New HyperSync tests: PocketWizard v3.700 for Nikon

studio460 · November 14, 2014

I wanted to create a new thread for the sake of clarity since the original thread is now missing many of its image links (I had inadvertently deleted the server-side images and did not have a back-up). Re-posted from the former thread ["Which monolights HyperSync to 1/8,000th with a Nikon D3s + Mini TT1/Flex TT5?"], Ian quoted the following statement from my website:

Why HyperSync? What are the benefits to increased shutter speeds while still being able to use your strobe? While most view HyperSync as a technique to “overpower the sun,” that’s only one of three <a id="itxthook5" href="/photography-lighting-equipment-techniques-forum/00YrJV?start=100" rel="nofollow">applications<img id="itxthook5icon" src="http://images.intellitxt.com/ast/adTypes/icon1.png" alt="" /></a> where HyperSync can be useful: 1. Increasing flash-to-ambient ratio to darken daylight exterior backgrounds. 2. Enabling large apertures when using strobes in daylight, without using an ND filter. 3. Employing high shutter speeds when using strobes in daylight to freeze motion.

</blockquote>

Ian said:

I would question the validity of 1... 'Increasing flash-to-ambient ratio to darken daylight exterior backgrounds.' The physics of hypersync is to allow a <a id="itxthook6" href="/photography-lighting-equipment-techniques-forum/00YrJV?start=100" rel="nofollow">faster<img id="itxthook6icon" src="http://images.intellitxt.com/ast/adTypes/icon1.png" alt="" /></a> shutter speed to be used with the SAME flash output. The SAME flash output (operating as 'ambient') is being reduced by shutter speed too. So where does this 'Increasing flash-to-ambient ratio' happen? Not with Speedlights, not with long duration flashes (except at max shutter speed - and that marginally) and not with short duration flash without banding or severe shutter curtain <a id="itxthook7" href="/photography-lighting-equipment-techniques-forum/00YrJV?start=100" rel="nofollow">intrusion<img id="itxthook7icon" src="http://images.intellitxt.com/ast/adTypes/icon1.png" alt="" /></a>.

</blockquote>

Strictly speaking, you're correct, Ian. I tried to qualify these statements throughout the article, indicating the shortcomings of the technique, while still trying to communicate that HyperSync does offer tangible benefits. Note that in most instances, I state that HyperSync permits increased recorded flash output (not increased flash output, per se).

studio460 · November 14, 2014

Here's a direct link to some HyperSync test images using a Quantum Qflash:

<a href="http://lightbasics.com/hypersync-test-series-1-nikon-d800e-quantum-qflash/">http://lightbasics.com/hypersync-test-series-1-nikon-d800e-quantum-qflash/</a>

rodeo_joe1 · November 15, 2014

I'm not sure what those "HyperSync" test images are supposed to prove or show, apart from the expected fading of flash power over time. It's a bit pointless showing every shutter speed from 1/500th to 1/8000th, since the shutter transit time remains constant at about 1/320th of a second with the camera used. So it's no surprise that the fading from top to bottom of the frame (inverted due to image inversion in the camera) remains almost identical across those higher shutter speeds.

Ian's point about flash power to ambient ratio is very valid. That leaves only the motion-stopping ability and flexibility of aperture choice as advantages of using Auto-FP, HyperSync, HSS or whatever name you care to call this technique. Once the shutter starts being used to alter flash exposure, then any differentiation between flash and ambient exposure is completely ruled out.

In Focal-Plane synch usage we have to get ourselves out of the X-synch mindset of "Flash is controlled by aperture - ambient is controlled by shutter speed" because that paradigm no longer applies.

