Please clarify a flash statement I found on-line ...

[xpatUSA]

... from http://www.photocritic.org/articles/manual-flash-basics

"The power output on a flash is typically displayed as fractions. Full power is 1/1; the least powerful setting is usually 1/128.Don't forget that if you set your flash to full power, it'll be working at its hardest and will take longer to refresh in between shots. When you're experimenting with your flash, you might want to set the power to 1/16, which is roughly half-power."

Which, taken literally,is nonsense - you know me. So, typically for photography, "power" has two meanings!

I do know that technically power is Watts - or Joules per second.

I would just like to understand his 1/16 power versus 1/2 power ... Ta!

[dje]

... from http://www.photocritic.org/articles/manual-flash-basics

"The power output on a flash is typically displayed as fractions. Full power is 1/1; the least powerful setting is usually 1/128.Don't forget that if you set your flash to full power, it'll be working at its hardest and will take longer to refresh in between shots. When you're experimenting with your flash, you might want to set the power to 1/16, which is roughly half-power."

Which, taken literally,is nonsense - you know me. So, typically for photography, "power" has two meanings!

I do know that technically power is Watts - or Joules per second.

I would just like to understand his 1/16 power versus 1/2 power ... Ta!

Ted I see the so-called flash "power output" more as an energy measurement rather than a power measurement. ie the product of power and duration. In photometric terms, I'd probably say Luminous flux (lumens) instead of power (watts). So the flash "power" is lumen secs. This translates to exposure in the camera after directionalty, distance, aperture and scene reflectance is accounted for. Typically for most "power" levels, luminous flux output remains the same and duration is varied.

I have some sample waveforms if you are interested. I'm afraid I can't see where he gets his "half power" figure from.

Dave

[xpatUSA]

Ted I see the so-called flash "power output" more as an energy measurement rather than a power measurement. ie the product of power and duration. In photometric terms, I'd probably say Luminous flux (lumens) instead of power (watts). So the flash "power" is lumen secs.

That all makes sense, thanks, Dave.

This translates to exposure in the camera after directionality, distance, aperture and scene reflectance is accounted for. Typically for most "power" levels, luminous flux output remains the same and duration is varied.

I have some sample wave-forms if you are interested.

I would indeed, especially if they come with units for the Y-axis. Please PM.

I'm afraid I can't see where he gets his "half power" figure from.

A pity - that was my main question. Maybe I can can get off my butt and work backwards from published data, since we're talking about ratios rather than absolute quantities of Joules, lumens, whatever ...

[pschlute]

I'm afraid I can't see where he gets his "half power" figure from.

Dave

I think he is trying to say is that on a Flash that has output settings down to 1/128, the 1/16 setting is roughly the mid point between 1/1 output and weakest 1/128 output.

His description of this as "half power" is potentially confusing as 1/16 is obviously 1/16 output, not half.

[dje]

I think he is trying to say is that on a Flash that has output settings down to 1/128, the 1/16 setting is roughly the mid point between 1/1 output and weakest 1/128 output.

His description of this as "half power" is potentially confusing as 1/16 is obviously 1/16 output, not half.

Possibly, particularly if you look at it in terms of logarithmic (exposure) settings.

[pschlute]

Possibly, particularly if you look at it in terms of logarithmic (exposure) settings.

I think he is just saying it is in the middle of the 8 possible settings

[xpatUSA]

I think he is trying to say is that on a Flash that has output settings down to 1/128, the 1/16 setting is roughly the mid point between 1/1 output and weakest 1/128 output.

Ah ... "half-way", not "half-power". Kind of like thinking in stops.

His description of this as "half power" is potentially confusing as 1/16 is obviously 1/16 output, not half.

Agreed. Elsewhere, it says for example:

"The maximum power output of a particular speedlite is written as a 1/1. This just means full power. All other power settings are written as a fraction of full power. The next full setting down is 1/2 or half power. Half power equates to half the brightness of the maximum flash output and, in photography terms, is known as 1 stop less. The next full setting down is 1/4 power, then 1/8, then 1/16 power, then 1/32 power, then 1/64 power and finally 1/128 power. "

Which all makes perfect sense and agrees with your good self! Now it is clear ...

