I was looking back on some audio theory i wasn't too sure about anymore and i have to admit your videos are probably the best on the subject right now, production quality is equivalent to channels that have millions of subscribers and the information is extremely well explained on subject that aren't that easy to understand, thanks for the reminder !
Very nice explanation and visuals. I’m using an fpga for a signal processing application, and it’s critically important that the signals I’m working with have no dc bias. Rounding errors from fixed point arithmetic can accumulate, so I use “fraction saving” in my filters to eliminate quantization noise at dc, and shift it to higher frequencies where it is less of a problem.
That last metaphor, comparing digital dithering to mechanical systems, being vibrated “to be in place“ (Mechanical quantization?), Was very helpful for really locking in this concept! Thank you for these incredible modules 🙏🙏🙏
@@akashmurthyDo you only use Dithering and noise shaping when truncating of bits 24/32 to 16 bit? Is there any benefits of using it with 24 bit audio? Thanks
@@joesmith4443 hey Joe, you'd generally only use dither when converting to 16bit. There is nothing wrong to dither the mix when converting to 24bit, but it's just not needed. There is not much advantage to it. In 24 bit audio, the dynamic range is high enough for the the quantization error not to be noticed. You can check out the 7th video in the series, and I've discussed the same there.
@@akashmurthy Thanks for getting back to me! I learn a lot from from your videos! Would love it if you talked about PWM and Sigma Delta (or direct me to where you do!). I’m more of a DSD guy than PCM. Does the truncation happened when you record from 24 to 16 via a coax/optical input and output?? Should you add dither to the 24 file? 24 bit SHOULD be the standard as CD’s and bit limitations is a thing of the past. I really don’t like the marketing of “High Res” when the science is different. The higher sampling rates do better relaxing and spreading the quantization error even further out into ultrasonic frequencies and the dac filters and prevent aliasing. All that said none will know what the higher bit and samplings can do when the dynamic range is soooo compressed that many believe vinyl “sounds” better. It’s a real shame!
One stop solutions for all doubts on Audio basics. Thank you so much Akash for providing the awesome content along with nice explanation of each term with visual effect. looking forward to adaptive filtering techniques using in audio domain.
Hey thanks a lot for the feedback! I've got a series on filtering in the pipes. Although it's more basic and fundamental rather than being mathematical and complicated!
@@bullikoti7586 oh, when I said it's in the pipes, I meant that it's on the way! In a couple of weeks I'll be posting them! Subscribe to get notified about it..
What a way to teach,Super awesome.Thanks How about MP3, though no renderer ask for a bit depth in MP3 but for example in general by math 128kbps audio are assigned 1.5 bit. Or is it in-built in the mp3 coding itself or why to even care, because the format is lossy anyways.
Thanks very much! Bit depth is sort of only relevant when dealing with uncompressed PCM data. When you convert it to MP3, there is no longer a need to store each sample at a fixed number of bits. The idea is to dedicate as little bits to store the data as possible. So the codec uses a different way of representing the data. I don't know much about the MP3 codec, but there are a lot of resources out there that are quite good to learn from.
Sir can you suggest me something on Sigma Delta ADC where oversampling and noise shaping are used? How they use and what impact they get I am interested to knwo
absolutely beautiful explanation, even for newbie
I was looking back on some audio theory i wasn't too sure about anymore and i have to admit your videos are probably the best on the subject right now, production quality is equivalent to channels that have millions of subscribers and the information is extremely well explained on subject that aren't that easy to understand, thanks for the reminder !
No worries mate! Thanks for the kind words..
Fantastic explanation of the POW-R noise shaping algorithms
Cheers! Thanks much!
Awesome content! Keep creating and they will become the best audio processing/recording content on UA-cam. 💪
Very nice explanation and visuals. I’m using an fpga for a signal processing application, and it’s critically important that the signals I’m working with have no dc bias. Rounding errors from fixed point arithmetic can accumulate, so I use “fraction saving” in my filters to eliminate quantization noise at dc, and shift it to higher frequencies where it is less of a problem.
That last metaphor, comparing digital dithering to mechanical systems, being vibrated “to be in place“ (Mechanical quantization?), Was very helpful for really locking in this concept! Thank you for these incredible modules 🙏🙏🙏
For me as well! It was fascinating to discover that.
Some of the best tutorials in sound/audio theory out there. Well done!
Thank you! Glad you think so!
@@akashmurthyDo you only use Dithering and noise shaping when truncating of bits 24/32 to 16 bit? Is there any benefits of using it with 24 bit audio? Thanks
@@joesmith4443 hey Joe, you'd generally only use dither when converting to 16bit. There is nothing wrong to dither the mix when converting to 24bit, but it's just not needed. There is not much advantage to it. In 24 bit audio, the dynamic range is high enough for the the quantization error not to be noticed. You can check out the 7th video in the series, and I've discussed the same there.
@@akashmurthy Thanks for getting back to me! I learn a lot from from your videos! Would love it if you talked about PWM and Sigma Delta (or direct me to where you do!). I’m more of a DSD guy than PCM.
Does the truncation happened when you record from 24 to 16 via a coax/optical input and output?? Should you add dither to the 24 file?
24 bit SHOULD be the standard as CD’s and bit limitations is a thing of the past. I really don’t like the marketing of “High Res” when the science is different. The higher sampling rates do better relaxing and spreading the quantization error even further out into ultrasonic frequencies and the dac filters and prevent aliasing. All that said none will know what the higher bit and samplings can do when the dynamic range is soooo compressed that many believe vinyl “sounds” better. It’s a real shame!
The best video I found for that topic, amazing. THanks a lot
One stop solutions for all doubts on Audio basics. Thank you so much Akash for providing the awesome content along with nice explanation of each term with visual effect. looking forward to adaptive filtering techniques using in audio domain.
Hey thanks a lot for the feedback! I've got a series on filtering in the pipes. Although it's more basic and fundamental rather than being mathematical and complicated!
@@akashmurthy Hi Akash. could you please provide the link to go through that series on filtering in the pipes. Thank you so much.
@@bullikoti7586 oh, when I said it's in the pipes, I meant that it's on the way! In a couple of weeks I'll be posting them! Subscribe to get notified about it..
You are a DSP legend.
Brillant videos! Thank you.
amazing content and presentation
Looking forward to deep dive filtering videos !
Coming soon!
What a way to teach,Super awesome.Thanks
How about MP3, though no renderer ask for a bit depth in MP3 but for example in general by math 128kbps audio are assigned 1.5 bit.
Or is it in-built in the mp3 coding itself or why to even care, because the format is lossy anyways.
Thanks very much!
Bit depth is sort of only relevant when dealing with uncompressed PCM data. When you convert it to MP3, there is no longer a need to store each sample at a fixed number of bits. The idea is to dedicate as little bits to store the data as possible. So the codec uses a different way of representing the data. I don't know much about the MP3 codec, but there are a lot of resources out there that are quite good to learn from.
Got it,will try to explore further.
Thanks Akash.
Like your videos!!! Top.
Cheers mate!
Sir can you suggest me something on Sigma Delta ADC where oversampling and noise shaping are used?
How they use and what impact they get I am interested to knwo
Sorry, I don't really have any video recommendations for you. You can read the books in reference section in the description to learn more about it.
Delta Sigma Modulation for a video! 💘
I'd love to do it at some point of time. It's fascinating..
Hey,
Where can I find your course on acoustics? I can't find the module.
Hey, that series is not done yet, at the moment there are 5 videos, it's from Loudness and Levels
@@akashmurthy Is there any ETA for availability of the acoustics module?
@@mohsens22 nope.
👂👍👍👍👍👍👍👍