Film Flatness Test: 3 Film Holders in the First Quantitative Comparison
This article documents an attempt to quantitatively compare film flatness in three DSLR scanning film holders. The method is based on depth-from-focus analysis with a macro rail. The measurement accuracy is limited -- the results show trends and relative differences, not absolute values.
Why film flatness matters
Film flatness is the most common complaint in the DSLR scanning community. The film curves in the holder, and the corners lose sharpness. The physics behind it is known: At typical scan settings, the depth of field is only fractions of a millimeter. A curvature of 0.3 mm is enough to cause visible sharpness loss in the corners.
Many opinions, no measurements
In forums and on Reddit, there are many statements about film flatness -- "The corners are soft," "Holder X is better than Y." What you won’t find: measurement data. According to our research, no one has quantitatively determined and published the deviation of a film in a holder. All comparisons are based on subjective assessment of scan sharpness.
This article is an attempt to change that.
The test strip
The test strip used -- Kodak Gold 200, shot with an Olympus OM-2n and Zuiko 50mm f/1.8 at f/5.6. A relatively dense negative with evenly distributed image information.
The same film strip and frame were used for all measurements. The choice of subject is relevant: The Depth-from-Focus method needs texture in the image (film grain, edges) to determine the focus point. A uniformly bright or dark subject -- sky, wall, snow -- provides no usable signal. Branches and leaves work well: dense, high contrast, evenly distributed across the entire frame.
Method A: The Reflection Check (qualitative)
The film surface reflects light. When you hold the clamped film under a ceiling lamp, unevenness appears as distorted reflection lines. This method is simple, shows fine details -- but cannot be translated into numbers.
The same film strip in four holders, each photographed under the ceiling lamp:
Ausgeknipst
Negative Supply
Valoi 360
Reference: The same film held by hand only, without a holder
What the eye can see: None of the holders keeps the film perfectly flat -- the reflection lines are distorted in all three. But all three clearly do a better job than no holder at all. Which one performs best can only be guessed from the reflection images. The curvature cannot be quantified -- whether the deviation is 50 or 500 micrometers cannot be read from the reflection. Hence the second method.
How the three holders guide the film
The three tested holders use different design principles to keep the film in position. This affects where and how much the film can bow.
Negative Supply: The base has a panorama format gate -- the film is guided only on the two long sides (top and bottom). There is no guidance lengthwise. There are masks that can be inserted from below (for Half Frame, 35mm, or Panorama), but even with a mask, the film is free lengthwise. This means: In the middle of the gate, the film has the most room to sag.
Valoi: For each film format, there is a dedicated holder cast from a single piece. The film is guided both in width and length. The construction of the lengthwise guidance is not visible from the outside as it is integrated into the housing.
Ausgeknipst: A combination of both approaches. The base has a panorama format gate like Negative Supply and guides the film only on the sides. On top, interchangeable tops are placed that also guide and press the film lengthwise from above. Without a top, the holder behaves like the Negative Supply approach (side guidance only). With a top, lengthwise guidance is added.
This difference in guidance is relevant for the measurement: Since the gate width varies with each holder, the evaluation was limited to the inner 80% of the film frame. The edge areas contain little image information and are cut off during scanning anyway -- they are not included in the comparison.
Method B: Depth-from-Focus measurement (quantitative)
The camera moves along a macro rail in defined steps through the film plane. At each position, a picture is taken. Each region of the image reaches its maximum sharpness in a different frame -- depending on how far it is from the lens. From the position of the sharpness maximum, the height of the film surface at each point can be calculated.
Setup
Measurement setup: Camera on macro rail, film holder on the light source. The Post-Its under the light table serve as shims -- they compensate for minimal height differences at the corners so that the film plane is parallel to the sensor.
Close-up: Macro rail with dial gauge -- 0.1 mm step size
The details:
- Camera: Sony ZV-E10
- Lens: Carl Zeiss Jena Tessar 50mm f/2.8 on macro bellows
- Aperture: f/2.8 (wide open, for maximum sensitivity to focus shift)
- Macro Rail: 0.1 mm step size, 21 shots per pass
- Alignment: Mirror Method
Mirror alignment: The lens reflection must be exactly centered so that the optical axis is perpendicular to the film plane
Evaluation
A Python script performs the analysis. It divides each image into a grid and determines for each cell in which frame the sharpness is highest. From this, the height of the film surface at each point can be calculated. Two corrections follow: First, the global tilt is removed (the sensor is never perfectly parallel to the film). Second, the lens field curvature is removed so that only the pure film topography remains.
Three runs per holder, mixed and averaged.
Technical details on data processing
For the evaluation, the script divides each image into a grid of 20 x 30 cells. For each cell, the Laplacian variance is calculated over all 21 frames -- a measure of how much high-frequency contrast (film grain, edges) the cell contains. The cell is sharpest when the focus plane lies exactly on the film surface.
The resulting sharpness curve has a peak per cell. Its position is determined by 3-point parabolic interpolation to achieve a finer Z-resolution than the step size (0.1 mm). Then a best-fit plane is fitted to the entire Z-map and subtracted (tilt correction). When comparing multiple holders, the average of all Z-maps is subtracted as system bias (common-mode rejection) -- this removes the field curvature of the lens, which is identical in all measurements.