[Manfred M]

sloppy writing for sure, but there really is no consistency in how flash output is rated.

Small flash relative power is quoted as a guide number and studio flash ratings are in W-s (Watt - seconds). In studio flash we have no idea what this really means; is it the power used to charge up the capacitor(s) or is it the output power delivered to the flash tube terminals.

When I shoot at one studio, that uses Photogenic brand lights, when I turn it to minimum power output, I get a reading of 22. The maximum, if I recall correctly is 577. Great to use once one has some experience with them, but in isolation, not terribly useful.

My Paul C Buff Einstein lights gives me two different relative stop measurements, a W-s reading and an EU number. EU 1 = 12.5 W-s, EU 2.0 = 25 W-s, etc. and is meant to provide consistency across different brands of lights.

Nicely said, once one works with the lights, one can figure this all out, but it can be challenging when one gets used to it.

[MrB]

I know nothing about the way electronic flash works. However, I think I read somewhere that the flashgun will get hot when used frequently in one session. This is simply evidence of typical use case of any electrical device - some of the enegy used is always converted to heat in the circuitry. As the writer of the quote in post #1 mentions the flash refresh time, perhaps the "half-power" he is referring to is actually the amount of battery energy used in charging the flash ready for a shot at 1/16 light output, compared with that required in preparation for a full light output flash.

Philip

[William W]

Ah ... "half-way", not "half-power". Kind of like thinking in stops.

I'm late. You've already worked it out. They are 1 Stop intervals. Like ND Filters: ND2 ND4 ND8 ND16 ND32 ND64 ND128.

You can use GN (Guide Numbers) to describe the steps of the Flash 'Power' settings. I think some of the early Electronic Flash Units did that; I can't find a reference to hand, though.

I have never been keen on the word 'power' to describe the Manual Settings for Electronic Flash. Typically I place the word in single inverted commas, as here.

WW

[xpatUSA]

Thanks all for your input.

Dave Ellis sent me some wave-forms from a photo-transistor illuminated by a flash at various manual settings. They cleared up the matter nicely! Here's three of them 1/8, 1/1, 1/2:



Please note the remarkable difference between them and the square waveforms and exact fractions so favored by flash tutorials!!

Be that as it may, there is a rough correspondence with common setting values for a flash in manual mode. Observe that the height of each pulse is the same, telling us that the peak flux is NOT varied by the flash unit - consistent with always starting with the capacitor at about full charge. On the other hand, the period (width) of each pulse is varied - but not strictly in accordance with the standard 1-stop intervals.

Bearing in mind the shape of the "full" pulse (sorta exponential) - it seems that the timing is perhaps trying to halve the area under the curve with each decrement. The area under each curve is, after all, proportional to energy - Watt-sec (Joules) or lumen-sec - whichever.

Also please bear in mind that the hardware nature of a flash circuit: a thyristor, a big capacitor, a xenon-tube operating at micro-second speed and high voltage - is not indicative of the digital precision that we might naively assume it has! I think that Dave's flash does pretty good, considering ...

So, rounding off the thread, the statement in the OP:

"you might want to set the power to 1/16, which is roughly half-power." ... is hogwash.

[rpcrowe]

I don't think we need to get too technical in this discussion. I would expect that it was a simply poor choice of words on the part of the author... in that 1/16 power is "half power".

A better choice of words would be that 1/16 power is the "median" power in that there are as many selections with less power as there are selections with greater power.

LESS POWER = 1/128 - 1/64 - 1/32

MEDIAN POWER 1/16 not HALF POWER!

GREATER POWER = 1/8 - 1/4 - 1/1 (or full power)

The definition of "MEDIAN" is: denoting or relating to a value or quantity lying at the midpoint of a frequency distribution of observed values or quantities. In this case 1/16 power is at the median of power values of most flash units which are seven (7) values from 1/128 to full power!

