KVR Audio

dangayle · Post by **dangayle** » Wed Dec 20, 2017 4:47 am

bmanic wrote:@rocolin: Just a friendly tip that it's not all that trivial to try to get an insight into how a heavily program dependent processor works with a few rather simple combined test tones. It takes quite a bit more. So what you are seeing doesn't necessarily correlate all that well with actual musical program material.

Been testing both analogue and digital processes for a long time now and I've yet to come to a good conclusion of how to truly test and compare severely program dependent stuff accurately so that you get some kind of correlation to what you hear.

The spec we were referencing earlier says specifically that it doesn’t apply to pure test tones, probably for exactly the reasons you state.

10bd01 · Post by **10bd01** » Wed Dec 20, 2017 6:11 am

Upgraded.

billcarroll · Post by **billcarroll** » Wed Dec 20, 2017 7:21 pm

Upgraded.

FabFilter Pro-L 2 sounds better to me than the iZotope Ozone Maximizer.

I'd much rather support FabFilter than iZotope's constant $ upgrades and inclination to kill products like Iris and others.

dangayle · Post by **dangayle** » Wed Dec 20, 2017 10:17 pm

billcarroll wrote:Upgraded.

FabFilter Pro-L 2 sounds better to me than the iZotope Ozone Maximizer.

I'd much rather support FabFilter than iZotope's constant $ upgrades and inclination to kill products like Iris and others.

Oh man, if FF could find a way to make their plugins communicate with each other like iZotope, that would be killer. RIP Iris.

xx JPRacer xx · Post by **xx JPRacer xx** » Thu Dec 21, 2017 6:25 am

plexuss wrote:If RoundTrip detects more potential for clipping then that is preferable to me because it means its doing a more conservative job which enables me to mitigate clipping better than if i used a tool that was more leanient.

Not only is RoundTrip giving false positives it's also missing some that could cause clipping.

Apple's AAC RoundTrip as used for Mastered for iTunes etc), simply seems to use the example detection filtering (4x oversampling) as described in the ITU-R BS.1770-4 specs. This method is generally considered to be sub-optimal, and can give quite significant under- and overreads.

So it's better than nothing but far from perfect. The FF implementation is just better.

plexuss · Post by **plexuss** » Thu Dec 21, 2017 6:58 am

xx JPRacer xx wrote:
plexuss wrote:If RoundTrip detects more potential for clipping then that is preferable to me because it means its doing a more conservative job which enables me to mitigate clipping better than if i used a tool that was more leanient.
Not only is RoundTrip giving false positives it's also missing some that could cause clipping.
Apple's AAC RoundTrip as used for Mastered for iTunes etc), simply seems to use the example detection filtering (4x oversampling) as described in the ITU-R BS.1770-4 specs. This method is generally considered to be sub-optimal, and can give quite significant under- and overreads.
So it's better than nothing but far from perfect. The FF implementation is just better.

Where's your proof? Let's see some evidence of your assertions:

1. RoundTrip giving false positives
2. RoundTrip... (is) missing some that could cause clipping
3. FF implementation is just better (than RoundTrip)

adl · Post by **adl** » Thu Dec 21, 2017 12:02 pm

Just curious: anybody tried Pro-L2 against Voxengo Elephant 4?

plexuss · Post by **plexuss** » Fri Dec 22, 2017 3:40 am

I'm going to do a live feed later on, on youtube. I am going to cover a feature of Pro-L2 that is unique and for me sets it apart from other limiter - audition - . I am also going to cover:

- a quirk of TBPro ISOL8
- comparison of Hornet Tape and U-he Satin
- a new feature in Logic Pro X 10.3.3 that I really wanted them to implement and they did
- how the different limiter styles in Pro-L2 sound
- and the unique audition feature of Pro-L2 I mentioned

