ISOs available in Auto that aren't choices in manual?

[DirtySocks85]

I usually spend most of my time shooting in manual, but today I popped over to full auto (green A) mode to get a couple of quick shots. Looking at the EXIF I noticed that it shot at ISO 500. In manual mode I normally can only select in full stops (100, 200, 400, 800, 1600, 3200, 6400). Seems a bit interesting. Any reason Canon doesn't allow me to select these "in between" ISOs manually?

(using a Canon 600D)

 

[Phrasikleia]

You should be able to go into your custom functions and change your ISO increments to 1/3 stop. Look in your manual's index for ISO Speed Setting Increments.

 

[alexxk]

Not all cameras have this setting available at least with the manufacture's firmware. t3i is an example.. I cannot manually set iso to 640.

 

[nburwell]

You should be able to go into your custom functions and change your ISO increments to 1/3 stop. Look in your manual's index for ISO Speed Setting Increments.

This is your best bet. When I got my 5DIII a couple weeks ago, I noticed the same thing. However, when I went into the menu settings in the camera, I was able to locate the 1/3 increment settings for ISO and that opened up more ISO option settings for me.

 

[Prodo123]

This is your best bet. When I got my 5DIII a couple weeks ago, I noticed the same thing. However, when I went into the menu settings in the camera, I was able to locate the 1/3 increment settings for ISO and that opened up more ISO option settings for me.

Not for the entry-level cameras; they are unable to do this natively.

For the 600D you should get Magic Lantern; it enables fine control over exposure settings, including 1/3-stop ISO control.

 

[VirtualRain]

Even if you have the option, I don't think it's wise to use it in all situations...

Following is a summary of the findings:

• The camera's hardware does not support the fractional ISOs, they are created by underexposing the previous lower respectively overexposing the following higher full-stop ISO,
• The +1/3 EV ISO steps, like 125, 250 reduce the dynamic range by 1/3 EV,
• The -1/3 EV ISO steps, like 160, 320 cause an "overmetering" by 1/3 EV, which may cause clipping,
• The ISO steps 100 and 125 cause about 1/4 EV higher exposure than it should be: ISO 100 is rather 119, thus the metering for ISO 100 may cause overexposure.

Thread on POTN here.

 

[DirtySocks85]

Not for the entry-level cameras; they are unable to do this natively.

For the 600D you should get Magic Lantern; it enables fine control over exposure settings, including 1/3-stop ISO control.

Thanks. I'll look into it. As far as I'm concerned more creative control is usually better.

----------

Even if you have the option, I don't think it's wise to use it in all situations...



Thread on POTN here.

Wow, thanks for the info. I'll probably try it myself and see if it works okay with me.

 

[snberk103]

Another quick ISO experiment - I think I learned it here in fact - is how to check which are the 'Natural ISOs' (for want of a better word) and which are the interpolated.

Photograph something with a very black and a very white areas. I used one of those grey card/exposure targets that has black, grey, white, reflective white on it.

Anyway...take a properly exposed photo at a middle of the range natural ISO using a steady and even light. Check the histogram to ensure that the peaks are all contained within the histogram. In my case I had 4 sharp spikes. Then take a series of photos that run through all the ISOs while making the corresponding exposure adjustments. In theory each exposure should show the same set of spikes, in the same place (or at least pretty close). In practice you'll see which ISOs are interpolated and which aren't. The interpolated ISOs will show less dynamic range - if memory serves the peaks get closer together. What I discovered is that ISO 100 was the slowest natural ISO... so I don't bother with ISO 50 much.

Another interesting test is to run through all the aperture/shutter combinations for a particular proper exposure. Use a nice blank wall of a mid-tone... a concrete wall works great. In theory each exposure should show the same tone.... In practice...... It doesn't always work that way. If you have a good steady tripod and are careful of the focussing you can pixel peep at different apertures and assess the sharpness of the lense at different settings. It's just a seat-of-the-pants test, but it will help identify an awful lense or perhaps just an few apertures to stay away from with that lense.

Luck.

 

[farbRausch]

Anyway...take a properly exposed photo at a middle of the range natural ISO using a steady and even light. Check the histogram to ensure that the peaks are all contained within the histogram. In my case I had 4 sharp spikes. Then take a series of photos that run through all the ISOs while making the corresponding exposure adjustments. In theory each exposure should show the same set of spikes, in the same place (or at least pretty close). In practice you'll see which ISOs are interpolated and which aren't. The interpolated ISOs will show less dynamic range - if memory serves the peaks get closer together. What I discovered is that ISO 100 was the slowest natural ISO... so I don't bother with ISO 50 much.

You mean the 'interpolated ISO's' have generally less dynamic range? For example, let ISO100 and ISO 400 be natural and ISO 250 interpolated. Hence, I have less DR in ISO250 as in 400 and 100? That would be a reason to avoid these ISO settings.

 

[snberk103]

You mean the 'interpolated ISO's' have generally less dynamic range? For example, let ISO100 and ISO 400 be natural and ISO 250 interpolated. Hence, I have less DR in ISO250 as in 400 and 100? That would be a reason to avoid these ISO settings.

That I don't know... but it is an easy test. Just shoot your target and check the histogram. My camera is pretty basic. I have ISO 50, 100, 200, 400, 800. And nothing else.