[xpatUSA]

LESS POWER = 1/128 - 1/64 - 1/32
1/16 = median power
GREATER POWER = 1/8 - 1/4 - [?] - 1/1 (or full power)

I'm sorta kinda confused - "one-half" is missing from your list of greater values. Is that normal for flash settings?

Had it been included, the median value would actually be 3/32 ...

[pschlute]

Thanks all for your input.

Dave Ellis sent me some wave-forms from a photo-transistor illuminated by a flash at various manual settings. They cleared up the matter nicely! Here's three of them 1/8, 1/1, 1/2:

Interesting that the 1/1 full power setting has a completely different form. (long tail)

[pschlute]

So, rounding off the thread, the statement in the OP:

"you might want to set the power to 1/16, which is roughly half-power." ... is hogwash.

I think we had already established that and surmised that he did not exactly intend to make that statement.

Very poor choice of words in the article. If he is going to present an article on something as potentially complicated as flash photography, he needs to check his wording carefully.

[pschlute]

I'm sorta kinda confused - "one-half" is missing from your list of greater values. Is that normal for flash settings?

Had it been included, the median value would actually be 3/32 ...

The median of the series may not available as a selectable setting on the flash. The author does acknowledge this when he uses the word "roughly". But he does go on to complicate matters by using the phrase "half power".

I think I know what he means, but agree with you it is not a well phrased article, especially in the context of the subject matter.

[xpatUSA]

Interesting that the 1/1 full power setting has a completely different form. (long tail)

As you probably know, the discharge voltage of a capacitor is an exponential decay when just a fixed resistance is involved.

Like so - but read 'discharge' where it says 'charging'



CiC doesn't like the above GIF - gets better if you click on it.

On the other hand, the xenon only conducts over a certain voltage albeit with quite a low resistance. So, even with some resistance in the discharge path to protect the xenon, the form will not be a classic exponential.

If you'll look again at the one-half 'power' curve, you'll probably notice the beginning part is similar to that for full until the thyristor cuts off.

Lesser 'power' curves begin to show other effects which I won't even guess at.

HTH.

[xpatUSA]

The median of the series may not available as a selectable setting on the flash. The author does acknowledge this when he uses the word "roughly".

Yes, I can see that but I'm still wondering why Richard left "1/2" out of his list. All two of my flashes have a 'half-power' setting and both go down to 1/128. Therefore their median is certainly 3/32, of course not available in their actual settings.

My feeling is that Richard discovered that a "real" median value only occurs in an odd-numbered series. But, for an even-numbered series, such as 1/1 down to 1/128 in stops, the two middlemost ones have to be averaged; so, to avoid that awkward 3/32 value and to put the aforementioned 1/16 in the middle, he just left the "1/2" out, without explanation.

[pschlute]

As you probably know, the discharge voltage of a capacitor is an exponential decay when just a fixed resistance is involved.

Ted, you are way overestimating my knowledge on the subject. You lost me at "discharge voltage"

[xpatUSA]

Ted, you are way overestimating my knowledge on the subject. You lost me at "discharge voltage"

Perhaps think instead about house temperature. Nice and cozy with 72 degs inside and freezing 32 degs outside. Suddenly the boiler quits. Heat then flows out thru the walls and roof at a rate proportional to the difference between inside and outside temperature. As the inside temperature goes down that difference becomes less and less and less and so therefore does the rate of flow of heat. So we get a "discharge of heat" curve of inside-temperature shaped just like the one I posted earlier. There's a formula which I won't bother you with but it does have an interesting property. The inside and outside temperature will never become equal - heat will flow out for ever! - Well, until the last quantum of heat energy departs from the room ...

It's called an exponential decay which, oddly enough, can have a "half-life":

https://www.mathsisfun.com/algebra/e...al-growth.html

(scroll half-way down)

Funny how the word "half" keeps popping up in this thread ...