Incidentally, I suspect that the fading would be much more noticeable with a camera having a longer shutter transit time. I believe that all of Nikon's Pro and Prosumer full-frame cameras have a fairly rapid transit time of < 1/320th second, whereas many other cameras have a slower time of around 1/250th second, and a pathetic ~ 1/180th second in the case of Canon's 6D. That extra millisecond or so might make a world of difference to the falloff in flash brightness. (Sorry Canon users, but you're at a distinct disadvantage in this area.)

studio460 · November 15, 2014

Rodeo said:

I'm not sure what those "HyperSync" test images are supposed to prove or show, apart from the expected fading of flash power over time. It's a bit pointless showing every shutter speed from 1/500th to 1/8000th, since the shutter transit time remains constant at about 1/320th of a second with the camera used.

</blockquote>

What these test images show is that you do not lose a significant amount of recorded flash output using this technique. The point in showing a range of shutter speeds is to both show any differences in recorded flash output using equivalent exposure values, and to determine the usable range of the technique.

ian_casement · November 16, 2014

The shutter speed effect is pertinent at the higher end range where the two shutters are following in very close proximity and the DURATION of the shutter ''open' time is the shortest. This means that more of the peak output can be captured.. and in some flash duration cases a greater amount of the flash duration within the frame at x-sync.

Incidentally, I suspect that the fading would be much more noticeable with a camera having a longer shutter transit time. I believe that all of Nikon's Pro and Prosumer full-frame cameras have a fairly rapid transit time of < 1/320th second,

</blockquote>

Interesting point for two reasons; As already pointed out the shortest DURATION is at maximum shutter speed AND that the best performers in your Hypersync tests have flash duration t.5 times of around 1/300s.

It makes total sense that an 'ambient like' flash needs to be sustained over the DURATION of the shutter time which is why also, shutter speeds the like of 1/320s will be worse (graduation wise) because they are open longer and include greater flash deterioration than faster shutter speeds of 1/8000s.

The importance of the step change shutter speed test would be to estimate at what point the graduations would be acceptable and indeed illustrate that there is a difference between shutter speeds.

rodeo_joe1 · November 18, 2014

"What these test images show is that you do not lose a significant amount of recorded flash output using this technique."

</blockquote>

Sorry Ralph, but those images exactly show that flash power is lost with increasing shutter speed. Otherwise the images would get brighter as the shutter speed was increased. As it is, the flash exposure behaves precisely the same as does a continuous source; i.e. the aperture needed changes by one stop for every halving of shutter time. In other words reciprocity is maintained. Just as it would be for "ambient" or continuous lighting. Therefore there is no real gain in ambient to flash ratio and no major advantage in using HyperSync for that purpose.

"It makes total sense that an 'ambient like' flash needs to be sustained over the DURATION of the shutter time which is why also, shutter speeds the like of 1/320s will be worse (graduation wise) because they are open longer and include greater flash deterioration than faster shutter speeds of 1/8000s."

</blockquote>

Ian, this statement shows a complete lack of understanding of how a Focal-Plane shutter and FP flash synch works. The shutter transit time - how long the shutter takes to travel across the frame - stays constant at speeds higher than X-Synch, regardless of the exposure time set. Therefore the flash fall off is shown over that transit time, not just for 1/1000th or 1/8000th of a second or whatever.

If you need proof of that, all you have to do is look at the test images linked to above. They clearly show that the flash fading remains constant at all shutter speeds once FP synch (HSS or HyperSync) is engaged. The fading can only be got rid of by having the shutter open for longer than the flash duration, not shorter. In other words by using X-synch and a shutter speed longer than around 1/500th second, or whatever the t~0.5 duration of the flash happens to be.

I'd disagree with the conclusion that the fading is a trivial drawback, since it can clearly be seen that it isn't.

studio460 · November 21, 2014

Rodeo said:

As it is, the flash exposure behaves precisely the same as does a continuous source; i.e. the aperture needed changes by one stop for every halving of shutter time. In other words reciprocity is maintained. Just as it would be for "ambient" or continuous lighting.

