Crash crash crash, but not for real REALS

October 19, 2017 12:40 PM

DestructionNation posted a series of videos where they drive virtual vehicles over 100 consecutive speed bumps at high speeds in the simulation video game, BeamNG.drive. Oddly relaxing and hypnotic.
posted by Brandon Blatcher (48 comments total)

God, I hope this can fill the hole left in my heart by Car Boys :(
posted by Merzbau at 12:51 PM on October 19, 2017 [8 favorites]

I really enjoyed that more than I thought I would. It made me think about how amazingly far computer graphics have come in my lifetime.
posted by 4ster at 12:58 PM on October 19, 2017 [1 favorite]

Interesting how the modern cars do a better job protecting the passenger cabin than the older models. Is the modeling that accurate?
posted by notyou at 1:00 PM on October 19, 2017

No idea how accurate the modeling is, but I did enjoy how realistic it is without being hyper-realistic. In some ways, the cars (and associated destruction) have a cartoonish edge to them that is subtle reminder that this isn't real, yet still allowing one to enjoy the various sighs and sounds of the the laws of physics and motion being applied to metal and steel. The vibrating jiggle of the wheels and body is a hilarious prelude to the glorious destruction you know is coming, but you don't know how exactly how bad things are going to go, so it's interesting to see what happens.

What, like you never had a crappy day at work?
posted by Brandon Blatcher at 1:05 PM on October 19, 2017 [2 favorites]

I really enjoyed how the [game? engine?] has determined "this piece of the car is what the camera follows" even when it's shredded down beyond a chassis or anything recognizable. I wonder if the camera is effectively attached to a single bolt somewhere near the engine compartment and it follows that one piece regardless.
posted by komara at 1:06 PM on October 19, 2017 [2 favorites]

God, I hope this can fill the hole left in my heart by Car Boys :(

or the hole left by the removed stair :-/
posted by numaner at 1:13 PM on October 19, 2017 [12 favorites]

I also liked how sometimes, you could see the engine fan still spinning after the hood had been torn off.
posted by 4ster at 1:18 PM on October 19, 2017 [1 favorite]

Excellent. Would be interesting to add a crash test dummy with rag-doll physics into the mix.
posted by Fizz at 1:20 PM on October 19, 2017

Sometimes when I don't want to slow way down before a speed bump I tap the brakes enough to dip then front end and then let up in time for the front end to start back up just before hitting the bump. That strategy would be completely useless here and somehow that pleases me.
posted by BrotherCaine at 1:20 PM on October 19, 2017 [1 favorite]

here's the link to the actual videos
posted by indubitable at 1:31 PM on October 19, 2017

Also, based on this and my expectations for the future of the nation’s roads, my next car will be an older model Jeep Cherokee with a hefty lift kit.
posted by notyou at 1:35 PM on October 19, 2017 [7 favorites]

Is the modeling that accurate?

I noticed that too - the yellow old 50's car (no idea what it is other than hideous yet also awesome) crumpled in a realistic enough way, and the Subaru cracking the windscreen due to torsion on the frame early on in the crash was pretty close too. I was also wildly amused by the exhaust on the Subaru (starts at 1:52) being almost unbent and bouncing enthusiastically at the end of the crash. Although it did bounce through the rear tyre, which is less realistic!

So the modelling is pretty accurate, but not extremely, from what I can tell. The Subaru is underdamped (from how it handles the normal bumps on the roads) and the cement mixer seems to be falling apart before anything even happens to it, and yet the biggest credibility stretch is the Jeep doing 125mph...
posted by Brockles at 1:37 PM on October 19, 2017 [5 favorites]

Great fun, but it looks like these vehicles are operating under something less than full earth gravity. Would they really remain in motion for that long after losing the use of their wheels? Does it even matter?
posted by Flexagon at 1:38 PM on October 19, 2017

I don't know how accurate the simulation is from the point of view of physics, but it seems to capture quite well the subjective feeling of driving over speed bumps, when you're in, say, a small sports car rather than the ridiculous SUV they were apparently designed for.
posted by sfenders at 1:41 PM on October 19, 2017

this whole channel is going to my drug for the month.

