2014 Japanese Grand Prix - Suzuka

[Phil]

With what tools are you guys working? I've tried virtualdubmod, but the container/codec isn't supported, so I am having trouble viewing individual frames. As I said earlier, the best way to determine the how long an individual frame is, is to count how many frames are there over the period of a second accoarding to the files timeline. Then you'll either end up with the 30 odd frames per second or possibly less or more, depending what the source file is or what was lost in the converting (* => webm) process. The more frames there are, obviously, the more accurate the calculations will be.

[thedutchguy]

I downloaded the original MP4 file and imported it in a 29.97fps timeline in adobe premiere. Very easy to calculate frames from there.

[lebesset]

The problem the FIA had with stopping the race, or putting a safety car out was the risk of looking biased. The race had literally just 2 laps before ticked over into being a full points race. To suddenly go "oh, and... we're done" or "and the safety car will lead them around for the rest", rather than letting them deal with changing tyres, and all the risks involved in that would have looked incredibly biased towards Hamilton.

It seems very petty now after what happened, but the FIA normally do need to think about whether they're advantaging one driver by making calls, and what the political fall out will be of making those decisions.

laps under the safety car count just like the others ; full points were already going to be awarded

[Richard]

The red flag was on lap 46. The rules disregard the last two laps when a race is stopped, so the results count from lap 44. The total race should have been 56 laps, hence actual race distance was 78%. Full points are awarded after 75%, ie 42 laps.

As it happens Sutil crashed on lap 43. Even if the SC was deployed instantly in front of Hamilton it would have come out on lap 44. So they would still have full points if the SC had led them into the pits for a red flag on that lap.

[Jolle]

Seeing at what speed Bianchi was traveling at double waved yellow (in the wet) looks like the the biggest reason why this all happened. There are also reports that, for example, Ericsson was also driving too fast at that section.
When the SC came out and Hamilton was slowing down, Kobayashi was still on full speed.

for me it looks pretty clear it's not a question if they should have red flagged it earlier or bring out the SC (seeing Koba's pace, it didn't slow him down) but the disrespect of double waved yellow. The yellow was out for almost a lap when Bianchi crashed, so the team had more then enough time to inform Jules.

So, why didn't they respect the flags? is it to do with the 10th place in the constructors championship? the small teams are under enormous pressure not to finish last. Racing under yellow as fast as you can, can make a difference worth millions.

Did the team say: Double yellow at sector X, watch out, or did they say: Push Jules

[mrluke]

"The arguments for sending a Safety Car out were that the conditions were worsening, the light levels fading and some cars (like Sutil and Bianchi) were out on worn intermediate tyres which had done over 20 laps and had little rubber or tread left. F1 engineers have told this website that in those circumstances when rain is falling a worn intermediate tyre loses temperature very quickly, in as little as a lap, and then the grip level drops dramatically."

The FIA could possibly at limiting the number of laps (20 laps or min. tread depth) for full wets and inters, again setting a safe limit to race.

If the drivers tyres were worn out they should have pitted for new ones. It is not the FIAs job to send out a safety car because some drivers have decided they do not need to pit for new tyres.

[thedutchguy]

I made a video to show how I got to my estimation of 80 kph on impact:

[Phil]

Excellent work!

(Please excuse the lame photoshop attempt)


Due to the possible angle, the messured distance he travels might actually be a bit lower. Though judging by the picture above, I think it's actually rather by an insubstantial amount.

[thedutchguy]

The way I see it, it doesn't really matter: even though it's at an angle, its (nearly) the same point on the ground where the front and back wheel are, five frames apart. It doesn't matter from which angle you look at it.

However, ff you'd want to do actual measurements on the car, you would have to factor in the angle.

[FW17]

How come no one is talking about the Hamilton DNF had it not been for the asphalt run off at turn 1

[multisync]

I made a video to show how I got to my estimation of 80 kph on impact:

https://www.youtube.com/watch?v=RHJE1huwAgQ

Looking at the tracking app it seems the car lost control at 213, stepped out of line, down to 210 then decelerated to around 95 on impact

but these things are not 100% probably nor is my analysis..

[FW17]

I made a video to show how I got to my estimation of 80 kph on impact:

https://www.youtube.com/watch?v=RHJE1huwAgQ

Looking at the tracking app it seems the car lost control at 213, stepped out of line, down to 210 then decelerated to around 95 on impact

but these things are not 100% probably nor is my analysis..

Surprising then that the cockpit crash structure and the roll hoop did not survive an impact at 80 / 95 kph

[Richard]

Due to the possible angle, the messured distance he travels might actually be a bit lower. Though judging by the picture above, I think it's actually rather by an insubstantial amount.

The angle doesn't matter. The calc measures many lengths the car travels, not the actual distance.

Lets say the camera is at an acute angle with a shortening effect so the car is 100 pixels long. If the car travels 2 lengths (ie 200 pixels) in 2 seconds, that is 1 length per second.

Next assume the camera is square to the car so the car is 500 pixels long. If the car travels 2 lengths (ie 1000 pixels) in 2 seconds, that is still 1 length per second.

[Phil]

I don't think there is much crash structure if any in the roll hoop? The crash structure at the front of the car hardly made contact and went more or less under the clearance of the tractor. I think most of the damage was done to the left sidepod (little resistance) as well as the roll hoop (which would be built to resist force from the top incase of a flip, but probably not a frontal impact) IMO.

[multisync]

the roll hoop would have been in sheer so the force of the car against the edge of the crane meant it had no chance. The roll loop is designed to withstand being loaded along it's length - ie when a car flips.