Dyn8 sounds phase-y to me

dLive dLive Troubleshooting
1

Hey all,
Does the Dyn8 sound “phase-y” to anyone else. Kind of a rhetorical question since I’m probably using it wrong. LoL. I’m trying to use the EQ portion as a d-esser but I just can’t get it to sound right.
Thoughts?

2

Using the Dyn8 does add a tiny bit of latency to the audio path. This alone is not a problem unless you also have a parallel path without the Dyn8 inserted. If you do have an identical parallel path, then you will need to account for the latency or you will get some phasing issues.

“Inserting a Dyn8 adds 4 samples on input / bus processing (42us).”

https://support.allen-heath.com/hc/en-gb/articles/25863491742353-Dyn8-and-FX-latency-dLive-Avantis

1 Like
3

AI answer that goes into more detail than I could in 2 minutes:

the Dyn8 dynamic EQ on the Allen & Heath dLive can introduce a “phasey” sound, especially when used aggressively or improperly. This occurs because Dyn8 employs minimum-phase filters, which inherently introduce phase shifts around the affected frequencies. These shifts become more noticeable with narrow Q settings, significant gain changes, or overlapping frequency bands.​

Minimizing Phase Artifacts with Dyn8

To reduce the “phasey” effect:

  • Use Gentle Settings: Opt for wider Q values and moderate gain adjustments to minimize phase distortion.
  • Avoid Overlapping Bands: Ensure that dynamic EQ bands do not overlap excessively, as this can exacerbate phase issues.
  • Careful Threshold and Ratio Settings: Set thresholds and ratios to apply processing only when necessary, avoiding constant or aggressive adjustments.
  • Bypass and Compare: Regularly bypass the Dyn8 to compare the processed and unprocessed signals, ensuring transparency.​Mastering Box

Best Practices for Using Dynamic EQ and Multiband Compression

  • Target Specific Issues: Use dynamic EQ to address specific problems like sibilance or resonant frequencies, rather than broad tonal shaping.
  • Combine with Static EQ: Employ static EQ for general tonal balance and reserve dynamic EQ for dynamic issues.
  • Mind the Signal Chain: Place dynamic EQs thoughtfully within your signal chain to prevent unintended interactions with other processors.
  • Monitor Latency: Be aware that inserting Dyn8 can introduce latency (~0.2 ms), which may cause phase issues if parallel paths are not properly time-aligned.
2 Likes
4

smacks forehead
Brilliant! I gotta remember the chat gpt reference library.

Thanks :smiling_face_with_sunglasses::+1:t3:

5

Other than what ‘Mr. AI’ said, you could also consider changing to the 3 band version of the comp for less x-overs (use the setup button), try the different slopes (6, 18, 24), and if you are only using the comp or the dynamic eq, disable the bands on the section you are not using.

Personally I only use the 3 band version when inserting the Dyn8 on ‘complex sources’ such as the main bus etc.

Ohhh, and never use it on ganged stereo sources. Only on true stereo or mono channels/busses.

1 Like
6

Thanks, runes.

Yeah it’s been a ‘best practice’ for me to disable bands I’m not using. I have all the EQ functions disabled since I’m really just using the comp as a d-esser.

7

I’m surprised that you are using the dynamic compressors as a de-esser. Personally I cannot get the compressor bands narrow enough to keep from negatively affecting the rest of the voice. For this reason, I use the dynamic EQ section when I need a de-esser. I can use much narrower cuts this way and really dial in the exact sibilant frequencies for each vocal without cutting a bunch of other frequencies that are not sybilant.

Perhaps what you are hearing isn’t really a “phase” issue, but just an overall loss of the higher frequencies because the dynamic compressor is cutting too wide of a set of frequencies…

1 Like
8

That very well could be. I’ve switched over to the EQ (disabling the comp) and I think I’m getting better results. The idea of using the comp came from a visiting engineer who was using it as a d-esser and as a way to control LF booming.

9

“Controlling LF blooming” would be a good use of the multiband compressor because you would generally want to control a pretty broad range of frequencies. For example, not just 195Hz, but maybe 80-350Hz. The multiband compressor would be a great option for that use case.

On the other hand, with de-essing I might want to hit just 4.6kHz and just 8.3kHz (I totally made those number up for this example) in which case the surgical precision of the multiband eq (set with a narrow q) works better IMHO.

For a generic de-esser (for a mic that is used by multiple people for example), I will often times use all four bands of the multiband EQ set with the narrowest Q possible, at 4.5kHz, 6.5kHz, 8.5kHz, and 10.5kHz. This way my “total coverage” is where most people have their sylibance, but there would never (OK rarely) be a time where a person would activate all four bands of the multiband EQ. So this way the de-esser should capture most sibilance, but only cut the offending frequencies and to the other high end frequencies.

Just keep in mind that none of this is “set and forget”. You will need to watch and adjust things (especially the thresholds) based on each speaker.

2 Likes
10

Yeah I’m starting to get in the flow of the Dyn8 now. Thanks everyone.

We are currently doing a conference so there’s a lot of adjusting and tweaking as we go. No “set and forget” here for sure. Multiple mic types and presenter types as well.