Please join. It will be posted as a video after the live stream. around 11pm EST

https://www.youtube.com/watch?v=2iRZHAbdUBw

TB-ProAudio · Post by **TB-ProAudio** » Fri Dec 22, 2017 7:14 am

plexuss wrote: - a quirk of TBPro ISOL8

Thanks for bringing this to our attention:

This is common to multi-band applications based on LR4 (IIR) filter design.
See here: https://www.gearslutz.com/board/showpos ... stcount=22

This is why we implemented the bypass button

mutantdog · Post by **mutantdog** » Fri Dec 22, 2017 8:32 am

I can confirm that the ISOL8 quirk is common to multiband plugins. I came across the same thing when i did my comparison of mono-bass plugins, those based on a multiband type architecture tended to exhibit the same behaviour (especially with steeper crossover filters) while those based around side filtering generally retained roughly the same peak levels. Definitely something to be aware of when using multiband compression.

xx JPRacer xx · Post by **xx JPRacer xx** » Mon Dec 25, 2017 6:20 am

plexuss wrote:Where's your proof

The proof is that 8x oversampling is more accurate than 4x. Other than that I don't really care about inaudible things.

FabienTDR · Post by **FabienTDR** » Thu Dec 28, 2017 2:50 am

Worth reading this table and the related chapter:

https://www.itu.int/dms_pubrec/itu-r/re ... df#page=22

The True Peak detection validity tests recommended by the docs have pretty large tolerances of +0.2/−0.4 dB.
The general "error" to be expected in resampled TP detection in it's own is surprisingly large at the 4x resampling recommended by the standard. It's beyond 0.5dB!

Measurement error is to be expected, they are inherent to the format. They definitely don't matter much.

Being too pedantic with these measures is classic "bike shedding" according to Parkinson's Law of triviality https://en.wikipedia.org/wiki/Law_of_triviality

bmanic · Post by **bmanic** » Thu Dec 28, 2017 4:53 am

dangayle · Post by **dangayle** » Thu Dec 28, 2017 4:56 am

FabienTDR wrote:Worth reading this table and the related chapter:

https://www.itu.int/dms_pubrec/itu-r/re ... df#page=22

The True Peak detection validity tests recommended by the docs have pretty large tolerances of +0.2/−0.4 dB.
The general "error" to be expected in resampled TP detection in it's own is surprisingly large at the 4x resampling recommended by the standard. It's beyond 0.5dB!

Measurement error is to be expected, they are inherent to the format. They definitely don't matter much.

Being too pedantic with these measures is classic "bike shedding" according to Parkinson's Law of triviality https://en.wikipedia.org/wiki/Law_of_triviality

Thank you for chiming in. Would you be willing to talk a little about true peak detection from your own experience?

Tappistry · Post by **Tappistry** » Fri Dec 29, 2017 1:50 am

plexuss wrote:I'm going to do a live feed later on, on youtube. I am going to cover a feature of Pro-L2 that is unique and for me sets it apart from other limiter - audition - . I am also going to cover:

- a quirk of TBPro ISOL8
- comparison of Hornet Tape and U-he Satin
- a new feature in Logic Pro X 10.3.3 that I really wanted them to implement and they did
- how the different limiter styles in Pro-L2 sound
- and the unique audition feature of Pro-L2 I mentioned

Please join. It will be posted as a video after the live stream. around 11pm EST

https://www.youtube.com/watch?v=2iRZHAbdUBw

Happy to see the Hornet Tape vs. u-he Satin comparison here. I was actually thinking about doing a similar comparison for YouTube land myself. Like a cheap vs. expensive tape plugin comparison.

Update: To my ears, the two tape plugins have very subtle differences. Can't necessarily say that u-he Satin is necessarily better in your example. I wonder if the differences would be more noticeable in a busier track with more instruments and vocals? Considering the price difference between the two, HoRNet Tape seems to be quite capable.

Will watch soon. I'll come back here and edit this post with my thoughts.

Pro-L 2 by FabFilter