Note tails when "re-amping"

[Adam Inglis]

I'm engineering and playing on a track that requires a fairly heavy rock guitar sound (not something I do a lot of) and after weeks of lacklustre results I'm realising what many others have found: it's difficult to judge the amount processing/distortion required for the final mix when you have your artist hat (er, phones) on!
So I've been doing my first experiments with 're-amping'.*
I record the "vibe" track - distortion, modelling, kitchen sink effects etc - in stereo, while simultaneously recording a clean DI'd version at modest levels.
The 'vibe' track helps me get the performance I need, and then later, with my mix engineers hat on, I send the mono track back to my rig to fine tune the sound.

I'm getting very usable results, and the scope for experimentation is vast, however, I notice that the note tails on my 'vibe' track just go on and on as you'd expect with any hi-gain processing chain, but when the DI track gets processed, the tails die away very quickly. So wailing lead breaks or sustained power chords etc sound quite different.
It doesn't seem to be a level related thing, so I'm wondering if it's an impedance thing?
I realise there are a bunch of tricks I could try to get around this, such as compressing or pre-processing the DI track, or adding audio edits to it etc etc but I'm wondering why it's not "seen" by the rig in the same way that it "sees" the live guitar?
i.e. do I really need a dedicated re-amping box to pull this off?

Thoughts?

* perhaps the term is technically not correct here. My 'rig' such as it is, involves a Sansamp Bass DI at the front (1 megOhm input. Footswitch is off, so it's not doing any stompbox stuff, just acting as an active DI). From here there are three outputs:
1) a hardwired thru from this input goes to a regular guitar Sansamp followed by a bunch of pedals going to a little submixer
2) The 1/4 inch output from the DI box goes to a Voicelive 3 guitar input (1 megOhm) for all the modelling and FX stuff, the output of which also goes to the submixer, and the stereo sub-mix output is recorded to the "vibe" track
1) XLR output from the DI box goes to a mic pre input on my MOTU 828 interface, recording to mono around no more than -12 dB peaks.

[Wonks]

It’s not an impedance thing. You’re dealing with active outputs which are fairly unconcerned about impedance. At worst, you’d get some EQing of the signal, but not a curtailing of the signal itself.

I’m a bit confused as to the recording chain. Is the ‘DI box’ the bass Sansamp or the guitar Sansamp or is there a separate DI box involved? What does the little submixer do?

Can you do a drawing, take a picture and post that?

Are you playing along via headphones or monitors or a normal guitar amp when recording?

How are you doing the reamping? With a miked amp or just software emulation in the DAW?

Is the reamping involving the same hardware effects chain or a different one or emulations of those in software?

I’d suggest that it’s probably a level thing, with the signal not hitting the effects hard enough. But as it’s all a bit vague as to what you are doing, that’s a wild guess.

[Wonks]

After reading about the Sans Amp module, I think I understand a bit more about the direct recording chain and the XLR output direct sound, but the two recording chains into the mixer leave me a bit baffled. Are you going for a wet/dry/wet mix?

Still need to know how the reamping is achieved.

[amanise]

That seems quite a complex arrangement to me. But that's just because I like simple things. For example, I wouldn't know (if it were me) whether I was making my mental comparison of the dry signal reamped against 1) or 2) above. I'd need to know that - again, that's just me.

There is also - as you're reamping - why make the comparison? Why not go for sounds from scratch with whatever the dry signal sounds like? Unless there's something you really like about 1) or 2) - why not just forget them and start tweaking all over again as part of the reamping process? Clear the decks, so to speak. Apply the IIWII (it is what it is) rule of zen defiance and build all new signal chains? It's not like you've got to get the vibe and play it all again - you've done that bit. Time for some knob twiddling fun!

[Adam Inglis]

After reading about the Sans Amp module, I think I understand a bit more about the direct recording chain and the XLR output direct sound, but the two recording chains into the mixer leave me a bit baffled. Are you going for a wet/dry/wet mix?

Still need to know how the reamping is achieved.

Well, that's why I put "re-amping" in inverted commas! It's really "reSansamping" ™

The submixer is simply a way of controlling what mix of pedals and Voicelive 3 modelling and FX actually goes to the record inputs. And that mix is what I'm calling the "vibe" track.

The "re-amping" is happening thus: the mono DI'd track in Cubase is assigned a dedicated analogue output on the MOTU interface (line level) and this plugs in to where the guitar was previously plugged in to the DI box (Sansamp Bass DI) and hence going out to those 3 outputs as delineated above.

