshop with a medium level of compression:

Look familiar?

It sure does! I thought that banding thing was caused because I was editing in PS first and then doing the noise reduction. It didn't seem as bad when I did the noise reduction first and then edited. Although I could be imagining the difference. I honestly don't know what I even did to save the photos. I was guessing at what my options should have been when it prompted me to save in PS.

You should notice that the "banding" problem gets better in each picture. You may see some banding as monitors are not perfect and sometimes will show their own...but this is typically much different looking than the one cause by JPEG compression.

And to answer your question I use a spyder2 with Coloreyes Display Pro software to calibrate my monitor. Yes, it will work on a laptop screen. How worth it it would be with a laptop screen is not up for me to decide...just know that they are often poor in quality.

Sorry if it's seems like a dumb question, but is a .png the name for a photoshop file? How do I save a file without using too much compression? When saving, I checked maximum, large file and set it to 10 but I wasn't sure what the rest of the stuff meant. (I could have saved at up to 12.) There is baseline standard, baseline optimized and progressive scans with the pulldown menu. I selected progressive scans and 5 - not even knowing what progressive scans are. Was that the wrong setting? I would rather save to get the best results with the least amount of compression. I know it must seem really dumb, but up until now I have been too intimidated to even attempt photoshop and didn't even know where to begin in asking questions for help, etc. Thanks for your patience...Oh and thanks for the calibration software info.

 

[NostalgicDad]

So if I had taken the photo with an aperture of say f14, it would be okay to do a 30 second exposure? (Depending on lighting conditions, of course)
Do you do Noise reduction first and then PP in Capture NX2 or do you do the reverse? I think that's part of what I'm confused about - what order do you do the processing?

Do you use a calibrated monitor? I'm amazed that no one else is seeing/saw what I saw. I guess I need to calibrate my monitor...

ETA: I just reread your earlier post where you said you wouldn't go past f11. So if I use f11 depending on lighting I could do a longer exposure and wouldn't have to worry about diffraction issues?

It's not that I absolutely would not go higher than f/11. Diffraction issues will vary from lens to lens. I simply use that as a rule of thumb but also become intimate with each of my lenses to know their limits. As far as how long to expose, that's simply going to depend on the available light and the scene, not necessarily your aperture setting.

Regarding your NR question.......For images where I think NR is needed, I typically perform that at the end of my PP. However, this is a point of contention with lots of folks. You will get differing opinions on this as to do it first or last. In Jason Odell's book on Capture NX2, he recommends performing NR prior to any other PP. I have simply had better luck doing the opposite, though. It's something you just need to practice and get comfortable with in your own workflow. I'm honestly not a fan of using NR anyway unless I think it is absolutely necessary.

 

[DisneySuiteFreak]

It's not that I absolutely would not go higher than f/11. Diffraction issues will vary from lens to lens. I simply use that as a rule of thumb but also become intimate with each of my lenses to know their limits. As far as how long to expose, that's simply going to depend on the available light and the scene, not necessarily your aperture setting.

Regarding your NR question.......For images where I think NR is needed, I typically perform that at the end of my PP. However, this is a point of contention with lots of folks. You will get differing opinions on this as to do it first or last. In Jason Odell's book on Capture NX2, he recommends performing NR prior to any other PP. I have simply had better luck doing the opposite, though. It's something you just need to practice and get comfortable with in your own workflow. I'm honestly not a fan of using NR anyway unless I think it is absolutely necessary.

Thanks Danny. I guess I just have to practice, practice, practice!

 

[VVFF]

It sure does! I thought that banding thing was caused because I was editing in PS first and then doing the noise reduction. It didn't seem as bad when I did the noise reduction first and then edited. Although I could be imagining the difference. I honestly don't know what I even did to save the photos. I was guessing at what my options should have been when it prompted me to save in PS.



Sorry if it's seems like a dumb question, but is a .png the name for a photoshop file? How do I save a file without using too much compression? When saving, I checked maximum, large file and set it to 10 but I wasn't sure what the rest of the stuff meant. (I could have saved at up to 12.) There is baseline standard, baseline optimized and progressive scans with the pulldown menu. I selected progressive scans and 5 - not even knowing what progressive scans are. Was that the wrong setting? I would rather save to get the best results with the least amount of compression. I know it must seem really dumb, but up until now I have been too intimidated to even attempt photoshop and didn't even know where to begin in asking questions for help, etc. Thanks for your patience...Oh and thanks for the calibration software info.

Saving at JPEG compression level 12 and the "optimized" setting will generally do a ok job of file size but no "banding" issues. .png is a completely different file format that has been around for quite some time. It is a lossless form of compression so you get all the quality of a bitmap without the file size. I use that format where quality is my primary concern.

Another solution would be to shoot in RAW and then convert to a file format. That way you are not compressing the file twice by saving in JPG and saving in JPG again.

 

[DisneySuiteFreak]

When saving, I checked maximum, large file and set it to 10 but I wasn't sure what the rest of the stuff meant. (I could have saved at up to 12.) There is baseline standard, baseline optimized and progressive scans with the pulldown menu. I selected progressive scans and 5 - not even knowing what progressive scans are. Was that the wrong setting?

