Can Di2 do...
 

Was just thinking about Di2. Can you:
- set it so it keeps you at a specific cadence? i.e. it changes gear automatically to keep you spinning at the same speed?
- set it so it automatically changes gear to the next highest / lowest? i.e. it automatically shifts the front and rear derailleurs to get the next gear in the sequence when you choose to shift up.

If not, surely this kind of thing should be the next step?

Daven

All it does is sub an electric switch in the brifter and put motors in the FD and RD along with a bit of chain position
sensing in the FD to trim the FD position. It is still up to the rider to decide when to shift. There are autoshifting
systems available but for 30-35# bikes and only a few speeds and they are mechanical, not electrical.

Not sure if Shimano is working on this (but they probably are), but Fairwheel Bikes has done your second:
http://www.cyclingnews.com/features/...ipped-hardtail
Integrating automatic shifting by cadence wouldn't be tough- Shimano already does that on the Nexus Auto-D:
http://cycle.shimano-eu.com/publish/...o_inter-3.html

I'm not certain I'd want this. It's kind of the difference between a manual and an automatic transmission in a car: with a manual, the driver can select the gear in anticipation of the approaching terrain or curve. With an automatic, it shifts only after it senses a change in load. (Although there are some incredibly sophisticated automatic trannys in cars now.)

Shimano was very specific when it introduced Di2 that it was NOT building an automatic transmission, just an electrically shifted standard gear train. It does shift very rapidly, automatically centers the front derailleur cage and offers multiple places to locate the shift buttons but it doesn't choose the "right" gear for you.

Not to say that function couldn't be provided and may be in the future, but not now.

I think that's called "maintaining plausible denialbility". If it doesn't work, it can be swept under the rug with a minimum of fuss.

Complete nonsense.

It's not CVT, and it's not automatic. Shifting is still user controlled. Nowhere does shimano claim, or is the group supposed to function as an automatic transmission.

As it currently stands, to answer the OP's question:

1) no
2) no, it is not a sequential shifting system, fairwheel bikes modification is extreme and is a prototype with no plans for market introduction. Sequential manual is a likely stock option for the future. Current Di2 and 2012 Ultegra electronic is 'manumatic', e.g paddle shifting on road cars that don't have true sequential shifting.

I agree with operator on this one. Shimano never intended Di2 to be an automatic transmission and it's target market wouldn't have accepted one anyway.

If Di2 didn't work it would be because it's shifting was unreliable or unpredictable. Shimano did a huge amount of prototype and field testing before releasing it to the market to assure it would work as advertised.

Well, I agree that Di2 doesn't do that right now, and their current market wouldn't accept automatic shifting per se. However, Dave N finished his query with:
"If not, surely this kind of thing should be the next step?"

IMO, if the mechanism is reliable*, then all that's really needed is the software to control it. Between Fairwheel and Shimano themselves, the software already exists. Integrating it to the level of a marketable consumer product is another kettle of monkeys entirely.

*The first time I got on a demo Di2 bike on a trainer, the front derailleur jammed in between the two chainrings. The Shimano rep was watching me, so I wasn't trying to do anything funny. When I got off, he fiddled with it for a couple minutes, then rolled the bike behind a curtain.

I'm with Jeff on this one; the capability is well within the ability of the current Di2, and there is a demand (as long as you call it a 'training aid') I'm sure this will happen eventually. It would be integrated with a power meter of some kind, and the sophisticated on-board computer will make all sorts of choices for you to optimize your performance.

I'm not saying that it's going to be wide spread, but since it seems to be only a trivial software problem away, I can't imagine something won't emerge.

Cool, thanks for the replies. Sounds like it could be something done in the future - probably not useful for pros as has been said, but quite handy features none-the-less.

Thanks

Daven

No.
No.

No question it could be done, either by Shimano or by a third party software provider, if someone thought it was marketable. Right now, Di2 isn't equipped that way and Shimano said it wasn't originally intended to be.

*Snork*!! He said "trivial software problem"... hee- hee!!

Well, maybe- in the sense that the tools are out there and there are apparently people with the skills to use them to who could make this happen. (See the Fairwheel Bikes link above.) That doesn't mean that reverse-engineering the software/middleware/firmware of the Di2 system is easy and/or straightforward. It takes people with talent and persistence to rewrite this stuff.

(FWIW: 22 years ago I met a lady who worked on chip-level code for GM-Delco. She'd have to debug things by working through 6-inch thick slabs of printout. Despite this, I liked her enough to marry her. Nowadays she does her magic at H-P. She doesn't do firmware anymore, though.)

How stupid do you need to ride a bike?

Probably stupider than this person: http://www.youtube.com/watch?v=ZbfD9OKwqBs

I didn't say it would be easy, especially as Shimano isn't know for their straightforward and open policies. However, software issues are much easier for a hobbyist with far too much time on their hands to work through than hardware. However, as the Fairwheel bike shows it can be done, and that there are people with talent and persistence working on it.

Or else, someone can just hack it together using an arduino, for street cred.