Fabfilter L2 vs. T-Racks (and others?)

[Rockatansky]

There are differences. I think a "mastering" plugin should keep the output spectrum as close to the input as possible. The IK cut happens at around 19 KHz, that's too far down into the spectrum. Just run some noise through your favourite limiter (at no reduction) and see for yourself what happens.

FabFilter Pro-L1 with 4x oversampling @ 0.4% - 0.5% CPU

Screen Shot 2018-04-18 at 23.19.11.png

IK T-Racks 5 (?) Brickwall Limiter with oversampling @ 0.7% - 0.8% CPU

Screen Shot 2018-04-18 at 23.21.00.png

IK T-Racks 5 (?) Stealth Limiter with 4x oversampling @ 1.0% - 1.1% CPU

Screen Shot 2018-04-18 at 23.19.30.png

I'm not the biggest FabFilter fan, but it's clear they know what they're doing.

[Caine123]

thanks so much, i wanna do such tests for myself too to sort out plugins and hear/see what they do, but i think there is missing somethin?
you dont show the input signal dry

[Rockatansky]

You do know what noise looks like?
But oh well, here's your dry input signal, simple Gaussian noise.

Screen Shot 2018-04-18 at 23.53.51.png

And just for kicks, here's my own 4x oversampling implementation @ 0.5% - 0.6% CPU
(Handwritten, not using the JUCE one, not using Muon halfbands.)

Screen Shot 2018-04-19 at 00.02.21.png

[Caine123]

thanks a lot!

so the t-racks plugins would be out of competition, cutting frequencies is a no go in my opinion.

but the fabfilter L2 one doesnt seem so clean in the lower frequency range?`

[Rockatansky]

thanks a lot!

so the t-racks plugins would be out of competition, cutting frequencies is a no go in my opinion.

but the fabfilter L2 one doesnt seem so clean in the lower frequency range?`

Well, I wouldn't say that right away.

You have to see it this way, the filter on the Pro-L1 (not Pro-L2, don't have that one) may look closer to the original spectrum, but since it appears to be a softer filter, it's possible (haven't tested) that when the limiter bends the signal and introduces harmonics in the ultra-high frequencies, some may not be cut off by the downsampling filter and mirror back into the audible spectrum.

With the IK downsampling filter, it's safe to say that you won't run into that problem ever.
(Just to be pedantic: that is, unless the generated harmonics are so far up that they surpass the highest possible frequency in the oversampled signal and manage to mirror back into the audible range at that point already, plus loud enough to not fall victim to the limits of floating point precision. But that would take a lot of serious distortion, I doubt either of the IK limiters could do that.)

None of them is universally "the better one" just judging by this analysis. My intention was merely to point out that there are significant differences, also CPU wise, and one should keep those in mind when making one's choice.

And I doubt that the FabFilter downsampling filter would have any negative effect on the lows, pretty sure that's just a measuring inconsistency.

Noise is noise, which means it's always changing and evolving, so a volatile spectrum graph won't tell you much about the Pro-L2's low-end. Set the scope up to hold the maximum peak value for a longer time, and maybe smooth the graph a bit (could use SPAN for that), the maximum/average curve collected over an extended period of time will give you better insight into what's going on in the lows.

And don't forget, spectrum analyzers use FFT, basically cutting up the real spectrum into little strips from left to right. The strips (also called "bins") in the lows appear to be wider for the same range of contained frequencies than in the highs, since the display of the frequencies is skewed to have the lower/midrange near the center.

[jbarish]

Curious if this is the same with TR5 One or Lurssen????

[Elektronisch]

thanks a lot!

so the t-racks plugins would be out of competition, cutting frequencies is a no go in my opinion.

but the fabfilter L2 one doesnt seem so clean in the lower frequency range?`

You mean no go because you can hear the frequencies bellow 20hz?

Come to think off it might be the reason why Stealth works so good for me, i always cut from 25-30 anyways From my expierience on various festivals and nightclubs with powerful system, you better avoid those if you dont want to loose your guts

[axb312]

No issues with the TDR Limiter 6 GE here so far...

[xx JPRacer xx]

For me the easiest to use with very good result is Boz The Wall. It's also the one that can go the loudest of all the limiters I've tried (if you're into that).

[plexuss]

nevermind lol

[plexuss]

double - server stale (eesh)