Strobe & Flash Comparison Testing: Part I (Output)

What Are These Tests?

Objective and controlled testing of power output from various strobes and flashes ranging from inexpensive entry level to higher priced pack and head systems.

What They Are Not

These are not equipment reviews or comparison of features, build or specifications. Those differences can be found in manufacturers product specifications or on various retailer sites and your needs and shooting style will dictate what’s important and what is not.

Method Used For Testing

Each strobe and flash were tested under exactly the same conditions. All strobes were tested bare bulb with no reflectors or light modifiers of any kind. The center of the flash tube was aligned with the center of the dome of a Sekonic light meter 6 feet from the flash tube, mounted to a light stand for constant measurement position. Strobes were all turned on and allowed to warm up at least 10 minutes before testing.

Two series of tests were conducted. The first was to test consistency when firing just after the strobe recycled and the second was done by firing the strobe 5 seconds after the strobe recycled. In each test the strobe was fired 10 times.

How The Results Are Presented

Results are given for each strobe at all power levels available. The first and fourth table show the greatest variation between any 2 pops, measured in 1/10 stops. The second and fifth table show standard deviation and indicate erratic behavior regardless of the amount of variation (Example: 2 strobes can each show no more than 1/10 stop variation but if one is alternating between 2 values and the other only fluctuates once in a series of 10 then the first will have a higher standard deviation). The higher the number, the less consistent. Next, the third and sixth table express consistency as a percentage. Finally, the last table gives you an overall percentage of consistency for each strobe, averaging the percentages from individual power levels.

TABLE 1

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

canon forum and no canon flash included in test?

neat chart. looks like most are quite consistent.

No Canon's?

OT: And by the way, when'd you switch your gear? and why?

Nope, no Canon's. The flash units were thrown in just to make some comparison between studio strobes and small flashes.

The tests were done primarily to compare various studio strobes.

picturecrazy wrote in post #5109354
canon forum and no canon flash included in test?

neat chart. looks like most are quite consistent.

Lighting Forum and Studio Lights are compared. Canon does not make a studio strobe.

TMR Design wrote in post #5109400
Lighting Forum and Studio Lights are compared. Canon does not make a studio strobe.

I see it was mainly studio strobes... but i was surprised to see the SB800 in there and not a 580. that's all.

picturecrazy wrote in post #5109430
I see it was mainly studio strobes... but i was surprised to see the SB800 in there and not a 580. that's all.

If I had a Canon 580EX or EX II I surely would have included it but I don't own one and didn't have access to one.

I should mention that the Dyna-Lite and the Metz flash results were contributed by our own Wilt.

Robert,

Thanks for posting these results. They are pretty much what I anticipated. Although the D-Lites are more consistent than the Alienbees, I think the difference is relatively minor. In your first table, you showed that all the way down to 1/8th power the greatest variation between flashes was 1/10 of a stop if fired immediately after recycling. The variation at the lower powers is consistent with what Paul Buff says about his lights. (He readily admits on his forum that at 1/32 power they are less consistent. Hence, the reason why White Lightning X1600 and X3200 have the dual power settings. Of course, the D-Lites do not have a 1/32 power setting.)

In the testing showing the consistency 5 seconds after recycling, the Alienbees B800 varied by no more than 1/10 of second all the way through 1/16 power. Again, at 1/32 power, it is less consistent.

I know you are extremely precise and a 1/10 of stop difference probably is less than acceptable to you, but I think the difference isn't terribly significant. I think most people would be hard pressed to see a difference in two photos taken with a 1/10 stop difference.

Finally, I think the standard deviation tables and overall % consistency table are accurate but misleading. A strobe that deviates by no more than 1/10 of a stop when it deviates is doing very well. To me, the amount of deviation is more significant than the number of times it deviates.

This does not mean that I don't agree that the D-Lites are more consistent and, if you absolutely have to have that consistency, the D-Lites are the way to go. For me, the Alienbees and White Lightnings (which I own) are certainly consistent enough on the power deviation.

I'll be looking for the color temperature deviation to see if I find that more significant.

Michael

Hi Robert,

Thanks for investing your time in doing all this testing and posting the results! The tables are a very useful and interesting read! I can't wait to see part 2.

Just one question, when you say in for example Table 1 that the AB800 has a 0.1 greatest variation between pops, does that mean that there is a 1/10th stop variation or that there is a 0.1*1/10 stop variation, i.e. 1/100 stop? It is just that the table description says that the units are in 1/10th stops and then the values are 0.1, etc so I am confused whether you meant to say that the units are in full stops and then you are measuring 0.1, i.e. 1/10th stop difference or whether they really are consistent to 1/100th stop in which case a 0.1 variation really is not very much...

Best regards,

Anton

Anton,

I'm sure it's 1/10th of a stop but I'll let Robert confirm.

