THSCC Forums

I didn't switch vids anywhere in that split-screen view- the top is all your afternoon Run 3, and the bottom is all my afternoon Run 3, or at least the files that I identified as such. I don't see any discontinuities in the vid for your run, either.

I synched them by looking for the first frame in each where I detected motion in the windshield view. Then I went back 15 frames (1/2 second) in each to provide just a bit of visual buffer.

From looking at the vid, though, it sure looked like you got a 2 tenth run on me going into the back section, which I took back by out-braking you.

I wonder why thet differences between the video and the GPS appear so large?

_________________
Carl Fisher

Be Cool to the Pizza Dude:
http://www.npr.org/templates/story/stor ... Id=4651531

Hmmm That is curious. You are right, it does appear from that video that I took a lead on you in the 1st 2 sectors. My 3rd run should have been CHN_0012.MPG while yours should have been CHN_0013.MPG. Are those the files you used?

It's hard for me to make out what is going on from the split-screen so let me take a look at the original files and I'll get back to you. At this point I'd be more inclined to trust the data logger as the video may have dropped some frames or something. I'm not sure how the recording software deals with that or if it was even a factor. It's also possible that the CF card I used to record the video wasn't fast enough but that is a only remote possibility as I think we would have seen other artifacts.

What software are you using to make the videos? That is one area I'm lacking in as I've never done much with video on a PC. I may have to check them out on my Mac when I get home but that won't be until this evening.

Jim

I couldn't load the files with XP's Movie Maker (improper codec but it sure gave an obscure error message) so I found another program, VirtualDubMod, which let me open and play with the files on the PeeCee.

That's right! VDubMod in 'da house, representing Deutschland!

Anyway, I trimmed both files to the point that the car first starts moving so they start at essentially the same place. If your software can seek by frame number, my car starts moving at frame 594 in CHN_0012.MPG while your car starts moving at frame 552 in CHN_0013.MPG. Well, they are both your car but you get the idea.

After that, I played each movie and tried to stop near the areas you were talking about with a decent reference point in each frame. Notice that I stopped the movies at exactly the same frame number each time so it should be as accurate as it can be. The images are a little big so I'm going to just post a link to them.

Jim heading into sector 3
Carl heading into sector 3

If you use the (fuzzy!) cone at the right side of the pictures as a reference, it looks like we are in a dead-heat at that point.

Jim leaving the loop
Carl leaving the loop

If you use the cones at the left side of the pictures as a reference, it appears you have a *slight* lead on me at that point. After seeing that, I went back to the data and put a new marker at roughly that point on the course. According to the data, I ran a 50.14 while you ran 50.01 up to that point so it looks to be spot-on to me.

Jim crossing the finish
Carl crossing the finish

For this final pair of images, I didn't pick the same frame number since you got there first. Instead, I tried to match up where we both crossed the finish line by using the finish cone on the left side of the pictures. If you look at the time stamps at the bottom of the pictures, you'll notice that your time is ~63.7 (1:03.664) seconds while mine is ~64 (1:04.031) seconds. That matches up very well with our posted times as well as the data.

For those that don't remember, keep in mind our times are longer here in the "virtual world" because we are using a different start and finish line. As I said before, it's the start that has the biggest impact on the discrepancy. According to the video, it takes ~3.7 seconds just to trip the start beam! So if we take Carl's real world time (ignoring the cone) from his 3rd run and add 3.7 seconds, what do we get? 59.985 + 3.7 = 63.685 seconds. That is very close to his "video time" of 63.7 seconds and matches up nicely with everything else.

I'd be inclined to say that the video didn't drop any frames and looks to be right as expected. I'm not sure what happened with your video but I did notice that it is very easy to be off by a few frames. Considering each frame is ~0.033 seconds, it can start to add up quickly.

I'm hoping to sync the data with the video and have some overlays in the next few days but it is low on the priority list at the moment.

Just for grins, I took a screen shot of where I would be on course when Carl just crossed the finish line. He was ~0.3 seconds ahead at the time and if you ever wondered what that looked like at roughly 35mph, here it is:

Carl is in the pits having a cold one while I'm still charging towards the finish.

Jim