zimmie

To be clear, Telegram is not a "secure messaging app". Group chats cannot be encrypted at all (except by TLS, which explicitly does not protect against Telegram the company). 1:1 chats are unencrypted by default. Even if you opt in to encryption in the 1:1 chat, it has some extremely questionable design decisions which make many cryptographers uncomfortable. One in particular described an issue as "the most backdoor-looking bug I’ve ever seen."

There is a formal proof of many relevant aspects of the current version of Telegram's cryptographic protocol. There is no proof of correctness of their implementations.

agilealtitude said:

Is BlackBerry still a thing? Learn something new every day...

BlackBerry is still very much a thing. They acquired QNX in 2010. It's a real-time operating system used in many cars' infotainment units, a bunch of core Internet routers, and some slightly more exotic industrial hardware.

I personally prefer seL4 or PikeOS, both of which have undergone formal verification to mathematically prove they are free of bugs, but QNX is very UNIX-y. As a result, there is a very large pool of developer talent to draw from when writing software for QNX.

If it's using the same GPU cores as the M1, clocked at the same speed, a 64-core GPU could do 20.8 TFLOPS, while a 128-core GPU could do 41.6 TFLOPS. For comparison, a GeForce RTX 3090 (the top consumer card from Nvidia) does up to 35.6 TFLOPS, and a Radeon RX 6900 XT (the top consumer card from AMD) does up to 23 TFLOPS.

Considering the RTX 3090 and RX 6900 XT still universally sell for more than double their MSRP, I wonder if Apple will have scalping problems. Their system of allowing backorders mitigates scalping, but doesn't eliminate it. With the added demand from blockbros, it may be difficult to get one for a year or so.

AppleZulu said:

zimmie said:

I'm extremely curious about the spatial audio versions. From the statement, it sounds like they're working with a whole separate sample track per channel, so music needs to be specifically remixed for it. Kind of like SACD or DVD-Audio.

Lots of songs dating back to the 80s are mixed as four or five channels, then encoded down to two with Dolby Pro Logic during mastering. An appropriate decoder unfolds the encoded audio back to four or five separate channels. If they gave AirPods Pro and Max the ability to decode Pro Logic, that would open up spatial audio to a bunch of existing media, including a lot already on Apple Music.

Seems unlikely. Studios and artists are going to want surround versions of their music to be mixed for that purpose, not reverse-engineered back up the chain. Mastering is the last step, is done by professionals, and is what makes a recording sound great, rather than just good or amateurish. Even if it were possible to automatically reach back to separate out tracks that existed prior to the stereo mastering, it wouldn't sound right, as no one ever intended that to be the final product. Apple Music's surround content will have been mixed and mastered for that purpose, not automatically tinkered with to simulate something never intended. 

Yes, I'm aware mastering is the last step. That's why I said that's where they take it from the four or five channels of the mix down to two. Look into Pro Logic. It's an intentional choice mastering studios make which lets a track sound good in stereo while still containing positional information which receivers can decode for surround. Specifically, it involves phase shifting the signals for the different channels such that when shifted back and combined at certain ratios built into the standard, they reinforce each other or cancel each other out.

Matrix encoding of quadraphonic sound down to stereo media, then decoding for quadraphonic playback has been around since the 60s. It's not a way to "simulate something never intended", it's an explicit intent by the mastering studio.

CarmB said:

bonobob said:

My ancient ears don’t really care about lossless audio.  They just aren’t capable of hearing the difference anymore.  Dolby Atmos, on the other hand, should be noticeable and much appreciated, especially when listening via Apple TV and my surround system. 

Don’t be so sure about that. There is a misconception that it’s about frequency range but it’s more complicated. If you’re ancient ears have a reduced ability to discern detail in sound, that’s one thing but if it’s the usual, namely losing the ability to pick up higher frequencies - myself, I can’t really hear anything above 13k hz - then a better grade of sound file still matters. It’s less about how high can you go and more about the quality within the range you are capable of picking up. Not saying the difference is so clear as to be undeniable but I would not assume that being older means it’s off the table to get something out of a higher quality audio format. 

To the best of my knowledge, nobody has demonstrated the ability to differentiate between 256-kbit AAC and lossless audio in ABX testing when the sources are sample-synchronized and controlled for volume. I don't pay too much attention to audio specifically, so someone may have managed it since I last checked.

44.1 kHz sample rate can perfectly reproduce any signal up to 22.05 kHz, and 16 bits can perfectly reproduce any sound quieter than a chainsaw a meter from your head.

I'm extremely curious about the spatial audio versions. From the statement, it sounds like they're working with a whole separate sample track per channel, so music needs to be specifically remixed for it. Kind of like SACD or DVD-Audio.

Lots of songs dating back to the 80s are mixed as four or five channels, then encoded down to two with Dolby Pro Logic during mastering. An appropriate decoder unfolds the encoded audio back to four or five separate channels. If they gave AirPods Pro and Max the ability to decode Pro Logic, that would open up spatial audio to a bunch of existing media, including a lot already on Apple Music.