Yes sorry, upon review it does seem rather complicated but this system has evolved over a couple of years to enable me to sing, play guitar, do both, and / or record all or part of the above at short notice!!

thx
A

[Adam Inglis]

That seems quite a complex arrangement to me. But that's just because I like simple things. For example, I wouldn't know (if it were me) whether I was making my mental comparison of the dry signal reamped against 1) or 2) above. I'd need to know that - again, that's just me.

There is also - as you're reamping - why make the comparison? Why not go for sounds from scratch with whatever the dry signal sounds like? Unless there's something you really like about 1) or 2) - why not just forget them and start tweaking all over again as part of the reamping process? Clear the decks, so to speak. Apply the IIWII (it is what it is) rule of zen defiance and build all new signal chains? It's not like you've got to get the vibe and play it all again - you've done that bit. Time for some knob twiddling fun!

Thanks Adrian, well yes, it's complicated... the price you pay for flexibility sometimes!!
But it's the issue with sustained notes - yes I can "clear the decks" but I really like those long sustained notes in certain areas... I suppose I could enlist a soft synth to add some of those notes on that bus, to fool the processor chain into "seeing" a smooth tail... I'm never adverse to cheating like that

[Wonks]

How are you setting the output level from the MOTU line out? Are you making sure it’s the same level as the original guitar signal?

[Adam Inglis]

How are you setting the output level from the MOTU line out? Are you making sure it’s the same level as the original guitar signal?

Aah... great question. Actually that probably gets to the heart of the matter.. I'm not really sure how to set this level, I'm basically doing it by ear.

But there didn't seem to be much difference in response despite large variations in output level - say, something like a 18 dB variation! - hence my assumption that it was not a level issue. But perhaps I need to look closer at this..

[The Elf]

Regardless of why you're getting these 'note tails', using the DI as a gate side-chain to the distorted track might give you a cleaner result.

[Adam Inglis]

Regardless of why you're getting these 'note tails', using the DI as a gate side-chain to the distorted track might give you a cleaner result.

Thanks Elf, but I'm not sure I want a "cleaner" result...
The note tails I'm talking about are when you have a decaying note in a high gain, highly-processed setting, there is a slow change in distortion and frequency spectrum as it decays so that you approach almost a sine-wave... the classic electric guitar feedback sound.
That is what is happening on the 'live' guitar recording but I can't seem to get it on the DI guitar 're-amped' track, despite going through what appears to be the identical signal chain.

[amanise]

But it's the issue with sustained notes - yes I can "clear the decks" but I really like those long sustained notes in certain areas... I suppose I could enlist a soft synth to add some of those notes on that bus, to fool the processor chain into "seeing" a smooth tail... I'm never adverse to cheating like that

You could have them from your previous recording session, on a track which you faded up with an envelope so they came up under the new processed dry track. That way you could pan them as well and have them coming up on the opposite side too. I do that a lot with vocals which I record in stereo and delay+flange one side and megaphone the other. Then I bring up some of the flanged side under the megaphone side just to get the nice tails . If you use a stereo digital delay synced to the tempo you can have one side with a longer delay than the other as well. Loads of fun, all in the box after the initial capture. Not everyone's cup of tea - but then obscurity is my thing as well.

[Wonks]

For level setting, feed the XLR DI from the bass Sans Amp into a Mic input on the small mixer (no other feeds) and set the input gain, channel and main output so the main meters are peaking at zero (or whatever you can get them to peak at).

Leave everything as it is but feed the Bass Sansamp from the interface line out playing a recorded clean DI guitar signal. You should now be able to see where the signal is compared to the guitar signal.

I have a feeling it will be significantly quieter than the guitar signal. As a result the signal won’t be compressing as much when put them through the drive pedals, so you won’t get the same level of sustain.

You can raise the gain of the output in the DAW, but you don’t want it peaking above say -1dBFS on a true peak meter. If you need to raise the level more, then you’ll need to use some gain in the analogue domain e.g. a clean boost pedal between the interface and the bass Sans Amp.

[Adam Inglis]

OK, well this is embarrassing.
Some IDIOT left a noise gate on the Cubase output channel for the DI track.
I don't know who it was, but when I find them, heads will roll :::facepalm:::

OK, switching the noise gate off, it's sounding much more as expected now. And you were right Wonks, it needed a lot more level on the DI output out to the rig.
I'm at zero on the output fader, and I could perhaps use a few more dB.
One option would be to switch the Sansamp DI into stompbox mode, and use the Blend pot for balance.. anyhoo, fun times ahead!
Many thanks all