Hassy H4D/40 Release

[etrigan63]

I got to play with one at the Launch Event in Miami. Very slick. True Focus works. Sounds like science-fiction but it actually works. Focus is very fast and the viewfinder is very bright. The TF button is on the back of the grip and the operation goes like this:

Compose and focus pressing the TF button on the back of the grip. Release the button and recompose. Depress the shutter half-way and the camera will adjust for the positional change. Shoot the photo.

Very easy to get used to.

 

[David Klepacki]

Read the thread over on LL as well as some of the online pdf's. Not sure why this technology is claimed to work better on wide angle lenses. Thought they typically had greater DOF than longer focal lengths. Not being a trigonometry guy I wonder what the actual adjustment in distance is when shooting (vertical) a full length shot of a model with a normal lens, i.e. the difference in distance between the focus on the eyes and the recomposed shot.

I think you answered your own question about wide angle lenses ... the DOF is the saving grace for compensating inaccuracy.

As for the amount of adjustment in distance for your model, a lot would depend on how still your model can be during your exposures so that the "True Focus" can function properly. For example, if you are shooting at F2.2 with a very shallow DOF, my understanding is that if the model has a tendency to rock slightly back and forth even by less than a centimeter, you will not be able to get the eyes perfectly in focus unless you have the uncanny ability to rock back and forth at the same frequency and in perfect sync with the model's motion, all while re-composing the shot. Another example is shooting infants or children or even pets and animals .... good luck in getting these subjects to remain perfectly still in order to reap the benefits of "True Focus".

Unfortunately, the Hasselblad "True Focus" mechanism only attempts to measure the camera movement and not that of the subject. I find the Canon 1D predictive focus technology to be much more effective in practice, since it does not matter who or what is moving. For example, the Canon technology can track something moving in its frame as fast as 100 mph at a distance of about 50 feet, and continuously track it at 8fps capture rate. IMO, that is a much better example of "true focus" capability.

 

[thomas]

I am really impressed of the ISO1600 posted here: http://luminous-landscape.com/forum/index.php?showtopic=41557&view=findpost&p=346549
Okay, it's just a single example, however that looks really promising.

 

[David K]

I think you answered your own question about wide angle lenses ... the DOF is the saving grace for compensating inaccuracy.

I went back and re-read the Hasselblad True Focus pdf (skimmed it too fast the first time) and now understand why they said this... see quote below.

"a. The closer you are to the subject, the worse the original problem becomes. Consequently, the need for True Focus solution becomes greater and its application thereby becomes more noticeable.

b. Short focal length ('wideangle') lenses naturally decrease camera to subject distances and therefore, following the point in (a), produce a greater need for True Focus adjustments."

As far as Canon's (or Nikon's) AF systems I'd agree with you that they are head and shoulders better than any MF system I've ever shot with. When I focus and recompose with my Nikon (typically portrait mode for people shots) I move the AF point as close to the eyes as I can get it which results in less of a "focus error" than using the center focus point.

 

[David Klepacki]

David K, Here is a puzzle for you:

The well-known 1/f rule for hand-held shutter speed indicates that the longer the focal length of the lens, the faster your shutter speed should be in order to overcome any shake/motion by the photographer. Yet, the Hasselblad "True Focus" literature implies the opposite in that adjusting for motion with longer focal lengths is not as critical as for wider focal lengths. Why is this? And, to be more specific, consider the following.

If Hasselblad's claim is true about needing less focus correction with telephoto lenses, then shooting a 500mm lens at 1/15 would not need as much focusing care as shooting a 15mm lens at 1/15. Yet, we all know this to be not true. The angular displacements that occur during 1/15 second with a 500mm lens are much more severe than those experienced with a 15mm lens in the same duration. So, if angular displacement errors are much less of a concern with wide angle lenses via this observation, then why is "True Focus" needed more?

Now, of course, if the shutter speed in my example above was changed to 1/500, then there would be much less of a visual difference, but the point of the example is that the camera movement here is simulated with longer exposure times, and more severe angular displacement of the subject occurs with a longer focal length (as born out via the 1/f rule), and this displacement occurs regardless of whether one is exposing during this displacement or not. So, why not a greater correction needed for telephoto lenses?

