f stops and sweet spots

[Josh Bass]

So I hear tell that "most lenses go soft around f1.6, so if you're going to shoot wide open, the widest open you should go is 2.0 or so."

I also hear tell that my cameras have a sweet spot around f5.6-8.0, where color reproduction is better, video is sharper, everything, in general, is hunky-dory-er.

so I did some simple, unscientific tests. Using my Sony PVM 14m2u NTSC monitor, 13", 500 lines of resolution, I couldn't really see any significance between f1.6 and 2.0 or 2.2, nor could I see anything significantly sweeter about the sweet spot than at other, allegedly more bitter/sour/salty fstops.

Is it me? Is it my monitor? Do these issues not apply to a noticable degree with prosumer miniDV cams (i.e. using Beta SP, DVC PRO, etc., you'd notice these issues much more)? What's the dealio?

All tests done with an XL1s. Also, I used the 16x manual lens.

[Andre De Clercq]

The sweet spot storry is allmost gone... It used to be experienced mainly in the (much higher resolution) photography especially on the larger negative formats and medium quality optics. The reason why it faded away is twofold: lower resolution in video masking potential lens resolution limits, advances in computer aided lens design resulted in better lenses. Remember however that high F-numbers with small target (CCD) dimensions can, due to diffraction, serously lower picture resolution.

[Don Palomaki]

Check some of the lens tests in Popular Photography for evidence of sweet spot with lenses. But video ls lower resolution, and smaller sensor then film, so as Andre niotes, they may be masked by the CCD resolution.

[Josh Bass]

Ok. Thanks.

[Jeff Donald]

Generally speaking, the "sweet spot' is two to three stops down from maximum aperture. So, on a f/2 lens the "sweet spot" would be from f/4 to f/5.6 The effects of diffraction would start to be noticed by f/8 and would be readily apparent by f/11. Stopping a lens down helps reduce several optical defects such as astigmatism and coma. However, stopping down the aperture on lenses with aspherical aberration will cause the focus to drift and may be what some XL1 users notice when they zoom and focus their lens.

As Andre and Don pointed out, the low resolution of video hides most of these optical defects. Many lower end cameras only stop down to f/8 or f/11 because the diffraction is so bad on these smaller lenses. HDV and its increase in resolution may make some of the optical defects more of a concern in the future. But for now I wouldn't be overly concerned.

[Dan Uneken]

Much more important is the amount of depth of field than the sweetness of the spot. This is unfortunately less noticable in video than using 35 mm film.
Beware of the f16 or f22 though. The XL-1 standard lens for instance goes entirely soft there, so much that it makes you wonder why those apertures are there at all.

[Josh Bass]

Does it go down to 22? I haven't used the auto lens in forever and the 16x manual lens closes after f16. I used f16 once for shooting a sunset, right into old Sunny, with the 1/64 (or whatever the more. . .uh. . .dense neutral density filter is).

[Dan Uneken]

Yep, you're right. I had to take out the cam and look at it. So that's f16 and f11 that one should stay away from. How is the f16 sharpness on the manual lens?

[Josh Bass]

Looked fine to me. . .but then again, I never "get" a lot of the comparison between this and that business, where people compare sharpness, and whatever else. How much "softness" am I supposed to notice? Are we talking subtle, or "your retarded grandmother could tell" kind of difference? If I see it on my 13" NTSC monitor, and it looks good, then I say it looks good. I might put that sunset footage up on my site, just cause it's real purty, and makes yer heart get all melty and runny and whatnot.

Never got that whole "XL1s' picture is too soft" business either. For about 9 months I edited, alternately, footage shot on my cam, and footage from PD150s/TRV900s, every week, and really couldn't see an appreciable difference. Maybe it's the manual lens on my cam. . .maybe I'm a fool. Maybe everything I do, even typing right now, is a dream. Anyway, only acknowledge the first paragraph here.

[Dan Uneken]

Oh man, the f16 on the XL-1 you notice big time, in the viewfinder even with your contacts out. Sometimes when you're working you just want to get a bit darker and toggle the iris wheel and then it all goes hazy and a bit more it all goes black !!! (Close iris) Great footage!!
One should be able to install a limiter on that switch. Perhaps they have it on the XL2.

[Josh Bass]

You're talking the XL1, for sure, not the XL1s? I use the terms interchangeably, sometimes, so I just want to make sure we're on the same page.

Either way, the footage I'm talking about looked great to me, very arty farty, demo-reel worthy kinda stuff. I'll post it when I feel not lazy.

Maybe they fixed that problem on the XL1s? Maybe, again, it's the manual lens? "They" say it's noticably sharper than the ISII, so. . .whatever.

[Jeff Donald]

The manual lenses seem to suffer less from diffraction than the stock lens.

[Dan Uneken]

XL1, not S.
I bet it's the manual lens. Maybe straight into the sun and a bit soft is just plain purty as you put it. Worth checking out on a less poetic subject, like a group of people. I find the results with f16 on the standard lens unusable.
Curious how that 20x standard lens for the XL2 hold up at maximum f-number. Will check out the grapevine there.

[Andre De Clercq]

Jeff, for a given target (CCD) size, diffraction effects are only f-number related. What sometimes happens and confirms your experience, is that for all kinds of reasons (tolerances, design constraints.. ) the physical axial position of the stop (aperture blades) is not perfect. This results in sub-optimal compensation for spherical aberations in lens systems, when used at small aperture settings. This causes resolution loss simular to diffraction effects.

[Jeff Donald]

Andre, I agree, I'm more used to seeing it in still cameras that may be focused at wide open aperture and the focus drifts when the lens is stopped down to taking aperture and the frame exposed. I've had students focus with the XL1 lens wide open then make adjustments to the lights. after adjusting lights they adjust exposure and start shooting without readjusting the focus. I think they are encountering the axial shift of the focus point because of spherical aberration.