What ? Eh ? Run that by me again ?

It used to irritate me somewhat, when someone would ask how to soundproof their recording room and most people answering would pile on them about it. I remember Miroslav often making the point that there was no need to soundproof a room.
But many of the people that would say this didn't live in apartments {or flats as we call them in England} with neighbours next door on both sides with whom one shared both walls and either directly above or below {and in some cases, both} whose ceiling you were above or whose floor you were below. If you wanted to play an actual acoustic drum kit {I use one} or a brass instrument {I've frequently used saxes and trumpets} or put an electric guitar up loud to take maximum advantage of an overegged amp, then soundproofing trumped room treatment every day of the week because you could always attend to the acoustic treatment afterwards.
I've had to take all kinds of measures to cut down on disturbing the neighbours over the years ~ but I don't whinge too much because I've become quite a diplomat 🧐 in the meantime !
This may sound stupid but soundproofing sounds expensive. Why do you not just do your demos in a studio?
 
It used to irritate me somewhat, when someone would ask how to soundproof their recording room and most people answering would pile on them about it. I remember Miroslav often making the point that there was no need to soundproof a room.
But many of the people that would say this didn't live in apartments {or flats as we call them in England} with neighbours next door on both sides with whom one shared both walls and either directly above or below {and in some cases, both} whose ceiling you were above or whose floor you were below. If you wanted to play an actual acoustic drum kit {I use one} or a brass instrument {I've frequently used saxes and trumpets} or put an electric guitar up loud to take maximum advantage of an overegged amp, then soundproofing trumped room treatment every day of the week because you could always attend to the acoustic treatment afterwards.
I've had to take all kinds of measures to cut down on disturbing the neighbours over the years ~ but I don't whinge too much because I've become quite a diplomat 🧐 in the meantime !
In my top floor apartment, I often think about using a combo bass amp and soundproofing is out of the question. When the time comes, I'll raise the amp above the floor on a chair and throw some heavy blankets around it. Not sure how I'll insulate the chair-floor vibrations. So sound/vibration damping - not soundproofing.
 
It used to irritate me somewhat, when someone would ask how to soundproof their recording room and most people answering would pile on them about it. I remember Miroslav often making the point that there was no need to soundproof a room.
But many of the people that would say this didn't live in apartments {or flats as we call them in England} with neighbours next door on both sides with whom one shared both walls and either directly above or below {and in some cases, both} whose ceiling you were above or whose floor you were below. If you wanted to play an actual acoustic drum kit {I use one} or a brass instrument {I've frequently used saxes and trumpets} or put an electric guitar up loud to take maximum advantage of an overegged amp, then soundproofing trumped room treatment every day of the week because you could always attend to the acoustic treatment afterwards.
I've had to take all kinds of measures to cut down on disturbing the neighbours over the years ~ but I don't whinge too much because I've become quite a diplomat 🧐 in the meantime !
So how did you soundproof the room?
 
This may sound stupid but soundproofing sounds expensive. Why do you not just do your demos in a studio?
Because I'm a home recorder.
You know, from when I first bought a bass guitar in '81, I played at home. When I first jammed with some guitarists, I jammed at home. When I started jamming with a drummer, we jammed at his home. At music practices with friends, we were always at someone's home. So when I eventually began multitracking 10 years after I'd began playing it didn't even occur to me not to record at home in my own flat. And it never has. Everywhere I've lived, I've utilized it when it comes to recording. That's not to say I haven't utilized other spaces as the need or opportunity arose, but comparatively, those are few and far between. The best description I've ever come across for what I'm about is a guerilla home recorder. It has never entered my mind to actually hire a studio. Why the heck would I want to do that ?
Soundproofing is expensive. But it has never been so vitally important that it actually prevents me from making music and recording and mixing it. In fact, nothing has ever prevented me from recording and mixing.
In my top floor apartment, I often think about using a combo bass amp and soundproofing is out of the question. When the time comes, I'll raise the amp above the floor on a chair and throw some heavy blankets around it. Not sure how I'll insulate the chair-floor vibrations. So sound/vibration damping - not soundproofing
I agree 100% with you. I do the same. I do everything I can to minimise noise and vibrations. I'll put amps in wardrobes, I've built a sort of iso cab which enables me to crank that amp loud, I'll play at low~ish volume and simultaneously mic 2 amps and blend or pan, I built a decoupler stage for the drums etc. We do what we can. But if the room was soundproofed, none of this would be an issue.
I play my drums when I see my neighbours have gone out
We have to be considerate of our neighbours. But inspiration doesn't work that way. My timetable isn't built around the movements of my neighbours. If there's going to be drums pounding during the day, I let them know. But I have great neighbours. I've been fortunate down the years, I always have had.
So how did you soundproof the room?
Well, I didn't. I couldn't. But that was my point to Spantini earlier, that if I could, I would. But I've learned to get around that anyway and have had thousands of sessions and hundreds of songs recorded.
I looked into the Esmono room that Slouching Raymond has got. I'd love one of those. I'd have recording sessions on through the night if I had one of those. But it is not doable in my flat. So I record in my kids' room, mine & my wife's room, the bathroom, the front room, anywhere I can, really.
Right now, if it was possible to soundproof this place, I'd do it.
 
Here’s one of those stupid questions, at least one that I have.

What’s the difference between 48k and 44.1k. Is one better?

For some reason unknown to myself I’ve always used 44.1
 
The glib answer is that the more the better, but really if you do the maths, 44.1 gave a frequency response above 20K, where we usually stop measuring, but there's also accuracy to take into account. You are taking a moment in time, and recording the level at that precise moment - and with adjustable accuracy - so our old 16 bit went to 24 and now 32 bit accuracy.

The analogy I used with students was to show them a moving head light. 8 bit, so 127 steps from off to max. Shining a light across the room, the 8 bit pan, spread over an entire circle would move nearly two feet when value 100 went to 101. You could not be more accurate than that. The lighting people solved it by adding in a second channel that used another 8 bits to fill in the two feet, so in the second channel, adding 1 was invisible, adding 10 produced a small movement. In audio terms, we had 16 bit, so from off to full was pretty smooth, but still steps. Now we have more bits available, these steps are smaller.

We can now record the live very accurately, so doing it more times every second does make sense as a practical thing to do. It must be better, but 44.1 to 48K I could not detect. 98K , and now double that offer better and more truthful sampling. You have to decide if it's worth it.

I do lots of video, which is 48K capture, so that's why I record at 48K in cubase - NOT because I can actually hear a difference. more bit depth and higher sampling = higher 'quality', but it's been pretty good at 16 bit 44.1KHz for a long time.
 
  • Like
Reactions: RFR
The glib answer is that the more the better, but really if you do the maths, 44.1 gave a frequency response above 20K, where we usually stop measuring, but there's also accuracy to take into account. You are taking a moment in time, and recording the level at that precise moment - and with adjustable accuracy - so our old 16 bit went to 24 and now 32 bit accuracy.

The analogy I used with students was to show them a moving head light. 8 bit, so 127 steps from off to max. Shining a light across the room, the 8 bit pan, spread over an entire circle would move nearly two feet when value 100 went to 101. You could not be more accurate than that. The lighting people solved it by adding in a second channel that used another 8 bits to fill in the two feet, so in the second channel, adding 1 was invisible, adding 10 produced a small movement. In audio terms, we had 16 bit, so from off to full was pretty smooth, but still steps. Now we have more bits available, these steps are smaller.

We can now record the live very accurately, so doing it more times every second does make sense as a practical thing to do. It must be better, but 44.1 to 48K I could not detect. 98K , and now double that offer better and more truthful sampling. You have to decide if it's worth it.

I do lots of video, which is 48K capture, so that's why I record at 48K in cubase - NOT because I can actually hear a difference. more bit depth and higher sampling = higher 'quality', but it's been pretty good at 16 bit 44.1KHz for a long time.
Thank you for taking the time to explain that.
 
For me, it's sum-to-mono. I'd love the ability to do this. But when I looked into it, I had to buy a special device for it. It's not a feature on any of the stuff I have.
 
Here’s one of those stupid questions, at least one that I have.

What’s the difference between 48k and 44.1k. Is one better?

For some reason unknown to myself I’ve always used 44.1
48 has become the most commonly used, but IME the bit depth is what really counts. 24 bit or better and dither as last step. Next to last if converting to MP3
 
Last edited:
  • Like
Reactions: RFR
I do not profess much if any great knowledge of 'digital theory', just what I have picked up over the years but here are my thoughts...
AFAIKT the "sliced up" then "reassembled" common idea of A/D, D/A processing is a gross simplification. The sampling theory tells us that provided the sampling rate is more than twice the highest audio frequency involved a waveform is reproduced 'perfectly' i.e. 20kH for a 44.1kHz sample rate. Whether we NEED a pass band above 20kHz is another question and applies to analogue systems, mics, tape recorders, speakers and always has.

"But! That is only a THEORY" some will cry. True but the limitations of real world digital processors, (jitter e.g.) are hardware issues, nothing to do with sampling rate. There is also a practical element and that is that D/A converters often use the same turnover frequency for 48kHz as for 44.1 so the apparent extra 'reach' to nearly 24kHz is illusory.

VERY early A/D converters needed a very complex anti-alias filter but it seems to me only slightly less difficult to make one for 48k as 44.1 ? There is some benefit in going 88k or 96k but many converters actually don't work as well up there as at the lower rates and file size is a bit of a problem. Up sampling has largely removed this problem in any case I understand?

Yes, 48kHz has become standard for video work. There is an idea this is due to it being related to the old 24 frames ps film rate ? Don't know. I do know that a lot of very low cost converters, UCA 202 and many USB mics are fixed at 48k.

Do we need to dither 24 bit recordings? The theoretical dynamic range is 144dB. Not that good in practice of course but still better than THE very best converters which only get a DR of the low 120dBs. I always understood that 24 bits 'dithered' on the unavoidable analogue noise floor.

Kevlar hat firmly applied.


Dave.
 
Here’s one of those stupid questions, at least one that I have.

What’s the difference between 48k and 44.1k. Is one better?

For some reason unknown to myself I’ve always used 44.1
The differences can be examined - for example - Frequencies are rolled off in accordance with the steepness of your filter’s slope (determined by your convertors). So if you apply too steep of a slope, you can experience a resonant peak or other unwanted side effects. At 44.1kHz, the anti-aliasing filter’s slope has to be pretty steep - and may start rolling off frequencies within the range of human hearing - 48kHz provides room to use a more gradual slope - thus much less a resonant peak - so you have a potential for a slight difference between a recording made at 44.1kHz and a recording made at 48kHz - but I think it's a difference one can hear - now jumping to 96 kHz one would think it's really great - but we have already exceed the level of human hearing - so not much gained in my book - so I record everything at 48 kHz and Bouncing my final mastered track to release at 44.1kHz.
 
The differences can be examined - for example - Frequencies are rolled off in accordance with the steepness of your filter’s slope (determined by your convertors). So if you apply too steep of a slope, you can experience a resonant peak or other unwanted side effects. At 44.1kHz, the anti-aliasing filter’s slope has to be pretty steep - and may start rolling off frequencies within the range of human hearing - 48kHz provides room to use a more gradual slope - thus much less a resonant peak - so you have a potential for a slight difference between a recording made at 44.1kHz and a recording made at 48kHz - but I think it's a difference one can hear - now jumping to 96 kHz one would think it's really great - but we have already exceed the level of human hearing - so not much gained in my book - so I record everything at 48 kHz and Bouncing my final mastered track to release at 44.1kHz.
Thanks for the lengthy reply.
What’s the benefit of bouncing the final mastered track back down to 44.1 ?
 
The differences can be examined - for example - Frequencies are rolled off in accordance with the steepness of your filter’s slope (determined by your convertors). So if you apply too steep of a slope, you can experience a resonant peak or other unwanted side effects. At 44.1kHz, the anti-aliasing filter’s slope has to be pretty steep - and may start rolling off frequencies within the range of human hearing - 48kHz provides room to use a more gradual slope - thus much less a resonant peak - so you have a potential for a slight difference between a recording made at 44.1kHz and a recording made at 48kHz - but I think it's a difference one can hear - now jumping to 96 kHz one would think it's really great - but we have already exceed the level of human hearing - so not much gained in my book - so I record everything at 48 kHz and Bouncing my final mastered track to release at 44.1kHz.

Can you give me an example of a converter/Ai that switches the filter corner frequency for the two lower sampling rates?

Dave
 
The sampling theory tells us that provided the sampling rate is more than twice the highest audio frequency involved a waveform is reproduced 'perfectly'
That assumes that you are taking perfect analogue samples. But we're not, we're digitizing them into steps.
 
Back
Top