tpolzer

9 ms·

61.
by tpolzer 6y ago
If you just want network between them, you should actually be able to connect them directly with a thunderbolt cable, no need for Ethernet in between.
62.
by tpolzer 6y ago
Key distribution is a problem, yes (as discussed in other comments here). Upload bandwidth isn't so much of a problem with the number of participants, as you of course wouldn't encrypt the whole data stream separately for each pee
63.
by tpolzer 6y ago
If you write C++ like it's Java, it will be slower, mostly for two reasons: - A C++ compiler cannot do speculative devirtualization. Which hurts if you split everything into small virtual functions. - If you are churning through lots o
64.
by tpolzer 6y ago
Just because 20ms is the default doesn't mean that it has to be that way. The chart you linked shows a big "bubble" for opus for a reason. Opus minimum frame size is actually 2.5ms: https://tools.ietf.org/html
65.
by tpolzer 6y ago
If all participants have good internet and are geographically close it should theoretically be possible to have delay not much greater than rtt/2 for everybody. With rtt < 20ms that should make musical performances possible. After a
66.
by tpolzer 6y ago
Indeed, the author is confusing things here: - It has a vastly smaller attack surface than e.g. OpenVPN, because it is much less complex. - Its performance is improved by being kernel based. - Compatibility is helped by it being in the main
67.
by tpolzer 7y ago
That's a typo on the diagram. The R1000 processors have Gen2 built in (and you always need both USB 3 and 2 - which explains the two arrows in the diagram).
68.
by tpolzer 7y ago
AES GCM should take insignificant amounts of power compared to eink panel refresh and make this secure enough (unless there's no way to reprogram the mcus in there?).
69.
by tpolzer 7y ago
It cannot do the same thing as an a6000, as it lacks a physical shutter.
70.
by tpolzer 7y ago
On the one hand that's correct. On the other hand you could completely avoid that if you did long exposures the way a Google Pixel (2+ I think) does it (in software): Take lots of short exposures and fuse them. If the device is handhel
71.
by tpolzer 7y ago
It has nothing to do with being a "Malthusian loony". Which of these do you prefer: - >10 billion people living on this planet, depleting resources so that inequality levels will rise while total (real) wealth will decline. - A
72.
by tpolzer 7y ago
Anything else that gets too popular will be aggressively blocked unless they comply with the censorship/give out logs.
73.
by tpolzer 7y ago
Have you actually used old Mono? I remember sluggish performance, especially of its stop the world garbage collection.
74.
by tpolzer 7y ago
What's the attack model under which this is better than what Google Photos does? - You still have to give the people who access the photos the key, they can share it (just like the secret link) - By entering it on the site, they also i
75.
by tpolzer 7y ago
It shouldn't matter _if_ all you're doing is reading. If you boot Linux, there's a ton of things that do small writes and then call fsync. That won't do anything on a RAM disk but is relatively slow, even on an SSD.
76.
by tpolzer 7y ago
Chrome is in an exceptionally hard place because of its JIT. Your language cannot tell you if it's safe for your JIT to omit a bounds check.
77.
by tpolzer 8y ago
It doesn't need to, but it would suffice in case your GPU doesn't manage.
78.
by tpolzer 8y ago
You cannot do syscalls, but you can read/write all memory, which means you can make some arbitrary (trusted) process your victim, executing code in a ROP manner and antivirus won't see a thing.
79.
by tpolzer 8y ago
There's a very big C inherited footgun you can avoid with signed sizes: implicit conversions to unsigned on comparison.
80.
by tpolzer 8y ago
You could say that there's no reason we're still using a stupid preprocessor hack of textual inclusion to import interfaces. As usual, the devil lies in the detail. If you want to make pragma once robust you need to checksum files
81.
by tpolzer 8y ago
At the cost of predictable performance though. With current implementations, the exceptional path will be several magnitudes (!) slower for cheap functions.
82.
by tpolzer 8y ago
I did some evaluation of (Linux based) OpenCL implementations two years ago (but I don't think anything changed significantly since then). I had a few takeaways: - OpenCL on GPUs and CPUs have little to do with each other (in terms of
83.
by tpolzer 8y ago
You're overpaying. Wingo will give you very good network coverage, a gigabyte of data per month and free calls for 25$ a month.
84.
by tpolzer 8y ago
Usually, offloading to the GPU will get you higher throughout but also higher latency. Although of course this assumes a sane software renderer, X11 API calls will take forever.
85.
by tpolzer 8y ago
Duplicate of https://news.ycombinator.com/item?id=17581988
86.
by tpolzer 8y ago
In a surprising move, the stock price disagrees with you.
87.
by tpolzer 8y ago
Have you ever compared Vanillin to real fresh Vanilla yourself? Yes, it replicates a large part of the flavor, but certainly not all of it.
88.
by tpolzer 8y ago
But that is exactly how the market would look from the consumer side if the airlines had perfect information.
89.
by tpolzer 8y ago
It's better than your strategy.
90.
by tpolzer 9y ago
If you really want that, just do a relaxed read into a local variable, increment there and relaxed store afterwards. At least clang will give you exactly the assembly you want, minus the technically undefined behavior: https://go

More ›