32 replies to this topic

[Cociani]

An article was written in the Atlas magazine about light activated launch systems which would be triggered by the start lights going off. Such a system would have hidden/and/or disguised sensors on the car itself or be remotly activated through the cars central computer. This concept would be very difficult to police if it were banned.

What do you think of this possible development. I think it would take a great deal of excitment out of the race start and take further control out of the hands of the drivers. It is reminisant of a video game race start.

Over to the rest of you.

[Amadeus]

printed on April 1st by any chance?

[Hi Test]

Yeah, when I read this I was imagining the lights going out, the car starting off on its own and suddenly the driver realizing the car in front of him has stalled!

Yikes - abort system! Hit the escape button. No good. Okay, CTRL-ALT-DEL (get blue screen on dash VDT). Damn. Okay, turn off ignition. WHAM.

[Cociani]

Amadeus, Printed april 18th 2001. and not at all beyond the realm of possibilities. Light sensors have been around for a long time, so has the radio. A machine is bound to be more consistant at launching a car than a human being.

[Cociani]

Should have done this before. Here is the URL for the article. www.atlasf1.com/2001/san/shoebotham.html

[Amadeus]

Originally posted by Cociani
Amadeus, Printed april 18th 2001. and not at all beyond the realm of possibilities. Light sensors have been around for a long time, so has the radio. A machine is bound to be more consistant at launching a car than a human being.

Of course light sensors have been around for years. One or two practical questions....

- How will your sensor tell the difference between lights going out and the sun?
- or a camera?
- How will it compensate for distance?
- How will it compensate for aborted starts?
- How on earth could you hide it? "Hmm loads of wires leading to a photoelectric panel cunningly hidden in the Prancing Horses eye. Ye gad it leads back to the ECU, where we find lots of code for a hidden launch system!! Nothing suspicious here - move along!!"
- Why bother? TC is legal in Spain
- And too many perfect starts (defined in most formulae as ah well, it did seem like a good idea......

[Cociani]

Amadeus, Read the article. The whole point is that the legalization of TC makes the whole thing possible and more effective. The sensor need not even be on the car it coud be set up remotly and could trigger the cars computer by radio control. I don't think any of this is all that technically farfetched. Traction control does not improve reaction time to a start light but a system like this could.

[Amadeus]

Just read the article, and he tries to cover a lot of my points.

However:
- How does <0.2sec head start translate to >8m? The presumption seems to either of a perfectly linear acceleration pattern (false) or that the car instantly reaches cornering speed (obviously false)
- Further, his assertion that "The cars will not be able to accelerate harder than the competition" shows he has never heard of gearing! Or the effects of aero drag and downforce....
- His asumption of a 0.2 sec gain is also false. If an athlete can react in 0.1 and the car reacts in 0.2 (his numbers) then optimal possible acceleration is humanly 0.3. In drag racing, as he states it is 0.4. The best possible gain you can make is by eleminating the 0.1 he says it takes an atlete to react - this haves the gain that can be made.
- The FIA would soon implement a minimum raection time anyway, totally negating the gain
- "Triggers would incorporate a time delay to provide launches in the minimum legal time, and they would hit this minimum time more consistently than the drivers could" And how easy is that to police!!! 'Ah, car x starts in exactly 0.40001 for the sixthrace in a row - not at all dodgy!'
- How on earth does a light sensor look like a pressure sensor?!! They are not that easy to disguise, not least since they must be in direct line of sight with the start lights!
- Radio comms of the type he mentiones are specifically banned and easily monitored

In short the gain is so small as to be not worth the huge lileyhood of being caught - it is too easy to find for the risk. This is an example of the worst kind of sloppy journalism - pure speculation with only a hazy grip on fact, scaremongering and raising groundless fears.

Frankly I expected better of Atlas than this kind of sensationalist rubbish.

[AD]

Actually when the new rules were realesed a few months ago, this light sensor system was specifically named and banned by the FIA.

