High speed synch versus stopping down

S

SamHH

Suspended / Banned
Messages
203
Edit My Images
No
Hi,

I understand that using a flashgun in high speed synch mode effectively reduces its maximum power output, as it has to spread its 'charge' across multiple flashes. How does this compare to stopping down so that the shutter speed falls within the maximum synch speed? This would also reduce the amount of flash light reaching the film/sensor, but by more or less than using high speed synch? Put another way, which of these would result in a brighter exposure:

- f5.6, 1/500, high speed synch mode, flash set to full power
- f8, 1/250, normal synch mode, flash set to full power

I'm assuming for present purposes that depth of field and other factors that vary with aperture don't matter. I'm just considering exposure.
 
•


The key ideas is ratio between two light sources
and and DOF, as everything else, everywhere else
in photography, a matter of compromises.

Here, the two effects (gain or loss) will be DoF and
light ratio between ambient light : flash light.

As in your example:
- f5.6, 1/500, high speed synch mode, flash set to full power
- f8, 1/250, normal synch mode, flash set to full power​
where the ƒ stop controls the ambient light and the
SS controls the flash.

If achieved correctly, the flashed part of the image
will be the same but the ambient will be darker at
ƒ8 and the DoF greater.
 
Hi Kodiak, thanks for replying. I'm not sure I understand why the ambient exposure would change in the two examples. The relevant variables, shutter speed and aperture, are precisely balanced, so why would there be any difference?

But anyway, my query wasn't really about ambient exposure, or about DOF. Maybe it's clearer to ask: for every stop of shutter speed above the maximum synch speed, does high speed synch reduce the maximum flash output by (a) one stop, (b) less than one stop, or (c) more than one stop?
 
The rule of thumb (and there's no precise measurement of this due to flashgun variability and actual shutter travel speed) you lose 1 stop of power as soon as HSS kicks in and a stop per halving of SS.

So for your example 1/500 would need f4.

However, when it comes to 'photography' @Kodiak Qc is right, it actually is all about the balance and just knowing the max power isn't much more than theoretically useful.

For a further technical tip though, you'll get a stop more flash if you use an ND instead of HSS, I carry a 3 stop for this.
 
SamHH said:
for every stop of shutter speed above the maximum synch speed, does high speed synch reduce the maximum flash output by (a) one stop, (b) less than one stop, or (c) more than one stop?
Click to expand...
•


These are waters I don't swim in as your question would
be more relevant when using a flash gun with low power
output. Since I work with multiple flash heads (B1: 500w/s
and B2 : 250 w/s) I don't have to worry about these.

Of course, changing the SS will affect the flash output as
it has to strobe more or less to match the SS but working
in TTL to control the visual balance quality of an obligatory
first test shot will tell you more then anything else.
 
Phil V said:
The rule of thumb (and there's no precise measurement of this due to flashgun variability and actual shutter travel speed) you lose 1 stop of power as soon as HSS kicks in and a stop per halving of SS.
Click to expand...

Great, thanks Phil, this is what I wanted to know.
 
Kodiak Qc said:
•


These are waters I don't swim in as your question would
be more relevant when using a flash gun with low power
output. Since I work with multiple flash heads (B1: 500w/s
and B2 : 250 w/s) I don't have to worry about these.

Of course, changing the SS will affect the flash output as
it has to strobe more or less to match the SS but working
in TTL to control the visual balance quality of an obligatory
first test shot will tell you more then anything else.
Click to expand...

I'm not sure what you mean by "visual balance quality" or first test shots. I was just asking about the technicalities of high speed synch, but I think Phil has answered that. Thanks for your replies though.
 
SamHH said:
Great, thanks Phil, this is what I wanted to know.
Click to expand...

I've heard the one-stop loss quoted a few times, and I don't think it's right. Rule of thumb, I would say a two-stops loss of effective brightness when switching to high-speed sync.

So I just checked it, with three flash guns at full power, and two cameras - a 5D2 (with quite a slow shutter cycle time) and 7D (relatively fast).

On the 5D2, moving from 1/200sec at normal sync to 1/250sec in HSS mode, dropped effective brightness of a Yongnuo 600EX-RT by 2.3 stops, 1.7 stops with a Canon 580EX, and 2.3 stops with a Canon 270EX. On the 7D moving from 1/250sec to 1/320sec, in the same order, the brightness losses were 2.3, 2.0, and 2.3. Thereafter, at faster speeds the brightness reduction was pro-rata with all cameras/guns.

Looking at the theory, in HSS mode the flash has to burn 3-4x longer to cover the whole shutter cycle time, corresponding to 1.5-2 stops less brightness from a single capacitor dump (very roughly) so it's hard to see how those figures could be improved by much.

