There are different values when measuring the loudness (LUFS, TP) in Renoise and outside of Renoise. Internal measurement is lower by round about 1 dB than external measurement with the same meter, whether it’s LUFS or TP. How come? I’m using Youlean Loudness Meter 2 inside and outside of Renoise. Songs that have -9 LUFS and -1 TP within Renoise do have around -8 LUFS and +0.2 TP outside of Renoise. Just why? Both wave and mp3 are affected, so it seems to be unlikely that it’s caused by compression. Any ideas?
1 Like
interessant question
I’m not good for mastering
I think free plugins are less reliable for this task
But “Melda” is reliable I think
Lossy compression could affect the integral ‘in the absolute’
I’m taking a wild guess here…
But the TruePeak measurement you see in Youlean inside Renoise refers only to the plugin output and does not take into consideration what happens at the Master fader output - which will be based on your output samplerate. This means that, depending on your samplerate, you might still get ISPs on Renoise output stage - which you can clearly see by having Youlean monitor Renoise’s output!
The samplerate will also affect Nyquist Frequency and, therefore, the amount of aliasing produced - which I believe can contain said ISPs (and correct me if I’m wrong).
This is actually something I have noticed since I started mastering: I’d have a clipper or limiter when exporting my song to master in Reaper, and set it at -1db TP. Then, when loading at Reaper, it almost always shows values above -1db tp.
That said, did you try other samplerates?
-
(A bit of rambling about True Peak, but) I honestly really don’t care about this small peak difference, mainly for 3 reasons:
-
When rendering the final masters in Reaper they will sit perfectly at -1db, which is what I usually aim for (even though CD can easily handle -0.3db).
-
Even if you do Master in Renoise, most modern consumer-level devices can easily handle up to 3db (occasional) true peak clipping without being damaged. So even if Renoise is giving you +0.2db TP, it’s not much of a big deal. Most plugins didn’t even use the concept of True Peak until like 5-10 years ago, and oversampling was extremely difficult a few decades before.
-
Forgot to mention this, but your song may still suffer soft clipping starting way under 0db when played, depending on the medium, making 0.2db TP even more irrelevant
ISP is for analog input
True peak is just a way of measuring intersample peaks, which happen in the digital realm in between taken samples.
PS: …and can produce “overshoots” in the dac
no.
Source: Audio Engineering Society (AES77-2023)
The terms are even used interchangeably
By the way, this is a very good video to understand it.
This is Skrillex - Bangarang, as show in the video:
I would never master this loud and don’t like Skrillex, but just showing you an example.
At 24bit…144db dynamic
-0.6db (-0.41%) from max volume and you get rid of
less than a true headroom
-3db at 24bit will be 23.5bit
Minimal headroom…-0.53db
true peaks go upper
I don’t like his music but I respect him
If you amplify the signal to get a shorter headroom after final mix via a bad algo you will have distortion
32bit export will reduce the distortion
Ok, I’ve noticed a false setting in the master fader, which were set to 1 dB for whatsoever. I didn’t do it myself. Anyway, I’ve corrected it back to 0 dB and at least the LUFS values of internal and external measurement only differ by 0.2 LUFS now. BUT:
Ok, makes sense, but shouldn’t the last device in the chain of the master track be the “crucial” one? It has to be possible somehow to render exactly what you’ve measured, right?
@taktik My suggestion: A native loudness meter after the master fader.
Same result for 44.1 and 48 KHz.
Internal measurement TP -1, external measurement TP +0.2.
Out of curiosity I checked what happens if you set TP to -2 internally. Comparison:
LUFS -9, TP -1 (internal) → LUFS -8.8, TP +0.2 (external)
LUFS -10.2, TP -2 (internal) → LUFS -10.2, TP -0.5 (external)
So basically it doesn’t work out. LUFS too quiet and TP still too high.
As far as I know ISP is just another term for TP. Inter Sample Peak = True Peak.
2 Likes
Why have ‘auto gain’ turned on after meter?
1 Like
In theory, yes, but more stuff happens during render stage, including dithering, resample aliasing, plug-in offline oversampling, etc… I just don’t know if that would be enough for a 1db difference.
This really caught my curiosity 
Auto gain is always on, but at this point it doesn’t do anything anymore. When I start mixing I use it to determine the max gain. I could turn it off now, but the result wouldn’t differ.
I don’t know either, but there has to be a way to get the “right” values after rendering. The closest I can get is this (internal left, external right):
There’s always a difference of 1.2 or 1.3 dB (TP).
Are you always talking about LUFS?
If you modify the waveform,There is 99.9% of chance that the integration will change
since the time,I know more or less you @TNT 