</blockquote>

Which is exactly what you don't get when not employing HyperSync, where virtually anything shot above x-sync is completely unusable. Here are the results you get without employing HyperSync:

studio460 · November 21, 2014

I'd prefer to make this HyperSynced exposure, over an all-black frame any day:

[HyperSync tests for the strobe used in creating the image above: http://lightbasics.com/hypersync-test-series-2-nikon-d800e-speedotron-force-10/ ]

ian_casement · November 21, 2014

I wrote:

"It makes total sense that an 'ambient like' flash needs to be sustained over the DURATION of the shutter time which is why also, shutter speeds the like of 1/320s will be worse (graduation wise) because they are open longer and include greater flash deterioration than faster shutter speeds of 1/8000s."

Rodeo Joe said:

Ian, this statement shows a complete lack of understanding of how a Focal-Plane shutter and FP flash synch works. The shutter transit time - how long the shutter takes to travel across the frame - stays constant at speeds higher than X-Synch, regardless of the exposure time set. Therefore the flash fall off is shown over that transit time, not just for 1/1000th or 1/8000th of a second or whatever.

</blockquote>

Not in my book it doesn't and FACTUALLY it isn't either.

I made a specific reference to DURATION in my text and nothing about shutter transit time. Shutter transit time is a constant, DURATION is a variable and is dependent on the distance between the shutters.

When two shutters transit a sensor with identical transit times and the distance between them reduces due to the shutter speed set, (i.e. at 1/8000s the distance - or gap - being smaller) then the DURATION of the shutter exposure reduces. THUS at 1/320s the DURATION of exposure is greater than at 1/8000s and therefore needs a longer sustained light level to achieve the same result as could be obtained at 1/8000s. The same would apply to your example of 1/1000s and 1/8000s since the travelling slit size at 1/1000s is greater than that at 1/8000s and would require a longer DURATION.

It is not I who "shows a complete lack of understanding of how a Focal-Plane shutter and FP flash synch works." It is you.

rodeo_joe1 · November 22, 2014

"THUS at 1/320s the DURATION of exposure is greater than at 1/8000s and therefore needs a longer sustained light level to achieve the same result as could be obtained at 1/8000s."

</blockquote>

Sorry Ian, but you still don't seem to be understanding how a focal plane shutter works. When the shutter becomes a slit, the effective exposure of the sensor or film is reduced, but the shutter still takes 1/320th or 1/250th (or whatever) of a second to travel across the entire frame. If the exposure DURATION really was limited to 1/8000th of a second, then you'd only see an area of the frame the size of the slit exposed. In fact Ralph's images above are a perfect example of an how the slit exposes the frame.

" The same would apply to your example of 1/1000s and 1/8000s since the travelling slit size at 1/1000s is greater than that at 1/8000s and would require a longer DURATION."

</blockquote>

No, it categorically wouldn't. With Focal-Plane synch, in order to expose the entire frame the flash duration has to be equal to or greater than the complete transit time of the shutter. That's regardless of the "speed" set on the camera. Also there's a fundamental difference in the timing of X and FP synch. X synch fires the flash as soon as the first blind has reached the end of its travel, whereas FP synch fires the flash as soon as the first blind begins its travel.

Ralph, the issue is not about whether Hypersynch allows higher shutter speeds to be used. I never denied that. What's arguable here is whether the feature is able to be used to substantially vary the ratio of flash power to ambient. And plainly it isn't.

"HyperSync can in fact deliver higher-than x-sync shutter speeds, allowing you to knock down ambient light levels to a considerable degree." - It's this statement that's being questioned. Since as soon as you raise the shutter speed in HyperSync you also have to open the aperture to compensate. Ambient exposure therefore stays the same.

What aperture was used in your flash Hypersynch example at 1/8000th Ralph? Because I see no reason why the same shot couldn't have been taken at X-synch shutter speed by maybe altering the ISO, the aperture, or using an ND filter to get exactly the same effect. Probably more to the point is why you'd want to point the camera at the sun like that in the first place. A much better use of that backlighting would be to hide the sun behind the subject's hair so that it formed a nice "halo" effect. Which would have been more pleasing in my view. As it is, you've just got a slight kicker effect on the right, and blue flare spots on the left-hand side of the picture. Or you could have just used a mirror or reflector to direct the sunlight itself back to the subject's face. No need for flash at all in that case.

ian_casement · November 22, 2014

No need to be sorry Rodeo Joe, you're still wrong.

