Recording two Bass channels

[Thenewexhibit]

I asked something similar to this a while back, but I wanted to get a little more in depth with this.

I’m doing a recording session today to track some bass, and I want to track a clean DI and a bass amp.

What is the best way to do this? Would I set up a chain with a bass amp, cabinet, maybe pedal before, and then set up a split after input 1 that feeds directly output 2? (I’ll be going to an interface via TRS cables)

My concern is phase issues. If I’m not mistaken, if you have a dual guitar amp chain, it can cause phase issues if there are certain or more blocks before one chain and less before another, no?

If that is the case, would that not cause phase issues if you were to have a chain of blocks for a bass amp tone, and then just a split for clean DI with no blocks feeding another output?

How does that work and what are your thoughts?

 

[jamesmarshall]

What you have mentioned is a good way to do it.
You will have to bump the DI forward in time slightly to line up with the effected track, in the DAW. Simple enough and should be easy to see when you zoom in a little on the wave form.

 

[Thenewexhibit]

What you have mentioned is a good way to do it.
You will have to bump the DI forward in time slightly to line up with the effected track, in the DAW. Simple enough and should be easy to see when you zoom in a little on the wave form.

Got ya! So, just for confirmation, this would add additional latency causing phase and timing issues between the two?

 

[jamesmarshall]

Got ya! So, just for confirmation, this would add additional latency causing phase and timing issues between the two?

Anytime you take a DI and also an affected signal you are going to add latency, not much you can really do about it. You don't want to add a bunch of blocks and bypass just to line things up for a DI. I think it is easier to just bump your DI in DAW to line it up with your bass that has the amp on it. You are not talking about much of a bump, it might not even be out of phase, but it might depend on how much latency the interface has.

 

[Thenewexhibit]

Anytime you take a DI and also an affected signal you are going to add latency, not much you can really do about it. You don't want to add a bunch of blocks and bypass just to line things up for a DI. I think it is easier to just bump your DI in DAW to line it up with your bass that has the amp on it. You are not talking about much of a bump, it might not even be out of phase, but it might depend on how much latency the interface has.

Got ya! That was gonna be my next question; if it’s worth using the same block to make sure everything is lined up. In a DAW, it’s easy to realign; you can even use the plugin Auto Align to do this. But, I was wondering; if you play live and take two signals, perhaps it would be worth having the second chain mimicked with the same blocks, only bypassed to make sure you have the same time alignment?

 

[jamesmarshall]

Got ya! That was gonna be my next question; if it’s worth using the same block to make sure everything is lined up. In a DAW, it’s easy to realign; you can even use the plugin Auto Align to do this. But, I was wondering; if you play live and take two signals, perhaps it would be worth having the second chain mimicked with the same blocks, only bypassed to make sure you have the same time alignment?

Also, Live you are not going to have the same latency as in a DAW because there is not going to be another A/D you are passing through. Yes there might be like 1/10 of an ms of delay or something along those lines and you should never notice that live.
Also, DO NOT use auto align, that thing murders audio! Just bump the track in the DAW that is slightly off in time. Using the DI will easily show the transients.

 

[Thenewexhibit]

Also, Live you are not going to have the same latency as in a DAW because there is not going to be another A/D you are passing through. Yes there might be like 1/10 of an ms of delay or something along those lines and you should never notice that live.
Also, DO NOT use auto align, that thing murders audio! Just bump the track in the DAW that is slightly off in time. Using the DI will easily show the transients.

For sure! I’m not too concerned about latency in the DAW as far as monitoring goes; I’m more concerned about phase/time alignment of the clean DI and bass amp combined sound in a live situation and if it even matters; if we’re talking about samples (a small amount can be negligible live), or ms (where a lot can make a big difference sonically.

 

[jamesmarshall]

For sure! I’m not too concerned about latency in the DAW as far as monitoring goes; I’m more concerned about phase/time alignment of the clean DI and bass amp combined sound in a live situation and if it even matters; if we’re talking about samples (a small amount can be negligible live), or ms (where a lot can make a big difference sonically.

You should not notice less than 10ms. The difference between the DI and your "amp" signal will only depend on the number of blocks and types of blocks. The more blocks and the more time based blocks the more latency. It should not be more than 1-3ms difference at a maximum.

