Leica S3 PDR on PhotonsToPhotos

[helged]

Advertisement (gone after registration)

23 minutes ago, John McMaster said:

The S(007) sensor is far superior to the M(240) one....

john

Agree about the superiority of the S007. But the dng-files files also depend on e.g. cover filter, sensor micro-lenses, processing, etc. 

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Jeff S]

1 hour ago, John McMaster said:

The S(007) sensor is far superior to the M(240) one....

john

Same sensor architecture, pixel pitch, etc.  This was apparently the first instance Leica created economies by sharing S/M sensor tech.  Of course there’s a lot more involved once micro-lenses, color arrays and filters, as well as any processing and/or software tech is taken into account (and any obvious benefits resulting from a larger sensor surface).

Jeff

Edited September 25, 2020 by Jeff S

[SrMi]

17 minutes ago, Jeff S said:

Same sensor architecture, pixel pitch, etc.  This was apparently the first instance Leica created economies by sharing S/M sensor tech.  Of course there’s a lot more involved once micro-lenses, color arrays and filters, as well as any processing and/or software tech is taken into account (and any obvious benefits resulting from a larger sensor surface).

Jeff

And the most important: the difference in sensor size. Frankly, I am surprised that there is not more difference between M10-R and S3 PDR results.

[Jeff S]

2 minutes ago, SrMi said:

And the most important: the difference in sensor size. Frankly, I am surprised that there is not more difference between M10-R and S3 PDR results.

As I already wrote... see last sentence... benefits from larger sensor surface.

Jeff

[Chaemono]

Am 24.9.2020 um 22:20 schrieb SrMi:

Yes, dual native ISO, dual base ISO, dual gain, Aptina's dual-conversion gain, all the same, AFAIK.

There must be a difference beyond just semantics. S1 looks like it has dual gain at ISO 800 https://www.photonstophotos.net/Charts/PDR.htm but doesn’t have the dual native ISO system of the S5.

[John McMaster]

15 hours ago, Jeff S said:

Same sensor architecture, pixel pitch, etc.  This was apparently the first instance Leica created economies by sharing S/M sensor tech.  Of course there’s a lot more involved once micro-lenses, color arrays and filters, as well as any processing and/or software tech is taken into account (and any obvious benefits resulting from a larger sensor surface).

Jeff

It is simply using (say) 6400 ISO, usually unusable on the M(240) due to banding but the S(007) files ate 12800 ISO are mostly usable.

john

[Jeff S]

Advertisement (gone after registration)

3 hours ago, John McMaster said:

It is simply using (say) 6400 ISO, usually unusable on the M(240) due to banding but the S(007) files ate 12800 ISO are mostly usable.

john

Whatever... still derived from same wafer.

Jeff

[SrMi]

7 hours ago, Chaemono said:

There must be a difference beyond just semantics. S1 looks like it has dual gain at ISO 800 https://www.photonstophotos.net/Charts/PDR.htm but doesn’t have the dual native ISO system of the S5.

DPR writes about Panasonic's dual native ISO (link😞

Essentially it's a way of exploiting and revealing the 'dual gain' sensor function that's become increasingly common in modern cameras. Dual gain sensors have two read-out modes at the pixel level: one that gives the widest dynamic range and a second, higher-gain step that reduces shadow noise at the expense of some dynamic range.

We will see how that affects still photographers once PhotonsToPhotos publishes results.

[BernardC]

Panasonic has been using dual ISO in their cine camera for a long time.

Here is an explanation from 2014:

https://www.newsshooter.com/2014/11/28/panasonic-varicam-native-iso-of-800-and-5000-how-do-they-do-it/

About the S3 and M10R sensor: the same sensor architecture was considered for the SL2. I remember that either David Farkas or Jonathan Slack mentioned this when it was released. The two of them them had a small wager on whether the SL2 would have 40 or 47 MP.

[Stuart Richardson]

I do notice a difference in baseline shadow noise between ISO 100 and 400 when the files are pushed. On my camera ISO 400 shows less banding in pushed shadows than ISO 100, which is odd, but it does make sense if the second ISO gain setting applies a different treatment to the files coming off the sensor. The banding is not gone, as it reappears at high ISO's, but it seems to be handled better at 400 than 100. It seems strange that the two base ISOs would be so close together, however...I would have expected something like 100 and 1600, for example.

Edited September 28, 2020 by Stuart Richardson

[SrMi]

2 minutes ago, Stuart Richardson said:

I do notice a difference in baseline shadow noise between ISO 100 and 400 when the files are pushed. On my camera ISO 400 shows less banding in pushed shadows than ISO 100, which is odd, but it does make sense if the second ISO gain setting applies a different treatment to the files coming off the sensor. The banding is not gone, as it reappears at high ISO's, but it seems to be handled better at 400 than 100.

Banding often disappears at higher ISO because of higher noise which smooths over banding.

[Stuart Richardson]

I don't think that is what is going on here, but I will see if I can post photos later. Right now my lightroom is stuck doing an exhibition for a client, so I cannot post anything!