All targeting solutions for sublight speed are computable.
It takes a hell of a lot more than a single clock cycle to run a targeting algorithm.
this post was submitted on 29 Sep 2023
174 points (92.2% liked)
NonCredibleDefense
3528 readers
57 users here now
Rules:
- Posts must abide by lemmy.world terms and conditions
- No spam or soliciting for money.
- No racism or other bigotry allowed.
- Obviously nothing illegal.
If you see these please report them.
Related communities:
founded 1 year ago
MODERATORS
Yep, around 200. The Missile meanwhile travels 0,2mm
Source? I'd estimate thousands to millions of cycles.
Not everything has to be computed in every computation cycle. Most things are already pre-computed and the operations to do the corrections would mostly be simple additions and multiplications.
Written in hand coded assembler language: Maybe 100 cycles. Actually Quake from 1997 used a very similar prediction for network coding and not joking, it took less than 60 cycles and the very same code has been used to guide the IRIS-T system - which by the way uses an 1.4Ghz ARM Quadcore.
Written in Dotnet: Your computer is to old. No matter which computer.
So they wrote Starfield in .net? /j
hypersonic during launch phase only, at the target it's more like Mach 3.5 if we're talking about Kinzhals downed by Patriot some months ago
TIL hypersonic is genuinely a distinct word from supersonic.
It's usually over mach 5, but originally it was used to refer to when you start needing to deal with heating up the air.
Also around the point where jet engines kinda just stop working right.
It's not about the targeting computers not being powerful enough. It's about the counter missile not being fast enough and the hypersonic missile being able to zig-zag like a cruise missile, but with similar speeds as ballistic has in it's decent phase.
They don't need to be as fast. They're not going in the same direction.
Then how does it react when the missile is going 3x as fast and sweeps right when the counter missile is sweeping to left towards the missiles previous position and misses, in that case it needs to be faster than the incoming missile itself. Yes, the speed would not be a problem towards a predictable trajectory, that's how ballistic missiles are intercepted even if they go super fast. it's basically a high school math problem in that case to calculate the point of interception in a firing solution. It's also fine if missile can change course, that's how cruise missiles are shot down, because the counter-missile can still race with them when they turn, but when the missile is fast and can change directions mid flight then it doesn't much help how fast the computer calculates if the hardware can't react fast enough. it's basically like if your mind were able to move at superhuman speed but your body is still human and you get shot and only thing you can really do is to watch the bullet approaching but being unable to dodge fast enough.
The issue with hypersonic missiles is that they go so fast that the air around them heats up enough to become a plasma. This prevents radio, microwave, and infrared based comms and radar systems from working. So any proper hypersonic missile is stuck with using either optical sensors (which require a ton of processing power and is slow) or inertial sensors (which aren't very accurate due to drift) in order to plot its course.
The Russians are the only ones with "hypersonic missiles" because they're using a different definition. Theirs are just ballistic missiles that move at hypersonic speeds, which by definition can't change course and have a huge IR and radar signature, so it's not that difficult to set a slower patriot missile to be in it's way to intercept.
Slower objects are more nimble. They don't have to fight inertia.
I can throw a rock at a moving car. It can swerve, but I can easily throw another and it can't swerve again quickly.
> hypersonichu
> ziz zag
Kek
okay, but for a missile 30km away radar signal gets 0.2ms to get there and back, that's like 40cm of missile movement
Just need to fire at a target 1.3 feet away from the launcher.
Checkmate Ukrainians
This 2hgz SHARC processor with redundancy, what is it's name?
Just to mention: The current MARS rover is using a Raspberry Compute Module 3. Several modern space probe also use off-the-shelf technology. Most Ukrainian drones use it too. The IRIS-T missile uses a 1.0Ghz Quadcore ARM. The times when such systems required extreme technology are far gone.
What? This is one of the most ridiculous things I've read in awhile and people are up voting this misinformation. No rover that's on Mars is using a rpi. Here are the actual specs: https://mars.nasa.gov/mars2020/spacecraft/rover/brains/
I googled around to see where you could have gotten this information from. You might be thinking of this educational rover: https://github.com/nasa-jpl/open-source-rover/blob/master/README.md
I tried looking for the Mars Rover that is using a compute module and couldn't find any that were using one which makes sense because it's not radiation hardened and they need to operate well outside of the temperature range that the modules actually tolerate. They seem to all be using a radiation hardened version of an IBM power PC 750 that is manufactured by a company name BAE. Ignoring that fact though the Mars Rover does not meet the same low latency processing requirements that this meme is referencing so we can move along further.
"Most Ukrainian drives use it too" while I can't quantify these numbers it does appear that the Ukrainian military uses orange pi's for their "kamikaze" drones which is good for that application as it's easy to source and doesn't require any heavy demand. As for everything else I was able to locate on the subject, it all appears to be typical DSP processors Analog Devices SHARC's and the TI equivalents such as the TMS320F28335 As for the Iris-T I could not find any information one way or another there. It's indicating high velocity and use of radar so I can't imagine they are using anything other than a DSP processor as you are going to need to shave every single microsecond of response time to maximize your effectiveness.