RX10 III: best IQ f-stop(s) consensus...?

Horacecoker · Dec 28, 2016

rapick said:
Jeff Greenberg said:

Horacecoker said:

1. there is no advantage quality wise to use any aperture smaller than F4 at ANY focal length.

2. your images at 600mm are of no value...for F stop comparison purposes due to varying degrees of heat haze.

Click to expand...

=====

Thanks for response!
1. Is there IQ disadvantage below f4?
(besides shallower DOF)
2. thought bottom edge details were close
enough to eliminate rising hot air issues -- sorry!!!
Will shoot at f5.6 or less except rare cases
requiring extreme DOF inches-to-infinitiy...

Click to expand...

Jeff,

I viewed your pics (thanks for sharing!!!) 1:1 on my 21.5" iMAC and came to the same conclusions as Horace: IQ starts to worsen above f/4 (degradation at f/5.6 is very slight, though).

On the other side I can't see any lower quality in the pics shot "wide open". can you?

I was surprised to view so much heat haze in the 600 mm pics. Very hot location? I can see a LOT of A/C condensers on the buildings roof.

--

Thanks for your input. I was quite confident to be a lone voice but it is always nice to have somebody on board who sees what I see.

You are right about the A/C condensers and this is what made me give up trying to determine any real differences in the 600mm focal length images. But I can only surmise the results will be identical to the other focal lengths. Especially as I know from experience that my RX10iii is pin sharp at F4.

David

Horacecoker · Dec 29, 2016

As promised here are some 100% crops from Jeff's test shots. I think I've successfully been able to demonstrate the best F stop for image quality on the RX10iii and also show the degrading effect to image quality due to diffraction on 1" sensors. In this case stopping down the lens from F4 by only two stops to F8 has blurred away detail that can clearly be seen at F4 but is next to invisible at F8.

At F4 most people should be able to see that the the coloured patterns above each of these three windows is not paint but a mosaic of small tiles.

At F5.6 people may now struggle to discern the tiles and at F8 it is virtually impossible.

At F8 you would think it was just paint!

I have included a same pixel size image of a shot taken at 200mm to prove they are indeed tiles for anybody who is doubtful.

100mm 1/1250 F4 ('sweet spot' aperture)

100mm 1/500 F5.6 (detail going)

100mm 1/250 F8 (detail gone?)

200mm 1/1250 F4

My findings aren't a surprise to me as people who read my posts know I've been recommending folk not to go above F5.6 with 1"sensor cameras unless they really need to due to the problem of diffraction. I've also stressed depth of field is plentiful at F4 and F5.6 on 1" sensor cameras and you also get the benefits of lower ISOs or faster shutter speeds by keeping to lower F numbers.

I'll just include a sentence from DPR review of the RX10iii that backs up all I've written on the subject.

The RX10 III's lens is clearly sharper, but it has another thing going for it: its faster maximum aperture helps it combat diffraction. Remember that F4 and F5.6 on 1"-type cameras are equivalent to F11 and F15, respectively.

Thanks for doing the tests, Jeff. Wish you hadn't started another thread though. Ideally it should have all been kept here.

David

elliottnewcomb · Dec 29, 2016

Thanks to both Jeff and David for all the detailed work on this. I am wondering if you could give us some further guidance, (either from rules you can copy and paste, or just best guidance from experience, iow, not asking for you to do further tests for exact answers):

if using f4.0 as smallest aperture as a general rule:

1. detail was lost at a distance in these examples, how much is lost due to diffraction at closer shots, typical indoors, street shots, i.e. subject within 25 feet?

2. when close, sufficient depth of focus using f4.0?

a. landscape, anything a distance more than ______? feet away, depth of focus will be enough

b. shooting indoors, f4.0, i.e. near a single kid, stay _______? feet away from their face to get their whole body in focus. for a group of kids playing a game on the floor, stay ____? feet away to get them all in focus?

3. stay f4.0 and use ND filter to reduce light when desired/needed?

this is where I want to add the 'good enough' factor. You are a perfectionist, I am not nearly as sensitive, or concerned about lost detail unless it is my primary subject, so, I could carry and use an ND filter to help tame too much light, like at the beach, or, I could simply use say f8. ???

......................................

the rx100's, same sensor, different lens:

safe to assume diffraction is already occurring at f5.6?, therefore better to try and limit no smaller than f4.0 like the RX10m3 examples shown here?

thanks in advance for any guidance.

--
Elliott

edform · Dec 29, 2016

Horacecoker said:
BTW your images at 600mm are of no value (that's a kinder way of saying worthless) for F stop comparison purposes due to varying degrees of heat haze.