Sorry Ian, but you still don't seem to be understanding how a focal plane shutter works. When the shutter becomes a slit, the effective exposure of the sensor or film is reduced, but the shutter still takes 1/320th or 1/250th (or whatever) of a second to travel across the entire frame.

</blockquote>

When the gap reduces between the first and second shutter SO TOO does the exposure DURATION.

DURATION is measured between the point the first curtain STARTS to open to the time the second curtain CLOSES.

Different durations are necessary - because - otherwise every exposure shutter speed set would be the same. Patently they're not.

Rodeo Joe wrote:

If the exposure DURATION really was limited to 1/8000th of a second, then you'd only see an area of the frame the size of the slit exposed. In fact Ralph's images above are a perfect example of an how the slit exposes the frame.

</blockquote>

Pardon me.. I never said anything of the sort. At 1/8000s the shutter DURATION is closer to the maximum x-sync speed of the camera (with the shutters travelling in close proximity) than when a 1/320s shutter speed is set (on a 1/250s x-sync camera) because the shutters are then furthest apart and force the DURATION to be longer.

E.G.

At 1/250s shutter speed the first shutter transits the entire sensor in 1/320s (your figure), then after the sensor is fully exposed the second curtain follows and takes another 1/320s (your figure) to transit the sensor.. total DURATION 1/320s + 1/320s = 2/320s = 1/180s (long duration)

At 1/8000s shutter speed the first shutter transits the entire sensor in 1/320s (your figure), then very, very, closely after (say 1/8000s) the second curtain takes 1/320s (your figure) to transit the sensor.. the sensor is never fully exposed.. total DURATION 1/320s + very, very, closely (+1/8000s) = near to 1/320s (short duration) !

Hence you will see that the DURATION of a shutter speed set to 1/8000s will take less DURATION than any other shutter speed set.

Why is this interesting in this thread? Because using a long duration flash, the flash DURATION will be closer to the shutter DURATION than any longer shutter speed (of the kind.. 1/250s - 1/4000s).

See what Ralph quoted as the longest t5 times... 1/300s...

Ian wrote:

Interesting point for two reasons; As already pointed out the shortest DURATION is at maximum shutter speed AND that the best performers in your Hypersync tests have flash duration t.5 times of around 1/300s.

</blockquote>

E.G. NOT at 1/180s (This is almost twice the DURATION and will receive a greater variance of flash output between bright and dim).

I hope that helps explain to you.

studio460 · November 23, 2014

Rodeo said:

'HyperSync can in fact deliver higher-than x-sync shutter speeds, allowing you to knock down ambient light levels to a considerable degree.' - It's this statement that's being questioned. Since as soon as you raise the shutter speed in HyperSync you also have to open the aperture to compensate. Ambient exposure therefore stays the same.

</blockquote>

The testing goal of the gray-card image series was to show a range of equivalent exposures to reveal any delta in recorded flash exposure as shutter speed was increased. The application goal is enable higher shutter speeds (holding aperture constant) to reduce ambient exposure. Rodeo said:

Probably more to the point is why you'd want to point the camera at the sun like that in the first place. A much better use of that backlighting would be to hide the sun behind the subject's hair so that it formed a nice 'halo' effect.

</blockquote>

The photo was only a test and had no intended "artistic" objective. The purpose of the photo was to frame the brightest part of the sky (which includes the sun) to demonstrate extreme daylight conditions. The objective was to compare the recorded output of a 1,000Ws strobe at a shutter speed of 1/8,000th. At that time of day, this part of the sky appeared pure-white to the naked eye.

ian_casement · November 23, 2014

Rodeo Joe wrote:

Ralph, the issue is not about whether Hypersynch allows higher shutter speeds to be used. I never denied that. What's arguable here is whether the feature is able to be used to substantially vary the ratio of flash power to ambient. And plainly it isn't.

