Oi, manufacturers of electronic gear shifting, you're doing it wrong!

[alexb]

Well they are for the most part aren't they?

If I was going to build a system from scratch, the last thing I'd do, or want is to copy what's already there.
When I'm riding down the road, I just want to be able to shift up and down gears. I don't want to have to remember if the middle chainring, smallest cog is higher or lower than outer chainring and third cog. In an ideal world there'd be two buttons and there aren't there are four.

My hub gear does it right and with all that fancy technology, that's what I'd expect Shimano and Campag to do. Have the system be intelligent and figure it out. It might take a calibration step, but after that the computer should know all the possible gear combinations and execute them correctly to provide a higher or lower gear on demand.

[Kim]

While I'm inclined to agree, I'm not sure I'd want to get a front shift when I expected a rear one.  There's an easing off on the pedals inherent in derailleur systems that makes them tricky to automate.

[Hot Flatus]

I've come to the conclusion that Di2/eps solves a problem I haven't got, and at great expense.

A gear cable is £2.

[Basil]

imo the "doing it wrong" bit is doing it at all. 

[Ham]

My hub gear does it right ....

...except that to do it right it adds massive weight, complexity and cost in order to achieve that.

As an aside, the only Di2 I am considering is Alfine Di2, as that would address the issue of imperfect shifting register on Versa shifters for drops

[jogler]

I've come to the conclusion that Di2/eps solves a problem I haven't got, and at great expense.

A gear cable is £2.

Similarly I see it as a solution to a problem that doesn't exist.

It's more of a stimulant to the manufacturers sales figures.

[contango]

While I'm inclined to agree, I'm not sure I'd want to get a front shift when I expected a rear one.  There's an easing off on the pedals inherent in derailleur systems that makes them tricky to automate.

You'd get a front and back shift at the same time if your current and target gears required changing both. I'm with you that I'd rather know what's going to happen when I change gear - if I flick one shifter the front moves and if I flick the other the rear moves. I like it that way.

[Cudzoziemiec]

While I'm inclined to agree, I'm not sure I'd want to get a front shift when I expected a rear one.  There's an easing off on the pedals inherent in derailleur systems that makes them tricky to automate.

So you just need to introduce an automatic clutch. And then you might as well have fully automated gearing. And then... I'm not sure what's next after that, but there will be something!

[Kim]

Similarly I see it as a solution to a problem that doesn't exist.

It's a solution to a couple of problems that do exist.  But one (hand disabilities) doesn't sell in quantity, and they seem to have forgotten about the other (shifters in multiple positions) entirely.

[Kim]

While I'm inclined to agree, I'm not sure I'd want to get a front shift when I expected a rear one.  There's an easing off on the pedals inherent in derailleur systems that makes them tricky to automate.

So you just need to introduce an automatic clutch. And then you might as well have fully automated gearing. And then... I'm not sure what's next after that, but there will be something!

Fully automatic gearing has been done, unsurprisingly with a hub gear.

[Ham]

I wonder what would happen if one molished one of these into the drive chain

http://www.ebay.com/itm/NEW-GO-KART-MINI-BIKE-30-SERIES-TORQUE-CONVERTER-CLUTCH-DRIVER-1-BORE-/330905160344

Anyone tried?

[tatanab]

Fully automatic gearing has been done, unsurprisingly with a hub gear.

With derailleur gears too.

As for an electronic sequential shift, I too would not want an  unexpected front change, especially not when I am changing to a lower gear and the electronics decides that is best performed by changing small to large on the front and then down a couple on the back.  That would really screw you up half way up a climb because of having to ease off a lot to allow the front shift.

I regard a double (other than a half step double) as being more like Land Rover gearing with high ratio for normal use and low for anything that high ratio does not cope with.

[Euan Uzami]

I've come to the conclusion that Di2/eps solves a problem I haven't got, and at great expense.

A gear cable is £2.

You're so wrong, it rules. Don't knock it till you've tried it.
A gear cable gets bunged up with water and mud and doesn't change up, even a gore fully sealed one after a bit.
But the front shifting of di2 really is very very smooth. I thought the blurb about the electronics timing the change to coincide with the chain going over the ramps and pins was bollocks but it obviously isn't.

[Euan Uzami]

However I agree that I still want control over the front and rear individually.