While the heat-induced mush in most of the area of the 600mm shots takes away the ability to judge stuff like corner versus centre IQ differences, the machinery on the roof in the left lower foreground shows the lens to be stellar in performance. Below is a 100% crop of that area...

Taking this into account with all of the other images, Jeff should be assured that the lens hanging off the front of his RX10-III is a very fine beast indeed and it would appear that its best performance is actually wide open at all focal lengths!

I'm out of the Sony world at the moment, having gone over to Nikon 1 to get the long focal length capabilities that I want and good enough focus accuracy and speed to combat my cack-handed tendencies, but the RX10-IV that looks likely to appear shortly with the hybrid focusing already seen in the RX100V is definitely going to tempt me back.

--
Ed Form

elliottnewcomb · Dec 29, 2016

Ed,

stay here for helping us, and stay in the mini-challenges, you will be back!

the rumored Nikon DL24-500 specs seem tempting, any comments?

--
Elliott

edform · Dec 29, 2016

elliottnewcomb said:
Ed,

stay here for helping us, and stay in the mini-challenges, you will be back!

the rumored Nikon DL24-500 specs seem tempting, any comments?

That's actually the camera I have my eye on. The focus ability and the high speed burst mode capabilities, which I've already experienced with my Nikon 1 V1 and V2 are very attractive indeed. But the 24-500mm lens will have to give equal IQ to the RX10 or I'll save up the extra and go for the RX10-IV. Jeff's lens is a peach and no mistake.

elliottnewcomb · Dec 29, 2016

I have my 'no bigger than jacket pocket size' rule, but, a 1" zoom is so desireable, and, the lack of an evf on the Nikon is a deal breaker, the optional evf might be the answer, as it is on my RX1r, but, it isn't even available yet.

Nikon is a bit smaller than the RX10m3, I will mess with one at B&H when/if they finally arrive. I held Bill's new RX10m3 when it arrived while I was visiting, it definitely is a big sucker, small for what the glass does to be sure.

In any case, I will wait for the price to drop like I did for the RX1r.

edform · Dec 29, 2016

elliottnewcomb said:
I have my 'no bigger than jacket pocket size' rule, but, a 1" zoom is so desireable, and, the lack of an evf on the Nikon is a deal breaker, the optional evf might be the answer, as it is on my RX1r, but, it isn't even available yet.

Nikon is a bit smaller than the RX10m3, I will mess with one at B&H when/if they finally arrive. I held Bill's new RX10m3 when it arrived while I was visiting, it definitely is a big sucker, small for what the glass does to be sure.

In any case, I will wait for the price to drop like I did for the RX1r.

I know about your jacket pockets and their ability to get clandestine cameras into entertainment venues, but I like loooooooooooooooong lenses and the DL24-500 has a viewfinder. It's the pansy short focal length models that don't.

Horacecoker · Dec 29, 2016

elliottnewcomb said:
Thanks to both Jeff and David for all the detailed work on this. I am wondering if you could give us some further guidance, (either from rules you can copy and paste, or just best guidance from experience, iow, not asking for you to do further tests for exact answers):

Don't get too hung up on these results, Elliot. Don't forget I had to pixel peep at 100% to see the diffraction induced quality reduction. It can't be seen at normal viewing sizes until you get beyond F8 and even then you wouldn't notice the detail loss if you weren't looking for it. You just need to be aware of it which anybody reading this thread will be in future.

elliottnewcomb said:
if using f4.0 as smallest aperture as a general rule:

1. detail was lost at a distance in these examples, how much is lost due to diffraction at closer shots, typical indoors, street shots, i.e. subject within 25 feet?

Diffraction takes no prisoners, the loss in detail will be exactly the same from macro distances to infinity. It is the way light rays react to smaller apertures that causes the problem and nothing to do with how far the light travels to reach the lens.

elliottnewcomb said:
2. when close, sufficient depth of focus using f4.0?

a. landscape, anything a distance more than ______? feet away, depth of focus will be enough

b. shooting indoors, f4.0, i.e. near a single kid, stay _______? feet away from their face to get their whole body in focus. for a group of kids playing a game on the floor, stay ____? feet away to get them all in focus?

I could provide you with answers for these two questions but why not find them out for yourself and much more by using this excellent tool.

DOF simulator - Camera depth of field calculator with visual background blur and bokeh simulation.

