Maximum f stop usable in the Canon XH A1

[Juan Parmenides]

Hi all,
Yesterday I took some sample clips to see how the A1 is working. I use TV mode in a bright light sunny day, no ND, speed 1/500. But all the samples were blurried, not so much, but blurried. Later at night I read in Manual´s page 34 (spanish) that smaller f stops can cause loss of sharpness, that´s all.
I believe is not a camcorder problem, because I have used it inside home with no sharpness problems.
My question is: What is the usable maximum f stop number to avoid sharpness problems?
I know that it depends on multiple variables, but as a general rule and based in your experience, can you, please, give me an idea?

PD: Next time I´ll use AV or A modes.

Thanks in advance.

[Tom Hardwick]

Originally Posted by Juan Parmenides

My question is: What is the usable maximum f stop number to avoid sharpness problems?

Look on it this way - all the apertures on the lens are usable, but some compromise the picture quality for various reasons.

Your Canon's zoom is designed to work best wide open, so in answer to your question f1.6 at wide angle and f3.4 at full tele. But shooting at maximum aperture you'll get vignetting - where the corners of the picture receive less light than the centre. Your zebras demonstrate this clearly, and you'll need to close down 2 stops to evenly light the frame.

At wide apertures you'll also get more internal flare, and this will gently cloud the image. The internal barrel of the lens cylinder should be a perfect matt black - and of course it never is.

Once you go smaller than f/4 (f/5.6, f/ 8 etc) then you'll gradually lose sharpness. It's not Canon's fault - it's the laws of optics at work, and diffraction will gradually soften the picture more and more.

If you're shooting HDV then don't stop down below f/4.5 if you want maximum sharpness, and at wide angle (where diffraction is even more of a problem) treat f/2.8 - f/3.4 as your smallest stop.

tom.

[Juan Parmenides]

Thanks a lot, Tom

A very usefull information from you. Coming from DSLR world my old rule of thumb is f8 for maximum sharpness, maybe f5,6 for fast lens. But, I reallly didn´t know the specific design rules of the Canon A1 zoom lens.

So, in this camcorder, recording in HDV, minimum f stop will be f4, as a rule of thumb.

Thanks again.

[Richard Hunter]

Hu Juan.

Tom's advice is very sound, but I think it would not be a good idea just to set the aperture in Av mode and then let the shutter speed take care of itself (in case that's what you are planning to do, based on your first post). If the shutter keeps changing, the feel of the video will change too, so I would recommend you try it out on several shots with movement and various light levels, and then decide that that is the effect you want.

Usually I will set a fixed shutter speed, and then use ND filters to get the aperture into the right range. The A1 has 2 ND filters built-in, and I have a couple of screw-on ND filters that I can also use if necessary.

Of course, it could just be that I'm a control freak. :)

Richard

[Mark Fry]

Originally Posted by Juan Parmenides

Thanks a lot, Tom

A very usefull information from you. Coming from DSLR world my old rule of thumb is f8 for maximum sharpness, maybe f5,6 for fast lens. But, I reallly didn´t know the specific design rules of the Canon A1 zoom lens.

So, in this camcorder, recording in HDV, minimum f stop will be f4, as a rule of thumb.

Thanks again.

Diffraction is more of a problem for the XH-A1 and similar video cameras than it is for DSLRs because the image detector is that much smaller. For good old fashioned 35mm film SLRs, it isn't an issue until you get to really quite high f-numbers (F22 perhaps?)

[Tom Hardwick]

Richard's right - for movies it's very important to lock down the shutter speed and use the camera in the shutter priority mode. I've seen footage shot in the aperture priority mode (PDX10) that showed the aeroplane's propeller coming to a stop and reversing as the Spitfire took off - all because the shutter speed and the engine's crankshaft were varying their speed and getting in and out of sync.

Mark's also right. A 35 mm SLR uses a huge 'chip' of 36 mm x 24 mm - whereas our tiny 1"/3 chips are but 4.8 mm wide. Diffraction affects us all the same of course, but the much longer focal lengths of the 35 mm camera mean you can 'get away with' smaller apertures. But then again, f/22 will be a lot less sharp than f/5.6 all other things being equal (like off centre elements and so on).

tom.

[Petri Kaipiainen]

35mm FILM cameras do not really suffer from diffraction even at the smallest apertures, as film is not as sharp as the sharpest digital sensors. But full frame dSLR:s like Canon EOS-1Ds Mk III start to loose real resolution if stopped down below f11.

