Measuring drop times. at DVinfo.net

Alastair Traill · November 2nd, 2012, 11:01 PM

I do not know how critical the drop time is in determining the performance of a stabilizer. However if accuracy and reproducibility are required counting one thousand, two thousand etc leaves a lot to be desired.

I thought that there should be a better way than counting so I recorded the “drop” in a mirror then played the clip on my editor setting the “in” point at the start of the drop and the “out” point when the column was vertical. The clip length thus became the drop time. The idea seems to work well and showed that what I thought was 1.5 to 2 seconds and another observer thought was over 2 seconds was in fact very close to one second – a considerable and maybe significant error?

Resolution depends on frame rates so at 50p I can resolve to 1/50th of a second. Remeasuring the same clip gave the same result so reproducibility was good also.

Dave Gish · November 3rd, 2012, 04:25 AM

Drop time is important.

With a short drop time (2 seconds or less), the bottom of the rig will tend to sway when you start, stop, accelerate, or change direction. Experienced operators can compensate by predicting this sway and feathering it out with the left hand, but this takes years to master.

Longer drop times (3 seconds or more) keep this swaying to a minimum, but then you have to be much more aware of keeping the rig level, which is particularly challenging if you're outside and there's wind.

I use about a 2.5 second drop time as a compromise, but I may go shorter if it's windy.

As for measuring drop time, keep in mind that you're typically not given a lot of time to set up, so counting one-one-thousand, two-one-thousand, etc tends to be the norm.

Hope this helps.

Alastair Traill · November 3rd, 2012, 06:54 AM

Hi Dave and thanks,

The aim of the exercise was to see how accurate or otherwise I was in my measurements which turned out to be not very. I realise the method I have described is somewhat roundabout and tedious but I only intend to use the one camera equipped with same accessories so really I am just calibrating my set-up.
I have been having problems with swaying so I will try longer drop times and see how I go. Thanks again.

Tom Wills · November 3rd, 2012, 02:04 PM

I think the important thing is to not get caught up in numbers. I know for a while I was obsessed with getting a 2.5 second drop time. Every time I set up my rig, I would try to get it to to that magic time, and start from there. Even though I had been operating for a while, I had just been taking for granted that 2.5 seconds was the magic number. On the feature I worked on a couple of months ago, I was doing a shot, and was having some issues with holding a tilt. The director (a former DP, and very knowledgable about Steadicam) asked me what my drop time was. I told him, 2 and a half seconds. His response? "I'm amazed you like your drop time that fast. My usual guy goes for 3 and a half."

And you know what, I tried it. I flipped my rig over, adjusted my gimbal, and I'll be damned, but 2.5 seconds just felt wrong compared to a much slower drop. I was relying on the numbers, rather than what my fingers were telling me. My operating improved from that point on, since I started listening to what my fingers were feeling on the rig.

If you find the rig's hard to hold tilts, or the pendulum action is killing you when you accelerate, slow your drop time up. If you're having trouble holding lockoffs, speed it up. Set it for every shot, and don't be afraid to not even count it - just give it a feel.

That's my two cents, anyways.

Alastair Traill · November 3rd, 2012, 05:32 PM

Hi Tom, thanks for your 2 cents worth.

I am at the stage where I do not know what my fingers are telling me. I guess that setting up a stabilizer is a bit like adjusting aperture, gain, shutter angle and ND filters to achieve a satisfactory exposure. There are many possible combinations but some are more appropriate for the occasion than others. However these parameters all have well recognized settings that you can return to with some accuracy at a later date.

I feel that I have made some progress in being able to accurately measure drop times and both yourself and Dave suggest that I should be aiming at longer times. Thanks again to you both.

