Wednesday, September 1, 2010

High Speed sync output tests....

(warning, this is a highly geeky post)
Ever wondered what the light coming out of your Nikon flash looked like output-wise when in Auto-fp High Speed Sync? Here are actual samples as taken with my pocket DSO and a photo-transistor.
I did this to put to rest some conjecture out on the internet. This is a proper method to understand what flash output is doing vs time. We can see output intensity and duration for a given shutter speed.

I tested this using a D90 I had lying around combined with an sb-800 flash set into manual power. I handheld the camera and aimed it towards the photo-transistor that was connected to the scope. The total output could be varying as I was handholding it and readjusting, but the graphs shape and duration are valid.

What the graphs will show are the following.

1) Output is practically continuous as the flash is pulsed for the duration necessary ensure a proper exposure.

2) When additional power is requested in auto-fp, the intensity of the flash is increased.

3) Full frame sensors use short auto-fp durations for the same shutter speed as compared to crop bodies.

Nikon D90, SB-800 1/128 @ 1/200th sec shutter (Clipped output going up, a regular flash pulse is a huge output of light that tapers off)

Nikon D90, SB-800 1/128 @ 1/250th sec shutter

Nikon D90, SB-800 1/128 @ 1/500th sec shutter

Nikon D90, SB-800 1/128 @ 1/1000th sec shutter

Nikon D90, SB-800 1/128 @ 1/2000th sec shutter

Nikon D90, SB-800 1/128 @ 1/4000th sec shutter

Now for a comparison of more power and also a full frame body.

This shows the D90 at 1/128th power and 1/32nd power, it is hard to make out but overall power is increased out. I noticed this when I shot it at full power, but didn't document it with a picture.

The last image shows how the light output waveform is different for the same shutter speed with a different body. I utilized my D700 for this test. In retrospect I should have made a test with my D300s as well to see if the two crop body shutters are different mechanically.

Nikon D90, SB-800 1/128 @ 1/1000th sec shutter

Nikon D90, SB-800 1/32 @ 1/1000th sec shutter

Nikon D700, SB-800 1/128 @ 1/1000th sec shutter (Note shorter duration)


  1. Very useful tests , thanks for taking the time !

  2. Would setting the flash power level at 1/2 make a substantial difference on the test results?

    How about 1/1?

  3. Cool Post. D90 vs. D700 effect might not be due to APS-C vs. FF but due to the D70o's faster shutter (1(250s vs. 1/200s X-sync)