[Cociani]

Point taken Amadeus. This article does further open the disscussion on the perils of attempting to police cheating in the age of electronic gagetry. Even the appearance of cheating justified or not, (endless debate on Ferrari/Benneton TC), does hurt the credibility of the sport and wears on my patience reading this BB. I don't have any perfect solutions for the problems presented too us by advanced electronics but I think it is important deal with and discuss these issues in a constructive manner; Perhaps we can come up with some ideas which eventually get implimented. I am starting to sound like an idealist, how pathetic ;)

[Hi Test]

Stick the sensor in the drivers helmet/visor because he will always be looking at the lights. Wire the sensor in with the radio commications channels - bingo bango done.

[P1 Pyrsol]

i agree, this system is not only feasable, but probable. won't tell you where ours are though

[Hi Test]

Originally posted by P1 Pyrsol
i agree, this system is not only feasable, but probable. won't tell you where ours are though

But, it will never see the light of day there

[desmo]

I don't think that it would be too hard to detect. It wouldn't be like TC that can be emulated using only legal hardware, you'd nee some pretty obvious bits or non-voice pit to car communication. The scrutineers aren't dumb.

[pbrett]

I must remember to take my laser pointer to the GP's in future.....

[f91jsw]

I agree with Desmo. An optical sensor with the ability to focus on the start lights and exclude all other light around would have to be fairly large, and it would be easy to detect. Also, it would be easy to intercept radio transmission from the pit and detect anything out of the ordinary.

[BRG]

Given the frequency with which the FIA starter seems to muck up the lights sequence (eg Imola F3000 - no lights at all), I am not sure I would want to rely on a lights triggered launch!!

[Dimo]

This is obviously too hard to police and should therefore be legalized. Preferably after the 4th GP of next year, which should give team red enough opportunity to win a few races with their "legal" light sensing system before the rules change.

Do I need a ;) at the end of that, or is that abundantly clear already?

[pbrett]

Originally posted by f91jsw
I agree with Desmo. An optical sensor with the ability to focus on the start lights and exclude all other light around would have to be fairly large, and it would be easy to detect. Also, it would be easy to intercept radio transmission from the pit and detect anything out of the ordinary.

Optical sensor - agreed, too big, easy to fool ( me and my lazer pointer could flood any CCD imager ).

Radio - no, very easy to hide. I worked for Marconi's for a few years working on Spread Spectrum radio systems. You can hide the signals below the general noise floor and still get a great connection / datarate.

On a sprectrum analyzer you see a very small overall increase in noise and that's it. I seem to remember a thread about teams radio systems months ago here ( I've been around a long time, just don't post much ) and someone said they used spread spectrum anyway.

Regards

Phil

[random]

Originally posted by f91jsw
I agree with Desmo. An optical sensor with the ability to focus on the start lights and exclude all other light around would have to be fairly large, and it would be easy to detect. Also, it would be easy to intercept radio transmission from the pit and detect anything out of the ordinary.

I think you're underestimating the current state of the art in radio communications and light sensors. While this stuff is still too large to install on the car itself, it doesn't need to be on the car.

An optical sensor could be held by a "fan" in the grandstands and easily disguised as a video or still camera. Aimed at a single light and adjusted for ambient conditions, it could be programmed to send a radio signal the instant the light started to fade.

As Phil has pointed out, spread spectrum digital signals would be very tough to pick up, let alone analyze and attribute to a single team. And of course frequencies can be changed every race. Keep in mind there is no need for a transmitter in the pits or car. All the FIA analysis would pick up is a very, very short stray radio signal from "somewhere". I think a signal burst this short would be nearly impossible to triangulate.

The only way scrutineering could try to defeat such a system is to look for a radio receiver in the car. But modern receivers of this sort can be very, very small. I've seen some Japanese ones smaller than a pea. Drop that pea sized receiver inside a hollowed out capacitor or chip on the ECU motherboard and it would be almost impossible to detect.

[IndyIan]

Wow, F1 meets electronic warfare Teams using radio bursts to control their cars. Other teams using jamming signals to leave drivers sitting on the grid wondering why their car is not doing a perfect start.
I guess F1 is high tech but this is a little crazy.

How about each teams has to give 5% of their budget to a FIA anti cheat squad, that way the FIA would have the resources to make sure teams weren't cheating know matter what the rules are.

This way the rules could be based on what makes interesting racing instead of what can be inforced.