SamHH said:
I'm not sure what you mean by "visual balance quality" or first test shots. I was just asking about the technicalities of high speed synch, but I think Phil has answered that. Thanks for your replies though.
Click to expand...

I'm guessing that Kodiak is referring to flash/ambient light balance, given that HSS is rarely used except in bright ambient light. There are two effects to note here: one is that if you're using HSS flash for a bit of fill-in, below the ambient light level, then just lifting the shadows takes a great deal less power than when you're trying to over-power the sun. Either way, the ambient light level can change the relative effect of the flash quite a lot.

The other thing is, while you lose an extra stop of flash brightness for every stop of increased shutter speed, raising the shutter speed also darkens the ambient light by the same amount so the flash/ambient ratio stays the same.

If you're not using HSS in bright ambient light but need a fast shutter speed to freeze movement, then you'll usually get better results using normal x-sync and turning the flash down a bit. This reduces the flash duration dramatically, and that becomes, in effect, your shutter speed. Eg, most speedlites will have a flash duration in the 1/5000-1/8000th region at 1/8th power, and with more brightness than in HSS mode.
 
Stopping down reduces both flash and ambient in equal measure. - similar to using ND filters

HSS effectively reduces the flash power without affecting ambient (It is the change in shutter speed that affects ambient and then causes further loss in flash power)

My tests with the Godox AD360 mk1 and Mk2 along with the SD600 lose at the power levels that Phil stated

The question has to be "what are you trying to achieve" and what are you trying to achieve it with?

Mike
 
mike weeks said:
It is with my Godox flashes, other brands may be different. The reason for this is that they have a long duration to start with at full power IMHO

Mike
Click to expand...

I've never been able to match your figures Mike, including with a Godox AD360. And what was also notable there was a fair amount of brightness fall-off up the frame at full power.

The long flash duration at FP has no bearing of HSS performance. What matters there is how much power is in the capacitor, and how long it has to strobe the tube for, while the shutter completes its cycle. I think you use Nikon though, and my limited experience there is that some models have faster running shutters than my Canons. That will make a difference if the gun has been tailored to exploit that advantage, but not by that much.
 
HoppyUK said:
I've heard the one-stop loss quoted a few times, and I don't think it's right. Rule of thumb, I would say a two-stops loss of effective brightness when switching to high-speed sync.

So I just checked it, with three flash guns at full power, and two cameras - a 5D2 (with quite a slow shutter cycle time) and 7D (relatively fast).

On the 5D2, moving from 1/200sec at normal sync to 1/250sec in HSS mode, dropped effective brightness of a Yongnuo 600EX-RT by 2.3 stops, 1.7 stops with a Canon 580EX, and 2.3 stops with a Canon 270EX. On the 7D moving from 1/250sec to 1/320sec, in the same order, the brightness losses were 2.3, 2.0, and 2.3. Thereafter, at faster speeds the brightness reduction was pro-rata with all cameras/guns.
Click to expand...

Thanks so much for testing this Richard. It is very helpful. By "pro-rata", I take it you mean one stop of reduction in brightness for every stop of shutter speed increase?

HoppyUK said:
Looking at the theory, in HSS mode the flash has to burn 3-4x longer to cover the whole shutter cycle time, corresponding to 1.5-2 stops less brightness from a single capacitor dump (very roughly) so it's hard to see how those figures could be improved by much.
Click to expand...

Ah, I didn't realise HSS involved a longer flash. I thought the flash pulsed. But either way, the consequence in terms of reduced maximum output is the same, I suppose.
 
SamHH said:
Ah, I didn't realise HSS involved a longer flash. I thought the flash pulsed. But either way, the consequence in terms of reduced maximum output is the same, I suppose.
Click to expand...
HSS is pulsed, Richard just went for shorthand. As far as the power is concerned, as you say the effect is the same.
 
mike weeks said:
The question has to be "what are you trying to achieve" and what are you trying to achieve it with?

Mike
Click to expand...

The situation I'm thinking of is using the flash to illuminate a shaded subject in otherwise bright ambient light. (I would call that fill-flash, but maybe I'm wrong.)

Say that the max synch speed is 1/250. And say that at f5.6:
  • you need a shutter speed of 1/1000 to achieve the desired brightness of the ambient-lit background;
  • you would, if you were not limited by the max synch speed, need a flash power setting of 1/4 to achieve the desired brightness of the subject.
What I thought, and I'm happy to be corrected, is that your choices are to either (a) stop down to f11 in order to get within the max synch speed, or (b) stay at f5.6 and use HSS. Because (a) reduces the amount of flash light reaching the sensor by two stops, in order to maintain the same subject brightness you need to increase the flash power by two stops, to 1/1. But if I understand Richard and Phil correctly, (b) effectively decreases the maximum output of the flash by more than two stops, meaning that even at a power setting of 1/1, the subject will be darker. Have I got that correct?
 
