If you give me the files I can test them side by side after running them through the app. DXO Pure RAW also cleans DNG files from any in-camera correction.
Which lens you want see, 21, 28 or 35? PM me and where I can load the file for you
Fred Miranda wrote:
Let's see the 35 FLE. Try at mid-distance like 3m or so using the same manual exposure for both cameras. Thanks for doing this, check your PM.
Forget I can put lens profile off for both on camera. Maybe I will post a quick results later and send you files in the meantime, let you do the deep stuffs
Jesko is Director of Product Division Photo, Leica Camera. I doubt that he tests himself . No details were mentioned, but what he said makes sense and is expected, unless the in-camera "resizing" is not messed up.
zhangyue wrote:
Did he mention how he tested? I can’t bare 2 hour for 10mins info video
LBJ2 wrote:
I tried the same with all the LED lights in my house too. What shutter speed did you use to see the scan lines? I'm thinking shutter speed should probably be lower than the LED light frequency?
I do not remember, but see the link to Jim Kasson's article for test details.
SrMi wrote:
Jesko is Director of Product Division Photo, Leica Camera. I doubt that the tests himself . No details were mentioned, but what he said makes sense and is expected, unless the in-camera "resizing" is not messed up.
I meant how Leica Store Miami determined no DR improvement, did they mean after resize? I understand what Jesko said, most likely at pixel level, shooting at 18M RAW will be cleaner. This has been discussed tons of time here on the forum or even this thread. I am just surprised that we still need visit it time after time
IIRC lens coding dependent corrections at time of exposure are for color shift correction and vignetting moderation and are baked in. Lens profiles in LR only apply distortion correction. At least this has been the case for the M9 and M240…
zhangyue wrote:
Did he mention how he tested? I can’t bare 2 hour for 10mins info video
It is a long video. I don’t think he mentioned a specific test approach just that both he and Josh spent a lot of time testing and comparing and both were confident he didn’t see any differences in the DR/ISO among the three resolutions using the most current firmware.
LBJ2 wrote:
Yes, I saw both videos. Maybe in a controlled lab environment? If I understood correctly, David was just not able to confirm real-world.
I did not watch David's video, I cannot comment on them. However, Jesko's statement is the expected behavior if you do not mess up when downsizing: the per-pixel noise is reduced when downsizing. If that were not the case, the lower resolutions would considerably disadvantage the L-DNG.
LBJ2 wrote:
It is a long video. I don’t think he mentioned a specific test approach just that both he and Josh spent a lot of time testing and comparing and both were confident he didn’t see any differences in the DR/ISO among the three resolutions using the most current firmware.
That is also correct if they compared the DR the right way: at the same sized output level.
zhangyue wrote:
Forget I can put lens profile off for both on camera. Maybe I will post a quick results later and send you files in the meantime, let you do the deep stuffs
If we think this scientifically, as long as camera profile changed with software used, it will affect exposure too.(default tone curve means adjust exposure histogram already) Well, you don't want to be an engineer when thinking this kind of stuff. That is why I prefer just shooting without dig too much. It makes you sleepless.
Vignette is all about delta between center and corner, in that sense I think it is OK. I can try to match them well enough using neutral profile maybe.
Now, A few images I took last Friday with 50mm APO used rangefinder only. I like its color overall. This time, I use standard profile instead embedded one.
zhangyue wrote:
I meant how Leica Store Miami determined no DR improvement, did they mean after resize? I understand what Jesko said, most likely at pixel level, shooting at 18M RAW will be cleaner. This has been discussed tons of time here on the forum or even this thread. I am just surprised that we still need visit it time after time
They compared 60MP (LDNG) vs 36MP (MDNG) at pixel level in LR and saw no visible improvement. (They also shared the side by side video showing this)
burningheart wrote:
It is unlikely. Although I only have Leica Monochrom's I have seen from the M9M, M246 to the M10M that the sensitivity to IR on each successive model decreased. IMO as Leica continues to improve their sensors and sensor filters I believe they are making changes to eliminate more IR bleed than previous M models. I am sure IR does get through but I think that it would require longer exposures and higher ISO settings.
I only know from experience that it works quite well between 5-10 sec exposure time and R72 filter on the lens with the M 240. But very true, intent is always so far to stop IR bleeding into color images and therefore use stronger ICF filters on the sensor each generation.
Fred Miranda wrote:
They compared 60MP (LDNG) vs 36MP (MDNG) at pixel level in LR and saw no visible improvement. (They also shared the side by side video showing this)
Do you happen to know the time?
I looked at ISO 1600 images, and at the pixel level: L-DNG looks noisier than M-DNG to me. I think Sean Reid had similar observations. If that were not the case, L-DNG would have much better image quality.
The M11 Triple Resolution discussion starts about 45:40 in the video:
They both disagree with marketing on this topic as well as what some other reviewers have stated on the same.
At about 49:47 Josh says "to state this clearly...as of current firmware and Jan 29, 2022...there is no quality advantage, noise, detail, DR using mDNG or sDNG however there are performance advantages..."
Fred Miranda wrote:
They compared 60MP (LDNG) vs 36MP (MDNG) at pixel level in LR and saw no visible improvement. (They also shared the side by side video showing this)
Did they test 18MP sDNG? Leica only claims an improvement with sDNG.
I've read all the posts references on this. I must be doing something wrong. Do you happen to remember the settings ( shutter speed) and the math you used to determine 1/10s M11 readout speed that you wrote on LUF ? I'm still trying to replicate the scanning lines with LEDs and computer/TV monitors and still not able to replicate these scanning lines while in electronic shutter mode.
