What is the ISO flash Guide Number methodology?

Getting an answer to this question is on my bucket list; but I'm darned if I'll pay the ISO 38 Swiss Francs to find out!

 

So has anyone handed over that extortionate amount, or has their employer, to buy ISO 1230:2007 'Determination of flash guide numbers for electronic flash equipment'?

 

I ask because there appears to be little relationship between makers' Guide Number claims and the amount of light actually output by any electronic flash. Particularly speedlights.

 

IME, the ISO 'standard' might as well say "Think of a vaguely believable number, and then double it." Or perhaps the methodology doesn't exclude the use of a narrow and white-painted (or mirrored!) corridor for the testing environment?

 

Whatever. I'm pretty sure the ISO haven't specified their methodology sufficiently rigidly to prevent a mighty big fudge factor being applied. But how would we know, when the tight-fisted ISO don't make their 'standards' readily open to user review?

It seems that the ISO determines to make money. I heard that if you paid 38 Swiss Francs for it you can read it for yourself but can't tell me. I too keep wondering about that and the ISO standard for ISO.

Some countries require their standards to be freely accessible. I sometimes download mechanical standards that are carbon copies of the ISO ones, but unfortunately the flash standards doesn't seem to be available. Maybe because nobody cares! You can certainly work it out yourself using ISO 100 and a 10' distance. In years past I found manufacturers guide numbers to be pretty accurate, because if they weren't, the little calc dial on the back of the flash would give the wrong exposure. Watch out for reciprocity failure with very short flash times. Hmmm... do image sensors suffer from that?

You can certainly work it out yourself using ISO 100 and a 10' distance. In years past I found manufacturers guide numbers to be pretty accurate, because if they weren't, the little calc dial on the back of the flash would give the wrong exposure.

A flashmeter, or simply chimping a correct exposure on a digital camera, would beg to differ with the 'calc dial' (how old are your flashes Conrad?) and makers lying published guide numbers. That's why I query the ISO's methodology.

 

Reciprocity? Only with film that leaks photons. No such nonsense with digital sensors.

I too keep wondering about that and the ISO standard for ISO.

 

For film or for digital?

The procedure for film speed determination is widely published, but not so for digital sensors. As far as I can tell, the digital sensor speed methodology is a lot less rigid than for film. It basically sets the base ISO at that exposure which gives 'best image quality', with some skant guidance as to how 'best quality' should be assessed.

 

I think it's pretty scandalous that the ISO has basically become a consortium of big-business interests, with said companies being the consultation bodies and effectively setting their own standards. Then veiling the details from public scrutiny via non-disclosure clauses, and obviously monitoring the internet for any unauthorised publication.

 

For example: Try finding something as simple as a dimensioned engineering drawing of an ISO hotshoe online. You'd think it would be easy, but no. All traces have apparently been meticulously eradicated. No wonder the cheeky sods have to charge 38 Swiss Francs. It must cost that to police their stupid secrecy!:mad:

So has anyone handed over that extortionate amount, or has their employer, to buy ISO 1230:2007 'Determination of flash guide numbers for electronic flash equipment'?

 

Not from my end, that I recall.

 

I HAVE seen older ANSI (American National Standards) though. Circa 1970s, the outfit where I worked decided to take a major step forward in its QC program, starting with copies of all the pertinent ANSI photo standards.

 

As some of these standards went out of date, I saved some of the copies, maybe several dozen, for posterity, or something like that. Looks like I have a couple of pertinent titles, "... Determining Exposure Guide Numbers for Photographic Lamps" and "... Method for Measuring and Designating the Performance of Photographic Electronic Flash Equipment." Book of these were current in the 1970s.

 

Or perhaps the methodology doesn't exclude the use of a narrow and white-painted (or mirrored!) corridor for the testing environment?

 

The ANSI method does exclude these.

 

The second standard I listed states in its appendix, "Due precautions should be taken to see to it that reflections from specular or light-colored surfaces in the area where measurements are made will not affect the readings." It further discusses use of flat black baffles such that "... all of the light produced is absorbed except that which travels directly from the source to the receptor."

 

As note, in my limited experience ANSI and ISO standards of the same time frame tend to be identical, or nearly so. When one is able to see a list of committee members, the ANSI team is generally also on the ISO committee.

 

Also, as a caution to would-be purchasers of standards, they are not necessarily stand-alone. There may 2 or 3 interconnected standards, such that the user needs more than one to deal anything useful.

How old are my flashes? Well, let's say I'd never use one on a modern camera without checking the trigger voltage.