it's weird to recognize scenes from Car Boys in their videos.
posted by numaner at 1:43 PM on October 19, 2017

"Would they really remain in motion for that long after losing the use of their wheels?"cars that crash at 125mph (not lose control, but actually crash at that speed) go for a REALLY long way... The switch to slow mo and back makes it feel all floaty, though, I think.
posted by Brockles at 1:43 PM on October 19, 2017 [1 favorite]

I used to own a Trooper similar to the one seen at 3:30 in the first video (100+ Consecutive Speed Bumps High Speed Testing #3) and I can confirm that it would spookily float over speed bumps if you hit them at just the right speed.

It's good to know that if I ever buy another Trooper (not too likely) and am being chased by the Highway Patrol through the desert (also unlikely), I need to find a stretch of road with at least 30 but no more 95 speed bumps in a row.
posted by flyingfox at 1:52 PM on October 19, 2017 [6 favorites]

The other part of these games that I love is the workaround for non-commercial/branded names for the vehicles. So generic and elevator music-esque.
posted by Fizz at 1:55 PM on October 19, 2017

I need to find a stretch of road with at least 30 but no more 95 speed bumps in a row.
I sense the start of a master plan.
posted by Brockles at 1:56 PM on October 19, 2017

It bothers me that there is no driver in the cars. There should be a little Donald Trump in there with a rag doll physics model.

That would make this a lot better. And more accurate, if you think about it.
posted by Naberius at 2:01 PM on October 19, 2017 [1 favorite]

“TOONCES, NO!!!”
posted by Huffy Puffy at 2:05 PM on October 19, 2017 [8 favorites]

This reminds me somewhat of Tricky Truck, an especially brutal game from yesteryear. Example video.
posted by matrixclown at 2:05 PM on October 19, 2017 [4 favorites]

I wonder how bias-ply tires would perform under these circumstances. It's interesting how something like radial tires made motor vehicles dramatically safer to operate, yet at the same time enabled faster speeds and riskier manuevering.
posted by Brocktoon at 2:07 PM on October 19, 2017 [1 favorite]

Also the car at 1:52 looks more like a Lancer/EVO than a Subaru.
posted by Brocktoon at 2:10 PM on October 19, 2017 [1 favorite]

These are great. The Land Cruiser was really nailing it right until the last bump.

Hope they can get this kind of realistic damage modelling in the next GTA, that would be sweet.
posted by turbid dahlia at 2:33 PM on October 19, 2017

Reminds me of the Offroad Tattoo stunt on Jackass (ft. Henry Rollins).
posted by stinkfoot at 2:55 PM on October 19, 2017

If you're being chased by Highway Patrol (4:49), take the road with the speed bumps.
posted by AFABulous at 3:30 PM on October 19, 2017

Somewhere near, virtual James Spader gets a series of boners.
posted by Kafkaesque at 3:34 PM on October 19, 2017 [5 favorites]

Also the car at 1:52 looks more like a Lancer/EVO than a Subaru.

Yes, it is, you're right. Around 15 years ago when they went all angles and straight edges, they kind of blended into one for me and I totally lost interest.

radial tires made motor vehicles dramatically safer to operate, yet at the same time enabled faster speeds and riskier manuevering.
Well. The moves are no more or less risky. The speeds are much higher because the tyres are better. It doesn't make a move more risky at a higher speed - you're still within the same percentage of the edge of grip of the tyre.
posted by Brockles at 3:48 PM on October 19, 2017

It bothers me that there is no driver in the cars.
Only one of the cars had a driver (the red race car-ish one in about the middle) and that was when I suddenly noticed none of the other cars had a driver. Weird.
posted by Brockles at 3:49 PM on October 19, 2017

It doesn't make a move more risky at a higher speed - you're still within the same percentage of the edge of grip of the tyre.

It does, in that the stakes are higher at higher speed. The move beyond the edge of grip happens faster, and the outcome thereof is potentially worse.

