At 12/20/2019 01:03 PM, Mike G wrote:
>AG (invariant) Schnozz writes:
>
>
><<The extended low ISO is almost always going to lose one stop of DR. I
><<would only use it if you need it for exposure control and an ND filter
><<is not available or practical.
>
><<Base ISO varies on the sensor and sensor design. Most of the sensors
><<now have around ISO 200 for base ISO. In reality, it's probably
><<anywhere from ISO 100 to 400, but the engineers and marketing
><<department have tuned it to 200. Ten years ago, the bulk of the
><<sensors were at ISO 100 for base ISO. Again, give or take a healthy
><<smidge.
>
>Thanks for thinking about this and commenting.
>
>Yes, that is the conventional thought on base ISO explained better than I did.
> Camera ISO is not "real ISO" has been known for awhile. I recall the hacks
>in CHDK and Magic Lantern I think would give "real ISO" as well as Raw Zebras
>if you want them.
>
>The new firmware is supposed to do something to handle detail better at
>extended low ISO's--only for jpegs as I recall. I dug a little deeper and
>the DXO graphs show DR the same with the extended low ISO (Hmmmm?) --they
>measure it a peculiar way though.
>
>See below: SNR is a tad better, though lose one stop of headroom but gain one
>stop or so in the shadows.
>
>https://www.dpreview.com/forums/post/62819260?image=0
>
>
>"Headroom is the difference between the camera's 18% grey level and maximum
>output (4095-255=3840 or 11.9 EV). For the normal ISO range the average
>headroom is 4.1 stops. For the low range it drops to around 2.4 stops. Total
>noise is higher in the low ISO range, but the signal level increases even more
>so the Signal/Noise Ratio continues to improve."
>
>So if expose for highlights carefully, perhaps the extended low ISO is a tad
>better? Capturing more photons can not be a bad thing in general. I can see
>the noise at base ISO with my Oly though it is essentially nonexistent with my
>Sonnie.
>
>Not sure what to tell Marnie yet or if it is worth fussing with, even if
>theoretically better.
>
>Pondering, Mike
I often check http://photonstophotos.net/ dynamic range curves. For instance
the E-M5 III peaks at 200, just as Ken mentioned, goes down a bit at 100.
http://photonstophotos.net/Charts/PDR.htm#Olympus%20OM-D%20E-M5%20Mark%20III
The sony a7's typically have peak at 100 and a secondary peak at 640 where they
switch out an extra capacitance on the individual pixels. The extra cap
basically allows the pixel to hold (integrate) more charge giving effective
lower ISO. So ISO 640 is same as ISO 400 in dynamic range. If they did not add
the extra cap, the base ISO would probably be closer to 200. Although, the GX85
best DR is at ISO 100.
12-Bit WayneS
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|