The boarding rate of a 3 door SD car is almost identical to a Sydney DD.  The DD doors have 3 channels: vestibule, Up & down.  The SD's have two: left & right.  The S sets and even T sets only have doors wide enough for 2 channels, but the newer sets have wider doors and largely address this issue.

For the same signalling headway and same boarding rate, line capacity is actually lower with SDs.

In the case of Paris,the boarding rate is faster: more doors and more channels, but at the expense of train capacity.  Line capacity is still higher than Sydney, but no by much (~5% after equalling out platform length).

The big advantage of a "new" system is the new signalling, not the "new" train shape.

The one format that does offer greater line capacity is a 4 door metro: where the boarding rate is much higher, and high enough to compensate for lower train capacity.

But in the end it all boils down to what you are trying to optimise for.  If it's minimal unionised staff, the automated metro is the way to go.

But we are not talking about same headway, You won't get 90 sec out of a Sydney DD stock which is the alt proposed to NWRL auto trains. As I said, you'd be lucky to get 2.5min.

Not one technology supplier is offering automated technology with DD stock and if you are limited to 20m, then third door option like RER is not a practical option.

4 door on 20m car length leaves little room for seating, hence why I think they went for 3 door. Its a part compromise.

What are they trying to optmise for is obvious
- De-unionise work force, although less of a driver than in past. But its the 21st century, build a line greenfield and you don't need to justify having an automated train. You need to justify why not having one is a good idea.
- Lower construction costs
- Lower operating costs.
- Better customer service with lower waiting and transit times.

regards
Shane