Thank you so much! I've struggled to understand this for YEARS and now it finally clicked after ~30 min. Hands down the best explanation, simplification and metaphors I've heard on the matter.
You are the engineering equivalent of MR FANTASTIC - THANK YOU so so so much. You're video should be played at Uni just before the lecturer sends us all to sleep trying to convolutedly explain Fourier transform.
This is actually a Discreet Fourier Transform. You explain that briefly at the end but it's an important distinction which is easily overlooked, especially if someone skips the technical details at the end. Otherwise, a very nice explanation
Thank you for creating this video. CZcams is full of "smart sounding" people who explain fourier with mumbo jumbo language. After two weeks of constant searching, I have finally found you. You explain it the best. Simply Genius! I can finally now understand fourier with your parables. Please keep posting more of your down to earth explanations. I will go through all of your videos.
dude, i just can't say thank you enough for making this tutorial. After watching this vid, me, who didn't learn math since high school, can program using DFT with intuitive understanding. Thank you so much you make the earth a better place!
You have an incredible talent of explaining complex topics in simple words plus a total mastering of the subject. Brilliant! this video really helped me a lot understand this complex topic. many many thanks.
Thank you so much for this comprehensive video!! I'm using FFT in my thesis and while I was able to comprehend the application of Fourier transform, I was having so much trouble with the theory/background explanation in the paper. This video has helped immensely!!!
Wow, wow this explanation was mind-blowing! You made such a complex topic feel so intuitive. Your teaching style is truly genius and your passion for teaching and clarity in explanation are truly commendable
Thanks Josh - I'd like to do one on LaPlace. The key intuition is that we're using exponential spirals (that both rotate and decay) vs. unchanging circular paths (that only rotate).
30 years on from using the word Fourier Transform in a meaningful sense I came across an analog - digital signal issue which caused a tickle in back of my skull. Yep your vid confirmed a FT is the right answer.
Love the videos, will send this to my son, a freshman at a tech college on the east coast of US. Though, after sending him to China for a couple of weeks, calculating time differences became an issue. So, I respectfully disagree with your use of a timezone analogy. New Delhi, India is ~12 hours ahead of San Francisco, California. That is not really anything like being 12 hours behind San Francisco. The sun may rise in the East, but it has to reach it's final destination in the West. Maybe just stick with "the 2 points cancel one another" or "the average of the points is zero." Thanks again for the intuition! Awesome series, love the periodic e-mails, and keep writing!
This is fantastic, thanks for the explanation. The only area I'd challenge is your starting position with Usain Bolt...I think he'd still win with a 50 ft handicap!
Thank you! I've got halfway through understanding the Fourier transform before, but I think you've carried me the rest of the way. The only thing I disliked was the smoothie metaphor at the start - ever since I was a kid I've found metaphors unhelpful in scientific explanations. I still remember how annoyed I was at high school physics trying to talk about electricity in terms of water going through pipes! Solid explanation though, and I now feel the urge to write some slow ugly code to make it happen in front of me.
Looks great,but will not expand to full screen and all links on the page are dead. Too bad as I was really looking forward to learning this interpretation.
I think the smoothie analogy is a bit confusing. I find it easier to stick to geometric/math interpretations. For anyone that is interested, 3 Blue 1 Brown does a more in depth and mathematically intuitive video on the subject.
Thanks the Intuitive explanation and interactive virtualization ... there is an inconsistency of your equation normalization between deferent sections ...
On the website I noticed the 0 cycle has the value of 1Hz and 1 cycle has the value of 1Hz. cycle ( amplitude) as {0, 1}...... time( frequency) as {1 -1}, i.e. how many times it goes around in for each one second. In the next line below you describe what it all means. You describe the 0 cycle ( amplitude) having 0Hz, and then say 0Hz means it stuck on X axis. What would it mean if you instead described the 0 cycle having 1Hz frequency instead, so it matches the value set for them to begin with? If 0 cycle had 0 frequency, wouldn't it mean that it was stuck at the start position of whatever the phase angle is. And 1Hz means it is stuck on the x axis right through? Say 1 cycle ( amplitude) having 0 Hz. Wouldn't it mean that 1 cycle is stuck at the start position? Also from the look of it the cycle is read from left to right but to correspond them to their frequency on the right, you have to go from right to left..... I can see from the following animation... {0 1 1} {2 -1 -1}. Isn't cycle another word for amplitude? Isn't time another word for frequency? If that be the case then { 0 1 1} mean amplitude 0 and amplitude 1 and another amplitude 1} And the {1 -1 -1} then would mean 0 amplitude is being repeated two times a second 2 Hz. 1 and the other 1 amplitude, both being repeated one time 1Hz...
How would someone like me who knows nothing about math i have memory problems basically pseudo dementia due to depression i lost alot of gray matter but I desperately want to learn this out of extreme curiosity and ocd please would u guide me on what in math to learn because i know basically nothing
A bit off topic but can you tell me how you make your videos where your speaking video, lower right corner- is simultaneously superimposed upon your computer screen- the background.
