>
> Note how you're basically proposing for the block interval to be decreased,
>> which has security implications due to increased orphan rates.
>>
>
> Note that the total transaction rate and block size don't materially
> change, so I don't
> see why the orphan rate will change.  Normal blocks are constrained to have
> all of the txs of the aux blocks, so propagation time should stay the
> same.  Am I missing
> something?
>

Ah, yes, I'm missing that the expected time to find each type of block is
halved, so the orphan rate doubles.