Don't shoot above ISO 1600!

Question for you all... high iso shooters out there... is there a difference you have seen in the sequence on final output of sharpening and noise reduction. Seems to me noise should be done first so that sharpening does enhance the effects of the noise. But I've see way too many men with no arm hair, and wondered if the sequence would help with that. Just wondering.... I general NR first, then sharpen, but with the masking tools, If gone to sharpen first because it yields in my eyes more natural looking skin... love to hear your thought.

Croasdail wrote in post #19080454
Question for you all... high iso shooters out there... is there a difference you have seen in the sequence on final output of sharpening and noise reduction. Seems to me noise should be done first so that sharpening does enhance the effects of the noise. But I've see way too many men with no arm hair, and wondered if the sequence would help with that. Just wondering.... I general NR first, then sharpen, but with the masking tools, If gone to sharpen first because it yields in my eyes more natural looking skin... love to hear your thought.

Regardless of noise-handling, I've always heard that sharpening is last. I think that's the rationale of why (for example) LR has a(nother) sharpening option in the Export dialog.

There is export sharpening, which can be applied depending on the medium you are exporting for. Prints can take more sharpening than display and even the type of paper can play a role.

But for sharpening as a part of processing I tend to go back and forth. I will apply sharpening, look to see what it has done to the grain/noise and then adjust my noise reduction. If applying the noise reduction changes the visual sharpness I go back and play with that again. In lightroom the two panels are next to each other so you can tweak an image until its just right

dasmith232 wrote in post #19080541
Regardless of noise-handling, I've always heard that sharpening is last. I think that's the rationale of why (for example) LR has a(nother) sharpening option in the Export dialog.

I believe LR documentation as well as Lightroom guides by experts indicate that the order does not matter.

Croasdail wrote in post #19080454
Question for you all... high iso shooters out there... is there a difference you have seen in the sequence on final output of sharpening and noise reduction. Seems to me noise should be done first so that sharpening does enhance the effects of the noise. But I've see way too many men with no arm hair, and wondered if the sequence would help with that. Just wondering.... I general NR first, then sharpen, but with the masking tools, If gone to sharpen first because it yields in my eyes more natural looking skin... love to hear your thought.

I make sure contrast is correct first along with white balance, and will often employ USM to add some micro contrast. Then comes noise reduction. Then comes all the other filters, like unsharp mask or smart sharpening, saturation, color, etc. I might run one final run of very mild noise reduction at that point. I have too many hours at post processing for noise over the years, and if you do any real sharpening up front, it makes NR more difficult later, and some cases nearly impossible.

Not all sharpening is created equal either, you must watch what type of sharpening algorithm you select during the process.

texshooter wrote in post #19078291
Try shooting at 160-multiple ISOs (640, 1280, 2560, 5120) instead of multiples of 100 (800, 1600, 3200, 6400). And never shoot multiples of 125 (500, 1000, 2000, 4000). You'll give up a little dynamic range, but your RAWs will have less noise, when shooting at high ISOs. I have not tested this theory out yet.

Can someone explain this?

Is this true on all cameras or just 5DII? How to know which ISOs a real and which are real with software help?

Are some ISO multiples better than others even if the number is higher.

For example, 640 is a good multiple, 500 is bad multiple, so I should choose 640 even it is higher than 500?

tekin112000 wrote in post #19080865
Can someone explain this?

Is this true on all cameras or just 5DII? How to know which ISOs a real and which are real with software help?

Are some ISO multiples better than others even if the number is higher.

For example, 640 is a good multiple, 500 is bad multiple, so I should choose 640 even it is higher than 500?

This was my question earlier as well...

tekin112000 wrote in post #19080865
Can someone explain this?

Is this true on all cameras or just 5DII? How to know which ISOs a real and which are real with software help?

Are some ISO multiples better than others even if the number is higher.

For example, 640 is a good multiple, 500 is bad multiple, so I should choose 640 even it is higher than 500?

Multiples of 100 are "real" ISOs. So, 100, 200, 400, 800.

125 is just 100 pushed 1/3 stop and will be noisier than 100.

160 is just 200 pulled 1/3 stop and will exhibit less noise than 200.

mike_d wrote in post #19080875
Multiples of 100 are "real" ISOs. So, 100, 200, 400, 800.

Powers of 2 x 100; not multiples of 100. 500 is a multiple of 100, but is a firmware push from 400 on most Canons.

Those would be expected with even an ideal camera. What is an anomaly is the fact that the very deepest shadows at ISO 125 could be noisier than the deepest shadows at ISO 160, despite 1/3 stop more light.

With such cameras, the headroom relative to metering is inconsistent, too. ISOs 160/320/etc have enough headroom for JPEGs, but the RAW data has 1/3 stop less headroom.

Here is some more propeller head reading on the subject....

https://photo.stackexc​hange.com …-iso-of-a-canon-5d-mk-iii

This is all interesting... but I find I never use any of it. I've never had anyone say - nice image but way too much noise.... can't remember the last time that happened. Unlike when I shot Canon which I did shoot ETTR, with Sony if I can't get it perfect, I shoot ETTL. Sony has great detail hidden in the shadows that can be recovered.

It just depends on the camera. And at the end of the day, what your shooting makes a lot more of a difference than what native ISO a camera shoots at. Just my 2 cents.

It gets more complicated because some cameras are ISO Invariant while others have Dual iSO. And some cameras vary by how much analog amplication they execute versus digital interpolation. We're going down a rabbit hole I'm affraid.

https://www.youtube.co​m/watch?v=Vbxz_Dn1ElY

John Sheehy wrote in post #19080885
What is an anomaly is the fact that the very deepest shadows at ISO 125 could be noisier than the deepest shadows at ISO 160, despite 1/3 stop more light.

But there isn't more light, just more amplification.

Pippan wrote in post #19080921
But there isn't more light, just more amplification.

Not if you are exposing for the ISO.

John Sheehy wrote in post #19080926
Not if you are exposing for the ISO.

But ISO is the 3rd most important side of the exposure triangle for most people. I know the DOF I need, I know the SS I need, and I know how much I need it exposed. ISO just completes the equation. There's a reason there's no Iv setting on the top dial.

mike_d wrote in post #19080927
But ISO is the 3rd most important side of the exposure triangle for most people. I know the DOF I need, I know the SS I need, and I know how much I need it exposed. ISO just completes the equation. There's a reason there's no Iv setting on the top dial.

This thread started out talking about choosing an ISO setting for a manual exposure, then early on, it switched to shooting at ISOs. I already explained the way 1/3-stop pushed and pulled ISOs affect both scenarios.