Exactly, the image is a 2d set of data (which the Fourier Transform can work on as well, though it's often used on 1d audio signals). Fixed up the Fourier Toy link to a version on archive.org.
I am not sure why you are doing all these delay and stuff when the explanation is a simple 0 result for dot product for non matching frequencies with sin and cos component and reconstructing the wave back from using Euler's identity.
Thank you so much! I've struggled to understand this for YEARS and now it finally clicked after ~30 min. Hands down the best explanation, simplification and metaphors I've heard on the matter.
You are the engineering equivalent of MR FANTASTIC - THANK YOU so so so much. You're video should be played at Uni just before the lecturer sends us all to sleep trying to convolutedly explain Fourier transform.
Well done! I'm an ME and this is helping me a lot with a signal processing project. Thanks!!
Can u help with mri machine
You do a good job of presenting a complex topic. Intuitive, clear, and digestable.
This is actually a Discreet Fourier Transform. You explain that briefly at the end but it's an important distinction which is easily overlooked, especially if someone skips the technical details at the end.
Otherwise, a very nice explanation
Your "Discreet" is actually spelled "discrete".
Thank you for creating this video. CZcams is full of "smart sounding" people who explain fourier with mumbo jumbo language. After two weeks of constant searching, I have finally found you. You explain it the best. Simply Genius! I can finally now understand fourier with your parables. Please keep posting more of your down to earth explanations. I will go through all of your videos.
Here from 3b1b recommendation.
lol, I went the other way. I'm guessing you mean this one: czcams.com/video/spUNpyF58BY/video.html
Both are awesome.
dude, i just can't say thank you enough for making this tutorial. After watching this vid, me, who didn't learn math since high school, can program using DFT with intuitive understanding. Thank you so much you make the earth a better place!
This is the best explanation of this on youtube. Thank you.
best explanation on the entire internet that i've seen so far
Amazingly clear explanation, you are professor material mate!
This is the best explanation of Fourier Transform that I have heard. Thank you very much!!!!
You have an incredible talent of explaining complex topics in simple words plus a total mastering of the subject. Brilliant! this video really helped me a lot understand this complex topic. many many thanks.
Thanks a lot for your fantastic explanation!!!
Thank you so much for this comprehensive video!! I'm using FFT in my thesis and while I was able to comprehend the application of Fourier transform, I was having so much trouble with the theory/background explanation in the paper. This video has helped immensely!!!
Wow, wow this explanation was mind-blowing! You made such a complex topic feel so intuitive. Your teaching style is truly genius and your passion for teaching and clarity in explanation are truly commendable
you've got incredible timing, I was planning to try and knuckle down and figure out how this works this weekend
that Homer Simpson visualisation changed my life
can you do Laplace next please?
Thanks Josh - I'd like to do one on LaPlace. The key intuition is that we're using exponential spirals (that both rotate and decay) vs. unchanging circular paths (that only rotate).
@@betterexplained wtf
30 years on from using the word Fourier Transform in a meaningful sense I came across an analog - digital signal issue which caused a tickle in back of my skull. Yep your vid confirmed a FT is the right answer.
incredible explanation, really appreciated all the aspects
Very good insight with quite enough clear and simple explaining method for a quite complicated subject. I found it quite helpful, thanks
Thank you so much. You explain the topic very well.
This was amazing!. Love the analogy, very helpful.
Love the videos, will send this to my son, a freshman at a tech college on the east coast of US. Though, after sending him to China for a couple of weeks, calculating time differences became an issue. So, I respectfully disagree with your use of a timezone analogy. New Delhi, India is ~12 hours ahead of San Francisco, California. That is not really anything like being 12 hours behind San Francisco. The sun may rise in the East, but it has to reach it's final destination in the West.
Maybe just stick with "the 2 points cancel one another" or "the average of the points is zero."
Thanks again for the intuition! Awesome series, love the periodic e-mails, and keep writing!
Excellent work. Thank-you for your time and knowledge.
THE COMMENTS BELOW SAYS IT ALL. YOU HAVE CHANGED LIVES.
Very well done. Please keep up the good work!
Yesss, finally a good explainer, thank you man great smoothie analogy
Wow dude, love your website and your work
Essentially this was quite helpful essentially
so essentially, what you're saying, essentially, is that, essentially, a function, is essentially the sum of a bunch of sinusoidal bits, essentially
This is fantastic, thanks for the explanation. The only area I'd challenge is your starting position with Usain Bolt...I think he'd still win with a 50 ft handicap!
This video is simply insane
Thanks for the lessons
Thank you, the topic is well explained.
Can you suggest the book which explains this topic in detail, probably from scratch?
Awesome explanation, thank you!