Last edited:
SamHH said:
The situation I'm thinking of is using the flash to illuminate a shaded subject in otherwise bright ambient light. (I would call that fill-flash, but maybe I'm wrong.)

Say that the max synch speed is 1/250. And say that at f5.6:
  • you need a shutter speed of 1/1000 to achieve the desired brightness of the ambient-lit background;
  • you would, if you were not limited by the max synch speed, need a flash power setting of 1/4 to achieve the desired brightness of the subject.
What I thought, and I'm happy to be corrected, is that your choices are to either (a) stop down to f11 in order to get within the max synch speed, or (b) stay at f5.6 and use HSS. Because (a) reduces the amount of flash light reaching the sensor by two stops, in order to maintain the same subject brightness you need to increase the flash power by two stops, to 1/1. But if I understand Richard and Phil correctly, (b) effectively decreases the maximum output of the flash by more than two stops, meaning that even at a power setting of 1/1, the subject will be darker. Have I got that correct?
Click to expand...

Re Phil's comment, I used the term 'strobed' rather than pulsed flash in HSS mode, where the gun fires a stream of reduced brightness flashes at incredible speed, like tens of thousands per second. The effect is like continuous light, though I have noticed, and others have too, that some of the bigger, higher powered IGBT-type units like the Godox AD360 appear to strobe at a slower rate and it's sometimes possible to see very fine darker stripes in some images. I need to take a closer look at that some time, but never noticed it with smaller hot-shoe speedlites.

In your example, it looks like you're on the cusp as to which method would be better. Well you're not actually, raising the f/number would give you more options with flash power to spare, but HSS is well worth a try. As I mentioned above, for fill-in flash you need a lot less power because, usually, the intention is to brighten the shadows that will be several stops darker than the highlights obviously. HSS has enough poke for that but bear in mind that the gun will be working hard and recycle times will be slow. The other method is to fit an ND filter (as Phil mentions above) that has the same effect as raising the f/number in dropping the shutter speed but doesn't effect depth of field. That's a popular technique when needing a low f/number, though two or three stops of ND is about the limit I find, before the viewfinder gets too dark.

In terms of flash output required, there's a world of difference between fill-in flash, and over-powering the sun. With fill-in, if you're a stop or so below ambient, that usually looks best. You don't want the flash to dominate. This puts it well within the scope of speedlites, especially as fill-in is one of the few occasions when on-camera on-axis flash looks good. This maximises the gun's efficiency and HSS retains all flash functions like auto metering, flash exposure compensation etc, and the camera will switch the gun in and out of HSS mode as and when required. Basically, it's dead easy :)

On the other hand, over-powering the sun needs the flash to be at least a stop brighter than ambient, preferably a bit more to get those darker moody skies in the background. So that's two stops more power required, minimum. Then when the flash is dominant, you need to control the shadows and make them attractive, so the gun has to come off-camera, maybe held up in your left hand, but shadows will still be harsh. So some softening is required, a bigger light source, ie a softbox of some sort and even the smallest and most efficient softbox will eat another stop or two of effective brightness. Add that all together and no speedlite really has a chance, hence the appeal of much more powerful products appearing, like the Godox AD360 that has 3-4x the power of a top-end speedlite, and the Godox AD600 with twice that, not to mention very high powered heads that can take two of those little monsters ganged together. Now you're talkin' :D
 
I got half way through typing a long post... - but what Richard said :agree:
 
SamHH said:
The situation I'm thinking of is using the flash to illuminate a shaded subject in otherwise bright ambient light. (I would call that fill-flash, but maybe I'm wrong.)

Say that the max synch speed is 1/250. And say that at f5.6:
  • you need a shutter speed of 1/1000 to achieve the desired brightness of the ambient-lit background;
  • you would, if you were not limited by the max synch speed, need a flash power setting of 1/4 to achieve the desired brightness of the subject.
What I thought, and I'm happy to be corrected, is that your choices are to either (a) stop down to f11 in order to get within the max synch speed, or (b) stay at f5.6 and use HSS. Because (a) reduces the amount of flash light reaching the sensor by two stops, in order to maintain the same subject brightness you need to increase the flash power by two stops, to 1/1. But if I understand Richard and Phil correctly, (b) effectively decreases the maximum output of the flash by more than two stops, meaning that even at a power setting of 1/1, the subject will be darker. Have I got that correct?
Click to expand...

Time is for me the critical factor in 2 aspects;

If you need to freeze action then higher shutter speeds and HSS come in to play

If there is no time to set this up and play then HSS is the answer

However if time is not critical you can do 1 of 2 things

A. Close down the Aperture

B. Use ND filters

in your example there I would be expecting approx 3 stops of loss, 1 for going to HSS and then 2 stops going from 1/250 to 1/1000

Mike
 
You must log in or register to reply here.
Back
Top