is signal processing comfy?
it looks pretty comfy
That looks like witchcraft to me, desu fampai.
>>54483577
it doesnt just plug everything in matlab and bamp your done
>>54483565
Signal processing is easy thanx to matlab's ident
Also you don't really use Fourier that much..in real life.GTFO of university.
>>54483636
You've never had to measure noise coming out of a circuit?
FFT is so important, any decent oscilloscope has a fucking button to turn FFT mode on.
>>54483636
EE who fails for the signal processing meme master here. Is it true that they're starting to standardizing everything soon and this field is becoming obsolete?
>>54483665
I never said they aren't important..but you don't really need to do the calculus yourself.
lmao looks like some shit that sheldon would do on the big bang theory
>>54483702
its actually really really easy, you just need to plug and chug using a table. Calculus is much harder.
>>54483682
Well, no, but it's very important to know about it to implement FFT efficiently in any algorithm. Sure, you still use FFTW or whatever in the end, but using it in the most naive way and actually being clever about what you are doing makes a huge difference in performance. Seriously.
>>54483728
Calculus is literally babby math and you get by just fine by memorizing that shit.
>>54483702
fourier is p easy and actually quite powerful in physics so yeah
>>54483785
Personally i hate signal processing...I studied signals and systems and this thing i hated it soo much.
I love identifying systems ,computing regulators etc..put this is damn messy.
>>54483785
Thats what he's saying, faggot. You dont do the calculation irl knowing how its done is enough. Fourier is much easier than it looks and you're not tough showing off how scary dem big symbols are.
>>54483565
confirmed terrorist.
>>54483873
indeed Fourier translates in a lame sum and all you have to know is the form of an and bn
>>54483873
I'm not OP.
I'm not sure what you are saying really, but knowing how to do those calculations and understanding the algorithm sufficiently to do anything with it is the same thing.
>>54483944
understanding the algorithm isnt the same as crunching these symbols all day to get the right answer
>>54483999
Are you somehow scared of theory? "These symbols" are just a convenient way to write it down so you can work with it, Jesus Christ. Sometimes you need to get back to that level to get worthwhile algorithms.
>>54484051
>Sometimes you need to get back to that level to get worthwhile algorithms.
No you dont. Enjoy teaching in community college kek.
>>54483565
I would not call it comfy, but I got pretty impressed more than one times studying it. Fourier is magical and ?numeric? (not sure 100% for the word) filters are just so fun to integrate. Currently doing image processing (computer vision) but I do miss signal a bit since I finished my degree.
I don't understand why we haven't switched to psuedocode for demonstrating math equations.
It's much easier to read and much easier to implement in code because 99% of psuedocode examples are already valid C code.
>>54484081
How would you define a limit, for instance, in pseudocode?
>>54484071
So you are basically admitting that you've literally never done anything more complicated than a few simple exercises.
Really, thinking about what exactly happens when using FFT can make the difference between "Takes two hours to compute" and "Twice real time speed" and that is absolutely important in a lot of applications. It helps A LOT if you know a bit of theory (and really, just because it's theory and involves integral symbols doesn't make it complicated).
>>54484081
>much easier to read
are you being serious? have a think about that again
Radio noob here. How do i learn to use gnu radio companion? What is the best book to get started in digital signal processing.
>>54484137
No you're just dumb and cant read.
'Knowing a bit of theory' doesnt translate to 'doing the calculations when it actually matters' like the poster above said and you tried to refute.
>>54484081
Because most of the people who do math are not impressionable CS students
Because most people actually need to do more with math equations than stupidly implement them
Because it's usually cumbersome to write pseudocode compared to math equations
Because not every math formula is meant as a numerical calculation, but rather as something abstract that can't be calculated numerically.
>>54484180
First a basic understanding of advanced calculus topics is required
>>54484187
Now you are just splitting hairs. Something about math seems to trigger you and I don't give a shit. It's important as hell for implementing signal processing algorithms, I know that. Cut your semantics bullshit and shut up.
>>54484081
>It's much easier to read and much easier to implement in code because 99% of psuedocode examples are already valid C code.
But that's fucking wrong.
>>54484208
Ok, would kahn academy suffice? Any specific topics i should focus on?
>>54484232
You dont know shit nerdlord, just reiterating what your jewish professor bullshitting you to earn money.
>>54484252
Don't do khan, you want patrickjmt
>>54483865
Zimbabwe
>>54484081
How is a 12 line algorithm, that is most likely a approximation anyway, easier to read than a one line equation?
>>54484071
Kek, this is sophomore level calc. Enjoy your BA and Starbucks job
>>54484081
All you had to do was suggest modernizing the symbols in math and people would agree. Instead you went full retard.
>>54483565
Back to /sci/
>>54483565
Take your nerd shit to >>>/sci/
