ISO question

[Martyn]

As a beginner I'm doing a lot of reading and slowly getting to grips with the relationship between shutter speed, aperture and ISO setting. Am I correct in understanding that a lower ISO setting (ie ISO 50) will result in a better quality picture and than a higher setting (ie 800)? 

[NannyFerret]

You would probably get camera shake if hand held because of the slower speed unless you use flash or have a really bright day which would allow a faster shutter speed..

[bugmeister]

With a digital camera, the ISO is a measure of the sensitivity of the sensor and is increased or decreased by changing the electrical charge applied to it. The higher the charge, the more sensitive the sensor.

 

Sensors in different cameras behave a bit differently and have a native range which is described by the manufacturers as the usable range (yeah right), and normally the low-end is 100 or even 200 ISO, using ISO 50 is possble on some and for landscape work it's often worth trying, although it does mean you will have slower shutter speeds and if anything moves like a person, car or even a cloud, you may get blurring.

 

The extra quality comes from the lower charge applied to the sensor, as opposed to raising the charge for higher sensitivity which will also increase the noise evident in the picture. Again different sensors behave differently. I had one camera where ISO800 was unusable because of noise, but with my current one I have used ISO3200 at a wedding with good results

 

So yes, in theory ISO50 would give you the best quality, but you also need to consider the shutter speed (maybe a tripod needed) and aperture, since a low sensitivity sensor may make the camera select a wide aperture with the accompanying shallow depth of field.

Edited May 11, 2013 by bugmeister

[JohnP]

Higher ISO often produces more noise on images, generally it's better to use the lowest ISO you can. Obviously there are times when you have to use higher ISO, an example would be somewhere in low lighting conditions where flash was not allowed. If you are taking a shot and you need a good depth of field and are for instance using f16, you may need to up the ISO in order to use a shutter speed at which you can hand hold the camera without risking a blurred image. Using a tripod would mean you could use a slower shutter speed which would then enable you to use a lower ISO. When taking shots in good lighting conditions an ISO setting of 100-200 is usually OK and with most cameras noise shouldn't be a problem.

[fuzzyedges]

Bigger the ISO grainier (noisier) looking the image but that may be what is wanted by the photographer

 

Quality is maybe not the correct word possably smoother might be nearer

 

As the three (iso aperture and shutter speed ) are all linked together Iso is generally the last thought

 

In aperture mode Av you would set the f stop to what you wanted and check the shutter speed if it was less than 90 (This is the area where camera shake can occur ) you would up the ISO

 

In shutter mode Tv (general for action shots) you would set the shutter speed and check the Av (usually for depth of field) then up or lower the iso to the desiered F stop

[colinb]

. Am I correct in understanding that a lower ISO setting (ie ISO 50) will result in a better quality picture and than a higher setting (ie 800)? 

 

that depends on your interpretation of a quality picture, as said above higher iso causes 'noise' but dropping it to 50 can cause other problems..

 

Martyn I know you said you had a simple point and shoot but which model? I would like to see or know does it have an aperture or shutter setting - I think you would be better trying to understand these perhaps using auto iso adjusting the aperture to see how that relates to shutter and iso?

Edited May 11, 2013 by colinb

[Martyn]

Thanks all for your speedy replies, I think I've grasped the basics of this.

 

Colin, my camera is a Rollei da1725 Prego. It has ISO settings of Auto, 50,100,200 and 400, that's the only control I have over exposure.

[colinb]

well you did say it was 'basic' - best thing is to take the camera, and setting the camera on a solid surface like a wall so movement is kept to a minimum, shoot the same scene using all the different iso settings and then compare and see which to you looks best quality.

Edited May 11, 2013 by colinb

[Martyn]

I actually own a tripod believe it or not. I got it to mount a Zoom H2 digital sound recorder on to record my fiddle playing 

[Black Pearl]

Another issue as ISO rises is a reduction is dynamic range - in basic terms the brightness range the sensor can record.

 

Essentially what happens with a digital sensor as you increase the ISO is the camera produces a pixel map that is underexposed then amplifies the resulting data to get it back to a nominal exposure - the result is what gets recorded and what you see.

