With the latest firmware updates the Phottix Odin TTL triggers for Canon and Nikon now have a great new feature called Over Drive Sync, or ODS.
ODS is a fast and simple, but also very precise, timing adjustment, allowing you to tune in the best results possible when using higher shutter speeds up to 1/8000th with suitable studio lights and monolights. Adjustments are made quickly and easily directly on the Odins regular transmitter interface.
ODS relies on flash units with long flash durations, so results will depend on the flash duration of the lights used (which also changes at different power levels), as well as the camera used. Full frame cameras with large sensors and slow shutters like the Canon 5D series are unfortunately the hardest to get good results with, but lights with long enough flash durations can still achieve clean frames all the way to 1/8000th.
Provided you have updated to the latest firmware, accessing the new ODS timing adjustment is just a simple matter of holding down the “MODE” and “+” buttons at the same time for 2 seconds. Little did we know the ODS feature was even already enabled in the Nikon version since the last firmware update back in April !
The ODS screen will then be displayed, with the default 0.0 timing setting flashing. Ready to be adjusted up and down with the “+” and “-” buttons.
There are 50 graduations from 0.0 to 5.0. Each 0.1 is one millisecond. So adjustment range is simply from zero to 50 milliseconds.
Its just a matter then of connecting a studio light to an Odin receiver via the supplied sync cord, setting the Odin TCU to HSS, and the camera to a shutter speed over X-sync (1/1000th etc), and taking a test shot.
You can then adjust the timing delay to see if that produces a brighter exposure, or removes shutter clipping (black bands) from the image. Its simply a matter of a nudging the brightest area into the frame (or shutter clipping out of the frame) as it suits best.
Below is an example using a manual speedlight at half power. The first frame at standard timing produced nothing but a black image. So without the new ODS timing adjustment available there would have been no result at all in this example. Adjusting the timing setting just a little to 0.3 some light already starts to appear at the bottom of the second frame. The third frame was around 0.8, and the fourth frame 1.8 (or 18 milliseconds delay).
The example above was an extreme example where it was not possible to avoid some clipping of the frame with the shutter curtains. This is because the flash duration of a speedlight at 1/2 power was simply much too short to light the entire frame. At full power the results would have been a lot better. But this does help illustrate clearly how you can at least make use of the most light available by simply nudging that light into the frame as it suits best.
Even when the standard timing produces a clean frame without any shutter clipping, ODS timing can still provide significant advantages in the exposure level, again by using the timing adjustment to nudge the brighter area of the flash duration into the frame. Here I used the 300Ws CL-360 Cheetah Light at full power (where its flash duration is longest) providing a full stop of extra light in the frame. An extra stop is equal to having a second ($800) flash unit firing as well. Gains may not always be this great (again depending on camera and lights used etc), but they can be quite substantial.
How Does ODS Work
ODS relies on flash units with long flash durations, and also what is known as an early Pre-Sync firing signal. The early signal fires the flash before the camera shutter has started to open, and the long slow burning tail of light (or long duration) acts like a constant light source, lighting the full image while the shutters are passing over. Hopefully the animated images below will help to illustrate this process more clearly.
Most TTL capable radio triggers do also provide this Pre-Sync firing signal, but the timing is fixed in most cases, and results can then be much more limited without any adjustment possible.
Results will depend on the flash duration of the lights used (which also changes at different power levels), as well as the camera used. Full frame cameras with large sensors and slow shutters like the Canon 5D series are unfortunately the hardest to get good results with, but lights with long enough flash durations can still achieve clean frames all the way to 1/8000th.
Please note – I’m not an electronics engineer, below is just a basic graphic illustration of how the process operates. The shutter curtains actually travel downwards on the camera, not upwards as shown bellow. I wanted to show how their motion affects the gradient and clipping seen in the resulting images captured, so the shutters are moving in the right direction in relation to the upright images (as the image is normally upside down in the camera).
Shutter Speed At or Below X-Sync
Firstly we need to see how a DSLR cameras Focal Plane shutter functions, disregarding flash for now.
The camera has two shutter curtains that cover the sensor. At or below the X-sync speed of the camera (1/250th etc), the first curtain opens, completely exposing the sensors entire surface area for a certain period, then the second curtain follows behind covering the sensor again. The fastest shutter speed the camera can fully open the shutter and expose the entire image at once is the cameras X-sync limit.
Shutter Speed Above X-Sync
At shutter speeds above the cameras X-sync limit, the first curtain starts opening, and the second curtain follows behind, creating a slit that moves across the sensor. So the sensor is never all entirely exposed at once.
At higher shutter speeds towards 1/8000th the slit moving across the sensor becomes narrower.