Calculates camera depth of field and background blur and simulates it on a photo for any lens, camera and distance combination with different types of lens blur (bokeh).

dofsimulator.net

I will just say you will be staggered by the amount of depth of focus you get at F4 with a 1" sensor for general photography and landscapes, think F11 on full frame!

Give me a shout if you struggle with the tool.

elliottnewcomb said:
3. stay f4.0 and use ND filter to reduce light when desired/needed?

This is an interesting question as only this morning I did a little experiment to find this out. In my earlier days with my original RX100, I shot mainly on P (program) mode and noticed the camera would almost never choose an aperture smaller than F4 but if pressed it would go to F5.6 Instead of a smaller aperture it would just increase the shutter speed if light levels demanded. Same with my RX100M3 and my RX10iii. Of course we now know why!! Musing on this I seemed to remember that the RX100M3 would engage the ND filter rather than stop down to a smaller aperture. I could tell it was doing this because of a slight mechanical sound the camera makes when the ND filter is engaged. I didn't like the idea of the ND filter engaging without my permission so I switched off auto ND and then forgot all about it till now.

That's the background and now to the experiment. I wanted to see if I could force my RX100M3 in P shooting mode to use the minimum aperture the lens is capable of. Of course I needed a lot of light for this and found it by using spot metering on a bare 60 watt light bulb. Because of this vast amount of light the camera realised its fastest shutter speed wasn't fast enough and in this instance it had no option but to stop the lens down to F11 to restrict the light for the correct exposure. Now the interesting bit - I then switched the Auto ND filter on. Aimed at the light bulb again heard the click as the filter engaged and the f stop was then F5.6 and not F11!

So Sony themselves know, due to diffraction, it's better to use the ND filter than go above F5.6 if light levels demand it.

I think that little story says it all and answers that particular question admirably.

elliottnewcomb said:
this is where I want to add the 'good enough' factor. You are a perfectionist, I am not nearly as sensitive, or concerned about lost detail unless it is my primary subject, so, I could carry and use an ND filter to help tame too much light, like at the beach, or, I could simply use say f8. ???

For someone who is not a perfectionist Elliot, you have asked an awful lot of questions in this post that seem to imply otherwise

Just reread my first paragraph for my feelings on this question.

elliottnewcomb said:
......................................

the rx100's, same sensor, different lens:

safe to assume diffraction is already occurring at f5.6?, therefore better to try and limit no smaller than f4.0 like the RX10m3 examples shown here?

Exactly

elliottnewcomb said:
thanks in advance for any guidance.

My pleasure

elliottnewcomb said:
--
Elliott

David

elliottnewcomb · Dec 29, 2016

rx10M3 DOES NOT INCLUDE nd FILTER, CORRECT? rx 10M2 DOES, CORRECT?

hAH, TYPOS GALORE!

Jeff Greenberg · Dec 29, 2016

Horacecoker said:
At F4 most people should be able to see that the the coloured patterns above each of these three windows is not paint but a mosaic of small tiles.

At F5.6 people may now struggle to discern the tiles and at F8 it is virtually impossible.

At F8 you would think it was just paint!

Wish you hadn't started another thread though.

=====

David-H & all,

Thanks for sensational report!
Am posting link to Alamy forum where
I posted same thread.

http://discussion.alamy.com/index.p...f-stop-mm-combos-for-100-inspection/?p=114727
Not asking you to post examples, did you
notice IQ change going wider than f4...?
E.g., going from f4 to (f3.5, f3.2, f2.8, f2.4)
similar to IQ loss for f5.6 or even f6.3?
Am shooting "P" with pop-up flash as filler
(homemade diffuser from plastic jug
prevents lens shadow) to get grab shots
going often back & forth inside-outside
& must concentrate only on compositions,
& that setting tends to select f2.8 for both
(ISO 3200-200, min. speed 1/30, handheld)
As stock shooter, must compromise, as
ruining shots via forgetting to change settings
much worse than getting shot with diffraction
f2.8 might produce...
As important as settings are, it is still possible
to walk in & out of places all day long whilst
traveling & forget one or more times to
switch settings, ergo, need for "P"...
Thanks in advance for any comments!

PS. thought it prudent to make lens test JPGs
its own thread title to get most responses.
Sorry!
PSS. just read your comments wrt "P"
& wider fstop preferences. Any way
to override on RX10-III so f4 is preferred
over f2.4-3.5 whilst in P with popup flash...?

PC Wheeler · Dec 29, 2016

elliottnewcomb said:
Ed,

stay here for helping us, and stay in the mini-challenges, you will be back!

the rumored Nikon DL24-500 specs seem tempting, any comments?

