Why the Boost in Volume & is it necessary?

[bigedawg]

First, I've been playing for over 20 years in a praise and worship environment. My switch to the Fractal was one driven by the need to find a solution to the stage volume issues that come with a traditional guitar->fx pedals->amp (mic'd) rig. Needless to say, of all the solutions available, Fractal is by far the leader of the pack. However, After using the Ultra for well over a year, there's a few items that still have me scratching my head. In attempt to answer the questions myself, I've decided to back up and start from scratch building a very simple patch. Unfortunately, simplifying my approach hasn't revealed any answers; rather, narrowed down my list of questions.

Simplified Patch...
Input -> Amp Model (Fender Blackface)->Output. I run output 1 thru an ART SLA-2 into Matchless 4x10 open back cab.
Next, I add delay1 block (mono, ingain 100%, Drive 0, Mix 35%, Level 0db) in parallel and there's a sudden boost in output volume.

Here's my Questions....Why and does it have to function this way? Each Additional fx block added in parallel brings additional boost in output volume.

I've got additional questions regarding the functionality of the delay blocks, but I'll ask those in the context of a new post.
Any insight is much appreciated!

Regards,
Chris

 

[FractalAudio]

If you add effects in parallel you need to set the mix to 100% and then adjust the level to get the desired effect amount. Otherwise you're adding direct signal to direct signal which will be louder.

 

[bongo]

A parallel path increases the volume because you are effectively doubling the amount of input seen by the amp stage. If you want the parallel paths to arrive at the same level as would a serial path, you need to bring down the levels such that they add up to 100%. You can use the level setting on the mix pages in each path, or you can route the paths through a mixer block before the amp.

You can also then assign the path levels to an expression pedal with opposite slopes, which allows you to keep the input level consistent while changing the mix. ( also a useful trick for amp morphing. )

 

[bigedawg]

Cliff, thanks for the quick response. I apologize, I provided inaccurate information. You're correct and I did set the mix to 100% when in parallel and adjust volume. The settings I provided are when I place the fx block in series, which is critical to my questions (not posted yet) regarding the delay block functionality. Again, I apologize.

I guess my question when placing an fx block, any fx block not just delay, in parallel why does it introduce an increase in output volume (6-8db)? This gain in volume is present with the fx block bypassed or active and only occurs when the an signal is run thru the fx in parallel? When placed before an amp block, the increase volume is doesn't appear to increase input signal to the amp and thus the amp block responding like it's slammed with a signal boost. The result appears to be an increased in output volume. If I place the fx block in series as apposed to parallel, there's no increase in output volume once the fx block is introduced into the signal chain.

Sorry if I'm not explaining it accurately. I can attempt to post the patch to clarify.

Here's what my routing looks like
(from 1st shunt)Delay1-> (connect to shunt b4 amp)
input ->Shunt -< shunt -> shunt ->Amp ->Output

 

[stevieray]

Cliff, thanks for the quick response. I apologize, I provided inaccurate information. You're correct and I did set the mix to 100% when in parallel and adjust volume. The settings I provided are when I place the fx block in series, which is critical to my questions (not posted yet) regarding the delay block functionality. Again, I apologize.

I guess my question when placing an fx block, any fx block not just delay, in parallel why does it introduce an increase in output volume (6-8db)? This gain in volume is present with the fx block bypassed or active and only occurs when the an signal is run thru the fx in parallel? When placed before an amp block, the increase volume is doesn't appear to increase input signal to the amp and thus the amp block responding like it's slammed with a signal boost. The result appears to be an increased in output volume. If I place the fx block in series as apposed to parallel, there's no increase in output volume once the fx block is introduced into the signal chain.

Sorry if I'm not explaining it accurately. I can attempt to post the patch to clarify.

Here's what my routing looks like
(from 1st shunt)Delay1-> (connect to shunt b4 amp)
input ->Shunt -< shunt -> shunt ->Amp ->Output

What is your bypass mode? If it is one that passes dry signal when bypassed you'll be summing parallel paths. Also, the amp block is mono so there is summation of L + R. IMHO the biggest drawback to the layout screen is that it offers no clue what is taking place mono vs. stereo or what the path through the effect algorithms are. You really have to know your algorithm block diagrams if you want to reign in all the details of the levels. It can be confusing at times. I've about pulled my hair out with the looper levels!

