Wildflowers

Here is the first photograph from the P45 that I posted on my site:




Phase One P45 on Hasselblad 503CW with 150mm Sonar with metering prism and handgrip motor.

ISO 100
5 seconds exposure
Custom white balance
Converted in Capture One 3.7 and optimized in Photoshop CS3.
Two captures blended together using Photomerge

Location: The Sonoran Desert North of Phoenix, Arizona near Lake Pleasant.

If you know the name of the purple flowers and of the plant behind them please post it here. I didn't look them up.

Hi Alain,

Although MF digital photography is way beyond my capabilities/budget right now, I can still enjoy the beauty captured with it. Thanks for showing.

PS: I don't know the name of the plants and/or flowers, sorry.

Cheers,

Cem

Bonjour Alain

nice flowers! and sweet velvet look of the leaves as well…
would you mind posting a 100 % crop so we an see the reaIQ?

Merci beaucoup !

Good idea. Here's an unsharpened 100% crop:

Thanks Alain!

Alain Briot said:

Good idea. Here's an unsharpened 100% crop:

Click to expand...

Alain, thanks for that crop. It does look a bit odd though, not what I'd expect that an unprocessed P45 image could look like. It has a mix of detail (spatial frequencies). Maybe it's detail suffered from diffraction due to a small aperture (looks like approx. f/22, although subject motion might also play a role) to compensate for the limited DOF due to the proximity. Depending on the sweetspot of the lens, I'd expext the 6.8 micron sensels will allow optimal performance at the pixel level at about f/8, and will progressively deteriorate at narrower apertures due to diffraction.

Just curious, what aperture did you use?

Bart

Bart_van_der_Wolf said:

Just curious, what aperture did you use?
Bart

Click to expand...

f32 with a 150 Hasselblad V (Zeiss Sonar).

I think Bart is on to something here. With that size pixel at f32 the light, (considered as a wave front), is sort of scraped by the edges of the aperture causing ripples across the film plane for each point of resolvable detail in the image on the sensor.

As an aperture circle gets larger, the "edge effect" is less and less until pretty well most light passes unhindered. When the circle is smaller, the edge assumes greater importance diffracting more.

I think what I've said is correct and in any case, Bart will set us straight!

When pixel diameter is smaller than the now expanded size of that fine detail, the detail is smeared over to an adjacent pixel. This can cause a bleed, for example of a green color to an adjacent white and a softening of the sharp edge.

With a progressively larger apertures, the interference with the light wave is less important for a given pixel size and spatial frequency of recognizable micro detail in the delivered image. Even though DOF increases with the tiny f32 aperture, the light smearing by the edge of the aperture diffracting (sort of scraping the light waves) really degrades the definition of what could be otherwise be observable fine detail when the image is finally delivered and observed at a suitable distance.

Asher

Alain,

The colors and the drawing of the lens are beautiful and the leaves have a great dimensionality. This Phase One back and the superb lens appear to be a killer combination.

It would be wonderful if you might do a series of shots on a tripod at various aperture to see where the sweet spot actually is.

Thanks for sharing.

Asher

Alain Briot said:

f32 with a 150 Hasselblad V (Zeiss Sonar).

Click to expand...

Ah, that explains, it's even a stop narrower than what I estimated. It might be worthwhile occasionally to experiment with focusstacking when the subject is motionless. Shooting several frames at f/8, gradually focused further away, will allow to be recombined into superior detail and DOF.

Bart

Bart_van_der_Wolf said:

Ah, that explains, it's even a stop narrower than I estimated. It might be worthwhile occasionally to experiment with focusstacking when the subject is motionless. Shooting several frames at f/8, gradually focused further away, will allow to be recombined into superior detail and DOF.

Bart

Click to expand...

Hmmm, as usual, interesting Bart, I like that idea! how would you combine the files?

Asher Kelman said:

Alain,

The colors and the drawing of the lens are beautiful and the leaves have a great dimensionality. This Phase One back and the superb lens appear to be a killer combination.

Asher

Click to expand...

Yes, the color, tone and contrast qualities are quite something. It's very different from 1DsMk2 files, and not only (or not so much) because of the higher pixel count. There's something about the sensor, or the way the files are recorded by the back, that creates a very unique color and tone quality. As you mention the Zeiss lenses most likely play a role in the look as well. I know that they are excellent for edge definition and for contrast. One of the qualities of the files is their beautiful contrast. Not too high and not too low, just right. I rarely have to adjust contrast with these, either during or after conversion.

During Raw processing far less work is required, the most important aspect being achieving good color balancing. The P45 is very sensitive to color balancing -- it has to have a very good color balance for the files to come out at their best.