November 2nd, 2012, 11:01 PM	#1
Alastair Traill Major Player Join Date: Nov 2008 Location: Melbourne Australia Posts: 626	Measuring drop times. I do not know how critical the drop time is in determining the performance of a stabilizer. However if accuracy and reproducibility are required counting one thousand, two thousand etc leaves a lot to be desired. I thought that there should be a better way than counting so I recorded the “drop” in a mirror then played the clip on my editor setting the “in” point at the start of the drop and the “out” point when the column was vertical. The clip length thus became the drop time. The idea seems to work well and showed that what I thought was 1.5 to 2 seconds and another observer thought was over 2 seconds was in fact very close to one second – a considerable and maybe significant error? Resolution depends on frame rates so at 50p I can resolve to 1/50th of a second. Remeasuring the same clip gave the same result so reproducibility was good also.

November 3rd, 2012, 04:25 AM	#2
Dave Gish Major Player Join Date: Feb 2008 Location: Riverdale, NJ Posts: 468	Re: Measuring drop times. Drop time is important. With a short drop time (2 seconds or less), the bottom of the rig will tend to sway when you start, stop, accelerate, or change direction. Experienced operators can compensate by predicting this sway and feathering it out with the left hand, but this takes years to master. Longer drop times (3 seconds or more) keep this swaying to a minimum, but then you have to be much more aware of keeping the rig level, which is particularly challenging if you're outside and there's wind. I use about a 2.5 second drop time as a compromise, but I may go shorter if it's windy. As for measuring drop time, keep in mind that you're typically not given a lot of time to set up, so counting one-one-thousand, two-one-thousand, etc tends to be the norm. Hope this helps.

November 3rd, 2012, 06:54 AM	#3
Alastair Traill Major Player Join Date: Nov 2008 Location: Melbourne Australia Posts: 626	Re: Measuring drop times. Hi Dave and thanks, The aim of the exercise was to see how accurate or otherwise I was in my measurements which turned out to be not very. I realise the method I have described is somewhat roundabout and tedious but I only intend to use the one camera equipped with same accessories so really I am just calibrating my set-up. I have been having problems with swaying so I will try longer drop times and see how I go. Thanks again.

November 3rd, 2012, 02:04 PM	#4
Tom Wills Major Player Join Date: May 2005 Location: Philadelphia, PA Posts: 326	Re: Measuring drop times. I think the important thing is to not get caught up in numbers. I know for a while I was obsessed with getting a 2.5 second drop time. Every time I set up my rig, I would try to get it to to that magic time, and start from there. Even though I had been operating for a while, I had just been taking for granted that 2.5 seconds was the magic number. On the feature I worked on a couple of months ago, I was doing a shot, and was having some issues with holding a tilt. The director (a former DP, and very knowledgable about Steadicam) asked me what my drop time was. I told him, 2 and a half seconds. His response? "I'm amazed you like your drop time that fast. My usual guy goes for 3 and a half." And you know what, I tried it. I flipped my rig over, adjusted my gimbal, and I'll be damned, but 2.5 seconds just felt wrong compared to a much slower drop. I was relying on the numbers, rather than what my fingers were telling me. My operating improved from that point on, since I started listening to what my fingers were feeling on the rig. If you find the rig's hard to hold tilts, or the pendulum action is killing you when you accelerate, slow your drop time up. If you're having trouble holding lockoffs, speed it up. Set it for every shot, and don't be afraid to not even count it - just give it a feel. That's my two cents, anyways.

November 3rd, 2012, 05:32 PM	#5
Alastair Traill Major Player Join Date: Nov 2008 Location: Melbourne Australia Posts: 626	Re: Measuring drop times. Hi Tom, thanks for your 2 cents worth. I am at the stage where I do not know what my fingers are telling me. I guess that setting up a stabilizer is a bit like adjusting aperture, gain, shutter angle and ND filters to achieve a satisfactory exposure. There are many possible combinations but some are more appropriate for the occasion than others. However these parameters all have well recognized settings that you can return to with some accuracy at a later date. I feel that I have made some progress in being able to accurately measure drop times and both yourself and Dave suggest that I should be aiming at longer times. Thanks again to you both.