[palmas]

What is the actual diference in reaction time in F1 pilots? very, very small, I would say.
What are the chances of finding the optical cheating system? or of the system failing the start? or messing up the guy in front (even if not a williams)? What is the real potential gain? and how big is the risk?

Now, really, would you invest in such a system?

I would is the is any possible gain!;)

[random]

A 1/10th second jump on the competition may not be worth it in itself. But combine that with near-perfect start consistancy and a vastly reduced occurance of jump-starts, and it may well pay for itself.

Of course the merits of such a system also have to be weighed against the risk of being caught ;)

[imaginesix]

Originally posted by Amadeus
Just read the article, and he tries to cover a lot of my points.

However:
- How does <0.2sec head start translate to >8m? The presumption seems to either of a perfectly linear acceleration pattern (false) or that the car instantly reaches cornering speed (obviously false)

The presumption is that the driver reaches 100mph before braking. That is explicitly stated in the article, and is a real-world figure. No other assumptions are necessary.


- Further, his assertion that "The cars will not be able to accelerate harder than the competition" shows he has never heard of gearing! Or the effects of aero drag and downforce....

'The competition' in this case refers to an equal car without the start-light system. basically he is saying the cars will not have extra grip, just extra time to accelerate.


- His asumption of a 0.2 sec gain is also false. If an athlete can react in 0.1 and the car reacts in 0.2 (his numbers) then optimal possible acceleration is humanly 0.3. In drag racing, as he states it is 0.4. The best possible gain you can make is by eleminating the 0.1 he says it takes an atlete to react - this haves the gain
that can be made.

He also explicitly mentions both 0.1 sec reaction time for sprinters, and 0.2 sec reaction time for drag car drivers. These are presumed figures, used for the sake of regulating starts. Which one is actually representative of a driver's reaction time is unknown, so he chose 0.2 sec as a working figure.


- The FIA would soon implement a minimum raection time anyway, totally negating the gain.

There are gains other than strict reaction times to be had from such a system, such as consistency and protection from false starts, neither of which would be lost even if the FIA did implement a minimum reaction time. That was also explicitly mentionned in the article.


- "Triggers would incorporate a time delay to provide launches in the minimum legal time, and they would hit this minimum time more consistently than the drivers could" And how easy is that to police!!! 'Ah, car x starts in exactly 0.40001 for the sixthrace in a row - not at all dodgy!'

What's dodgy? If the regulation is 'minimum 0.4 sec reaction time', car x is fine, if the regulation is 'no start-light system', then it is suspicious, but that is the whole point of the article!


- How on earth does a light sensor look like a pressure sensor?!! They are not that easy to disguise, not least since they must be in direct line of sight with the start lights!

I agree, such a sensor might be difficult to hide, but I don't possess enough knowledge of these devices to rule it out.


- Radio comms of the type he mentiones are specifically banned and easily monitored.

That is not true, the kind of transmission such a system would need would be nearly impossible to even notice.


In short the gain is so small as to be not worth the huge lileyhood of being caught - it is too easy to find for the risk. This is an example of the worst kind of sloppy journalism - pure speculation with only a hazy grip on fact, scaremongering and raising groundless fears.

If Coulthard had such a system in Imola, it could have prevented him from loosing the race. Is that worth enough?
This is an example of the worst kind of sloppy criticism - pure contradiction with only a hazy knowledge of the actual text, self-righteousness and inventing the facts to fit the argument.


Frankly I expected better of Atlas than this kind of sensationalist rubbish.

Well now we know who you are to be the judge.

[desmo]

The more I think about it, the harder it seems it will be to preclude the use of such a system. With the allowal of TC/LC & auto shifts, the only driver input between launch and the first turn may be the simple press of a button to initiate launch and whatever steering input is required to block or avoid hitting other cars.

If the light sensing hardware is located off the car and all that is necessary to initiate launch is a simple burst of coded data (this at least I hope to avoid an accident due to a failure of the system), it will be hard to preclude the use of some sort of signal being transmitted to the car to "push the go button" for the driver.