</blockquote>

I agree that 'doing it' is one thing, but getting some advantage from it is another. The 'getting some advantage from it' is the hardest part since as described previously the use of ND filters could adequately be used to attain some specific wider aperture whilst retaining flash benefit ratio, but Hypersync use can provide higher shutter speeds - but not always retain ratio. Ideally the higher shutter speeds should be provided along with retained ratio, but alas that isn't always the case.

Even better, would be higher shutter speeds and increased flash benefit ratio but those cases are rare and don't need 'Hypersync' to be attained. I believe Ralph's research has this objective in mind with searching for a lighting/trigger/camera combination effecting the best result is whats sought.

rodeo_joe1 · November 24, 2014

Ian, the transit time and exposure time (DURATION as you insist on calling it) at the maximum X-synch speed are almost one and the same. At X-synch speed the second blind begins to close almost as soon at the first blind has fully opened. The frame is effectively fully exposed for only the transit time of the shutters, not twice that time. Otherwise the X-synch speed wouldn't be 1/320th of a second would it? OK, maybe the shutter blinds travel a little faster than the maximum X-synch speed would indicate, but not by much margin. Any shading or loss of exposure due to the time the blinds take to travel across the frame is taken into account by the marked shutter speed. So the marked 1/320th second (or whatever the camera's maximum X-synch speed may be) probably actually takes a little longer than that. I took that figure at face value to illustrate my point, not as an accurate measurement of the actual transit time.

At 1/640th of a second the second blind begins it's travel when the first blind is only halfway across the frame. But it still takes approximately 1/320th of a second for both blinds to have crossed the frame. At this point the shutter effectively becomes a travelling slit, and the time taken for the whole frame to be exposed remains constant. You can plainly see this is the case in Ralph's examples at different X shutter speeds. Interpolating between the 1/500th and 1/1000th shots; at 1/640th the shutter shadow would be almost exactly halfway across the frame.

The same happens all the way up to the maximum shutter speed. At 1/8000th of a second the blinds take exactly the same time to transit the frame as at 1/500th or 1/1000th. The effective exposure time is pretty much irrelevant, since the flash has to output a pretty much constant power over that transit time. That's the point I was making. That a camera with a longer X-synch speed (and hence transit time) will show more fall off of flash power over time. In other words the t~0.5 duration of the flash needs to be longer with those "slow" cameras. And not many flashes have such a long flash duration.

Ralph. I still don't see how it's possible to keep the same aperture while varying the shutter speed, and at the same time keep the flash exposure constant while the ambient exposure reduces. Say your X-synch exposure is 1/250th @ f/8. Moving to HyperSync and using a shutter speed of 1/500th requires a change in aperture to f/5.6 to keep the flash exposure constant. That change in aperture also doubles the brightness of any ambient light at the camera sensor - and so on up to 1/8000th and f/1.4. So no net gain in flash power, nor loss of ambient light.

Yes, HyperSync or FP-synch allows you to use a bigger aperture, but the ratio of ambient to flash remains very much the same. In fact it's less flexible than X-synch because the flash has to remain on full power and can't be turned down by shortening its duration.

studio460 · November 25, 2014

Rodeo said:

Ralph. I still don't see how it's possible to keep the same aperture while varying the shutter speed, and at the same time keep the flash exposure constant while the ambient exposure reduces . . .

</blockquote>

Here's an example: If shot at 1/250th (i.e., x-sync) instead, the background would've been brighter at same aperture.

The gray card tests shown here: http://lightbasics.com/hypersync-test-series-2-nikon-d800e-speedotron-force-10/ reveal that there is not a significant difference (other than slightly uneven exposure, which goes unnoticed in this image) in overall recorded flash output for this particular strobe when comparing shutter speeds of 1/250th (x-sync), and a range of higher shutter speeds (where 1/640th is a fairly moderate shutter speed).

ian_casement · November 25, 2014

the transit time and exposure time (DURATION as you insist on calling it)...

