Understanding Aperture Diffraction
Understanding Aperture Diffraction
Hi guys. I have some questions about aperture diffraction as well as a possibly naive suggestion.
Whilst trying to wrestle a clean and accurate render of my rings scene I have started asking myself questions about aperture diffraction as it is implemented in indigo and how it compares to the real world.
So far whenever I use AD my renders get blurred. They always seem slightly out of focus. Now in the real world where AD is always in effect it is quite possible to have completely sharp images but also having the hot spot "flares" caused by AD so why does indigo seem incapable of achieving the same?
I was also thinking about obstacle maps, in particular the generated blades. At present you can customize the blade generation to suit yourself however I would have thought that the generated size of the opening would be directly tied to F Stop and so therefore would be automatic because the generated blade obstacle map is meant to be representative of the internals of a real camera right? Now this may be complete ignorance on my part and maybe this is already the case but was wondering if a mismatch between F Stop and the size of the opening in a generated obstacle map might have something to do with this blurriness problem...?
Whilst trying to wrestle a clean and accurate render of my rings scene I have started asking myself questions about aperture diffraction as it is implemented in indigo and how it compares to the real world.
So far whenever I use AD my renders get blurred. They always seem slightly out of focus. Now in the real world where AD is always in effect it is quite possible to have completely sharp images but also having the hot spot "flares" caused by AD so why does indigo seem incapable of achieving the same?
I was also thinking about obstacle maps, in particular the generated blades. At present you can customize the blade generation to suit yourself however I would have thought that the generated size of the opening would be directly tied to F Stop and so therefore would be automatic because the generated blade obstacle map is meant to be representative of the internals of a real camera right? Now this may be complete ignorance on my part and maybe this is already the case but was wondering if a mismatch between F Stop and the size of the opening in a generated obstacle map might have something to do with this blurriness problem...?
"A foolish consistency is the hobgoblin of little minds..." - Emerson 1841
Re: Understanding Aperture Diffraction
Hi WytRaven,
First off - I've been checkign out some of your renders and they're awesome - lovely attention to detail!
I was wondering a bit about A.D-vs-"focus" as well - especially after your comments with your Renderer's Rings.
Anyway, this is what I found:
http://www.bobatkins.com/photography/te ... ction.html
I found this article very informative - and, something it mentions made me think of something to do with Indigo's auto focus feature (which I'm sure most people make use of) - if I read it correctly (I just skimmed through it) they seem to imply that the (A.D-induced) "spot size" is related, at least partially, with focus (as, I would call it, "the focus spot") - I may be totally off on this 'observation', but, perhaps there is a slight connection and if this is true then, in cases of high A.D with Indigo (assuming it's model is 100% accurate/based on reality - which I suspect it is indeed at least *close* to in this regard,) it could be possible that Indigo's auto-focus feature is a lilttle off (from the otherwise expected Focal Length) if Indigo were not to take into account this hypothetical variable (which would mean that, "until it did", slight manual focus adjustment "might remedy this"). Anyway, all this aside, as can be seen form the article, A.D. does indeed cause quite some bluriness!
Azure'
First off - I've been checkign out some of your renders and they're awesome - lovely attention to detail!
I was wondering a bit about A.D-vs-"focus" as well - especially after your comments with your Renderer's Rings.
Anyway, this is what I found:
http://www.bobatkins.com/photography/te ... ction.html
I found this article very informative - and, something it mentions made me think of something to do with Indigo's auto focus feature (which I'm sure most people make use of) - if I read it correctly (I just skimmed through it) they seem to imply that the (A.D-induced) "spot size" is related, at least partially, with focus (as, I would call it, "the focus spot") - I may be totally off on this 'observation', but, perhaps there is a slight connection and if this is true then, in cases of high A.D with Indigo (assuming it's model is 100% accurate/based on reality - which I suspect it is indeed at least *close* to in this regard,) it could be possible that Indigo's auto-focus feature is a lilttle off (from the otherwise expected Focal Length) if Indigo were not to take into account this hypothetical variable (which would mean that, "until it did", slight manual focus adjustment "might remedy this"). Anyway, all this aside, as can be seen form the article, A.D. does indeed cause quite some bluriness!
Azure'
Re: Understanding Aperture Diffraction
Hi Wyt,
I've noticed the (possible) blurriness issue with A.D as well. I'll look into it as soon as circumstances allow.
I've noticed the (possible) blurriness issue with A.D as well. I'll look into it as soon as circumstances allow.
Re: Understanding Aperture Diffraction
Ok so it's not neccessarily just user error then?
Thanks for the update
Thanks for the update
"A foolish consistency is the hobgoblin of little minds..." - Emerson 1841
Re: Understanding Aperture Diffraction
No, I don't think it's User error.
Aperture diffraction definitely blurs things in real life somewhat, but, as you said, it should still be possible to achieve sparkles / glare-y highlights while maintaining a decent sharpness on the rest of the image.
By the way, when you construct an aperture shape, that aperture is subsequently scaled down to the appropriate size, depending on the camera aperture radius (e.g. f-stop).
Aperture diffraction definitely blurs things in real life somewhat, but, as you said, it should still be possible to achieve sparkles / glare-y highlights while maintaining a decent sharpness on the rest of the image.
By the way, when you construct an aperture shape, that aperture is subsequently scaled down to the appropriate size, depending on the camera aperture radius (e.g. f-stop).
Re: Understanding Aperture Diffraction
I did a bit of work on this issue today, and I'm pretty sure I've found the cause of the blurring issue, and somewhat sure that I have successfully fixed it.
I'll do some more testing on it tomorrow.
w00t!
I'll do some more testing on it tomorrow.
w00t!
Re: Understanding Aperture Diffraction
Awesome
"A foolish consistency is the hobgoblin of little minds..." - Emerson 1841
Re: Understanding Aperture Diffraction
Thanks for that Azure. Interesting indeedAzureSky wrote:Anyway, this is what I found:
http://www.bobatkins.com/photography/te ... ction.html
I found this article very informative
"A foolish consistency is the hobgoblin of little minds..." - Emerson 1841
Re: Understanding Aperture Diffraction
Ok, this bug is fixed now.
You can see the (rather subtle) difference in the images below: the 2.0.10 image is blurrier.
You can see the (rather subtle) difference in the images below: the 2.0.10 image is blurrier.
- Attachments
-
- Fixed A.D. in 2.0.11
- im1248231947.png (763.52 KiB) Viewed 4799 times
-
- blurry A.D. in 2.0.10.
- im1248232504.png (697.59 KiB) Viewed 4800 times
Re: Understanding Aperture Diffraction
Thanks Ono
Once 2.0.11 is released I'll use my rings to do a high res real-world test.
Although 2.0.10 is obviously too blurry when using A.D. I find that with it off altogether the images are unnaturally sharp so hopefully this fix finds the happy medium
Once 2.0.11 is released I'll use my rings to do a high res real-world test.
Although 2.0.10 is obviously too blurry when using A.D. I find that with it off altogether the images are unnaturally sharp so hopefully this fix finds the happy medium
"A foolish consistency is the hobgoblin of little minds..." - Emerson 1841
Who is online
Users browsing this forum: No registered users and 42 guests