 

IMO, any standard that carries the weight of law, and many do, should be available for free. Why should you have to pay to find out what laws you have to obey? All the CE mark requirements are like that. Fortunately you can't get into much trouble with guide numbers or hot shoes.

A flashmeter, or simply chimping a correct exposure on a digital camera, would beg to differ with the 'calc dial' (how old are your flashes Conrad?) and makers lying published guide numbers. That's why I query the ISO's methodology.

 

Reciprocity? Only with film that leaks photons. No such nonsense with digital sensors.

 

 

For film or for digital?

The procedure for film speed determination is widely published, but not so for digital sensors. As far as I can tell, the digital sensor speed methodology is a lot less rigid than for film. It basically sets the base ISO at that exposure which gives 'best image quality', with some skant guidance as to how 'best quality' should be assessed.

 

I think it's pretty scandalous that the ISO has basically become a consortium of big-business interests, with said companies being the consultation bodies and effectively setting their own standards. Then veiling the details from public scrutiny via non-disclosure clauses, and obviously monitoring the internet for any unauthorised publication.

 

For example: Try finding something as simple as a dimensioned engineering drawing of an ISO hotshoe online. You'd think it would be easy, but no. All traces have apparently been meticulously eradicated. No wonder the cheeky sods have to charge 38 Swiss Francs. It must cost that to police their stupid secrecy!:mad:

 

I would like to have the ISO standard for digital. That guy Northrup said ISO is fake and I do not think so. I think it's not fake but the standard is so slack that each manufacturer can pretty much determine their sensor ISO their own way. As for the GN I read a document somewhere from Metz that they said they used a 50% reflectance target for determining the GN.

The ANSI method does exclude these.

 

The second standard I listed states in its appendix, "Due precautions should be taken to see to it that reflections from specular or light-colored surfaces in the area where measurements are made will not affect the readings." It further discusses use of flat black baffles such that "... all of the light produced is absorbed except that which travels directly from the source to the receptor."

Then that definitely begs the question - 'where does it all go wrong?' - and how can speedlight makers continue to exaggerate their guide numbers by at least one whole stop? Compared to any flashmeter reading you care to take, or to a correctly exposed image got by empirical means.

 

The exaggeration (hyperbole or lying, let's call it by its real name) goes way back. I bought my first decent 'pro quality' electronic flash back in 1969 or thereabouts, and immediately noticed that exposures were under by about one whole stop when using the given guide number. I've never trusted published guide numbers since, and never had cause to revise my suspicion of them.

 

So if a 'standard' doesn't actually work in practise, or is patently ignored; what good is it? And I suspect that flash guide numbers aren't the only ISO 'specification' that manufacturers manage to run rings around.

Then that definitely begs the question - 'where does it all go wrong?' - and how can speedlight makers continue to exaggerate their guide numbers by at least one whole stop? Compared to any flashmeter reading you care to take, or to a correctly exposed image got by empirical means.

 

The exaggeration (hyperbole or lying, let's call it by its real name) goes way back. I bought my first decent 'pro quality' electronic flash back in 1969 or thereabouts, and immediately noticed that exposures were under by about one whole stop when using the given guide number. I've never trusted published guide numbers since, and never had cause to revise my suspicion of them.

 

So if a 'standard' doesn't actually work in practise, or is patently ignored; what good is it? And I suspect that flash guide numbers aren't the only ISO 'specification' that manufacturers manage to run rings around.

I've had similar experiences with battery powered flashes, but both sets of White Lightning studio strobes that I have owned going back 30 years now have put out the power that they claimed that they did. I have never depended on ISO determined guide numbers for these since I bought a flash meter when I bought the studio strobes, but the guide numbers published by this manufacturer correspond well to flash meter readings and finicky color transparency film.

Two points . . . Don't evaluate the accuracy of the guide number with a flash meter and the flash set to any automatic mode. I'm saying that anyone did this, I'm just saying not to. But, the second point is pretty clear to me as an engineer: Don't rely on experiments (like determining guide numbers) done in labs to match what you see in the real world.

 

Having said that, remember when you check the guide number of a flash, that the manufacture includes the zoom setting and usually provides the guide number at the tightest setting for obvious reasons.

Don't rely on experiments (like determining guide numbers) done in labs to match what you see in the real world.

Well, if measured properly, in a totally black-painted room, to only read the light directly incident, then guide numbers should tend to overexpose in the real world. But they don't.

remember when you check the guide number of a flash, that the manufacture includes the zoom setting and usually provides the guide number at the tightest setting for obvious reasons.

That seems to be another big fudge by flash makers, since the tightest 'zoom' setting GN is usually even more exaggerated than those given at mid coverage settings.

