Under the assumption that 41kb space for transaction with zero fee should be enough, it does not break CPFP: The miner can throw out other transactions to make space for all parent transactions he needs. Please note: The incentive for miners is not to have 41kb filled with zero fees. The incentive is to fill all space with highest possible fee per byte transactions. (if necessary we could change the constant in the formula from 1.1 to 1.2 that would modify the space for zero fee transactions to 79kb)
 
Regards,
 
Gesendet: Montag, 08. April 2019 um 00:11 Uhr
Von: "Bernd Jendrissek" <bitcoin@bpj-code.co.za>
An: simondev1 <random@gmx.ch>, "Bitcoin Protocol Discussion" <bitcoin-dev@lists.linuxfoundation.org>
Betreff: Re: [bitcoin-dev] new BIP: Self balancing between excessively low/high fees and block size
On Sun, 7 Apr 2019 at 17:45, simondev1 via bitcoin-dev
<bitcoin-dev@lists.linuxfoundation.org> wrote:
> ==Implementation==
> Sort transactions by FeeInSatoshiPerByte (lowest first)
> For each transaction starting from lowest FeeInSatoshiPerByte: Sum up the bytes of space used so far. Check if summed up bytes of space used so far is smaller or equal than the formula result.
> If this is valid for each transaction then the blocksize is valid.

Doesn't this break CPFP? I think to avoid that you'll need to rework
your proposed algorithm to treat chains of transactions as a group.
(And note that you could have multiple transactions in one block that
depend on the same "parent" transaction, also in the same block.)