So the speed the shutter curtains are traveling is not actually changing at higher shutter speeds. They still take close to the original 1/250th etc X-sync speed to move across the sensor.
Shutter Speed At or Below X-Sync with Flash
As shown above, at or below X-sync the first shutter curtain opens completely exposing the entire sensor for a certain period. The flash waits for the first curtain to completely open, then fires as soon as the complete sensor is exposed. So the entire frame is lit at once with a regular flash duration. And then the second curtain then covers the senor again.
FP HSS – High Speed Sync
So as seen above, at shutter speeds above X-sync there is only a narrow slit of the senor exposed at any one time. So firing a flash with a relatively fast duration would only light one narrow slit or band across the image.
But Canon and Nikon etc speedlites which have an FP HSS (High Speed Sync) feature are able to work around this problem by firing many small flash pulses as the slit moves across the sensor. This provides an even exposure across the frame, as the pulses are firing so fast (50Hz) they appear as a constant light source to the camera. Because the flash has to fire many times, that has to be at lower power output than just firing once.
Now the speedlite FP HSS process shown directly above is where the early Pre-Sync fire signal used in ODS originates from.
As shown further above, at or below X-sync the camera waits a while for the first shutter to completely open before firing the flash. But with the FP HSS process the flash needs to start firing before the first shutter even starts opening. Which is well before the regular X-sync fire signal.
So this early fire signal is the Pre-Sync signal used in both FP HSS, and ODS Long Duration style Sync.
Long Duration Sync – ODS Over Drive Sync
Now most studio lights do not have the ability to fire many small pulses of light at high frequency like lower powered speedlites do. So instead, to cover the entire frame at higher shutter speeds, ODS simply relies on a long flash duration to act as a constant light source for the short time it takes the slit to more across the camera sensor.
A flash pulse from a studio strobe may look very fast to the human eye, but its actually often a relatively long process, consisting of a bright initial burst, followed by a long slow exponential decline, or long tail of light. (Think of a firework that goes off with a bright burst and then slowly dissipates). So flash duration curves will often appear as shown below, a short peak followed by a long tail of declining light.
Flash durations are indicated by t.1 and t.5 times.
- t.5 is the time it takes for 50% of the flash power to dissipate
- t.1 is the time it takes for 90% of the flash power to dissipate
t.1 is the relevant time here, as that includes the tail of light which can be used to help light the frame.
ODS and Long Duration Sync then use the early Pre-Sync signal discussed above, to fire the flash just before the first shutter starts opening. So the peak of the flash pulse goes off just before the shutter starts opening. The long tail of light then lights the rest of the frame as the slit moves across.
As before, as the shutter speed increases towards 1/8000th, the slit travelling across the sensor narrows, and the exposure level in the image will be lower as less light is able to reach the sensor through the smaller opening.
Also as the light is declining towards the top of the frame, this is where you may see a lower exposure or gradient in the image. The longer the duration of the flash, the more this curve may appear flatter, and more even across the frame.
As mentioned previously, the speed the shutter curtains travel across the sensor does not actually increase at higher camera shutter speeds. So even at 1/8000th it still takes around the cameras X-sync speed for the shutter curtains to cross the sensor (depending on cameras sensor size and shutter speed etc). So this is where we get an indication of the required length of flash duration, or the t.1 time required to light the entire frame without shutter clipping the frame (black bands showing at the top and/or bottom).
There are basically 2 types of flash units, Variable Voltage, and IGBT, control of flash power. The majority of studio monolights are Variable Voltage, while most lower power speedlites are IGBT. Though more higher power IGBT studio lights are a now becoming available as well (like the PCB Einstein).
Variable Voltage – control the flash power by varying the voltage to which the flash capacitors are charged.
IGBT – fill the capacitors, but control the flash power by abruptly shutting the flash tube off once the desired amount of light has been emitted.
At full power both flash types have a similar power curve as seen previously with long tail.
But at half power (and lower) they differ dramatically. Variable Voltage retain a similar slowly declining curve at all power levels, and durations actually become longer at lower power levels, as seen here.
IGBT are very different as the flash power is abruptly cut short at anything below full power, causing much shorter flash durations as shown below.
So at 1/2 power, what was a nice long 1/200th t.1 at full power, is now reduced to a t.1 of approximately 1/600th. Which is likely too short to cover the complete frame, and we would see results like the original examples shown with an IGBT speedlite at 1/2 power.
So IGBT flashes are not generally ideal for ODS or Long Duration Sync at any power level below full power. And if like the Einstein they have very fast duration even at full power (t.1 – 1/540), they may not be able to achieve a frame completely free of some shutter clipping at all at higher shutter speeds.
