>>3081369The histogram is often a more important tool than the image preview, but of course both should be used to judge the scene, otherwise you may end up with a lot of under/over exposed scenes in general.
>>3081371This also depends on the camera. One thing to keep in mind is that almost always the histo is generated based on the jpeg preview and not the RAW data, so you have to take into account the way your camera renders jpegs. You should experiment to see at what levels you camera actually clips highlights compared to the amount displayed. Also see if you can take advantage of picture styles to edge the jpeg rendering closer to what the RAW files is capable of, usually by selecting a neutral setting and lowering contrast.
Since the preview is always an 8-bit image, the amount of dynamic range shown is extremely limited and makes matters worse on cameras that near 15 stops. For example, on one camera there may be an even amount of "leg room" to either end of the histo, which makes sense for ETTR, because there is actually 1.5-2 stops of DR past the limit of the graph and you'll often be under-exposing if you believe that the right edge of the graph is the clipping point. On another model, the histogram may actually be more representative of where you'll clip highlights, and so you should still ETTR, but avoid hitting the right-hand side and not worry about shadows, since you can always bring them up in post.
>>3081395Looks perfect now. Next step is to learn to image-stack, if you're keen on wanting to photograph small objects and actually have stuff be in focus. As long as you're on a tripod and want to have more in focus, don't bother going over f/8, just rack focus, take several shots and use some software capable of image stacking.