The specs are here http://www.nikonusa.com/en/nikon-pr....html#tab-ProductDetail-ProductTabs-TechSpecs

The 24-500 mm f/2.8-5.6 lens seems awfully slow on the long end. Makes for a smaller camera, but I would expect the RX10 III (and even the Panasonic FZ2500) to have better IQ due to the faster lenses.

Horacecoker · Dec 29, 2016

elliottnewcomb said:
rx10M3 DOES NOT INCLUDE nd FILTER, CORRECT? rx 10M2 DOES, CORRECT?

hAH, TYPOS GALORE!

tbcass · Dec 29, 2016

Quite frankly I find the lens sharp from wide open to about f8 where diffraction begins to set in.

rapick · Dec 29, 2016

elliottnewcomb said:
Thanks to both Jeff and David for all the detailed work on this. I am wondering if you could give us some further guidance, (either from rules you can copy and paste, or just best guidance from experience, iow, not asking for you to do further tests for exact answers):

if using f4.0 as smallest aperture as a general rule:

Here is my humble contribution...

elliottnewcomb said:
1. detail was lost at a distance in these examples, how much is lost due to diffraction at closer shots, typical indoors, street shots, i.e. subject within 25 feet?

I'd say: in my understanding, diffraction occurs inside the lens, irrespective of focus distance. Therefore the image details degradation effect is always the same.

elliottnewcomb said:
2. when close, sufficient depth of focus using f4.0?

a. landscape, anything a distance more than ______? feet away, depth of focus will be enough

Assuming focal length 8.8 mm (equivalent to 24 mm FF), hyperfocal distance for 1" size sensor is around 6 ft, allowing everything between 3 ft and infinity to appear "in focus". This means: no need to further stop down when shooting landscapes (assumption of using full WA for shooting landscape is reasonable.

elliottnewcomb said:
b. shooting indoors, f4.0, i.e. near a single kid, stay _______? feet away from their face to get their whole body in focus. for a group of kids playing a game on the floor, stay ____? feet away to get them all in focus?

Here, it depends on the focal length you use. With f.l. = 25.7 mm (70 mm eq. - that is reasonable for indoor "portraiture") evaluated DOF is:

- 1.0 ft for subject distance = 5 ft

- 2.6 ft for subject distance = 8 ft

If these can be assumed as enough, it depends on specific circumstances.

elliottnewcomb said:
3. stay f4.0 and use ND filter to reduce light when desired/needed?

elliottnewcomb said:
this is where I want to add the 'good enough' factor. You are a perfectionist, I am not nearly as sensitive, or concerned about lost detail unless it is my primary subject, so, I could carry and use an ND filter to help tame too much light, like at the beach, or, I could simply use say f8. ???

Does the ND filter introduce any image quality degradation? How much "bright" shall be the scene for the maximum available shutter speed not fast enough? I don't know.

elliottnewcomb said:
......................................

the rx100's, same sensor, different lens:

safe to assume diffraction is already occurring at f5.6?, therefore better to try and limit no smaller than f4.0 like the RX10m3 examples shown here?

Waiting for someone here willing to repeat with an RX100 the same good work made by Jeff with his RX10... since diffraction (AFAIK) depends on the physical size of the iris "hole" (this is why f/16 on a 1" camera lens is NOT the same as f/16 on a FF lens), my reply is YES.

elliottnewcomb said:
thanks in advance for any guidance.

Lightpath48 · Dec 29, 2016

I had the RX10Mk1, with the same size sensor. To my eyes, diffraction started degrading the edge acutance beyond f/4. So f/4 was my landscape aperture. I often shot at 2.8-4 otherwise, and the images were very good.

elliottnewcomb · Dec 29, 2016

Horacecoker said:
elliottnewcomb said:

rx10M3 DOES NOT INCLUDE nd FILTER, CORRECT? rx 10M2 DOES, CORRECT?

hAH, TYPOS GALORE!

--
Elliott

Click to expand...

Thanks for the thanks.

I apologize, my bad, I am always thankful for your shared experience, and others. I try to say thanks to everyone for what I have learned here these past 4 years, and, let new members know how helpful and experienced the members are here. 1" sensor has brought so many DSLR shooters, with great knowledge here.

I often ask and write for the others, often silent, who may be following, knowing the OP or you know most or more of what I write. So I think of your answers as more 'to everyone' than just me personally. Of course, an apology with explanation is not an apology, just letting you know my approach of trying to give back or pull knowledge from others for everyone, keeping this forum the best! You are a great addition. The Oly fjorum is dead comparitively, and I try not to be too disappointed or critical about it. One reason is they have no 1" sensor cameras, so fewer DSLR shooters are there.