Saving at JPEG compression level 12 and the "optimized" setting will generally do a ok job of file size but no "banding" issues. .png is a completely different file format that has been around for quite some time. It is a lossless form of compression so you get all the quality of a bitmap without the file size. I use that format where quality is my primary concern.

Another solution would be to shoot in RAW and then convert to a file format. That way you are not compressing the file twice by saving in JPG and saving in JPG again.

Thanks again for the info. One more question regarding above, when saving, could I have selected baseline standard, baseline optimized or progressive scans 5 (which is what I did select). What are 'progressive scans' ?
I eventually want to shoot in RAW, but I have to upgrade my software and don't want to spend the mney right now. So every time I save my JPEG file it compresses the file? That sucks. So if you save as a .png file, in order to post on the web or in this forum even, wouldn't you have to then convert the .png file into a JPEG file?
Thanks again!

 

[VVFF]

Thanks again for the info. One more question regarding above, when saving, could I have selected baseline standard, baseline optimized or progressive scans 5 (which is what I did select). What are 'progressive scans' ?
I eventually want to shoot in RAW, but I have to upgrade my software and don't want to spend the mney right now. So every time I save my JPEG file it compresses the file? That sucks. So if you save as a .png file, in order to post on the web or in this forum even, wouldn't you have to then convert the .png file into a JPEG file?
Thanks again!

If you really care about quality and are editing a picture...save the resultant file as a png. Pngs are entirely lossless but still compressed and are the best bet for photos. Many photo sites and programs can handle png so it shouldn't be an issue for you when you decide to go and upload that file. For instance, photobucket accepts png files.

As for jpeg format. I have the best luck with jped optimized. It is slightly smaller than the standard but typically looks the same. There have been some programs that have trouble with optimized but its very rare and I've never encountered this problem. Progressive is the worst option actually.

 

[Groucho]

Groucho, I am using my laptop which has an LCD monitor. My regular computer has an LCD screen as well. What to do?...CRT monitors are the older hulking heavy ones, right? What do you recommend? I actually gave my big monitor to my son for his computer so I could use a smaller LCD.
Glad you are not seeing much noise in the photos! YAY!

Well, laptop LCDs are just about the worst for judging noise and such... look at them just a tiny bit above or below straight-on and you'll get color shifts, some of which will unrealistically exaggerate the noise in a photo.

Desktop LCDs are generally better, but the strength of LCDs (the extreme crispness) can be a detriment when actually looking at photos! Each pixel is a fairly well-defined square and so you're more likely to notice jagged curves, and pixels of similar color will not "blend" as much as you'd see on a CRT monitor. Cheaper and older LCDs also often don't show colors as well as CRTs and better LCDs.

I'm not suggesting throwing away your LCD for a, yes, big old hulking CRT, but I would remember that how you're seeing it on the screen is not quite how it'd look if you printed it.

FWIW, I'm using calibrated displays; I don't think that would make much of a difference when you're looking at noise.

 

[VVFF]

Well, laptop LCDs are just about the worst for judging noise and such... look at them just a tiny bit above or below straight-on and you'll get color shifts, some of which will unrealistically exaggerate the noise in a photo.

Desktop LCDs are generally better, but the strength of LCDs (the extreme crispness) can be a detriment when actually looking at photos! Each pixel is a fairly well-defined square and so you're more likely to notice jagged curves, and pixels of similar color will not "blend" as much as you'd see on a CRT monitor. Cheaper and older LCDs also often don't show colors as well as CRTs and better LCDs.

I'm not suggesting throwing away your LCD for a, yes, big old hulking CRT, but I would remember that how you're seeing it on the screen is not quite how it'd look if you printed it.

FWIW, I'm using calibrated displays; I don't think that would make much of a difference when you're looking at noise.

The reason I brought it up in this case is that the noise he is referring to is in a very dark area. If you haven't calibrated your display this entire area could be clipped and make it impossible to see the gradients that do exist in the image.

 

[DisneySuiteFreak]

If you really care about quality and are editing a picture...save the resultant file as a png. Pngs are entirely lossless but still compressed and are the best bet for photos. Many photo sites and programs can handle png so it shouldn't be an issue for you when you decide to go and upload that file. For instance, photobucket accepts png files.

As for jpeg format. I have the best luck with jped optimized. It is slightly smaller than the standard but typically looks the same. There have been some programs that have trouble with optimized but its very rare and I've never encountered this problem. Progressive is the worst option actually.

Thanks! Back to the drawing board!

 

[DisneySuiteFreak]

Well, laptop LCDs are just about the worst for judging noise and such... look at them just a tiny bit above or below straight-on and you'll get color shifts, some of which will unrealistically exaggerate the noise in a photo.

Desktop LCDs are generally better, but the strength of LCDs (the extreme crispness) can be a detriment when actually looking at photos! Each pixel is a fairly well-defined square and so you're more likely to notice jagged curves, and pixels of similar color will not "blend" as much as you'd see on a CRT monitor. Cheaper and older LCDs also often don't show colors as well as CRTs and better LCDs.

I'm not suggesting throwing away your LCD for a, yes, big old hulking CRT, but I would remember that how you're seeing it on the screen is not quite how it'd look if you printed it.

FWIW, I'm using calibrated displays; I don't think that would make much of a difference when you're looking at noise.

Thanks Groucho!