 

[bigedawg]

A parallel path increases the volume because you are effectively doubling the amount of input seen by the amp stage. If you want the parallel paths to arrive at the same level as would a serial path, you need to bring down the levels such that they add up to 100%. You can use the level setting on the mix pages in each path, or you can route the paths through a mixer block before the amp.

Thanks! That makes sense (kinda / sorta), but my question is more why and does the axe have respond to parallel routing this way? In traditional environment, does the signal level double when running fx in parallel? I guess I'm expecting to insert an fx block in parallel @ unity gain, but that's not how the axe responds. I don't understand why and I don't have a solution without introducing more fx blocks or tweaking the output volume each time I add/subtract a new parallel block.

Also, with the fx block run in parallel, the level setting on the mix page is only changing the effect level because I've set the mix to 100%, no? Also, this volume increase remains with the parallel fx block in bypass mode. If I understand the functionality correctly, all controls within the fx block are bypassed when the fx block is bypassed. Therefore, If I adjust the level @ the mix page to compensate for the increase volume, the adjustment is only effective when the fx block is active. Is there a level control in an fx block that is independent of the bypass mode?

I really do appreciate the input. This is one of the few items that have puzzled me since I started on my axe journey. I thought with time, I'd figure it out. But I haven't, yet. The reason it creates an issue for me, is I start building my patches from the ground up (input-amp-output). I work real hard, probably too hard, to get the amp model sounding right to my ears. Once I'm satisfied, I then add fx to taste. When I start adding fx, I'd like the output volume to remain constant so I can focus only the fx being added. With the increase in volume due to parallel routing, I'm now chasing volume levels (which dramatically influence the sound) AND trying to add/tweak the fx being added. Just complicates the process for a methodical/process driven person like myself.

 

[steverosburg]

Thanks! That makes sense (kinda / sorta), but my question is more why and does the axe have respond to parallel routing this way? In traditional environment, does the signal level double when running fx in parallel? I guess I'm expecting to insert an fx block in parallel @ unity gain, but that's not how the axe responds. I don't understand why and I don't have a solution without introducing more fx blocks or tweaking the output volume each time I add/subtract a new parallel block.

Well, in the "real" world, splitting the signal chain in the first place would reduce the signal strength (assuming no signal amplification is involved), but that doesn't happen in the Axe-Fx. It helps me to think of it more as if they were "sound" blocks rather than electrical signals -- the more parallel blocks you have feeding the next link in the chain, the stronger ("louder") the signal. Obviously, this is over-simplified since you have to figure how much wet vs. dry signal is being added, but generally whenever I'm combining multiple blocks into one without using the mixer, I reduce the level on each block accordingly.

 

[bigedawg]

Well, in the "real" world, splitting the signal chain in the first place would reduce the signal strength (assuming no signal amplification is involved), but that doesn't happen in the Axe-Fx. It helps me to think of it more as if they were "sound" blocks rather than electrical signals -- the more parallel blocks you have feeding the next link in the chain, the stronger ("louder") the signal. Obviously, this is over-simplified since you have to figure how much wet vs. dry signal is being added, but generally whenever I'm combining multiple blocks into one without using the mixer, I reduce the level on each block accordingly.

I'm not sure I'm traveling with you on the first part, but ok. Again, intuitively I'm expecting the parallel fx block to function @ unity gain, but the axe doesn't. Maybe that should go on the wish list, but that's a different topic all together.

So what I need to know is...is the increased signal output a fixed amount? If so, I can compensate at the final output level the fixed db x # of parallel fx blocks. Also, when in parallel, what's the recommended bypass mode?

Thanks!
Chris

 

[DNW]

What he said about splitting and reduced volume is right as far as I know. Unless you've got something active in the path that will split the signal without reducing volume, as passive split can only take that initial signal and share the level across however many splits. Hence a decrease in volume.

In terms of the Axe, think of it this way. If you run one row of shunts start to finish, you've got your raw signal coming through. If you run shunts across all 4 rows start to finish, you're getting four lots of that exact same signal passing from the input to the output at the same time, hence they will stack up and each add to the volume.

The same will happen if you add another row part way through for a parallel effect. The signal will split (in the non-volume dropping way), and the signal passing through that new chain will stack on top of your original sound. If you run that parallel block at 100% wet though, you will have only the processed FX signal passing through the block and zero dry signal. With no dry signal passing through the FX block, you shouldn't get an increase in volume, unless say you're running your delay volumes so they come out louder than the initial sound. But your dry signal should not be affected by the level of a 100% wet block as there is no dry signal coming out of it to cause that effect.

