What happened to ISO 100 on My T1i?

[photography by a.f. smith]

<p>I recently activated the ISO expansion on my T1i and noticed that when I did my minimum ISO became 200. I went back in and turned it off, but guess what, ISO 100 didn't come back. What am I missing?</p>

[zml]

<p>HTP set to on..?</p>

[phule]

<p>Indeed, you turned on Highlight Tone Priority. </p>

<p>http://www.dpreview.com/reviews/canoneos500d/page17.asp</p>

 

[mark_t5]

<p>yeah..what those 2 said</p>

[fotoscopio]

<p>Yes, it's highlight tone priority. It will make your lowest ISO 200... BUT</p>

<p>¿Can anyone elaborate on when to use it and when not to use it?</p>

[robin_sibson1]

<p>There are two schools of thought about what HTP does. Assume you are shooting in RAW. One says that it reduces the ISO by one stop from what you have actually set (so that you are under-exposing) and then doubles the recorded value to compensate by 1 stop, and that's it. That's why you can't go below ISO200 when HTP is active. The effect is to give more headroom for the highlights but, because you are pulling the shot up by a stop, at the expense of more noise. If this is correct, you could achieve exactly the same effect manually, and there would be very little point in having HTP.</p>

<p>The second school of thought says no, it's a bit more subtle than that. All of the above happens, but above a certain point the relationship between the charge captured in the photosite and the digital value recorded in the RAW file is tapered off to give a changed relationship between scene brightness and image brightness in the highlights. If that is correct, HTP has a real effect on the RAW file (unlike, say, colour temperature, which is just a recipe applied later) and that effect cannot in general be reproduced by manual adjustment and post-processing, so it is potentially useful in some circumstances.</p>

<p>When HTP was introduced, someone looked at RAW files at the binary level with and without HTP and it appeared to them that the first explanation was true, because the histogram with HTP hit only every even value, and otherwise matched the shape of the non-HTP histogram. But as far as I could see the histograms stopped so far short of the highlight end of the scale that they may well not have explored the region where, on the second explanation, HTP would be doing something different, and I did not regard the question as having been resolved one way or the other. Canon's own literature seems to suggest that the second explanation may be the correct one, but it's far too vague for certainty. It would be nice to know the truth of the matter. Unfortunately many of those who post confident assertions (usually that the first explanation is correct) seem simply to be echoing what they have read somewhere rather than offering real evidence.</p>

<p>When to use HTP, if at all? I've experimented with it quite a bit, and reached the conclusion that it is occasionally useful but not to be used unless it is needed. I use it if I am photographing a normal to reasonably bright scene (so I don't need to worry too much about noise in the shadows anyhow) which contains a few very bright highlights. In particular, some of my botanical work involves flower colours with an amazingly powerful red response that easily blows out the red channel. This can be very difficult to overcome. HTP can help here, but further under-exposure may be necessary over and above the effect of HTP. I guess that the needs of wedding photographers may have been in Canon's mind when providing HTP, and it would be interesting to know whether it is proving to be of any value for such work.</p>