Shimano Di2

Questions about bike hire abroad and everything light bike related. No off-topic chat please

Moderator: Moderator Team

Super_fast
Posts: 341
Joined: Tue Nov 01, 2005 9:36 am

by Super_fast

lordconqueror wrote:Any speculation on how much it would cost to get your Di2 programed for sequential shifting?


Re-programming probably won't work, but we haven't tried it yet. Normally the fuse bits of the microcontroller are programmed so it can be reprogrammed in a production process. But we have decoded the complete working of the DI2 group, so we can program our own microcontroller. Complete sequential shifting is an option (but as a racer it wouldn't be my choice I think), shift the front-derailleur with a single button, the shifting can be made wireless or for example let the derailleur shift multiple gears when you keep the button pressed. The sky is the limit :wink:

by Weenie


User avatar
stella-azzurra
Posts: 5072
Joined: Mon Jan 23, 2006 6:35 am
Location: New York

by stella-azzurra

And that's what I'm waiting for: wireless shifting. Pointless in investing in a wired version. Then no more hacked frames, or the specific di2 frame needed.

Although now I could been on someones frequency and shift their rig before or during the sprint :twisted: :twisted: :lol: :lol: :lol:
I never took drugs to improve my performance at any time. I will be willing to stick my finger into a polygraph test if anyone with big media pull wants to take issue. If you buy a signed poster now it will not be tarnished later. --Graeme Obree

User avatar
prendrefeu
Posts: 8634
Joined: Wed Oct 31, 2007 10:32 pm
Location: Los Feliz, Los Angeles, California
Contact:

by prendrefeu

Or opposing teams having a 'hack' on other teams during important races.
Anything wireless can be hacked, no matter the security.
Teams or individual racers *ahem*LanceArmstrong*ahem* are willing to pay for success if it does mean cheating.
Exp001 || Other projects in the works.

User avatar
stella-azzurra
Posts: 5072
Joined: Mon Jan 23, 2006 6:35 am
Location: New York

by stella-azzurra

I'm up for it: no holds barred racing.

1. Juiced
2. Hacked
3. Incommunicado

SaWEET!! :smartass: :smartass: :smartass:
I never took drugs to improve my performance at any time. I will be willing to stick my finger into a polygraph test if anyone with big media pull wants to take issue. If you buy a signed poster now it will not be tarnished later. --Graeme Obree

jeffr
Posts: 518
Joined: Sun Oct 16, 2005 11:50 am
Location: Maui
Contact:

by jeffr

Modern microcontrollers are fast enough to implement encryption schemes which would take longer to crack than the length of an average road race. Selecting a 64bit key with a LFSR polynomial would be sufficient. Wireless encryption is typically broken by brute force with poor key management and takes some time and a lot of traffic. A shifter would send too little data to analyze to discover the key. Although it has the disadvantage of little entropy and predictable cleartext. Anyway, I don't think it's a realistic scenario.

More to the point, while riders will obviously cheat by boosting their own performance, I really doubt anyone would risk the safety of others by purposefully mis-shifting another rider. Another point is that new wireless protocols are digital, in contrast to things like legacy heart rate and even power meters. It would be virtually impossible for random interference to cause a shift. Although it would be possible for prolonged interference to prevent a shift.

Super_fast, may I ask who you are working with/for on reverse engineering di2? You've more or less described all of the features fairwheel demonstrated at interbike.

cryoplasm

by cryoplasm

Oh wait, telepathic shifting happens to be around the corner too.

ty-ro
Posts: 670
Joined: Sat Nov 22, 2008 11:15 pm

by ty-ro

I think (or fear) that the direction this is all headed is some sort of cyborg shifting setup where your shifting is no longer in your control. Of course you could override the system if needed. The system 'knows' from your many year history of power output, heart rate, and other inputs and puts you in the optimum gear, cadence, etc. You would use GPS to program the route/race profile and the series of algorithms would calculate on the fly based on real time changes, blah, blah, blah.

To me, this would really suck the fun out of it.

(I'm only half serious with any of this.)

User avatar
ave
Posts: 1844
Joined: Sun Dec 28, 2008 10:15 pm
Location: Hungary

by ave

Well, it might be hard to crack the system, but jamming the signal might be easy enough. (?)