 

[Bakerman]

if you play live and take two signals, perhaps it would be worth having the second chain mimicked with the same blocks, only bypassed to make sure you have the same time alignment?

I don't think bypassed blocks add any delay but you can use a block like flanger (100 % mix, 0% depth) or classic enhancer (width = 0 to 20 ms delay on left channel) for precise alignment.

 

[jamesmarshall]

I don't think bypassed blocks add any delay but you can use a block like flanger (100 % mix, 0% depth) or classic enhancer (width = 0 to 20 ms delay on left channel) for precise alignment.

The issue is your DI should be mono, so you should only have a DI with a mono send and not stereo (the L flanger could set that off, unless the output of the block is set as mono and set that way). As long as you don't have some crazy effects on the main effected chain like 2 chorus blocks (one on each side set to mono), and a delay or two... you should notice almost no latency. Test it out for yourself and see. You should feel it right away.

 

[Bakerman]

I was thinking OP intends to blend drive/amp and DI signals. There can be audible tone changes from misalignments well under 10 ms when doing that.

 

[Thenewexhibit]

I was thinking OP intends to blend drive/amp and DI signals. There can be audible tone changes from misalignments well under 10 ms when doing that.

That’s what I’m getting at. 10ms is pretty massive in terms of sonic change and shift in the low end, so that’s why I’m wondering if this is on the sample range or ms range, and if it’s worth using the same blocks with the mixes turn all the way down. Is that a fact with Fravtal that if the block is bypassed, it doesn’t add latency? I thought I read that you should add the same blocks (in a guitar instance) but just have them bypassed if you’re doing two amp setups?

 

[jamesmarshall]

I was thinking OP intends to blend drive/amp and DI signals. There can be audible tone changes from misalignments well under 10 ms when doing that.

No, I said you won't feel latency until About 10 ms. You will be out of phase by about 1.2ms, usually, but could be less depending on signal chain between affected signal and di.

Just think... You use about 7ms when trying to double a guitar across the stereo field.

 

[Bakerman]

Is that a fact with Fravtal that if the block is bypassed, it doesn’t add latency?

Yes, at least on the Axe-FX II. One other option for alignment: compressor engaged at 0% mix, using the lookahead parameter to set delay.

 

[Thenewexhibit]

So, I got to tracking (it’s for my father’s band), and it was an interesting find! The bass chain was input 1 > drive block 1 (bypassed) > Amp block 1 SV Bass 2 (w/ bright on) > Cab block York Audio Ampeg mix 1 > Output 1.
The clean DI came through input 1 and split off to output 2 which was positioned right under output 1.

When recording both channels, the bass amp tone was about 50 samples late, which is to be expected. Now, 50 samples is not even 1ms at 48k, however, it DEFINITELY makes a difference in the midrange clarity and low end tightness when aligned properly. To do this in the DAW, I used Sound Radix’s Auto-Align, and it’s pretty spot on when measuring manually, and works great.

The interesting thing though, is we used a different tone for some other tunes (more of a Paul McCartney type bass tone with a Hofner bass) and the chain remained the same, but the amp and cab changed in a different scene to a Fender Bassman (the 65 Bassguy; Bass or Normal, I can’t remember) and a stock 2x12 Bassman cab with the 67 mic, as this was closest to what we quickly searched on McCartney’s recording tone, and the delay for this came up different, more around 100 or so samples. Wether that’s correct or not, I’m not sure, but the Auto-Align plugin really makes it easy to get the phase and timing dialed in!

I thought I would share my findings!

On that note, it would be awesome if there was a block or parameter developed that allowed you to phase and time align two signals in the Fractal at a sample level with a visual display like how you can in the cabinet block; almost like an Auto-Align style block in the Fractal would be amazing!

For example, (speaking theoretically) take a “send” block of some sort at the very end of the amp chain before it’s output block, and split that to both the DI chain’s output block and amp chain’s output block. Then, in the output block of the DI chain (or less delayed chain for that matter), a new parameter/tab in the output block that has a sidechain input from the “send” block that reads the incoming signal, and then Auto-Align’s the DI (or earlier signal) and then becomes a set and forget after it analyzes your playing, with the option to also fine tune on a sample level.
Sounds mega complex, and very potentially not feasible but just speaking and dreaming

 

[Luca9583]

@Thenewexhibit is there a specific reason for using the interface after the Axe to record the tracks? You could just use the Axe as the interface to avoid two rounds of conversion for both the DI and Amp Block paths.