Thank you! I've got halfway through understanding the Fourier transform before, but I think you've carried me the rest of the way.
The only thing I disliked was the smoothie metaphor at the start - ever since I was a kid I've found metaphors unhelpful in scientific explanations. I still remember how annoyed I was at high school physics trying to talk about electricity in terms of water going through pipes! Solid explanation though, and I now feel the urge to write some slow ugly code to make it happen in front of me.
Marvelous, thank you!
love you !! thanks so much
Looks great,but will not expand to full screen and all links on the page are dead. Too bad as I was really looking forward to learning this interpretation.
Presenting the wave form under the circle, vertically, would be even better visually.
Excellent explaination - thank you
This was beautiful - thank you.
dude god bless you
I think the smoothie analogy is a bit confusing. I find it easier to stick to geometric/math interpretations. For anyone that is interested, 3 Blue 1 Brown does a more in depth and mathematically intuitive video on the subject.
As a more visual learner, I liked the smoothy concept. At least to start with...
@Mark Taronji I agree Blue 1 Brown's video is far more intuitive and it is much better explained.
To watching this video please turn down the play speed to 0.75 will be helpful....XD
Very nice explanation!
Amazing explanation!
Great explanation.
Highly insightful!
great job!
Dude, you rock. This is great.
I agree with Amanda 7913. However I was luckier, this is my first attempt to learn it. So I got lucky- the magic of the internet
Great video!
Thanks the Intuitive explanation and interactive virtualization ... there is an inconsistency of your equation normalization between deferent sections ...
I wish i could give this more than one thumb up... Thanks!
I did it for you
May you consider explaining in your 'easy-to-understand' terms and visualizations the Nyquist Shannon Theorem?
THANK YOU!
nice xplaner! wondering what if the signal is not sampled at uniform interval?
Simply the best
Interesting. Thanks. The phrases [kind of, sort of, kind of like] were used often.
I just `want to say, Thank you...
perfect thanks
凄い👍
On the website I noticed the 0 cycle has the value of 1Hz and 1 cycle has the value of 1Hz.
cycle ( amplitude) as {0, 1}...... time( frequency) as {1 -1}, i.e. how many times it goes around in for each one second.
In the next line below you describe what it all means.
You describe the 0 cycle ( amplitude) having 0Hz, and then say 0Hz means it stuck on X axis.
What would it mean if you instead described the 0 cycle having 1Hz frequency instead, so it matches the value set for them to begin with?
If 0 cycle had 0 frequency, wouldn't it mean that it was stuck at the start position of whatever the phase angle is. And 1Hz means it is stuck on the x axis right through?
Say 1 cycle ( amplitude) having 0 Hz. Wouldn't it mean that 1 cycle is stuck at the start position?
Also from the look of it the cycle is read from left to right but to correspond them to their frequency on the right, you have to go from right to left.....
I can see from the following animation... {0 1 1} {2 -1 -1}. Isn't cycle another word for amplitude?
Isn't time another word for frequency?
If that be the case then { 0 1 1} mean amplitude 0 and amplitude 1 and another amplitude 1}
And the {1 -1 -1} then would mean 0 amplitude is being repeated two times a second 2 Hz. 1 and the other 1 amplitude, both being repeated one time 1Hz...
Really good explanation! Thanks for that :)
still a lot confused.
What are you the math tutor prophet? I was just looking into a solid resource a few days ago for this due to NLP research
very good lecture.
THANK GODDDDDDDD!!!!
whoa! so good explanation. wish I got this explanation before T.T
Thanks awesome
How would someone like me who knows nothing about math i have memory problems basically pseudo dementia due to depression i lost alot of gray matter but I desperately want to learn this out of extreme curiosity and ocd please would u guide me on what in math to learn because i know basically nothing
do you have the homer simpson visualisation as a gif or video for me to put in a presentation please?! awesome video!
Yep, it's from czcams.com/video/QVuU2YCwHjw/video.html
Would this be used to control a motor in a car or machine?
A bit off topic but can you tell me how you make your videos where your speaking video, lower right corner- is simultaneously superimposed upon your computer screen- the background.
Use Open Broadcasting Software, OBS
Is the circle representation basically using vectors then?
Is the Simpson pic in the complex plane? [btw the Fourier Toy link is broken 🙁 ]
Exactly, the image is a 2d set of data (which the Fourier Transform can work on as well, though it's often used on 1d audio signals). Fixed up the Fourier Toy link to a version on archive.org.
Awesome. Thanks!
For me, it is hard to understand the maths behind FT from this video.
I am not sure why you are doing all these delay and stuff when the explanation is a simple 0 result for dot product for non matching frequencies with sin and cos component and reconstructing the wave back from using Euler's identity.
PogChamp
Totally lost you at 1:21 and on
You forget to add some curry in your recipe
similarly you forgot to add a valid point in your racist rot existence
your head was annoying alwasy be on