Horacecoker said:
I went to all that trouble to answer your questions and that's your reply?

There are no typos read it again.

You do take issue unnecessisarily, it's my typos I am talking about, and I am warning OTHERS, MY 2007 POSTS wILL BE FUL OF THEMM.

Horacecoker said:
I deliberately differentiated between the RX100M3 and the RX10iii my typing them as M3 and iii. I know full well the RX10iii has no ND filter that's why I used my RX100M3 on the ND filter experiment.

aND, i WANTED TO CONFIRM THAT FOR ANYONE ELSE THAT MIGHT NOT KNOW, because, knowing the RX10m2 had ND built in, many could assume RX10m3 had one. One member bought an RX1rII, and was quite upset that it does not have a built in flash. I didn't realize that, and knowing it, it is one reason I prefer the m1 to the m2.

you know plenty, you donn't need to convince me of that. I am tempted to add Silly! but don't want ot offend, so I wonn't.

as for differentation, I started using lower case for rx100's and upper case for RX10's and RX1's.

Horacecoker said:
David

ps, I asked Jeff to rename his images to make it easier for you and others who I know would chase it down, I had no intention other than to view them, to see what I could see at posted image sizes.

I think he knew that from the starrt.

Elliott

Horacecoker · Dec 29, 2016

rapick said:
Waiting for someone here willing to repeat with an RX100 the same good work made by Jeff with his RX10... since diffraction (AFAIK) depends on the physical size of the iris "hole" (this is why f/16 on a 1" camera lens is NOT the same as f/16 on a FF lens), my reply is YES.

I'm sorry but Jeff just took the pictures at my suggestion and it was Horacecoker who did all the 'good work'. I don't really expect any praise but this is ridiculous. What with Elliot's unpleasant reply and now this. I wish I'd never bothered!!

David

rapick · Dec 29, 2016

What you discovered about the P program "strategy" in RX10 / RX100 cameras is interesting and enlightening. I already noticed on my RX100 III that it is almost impossible to force aperture values beyond f/5.6 in P mode.

I substantially agree with your thoughts, as you can see in my own reply to Elliott.

Thanks.

Horacecoker · Dec 29, 2016

Jeff Greenberg said:
Horacecoker said:

At F4 most people should be able to see that the the coloured patterns above each of these three windows is not paint but a mosaic of small tiles.

At F5.6 people may now struggle to discern the tiles and at F8 it is virtually impossible.

At F8 you would think it was just paint!

Wish you hadn't started another thread though.

Click to expand...

=====

David-H & all,

Thanks for sensational report!
Am posting link to Alamy forum where
I posted same thread.

http://discussion.alamy.com/index.p...f-stop-mm-combos-for-100-inspection/?p=114727
Not asking you to post examples, did you
notice IQ change going wider than f4...?
E.g., going from f4 to (f3.5, f3.2, f2.8, f2.4)
similar to IQ loss for f5.6 or even f6.3?
Am shooting "P" with pop-up flash as filler
(homemade diffuser from plastic jug
prevents lens shadow) to get grab shots
going often back & forth inside-outside
& must concentrate only on compositions,
& that setting tends to select f2.8 for both
(ISO 3200-200, min. speed 1/30, handheld)
As stock shooter, must compromise, as
ruining shots via forgetting to change settings
much worse than getting shot with diffraction
f2.8 might produce...
As important as settings are, it is still possible
to walk in & out of places all day long whilst
traveling & forget one or more times to
switch settings, ergo, need for "P"...
Thanks in advance for any comments!

PS. thought it prudent to make lens test JPGs
its own thread title to get most responses.
Sorry!
PSS. just read your comments wrt "P"
& wider fstop preferences. Any way
to override on RX10-III so f4 is preferred
over f2.4-3.5 whilst in P with popup flash...?

I'm sorry Jeff, but I'm opting out of answering anymore questions on this thread. Not your fault but I think you'll understand.

David

RX10 III: best IQ f-stop(s) consensus...?

Veteran Member

Veteran Member

Forum Pro

Veteran Member

Forum Pro

Veteran Member

Forum Pro

Veteran Member

Veteran Member

Forum Pro

Senior Member

Forum Pro

Veteran Member

Forum Pro

Veteran Member

Lightpath48

Guest

Forum Pro

Veteran Member

Veteran Member

Veteran Member

Similar threads

Keyboard shortcuts