[Rhys W]

My finger is a mechanical lever. My gear shifter is a mechanical lever. My derailleur is a mechanical lever with a couple of pivots. Replacing what goes in the middle of that lot with a rechargeable battery, microprocessor, electric motor etc is just madness.

Campagnolo put a stronger spring in their pushbutton switches so it would feel like an ergolever. 

[Rhys W]

A gear cable gets bunged up with water and mud and doesn't change up, even a gore fully sealed one after a bit.

Ever heard of something called bike maintenance?

Keep your bike clean and lubed, with new cables every year or so and you won't need an electronics engineer when things go wrong.

[Kim]

But the front shifting of di2 really is very very smooth. I thought the blurb about the electronics timing the change to coincide with the chain going over the ramps and pins was bollocks but it obviously isn't.

Okay, points to them for that.

Also for eliminating the trimming of the front shifter problem.

Bung a strain gauge in there and you could delay shifting until the load eases up sufficiently.

All good stuff, but marginal gains at best.

[BrianI]

I've come to the conclusion that Di2/eps solves a problem I haven't got, and at great expense.

A gear cable is £2.

You're so wrong, it rules. Don't knock it till you've tried it.
A gear cable gets bunged up with water and mud and doesn't change up, even a gore fully sealed one after a bit.
But the front shifting of di2 really is very very smooth. I thought the blurb about the electronics timing the change to coincide with the chain going over the ramps and pins was bollocks but it obviously isn't.

A gear cable won't suffer from flat battery / electrical gremlins!   

[Euan Uzami]

My finger is a mechanical lever. My gear shifter is a mechanical lever. My derailleur is a mechanical lever with a couple of pivots. Replacing what goes in the middle of that lot with a rechargeable battery, microprocessor, electric motor etc is just madness.

Campagnolo put a stronger spring in their pushbutton switches so it would feel like an ergolever. 

I just don't think the idea of a bike that relies on the strength of a spring to pull a cable through its outer is seamless enough for me. Sure it can always be got to work  eventually but it misses changes which riles me.
Road, mountain, shimano, sram, fully sealed, lubricated - they all fail to upshift at the rear at times, eventually.
It might be different if you are the sort of cyclist who is prudent enough to only ride their "summer" bike when it's dry, and their "winter" bike is fixed. Neither of which applies to me

[Euan Uzami]

But the front shifting of di2 really is very very smooth. I thought the blurb about the electronics timing the change to coincide with the chain going over the ramps and pins was bollocks but it obviously isn't.

Okay, points to them for that.

Also for eliminating the trimming of the front shifter problem.

Bung a strain gauge in there and you could delay shifting until the load eases up sufficiently.

All good stuff, but marginal gains at best.

Oh yeah , and the chain NEVER rubs the front mech

Continue denying the advantages without having tried it for as long as you like, you're fighting a losing battle

[Greenbank]

I've come to the conclusion that Di2/eps solves a problem I haven't got, and at great expense.

A gear cable is £2.

You're so wrong, it rules. Don't knock it till you've tried it.
A gear cable gets bunged up with water and mud and doesn't change up, even a gore fully sealed one after a bit.
But the front shifting of di2 really is very very smooth. I thought the blurb about the electronics timing the change to coincide with the chain going over the ramps and pins was bollocks but it obviously isn't.

A gear cable won't suffer from flat battery / electrical gremlins!   

There's an easy solution to bunged up gear cables. (Well, someone has to mention it in every thread about gears...)

[Toady]

I'll let you know what I think once I've figured out what the problem is with my down tube friction shifters

[bobb]

Continue denying the advantages without having tried it for as long as you like, you're fighting a losing battle

To be honest, Ben - you're fighting a losing battle against a bunch of luddites who think the bicycle reached the pinnacle of its design in about 1954 

[orienteer]

I wonder what would happen if one molished one of these into the drive chain

http://www.ebay.com/itm/NEW-GO-KART-MINI-BIKE-30-SERIES-TORQUE-CONVERTER-CLUTCH-DRIVER-1-BORE-/330905160344

Anyone tried?

Torque converters only work at upwards of 1000rpm, rather beyond even Chris Froome's cadence 

[Greenbank]

Has no-one mentioned the NuVinci (or other Continuously Variable Transmissions) yet?