There has been talk about members on this forum using a digital camera to time? i do not understand this phenomenon one single bit. Can someone explain how this is done?
There’s a method for counting frames- No 2 knows how it works. Number 2?
woahh…
hahhaa… yes number 2… please explain.
now its 10 times even more complex than i thought!
With my digital camera there is a cronograph that goes to the 100th of a second. In slow motion you can either watch the smoke from the gun, or go by the athletes first motion and get a fairly accurate time. Still not as acurate as electric timing.
in North America, the format is NTSC so that is 29.97 frames per second.
You can do a frame by frame analysis(i.e with Adobe Premiere), so if you do a 40yd in 4 sec and 12 frames, that’s 12/29.97 or 4.40, based on movement.
When filming meets, I try to include the starter to see the smoke from the gun. This gives a more accurate time.
DCW23 wrote a more extensive article…
Fjlee beat me to it, but yes, you simply have to read the timecode on the camera and do the math. If there are 29.97 frames per second, that means each frame is worth about 0.0333667 seconds. Let’s try an example:
An athlete runs a time (as read on your DV Camera timecode) that starts from 01:22:28 (from the gun smoke). Assuming you’ve properly lined yourself up with the finish line, you take the finishing time - in this case, let’s use 01:33:06 on your camera timecode.
That means it took your athlete 10 seconds and 8 frames to run the 100m. What does that mean? Well obviously it took ten seconds and some change. So simply multiply the 8 frames x 0.0333667 = 0.2669336 seconds. So, your athlete ran a superb 10.27 seconds in the 100m (rounding up of course).
Does that make sense? It comes in handy at meets when you don’t have electronic timing or, in my case, the timing breaks down and you only have your camera to give you proper feedback. That’s why I always line up on the finish line at meets, and have the starter in the frame (to see the gun smoke) at the start.
ok first of all what camera do you use?
now let me get this straight? you stand by the finishline and look through the camera at the starter and start the camera when you see the smoke? then what?
sorry, im just trying ot understand this… its making sense little by little.
It must be a digital video camera. All DV cams have a running timecode. For every frame taken, there is a counter that records how many frames, and how many seconds have passed. In the case of NTSC cameras (North American format), it will be 29.97 frames per second.
Now that you know this, you just tape all your races as you would normally. I just identified how you would do it to find out your times in a race. Obviously, you want to be right at the finish line (to get an accurate finish time). We also said that the best way to get the start time accurately is to watch for gun smoke (as light travels faster than sound).
As you review the tape after the race, you can pull off the start time (from the counter) and the finish time. The differences between the finish time and the start time is your race time. It’s not like a stopwatch that you start when the gun goes and stop with the finish.
You may need a practical demonstration to get a real feel for what we are talking about. Do you know anyone who has a DV camera?
This is important, as you can kill two birds with one stone. Video analysis and electronic timing for one price and far less set up time- wires, photo eyes, etc.
What a GREAT idea!! You can just start filming the race when the athlete is on your marks (keeping the starter in the frame) and then take note when you see the smoke. Follow the athlete through the race and concentrate on the finish line. great great idea!!
It is very good. During meets I generally get up as high as possible and set up about the 60m mark for a 100. I get the most bang for my buck this way.
It is even better for those of us with PAL, because each frame is made up of two fields, each = exactly 0.02 seconds. If you get the time of the winner from the clock, you can advance field by field to get the times of the placings before they are posted.
Another top tip if you are buying a mini DV camera, go for the maximum OPTICAL zoom. This may mean that the camera is a little bigger in size, but it makes all the difference. Mine has 20x optical. It means that you can video from further away without degredation in quality. The further away, the less perspective will interfere with your analysis.
Good points DCW. (Have you ever noticed when watching car races etc, it always looks slower the farther the camera is away.)
here’s a great site that explains everthing about using a DV camera for timing…
http://www.trackinfo.org/videotiming.html
Poor Man’s
Automatic Timing System
By rule, timing qualifies as automatic timing if timing is started electronically without human reaction, it is taken with a CCD camera and you photograph the athletes crossing the finish line perpendicular to the finish line. Most current home cameras use CCD’s (very old cameras might not qualify–they will say things like vidicon, saticon, plumicon, which mean they are a tube camera). If you understand what conventional automatic timing does, you will be able to place your camera at the same angle (elevated, unobstructed, right on the finish line) to take the same kind of picture of the athletes crossing the finish line. This timing is based on the video frame rate, which is 29.97 frames to a second. That’s what limits the accuracy to .03 or .04 of a second. I have extrapolated the rate (including the necessary rounding up factor) out on the chart at the end of this article. New, commonly available video editing programs can capture video and then allow you to analyze the video, frame by frame. Personally I use Final Cut Pro (FCP) on MacIntosh, but there are several dozen other programs on other platforms that can do this (I Movie, IMmix, Media 100, Avid, Premiere etc.). You must capture at full speed–the same speed as the camera is shooting, 29.97 fps (some programs might simplify this to 30 fps but your timing is not valid unless you can certify which rate it is counting at–15 fps or 10 fps ARE NOT ACCEPTABLE).
