LOL
Yes it is.
For me the argue is over.

Is that supposed to be an argument? Since you disagree with my statement

you must be able to break any problem instance in that domain. How about I give you a RSA4096 public key and you factor it for me?
Quite obviously, you won't be able to do so, and I don't think anyone can arguably disagree with that (without trolling).

@mcp
You can not proof it because nobody knows which attacks appear tomorrow. Just suppose in 10 years quantum computers appear. And almost all current crypto would be trash.
Years ago DES looks uncrackable. Nowadays it is weak.
The same to RSA. Ten years ago RSA-512 was strong. Now it is weak.
Could you make the RSA less or equal to 512 bits which we can not crack? Sure, you cant. Most algos add more rounds to be stronger or increase key sizes or other params.

HE libraries are very rough. Limited in the operations to Add and Mul in most. Also, it is hard to imagine the use cases which help to protect applications. Could you describe any?
To operate in HE you need both numbers encrypted with private key. To decrypt result you need also public key. Would you store both keys in the software? Or how do you plan to make protection?

Of course there is exactly one crypto scheme which is provably secure against any attack (OTP) but I was just arguing against the claim of being able to break any instance of those problems. And that's obviously not true.

Yap, there's a reason why not "everything" just simply switches to (F)HE schemes. There are multiple reasons: a) it is slow as hell b) full HE isn't trivial, most libraries limit themselves to addition and/or multiplication as you said.
Take for example the use case that you want to compute something which must not be revealed to the public, still the computation has to be made on every consumer's device, and the consumers must not know how the computations inner workings look like.
Then again, the weak points of FHE are the input and output values: if these are to be used in other non-HE parts of the program, these clearly must be decrypted.

As always in security, you have to be aware of the "attacker model": FHE per se cannot be used to create any kind of "unbreakable" protection, and no sane person would ever claim that. On the other hand, I strongly disagree with the statement that "everything made by man can be broken". That's too broad of a statement and is simply not true in general.

Sure, it can be. But nobody guarantees it will work after that or even be valid. :D:p

Whilst I agree, I don't see how that is relevant to software protection. Your analogy is like saying I let anyone download the fully registered version and say its uncrackable as the executable is encrypted by a 4096-bit RSA private key. Sure, it will take many years (maybe longer) to 'crack'.

However, make a piece of software runnable only with a license file, protect the license file with an RSA 4096-bit private key and I guarantee you it will be broken and fully registered versions available within 24 hours.

Same goes for if the license checking is built into a dongle. If you have access to the dongle, the software be made to work without it. I have done this many times for people who worry about the security of their software dongles.

How about Senselock and other dongles which are built on smart card technology?

If you read my answer in the context of the original claim that "anything made by man can be broken by man", then my answer makes sense again. It was simply a counter example to that claim, not necessarily related to copy protections in generl.

Hi mates,
How would you rank latest SolidShield? Thanks!

What's some modern software that uses it? It pretty much faded into obscurity after AAA games dropped it sometime over a decade ago
Did they develop an x64 protection? Sometimes it's a huge step back

It's pretty lame protection compared to today's standards. Can be cracked quite easily.

I believe Safenet Sentinel LDK protection - its very hard to unpack files protected by this software and maybe few has ever to do it but I never know someone.

True. Only the paid ones post the solutions these days for the Sentinel LDK. No free tuts!

I think Wibu CodeMeter can also occupy a place，it can be the best one.

It is the worst one in dongle. See how many codemeter apps are cracked and released on regular basis. I agree that it could be implementation problem in some cases, but still...