Yongnuo like flash with SHORT flash duration (for action)

asamimasa wrote in post #14414807
So speedlights and the PCB Einstein use an flash system called IGBT.
As far as flash duration is concerned, these flashes will get slower as you power up.

In contrast, a typical studio strobe (any other PCB strobe, elinchroms, even cheaper Calumet/Flashpoint strobes) gets a faster flash duration as you increase power.

If you want to continue using strobes, you may just have to lower the power as much as possible and bump up the ISO within reason. And get the lights in closer. Speedlights should get some pretty darn fast maximum flash durations, but that only happens at the lower 1/64 ~1/128 region

Ok thanks. How fast is pretty darn fast? What would be the speeds are 1/64-1/128?

TijmenDal wrote in post #14415192
Ok thanks. How fast is pretty darn fast? What would be the speeds are 1/64-1/128?

He was referring to POWER LEVEL, to achieve the brief flash duration times. Duration specs on speedlight units might be 1/200 (full power) to 1/20000 (1/128 power)

Wilt wrote in post #14414067
Canon is so secretive about the duration of the 580EXII, it does not bother to include information in the user manual or on the Canon USA web site.

http://www.waynesthisa​ndthat.com/flashdurati​ons.html

Canon's 580EX manual and 580EX II manual state that normal flash duration (I am assuming at full power) is 1.2milliseconds (1/833 seconds) or shorter.

580EX:
Full power = 1/833 seconds
1/16 power = 1/15,000 second
1/32 power = 1/19,000 second
1/64 power = 1/31,000 second
1/128 power = 1/35,000 second

The 430EX and 550EX also have similar flash durations.

The Vivitar 285HV is probably the grandfather of the "Yongnuo like flash" and should be affordable and sufficient for the OP's intent...

Full power = 1/1000 seconds
1/2 power = 1/2000 seconds
1/4 power = 1/4000 seconds
1/16 power = 1/16000 seconds

Andy Gock measure the Yongnuo 560 compared to other flashes:
http://www.gock.net …-durations-small-strobes/

clarence wrote in post #14416020
http://www.waynesthisa​ndthat.com/flashdurati​ons.html

580EX:
Full power = 1/833 seconds
1/16 power = 1/15,000 second
1/32 power = 1/19,000 second
1/64 power = 1/31,000 second
1/128 power = 1/35,000 second

The 430EX and 550EX also have similar flash durations.

The Vivitar 285HV is probably the grandfather of the "Yongnuo like flash" and should be affordable and sufficient for the OP's intent...

Full power = 1/1000 seconds
1/2 power = 1/2000 seconds
1/4 power = 1/4000 seconds
1/16 power = 1/16000 seconds

Andy Gock measure the Yongnuo 560 compared to other flashes:
http://www.gock.net …-durations-small-strobes/

Wow! Thanks for the post! I've thought about Vivitar 285's, but the pics I've seen from them always look very yellow.
That was REALLY helpful. Especially the flash duration of the YN-560's. Thanks a lot!

EDIT: Looking a little more into flash duration, I see that the 1/2 power of every flash in the test shows a significant exponential increase over full power. Whereas at full power *all* flashes shoot slower than 1/300, at 1/2 power, they all shoot at 1/900 and faster! Guess I should just use it at half power and I'll be fine! Thanks a lot guys!

EDIT2: Clarence, in your post you say that the 580ex and Vivitar shoot at 1/833 and 1/1000 at full power respectively, but the tests Andy Gock did on his site tell otherwise. Not that I was planning on getting a 580ex, but the Vivitar flash duration is quite a differnce. 1/1000 vs 1/275 on the site you posted.
Thanks.

The thing with flash duration is that if one want to get faster than 1/200s (1/160s- 1/250s) one has to use high speed sync. This HSS significantly reduces the flash power. This might be OK for portraits against sun during daytime but for action shoots one will need more power. So 3 - 4 580 incl. TTL triggers will server enough light on a reasonable distance for late night skate board or bike sessions. The total equipment will be quite pricey and often accidents happen.

A good alternative is a camera with an electronic shutter like a Canon G12. This shutter allows full flash power even at 1/8000s and is used in many professional night action shoots.

TijmenDal wrote in post #14414796
It's true most flashes have shorter flash durations when not used at full power, but most speedlights have outputs of 1/800 and faster...

Flash duration and minimum shutter speed are diffenrent things. The power of flashes is controlled by the flash duration. Means to reduce the flash output the time it flashes is reduced.

All DSLR use a mechanical shutter. This is a kind of a slot wandering over the sensor from top to down. The shutter speed is controlled by the size of this slot, not the speed the shutter passes the sensor. The sync time is the duration during the shutter slot wanders over the sensor while the flash is still on.

Thats why a shorter flash duration is not relevant for having a short sync speed. Only parts of the sensor would be exposed to the flash light. A technic called high speed sync is used to reach shorter shutter speed during flash. It works by sending out a couple of short flashes so that the complete sensor is exposed under the moving shutter slot. As one can see from the posted tables the flash power is reduced significantly while useing this. The many flashes send out during HSS are not summing up because they alll expose a different area on the sensor.

As posted before electronic shutters as found on cameras like the G11/12 do not have this effect, since no mechanical shutter moves over the sensor. They can offer short sync time even at full flash power.

TijmenDal wrote in post #14423850
EDIT2: Clarence, in your post you say that the 580ex and Vivitar shoot at 1/833 and 1/1000 at full power respectively, but the tests Andy Gock did on his site tell otherwise. Not that I was planning on getting a 580ex, but the Vivitar flash duration is quite a differnce. 1/1000 vs 1/275 on the site you posted.
Thanks.

I think it depends on what timing you're actually using. That site seems to include the whole output curve (before and after peak power), whereas the faster numbers are probably only measuring the time its actually at the required power output (or within x% of it), which is what actually contributes the most to the exposure.