 

[Nick-T]

True Focus has nothing to do with shutter speeds, it is about compensating for focus errors caused by focus recompose. With a wide angle lens the change in angle (when re-composing) is greater than with a telephoto and therefore the correction needed is greater.
Nick-T

 

[David Klepacki]

True Focus has nothing to do with shutter speeds, it is about compensating for focus errors caused by focus recompose. With a wide angle lens the change in angle (when re-composing) is greater than with a telephoto and therefore the correction needed is greater.
Nick-T

My point is about camera movement and corresponding relative angular displacement of the subject, which has everything to do with True Focus. In fact, a telephoto lens has a much narrower angle of view. For example, in 645 format a 200mm lens has roughly a 20 degree angle of view, while a 35mm lens has roughly a 90 degree angle of view. So, if the camera movement (when re-composing) is say 10 degrees, this would constitute a 50% change in angle relative to the 200mm lens, and only 11% change in angle relative to the 35mm lens. So, in fact the wide angle lens does not experience a greater change in angle.

And, indeed this is the basis of the 1/f rule having to do with shutter speed selection and why a wide angle lens can be shot with greater clarity at the same (slower) shutter speed as compared to a telephoto lens. It is precisely because there is much LESS relative angular displacement that occurs with a wide angle lens due to camera movement. Whether this camera movement is due to mirror slap, shutter vibration and/or hand shake, it is a movement nevertheless and not any different than movement due to re-composing.

 

[jlm]

movement due to re-composing effectively results in a changing camera to image distance of the point of interest only

movement occurring during the open shutter is omnidirectional

 

[Nick-T]

David I'm not going to argue with you. I'm well aware of the 1/ rule for choosing shutter speeds and have had good success with it in 25 years of shooting for clients. I have also extensively tested the H4D40 and true focus (I'm shooting with it every day) and can assure you that the recompose focus error is much more evident when working close in with wide angle lenses than it is with a longer lens further away ie two different lens and the subject filling the frame to the same degree. Hope that makes sense.
Nick-T

 

[David Klepacki]

Hi Nick, I did not mean to argue, so I hope you do not perceive it this way. I was only having some fun by stimulating some thought about the process. The important thing is that if it works for your type of shooting, then that is all that matters.

 

[David Klepacki]

movement due to re-composing effectively results in a changing camera to image distance of the point of interest only

movement occurring during the open shutter is omnidirectional

Yes, there may or may not be omnidirectional movement. The question is why should any movement be less critical for a telephoto lens than for a wide angle lens, especially when the relative angular displacement of a telephoto lens is always greater than for a wide angle lens for the same amount of camera movement.

 

[David K]

David, I appreciate your thoughts on the subject. I found this aspect of the True Focus counter-intuitive to how I thought it would work and wanted to understand it better.

Nick, thanks for your input too. However it works Hassy get's my kudos for an innovative feature that has real world application for how and what I shoot. I'm looking forward to trying it for myself.

 

[David Grover / Phase One]

Hi All,

As Nick says, the shutter speed has nothing to do with true focus. The True Focus correction simply applies an offest to the AF drive before the image is captured.

The reason why it is less noticeable at longer lenses / distances, is due to the trigonometry. As you move further back / increase focal length, the angle Ø decreases and therefore so does δ.

See attachement!

David

 

[fotografz]

Okay, help me out here RE: the True Focus concept.

Isn't off-center recompose a known and well documented issue? Isn't it why the 35mm DSLR makers employ movable multiple focus points, and have been striving to improve the sensitivity of the off-cented focus points?

The downside of the multiple focus points continues to be that they really are not out far enough from the center, don't always match where you want the focus point, and they still are not as sensitive as the center point ... and if solving all that were an easy task, wouldn't Canon/Nikon have done it already? Doesn't that indicate how difficult it would be to apply multiple focus points to Medium Format which is way farther away from the center?

I use the multiple focus points on all of my 35mm DSLRs for thousands of wedding shots a month, so I'm very fast at it. Yet, it is an imperfect solution at best. You have to wheel the focus point to the subject ... which may or may not be where you want it. So you get as close as you can then re-compose anyway ... only now at least it's closer in terms of distance. But, again, I still almost always have to re-compose. And subject movement or camera movement is the same either way when using single AF point.

This "True Focus" concept seems to solve those issues, at least in part, since it doesn't matter where the subject is in the composition ... you lock and re-compose without the time needed to wheel to a different focus point, then lock and re-compose like you do with a 35mm anyway. True Focus seems faster to me, and I dearly wish the 35mm DSLRs had this feature.

That it is now available in Medium Format with it's more shallow DOF seems pretty useful to me.


What I don't quite grasp is the role DOF plays in True Focus ... I get the differences between field-of-view of WA verses longer lenses, and made myself a "teaching aid" to show it ... where clearly the distance to off-center subject is greater with a WA field-of-view. On that chart are yellow lines that indicate distance ... what is missing is an overlay showing the effect of Depth-of-Field inherent with WA lenses verses longer lenses. I guess that would be possible, but you'd have to be pretty specific as which lenses, then map out DOF for each.

The other thing is flatness of plane which seems more critical an issue with WA than long lenses. Wouldn't hitting the off center subject exactly with your focus point aid in that?

Just wondering out loud, since I haven't actually tried True Focus and will reserve judgement. I just figure it can't hurt and I can focus as usual in some cases where the subject is close to the center.

-Marc

 

[David Grover / Phase One]

Marc - did we post at the same time?

Have a look at this link on the Hasselblad site, and this will show what you are looking for...

http://www.hasselblad.com/media/2234814/when true focus makes a difference.pdf

 

[thomas]

Maybe this one also helps to understand why recomposing in general leads to "backfocus"...
http://mhohner.de/images/recompose.gif
... and to understand that the larger the movements of the camera the more apparent the focus error will be.
I understand that "True Focus" only compensates for "tilt", i.e. vertical recomposing (ist that correct?), however the basic principle is the same.

 

[David Grover / Phase One]

True Focus will compensate for landscape recomposing as well. As long as it is rotational it doesn't matter what plane it is in.

 

[thomas]

True Focus will compensate for landscape recomposing as well. As long as it is rotational it doesn't matter what plane it is in.

ah okay - thanks for clarifying!
So on a tripod with non moving subjects it should work very, very well, I guess. Handhold you have to live with the movements of the subject and yourself... which is still one error factor less than with "regular" AF.
If it works as it is supposed to be (obviously it does) this feature seems to be one of the most usefull recent "inovations" in MFD... IMHO.

 

[David Grover / Phase One]

ah okay - thanks for clarifying!
So on a tripod with non moving subjects it should work very, very well, I guess. Handhold you have to live with the movements of the subject and yourself... which is still one error factor less than with "regular" AF.
If it works as it is supposed to be (obviously it does) this feature seems to be one of the most usefull recent "inovations" in MFD... IMHO.

Yes, you are exactly right on all points.

 

[fotografz]

Marc - did we post at the same time?

Have a look at this link on the Hasselblad site, and this will show what you are looking for...

http://www.hasselblad.com/media/2234814/when true focus makes a difference.pdf

Thanks David, that explains the DOF part pretty clearly. Since I am a wide open aperture shooter this could help a lot.

I also cannot wait to try it with the 100/2.2 which I off-center focus wide open with a lot ... usually really close.

I understand the issue of body sway or subject movement, but that's there anyway no matter what even if you manually focused and recomposed ... and it seems that this would mitigate that issue somewhat depending on which direction the movement is. Lots to explore with
my new H4D/60 when and if it ever gets here (sorry, couldn't help it, I'm "so waiting" for the darned thing).

-Marc

Okay, here's another focus suggestion for you to forward to the Elves in Sweden or Denmark or where ever they now work ... dual select focus so you can pick a back point and a front point and the camera selects the correct f stop. Nothing new, but would be VERY useful ... and maybe that could even work with Manual focus using the HTS/1.5 huh?

Yeah, I know, we're never satisfied ... :ROTFL:

Marc