Here is a semi-scientific paper about diffraction: Do Sensors “Outresolve” Lenses?

At the end is a table showing the best possible resolutions at each sensor size and f-stop. Those sensors are for still cameras, but it is easy to extrapolate and realize, that small compact cameras with slow zooms can not resolve 10-12 MPix as promised, and that 1/3" sensor HD video camera is actually impossible to make, unless you have f2 lens throughout the zoom range.

[Wayne Dupuis]

Take the camera out of autofocus, and manually focus it on your subject.
Use ND fileters outside to set an f stop that applies to what you want in focus, and or where you want bokeh to appear. The XHA1 is very responsive. A number of other things. Do a search on "sharpening + SLR" in Google and look at what most shooters understand about in camera sharpening. Use Wolfgang's -3db custom settings and read what he has to say about this.
If you point and shoot this camera at a subject in auto mode you run the risk of your subject, being in the middle of the screen, going in and out of focus because the camera does not know to follow the action. Go manual or suffer.

Wolfgang's hp blog - Presets for Canon XH A1

[Tom Hardwick]

C'mon Petri - saying film cameras don't suffer from diffraction is only correct if you don't use lenses on them. Even a pin-hole camera suffers from diffraction (very much so).

It's the iris diaphragm that causes diffraction losses, and the sharpness of the film (K25 anyone?) has no effect on it at all - it's simply the laws of light being brought into play.

[Denise Wall]

Coming also from high end still digital camera/lens background, I'm surprised there are no resolution charts of stills pulled at the various apertures and zoom distances with a fixed optimal shutter speed. The "pixel peepers" in DSLRs would have been all over this for us ;)

If anybody knows if such a thing for the the A1 with its standard 20X lens, I've love to see the link.

[Petri Kaipiainen]

Originally Posted by Tom Hardwick

C'mon Petri - saying film cameras don't suffer from diffraction is only correct if you don't use lenses on them. Even a pin-hole camera suffers from diffraction (very much so).

It's the iris diaphragm that causes diffraction losses, and the sharpness of the film (K25 anyone?) has no effect on it at all - it's simply the laws of light being brought into play.

Of course the diffraction is a lens thing, but the fact is that the sharpest digital sensors have far surpassed film resolutions. For that reason diffraction was not of much concern in the good old days of film. Now 20+ MPix full frame 35mm sensors are much better than lenses, same with mid format backs approaching 50 MPix resolutions. But the worst are 1/2" sensor 12 MPix compact cameras, where diffraction starts to kick in the moment you touch that zoom rocker.

Only after 39 MPix (sub-)midformat digital backs came to being did major field camera lens makers like Rodenstock and Schneider start making new "diffraction limited" digital lenses, which start to loose resolution already from f5.6, which paradoxically, means they are of best quality. In the times of film there was no need.

[Juan Parmenides]

Well, I did a test just a moment before. Iris full open (f1,6), then f 2.0 and f 3.2 and sharpness was really fine. more than enough for me. Several zoom positions.

[Juan Parmenides]

My DSLRs are Nikon D200 and D300 and all my lenses are fast lenses all of them, more than $1000 each, so I know what I am talking about. In DSLR world most of lens sharpness is in the middle of the f stop numbers, so I can imagine, by simple optics laws, that in a different range of f numbers can change the rule. The Canon is f1,6 -9,3, theorecatelly, the middle is about f4, but, of course this is not a scientific theorema.
Sorry but I can´t express myself as well as I would like, my native idiom is spanish.

Thanks all. Really nice forum.

[Tom Hardwick]

Sorry Petri and with respect, but what you're saying just isn't true. Diffraction was certainly of GREAT concern in the good old days of film, as anyone shooting Standard-8 or Super-8 film (5.4 x 4.1 mm frame) found out to their cost.

Then you say that, 'diffraction starts to kick in the moment you touch that zoom rocker', and of course this is nonsense. The focal length of a lens doesn't cause diffraction on its own, it's small apertures that cause diffraction. The shorter the focal length the more diffraction affects the sharpness, but it's still aperture dependant.

It's the reason the Z1 allows you to limit how far the diaphragm blades stop down - Sony knows what shooting at small apertures will do to their reputation.

tom.

[Chris Hurd]

Same with the Canon XH and XL H series... f/9.5 is about as small as you're allowed to go in most program modes. You can force it into f/11 to f/22, but at those apertures diffraction is indeed a serious problem affecting focus.