The evaluation is done on the inner 80% of the film frame. The edge areas are discarded -- they contain little to no image information and are cut off anyway during scanning.
The scripts are written in Python (numpy, opencv, matplotlib). Anyone who wants to check the code or recalculate the raw data can get in touch -- if there is justified criticism of the method, we will adjust the evaluation and publish the corrected results.
Limitations and caveats -- please read
This is not a scientifically accurate measurement.
The achievable measurement accuracy is about 100 micrometers. Typical film curl ranges from 80 to 500 µm. The measurement thus operates at the lower limit of its resolution.
The published values must not be read as absolute. They show trends and relative differences.
What was missing: A reference measurement with film clamped between two Newton glasses (perfect flatness = zero point). We would have needed that, but we didn’t have one on hand. Instead, we included a worst-case check: The film was guided only by the sprocket holes, without a cover plate. If our method works, this value must be significantly worse than with the proper holders.
Worst-case reference: film guided only at the edges, without cover plate -- maximum sag
Results
Three runs per holder, averaged. Measured on the inner 80% of the film frame (edge areas excluded, as the holders guide the film at different widths -- see above). Lens field curvature removed by common-mode rejection.
| Holder | PV (um) | RMS (um) |
|---|---|---|
| Ausgeknipst | 1102 | 163 |
| Valoi | 1382 | 175 |
| Negative Supply | 1708 | 202 |
| Sprocket (control) | 2309 | 381 |
PV = Peak-to-Valley: the largest deviation between the highest and lowest point.
RMS = Root Mean Square: the average deviation -- much more robust than PV because a single outlier does not distort the result.
Heatmaps
The heatmaps show the topography of the film plane. Red means: the film bulges toward the lens. Blue: it dips away. White is the ideal plane. The scale shows micrometers.
3-holder comparison: Red = film closer to the lens, Blue = further away
Validation: The sprocket holder (left) shows significantly more deviation than the holders with cover plate
And the individual heatmaps, each averaged over three runs:
Ausgeknipst -- averaged over 3 runs
Valoi -- averaged over 3 runs
Negative Supply -- averaged over 3 runs
What the data says
Validation: Does the method work?
The sprocket control (film without cover plate) shows 2.3x higher RMS values than the best holder. This is the most important data point of the entire measurement: it confirms that the method resolves real differences in film flatness and that the results are not lost in measurement noise.
Ausgeknipst (RMS 163 um)
The lowest RMS value in the test. The heatmap shows a relatively even distribution without dominant hotspots. The interchangeable top presses the film both on the sides and along the length, which is reflected in an even guidance. The run-to-run variation (how different the results are when reinserting the same film repeatedly) was 176 um -- comparable to Negative Supply.
Valoi (RMS 175 um)
Just behind Ausgeknipst. The heatmap shows slight wave patterns that could be due to the S-curve guidance of the Valoi channel design -- the film is guided through a curved channel when inserted. The run-to-run variation was slightly higher at 210 um compared to the other two holders. Whether this is due to the design or the way the film is threaded cannot be determined from the data.
Negative Supply (RMS 202 um)
The highest RMS value of the three holders. The heatmap shows more contrast than the other two -- areas with stronger curvature are more pronounced. Negative Supply only supports the film at the long edges, not along the length. The missing longitudinal support could explain why the deviation in the middle of the gate is somewhat higher. At the same time, the run-to-run variation at 175 microns was the lowest in the test -- the film sits consistently in the same place each time it is loaded.
The comparison
The factor between the best and worst holder is 1.2x (163 vs. 202 microns RMS). In absolute numbers: 39 microns difference. That is less than the thickness of a human hair.
For context: At f/8 -- the aperture most people scan at -- the depth of field at the negative is about 500 microns. All three holders keep the film within this tolerance. The 39-micron difference will not be visible in a finished scan at this aperture.
At wider apertures (f/4 or f/2.8, as found in high-end scanning setups), the depth of field shrinks to under 200 microns. In this range, the measured differences could become relevant -- but even then, the effect is hard to separate from other error sources (sensor alignment, lens field curvature, film curl of the specific film strip).
Conclusion
All three holders keep the film measurably flatter than an unsupported film strip. The differences between them are small -- the factor between the best and worst result is 1.2x.
The products differ in many other aspects (material, workflow, compatibility, price). Film flatness is just one factor. For this one factor, the three tested holders are close to each other.
Note on the setup
Measurements were taken at f/2.8 (wide open). This is not common for scanning -- image quality decreases at wide open aperture, especially in the corners. The reason for the wide aperture: The depth of field must be small enough to produce measurable differences at the film plane. At f/5.6 or f/8, the DoF would be too large to resolve film curl. A 100mm macro at 1:1 magnification and slightly stopped down would have been a better measuring instrument -- but was not available.
To the community: Help us measure better
This experiment was a first attempt with limited equipment. The method has weaknesses, which are documented above. If anyone in the community knows a more precise, affordable method -- laser interferometry, moiré topography, or something else -- we would appreciate the tip. The tests will be repeated, and the raw data published.
The goal is not a marketing comparison. The goal is to improve the design based on measurement data.