Automatic Gain Protection (AGP) vs Intersample Overs

Submit your proposals and ideas here.
EQenthusiast
Posts: 24
Joined: Sat Jul 09, 2022 7:43 am

Re: Automatic Gain Protection (AGP) vs Intersample Overs

Post by EQenthusiast » Mon Jul 11, 2022 5:58 pm

dmitrykos wrote:
Mon Jul 11, 2022 9:39 am
> add an optional intersample over detection

It would be excessive load on CPU because core would need to check amplitude of every and the benefit of that is miserable. To my view strategy of manual decrease of Preamp to the negative value of the loudest low-frequency band would be the best approach. Otherwise it is better to Normalize (Peak type) album with such EQ preset to guarantee 0 clipping.

I will consider adjusting RMS's window length to make AGP more reactive to a short bursts of exceeding sample values.
Fair enough. That could be a good idea. I know I myself was confused over just how much you could rely on AGP to catch all digital overs, and I had initially thought that AGP would indeed "catch everything", so there's probably a fair amount of users that are assuming it's catching everything, when in fact it's not. My ethos towards the digital side of audio is that I want everything to be covered & transparent at that part of the audio chain, so it was important for me to have found out that AGP doesn't cover everything, and I thank you for letting me know that in this thread earlier.....so indeed I can now instead make sure I'm using a large negative preamp to cover everything rather than using AGP. So I'm totally fine with that, but I'm thinking there's a lot of users of your software that have a similar mindset to me & want "everything to be covered" and wrongly assume that AGP is doing just that.....so it might instead be an education thing to your users rather than changing AGP to catch more, unless you change it to catch everything.

EQenthusiast
Posts: 24
Joined: Sat Jul 09, 2022 7:43 am

Re: Automatic Gain Protection (AGP) - a technical question

Post by EQenthusiast » Mon Jul 11, 2022 6:09 pm

blaubär wrote:
Mon Jul 11, 2022 5:52 pm
EQenthusiast wrote:
Mon Jul 11, 2022 5:41 pm
I'll still choose to apply cover for EQ boosts & intersample overs using the negative preamp then.
Right. If one uses oversampling and agp then factor 2 should be enough, see here they seem to suggest that factor 2 should suffice to identify intersample overs.
[...] all tracks was rescanned with the replaygain scanner in v1.3.9 with the scanner set to 2x oversampling for peaks (anything higher seemed to report the same peaks [...]
Screenshot 2022-07-11 195436.jpg
To be honest I don't really understand oversampling, I've not looked into it. But the easiest solution seems to be just running a large enough negative preamp to ensure that EQ boosts & intersample overs are covered, rather than using AGP combined with oversampling? And I'm also assuming that AGP + Oversampling still wouldn't cover my EQ boosts 100% anyway.

blaubär
Posts: 3326
Joined: Tue Apr 02, 2019 6:48 am

Re: Automatic Gain Protection (AGP) - a technical question

Post by blaubär » Mon Jul 11, 2022 6:45 pm

EQenthusiast wrote:
Mon Jul 11, 2022 6:09 pm
I don't really understand oversampling
For a brief description see here. In short: oversampling increases the sampling rate, typically by an integer multiple and in a resource-efficient way.
EQenthusiast wrote:
Mon Jul 11, 2022 6:09 pm
But the easiest solution seems to be just running a large enough negative preamp to ensure that EQ boosts & intersample overs are covered, rather than using AGP combined with oversampling?
If you know the amount of headroom you need, you can get it by decreasing preamp or software gain, yes.
EQenthusiast wrote:
Mon Jul 11, 2022 6:09 pm
And I'm also assuming that AGP + Oversampling still wouldn't cover my EQ boosts 100% anyway.
AGP alone will cover your EQ boost. Oversampling comes into the discussion only when you want to detect intersample overs and get AGP to prevent them.

blaubär
Posts: 3326
Joined: Tue Apr 02, 2019 6:48 am

Re: Automatic Gain Protection (AGP) vs Intersample Overs

Post by blaubär » Mon Jul 11, 2022 6:53 pm

But, whatever, you know what you'll be doing, and the developer might eventually let us customize AGP's latency, so I suggest to terminate this discussion. Good luck with the preamp !

EQenthusiast
Posts: 24
Joined: Sat Jul 09, 2022 7:43 am

Re: Automatic Gain Protection (AGP) vs Intersample Overs

Post by EQenthusiast » Mon Jul 11, 2022 7:48 pm

blaubär wrote:
Mon Jul 11, 2022 6:45 pm
EQenthusiast wrote:
Mon Jul 11, 2022 6:09 pm
I don't really understand oversampling
For a brief description see here. In short: oversampling increases the sampling rate, typically by an integer multiple and in a resource-efficient way.
EQenthusiast wrote:
Mon Jul 11, 2022 6:09 pm
But the easiest solution seems to be just running a large enough negative preamp to ensure that EQ boosts & intersample overs are covered, rather than using AGP combined with oversampling?
If you know the amount of headroom you need, you can get it by decreasing preamp or software gain, yes.
EQenthusiast wrote:
Mon Jul 11, 2022 6:09 pm
And I'm also assuming that AGP + Oversampling still wouldn't cover my EQ boosts 100% anyway.
AGP alone will cover your EQ boost. Oversampling comes into the discussion only when you want to detect intersample overs and get AGP to prevent them.
Thanks for the link on oversampling. AGP doesn't cover the EQ boost 100% though, just as dmitry said earlier (as it's not super strict).....but yeah I'll use a large enough negative preamp instead. (And understood that your oversampling comment was with regard to intersample overs).
blaubär wrote:
Mon Jul 11, 2022 6:53 pm
But, whatever, you know what you'll be doing, and the developer might eventually let us customize AGP's latency, so I suggest to terminate this discussion. Good luck with the preamp !
Yes, indeed, thanks very much for some of your facilitation on these matters, and your replies. I'm pleased I brought up the topic & created this thread, I suppose feedback is always of some use.

Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests