ISO 100-800...any big dif in IQ?

charlie76 · Aug 7, 2012

greybeard said:
My D7000/D5100 operates best at an ISO of 100-400, 500-1600 is descent, 2000-3200 is acceptable, 4000 and beyond will have to do in a pinch. It really depends on the amount of crop, the amount of enlargement, and your ability to post process. Light Room's noise reduction can do wonders. Older digital cameras were a lot more picky about ISO. My old Sony F828 is good 100-200 and that is it, period. Anything higher is pretty rough.

Hey thanks for the feedback! I have found the same with my camera...low noise all the way up to 600-800! But I still use 100 whenever possible.

usayit · Aug 7, 2012

Nah.. I think it base iso is a factor of the sensor design.... (and too much credit given to Nikon).

There are tons of cameras whose base iso is 200 and not designed for action/sports etc...

markj · Nov 3, 2012

I was told by a self proclaimed photo journalist I met last week while shooting landscapes, that color rendition at 200 iso was the optimum setting for color. Can anyone confirm or depute that thesis?

MLeeK · Nov 3, 2012

markj said:
I was told by a self proclaimed photo journalist I met last week while shooting landscapes, that color rendition at 200 iso was the optimum setting for color. Can anyone confirm or depute that thesis?

I believe Nikon's native, base ISO is 200. Might that have something to do with it?

KmH · Nov 3, 2012

Not all of Nikon's DSLR's use ISO 200 as their native base ISO.

markj, the PJ you talked to was blowing smoke up your

unpopular · Nov 3, 2012

markj said:
I was told by a self proclaimed photo journalist I met last week while shooting landscapes, that color rendition at 200 iso was the optimum setting for color. Can anyone confirm or depute that thesis?

If the base ISO is 200, then the maximum dynamic range will be at 200. Because RGB color and tone are intrinsic, greater dynamic range means better color depth.

markj · Nov 3, 2012

My base with a D7000 is 100 iso. So your saying then that the MDR will be 100 iso? KmH, how did you possibly know we were smoking???

unpopular · Nov 3, 2012

^ you do know that the base/native ISO is not the same as the minimum ISO, right? I don't know what the base ISO is, it could be ISO 100, but I am betting on 200.

ETA: on the D7000, there is no dynamic range difference from ISO 100 to ISO 800, according to dpreview. So in this regard, color should be similar. Noise OTOH may play a roll.

rexbobcat · Nov 3, 2012

Nikons tend to have a base ISO around 200 while Canon's is around 100, however there is some variation. One example is the 60D, which apparently has a base ISO around 160.

markj · Nov 3, 2012

Sorry, I thought the base number was the lowest available iso number. Could someone through a simple definition this way. Thanks.

rexbobcat · Nov 3, 2012

With film the speed was already set for the specific film you were using.

With digital, there are certain ISOs that are native to the actual sensor and its sensitivity.

Like, if the base ISO of the actual sensor for your camera is 100, then all of the multiples of 100 are also native to the sensor.

IE: 100, 200, 400, 800, 1600

But, say, you use ISO 160. Then the sensor is actually using software to simulate that ISO by using the native ISO of 200 and pulling down the exposure to match what 160 ISO would look like.

And with an ISO of 125 it would push the exposure of ISO 100, meaning more noise.

I think all of that is correct. Does it make sense?

usayit · Nov 3, 2012

Base iso is not necessarily the lowest iso setting. Some cameras have iso settings that are less than the base iso. Shooting at those lower iso settings also reduces dynamic range.

Digital cameras are essentially iso-less. I would not be caught up in he technical aspects so much as its only applicable in the sense that it approximates he sensitivity of film iso. It was intended to make the transition from film to digital easier. Digital sensors only operate optimally at a single setting... its base iso.

unpopular · Nov 3, 2012

rexbobcat said:
With film the speed was already set for the specific film you were using.

With digital, there are certain ISOs that are native to the actual sensor and its sensitivity.

Like, if the base ISO of the actual sensor for your camera is 100, then all of the multiples of 100 are also native to the sensor.

IE: 100, 200, 400, 800, 1600

But, say, you use ISO 160. Then the sensor is actually using software to simulate that ISO by using the native ISO of 200 and pulling down the exposure to match what 160 ISO would look like.

And with an ISO of 125 it would push the exposure of ISO 100, meaning more noise.

I think all of that is correct. Does it make sense?

Kind of. First of all, film manufacturers lie ... but that another topic.

With digital, the signal gain is typically done on the analog side of the camera, not the digital (software) side. This is a pretty important distinction because digital data is discrete while analog data is continuous. So it's not exactly like bumping up the exposure knob.

OTOH, I have heard a rumor that the D7000 handles ISO on the software end.

But yes. The sensor in your camera has a fixed sensitivity whose signal is amplified or attenuated to meet a given equivalent sensitivity of higher or lower ISO value. What ever this sensitivity is is the "base ISO".

Dikkie · Nov 13, 2012

With film, I could see difference in ISO by the graininess.
I had used slow films from 25 iso to fast speed films at 3200 iso.
The lower the iso, the less grain (now noise), on my prints. This theory was standard and reliable.

But since I have a digital camera (5 years old DSLR), I see lots of grain at 800 iso, but with my brothers newer digital camera (DSLR), there is no grain at 800 iso.
So that digital noise-by-iso theory isn't reliable anymore.

unpopular · Nov 13, 2012

^^ I think at this point, starting at around 2005 the signal to noise ratio of most DSLRs out performs film. There is a lot of hoohah about noise/grain, I think it's a bunch of malarky... Ok, maybe not entirely malarky. While the two qualities are functionally and aesthetically different, they represent the same limitation to SNR, and can be statistically measured in the same way.

The way you make a film more sensitive is by adding dopants that physically change the surface of the silver halide such that it can more readily develop out; sensitivity is a function of post-exposure, and silver halide has the same quantum yield (right word?) regardless how the crystal is shaped. Provided that you applied this process linearly and predictable under similar development conditions, any similar emulsion formula will likewise have predictable grain. However, variation between films does exist, and so it's not really possible to say "If I use Kodak Royal Gold 400 it will be just as grainy as Fuji Supra 400" And slide film is entirely different yet, with grain that is merely a product of where silver used to be.

With digital, it's really no different except that the "film" is in the camera and that technolgy is still being improved upon, so variation is going to be more apparent. Eventually, as with film, manufacturers will be pressed up against what is economically and physically possible or practical and camera manufacturers will have to go back to developing and marketing cameras.