 

If you have set the camera to record a jpeg then generally the processor will apply some sort of noise reduction (smoothing of the variances) to minimise the appearance of the noise. This is all well and good but the flip-side is the processor also ends up smoothing away fine details resulting in a drop off in sharpness. Various brands do this to varying degrees of success with the varieties ending up being one of the hot topics of discussion and arguments on many forums. If you set your camera to record a RAW file however (assuming it has the option) then the file when opened in an appropriate RAW Converter will have no noise reduction applied, this must be done manually. The advantages are that the photographer gets to choose the level of fine detail smoothing balanced against the level of sharpness require for final output to a very precise degree. 

 

There's nothing you can really do about the dynamic range loss but by ensuring you correctly exposure when using high ISO's the noise should remain minimal or very often invisible under normal viewing conditions. Problems occur when images shot at higher ISO's are still too dark and have to be lightened further and when they are viewed at high magnifications on screen. The term Pixel Peeping is where people of a certain persuasion seem obsessed with viewing everything they shoot (more often a test chart) at 100% and fretting over the results. Its a disease that is all to common and one I'd love to eradicate. 

[grounded]

HEY, slow down you guys, Martyn is only a beginner and will hardly understand all your technical stuff. keep it simple for him.

[Guest DaveW]

Most has already been said, I would just like to add that it is not necessarily the lowest ISO on the camera that produces best quality images, but the native ISO of the sensor (the ISO it was originally optimised for). Usually this is the lowest ISO on the most cameras, but some newer cameras have simulated ISO's like 50 allowing you to set below the sensors native ISO which is say 100, just as they have simulated settings above their highest normal ISO like 102400 and 204800. These are really for exceptional purposes, not for day to day use since image quality always suffers to some degree using them.

http://photographyforrealestate.net/2011/03/01/new-revelation-about-noise-iso-160-may-have-less-noise-than-iso-100/

However it is sometimes disputed that working in multiples of the base ISO produce better images than other ISO settings above it, but not that the base ISO the sensor was designed for produces the best image.

http://dpanswers.com/content/tech_iso.php

You may find this camera simulator a way to study the different relationships between aperture, shutter and ISO on the camera Martyn, though I am sure others will like playing with it too!:-

http://camerasim.com/camera-simulator/

Edited May 12, 2013 by DaveW

[Black Pearl]

HEY, slow down you guys, Martyn is only a beginner and will hardly understand all your technical stuff. keep it simple for him.

 

Don't think anything above is 'too technical' in fact I believe that the best way for a beginner to learn is to read as much as possible about a subject and try to fully understand it.

In his original post Martyn actually states he is 'doing a lot of reading' so the explanations above can only help to further his understanding.

[Phil]

However it is sometimes disputed that working in multiples of the base ISO produce better images than other ISO settings above it, but not that the base ISO the sensor was designed for produces the best image

 

The 7D noise test is fascinating. 

[Martyn]

HEY, slow down you guys, Martyn is only a beginner and will hardly understand all your technical stuff. keep it simple for him.

Stuff that I don't understand I just pass over: I come back to it later and find I can understand more than the first time after I've picked up more info!

 

Dave, there's some great links there! I love the idea of buckets catching water to illustrate the relation ship between light and sensor size. That Camera Simulator is brilliant!

[Guest DaveW]

I did say it was disputed Phil that you can extrapolate that the best ISO settings are always in multiples of the base ISO:-

 

http://indigoverse.com/the-truth-about-native-iso-for-canon-dslrs/

 

Also different cameras have different base ISO's ranging between ISO 100 to ISO 200 depending on sensor design, sometimes even from the same camera manufacturer, so you need to do your own tests in the ISO 100-200 range. However most of us will see virtually no difference anyway and the old idea that the lowest ISO setting on the camera produces the best image usually holds good, provided it is not a simulated one like say ISO 50.

 

You even get similar arguments for Nikon:-

 

http://improvephotography.com/2493/does-iso-200-produce-less-noise-than-iso-100-on-nikon-cameras/

 

As pointed out in the following link changing the ISO does not change the sensitivity of the sensor, it is simply gain applied to the signal from the sensor by the image processing chips in the camera around the sensor as I understand it?  Here's the heavy explanation if you want it

 

http://www.clarkvision.com/articles/iso/

 

I think being dogmatic about theoretical differences that are generally only shown shown by contrived tests is pointless, it's only the general principles we need to know.  The old saying regarding theoretical lens resolution, image noise and other things shown in laboratory conditions applies, "if you can't see the difference in real life photography, don't worry about it!".

Edited May 12, 2013 by DaveW