At X-synch speed the second blind begins to close almost as soon at the first blind has fully opened. The frame is effectively fully exposed for only the transit time of the shutters, not twice that time. Otherwise the X-synch speed wouldn't be 1/320th of a second would it?

</blockquote>

DURATION is not exposure time.

It takes 1/320s for the first shutter curtain to traverse the sensor to fully expose it, it then takes 1/320s for the second curtain to traverse the sensor and complete the exposure. Thats 1/320s + 1/320s = 2/320s = 1/160s.

1/160s is the DURATION of the exposure. 1/320s is the EXPOSURE time.

I already stated that.

For a 1/8000s shutter speed, the first shutter takes 1/320s to traverse the sensor with the second curtain travelling behind it at a fixed interval of 1/8000s. You then get a DURATION of 1/320s + 1/8000s = ALMOST HALF THE DURATION OF THE PREVIOUS EXAMPLE.

When you're discussing how to obtain the most from a cash with a fixed t.5 time then you need to match your shutter duration to the flash duration as closely as possible for 1) Greater Output 2) less banding. The faster shutter speed does this because its DURATION is shorter and more closely matches the t.5 times discussed.

What do you not understand?

rodeo_joe1 · November 28, 2014

"The faster shutter speed does this because its DURATION is shorter and more closely matches the t.5 times discussed."

</blockquote>

But plainly it doesn't. As you can see from Ralph's example HyperSynch tests, the light falloff from bottom to top of the images is almost identical for all speeds from 1/500th to 1/8000th. In fact the falloff appears greater at the shortest shutter speed. Explain how that gels with your changing DURATION theory Ian!

"If shot at 1/250th (i.e., x-sync) instead, the background would've been brighter at same aperture."

</blockquote>

That really isn't the point Ralph, and nor is it true. If you'd used the X-synch speed of 1/320th, then you'd have had to stop the aperture down to f/6.3 to keep the same flash exposure, and the background exposure would therefore be identical. OTOH if you'd kept the same aperture the flash exposure would have been brighter as well.

What I'm saying is simply that the ratio between flash and background illumination stays the same at all HyperSync shutter speeds, as well as with the maximum X-synch speed. The fact that you have to alter the aperture to compensate for a shutter change proves that. If you could keep the same aperture or change the flash power, then it'd be a different story.

studio460 · November 28, 2014

Rodeo said:

That really isn't the point Ralph, and nor is it true. If you'd used the X-synch speed of 1/320th, then you'd have had to stop the aperture down to f/6.3 to keep the same flash exposure, and the background exposure would therefore be identical. OTOH if you'd kept the same aperture the flash exposure would have been brighter as well.

</blockquote>

That's simply not true. Changing shutter speed doesn't affect flash exposure, it changes ambient exposure. Can you post any photos which support your claims?

ian_casement · November 29, 2014

I notice a few things in Ralph's examples:

Using HSS/FP mode enabled the 'normal' inversion of graduation between max x-sync flash sync and beyond x-sync flash sync is in existence.

The highest output provided by the beyond x-sync example is greater.

I also notice that when HSS/FP mode is enabled that the 'normal' inversion of graduation between max x-sync flash sync and beyond x-sync flash sync is in existence.

And again, that the highest output provided by the beyond x-sync example is greater.

The common factor (besides the shared inversion of graduation) is increase in exposure.

Since these examples are using an optimised system which optimises for each individual shutter speed - then you won't see very much difference between each shutter speed (maybe an advantage to using Pocket Wizard ControlTL system) but using the regular and existing camera HSS/FP mode retimed shutter sync then you will see these differences I pointed out as I have already tested, tried, compared and wrote about, which I illustrated back in 2011. http://www.commercialphotographer.co.uk/blog/?p=274

Sign In

New HyperSync tests: PocketWizard v3.700 for Nikon

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in