A good reason for not using hyperfocal distance ???????????

[ajohnw]

Stumbled on while looking for something else.

http://www.trenholm.org/hmmerk/DOFR.html

Sounds rather good to me. And where the HFD calculations came from is absolutely correct.

John
-

[jcuknz]

It saves you having to learn them, or buying the app for your cellphone etc

[Glenn NK]

Another of his articles (which includes some info on HFD):

https://app.box.com/s/04y4qncm7fxdxgzt9cqb

[Manfred M]

Frankly, in the old days of film, where every shot was precious due to the cost of processing the film, shooting prime lenses with the HFT scale debossed into the lens barrel and a very large throw (the better part of 300 degrees) to focus a lens; yes, I paid attention to this.

Now, in the modern world where I shoot with a DSLR, use zoom lenses that autofocus; where any manual focus is a bit more hit and miss because of the lens design; I tend to not even worry about this; for my shooting needs. I bracket shots and guesstimate the in-focus distance. For the type of work I generally do, this works very well.

[ajohnw]

Another of his articles (which includes some info on HFD):

https://app.box.com/s/04y4qncm7fxdxgzt9cqb

That PDF is about on the site I linked to as well.


It's interesting to read his comments about lens scales and larger than 10x8 prints. That's one of the reasons I have mostly regarded them as a bit of a waste of time. I didn't realise why at the time but there is more info about on how the scales were obtained now. 10x8 was regarded as the max for 35mm film but even that aspect is a little odd as the Europeans reckoned that smaller circles of confusion where needed so had different scales to the Japanese. Most people I know of at the time focused to infinity or some what but not much short.

A more realistic limit for film was 20x16 even for colour. Never went to that size on colour as learning was just too expensive and time consuming so went to slides.

He has used approximations in his method but I feel it's worth some thought as often foreground detail is the part that needs consideration. Not much thought by me at the moment as my head is full of GIMP but there are signs of a method of maximising detail definition over any distance there.
John
-

[ajohnw]

This might help clear this area up. It's an approximation just like the usual hyperfocal distance calculations. Zero distance in practice will be some multiple of the lenses focal length, probably low. The other aspect is what resolve means. Say it worked out as per the example 1.75mm. Something of that size will be captured but can not contain significant detail. The other problem is that the camera may not be able to actually resolve it, lens neither. At least it will as good as it can be in those circumstances. Using the actual size of a lenses aperture makes some sort of sense as the angular resolution of an optic is set by it's diameter and the focal length just sets the image scale.



I have to do something like this to get my head round it as simple as it is.

Actually I suspect that the figures it comes up with are dead safe but am not going to try and work it back from angular resolution - that one based on diameter is not an approximation.

This sort of thing is probably of great interest to people who make large landscape prints in particular. Actual resolution is more important then. HFD assumes a 10x8 and states nothing about resolution as the article points out. Also shooting wildlife and portraits. That's why there is a 0.9 of the distance in the graph. In case no one notices each entry adds up to 1.

John
-

[Brev00]

I have never been dissatisfied shooting in my own quasi-hyperfocal method (using good guesses from having read charts but not anything so precise as an actual chart or app in the field). Then again, I don't print 20 x 30 so the 8 x 10 standard is fine with me. That is, if I am reading (skimming) this article correctly. Quasi-reading, I suppose. I tend to skip past stuff that is hurting my brain. Which is why I go by my own results rather than extensive, mathematical treatises. I prefer to let others sum things up for me if that is needed. Maybe I modify things because distant sharpness on my quasi-hfd shots seems fine to me. Gotta say, I love shooting the old-fashioned way. The clarity always feels like a nice surprise when the dof is not there in the viewfinder the instant before the shot.

[ajohnw]

The problem with finding something like this even with the maths and the fact that it used to be in a book doesn't mean that it's correct. The results he shows though suggest it is correct.

Focusing on the horizon and picking an aperture based on previous results used to be a pretty normal way of shooting landscapes. Same with focusing on some object in one. This method gives some idea of what can be resolved and doesn't need distance measurement. Just fractions of some distance which is much easier to guesstimate.

The book can be downloaded now. He gives some interesting examples of the use of the math - resolving stripes in a tee shirt and not lettering on a vehicle. Seems to me that this could be used to set back ground blur and is a lot simpler than the usual way of doing it. The chart can be used for that as well. It could also be used to set a max aperture for a given situation.

http://www.trenholm.org/hmmerk/download.html

The thing I like about it is that it doesn't even remotely relate to how big a print is produced really, only what can be captured. He explains why he went this way towards the end of the book. Also a fact that in the distance the hyperfocal method does limit what can be resolved. To improve things there is a need to focus past it and close up the same thing happens limited by the sums he provides = the aperture. Then there are pixels in digital cameras which set another limit so people might find they can't resolve it anyway. Then there is the lens itself.

John
-