Okay this is bugging me. Imagine we have sampling rate Fs. Then we define our nyquist frequency as Fn = Fs/2, where the implication is that this is the highest frequency we can observe without aliasing/and/or our system bandwidth. However in the case of quadrature detection and the complex fourier transform, we can have a message signal occupying anywhere from -Fn to +Fn, representing a total bandwidth of Fs, NOT Fn!
This is one of those problems that I feel DSP 101 students stumble over, but somehow its tripping me up now.
Was going through the announcement and informing a lot of y’all but so many have this wrong I would have to spam like hell to tell you all.
By sampling rate I mean how many times per second? Minute? Does it take temp or humidity samples.
Also I think there is a 160 mhz and 240 mhz versions. Does the mcu clock rate determine the sample rate or is that the sensor that does that?
Tia
Using my Port to my receiver… receiver says that input is 44.1 kHz but seems like that’s just the connection to the Port, not necessarily the quality of sound that Sonos is receiving from the streaming source
On my Arc I can tell the audio/video quality… looking for something similar from the port. Can this be found in settings somewhere?
I definitely understand that there are very good reasons to over sample when using plugins in a DAW, especially those whose effects are non-linear like compression, saturation and distortion to avoid aliasing, but if you’re someone like me who basically uses the computer like tape — as a medium to record onto — while using my mixer and a few pieces of outboard gear that I have to actually do the mixing, is there a reason to use a sample rate greater than 44.1KHz? The Nyquist limit at that sampling rate is 22.05KHz and the upper limit of what is audible to a child with perfect hearing is 20KHz, and furthermore the highest frequency that one will ever actually hear in music (and not often) is C8, the 88th key on a piano. which in standard A4=440Hz pitch and 12 tone equal temperament is 4186Hz, and in terms of vocals the extreme top end of the whistle register for a soprano is D6, which is 2349Hz.
I’m honestly quite new to the world of digital recording and it seems that the universal recommendation is to use a sampling rate of 96KHz, so that’s what I’ve been using. However, a file with 96,000 samples per second, each sized 24 bits, is humongous, and my extremely rudimentary knowledge about digital signal processing, the limits of the human ear and the range of instruments suggest to me that more than 50,000 of those samples are totally redundant. Am I missing something? Is there a reason to record at a sampling rate higher than 44.1KHz when simply recording and not working in a context where aliasing might become an issue?
From a thread on Mike Dean’s Twitter
This is bullshit right? I’m no professional and didn’t help produce some of the most critically acclaimed albums of the 21st century but… wtf?
This is a follow-up to a recent post of mine where I couldn't get my Yamaha A-S801 to display the sampling rate of the inbound file - it only displayed, statically, the setting in the Steinberg driver. Hope this helps other 801 owners!
________________________________________
Yamaha Tech Support really came through for me. There is a way to get the behavior stated in the owner's manual, and it's really useful! Here's the information from Yamaha posted as a PSA for other 801 owners. This works perfectly...
You have to add an ASIO component to Foobar before it will behave as you describe. You can download the component here:
https://www.foobar2000.org/components/view/foo_out_asio
After downloading, launch Foobar and select "Preferences">"Components">"Install" and point it towards the file you downloaded. After installing the component, select "ASIO: Yamaha Steinberg USB ASIO" as your playback device. After doing this the sampling frequency indicators will change depending on the song being played.
I'm using Limbo PC Emulator, running off QEMU 5.1. The change to the sampling rate to 22050 from 44199 now distorts my audio where it previously didn't. I'd like to adjust this sampling rate back but I'd like to know if it's possible to set sampling on boot using commands. On the QEMU docs it doesn't list any commands related to setting the sampling rate.
Sometimes I have to use my fiio m6 player as a DAC for my computer. The problem is that maybe due to driver problems there is a noticeable delay at 44,1 kHz and almost no delay at 192kHz. In the browser when I watch videos it is very noticeable so I have to use high sampling rate. But also I use Spotify which outputs 44,1 kHz and I think because of the upsampling it may cause sound deterioration. If I set 176,4 kHz as standard in windows will the quality degrade when using windows upsampling? ( Spotify 44,1 kHz -> Windows Output 176,4 kHz).
Sorry for my English. I hope you understand what I mean.
In a basic definition, the sampling rate is the number of times the screen can register the user's touch. This rate is usually measured in 1 second, and the Hertz or Hz measurement is also used to indicate frequency.
So way back in late 90's/early 2000's I was in a Sam's Club and they had an iMac G3 on display probably running Mac OS 8. This was the first Mac I'd ever seen in person, let alone interact with.
Other than how completely foreign the interface was, what stuck with me was how smoothly the mouse moved around on the screen. I wanted to replicate this smooth movement on my Windows 98 computer at home and there was some software you could download that ran in in the background that would increase the sampling rate or refresh rate of your mouse on the screen so it would appear to move a lot smoother.
Was this a common thing people did at the time, or was I the only one who got satisfaction from smooth mouse movement? Did Windows XP or Windows 7 end up changing how the mouse movement looked, or did I just eventually get used to how it looked on Windows?
I'm a bit frustrated at this point, so hopefully someone here can shed light on the situation.
Many, many years ago I purchased the M570. And I've been using it to this day. It's great. But one of my biggest frustrations is that it completely stops tracking if I spin or flick the ball too fast. I always hated that and planned to upgrade because of it.
I tried to replace it with the Elecom Deft, and while that actually didn't stop tracking when spun too fast (at least on the "high" radio signal), I really disliked everything else about that trackball, the jumpy ball itself being the main issue. And the replacement Perixx ball I purchased barely helped.
So back to the M570 I went.
One thing I was always curious about was the Kensington line. The expert/slimblade were too expensive so I never got one. I wanted an Orbit, but considering it's age I was worried it'd have the same problem.
Then I saw they did a soft refresh on the Orbits and introduced the white version. What sold me was the wireless one (https://www.amazon.com/gp/product/B09DGMLRCB/) - The fact that it had adjustable DPI seemed like a great sign - It must be a new sensor. Surely it wouldn't have this problem, right?
Sadly... Wrong. The problem is actually worse on the Kensington than the M570. If I go even slightly fast the cursor just stops dead in it's tracks. You have to stick to slow movements and not rely on flicking. I was able to tolerate it on the M570, but it's pretty much unacceptable on the 2021 Orbit Wireless w/ Scroll Ring. Moving closer to the receiver doesn't have any effect either.
I'm at a loss here.
Does the Expert line also do this? I've read some reports that it does. If so, should I try the wired Orbit instead? Maybe the current wired Orbit has better tracking? Is it just a problem with wireless trackballs? Do they ever update these sensors?
Or should I just give up on fingertip and get an MX Ergo or M575? Do they do this too or has there been improvements? The MX Ergo seems like the next step if the wired Orbit doesn't work.
If anyone knows it'd be super helpful. Especially about the wired Orbit scroll wheel. I can still exchange and of course the Wired version is a bit cheaper, so hey, fair trade if it doesn't do this. If you have the wired version please let me know!
Thanks for any help, I appreciate it a lot.
My super basic understanding is that when sound is digitized, the sample rate has to be double the source frequency, so the frequency is doubled. And the reverse happens when it's turned back into an analog signal.
So in other words, the DAC in my computer takes the 44.1 KHz audio from a Blu-ray and sends 22.05 KHz analog signal to my headphones. I checked and my headphones have a frequency response of 15 - 28,000 Hz, therefore all of the sound will be reproduced (whether I can hear it all or not).
Is this correct?
Latest Signal desktop, Windows 10. Creative G6 USB sound card/DAC.
I've noticed no audio is heard on both sides with 32 bit/192 kHz. Once I change the setting to 16 bit/48 kHz everything is fine. Is this known or is it a random quirk?
When I am considering a new SDR is a higher sampling rate better? For example 20MSPS or 10MSPS using a 12 bit ADC.
Not been able to find any info on it anywhere. Now that some people are getting phones delivered it'd also be helpful if people can suggest an app or website that will allow this to be tested.
Hello everyone, in my local area the fires are a common thing and so many families of firefighters wants to listen to their radio conversations. My question is that if I'm able to reduce the sampling rate on my RTL-SDR V3 dongle in order to decrease the overheating it suffers, so then I can re-broadcast via Zello their conversations, to the families of them.
My actual setup is a windows 10 labtop conected to the internet via an external WiFi antenna, SDR#, and RTL-SDR V3 dongle at 0.24 MSPS connected to the high speed port (the blue-one).
If anyone can give me a solution I woluld really apreciate the help.
Hi everyone, hope you're having a good day.
I am new to all things espressif and just looking for a minimal solution for ESP32 as a fast TCP Client to transfer ADC data.
I have created an Arduino based client, but getting slow speeds (around 10 kilobytes per second). I have tried sending larger packet sizes but 2048 bytes per packet causes packet fragmentation. I have tried iperf and a few other examples from esp-idf, but they all seem WAY to complex for my application.
Does anyone have a bare minimum implementation for a fast TCP client? Do I need to change config settings to speed up device (CPU speeds etc.)? Is Arduino IDE unsuitable for high speed TCP?
Please spell it out for me as a lot of previous forum posts have been very confusing!
Thanks in advance,
Will
I have noticed that some albums show a different bit/sampling rate on Apple Music iPhone vs. Mac. Can anyone explain why?
Both are set to lossless. Here is one example where iPhone indicates 24/48 yet Mac shows 16/41. And the Mac is connected to an external DAC.
iPhone set to Lossless (not HR as not using external DAC)
...but similarly set to lossless (HR as running external DAC)
do we know it? i couldnt find anything with google. or am i to assume its the same as the screen refresh rate?
Hi everyone !
I'm doing some data collection with LabView and I have a hard time understanding how DAQmx works.
I'm collecting 6000 data at 100 Hz using continuous collecting (and yes I will not changes it to finit for some more complicated reasons). I'm reading 4 data at the time, will do some calculations and will be sending the results to separate loops (producer consumer model) to log is and display it on the front panel.
Here is an example
https://preview.redd.it/us67eo03bjm71.png?width=1363&format=png&auto=webp&s=f28bcbb2f936fbd1ad482cbfe5787392266fc220
I know that there can be some delay with the front panel update or logging. I don't care with that, it's minimal and working well with notification and queue.
My questions are the following:
- I've read that sampling rate can be slow down when the is memory issue. I know that graph updates, for instance, can slow down a loop, but could someone explained how it can slow down the DAQ card sampling rate ? I thought that, with the buffer, even if the loop is slowdown, the DAQ is still collecting data base on his own sample clock. Thus a loop of 10 ms could be slowdown, for some raisons, to 100 ms, while the DAQ will put in his buffer 10 data even if there was only one iteration. In other word, data collection is not affected thanks to the buffer, but it's the reading that is affected. At leas, this is my understanding. I'm I wrong ?
- If sample rate can indeed be slowdown at the DAQ level, is there a way to verify the precision of my sample rate ?
Thanks everyone for your time !
I'm looking for an MCU with 12-bit ADC that has sampling rate higher than 15 MSps. I couldn't find a tool on TI, microchip, NXP, ST and Cypress that filters based on the MCU sampling rate. Can someone please help? Thanks
Hi everyone,
I am designing an ECG sensor and looking to stream data from the ADC wirelessly to a PC at a minimum sampling speed of 8kHz (1kHz x 8 channels).
I have tried Wi-Fi and Standard Bluetooth (Not BLE) on the arduino nano IOT, however it has proven to be slow for Wi-FI and unreliable for Bluetooth (lots of data loss). I have also tried interfacing the HC 05 BT module, however that is painfully slow (100Hz).
Can anyone suggest a RF module that can meet my needs? It does not need to be compatible with Arduino.
Any suggestions much appreciated. Thanks in advance, Will
New to hi-end audio gear. Trying to decide on some new headphones. What is the difference between the Frequency Response of the headphones and the Sampling Rate of the DAC and how do they work together to produce the best / better sounding music experience?
I notice a lot of Sony headphones have a 4 to 5Hz-40,000Hz and their bass response is deep and clear. And then I see that some “high-end” headphones have a 15Hz/20Hz to 28,000Hz and don’t seem to produce the same range of sound. Not sure how to match the headphones with the Sampling Rate for some DAC’s or Headphone AMPs.
Any links to do my own reading would be great - don’t expect anyone to just hand me the answer. Thanks!
I read on this Arduino Forum post that the digitalRead function has a sample period of 4 us (4 microseconds) on the Arduino Uno. However, when feeding my Arduino Uno with a function generator signal with the following settings...
- 5V 1Hz Pulse
- 10 us (10 microseconds) pulse width
- Max baudrate
... I am not able to detect any of the pulses. I'm therefore inclined to think that the sample period mentioned above is in fact higher. Could this be true? And if so: is there any way to achieve a faster sampling rate on the Arduino Uno (or ATmega 2560). I really need to detect there 10us pulses...
Edit: aside from detecting the incoming pulses, I would also like to be able to measure their width
I've been reading up on several articles but none of them really seem to answer the question I have in my head...
So recently I've been playing around with changing the sampling rate on my interface from 44.1 - 192kHz, and I noticed that the lower it is, the more 'raw' it sounds (but not gross or harsh) and the mid frequencies 'tickle' my ears more (which I like from the music I listen to) and seems more punchy. The higher the sampling rate, the more 'flat'/'bland' of a response curve it seems to be, whilst also making it quieter (I have to turn the volume knob up everytime I go up a step of kHz to get to the same level of loudness).
I'm sure someone has a scientific explanation behind this/can give me closure on what is actually happening when the sampling rates are changed.