I love Beam NG.Drive. Not enough to actually buy it myself, but stuff like this and Car Boys is awesome.
posted by Dysk at 4:10 PM on October 19, 2017 [1 favorite]

Somewhere near, virtual James Spader gets a series of boners.

100 consecutive boners.
posted by turbid dahlia at 4:13 PM on October 19, 2017 [1 favorite]

This thread is making me wish for a version of BeamNG.drive called "Tuner's Delight" that lets you do things like rebore your cylinders and then deal with your engine overheating eventually because your bore was too aggressive and now the coolant has leaked out through the resulting crack in your block.
posted by invitapriore at 4:31 PM on October 19, 2017

It does, in that the stakes are higher at higher speed.

My point was that the risk of losing control is the same, but obviously the consequences of the loss of control would be higher. You are not at any more risk of having an accident with radial tyres - they are a safer tyre - which the original comment could have been interpreted as. The risk is whether or not you have an accident, not your chances of having a BIG accident.

Driving at 90% (for instance) of the available grip of the tyre is the same in both circumstances.
posted by Brockles at 4:33 PM on October 19, 2017

Right, the tyres are safer, but the idea that taking the same corner at higher speed with a grippier tire (with speed increase consistent with grip increase) is no less safe is just untrue. Any changes to conditions, road surface, any maneuver you may have to make mid corner to correct something, to avoid a deer running into the road, whatever will have to happen faster, weight transfer thus happens faster and more suddenly, making it harder to not have the accident, as well as the accident being worse.
posted by Dysk at 4:50 PM on October 19, 2017

90% of grip is 90% of grip, though. Both cars have ten percent grip to recover from an eventuality. The scenario you are suggesting requires the car travelling faster to be nearer the edge of grip than the slower car for it to be a problem. I'd also wager that the bias ply car would actually be less safe at 90% grip, actually, because the tyres ability to recover from changing load is lower through the inferior construction.

The only difference between the two scenarios that could be an argument is the speed of correction to a changing condition - that's not a tyre thing, that's driver or vehicle handling ability. Faster doesn't automatically mean less safe - that's a fallacy - unless that extra speed also crosses a separate and unrelated parameter such as driver skill or vehicle competence. If you assume that the driver and vehicle are competent and capable of the extra speed, the risk is the same at the higher speed if we are judging the effect of a different tyre on the risk factor. Otherwise, your actual risk relates to the driver not being capable of going X percent faster.

If the bias ply tyre is at 90% grip on a freeway curve in the middle of the day in a desert, but the Radial is at 90% grip on a dark stormy night on a twisty mountain road with a tired driver driving through a deer migration route, then sure the Radial tyre is more risky but it is nothing to do with the tyre construction. But for a given scenario where tyre change is the only scenario, you are just as likely to lose grip and have an incident if you are at the same percentage of the tyres grip in each case.
posted by Brockles at 5:05 PM on October 19, 2017

Human reaction times are finite. In the time it takes you to react, you've travelled further if your travelling faster, possibly requiring a larger correction as a result, etc. At faster speeds, the change from 90% to exceeding 100% happens faster. Unless the tyres somehow affect your reaction times, that's still riskier.

There are all kinds of reasons modern radials are far safer. But higher speed always makes anything riskier. Driver skill isn't some magic thing that you're either inside of or beyond. Any driver is more likely to catch a slower speed incident than a faster one, even if the grip conditions are such that you're the same arbitrary percentage from the edge of grip.
posted by Dysk at 5:23 PM on October 19, 2017 [1 favorite]

Speed bump reality. Back in the yesteryears I was walking to high school. There was a lane with speed bumps up the side of the school to the back parking lot. As I was crossing the lane, a police car came by and stopped with its front tires right at the start of the first speed bump. He then stomped on the gas and the front of the car rose over the bump, tires off the ground, and then the front came crashing down on the pavement. The cop looked over at me with that if you say anything I'll kill you look. He then slowly drove up the rest of the lane.
posted by njohnson23 at 5:32 PM on October 19, 2017 [1 favorite]

It's a pretty good video, but so far my main takeaway from this post are some words related to tires that I didn't know before from a side argument I don't fully understand the disagreement in. That's metafilter in a nutshell right there.
posted by neonrev at 5:39 PM on October 19, 2017 [3 favorites]

Unless the tyres somehow affect your reaction times, that's still riskier.
In some ways it does, though. Because the way the car loses grip and (crucially) the way it regains grip and how quickly is a major factor in the difference between radial and bias ply. The Radial tyre helps by having a better grip loss character - arguably, you'd have to have better reactions on a bias ply than on a radial to recover for the same grip loss.