Michael

By the way, just to quantify it, here's an example of the difference 1/10th of stop makes. I'll use shutter speeds because it's easier to illustrate even though you don't use the shutter speed to control exposure with strobes. You use a shutter speed of 1/60th of a second. If your lighting were to require an addition 1/10th of stop to compensate for lower light, than you would need to keep the shutter open an additional 1/600th of second (1/10 x 1/60.)

magicmikey wrote in post #5109542
Finally, I think the standard deviation tables and overall % consistency table are accurate but misleading. A strobe that deviates by no more than 1/10 of a stop when it deviates is doing very well. To me, the amount of deviation is more significant than the number of times it deviates.

I don't believe they are misleading at all Michael. Many people have chimed in on various other threads indicting that they are forced to use their strobes at those lower power levels. If you're working at 1/16 power and there is a 3/10 stop variation at any given time then that is a recipe for inconsistent exposures.

If you had a smaller standard deviation it might mean that 1 or 2 in 10 pops are off by 3/10 stop, or if you knew that the first pop was always going to be off then it wouldn't be so bad but if you are getting readings that vary by 3/10 stop and it jumps around on you it's a mess. With reference to that exact situation.... when I first started working with lighting and I got a light meter I couldn't figure out what I was doing wrong and why I was getting different readings when taking the measurement in the same place. Once I figured out that the strobe was the problem and was not consistent at those lower power levels it cleared a lot of things up but it drove me crazy for months.

If you recall, this all started by talking about the inconsistent behavior of Alien Bees at lower power levels and the fact that many of us need and want to use those power levels. I use my Elinchrom's on the minimum setting whenever I need or want and get the same consistency as when I have it on full power.

It's ridiculous to have a strobe that boasts a 6 stop range when the usable range is only 4 stops. The Alien Bees strobe is the only strobe to show that kind of instability at lower power levels when every other strobe remains within that 1/10 stop tolerance from maximum to minimum regardless of popping on recycle or waiting.

aia21 wrote in post #5109612
Hi Robert,

Thanks for investing your time in doing all this testing and posting the results! The tables are a very useful and interesting read! I can't wait to see part 2.

Just one question, when you say in for example Table 1 that the AB800 has a 0.1 greatest variation between pops, does that mean that there is a 1/10th stop variation or that there is a 0.1*1/10 stop variation, i.e. 1/100 stop? It is just that the table description says that the units are in 1/10th stops and then the values are 0.1, etc so I am confused whether you meant to say that the units are in full stops and then you are measuring 0.1, i.e. 1/10th stop difference or whether they really are consistent to 1/100th stop in which case a 0.1 variation really is not very much...

Best regards,

Anton

Hi Anton,

1/10 stop is expressed as .1 and vice versa. Since nothing we have can measure or operate in hundredth's of stops that wouldn't make much sense.

Well now I don't really know what to do! I'm new to studio lights and I was set on getting a set of AB400 next week. From the looks of it, they AB's don't perform that well in the low end. But the thing is I don't really understand what it means when you say they are .7 off of variation. Does that mean that they are not bright enough, too bright? Should I still go with AB's? Should I move to Photogenic? Looks like a couple more long nights of research ahead of me....

Hi Cougar548,

What it means that your exposure will be off. Say you set everything up and take a measurement with your light meter whilst firing your lights. Then you adjust your camera to match the reading (or change power on the lights if you want to keep your camera setting it doesn't matter for this example), anyway, lets assume you take a light meter reading and it shows everything is set up correctly for perfect exposure.

If the light shows a .7 variation it means that when you take the actual photo your light might produce up to 7/10th of a stop difference to the time when you measured with your light meter. That would mean your photo would be up to 7/10th of a stop either underexposed or overexposed. And 7/10th of a stop is rather a lot of exposure variation. Basically two "steps" of exposure compensation on your camera (assuming your camera uses 1/3 stop steps) which makes a real visible difference.

Hope this helps...

Best regards,

Anton

Hi Cougar548,

If you're looking for the best alternative to Alien Bees I would recommend the Elinchrom D-Lites. The Photoogenic's performed very well but one drawback with the Photogenic's, primarily in the studio, is the lack of proportional tracking of the flash and modeling light. The modeling light is either on (full power) or off. If you don't need or care about modeling lights or are using them outside where you would not use modeling lights, then the Photogenic's are really great. I like them quite a bit.

Another very nice thing about the Photogenic is that if you plan on doing any location shooting with 1 strobe and a reflector or perhaps 2 strobes at the most, you can get the StudioMax III in an AC/DC version that has a Turbo power jack and is supplied with the Turbo cable so you can plug into any Quantum Turbo battery or the branded Photogenic battery pack. Battery life and number of pops is astonishing and you don't have to drag around a 20 pound Vagabond, Tronix, etc.