 

The makers appear to use a theoretical multiplication factor based on the calculated area of light coverage. This obviously doesn't hold true in reality. I now see ridiculous claims of GN 60 (metres/100 ISO) for a little hotshoe mounted speedlight. There is no way on this earth that such a claim can be substantiated by metering or actual exposure.

 

All I want to know is: Where in the ISO methodology is the 'getout clause', or what are the 'weasel words' that let flash makers get away with their obvious inflation of a true guide number?

 

As things stand, the unsuspecting get a less powerful flash than they bargained for, and the rest of us have to do a bit of mental arithmetic to divide published GNs by (at least)1.4 in order to find the real exposure - should we find ourselves reliant on a GN calculation.

but both sets of White Lightning studio strobes that I have owned going back 30 years now have put out the power that they claimed that they did.

Yes, my rant about guide numbers mainly applies to hotshoe or bracket-mounted flashes - hammerheads included. Proper studio flashes tend to be more accurately specified. Probably because their makers know that the first thing their equipment's new owner is going to do is show it a flashmeter.

I've not seen that one but I do work with a few ISO procedures in non photographic realms. The procedures are generally fairly comprehensive on obvious matters such as the reflective surfaces here, but can still have vague sections, that sometimes effectively allow manufacturers to quote generous values.

Unlike ASTM standards (some of which are altered several time a year) ISO ones are very hard to change they are reviewed on a 5 year cycle, but the review process is very conservative. National bodies can add comments to the standards but have to publish the actual text unchanged.

 

FWIW Knowing the amount of work that goes into producing & testing these methods I wouldn't consider 38 Swiss Francs to be much for a standard like this. They are not intended as documents for general interest to geeky users (the summary on the ISO website can be enough for this) The methods are designed for manufacturers who probably end up buying several books of related standards for their field every year. IIRC the annual budget for standard test methods where I work is about £2000 and that with us getting members discounts.

 

One extra factor in your case could well be that as they age flash units may give less powerful light output than when they were new.

One extra factor in your case could well be that as they age flash units may give less powerful light output than when they were new.

Could be, but it isn't.

 

Even new from the box units suffer from guide number inflation - or deflation, depending how you look at it.

 

And I have flash units many years old that output the exact same amount of light as their current replacement model.

Sooo. Nobody knows.

 

I'm beginning to wonder if anyone, least of all any maker of speedlights, has ever bothered to read the ISO standard.

 

One would hope that any standard worthy of the name would specify its methodology tighter than a gnat's orifice, and deny any wriggle room for varying results. Therefore it beggars belief that flash makers rigidly follow an ISO standard, only to arrive at a figure that's at complete variance with a simple flashmeter reading. And certainly a figure that can't be relied on to give a practically useable exposure.

 

38 Swiss Francs for a document that appears to be meaningless? Is this what the world's supposed standards body is peddling these days? Time to bring back the BSI, DIN, ANSI, JASC, etc. if that's the case, and sack those Swiss bean counters.

I don't think that you really understand how any of this works: You expect that a laboratory standard is going to be directly relatable to real world situations. That's not what it's supposed to do.

 

The idea of a standard is that it create reliable, repeatable, results that allow comparison of products (or samples). If one flash has a published guide number or 60 and another of 120 you can assume that they will reliably produce one stop of light difference under most situations. That doesn't mean that YOUR experiment (if you don't purchase and read and apply the standard) will result in the same number.

 

Most of the time that guide numbers were used by photographers was when photographers where shooting film. On print film, getting the laboratory results wasn't that important since the negative film has a few stops of leeway in either direction. Photographers who shot slide film knew that they needed to work out their exposures and, for important shots, they bracketed. This has always been something that photographers have needed to work out on the own to get truly accurate exposures.

 

This is also why Auto and TTL flash were developed . . .

 

Don't even get me started on the accuracy of the indicated f-stops on most camera lenses.

You expect that a laboratory standard is going to be directly relatable to real world situations.

Yes. Otherwise it's useless.

It would be like specifying the pitch, angle and profile of a screw thread for a bolt, and a different pitch, angle and profile for a matching nut.

 

The 1963(?) ASA/BS/DIN co-operative standard for film speed worked in the real world. Why shouldn't the specification for measuring flash output?

 

Lab standards, especially industrial standards, are all about real world applications. Otherwise they're only of theoretical interest, and nobody is going to pay for one - unless they're really desperate for toilet paper!

If one flash has a published guide number or 60 and another of 120 you can assume that they will reliably produce one stop of light difference under most situations.