User avatar
stella-azzurra
Posts: 5072
Joined: Mon Jan 23, 2006 6:35 am
Location: New York

by stella-azzurra

ty-ro wrote:
To me, this would really suck the fun out of it.

(I'm only half serious with any of this.)


Errm similar conclusion I had on page 1
I never took drugs to improve my performance at any time. I will be willing to stick my finger into a polygraph test if anyone with big media pull wants to take issue. If you buy a signed poster now it will not be tarnished later. --Graeme Obree

maxxevv
Posts: 2010
Joined: Wed Feb 09, 2005 4:51 am

by maxxevv

ty-ro wrote:I think (or fear) that the direction this is all headed is some sort of cyborg shifting setup where your shifting is no longer in your control. Of course you could override the system if needed. The system 'knows' from your many year history of power output, heart rate, and other inputs and puts you in the optimum gear, cadence, etc. You would use GPS to program the route/race profile and the series of algorithms would calculate on the fly based on real time changes, blah, blah, blah.

To me, this would really suck the fun out of it.

(I'm only half serious with any of this.)


There is already such a thing in prototype form for a long, long time already. No, I'm not kidding.

ty-ro
Posts: 670
Joined: Sat Nov 22, 2008 11:15 pm

by ty-ro

maxxevv wrote:There is already such a thing in prototype form for a long, long time already. No, I'm not kidding.


Yeah, I figured they were already working towards this. :roll:

mrlobber
Posts: 994
Joined: Sat Oct 23, 2010 9:36 am
Location: Where the permanent autumn is

by mrlobber

Reminds me of some science fiction stories where all the highways were "automatical", but some adrenaline seekers hacked their vehicles to be invisible for central traffic regulation system (and thus for the other vehicles on the highway as well), and drove around manually.

I think we can't ignore that electric shifting is going to evolve in the direction of more automated, integrated, real time data analyzing solutions, although I see that for such "oldies" as weenies here, this certainly might not be the way we'd like to see things moving.

However, a paradigm shift also happens faster than one might think.

With advancements in sports science, human psysiology (and chemistry :D), sports are slowly moving towards proper programming of the athlete for the big event. There still exists a significant part of unknown factors for a particular human being, but this part is getting less each year, and eventually, with all the possible interaction factors well explored and modelled, the winning strategy will be calculated with high probability by some powerful (but small) supercomputer, just as game theory allows it.
Retired bikes: Cervelo S5 2015 / Felt AR FRD 2014 / Cannondale SS HM 2014 / Scott Addict SL 2014 / Scott Plasma Premium 2014 / Orbea Orca 2008 / Look 596 /

xnavalav8r
Posts: 2643
Joined: Mon Jan 21, 2008 12:09 pm

by xnavalav8r

When Di2 was first introduced a friend of mine got his hands on an early group. He was, at the time, an engineering grad student, and had worked out a way for his Di2 group to accept cadence input from his cycling computer which produced the appropriate up/down shift to maintain his desired cadence. Last I knew, he had it working pretty well. I believe his was sequential shifting as well. I'm not entirely positive. I do know the biggest trick for him to work out was how to delay shifts to allow for coasting. I know he tried to get a job in the cycling industry but is now working for an aerospace company and spends more time on a mountain bike than on the road.

maxxevv
Posts: 2010
Joined: Wed Feb 09, 2005 4:51 am

by maxxevv

xnavalav8r wrote:When Di2 was first introduced a friend of mine got his hands on an early group. He was, at the time, an engineering grad student, and had worked out a way for his Di2 group to accept cadence input from his cycling computer which produced the appropriate up/down shift to maintain his desired cadence. Last I knew, he had it working pretty well. I believe his was sequential shifting as well. I'm not entirely positive. I do know the biggest trick for him to work out was how to delay shifts to allow for coasting. I know he tried to get a job in the cycling industry but is now working for an aerospace company and spends more time on a mountain bike than on the road.


For accepting coasting, to prevent unnecessary shifts, he just needed to add the input of current bike speed as a comparison to cadence.

by Weenie


nitropowered
Posts: 1146
Joined: Fri Jul 17, 2009 4:10 am

by nitropowered

ave wrote:Well, it might be hard to crack the system, but jamming the signal might be easy enough. (?)


Yeah but then your own team would get screwed too. Unless you are running a mechanical group. Then it would be pretty easy narrow down who instituted the attack

Post Reply
  • Similar Topics
    Replies
    Views
    Last post