331mph SSC Twatara

[mik]

Interesting road too.

https://www.topgear.com/car-news/superc ... ssion=true

[DeskJockey]

Scary how completely effortless that looks.

[mik]

Particularly when it is holding around 200mph, waiting for the appropriate time to apply full-beans.

I guess you’ll get back quicker from your “out to pick up a pint of milk” run.

[DaveE]

One-handed, punching-the-air celebrations.

At 200mph.

That was boring to watch, which is the incredible thing.

it looked absolutely rock solid though.

Nuts.

[GG.]

Mental.

[nuttinnew]

The climb from 200 mph would have been impressive enough if it was in kmh.

[mik]

Oops....

[mik]

Apparently the gearing would not have allowed anything like 331mph.... small outfit - accepted - but you’d have thought they would notice that kinda detail eh?

[Beany]

Racelogic founder went a bit nerdy with his analysis - I like it.



As he points out, if his number crunching is correct, 223mph and still pulling 0.1G suggests it still has a fair bit left on the table. Perhaps not 300mph though...

[V8Granite]

I’d been following it since it was first questioned on Misha Charoudins channel on YouTube.

It amazes me they got it so incredibly. It’s the equivalent of a Saco driver aiming for 120 and hitting 150 and just not questioning how

Dave!

[Rich B]

So I've seen the proof for how far they were out, but how did they get it so wrong?

[RobYob]

Racelogic founder went a bit nerdy with his analysis - I like it.

*Snip*

As he points out, if his number crunching is correct, 223mph and still pulling 0.1G suggests it still has a fair bit left on the table. Perhaps not 300mph though...

You'll know more about this Beany, won't Python do OCR straight from a windowed part of the screen rather than that blokes clipping the video and screen capping it?

I imagine Dewetron are not pleased at being associated with this FUBAR, the "error factor" of 1.4 sounds suspiciously close to the 1.6 for correcting mph to kph and the programmer didn't realise they were already working in kph...

Shocking that they never bothered with the simplest of calibration runs to check their equipment.

[RobYob]

Thought just sprang to mind that with the data they already have it'd be relatively straight forward to drop out good estimates of drag coefficient and power to the road and theoretical top speed.

[Sundayjumper]

Not sure about the 1.4 / 1.6, they’re actually quite a long way apart. More than 10%.

However, 1.4 is the square root of 2, and if you’re dealing with speeds and accelerations there are powers & roots of two involved.

[dan]

Americans in exaggerating shocker.

[RobYob]

Good point, more than believable than someone coding was thinking of something to do with RMS and stuck a root 2 in where it shouldn't have been.

[duncs500]

I don't really believe that one of their engineers who knew the gearing didn't point out that something is not right, perhaps management didn't want to listen.

If there was anything nefarious going on, then it was a pretty stupid attempt at deception.

Is it too much to ask to just use a VBOX? You'd think it would be standard practice.

[RobYob]

Is it too much to ask to just use a VBOX? You'd think it would be standard practice.

Judging by this somewhat outdated blurb STILL up on the SSC website they used Dewetron for everything. Fair enough it's a well know data acquisition company. I've used them myself.
https://www.sscnorthamerica.com/news/de ... -top-speed

October 26, 2020
“Our planning process for the record run was extensive, and required thorough validation to ensure everything was captured accurately and legitimately,” Jerod Shelby, CEO of SSC North America, explains. “SSC was committed to using DEWETRON’s measurement system; we have been incredibly impressed by the accuracy, as well as the direct support that their team has provided to ensure proper implementation and data verification.”

Now there will be some spectacular finger pointing going on out of public view about what flashy numbers were actually derived and displayed from the Dewetron data.... Oh to be a fly on the wall

[Beany]

Racelogic founder went a bit nerdy with his analysis - I like it.

*Snip*

As he points out, if his number crunching is correct, 223mph and still pulling 0.1G suggests it still has a fair bit left on the table. Perhaps not 300mph though...

You'll know more about this Beany, won't Python do OCR straight from a windowed part of the screen rather than that blokes clipping the video and screen capping it?

What do you think I am, a developer? there is a PIP* for everything though so while it might be possible, it's also entirely likely that Racelogic chap is more a basic data wrangler than that, and did the work on his own time rather than tasking his development team to shit on SSC

As for the RMS stuff you and SJ mention, that does sound more likely, but as you say, you'd think they'd do some basic calibration or A/B testing before heading out to do a proposed world record run with cameras etc.

*PIP, IIRC is basically a Python module manager - "pip install <module>" and bosh, it sets it up, saves all that tedious mucking around with source code

[RobYob]

What do you think I am, a developer? there is a PIP* for everything though so while it might be possible,

Yes master!

I saw a video of some dude doing windowed OCR from a game because when I have a spare life I'd like to have a go at programming a Python bot for Elite:Dangerous to automate some of the busywork.

I've gotten as far as dropping a screengrab from the Vbox blokes video into Excel and scibbling on it, then got distracted by Smith and Sniff videos.