Perhaps requiring all the com hardware to be helmet mounted and not connected to the car might make it more difficult. I've read that the Israelis send out bursts of EM energy to try to initiate the detonation of any terrorist radio-controlled bombs while still in the hands of their makers and have actually had some success using this but I can't see it as a good idea to be trying that as the cars sit revving on the grid waiting for the lights to go out.

Any bright ideas that might make enforcing this rule feasible?

[random]

EM bursts probably work for the Israelis becuase they're going up against low-tech analog radio controlled devices. The F1 boys are already using digitally encrypted spread spectrum pit communication systems. If the receiver were looking for an exact frequency and a digial code to be sent, it would be a billion to one chance that any other signal would activate the start-system.

The other possibility is FIA radio jamming. But digital spread-specturm technologies were initially developed by the military to preclude jamming. You basically have to jam the entire EM spectrum to block a good spread spectrum broadcast.

Without major changes in the starting process, I think the FIA is up against it if the teams decide this is a technology they need.

The only fix that comes to mind is for the total removal of the current lights. Put a sealed-unit FIA LED system on the dash of every car. Activate it with a digital spread spectrum encrypted channel and turn on everyone's lights at the same time. It's not jammable and since it's constantly communicating with the cars on an encrypted channel, there is no tell-tale for the cheaters to base a start system on.

The main problem here is activating all the lights on all the cars at the same exact moment. This is solvable however with a wireless network and very accurate clocks in each unit. The GO signal would be sent for the cars to activate the lights at an exact time a few seconds in the future. The network server would wait for an affirimative from each car, if it didn't receive a response from each car within 1 second the abort signal would be sent. If all the responses were received, no further signals would be sent and the on-board-systems would illuminate all the lights in all the cars at the exact same time to within a milisecond.

[Amadeus]

Originally posted by Amadeus
Just read the article, and he tries to cover a lot of my points.

However:
- How does <0.2sec head start translate to >8m? The presumption seems to either of a perfectly linear acceleration pattern (false) or that the car instantly reaches cornering speed (obviously false)

The presumption is that the driver reaches 100mph before braking. That is explicitly stated in the article, and is a real-world figure. No other assumptions are necessary.
I’m sorry, I disagree. He states “But is this enough to be significant? A 0.2 second advantage into a medium speed first turn at 100 miles per hour (160km/h) would result in a distance advantage of 8.9 meters.”
In other words the cars, at 100mph cover 160 kilometers in an hour. That is 160,000m in an hour, or 44.44 recurring m/second. Multiply that by 0.2 and you get 8.9meters. In other words at 100 mph the cars cover 8.9 m in 0.2 seconds. Therefore they must reach 100mph instantly to generate a 8.9 m gap.

- Further, his assertion that "The cars will not be able to accelerate harder than the competition" shows he has never heard of gearing! Or the effects of aero drag and downforce....

'The competition' in this case refers to an equal car without the start-light system. basically he is saying the cars will not have extra grip, just extra time to accelerate.
“If every car on the grid has well-optimized traction control then start line accelerations will be very similar”
He explicitly states that he is referring to “every car on the grid”. He is ignoring drag, gearing, engine and fuel load and effectively saying that if you remove driver input and add traction control then all cars will accelerate equally. I am afraid I do not agree.


- His asumption of a 0.2 sec gain is also false. If an athlete can react in 0.1 and the car reacts in 0.2 (his numbers) then optimal possible acceleration is humanly 0.3. In drag racing, as he states it is 0.4. The best possible gain you can make is by eleminating the 0.1 he says it takes an atlete to react - this haves the gain
that can be made.

He also explicitly mentions both 0.1 sec reaction time for sprinters, and 0.2 sec reaction time for drag car drivers. These are presumed figures, used for the sake of regulating starts. Which one is actually representative of a driver's reaction time is unknown, so he chose 0.2 sec as a working figure.
He does explicitly mention both sets of numbers. My point is that he has contradicted himself and confused the issue by saying that Drag Racers react in 0.2 but athletes in 0.1. If the car takes 0.2 to respond then the difference between 0.2 and the actual start is the human reaction time.
If an F1 driver can react as fast as a sprinter (not an unfeasible assumption) then the fastest possible start is 0.2 (car) + 0.1 (human) = 0.3.
He states that a car reacts in 0.2. The difference between 0.3 and 0.2 is 0.1 seconds. This is half the difference that he is claiming, and is based on the numbers he himself has given in the article.

