Hacker News Reader

Comment by hulitu | Hacker News Reader

> it is suspected, Potential, scientists at Harvard have now invented and simulated

So another hype. I thought IEEE are serious people eveb if they charge a lot of money for their publications. You can simulate _anything_. All you need is a simulator and a model. Reality is a bit different.

View on HN

God no they are a shit show. Having been an EE years ago I stopped reading any journals and research. 99.9% of it is garbage which is engineered to sell attention rather than a better future.

Anything worth giving attention to turns up in EDN when a product comes to market after someone took it serious enough to actually invest in it. Then you pay attention.

My favourite recent thing is chip scale atomic clocks: https://www.microsemi.com/product-directory/clocks-frequency...

You can actually buy them!m

Edit: the worst garbage is actually IET’s E&T Magazine. That’s horse dung.

I didn't really understand IEEE until I realized that they're more of an academic society than a professional society, despite what their marketing suggests. The academia mill is a weird way to fund training engineers for industry, and IEEE is just a forum for it.

> You can actually buy them!

For $5,200.

I wonder how many they sell. They would be useful in things like rangefinding and positioning (and maybe encrypted packet comms).

Sounds like a military market.

They almost entirely are military market. A chipscale atomic clock is really only good for GPS-denied environments whether that means a UAV looking for a jammer or some very small device underwater that doesn't have room for a traditional rack mount time and reference frequency source. Space (civil and military) is another application where you may have a satellite that is not in optimal range of GPS satellites and may need a backup reference. There could be some mining and petro industry applications but again, something deep underwater or underground. Anything other situation and you'd have clear sky for GPS.

Actually they’re $1750 for the module which is quite affordable if your hobby is timenuttery.

http://www.leapsecond.com/

Fair ‘nuff. I probably misread the ordering page (renders like crap on mobile).

In particular you need to know what to simulate, because when you simulate you leave lots of things outside the simulation.

The only thing that proves that you made a good work ignoring those things is prototyping on the real world.

Without that it is just vaporware.

Do you suppose that modellers are unaware of the differences?

I don't think they are not aware. But for fame, they ignore it.

Reply on news.ycombinator.com