But don’t rule out IGBT flashes completely, because they can potentially offer the best of both worlds, long duration at full power for use with ODS/Long Duration Sync, and fast durations at lower power levels, which can be used for freezing motion when there is no ambient light in the image (in studio etc). Many of the inexpensive Chinese brand IGBT lights still have longer flash duration at full power.
Below is a side by side comparison with 3 of the most common TTL triggers which provide a fixed Pre-Sync signal, to see what advantage the ODS timing adjustment could offer. Results and potential gains are always going to vary depending on the lights and camera etc used. Here I’m using a Canon 50D which is a 1.6x crop sensor camera, and a Cheetah CL-360 300Ws IGBT flash unit, which still has a suitable t.1 flash duration at full power.
Even with the standard timing the YongNuo YN-622C and Pixel King/Pro provide a completely clean frame without any shutter clipping here. The frames shown below are at 1/1000th shutter speed, but even at 1/8000th there was still no clipping (and coverage was more even at higher shutter speeds).
As seen above the YN-622C and Pixel King/Pro where virtually the same result. The Godox Cells II was about 1/3rd of a stop brighter. And the Odin with an ODS setting of 1.2 managed approximately another 2/3rds of a stop of light in the frame. So that’s a full stop advantage over the YN-622C and King/Pro, which is equivalent to having a second flash unit worth of power on hand!
So the simple ODS timing adjustment can be a very valuable tool!
Adjusting ODS Timing
With the fast timing adjustment on the Odin TCU interface you could certainly adjust a light on location if needed. But if using regular known lights and camera, ideally it would be best to do some testing at home or in a studio, noting down some timing settings which could then be quickly set on the Odin TCU when needed.
Shutter clipping (or black bands in the image) are always going to be fairly easy to see (provided there is no ambient light in the image). But if you want to see more subtle differences in exposure and gradient in the frames, the trick is to achieve a fairly evenly lit frame to start with below X-sync. The way I do this is simply to bounce the flash off the ceiling, and shoot the more even light bounced onto a wall (as shown in the test set up image above).
Ideally start at a shutter speed of 1/125th, ISO set to minimum (ISO 100 etc), and with the flash turned off make sure there is no ambient light in the image (a black frame). Then turn the flash on (likely to full power) and adjust the camera aperture until the spike in the histogram is close to the right side. A nice narrow spike in the histogram shows you have a fairly even frame then to start with, as shown below. You know then any gradient seen in subsequent images at high shutter speeds is purely due to the flash duration curve. That’s easily hidden or harder to gauge if you don’t start with an even frame (which is a good thing as it means it often won’t be too noticeable in your regular images).
Then set the camera to a desired shutter speed over X-sync (1/1000th) etc, and take a test shot. Asses the image for shutter clipping etc, and adjust the the ODS timing setting and see if that makes any improvement.
If the default 0.0 produces a black frame, its likely best to start with coarse adjustments of 1.0 (10ms) at a time to see if that makes any difference, and then fine tune from there.
The ideal timing setting will change from shutter speeds close to X-sync, to higher shutter speeds like 1/8000th. So you may like to note a few good settings across the shutter speed range.
If you were in the field with an unfamiliar light, and you just wanted to quickly adjust any shutter clipping out of the frame (if possible), you could simply shoot a grey card or similar, holding the card fairly close to the light so its blowing out the exposure and frame completely. Blowing out the frame would still clearly show any shutter clipping in the image, so you could then try to adjust that out.
The main thing is that you don’t want any ambient light showing in the test shot, otherwise you can’t easily see how much of the frame the flash is actually lighting. So again start with camera settings that provides as close to a black frame as you can get with the light off, then only light the card with the flash.
Its not always important to have flash lighting the complete frame though either. If you know your subject is only going to be in part of the frame (like this for example) shutter clipping in part of the frame may not always be an issue. So it may be preferable to move the most light available into the are of the frame where the subject is, even if some shutter clipping occurs elsewhere in the frame.
What is ODS or Long Duration Sync Useful For?
Often people unfamiliar with using high sync speeds in this way, assume the high shutter speeds are going to give a big advantage in underexposing the ambient background and sky etc. Unfortunately the results there are actually generally similar to the other option available which is placing ND filters over the camera lense to reduce light levels. With both methods ambient and flash light is reduced equally. So raw flash power is really the main thing that is going to allow you to underexpose the ambient or sky etc.
Compared to using ND filters, Long Duration Sync methods like ODS have their positives and negatives, but ND filters do already offer an option for underexposing ambient, or using wide apertures for narrow depth of field in portraits etc in bright ambient light.