No. A doubling of Guide Number would indicate a fourfold increase in light output, and a two stop difference.

Edited August 10, 2019 by rodeo_joe|1

Joe . . . On point #1, you're wrong. Lab standards are designed to allow for repeatable results . . . IN A LAB SETTING! That's why the standard defines the entire set up and test procedure. It doesn't do ANYTHING to ensure that the same measurements will ever be met in the real world.

 

On point #2, you're correct . . . My mistake . . .

Joe . . . On point #1, you're wrong. Lab standards are designed to allow for repeatable results . . . IN A LAB SETTING!

Oh, OK then. We should just ignore the standards laid down for the Metre, Kilogram, Litre, Second, Ampere, Volt, Ohm, Inch etc. etc? Since they're just defined by some laboratory somewhere, and therefore have no practical application in the real world.

One of my hobbies is metrology and if my voltage standards are off by 10 ppm, I consider that serious. Photography isn't so fussy but the guide numbers should be at least very close. I wonder if the manufacturers measure "balanced and blue-printed" units to get the numbers, then lose it in production. I've found capacitors tend to be low in value these days (probably saves materials so long as the tolerance is met) and the quality of reflectors and optics might suffer from constant cost reduction.

One of my hobbies is metrology and if my voltage standards are off by 10 ppm, I consider that serious. Photography isn't so fussy but the guide numbers should be at least very close. I wonder if the manufacturers measure "balanced and blue-printed" units to get the numbers, then lose it in production. I've found capacitors tend to be low in value these days (probably saves materials so long as the tolerance is met) and the quality of reflectors and optics might suffer from constant cost reduction.

 

As Joe first post, none of us here would spend the money to buy the ISO standard so we don't know. My thinking is that manufacturers do follow the ISO standard but because the way the standard were set they have a lot of lee way. I don't think they lose it in the production vs design. Capacitors have very loose tolerance so I think they make them low in value and yet still within specs.

Oh, OK then. We should just ignore the standards laid down for the Metre, Kilogram, Litre, Second, Ampere, Volt, Ohm, Inch etc. etc? Since they're just defined by some laboratory somewhere, and therefore have no practical application in the real world.

 

Until we reach relativistic speeds, a meter is a meter is a meter but I can promise you that the meter stick that you buy at the home center isn't accurate to the tenth of a wave length. Your kilogram of coke isn't accurate to the hundredth of a gram, most clocks loose or gain time and don't even get me started on electricity. Do you have a volt meter that is accurate to the electron? Have you even looked at schematic for a Fender guitar or bass amplifier? "All values +/- 20%. Find me a 100 ohm resistor that measures exactly 100 ohms?

It may help for you to read the old 1998 version of iso1230 - which is available free its it's Indian version (technically equivalent).

Until we reach relativistic speeds, a meter is a meter is a meter but I can promise you that the meter stick that you buy at the home center isn't accurate to the tenth of a wave length. Your kilogram of coke isn't accurate to the hundredth of a gram, most clocks loose or gain time and don't even get me started on electricity. Do you have a volt meter that is accurate to the electron? Have you even looked at schematic for a Fender guitar or bass amplifier? "All values +/- 20%. Find me a 100 ohm resistor that measures exactly 100 ohms?

Each of these values are simpler than the flash one too. With light you have a range of wavelengths present. If your light meter measures a slightly different range of wavelengths than that measured in the guide number determination that will introduce a difference. The details of flash spectral distribution seem to be in ISO3028 which I've not found an equivalent version of, but iso 2827 seems to be relevant and refers to '“photographic daylight D,,”...

but iso 2827 seems to be relevant and refers to '“photographic daylight D,,”...

 

Probably a CIE illuminant known as "D55." (CIE is roughly "The International Commission on Illumination," based on the French.) D55 is roughly equivalent to daylight with a color temperature of 5500K, which is roughly what you get midday in the sun; it's a combination of the direct sunlight plus "skylight" from the blue sky. This is where daylight films are typically balanced. (I'm going from memory, so best to cross check me before quoting, etc.)

Probably a CIE illuminant known as "D55." (CIE is roughly "The International Commission on Illumination," based on the French.) D55 is roughly equivalent to daylight with a color temperature of 5500K, which is roughly what you get midday in the sun; it's a combination of the direct sunlight plus "skylight" from the blue sky. This is where daylight films are typically balanced. (I'm going from memory, so best to cross check me before quoting, etc.)

Yes its CIE, the subscript is pretty much illegible at the end of the quote, but I thought it looked like 56. ~5500K would certainly make sense for a typical electronic flash.