blog.kasson.com
blog.kasson.com
blog.kasson.com
It looks to me like the conversion gain changes on the transition from ISO 100 to ISO 200. That would explain it.
Shutter speed was 1/800 second
Some takeaways:
I will test at 16 bit raw precision.
- EDR is never as large as 15 bits.
- Both ISO 100 and ISO 64 are lower than you'd expect from the other ISOs.
- ISO 100 is even lower than ISO 64
blog.kasson.com
Jack does good work, and is an all-around nice guy to boot.
Clear and concise writing which is definitely worth reading for me.
There seems to be no benefit in shooting 16 instead of 14 bits. I briefly tested lifting shadows and did not notice any difference, but I hope you can do a proper test.
What are your thoughts on applying the conversion gain change earlier?It looks to me like the conversion gain changes on the transition from ISO 100 to ISO 200. That would explain it.
Shutter speed was 1/800 second
Some takeaways:
I will test at 16 bit raw precision.
- EDR is never as large as 15 bits.
- Both ISO 100 and ISO 64 are lower than you'd expect from the other ISOs.
- ISO 100 is even lower than ISO 64
It's a tradeoff. The higher the ISO when the change occurs, the more of a conversion gain boost you get. But it costs you DR in the higher ISOs before the change. For a medium format camera, where you need more light than with an FF camera to get the photon noise advantages, I think it makes sense to do it a lower ISO than you'd use for a FF camera.What are your thoughts on applying the conversion gain change earlier?It looks to me like the conversion gain changes on the transition from ISO 100 to ISO 200. That would explain it.
Shutter speed was 1/800 second
Some takeaways:
I will test at 16 bit raw precision.
- EDR is never as large as 15 bits.
- Both ISO 100 and ISO 64 are lower than you'd expect from the other ISOs.
- ISO 100 is even lower than ISO 64
blog.kasson.com
blog.kasson.com
blog.kasson.com
You know I love what you do and read everything you post here and your blog. But I got a he honest, the only thing I know of that sounds cooler than Fourier transforms is Flux capacitors.
Not quite textbook performance, but not far off. I'll look at some Fourier transforms to figure out if there's any spatial filtering, but there aren't the jumps that you see with some cameras. I colored the lines to correspond to the raw channels.
--
https://blog.kasson.com
My grandfather used to do furrier transforms. He was a taxidermist.You know I love what you do and read everything you post here and your blog. But I got a he honest, the only thing I know of that sounds cooler than Fourier transforms is Flux capacitors.
Not quite textbook performance, but not far off. I'll look at some Fourier transforms to figure out if there's any spatial filtering, but there aren't the jumps that you see with some cameras. I colored the lines to correspond to the raw channels.
With FFTs (Fast Fourier Transforms) Jim can check if Hasselblad is doing tricks, like spatial filtering, on raw files.My grandfather used to do furrier transforms. He was a taxidermist.You know I love what you do and read everything you post here and your blog. But I got a he honest, the only thing I know of that sounds cooler than Fourier transforms is Flux capacitors.
Not quite textbook performance, but not far off. I'll look at some Fourier transforms to figure out if there's any spatial filtering, but there aren't the jumps that you see with some cameras. I colored the lines to correspond to the raw channels.
[Rim shot]
blog.kasson.com
Jim has published similar measurements for GFX100 and GFX100S.With FFTs (Fast Fourier Transforms) Jim can check if Hasselblad is doing tricks, like spatial filtering, on raw files.My grandfather used to do furrier transforms. He was a taxidermist.You know I love what you do and read everything you post here and your blog. But I got a he honest, the only thing I know of that sounds cooler than Fourier transforms is Flux capacitors.
Not quite textbook performance, but not far off. I'll look at some Fourier transforms to figure out if there's any spatial filtering, but there aren't the jumps that you see with some cameras. I colored the lines to correspond to the raw channels.
[Rim shot]
It seems that DR drops significantly with long exposures. That should have been self evident to me, but I have not seen it demonstrated that clearly.
I like how Jim's work here and on his blog separates the wheat from the chaff.
