Menu
What's new
The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Why are Native ISOs so much better in all dSLRs?

C

clark666

New member
http://www.zacuto.com/shootout

The GH-1 is fine at low ISOs, the Nikon D3s ----

What about the native ISO statements: 160, 320, 640, 1250, 5000, 10,000
25,650 51,200 and up? Why are these ISOs so superior?

I have heard this from others that the oscilloscope demonstrates that these ISOs are better than the others that are electronically produced from the native ISOs.
 
C

clark666

New member
Thank you for the link. I still don't understand why all dSLRs would have the same effect, but I guess the design specifications are the same so the imaging sensors all reflect their specs?

I wonder if this means that the 160 multiples are better for still photography so that if I need ISO 6400 according to the camera metering system, I should manually set 5000 instead? Something to try in the dark some day!
 
ggibson

ggibson

Well-known member
clark666 said:
I wonder if this means that the 160 multiples are better for still photography so that if I need ISO 6400 according to the camera metering system, I should manually set 5000 instead? Something to try in the dark some day!
Click to expand...
What did you see that makes you think that 160 multiples are better for still photography? The way I read is is that ISO 160 is actually just ISO 200 with some calculation applied. For all intents and purposes, there's no difference between shooting at ISO 160 vs shooting at ISO 200 and then post-processing. Am I interpreting that correctly?

The aperture scaling from that link is interesting. It suggests that when using a manual lens with a wider aperture than f2.8, the camera will not apply a scaling percentage to compensate. It is slight though (~3% per stop), so it may be hard to notice.

But all of this research was done with Canon cameras, so I'm not sure how much can be applied to m4/3 cameras. I think it's clear that there are native ISOs and pushed ISOs for all digital camera sensors. But even manufacturers using the same sensor have different methodologies for implementing ISO (like the E-P1 vs GF1).
 
ggibson

ggibson

Well-known member
I figured, but as much, but didn't know if the "still photography" part had any significance. Anyways, the author prefaced that with the following:

Because of this "stretching" and "compressing" ISOs like 160 appear very good in the following noise graph and 125 and the like appear bad. This is however deceptive - it is a purely mathematical manipulation of the data and therefore when shooting raw is useless - the same effect can be achieved by shooting at whole ISOs and pushing or pulling the image during raw conversion. When shooting JPEG on the other hand the intermediate settings can be useful.
Click to expand...
It is interesting how in some instances (like the one above) ISO 200 is above ISO 100. The author states that this is not because ISO 100 is noisier, but rather that ISO 200 has a higher dynamic range in the example (see the full explanation here 30D).
 
pellicle

pellicle

New member
It is interesting how in some instances (like the one above) ISO 200 is above ISO 100. The author states that this is not because ISO 100 is noisier, but rather that ISO 200 has a higher dynamic range in the example
Click to expand...
the stuff in the raw files represents what the Analog to Digatal Converter (ADC) made of it after signal processing, if it added a base biase to the signal and then clipped it (or infinity to one compressed it) then that will reflect in the signal to noise ration (which was plotted in that graph) and if the dynamic range is lower (for the same sensor) then you can bet that means more signal processing on the bottom end..

If you're interersted try reading my exporaiton here. I noticed how much later (higher numbers) the G1 was beginning recording the signal when bracketing exposure; that would result in a reduced dynamic range.

notice what level number data starts at here


and then here


note also the added clustering around the 3988 value? I was never able to get it to 4094 when dealing with this sensor.
 
Jonathon Delacour

Jonathon Delacour

Subscriber Member
Is the "optimum" 160, 320, 640, 1250, 2500, etc ISO sequence shown above applicable to all sensors? IANADSE (I am not a digital sensor engineer) but I would have thought that different sensors would have different native ISOs (eg 80 or 100 or 125 rather than 160). Or that the same sensor in different camera models (either from the same manufacturer or different manufacturers) would have a different native ISO, depending on the camera. And, therefore, that the optimum ISO sequence would vary from one camera/sensor to another.

Or am I simply revealing my ignorance? (It wouldn't be the first time.)
 
C

compositor20

Guest
with my e-p1 the iso 2500 is much better than iso 320 and iso 5000 much better than iso 6400... try it and you will agree

i have my auto-iso to iso 1250 since if i need one more EV i can dial iso 2500 (the image quality limit that you should use to get good a4 prints)

comparing iso1250 to iso 1000 you get more grain luminance at 1250

comparing iso 1250 to iso1600 you get just a little more grain but noticeable less tonal range in the shadows so much less dynamic range then 1600

i have made this test with real life samples and they show this
 
T

Tonygamble

Member
I have read this several times and still not fully understood it.

Next week I have an available light shoot for which I'll use my Canon 5d. Am I being recommended to choose 1250 rather than 1600? The original 5d stops at 1600 (it has a H setting but I gather that is equally artificial).

Tony
 
You must log in or register to reply here.
Top