The difference with using ND filters, is that the camera stays at X-sync, which is usually relatively low 1/200th – 1/250th shutter speeds. So the thing ND filters can not offer is motion freezing shutter speeds (up to 1/8000th) in bright ambient light. You can use flash with fast duration to freeze motion with no ambient light in the image, (regardless of using low shutter speeds), but that doesn’t work in bright ambient light, as that continuous ambient light will allow the motion to blur. So the fast shutter speeds are needed to freeze the motion in images also containing bright ambient light.
So ODS and Long Duration Sync methods of higher shutter speeds can offer an alternative to ND filters, but there main “unique” advantage is for freezing action in bright ambient light. Joe McNally often shoots images like this. That example is actually using speedlights with FP HSS, but as explained above FP HSS has the same effect of becoming a continuous light source as ODS and Long Duration Sync methods. A relatively small 300Ws studio light using ODS could provide more light than all those Nikon speedights in HSS (Canon speedlights fair better in HSS than Nikon).
Dave Black also has lots of great inspiration for freezing action in bright ambient light using high shutter speeds –
People will often argue that High Speed Sync etc, does not freeze motion, because the camera sensor is still exposed in parts for 1/250th while the slit between the shutter curtains passes over. That is true to some degree, and the issue can be distortion of the image more than motion blur. But the reality is HSS acts like a constant light source, so motion will be frozen to exactly the same degree as high shutter speeds freeze motion in ambient light without any flash. So the subjects in the sports images above for example will still be frozen much sharper at 1/1000th +, than at very slow X-sync shutter speeds like 1/200th.
Flash units with long flash durations are going to provide the best results with ODS. Ironically lower priced flash units often have longer flash durations providing a wider range of good results, as higher end lights are often actually trying to engineer out these long durations.
So a good indication of which lights are going to give reasonable results can be found by looking at their flash durations. Most manufacturers specs are given in t.5 times, but rough guide to the t.1 time can be found by multiplying the t.5 by 2 to 3 times. So a 1/600th t.5 would likely be around the 1/300th – 1/200th t.1. The closer the t.1 times get to around the X-sync speed of the camera the better the results should be.
The PCB Einstein for example has a maximum t.1 time of just 1/540th. While the AB1600 with a maximum t.1 of 1/300th is likely going to provide better results. The White Lightning X3200 should be better still as the durations getting even longer towards 1/150th at minimum power. Outside the US , the Jinbei Discovery is often quoted as a portable flash option which provides good results with Long Duration Sync methods like ODS.
IGBT Flashes – As mentioned previously above, IGBT flashes should not be ruled out as many often still have a long enough durations if used at full power. There are many variations of the popular compact portable IGBT strobes like Phottix’s own PPL-400, some of which still have durations as long as 1/125th at full power. These can potentially offer the best of both worlds, long duration at full power for use with ODS/Long Duration Sync, and fast durations at lower power levels, which can be used for freezing motion when there is no ambient light in the image (in studio etc).
Speedlites – (also IGBT) often have long enough durations at full power as well. But firing speedlites at full power all the time can quickly overheat them, and power loss is similar to regular FP HSS (which is often fairly limited with small speedlites). So manual speedlites can be used with Long Duration Sync methods like ODS, but much higher powered studio lights are generally much more capable.
Which flash units exactly are going to give the best result with your particular camera etc, is otherwise the million dollar question. This is something that can only be answered from people experimenting and sharing their results and experience.
Chris Valites from PocketWizard has also put together an excellent set of test results with a range of cameras and lights HERE under the heading “What HyperSync results can I get with my camera and flash?” Those results are using PocketWizard’s HyperSync timing adjustment, so results may differ with ODS, but they should at least give an indication of the difference in results to be expected relative to one another between the lights and cameras tested there.
But please DO NOT contact Phottix asking which lights work, or why your light does not give the results you expected !!
Phottix have been sitting on this ODS timing feature for quite a while, with a well founded concern releasing this could send their customer service system into meltdown with a barrage of questions like this. ODS was released as a beta function for people to experiment with for themselves and make the most of the results they can. For experienced users this is a highly valuable feature, so it would be shame to loose it in future firmware. Phottix value any feedback which can help to improve any products or features including ODS, but please don’t tie up their customer service with ODS results and compatibility questions !!
So have fun experimenting, and feel free to share your results here.
Lastly, don’t forget to set the ODS back to 0.0 when you’re finished though (or hold the clear button for 2 seconds), otherwise the timing change may effect your speedlites regular FP HSS results.
Full Phottix Odin Review HERE
Odin Price and Availability –
Prices range from around $320 for the Tx & Rx set.
See also the Phottix Mitros + flash unit, with Odin radio transceiver unit built-in.