The procedure is to to a continuous recording starting with a picture of the starters gun as the race is started (the picture of the flash or smoke exiting qualifies for the automatic start because human reaction is removed from the timing process). After the gun, reframe the camera to the finish line shot and record the entire time the race in progress (photographing each athlete as they cross the line). You can do it the slow way and record to video tape (so you have to calculate times later), or with newer computer editing systems you can go directly into the computer. When you capture directly into the computer, still record the video tape for back up–you might also find you can’t get your computer set up to record fast enough to keep up with the meet operations, so the video tape will save you from missing a race. To capture live you will probably need to disable transport control in some fashion, FCP allows that in the “Device Control” menu under “Log and Capture.” For ease of operation, in your set up menu you should set the start of your time line to either 00:00:00:00 or 01:00:00:00. By the way, 1:00:00:00 is traditionally used in the video industry because many of the early computers upon which this is based, did no understand the concept of midnight and coudn’t count properly past 23:59:59.29 (in other words, it couldn’t go back to 0).
One more thing to understand before you can use the chart. There are two kinds of “Time Code” and you must use the chart for the system your computer is using. If you don’t know which yours is doing (it probably is doing what is called “Non-Drop Time Code.” If you can’t find it in a menu, you can find out by going to the top of the first minute. You have to locate the counter (lower left on FCP but it could be anywhere on the different systems). As you advance throug the frames, if it counts 59.29, 1:00.02 (missing 00 and 01) you are in Drop Frame mode, if it counts 59.29, 1:00.00, 1:00.01 you are in Non-Drop Frame. Non-Drop makes math easier but the conversion to real time will not. Drop Frame converts to real time easier.
Locate the video time, then use the rounded time. I have visually done the rounding, following IAAF/USATF procedures. There may be some mistakes in here, so I included the actual calculations in the column next to the rounded time. Please let me know if you see an error in the chart, so I can correct it. Because of the volume of data required, I’ve only got the first 3 minutes–I’ll do more on other pages later.
Drop Frame Chart Non-Drop Frame Chart
I’ll describe the process using FCP: Open “Log and Capture” Your camera goes into firewire in (adaptor boxes are available if you need them to do this), device control is set to non-controlable device if you do it live (or use the machine control if you are working from a tape). From the finish line, as the starter is calling them to their marks, hit “Now” to begin recording (check to make sure it really is doing it–it may take a few seconds depending on your drive and other set ups) and shoot a close up of the starter. After the gun is fired, frame up on (where the athletes will cross) the finish line. Record all the way through the last runner. Escape to stop recording. Save it (obviously give it a unique file name you can figure out). Open the file in FCP. Scan forward to the start. Back up (reverse arrow) one frame (the flash or smoke is the first frame AFTER the gun has sounded). Hit the letter “i” (the “in” marker). Drag that file into your timeline sequence, aligned with the start of the timeline. Scan forward to the frame the athlete first touches the finish line. Read the position time in the lower left. Look up that time on the chart and you have the automatic time.
Doing this in a rush in bright daylight is a challenge. If you can’t do it fast enough for the meet, I’ll recommend a second computer to read and a couple of external (firewire) drives to exchange data between. All in all, this is commonly available equipment (that more and more people already have) and can be had for a few thousand dollars rather than upwards of five figures.
My qualifications: I am a USATF Certified Official, with a specialty in AccuTrak/FinishLynx timing–currently with 32 years of experience. Professionally I own a TV Production company that does nationally seen TV shows and infomercials. In other words, I understand both sides of these systems.
thanx… interesting…
You couldn’t time a fly accurately though, sine the camera can’t sit at both the start and finish at the same time. Maybe if you’re a trigonometry whiz you could figure out exactly where the start/finish lines are relative to the position of the camera… but that would be pretty hard
thats what a speedtrap 2 is good for!!
Originally posted by kevinG
You couldn’t time a fly accurately though, sine the camera can’t sit at both the start and finish at the same time.
The better the zoom the narrower the angle. It’s pretty easy to get a very accurate fly. You need to use four cones and it becomes easy to see the point that the athlete crosses the line.
Also, if you are using mini DV, it is better to get the timecodes off the camera, not your PC. Each frame is made up of two fields. In PAL, each of these fields are 0.02. You just need to type these codes into, say, a Pocket PC and the split is calculated in an instant.