P2PK is at risk.
The whole Bitcoin would be at risk not just certain outputs. You see if some day an algorithm is found and hardware improves so much to solve ECDLP at reasonable time it would just be a matter of [short] time before they can reverse any public key including the public keys revealed in the transactions waiting in the mempool for a couple of minutes to be confirmed.
i dont think you understand the math
lets use the announced (yeas ago) RSA 768 'crack'
(to easy display the magnitudes needed to get from a 2 year crack to under 4 minute crack(if all went well and budgets of labs were unrestricted)
RSA 786 takes one binary pc trillions of years natively without any special treatment
but factoring in some efficiency algo's and other things. they can get it down to
under 600 years on 1 binary PC
and creating an algo to use ~
300 pc's(288pc of 64bit = 18432bits) brings that down to a couple yearsthen
![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FwXRQpJ0.png&t=665&c=QDj7R7D-Jny0bg)
its not simply going from 64bit to 65bit to 2x factor the efficiency(2 gates per bit)
its not simply going from 64bit to 128bit per PC to 256bit per PC to 512bit per PC to 1028 bit per PC
meaning its not 4 evolutions of binary cpu architecture to get down to a 45 day hack in a single PC
its buying/creating 300pc's of 64bit pc's(plural) for 2 year test
then 600 pc's of 64bit pc's (PLURAL) for 1 year test
where they only had the budget to buy ~300 PC's(~$100k) to prove the 2 year crack experiment. and not over $20billion to prove they can do it in 4 minutes
now do the same math of quantum to get from 8 hours to 2 minutes in a factor 4gate logic
and then do the math of how many QC's are needed
oh.. and once you do the math. realise that simply making multiple QC's of 2048qubit still has algo problems of syncing all QC together to get such efficiency factor
and then. final task. work out the total cost of all them QC's synced together
oh one last thing. if you still dont get it yet
most of the efficiency is not just native bit/qubit counting
its mulltiplying the amount of whole machines and also having the right algo's to make it all work.
and having other efficiency algos such as shor and schnorr to cut down on that time too
oh and a reminder. the new tx format uses schnorr already. so a QC cant schnorr a schnorr, thus it cant use schnorr to gain trick efficiency if a bitcoin format is already using it to stay ahead
thus p2tr is already an efficiency stage ahead of QC making it less easy for QC to break p2tr compared to p2pkh
so once you can compute all that.. sit back have a nice day