Question, what stops you from modifying the GPU driver to make it so the command to set a draw-context to be DRM'd is a no-op? That way the gpu would happily let you snapshot the context, and the OS would be none the wiser. Or is gpu driver tampering something nvidia explicitly designs against?
EDIT: oh wait, i bet the driver is signed and the OS checks it and only allows DRM content to play if the driver is signed by nvidia. That would make sense.
>I'm confused. How does a home invader know if I am armed?
They don't, but they if you're _not_ if you live in a gun-hostile nation. And knowing you might be is a deterrence, for the same reason knowing you might get caught by the police is.
>How is gun ownership supposed to prevent theft or violence?
Well, because of the risk of being killed if you try to commit a crime against someone with a gun. The cause-effect there is pretty clear.
Can you simply just not pay? What are the legal consequences in that case? Effectively some random 3rd party is requesting you pay them for something you didn't ask for or contract them to do.
Sure, you think that, but then the one time it does happen and you don't give them money and someone gets serious hurt or dies, you never forgive yourself ever again. And out of worry that that may happen, a lot of people will be safe rather then sorry. Since money is less important then lives.
No. The Supreme Court exists primarily to interpret the law, reconcile laws with other laws, and address vague language or other inconsistencies. Overriding the will of senators, immediately after a bill’s passing, is very rare and threatens their perceived legitimacy. Senators can always impeach SCOTUS judges, and to do such an act begs for it.
Considering Congress passes almost 500 laws per year (1229 from 2019 to 2021, 1234 [sic] from 2021 to 2023), it is a little less than 1% of cases.
But that’s just the federal branch. Your list includes states and local municipalities - which makes the real number much, much lower. Much closer, arguably, to 0.1% or less of all legislation.
> Much closer, arguably, to 0.1% or less of all legislation.
You're missing the giant flip side of that coin; that local, state, and Federal courts can all determine laws conflict with other laws or the Constitution, and do so regularly.
If your town passes a "no black people allowed after 10pm" law, it'll never get to SCOTUS, but it's still unconstitutional. The first level of courts it encounters will immediately overturn it.
(Two other notes: clearly unconstitutional laws tend to be a bit rare, because they're a bit embarassing when they get readily overturned. A significant portion - as much as 20% - of those 500 average bills is naming post offices; for example, https://www.congress.gov/bill/117th-congress/house-bill/203. Much of the rest are similarly void of thorny constitutional questions.)
The fact remains that it is entirely within the role of the court to overturn unconstitutional legislation, which the court does regularly.
The idea that the court would not give proper scrutiny to a legislative act -- one which may violate the first amendment, fifth amendment, and may amount to an illegal bill of attainder -- just because it had support from the senate, well, that's just preposterous.
Anyone who is legally knowledgeable know how this works? The text says "It shall be unlawful for an entity to distribute..a foreign adversary controlled application by carrying out..internet hosting services to enable the distribution", and it specifices that this includes source code. But https://en.wikipedia.org/wiki/Bernstein_v._United_States specified that the publication of source code was protected by the first amendment, so the government would need a compelling reason to prevent that publication. Is there a conflict here? Is preventing the publication of an apk likely to withstand court scrutiny?
EDIT: My best guess currently is that the government is claiming there is a compelling state interest, that of national security, and that is why they should be allowed to ban the publication of this app and it's code. And even if it were to go to the court the courts don't like telling the government was is and isn't national security related, so they would probably just ok this.
Reading this story and doing some googling sadly makes me pretty confident that this linked one (about boat jousts) is fake. The ending details an explosion and there is no news article of that anywhere in the specified time period (which he says is a few months before the invasion of Iraq). Would love to be wrong though.
Does that tract also eventually lead towards increasing corruption in the long run, though? By not punishing corruption, then it attracts those sociopaths who want to do bad deeds and get away with it. They know no one is going to fight back for exactly those reasons you gave.
I'm really disappointed this doesn't include technical details, especially the bit about "The game can not run without the token, as it is used to e.g. decrypt certain values at runtime and similar things."
Is the game installed encrypted? What encryption schema allows you to decrypt the same set of data using multiple different keys in this way?
If the game isn't installed encrypted it seems easiest just to not allow it to encrypt itself before running.
My vague understanding is that Denuvo generates many redundant diverging paths through the executable, which are chosen at runtime based on pieces of the users hardware signature. The keys you get from the server only work for the paths your hardware is supposed to follow, so if you take those keys to a different machine it will branch in different ways and crash when it's not able to decrypt that path. That redundancy is probably why the protected executables are so large.
That makes sense, but how are these paths encrypted? If it's at runtime, you should be able to just not encrypt them. If it's at install time, do they generate a custom encrypted exe just for you per-download?
It's all done ahead of time when the game is packaged by the developer. The executable you download contains every encrypted path that might be executed, but the authentication handshake only gives you the keys for a subset of those.
If it were possible to fully sandbox an exe, so that it will always be presented the same hardware/system configuration, would that function as a universal denuvo bypass?
From my understanding of looking at denuvo, yes, but at the moment that would require something close to full system emulation -- it's hard to "hide" everything about your 3d hardware in particular, because engines often include different code depending on your exact card, it's GPU power, amount of memory, etc. Emulating a "standard" 3d card would introducing a fairly huge overhead.
Feels within the cost reward for comercial pirates to activate a few copies of the game across every model of GPU. Full emulation with a GPU passthrough...
Ohhh, gotcha! How do they prevent replay attacks? Assuming you were only interested in cracking the game on your specific hardware, could you not just record the decryption key you get and use that offline? Then in theory you could just enumerate every hardware combo and aquire all the keys.
Question, what stops you from modifying the GPU driver to make it so the command to set a draw-context to be DRM'd is a no-op? That way the gpu would happily let you snapshot the context, and the OS would be none the wiser. Or is gpu driver tampering something nvidia explicitly designs against?
EDIT: oh wait, i bet the driver is signed and the OS checks it and only allows DRM content to play if the driver is signed by nvidia. That would make sense.
reply