- The FIA would soon implement a minimum raection time anyway, totally negating the gain.

There are gains other than strict reaction times to be had from such a system, such as consistency and protection from false starts, neither of which would be lost even if the FIA did implement a minimum reaction time. That was also explicitly mentionned in the article.
I was referring to the reaction time gain. I agree that the other benefits exist

- "Triggers would incorporate a time delay to provide launches in the minimum legal time, and they would hit this minimum time more consistently than the drivers could" And how easy is that to police!!! 'Ah, car x starts in exactly 0.40001 for the sixthrace in a row - not at all dodgy!'

What's dodgy? If the regulation is 'minimum 0.4 sec reaction time', car x is fine, if the regulation is 'no start-light system', then it is suspicious, but that is the whole point of the article!
My point was that this is concrete and measurable, as opposed to TC which was subjective. There are ways around it by modulating the delay trigger, but that then starts to negate the benefit. The fact that you can see that a car is always starting with a consistency that no human could achieve proves a launch control. With TC it was always a “well, his wheels don’t spin because he’s good” subjective and unmeasurable thing.


- How on earth does a light sensor look like a pressure sensor?!! They are not that easy to disguise, not least since they must be in direct line of sight with the start lights!

I agree, such a sensor might be difficult to hide, but I don't possess enough knowledge of these devices to rule it out.


- Radio comms of the type he mentiones are specifically banned and easily monitored.

That is not true, the kind of transmission such a system would need would be nearly impossible to even notice.
The military level of F1 radio comms does make it very hard to police. However the rules only allow 2 way voice comms, with no other 2 way communication. The rules would only need a simple tweak to ban the hardware.

In short the gain is so small as to be not worth the huge lileyhood of being caught - it is too easy to find for the risk. This is an example of the worst kind of sloppy journalism - pure speculation with only a hazy grip on fact, scaremongering and raising groundless fears.

If Coulthard had such a system in Imola, it could have prevented him from loosing the race. Is that worth enough?
Agreed. However, TC was a ‘soft’ cheat system – it could be implemented through software using pre-existing and legal hardware. This is a ‘hard’ cheat system needeing very complex and advanced equipment to make it function correctly. The fact that this equipment would be pretty obvious to an F1 official means that the teams will do a simple cost/benefit analysis. Gain: 0.1 – 0.2 at start and a vastly lower chance of missed starts. Cost: Very high chance of being caught, development time will be at expense of other, possible legal ways of making other gains (if I had an F1 team I would rather gain 0.05 per lap legally than 0.1 illegally at the start). While DC would possibly not lost the race at Imola with such a system when (not if) the launch trigger was found McLaren would be removed from the championship – where is the bigger risk?
This is an example of the worst kind of sloppy criticism - pure contradiction with only a hazy knowledge of the actual text, self-righteousness and inventing the facts to fit the argument.
Actually I did read the text – the hazy section was because I just jotted down 1st impressions. Having gone through in finer detail I stand by my original feelings – the text has no solid grounding in fact, carries unsubstantiated speculation and is a scaremonger piece in the lowest tradition of tabloid journalism. The posts in this thread (particularly those on data transmission) have been intelligent, well thought out and reasoned. Based on fact rather than sensationalism they are everything the article is not.

Frankly I expected better of Atlas than this kind of sensationalist rubbish.

Well now we know who you are to be the judge.
Expression of opinion is a free right. The author who put that piece in the public domain invited criticism and comment (as I am by putting this in public). If you don’t wish to read comments that contradict your opinion I suggest you stay away from BBs!

[Mellon]

- How does <0.2sec head start translate to >8m? The presumption seems to either of a perfectly linear acceleration pattern (false) or that the car instantly reaches cornering speed (obviously false)
The presumption is that the driver reaches 100mph before braking. That is explicitly stated in the article, and is a real-world figure. No other assumptions are necessary.
I’m sorry, I disagree. He states “But is this enough to be significant? A 0.2 second advantage into a medium speed first turn at 100 miles per hour (160km/h) would result in a distance advantage of 8.9 meters.”
In other words the cars, at 100mph cover 160 kilometers in an hour. That is 160,000m in an hour, or 44.44 recurring m/second. Multiply that by 0.2 and you get 8.9meters. In other words at 100 mph the cars cover 8.9 m in 0.2 seconds. Therefore they must reach 100mph instantly to generate a 8.9 m gap.