There's far less need to reduce the amount of blacks during conversion because the shadows remain open even though there's a significant amount of black generated during raw conversion.

I'm working on converting the photographs from my commercial shoot in France this past week as we speak. They want them now which means I'll have to recover from jet lag later on! I brought back nearly 60gb of Raw files, half P45 and half 1DsMk2 (30 gigs of each). The P45 files are superb. I used the back on the SWC-CM, the super wide Hasselblad with the Biogon 38mm lens, which has by far the sharpest of all the Hasselblad V lenses that I have.

The back is very easy to use and can be positioned either vertically or horizontally, without having to tilt the camera. That's nice because there's no RRS L bracket for V Hasselblads. The back senses which way it is turned and rotates the preview on the LCD screen accordingly.

I'll have a sample from the shoot shortly, all from the Mediterranean coast near Montpellier. Lots of seascapes and dunes and also canals with reflections.

Nicolas Claris said:

Hmmm, as usual, interesting Bart, I like that idea! how would you combine the files?

Click to expand...

I don't want to hijack this thread, so I'll start a new one in a couple of days, but I'm currently testing TuFuse (similar to Enfuse), in combination with PTassembler (a Windows based Panostitcher). The great thing about TuFuse is that it automagically blends exposure and focus differences. Enfuse can also be used for 'focus-stacking', with the right parameters, and a Mac version is available.

The only issue may be that one needs to align the images which, especially for close-ups, may require resizing the images so they align perfectly. For a Pano application that alignment/resizing shouldn't be too difficult. The pano application just needs to be able and hand-off the individually aligned images to TuFuse/Enfuse for automatic focus blending.

Bart

Asher Kelman said:

I think Bart is on to something here. With that size pixel at f32 the light, (considered as a wave front), is sort of scraped by the edges of the aperture causing ripples across the film plane for each point of resolvable detail in the image on the sensor.

As an aperture circle gets larger, the "edge effect" is less and less until pretty well most light passes unhindered. When the circle is smaller, the edge assumes greater importance diffracting more.

I think what I've said is correct and in any case, Bart will set us straight!

Click to expand...

That's essentially correct. I've just posted an example of the effect at different apertures here. The effect is very clearly shown with the lens I used there, but the principle is the same for other lenses. Too small an aperture (in an attempt to gain DOF), may reduce total image quality more (due to total loss of microdetail) than some OOF areas would do.

Bart

Hi Bart,

How would someone go about taking an image, such as Alain's, e.g. flowers outside, in the natural environment, to apply, say Tufuse?

I'm thinking about the time taken, wind, etc. Presumably the focus on most cameras would be set manually between each shot, and you would need a very solid setup, and wind breaks.

Best wishes,

Ray

Ray West said:

How would someone go about taking an image, such as Alain's, e.g. flowers outside, in the natural environment, to apply, say Tufuse?

I'm thinking about the time taken, wind, etc. Presumably the focus on most cameras would be set manually between each shot, and you would need a very solid setup, and wind breaks.

Click to expand...

Hi Ray,

Well that's the dilemma we face with these subjects. Whatever solution we choose, there is usually an implied downside. You're damned if you do stop down (too far), you're damned if you don't. It therefore requires a fine balance between gaining DOF and losing resolution.

Alain's image, because of the large number of pixels, requires less magnification to reach a certain output size. However, in an attempt to gain DOF, he lost microdetail and therefore would need to downsample to regain per pixel sharpness, or print smaller (use a higher PPI), which increases apparent DOF. IMHO, he lost more resolution than he gained from increased DOF, by stopping down too far. My feeling is that he might have gotten a better trade-off at f/11 (- f/16 maximum), where some of the diffraction can be countered by deconvolution sharpening.

In my focus blending/fusion experiments with TuFuse, which look very promising, I ran into an issue (a bug IMHO) which is now on the to-do list of Max Lyons (the author of TuFuse). I hope he fixes it soon, so I can share the results.

The DOF/aperture/diffraction dilemma remains, unless we can fall back on the multiple focus exposures, which may raise a new issue in the shape of subject motion. In my estimation the specific species of plants (they look like succulents) that Alain shot wouldn't pose too much of an issue in mild wind, while other plants/grasses would usually require a single exposure to avoid 'ghosts'.

To wrap up, maximum detail&DOF requires a fine balance between diffraction and sensel size, or stationary subjects and focus blending. Having more (quality) pixels, helps to preserve the inherent image quality at larger output sizes.

Bart

Hi Bart,

I just realised that it was a 5 second exposure anyway, so I would guess there was no wind for that particular period of time, and I expect it would be quite possible, for this location and other specific situations, as you say. Used to less calm air here ;-(

Best wishes,

Ray