Motion Techniques and Terminology

When building a motion control rig for timelapse, gigapano, or other photographic motion control, it is important to first establish which type of motion technique will be appropriate for your project. You must know which technique(s) you wish to employ before beginning motor selection and overall system design.

While some of the techniques described here will be well-known and agreed upon by the time-lapse community and others, some are purely hypothetical and crafted to encourage experimentation and thought on the subject.  Please feel free to add any additional techniques, or discussion as a comment to this article.

Continuous Motion

The Continuous Motion technique implies that during the motion phase of a sequence, the motion is constantly happening.  This implies that the rig will be continuing to move in some fashion while the camera is exposing.  This is a technique commonly employed when using many of the available commercial telescope mounts. 

Motorized telescope mounts are primarily designed to compensate for the earth's movement when viewing objects through a telescope.  As a result of this design, those with adjustable speeds tend to provide speed setting relative to that which the earth rotates - such as 2x sidereal, 4x, etc.

There are two forms to this technique: continuous motion for a time period, and continuous motion for a spatial period.

This technique employed over a time period is considered one of the most simple to implement.  When we say over a time period we mean that the motion is controlled by how fast it runs and how long it runs.  Your time period could be a period of one hour, starting at some point before or after shooting begins, or it could be for the duration of the shot cycle.  The simplicity of this technique lies in that you only need to know three things: when to start motion, how fast to move, and how long to run for.  As the only control to the motion aspect here is how fast to run the motors, it requires a bare minimum of components, which can usually be sourced easily and assembled by even the greenest of novices.

The spatial period form of this technique differs from the time period in that instead of knowing only when to start and stop, you must also know where to start and stop.  For example, a spatial period could be defined as a movement from 0 degrees to 45 degrees on the pan axis.  The previously mentioned factors of when to start and how fast to move still are applied.  This form offers greater complexity than the time period form due to the additional requirements of recording and translating position of either the motor shaft or final output shaft of your rig to your controlling device. 

Some time-lapse shooters using dSLRs prefer this technique as it is believed that the small amount of blur introduced during exposures while the motors are moving add a more film-like feel to the final output video.  As both forms of this technique require no knowledge of when the camera is firing, this technique can easily be employed into an existing time-lapse workflow, without need to integrate intervalometer and motion control functions.

Pros:

• Simple to implement
• Requires only a few inexpensive parts
• No need for integrated intervalometer and motor driving functions
• Excellent range of functionality applicable to most time-lapse shooting

Cons:

• May not be suitable for telephoto or macro shooting due to blurring during shot
• Requires very solid rig to prevent vibrations through camera while exposing
• Motor speed controls available limit shot intervals and lengths of timelapse videos
• Not generally suitable for still re-combination, like gigapano shots

 Shoot-Move-Shoot

The name of this technique, shoot-move-shoot, implies that movements only occur between exposures.  That is to say, an exposure is taken, and after completion of the exposure, any motor movements are then executed.  This technique is commonly employed in stepper-motor based rigs, where fine-grained movements can be executed in a controlled manner.

The most important aspect of this technique is that the moco rig must be fully aware of exposure time, whether its being managed by the intervalometer or the camera, and be able to assure that the motors are not moving while the camera is exposing.  Of course, in some systems this feature is over-looked, without it the primary benefit of lack of any blur cannot be promised.

There are three primary reasons why this technique could be employed: first, for critical-focus shots movement while exposing is prevented.  Secondly, for movement during long-term timelapses, where the rate of movement is far slower than a continuous motion rig has been designed for (such as 1/100th degree of movement per day using a system designed for moving 50 degrees in one hour of shooting).  The third primary reason for employing this technique is the achieve complex, non-linear, speed-ramps (such as S-Curve) in the output video that could not be achieved using a continuous motion rig.

The number of steps between shots (s) is calculated as the spatial distance per step (d), the total distance to be moved (D), the seconds of video time to make the move over (T), and the frame-rate per second (f) of the output video.  Or: s = (D / d)/(T / f).

For time-lapse, the camera will often be placed off-nodal-center, such that a natural looking pan (similar to how a video camera would be operated by hand) or tilt can be achieved.  For gigapano shots, the camera will often be placed on-nodal-center, allowing for non-distorted horizons when stitching multiple images together. 

Some time-lapse shooters prefer this technique as it gives them most of the control benefits of the continuous motion technique while also being much more flexible on length, speed of motion, and focal control.  This technique is inherently more difficult, requring complex circuitry and more expensive hardware to understand travel rate, current position versus expected position, and complete integration with the intervalometer to understand when the camera is exposing.

Pros:

• Ideal for close-focus and telephoto shots
• Practically requisite for gigapano shots
• Much more control over movement rates relative to real time and video time
• Better ramping controls

Cons:

• More expensive to implement
• Requires more complex electronics and components
• Can look too crisp (aka 'digital') in final output video for some applications
• Requires greater setup and prep time to define movements

Move-and-Shoot (and then maybe move some more...)

This technique is a form of the shoot-move-shoot technique that attempts to re-create the slightly blurred nature of the continuous-motion technique, while gaining all of the advantages (and increased cost) of the shoot-move-shoot technique.

Effectively, this technique requires one to know the distance to move while taking an exposure to create the desired level of blur, and the distance to move between exposures to create the desired output speed of movement in video time. 

To calculate the desired level of blur on a given axis, one must first take into account the field-of-view of the camera, the number of pixels available in the sensor, and the minimum movement amount of the rig.  To use a stepper model: we would calculate how many steps have to be taken while the camera is exposing to create the desired blur and then reduce the number of steps taken between shots by this amount. 

Pros:

• All of those inherited from Shoot-move-Shoot
• A more natural film-like blur during movements
• More fine-grained control of blur, regardless of video time motion speed

Cons:

• All of those inherited from Shoot-move-Shoot
• Even more complex driving logic