No. Car A with launch control starts 0.2s and car B without launch control starts 0.4s after the lights. For arguments sake assume the cars accelerate equally. Draw a graph of the cars' speed vs. time. Now shift car B's graph -0.2s. The graphs will be identical except for car A having travelled 0.2s longer at ~160km/h => car A has gained 9m.

[desmo]

Mellon, thank you for sparing me from having to make the same point.

"The fact that you can see that a car is always starting with a
consistency that no human could achieve proves a launch control. With TC it was always a 'well, his wheels don’t spin because he’s good' subjective and unmeasurable thing."

Attempt to define "consistancy that no human could achieve" and I think you'll see the folly of basing il/legality on such vague criteria.

"However the rules only allow 2 way voice comms, with no other 2 way communication. The rules would only need a simple tweak to ban the hardware."

Yes you could certainly ban the hardware but then we're up against the old bugaboo of enforcability. Others have mentioned some potential scrutineering nightmares that could result.

I thought the article was quite good and found it thought provoking. It certainly brought up a lively debate here didn't it?

[imaginesix]

Amadeus:

1-Any driver who can find a 0.2 sec advantage will maintain that exact advantage over the course of the lap, everything else being equal.
At the very start of a race, that may only equate to a one-foot advantage if the driver can get up to 5km/h in that amount of time, but the difference increases as speed increases, because the time advantage stays the same. That is why you see the cars close up on each other in the slow bits and distance themselves again in the straights, though the time gap between them remains virtually unchanged.

2-I see your point about the description of the cars accelerating at equal rates. It would have been more accurate to say that the cars will simply not have the opportunities they had previously to close up on, or pass, a competitor on a race start. I feel that to insult the author for trying to be concise, on an issue that really has no huge impact on the overall argument, is juvenile.

3-So he chooses to use 0.2 sec, and you choose to use 0.1 sec. They are neither contradictory nor confusing. Just like speed limits are different in different countries, these figures are only intended to approximately represent human reaction times and are sourced from two different regulatory bodies.
0.1 sec still equates to one car length at 100 mph, and remains significant. I am sure you are aware of the extent teams go for an extra 0.1 sec here and there.

4-I fail to see how the ability to precisely measure response time 'proves' anything. There are a million problems with that concept.
Hehe, not that that has ever stopped the FIA!

5-As far as the banning the hardware needed to receive such a signal (or banning it's operation during a race start), again I have to revert to my ignorance of such matters.
You assert as fact that scrutineers would have an easy time finding any extra receivers or light sensors. If that is correct, it may provide the enforceability that puts an end to start-light systems, but pardon me if I don't take your word for it.

6-"...I stand by my original feelings – the text has no solid grounding in fact, carries unsubstantiated speculation and is a scaremonger piece in the lowest tradition of tabloid journalism."
I respect your feelings. But before you made that clear, I felt it important to point out that you have not supported your impressions by identifying even the slightest iota of unfounded argument, unsubstantiated speculation or scaremongering.

"If you don’t wish to read comments that contradict your opinion I suggest you stay away from BBs!"
I second that, Amadeus!

[DangerMouse]

Back to 93-94 there were teams which alledgedly used such a system to automatically launch the car by watching the lights, an amendment then full rule came into effect in 94 specifically banning such systems.

Who's to say the sensor couldn't be remote with the sensor positioned anywhere in view of the lights? it would be easy to hide a radio receiver in the hardware on the car and link it to the clutch system!

[desmo]

I am being halfway facetious, but how about we go back to a flagman starting the race? I think I've still got a "When the green flag drops, the bullsh*t stops!" T-shirt from the '70s somewhere.

[imaginesix]

Yes desmo!!!!

Of course, it would have to be an automated 'flagman' in order to pre-empt accusations of favoritism, and to avoid the possibilty of the poor chap's head being lobbed off in a start-line crash!