DxOMark's measured ISOs vs. manufacturer ISOs

Started Mar 13, 2013 | Discussions thread
Jack Hogan Veteran Member • Posts: 6,983
Ode to Hsat

xpatUSA wrote:

Haven't found any mention of lux-seconds in this thread, maybe it's there somewhere.

Here, you have to look for it

My opinion is that the Organization, in pandering to the obfuscation needs of the manufacturers, has done us customers a great disservice

Hard to disagree in this case

Ssat = 78/Hsat . . . .

That's one I actually find very useful backwards, independently of the constant, which is really immaterial for our purposes, because it and DxO tells us at what exposure the sensor in question saturates the Raw data:

Ssat = 107 --> Hsat = 0.73 lx-s
Ssat = 29 ----> Hsat = 2.69 lx-s

DxO actually measures Hsat and calculates Ssat through the formula, so these are not synthetic values. Also, the constant 78 is not as arbitrary as it looks:

1) Average Exposure H = Et = pi/4 * Lt/f^2 lx-s

But the typical reflected light meter in a DSC will produce f and t settings according to this equation

2) Lt/f^2 = K/S, so substituting in 1) above we get
3) H = pi/4 * K/S with K the reflected light meter calibration constant or

4) Average Exposure H= 0.78 * K/S lx-s

If K for CaNikon is 12.5 and we assume that the average tone is recorded at 12.5% of Full Scale...

Jack

Post (hide subjects) Posted by
tko
tko
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow