Is realtime oversampling a dirty/destructive process?

VST, AU, AAX, CLAP, etc. Plugin Virtual Effects Discussion
Post Reply New Topic
RELATED
PRODUCTS

Post

bootsie wrote:
DaveGamble wrote: - best solution: find ways to avoid generation of aliasing without oversampling
word!
correct. but find me a way to do this for f.e. analog like resonant filters with a tan-h (or whatever approximation for the drive) in the feedback path... imo the only way in this case is oversampling, if you would want the satureation/harmonics to be applied to the full frequency response of the filter... no? please correct me if i'm wrong....
regards,
brok landers
BIGTONEsounddesign
gear is as good as the innovation behind it-the man

Post

in practice there's not too many signals, especially transient rich signals, that sound good when the transient is reaching up to 21khz
I'd like to hear just 1 bit of sound (anything) that sounds different with its content above 20khz fully filtered.

I'd even suggest 16-17khz, but I'm 37 & there may be kids who can really hear the difference. But with 20khz it's pretty safe.
DOLPH WILL PWNZ0R J00r LAWZ!!!!

Post

tony tony chopper wrote:
in practice there's not too many signals, especially transient rich signals, that sound good when the transient is reaching up to 21khz
I'd like to hear just 1 bit of sound (anything) that sounds different with its content above 20khz fully filtered.
That may be true, but I don't have access to such filtering technology. Fact is, even the steepest filters ALWAYS have a smooth transition. Keeping this transition away from the audible band is the reason why you want to tune your nyquist filters a little higher (or better: as high as possible). Just try it out. Try to build a filter that fully filters content above 20kHz. :)
Fabien from Tokyo Dawn Records

Check out my audio processors over at the Tokyo Dawn Labs!

Post

Try to build a filter that fully filters content above 20kHz.
Why not? You can make a FIR lowpass as giant as needed.
I was only saying that whatever is above 16-17khz hardly matters.
DOLPH WILL PWNZ0R J00r LAWZ!!!!

Post

tony tony chopper wrote:
Try to build a filter that fully filters content above 20kHz.
Why not? You can make a FIR lowpass as giant as needed.
I was only saying that whatever is above 16-17khz hardly matters.
well, i wouldn't go that far and say it doesn't matter, but i for one would doubt to hear a change of -3db above 20khz, let alone the phase shift above 20khz. and even if i would hear it, i assume i'd actually prefer the shifted, minus-3db-at-21khz signal.... :)
regards,
brok landers
BIGTONEsounddesign
gear is as good as the innovation behind it-the man

Post

tony tony chopper wrote:I was only saying that whatever is above 16-17khz hardly matters.
Again, that may be fine for some signals. But in practice, the process of filtering "whatever is above" will ALWAYS affect "whatever is below" as well. That's my point. And it becomes clear as soon you leave theory and try it yourself.
Fabien from Tokyo Dawn Records

Check out my audio processors over at the Tokyo Dawn Labs!

Post

But in practice, the process of filtering "whatever is above" will ALWAYS affect "whatever is below" as well.
how?
That's my point. And it becomes clear as soon you leave theory and try it yourself.
but I've done it many times

can you post an example?
DOLPH WILL PWNZ0R J00r LAWZ!!!!

Post

mmmm...I wonder if this would affect bat range hearing.... :wink:
Barry
If a billion people believe a stupid thing it is still a stupid thing

Post

tony tony chopper wrote:
But in practice, the process of filtering "whatever is above" will ALWAYS affect "whatever is below" as well.
how?
Beside obvious things such as filter passband ripple, the smooth transition region, and maybe phase shift, don't forget the Gibbs phenomenon. The latter isn't audible, but its time-domain implications can turn audible (e.g. via overloads).

It makes no sense to over generalize these things. Reducing the bandwidth introduce side-effects as a matter of fact. No matter if you like them or not.
Last edited by FabienTDR on Mon Mar 04, 2013 3:06 pm, edited 1 time in total.
Fabien from Tokyo Dawn Records

Check out my audio processors over at the Tokyo Dawn Labs!

Post

Beside obvious things such as filter passband ripple, the smooth transition region
but with a massive FIR that would be less than neglectable.

Again I'm not arguing about filtering, but about this sentence
in practice there's not too many signals, especially transient rich signals, that sound good when the transient is reaching up to 21khz
which has nothing to do with the question whether you can perfectly filter above 21khz (and the answer is yes, you can, you just can't do without a massive latency but you can)

The latter isn't audible, but its time-domain implications can turn audible
obviously yes what's at 20khz matters if you process it further, but that wasn't really the question.
DOLPH WILL PWNZ0R J00r LAWZ!!!!

Post

bmanic wrote:I can't quite put my finger on it but it's a similar constant annoyance like when a client delivers some song for mastering and he/she hasn't used any dithering when it's required but instead has truncated everything. It's a nasty harshness, usually manifesting itself from about 2kHz onward. Especially noticeable once you start hitting the final mastering limiter/ADDA loop hard.
Have you ever verified that impression in a blindtest?

You really had to mix at low levels and render out in 16bit to ever have an issue with truncation errors.
And nobody would ever send a 16bit file to mastering.
bmanic wrote:Lets say you have quite a few tracks in a production.. a typical number would be perhaps 50 tracks playing simultaneously (this is quite common for most pop/rock things with a lot of multitracked content). Now you have some compression on most tracks at some point in time. They all have annoying aliasing/nasty stuff below -100dBFS.. at some point in the song you will have some of these inharmonic peaks cancel each other out and at some points they will reinforce each other, basically doubling their volume.
You forget two things here:
1. Loudness of unrelated signals won't add up indefinetly.
2. The loudness of the actual audio signals (100dB louder than your "nasty stuff") will add up as well. So the relative loudness of your errors stays about the same.
tony tony chopper wrote:But if your compression envelope is too fast.. it gets in the audible range.. and I don't think that you want a compressor produce "sound" by itself.
Huh?
I thought, that's what compressors are for.

I totally love the snappy sound with shorter attack-times of some compressors.
tony tony chopper wrote:I'd even suggest 16-17khz, but I'm 37 & there may be kids who can really hear the difference. But with 20khz it's pretty safe.
+1
I'm 27 now, and stuff above 20kHz is inaudible to me.
(And I use earplugs to protect my ears in nightclubs.)

Post

bmanic wrote:
DaveGamble wrote: Dynamics processing doesn't need oversampling by more than x2 in the audio path.
Not so sure about this. I've done a lot of testing of this and feel that it very much depends on many things:

[snip]
Finally, I agree that the most critical component is to oversample the actual sidechain/VCA block thing.. not sure if there is much benefit in oversampling the actual audio path too much.
Ah, to clarify: My original statement wasn't something that one can have an opinion on - multiplying together two full bandwidth signals only generates intermodulation signals that go up to twice nyquist, so x2 in the audio path is all that's necessary. This is assuming, of course, that your VCA is just that - a variable gain stage.

But, as you say, oversampling the process that determines what gain the VCA should apply does indeed appear to offer benefits.

Dave.
[ DMGAudio ] | [ DMGAudio Blog ] | dave AT dmgaudio DOT com

Post

brok landers wrote:
bootsie wrote:
DaveGamble wrote: - best solution: find ways to avoid generation of aliasing without oversampling
word!
correct. but find me a way to do this for f.e. analog like resonant filters with a tan-h (or whatever approximation for the drive) in the feedback path... imo the only way in this case is oversampling, if you would want the satureation/harmonics to be applied to the full frequency response of the filter... no? please correct me if i'm wrong....
Well, this is precisely the thing. It would be pessimistic to imagine that we'll never have good solutions to this problem.
You already might use an RK-based strategy?

But finding you a way to do this, even though there isn't an obvious way now, is likely going to be one of the big research topics for us all over the next few years.

Either that, or we'll all sink into oversampling complacency until some bright spark comes along with something clever.

Dave.
[ DMGAudio ] | [ DMGAudio Blog ] | dave AT dmgaudio DOT com

Post

Nokenoku wrote:
bmanic wrote:I can't quite put my finger on it but it's a similar constant annoyance like when a client delivers some song for mastering and he/she hasn't used any dithering when it's required but instead has truncated everything. It's a nasty harshness, usually manifesting itself from about 2kHz onward. Especially noticeable once you start hitting the final mastering limiter/ADDA loop hard.
Have you ever verified that impression in a blindtest?
Many many times. I do ABX tests all the time. A shootout between 3x tracks of The Glue set to 64x oversampling versus the same 3 tracks set to no oversampling is very easy to pick out.

However, it might be also due to the oversampling itself.

I've also done plenty of blind "taste" AB tests where I do not concentrate on the differences but simply select the one I feel is better sounding. These usually end up being the heavily oversampled ones.. but not always. There is a lot of material that does suffer from the actual oversampling process. Sometimes things can sound a bit strained or less punchy. Not sure why this happens either as the filtering that needs to be done when oversampling is at such high frequencies. :shrug:

Cheers!
bManic
"Wisdom is wisdom, regardless of the idiot who said it." -an idiot

Post

I thought, that's what compressors are for.
to me a compressor is an "auto gain", but if the gain envelope is going at such a high rate that it enters the audible, it's not just a gain anymore, it colors the sound. Which may not be a problem, but I wouldn't go as far as saying that it's what compressors are for.
DOLPH WILL PWNZ0R J00r LAWZ!!!!

Post Reply

Return to “Effects”