Why RF doesn’t have more focus points?

[Epimetheus]

Is there a technical reason why a rangefinder camera doesn’t have more focus points? I don’t mean like 51 focus points or anything like that but like two more, one for each side. It would reduce focus problems caused by focus-recompose technique at close range with wide aperture. Wouldn’t it?

 

[Pherdinand]

Heh
I had the very same question some time ago. Exactly with the same reasoning: why should be only one in the middle whiuch is clearly unfavourable for the composition.

I got a reasonable answer. It said, the finder with the double images would be way too crowded and would be difficult to quickly find the way through it.
Besides, it's already difficult to align the rangefinder spot (one spot) in such a way that the top and the bottom part shows focus in the same time. I can imagine, with multiple spots it would need reeeeal good alignment.

I must say i never had focus error from recomposing, only from bad focusing itself.

 

[Epimetheus]

I must say i never had focus error from recomposing, only from bad focusing itself.

Problems from focus-recompose technique sure can be corrected by experience and careful focusing. I was just wondering if there was something in the mechanics of the rf that prevents multiple focus points but I guess it’s more practical reason then technical.

 

[ywenz]

51 focus patches in the little RF window? no thanks.

 

[mervynyan]

focusing is only on one focal plane, that is given, double image focusing can't be adjusted indenpendently even you have 2 or 3 focus points. Plus the viewfinder mag is a fixed ratio, you can't really observe to be sure it is in focus.

 

[payasam]

Jenni, you've said exactly why there is a single focus mechanism. Pherdinand, trouble is unlikely except when the lens in use has a curved field.

 

[JNewell]

Further, although this is YMMV territory, I am a little skeptical about the real life value of multiple focus points. I have a DSLR (sorry!) that has multiple focus points and I think I've used that "feature" exactly once. In general, either it isn't a problem to focus/lock/recompose or it's actually a disadvantage. Further, because I don't think it would be possible to turn "off" multiple patches in an optical/mechanical range/viewfinder, I think it'd be a real pain for most users most of the time.

 

[dazedgonebye]

Four patches, one on each of the 1/3rd intersections.

 

[einolu]

because only one thing can be in focus at a time... anyway rf's require you to pick what you want to focus on before you compose.

 

[dberger]

Four patches, one on each of the 1/3rd intersections.

Close!

5 pathches -- 1 in the middle (like normal) and four with their inside corners at or just outside the 1/3 intersections. In principle, I do not see why a viewfinder-sized rangefinder window plus a masking system couldn't be used to turn on and off selected patches. Ideally, one would be able to choose:

1 - All off (i.e., no rangefinder)
2 - Center patch on
3 - All patches on
4 - Corner patches on

Patches would be switchable with a horizontal thumb wheel on the right back of the top deck (like the focus wheel on a barn-door Vitessa). While it would be possible to use the entire viewfinder image as a rangefinder (as the NHSwebmaster points out), that would be a bad choice in my opinion, because in many circumstances most of the view would be out of focus at all times. (Clearly I've given this too much thought.)

Of course, this would not be mechanically trivial. So, I expect to see a working prototype by 1Q 2008.😀

cheers,

David

 

[JNewell]

Of course, this would not be mechanically trivial. So, I expect to see a working prototype by 1Q 2008

That's an understatement. Does the second sentence mean that you'll have a working model ready for us by Q108??? 😉

 

[oscroft]

Four patches, one on each of the 1/3rd intersections.

But the 1/3rd intersections move when you change the lens and use a different VF frame

 

[dberger]

But the 1/3rd intersections move when you change the lens and use a different VF frame

Ahh, quite right.

But, that's OK. If the outside RF patches are too crowded with longer-lens frame lines, just use the central patch. The Multi-Patch RF System™ would be most useful with 50mm to 35mm lenses, where the desired focus point is often off-center. The system would be less necessary with 28mm lenses, where DOF could cover many focus-recompose errors.

Ciao,

David

 

[Pherdinand]

Honestly...do you have the time to screw around selecting / switching on/off the desired patch when shooting with a rangefinder?
Maybe when shooting flowers, or landscape, yes.
I myself hardly have time to push the shutter button sometimes.

 

[Pherdinand]

I have such multiple AF points in my minolta SLR, but i either leave it to the computer to pick one (e.g. wqhen i use a small aperture and large distances) or i just use the central spot switched on, the rest are off.

 

[wyk_penguin]

The first things that come to mind are:

1. Masking different patches should be simple enough, but I doubt that it would be useful for the street shooter simply because it will take up another fraction of his/her time. You would have to view the scene, compose, and then choose which patch to use, focus, recompose (half of my photos have moving subjects), (re-focus) and finally shoot. And that's assuming you have already set the exposure. But for a 1 patch system, you focus on the subject and then worry about composition later.
2. I find myself using the center AF spot of my DSLR most of the time. The time that I use the other spots is when I have my DSLR on a tripod where it is difficult and awkward to point the camera at the subject recompose afterwards. On a tripod, my workflow is to tweek the composition and then focus on the ground glass of my DLSR manually (like using a view camera and often I wonder if I need the 9 point AF.)
3. The RF mechanism would be obscenely large since you are now talking about the RF patch virtually covering the whole VF (though a large part of the RF image would be masked at any one time). My guess would be the RF mechanism would have to be 9 times as large as it is now, since the RF patch in most cameras is about 1/3 of the VF in width and height.
4. Everything else set aside. I would say it is uneccessary to have multiple RF patches simply because you don't really need you subject smack in the middle of the RF patch to focus on it. For example, a face in the 1/3 position could be focused with the corner/edge of the RF patch where the corner of the persons eye could be used as the target which would give acceptable results if your focal length is not too long (90mm or less) and you use something smaller than f/4. You could argue about the accuracy, but using an off-center focus point doesn't really solve the problem since we don't have lenses with a perfectly flat field (and a corner resolution equal to that of the center) at 90mm at f/1.2 yet. Therefore we would still be using depth of field.