Another thing to try if your interface is providing a specific sonic benefit is to get a DI box to record the DI and then split from that to the Axe.

Are you using the low end from both your DI and Amp Block paths in your mix? It might work better, after time alignment, to choose one for the lows and the other for mids but i'm sure you've already got that going on.

 

[Thenewexhibit]

@Thenewexhibit is there a specific reason for using the interface after the Axe to record the tracks? You could just use the Axe as the interface to avoid two rounds of conversion for both the DI and Amp Block paths.

Another thing to try if your interface is providing a specific sonic benefit is to get a DI box to record the DI and then split from that to the Axe.

Are you using the low end from both your DI and Amp Block paths in your mix? It might work better, after time alignment, to choose one for the lows and the other for mids but i'm sure you've already got that going on.

Thanks for the reply! I am just using it that way so I don’t have to use an aggregate device. I inly use aggregate when I’m reamping, which doesn’t happen too too often. I definitely have done the clean DI for low end and bass amp for midrange trick! Fab filter Pro q3 with 24db/octave cutoffs work pretty great in post, but I don’t always do that. It just depends on what I’m looking for. I think the McCartney tones I captured for some of the songs will benefit from both being blended together with most of it’s sonic information.

I considered doing the DI before the Axe Fx, but wasn’t sure I could see a benefit to doing that unless I used a specific high end DI box like the Reddi or something, and it seemed like there would be less delay between the two, or at least I assume so.

 

[Luca9583]

Thanks for the reply! I am just using it that way so I don’t have to use an aggregate device. I inly use aggregate when I’m reamping, which doesn’t happen too too often. I definitely have done the clean DI for low end and bass amp for midrange trick! Fab filter Pro q3 with 24db/octave cutoffs work pretty great in post, but I don’t always do that. It just depends on what I’m looking for. I think the McCartney tones I captured for some of the songs will benefit from both being blended together with most of it’s sonic information.

I considered doing the DI before the Axe Fx, but wasn’t sure I could see a benefit to doing that unless I used a specific high end DI box like the Reddi or something, and it seemed like there would be less delay between the two, or at least I assume so.

All sounds good. Yeah i think direct to DI to interface will have less latency from the performance itself than direct to Axe and then to interface, so i reckon the best way to not get the DI late is to either run it DI box to interface or direct to Axe with Axe as aggregate. It's just that extra conversion happening between Axe and interface that isn't needed given that the Axe would be a perfect interface in this case.

Another great trick is do away with a bass amp altogether and blend the DI with a very midrangey guitar amp model and speaker that has no low end, but for the Macca style tones, you probably just want the amp tone alone without a DI. Interesting if we think about pre digital era recordings and phase when blending sources. Before hardware digital delay units existed to line up DIs and amps for analog recording, i suppose it was just a case of flipping phase and removing all the low end from one of the sources.

 

[jamesmarshall]

All sounds good. Yeah i think direct to DI to interface will have less latency from the performance itself than direct to Axe and then to interface, so i reckon the best way to not get the DI late is to either run it DI box to interface or direct to Axe with Axe as aggregate. It's just that extra conversion happening between Axe and interface that isn't needed given that the Axe would be a perfect interface in this case.

Another great trick is do away with a bass amp altogether and blend the DI with a very midrangey guitar amp model and speaker that has no low end, but for the Macca style tones, you probably just want the amp tone alone without a DI. Interesting if we think about pre digital era recordings and latency when blending sources. Before hardware digital delay units existed to line up DIs and amps for analog recording, i suppose it was just a case of flipping phase and removing all the low end from one of the sources.

No real need for latency adjustment back then (minus normal air movement and length from source). You didn't have AD/DA or conversion adding in latency. Tape was instant, same with ADAT tape, and such. The way you lined up a DI and amp is to adjust the microphone until they were in or out of phase, easily heard with headphones on while doing it. Something I did for many years on many different instruments.

 

[Luca9583]

No real need for latency adjustment back then (minus normal air movement and length from source). You didn't have AD/DA or conversion adding in latency. Tape was instant, same with ADAT tape, and such. The way you lined up a DI and amp is to adjust the microphone until they were in or out of phase, easily heard with headphones on while doing it. Something I did for many years on many different instruments.

Exactly