SRAM goes electric....(almost official) Has Landed

[maxxevv]

And you think SRAM have seemingly missed all this?

"seemingly miss" and "blatantly choose to ignore" are one and the same in the eyes of marketing profit if you must split hairs. And nobody will admit to the latter even if confronted.

But we will have to see how it pans out. Who knows, it might really be as good as people think it should be.

[p3dalfaster]

Let's take a sport that is already smothered in tech, namely F1. If some hackers wanted to wreak havoc, F1 would be one of the primary targets. Has it ever been targeted? Not to my knowledge? Will it be? I have no idea. What I do know, is that SRAM are not a stupid company, I am sure they have employed industry experts to pen test their wireless system and I am sure it wouldn't be out in the wild if they thought for one second it could be compromised.

[MRM]

There is no absolute safety. Technology companies and their products get hacked as well. If you want the positives of new tech, you need to take the down sides as well until these get resolved (if that is even possible to do). If you like it, not making use of it, because of fear of groupset hacking seems like throwing the baby out with the bath water.

[KWalker]

When mobile BTLE beacons and protocols started to be developed about two years ago hacking was more of a concern in terms of extortion. People figured out how to root out insecurities in transmissions, track connected devices, etc. and often used it as leverage to get money from companies in various ways. Not saying this WILL happen, but it does happen more than people realize.

[eric]

Unfortuntatly some engineers do not realise that security is a very special dicipline and that the hackers are moving very very fast at the moment , so you need to keep up with what hackers are capable of. Wrong usage of seemingly strong security measures are one of the more common reasons to security failing

Yes. Doing that stuff right is how I make my living.

When I read the Sram patent and then the documentation for the chip and protocol they're using, I was prepared to find obvious design flaws as often happens. I was surprised that they seemed to be well designed from a security standpoint.

Of course that's just a start. The protocol leaves out the key exchange, which is a hard problem. Sram (or users) would need to do some sort of pairing between shifters and derailleurs in order to securely exchange keying material. But you need to do that anyhow so the shifters do not accidentally send commands to your teamate's derailleurs. And there could be flaws in the protocols that I did not see, or flaws in the implementation.

But it appears that Sram did a decent job of thinking these problems through, selecting good technology to solve them, and improving on it where needed. We'll see how well it works in real life when they're finally made available.

[pdlpsher1]

One potential problem is electromagnetic interference, such as those coming from high tension power lines. Do you recall those days when your HR monitor reads 267 while riding near a power line? Car manufacturers subject their test mules to very high levels of electromagnetic fields to make sure the drive-by-wire/ABS/traction control systems work as designed. So to think that a wireless system won't have any issues when wired systems can be subjected to interference is naïve.

[Calnago]

Truth is, some jammers on the side of the road or elsewhere would really be the last thing I'm worried about on a bike, even if it could be done, and I'm sure it will (kids will be kids). I'd be more concerned with things like what @pdlpsher1 just posted above. I want my system to work reliably 100% of the time. As soon as it doesn't work once due to some electrical or electrical magnetic interference I would lose confidence in it. Knowing it "could" happen would be enough for me to shy away... there's powerlines all over the place in urban areas. There's certain intersections where things go crazy with my electrical stuff. I don't worry if it shows an erratic heart rate or something in those areas because I know the cause and it's not affecting what my bike is doing. With wireless I'm not sure it wouldn't affect the bike. SRAM has got to be thinking about this stuff... whether they've resolved enough of these issues satisfactorily will be the question. I do like the idea of wireless shifting on tt setups however. And if it works flawlessly, then using it anywhere would be fine. But working "flawlessly" seems to be pretty lofty goal with a wireless system. We'll see.

[HammerTime2]

But you need to do that anyhow so the shifters do not accidentally send commands to your teamate's derailleurs.

I'm sure if SRAM had had this in 2009, Lance would have enlisted technical help from one of his homies to enable him to accidentally sent commands to Alberto's bike.

[tinozee]

Is there no way to require authority to send instructions from shifter to derailleur, and then encrypt that? Or to design so that so that the mech only responds to that one device's unique ID? There must be some way to do that for other wireless controlled things like drones etc.

[weenie]

Is there no way to require authority to send instructions from shifter to derailleur, and then encrypt that? Or to design so that so that the mech only responds to that one device's unique ID? There must be some way to do that for other wireless controlled things like drones etc.

you probably don't understand the way radiowaves work and thats ok.
you can jam anything, drones included of course. in fact, many lose control because they fly in areas where theres a lot of interferences (ie cities/nearby cities). But drones are hundred of meters or kilometers away from the controller. the derailleur is a meter away from the handlebar controller. or less.

the radio space has a finite amount of space. its a bit like sound waves you dont hear because of the frequency its vibrating at.

imagine you're talking with someone and someone calls and yells non-stop (the jammer). the only "authority" you get there is punching him in the face or putting a wall. neither are very practical. same goes for radiowaves.

otherwise:

encryption only ensure no one elses knows what order you sent to the derailleur. but they know you sent an order. and also they can "yell louder" so that the derailleur cant hear the order from your handle bars buttons.
also, authentication ensure only your orders are the ones the derailleur will follow.

sram will most certainly do encryption and authentication - your wifi and many (not all!) drones do that too.

i would also put the jamming into perspective, while its possible, its also easy to piss off your neighbors by jamming their wifi with your laptop. its not very hard. but how often does this happen in reality? exactly. nobody aint got time for that.

there's really little risk that it happens to your bike, and if happens at races, the police will be there real quick and start monitoring people doing it (since you're "yelling" when jamming you can actually be quickly located by triangulating the signal power - some radio amateur actually do that for fun)

[tinozee]

Tbh I didn't read all of the detail about the specific type of attacks in this thread. I was thinking there was some sort of hack involved. I never really considered a jammer or that you could even do that and that you can't protect it. But that was a nice clear breakdown of how it could work. Jam on.

[damond]

https://www.youtube.com/watch?v=bVufkFjYX8w

Note from the 2:10. Wasn't the rider supposed to shift heavier gear on that sprint scenario? It is clear that he shifts several times the left shifter. The fact of him shifting the left shifter that amount of times may indicate that the rumors (left shift = higher, right shifter = lower and both = front up/down) may be confirmed!

EDIT: stage finish was climbing tough, he was certainly shifting higher gearing!

[743power]

I was looking for FCC info on this, but found nothing. Did find a couple etap trademarks though. What I also found with the trademarks was a "qollector" trademark and corresponding FCC ID https://fccid.io/document.php?id=2590547

Looks like the Quarq data logger will finally see the light of day. These filings are from this spring so it certainly seems like a current interest.

[bikedoc]

Seen it, tried it, staying with DA mechanical

[hasbeen]

oh do tell!