There are all kinds of reasons modern radials are far safer. But higher speed always makes anything riskier.
Either they're safer or they're not. It can't be both.

At faster speeds, the change from 90% to exceeding 100% happens faster.
I think you're imagining a much bigger speed differential between radial and bias ply tyres than actually exists so you're applying a differential that would exist for a wide range in speed and assuming it is also a tangible difference at a smaller rate of speed. Plus 10% of grip away from the limit is actually pretty huge. You can do a lot of things wrong within 10% of the limit and not have an accident. Also, vehicle design makes most of this moot anyway - Bias ply tires on a 1980's car at 60mph is far less safe than a 2017 car on radials at 80mph. But the 1980's car itself on radials would be less safe at 60mph also.
posted by Brockles at 5:41 PM on October 19, 2017

Either they're safer or they're not. It can't be both.

It kind of can. Ceteris paribus, they're safer, but once you introduce an additional element of risk (be that speed, conditions, whatever) you can no longer make a direct comparison - for some purposes, the safer tire might compensate for the extra speed, or wetter road or whatever, but that samme difference could overwhelm the impact of the tires in different scenarios - are we on a straight or a corner, RWD, FWD or 4WD? Is it we back stepping out, or the fronts washing? Or is it a brake lockup, and if so, which wheel? The balance will not be the same in each set of circumstances. All you can universally say is that radial tires are always better than bias ply, and slower is always better than faster. The precise balance between them isn't something you can meaningfully quantify or determine at a fixed difference.
posted by Dysk at 5:59 PM on October 19, 2017

Am I the only one who's totally unsurprised that the taxi cab in the first video managed the whole strip at 90 miles an hour unscathed?
posted by ardgedee at 7:32 PM on October 19, 2017 [2 favorites]

To be fair, those speed bumps are all crazy high. In my neighborhood we have a couple that are much lower and much more forgiving, although I'm not willing to try them out at 100mph. Let alone 100 of them in a row.
posted by monospace at 7:39 PM on October 19, 2017

you know what they, when you get in a fight with mr. floor...
posted by wibari at 11:07 PM on October 19, 2017 [1 favorite]

It's a disappointment that somebody went through the effort of designing www.ibishudreidtopspeedteam.com livery for their tuned sedan and didn't follow through with registering the domain.

Brockles: BeamNG hasn't licensed any real car marques, so the cars in the game are recognizable knockoffs with contrived names. For example, most of the Japanese cars, whether they're mock-Honda or mock-Subaru or whatnot, are "Ibishu", American cars and trucks are "Gavril", and so on. The game allows fans to create their own environments and vehicles, so the occasional actual Lamborghini or BMW badges or livery you see in these videos are third-party mods.
posted by ardgedee at 3:06 AM on October 20, 2017

The DestructionNation YouTube account that provides the videos for the aggregation site the OP links to has become my background noise for the weekend while I get deskwork done, and holy crap this boy has been busy making many, many hours of cars smashing into things over and over again, repeatedly and obsessively. You can watch 11 minutes of The Stigs swarming Fiat 126es through all kinds of catastrophes. Or the seeds of a potential e-sport: Automotive curling
posted by ardgedee at 4:47 AM on October 22, 2017

Just found this video comparing real crashes replicated in BeamNG.Drive, and it's pretty accurate, just much less small debris and slightly floatier physics, as noted above.
posted by numaner at 6:04 PM on October 23, 2017 [1 favorite]

This early Destruction Nation video uses a cannon and Busto 2.0 to make some momentarily beautiful footage out of glitching the then-current version of BeamNG.Drive.
posted by ardgedee at 4:35 PM on October 24, 2017