:-) I know what inference is. It's just that the speed of inference very much depends on the model you're doing the forward pass on, and the phrase "inference can be run at 10fps" is non-sensical without also specifying the model.
That's not the case. Today's translation and speech recognition systems are considerably larger than even the beefiest phones can sustain. There are some simple OCR models and word (not phrasal) translation systems that run on the device, but not speech recognition.
Could be. But I bet this is just a pet project that a couple of engineers do in their spare time, so I wouldn't expect too much from it. I just don't see this as something Larry would care about one iota. And things that Larry doesn't care about tend to eventually die at Google. Case in point: Social.
Since the article doesn't have any information whatsoever, here's my prediction: by "bringing AI to Raspberry Pi" they mean being able to call their cloud APIs from there.
TensorFlow is not suitable for anything practical on the Pi. You can certainly get it to run there, but CPU vector math on resource constrained devices is not going to be a forte for a framework designed primarily for quickly iterating over models on a GPU workstation or a multi-GPU server. TF very much likes to have a very beefy GPU.
Yes, my interest is basically at a level slightly above the OS, not bare metal embedded or OS development per se. And the OS (at least Linux) will tell you which features a given CPU supports. It's just that the variety on the 32-bit side is so enormous, it's hard to be sure of anything. I.e. your particular chip might not even have hardware FP. With aarch64 even the lowest common denominator is better than just about any arm32 (modulo the specialized extra hardware the particular chip might have).
I thought that too initially, regarding the takedown procedure. There are even videos on YT of people shooting their walls. Nowadays removing the mag and racking the slide is muscle memory. It also happens to be the only way to make sure the gun is well and truly safe.
Glocks have triggers that are pretty heavy, and you don't have to carry with a round in the chamber unless you're in _really_ bad places. Safety has its own issues: what if you don't completely disengage it or forget to do it at all? What if you forget to engage it and do something stupid because you thought your gun was on safety?
Well, I'll admit I haven't yet tried the P320. I have no doubt Sig can make a great gun, especially if they decide not to overengineer the heck out of everything.
IMO it's a mistake to develop a board that's not capable of running 64 bit ARM code these days. While you might not need the address space capabilities, aarch64 standardizes a lot of things that would otherwise bite you in the ass if you develop software that needs to run on several flavors of ARM. I.e. all 64 bit ARMs have NEON and hard float. And not just any NEON, but NEON specific to 64 bit ARM, with newer instructions. You get a _very_ nice lowest common denominator to code against. Nicer, in some regards, than Intel. This is especially egregious considering that Raspberry Pi 3 _does_ have 64 bit chip, even though they run it in 32 bit mode. I guess what you're really buying in this case is a hardware 4K video decoder.
Why not Glock, I wonder. If you're going to go striker fired anyway, Glock is the Kalashnikov of handguns. Simple as can be, and just keeps going no matter what. I have a Sig P226, it's a fine gun, I'm much more accurate with it (shooting stationary paper targets), but if you asked me what I'd take to a war zone, I would take a Glock 17 or a Glock 19. Much less of a chance that it'll crap out on me at the worst possible moment, which is an important feature, when your life is on the line, IMO.
Yes, the covered surface is much larger. Frankly I don't get why MS insists on keeping Windows closed source. I get why you'd want to keep something others might be able to easily reuse closed source, but I don't get why Windows developers don't get to see Windows source code. In fact, having worked with Linux extensively for the past decade, I don't think I'd be comfortable doing serious work on a system where I don't have source code for debugging and documentation.
I see. "I refuse your reality and substitute my own". If you can't see that Gigapixel shows radically more people, then I'm afraid I can't help you. :-)
I've done the opposite. For me it doesn't matter how fast the uploads are (within reason), it matters whether I can restore reasonably quickly if something craps out. CP does have a disadvantage though: their client is written in Java and if you have a lot of files it takes up a ton of RAM.
He will be out in 2 years or less. The problem with Microsoft is it has a massive old boys network. Once Ye Olde Microsoft mafia moves in, people who haven't been with the company for 15-20 years tend to move out, irrespective of their position. This is also why all Microsoft acquisitions where old boys network was allowed to intrude (basically all but Skype) have been promptly run into the ground within 2-3 years after acquisition.
At the moment I'd pay a lot of money for a car that I drive mostly myself, but that also had an advanced safety system which makes it much harder to get into an accident. I.e. fix my driving mistakes and lapses in judgment for me while I'm in control. I'd be willing to pay a hefty premium for this.