When you bypass that block, you want it to be set to one of the mute functions. That way when the block is muted, it doesn't just pass the dry signal through where it would sum with the dry path and increase the volume. Set to thru, the block will effectively be acting like a shunt, so that path would basically be acting like a doubling of the dry path, hence the increase in volume.

 

[Jay Mitchell]

Again, intuitively I'm expecting the parallel fx block to function @ unity gain,

And it does. A parallel signal block has two paths, however, and both of them are at unity gain. At the point where the parallel paths meet, the two signals are added to (i.e., mixed with) each other. If the signals are identical - e.g., if you place a shunt in parallel with another shunt - the result will be a signal of twice the original amplitude (1+1=2), or +6dB. If you place an FX block in parallel with the dry path and set the mix to anything less than 100%, dry signal will be added when you activate the block.

The trick with parallel signal paths has two parts: the effects mix should be set to 100% - so you're not adding dry signal to the dry path that's in parallel with the block - and the bypass mode should be set to something other than "Mix=0%," because that setting would create a parallel shunt (i.e., boosting the signal by 6dB) when the block is bypassed.

but the axe doesn't.

It does, you just have to recognize that unity plus unity = twice unity.

Maybe that should go on the wish list,

This has come up a few times. It most definitely does not need to change. A parallel signal path in the Axe-Fx works exactly as any active parallel analog path would work.

 

[bigedawg]

Thanks Jay/All!
At the risk of being redundant, I don't completely understand the why(s), but I don't really have to. That being said, I understand it better now than I did before I started this post.

1) I'm good with the mix % needing to be 100% when i parallel. I understand that less than 100% introduces dry signal
2) I better understand the summing effect of the parallel signal. Not 100%, but I'll study up on my own.
3)The sum'd parallet = +6db is exactly what I need.
4) The bypass mode was likely adding more confusion than I realized. I did originally have it set to mix=0%. As this topic developed, I was working with my axe, started changing the bypass mode and auditioning the results. I think I've landed on mute fx-in for the delay fx. This allows the delay spillover effect I'm looking for.

Again, thanks to everyone who contributed!

Regards,
Chris

 

[Smilzo]

What he said about splitting and reduced volume is right as far as I know. Unless you've got something active in the path that will split the signal without reducing volume, as passive split can only take that initial signal and share the level across however many splits.

Anyone of you have ever heard of passive delay stompbox? Or passive chorus stompbox or rack fx? If you add two parallel path you double the signal level (minus loss/impedance problem). The parallel works just like it does in real life.

 

[yek]

4) The bypass mode was likely adding more confusion than I realized. I did originally have it set to mix=0%. As this topic developed, I was working with my axe, started changing the bypass mode and auditioning the results. I think I've landed on mute fx-in for the delay fx. This allows the delay spillover effect I'm looking for.

Just setting Bypass mode to (f.e.) Mute In solved all volume issues for me when running fx in parallel at 100% mix.

 

[hippietim]

A parallel signal path in the Axe-Fx works exactly as any active parallel analog path would work.

Not exactly, the Axe-FX doesn't have the phase problems that you normally get :lol:

 

[Jay Mitchell]

A parallel signal path in the Axe-Fx works exactly as any active parallel analog path would work.

Not exactly, the Axe-FX doesn't have the phase problems that you normally get :lol:

I know your quip is tongue-in-cheek, but a parallel analog signal path doesn't create latency ("phase") problems. It's only when you introduce conversions (i.e., one or more parallel branches are digital) that the possibility for interference becomes a significant issue.

 

[steverosburg]

What he said about splitting and reduced volume is right as far as I know. Unless you've got something active in the path that will split the signal without reducing volume, as passive split can only take that initial signal and share the level across however many splits.

Anyone of you have ever heard of passive delay stompbox? Or passive chorus stompbox or rack fx? If you add two parallel path you double the signal level (minus loss/impedance problem). The parallel works just like it does in real life.

Yes, combining parallel signal paths in a passive system will increase your signal strength, but you first have to split the signal, which will weaken it commensurately, excluding any system loss. In the Axe-Fx, when you split the signal into parallel paths, there is no such drop in signal strength for each branch as there would be in a passive system. All we were trying to do was help the OP understand why he needs to reduce the volume when adding parallel blocks. I think that has been achieved.