Jump to content
Sign in to follow this  
radatats

Mixer Blasphemy!

Recommended Posts

OK so while we have all been playing around with different inputs, outputs, topologies and other settings I decided to work a little with the mixer.  After really studying perperra's schematics and understanding what and how the blocks work, I started to question why the mixer always seems panned 100% left and right.  Seems that pretty much every patch I download and even the ones it came with are all panned left and right.

 

The Pan knobs in the mixer actually function more like Balance knobs for Path A and path B separately.  Knowing that all amps are MONO blocks, it didn't really make sense to me why there would be any need to pan the amp outputs.  All it seems to do is weaken or thin out the signal to my ears.  What I am essentially doing is feeding half a stereo signal to each side of the mixer rather than a full stereo signal.  If the amp ouputs an equal stereo signal on each of the Paths, then by panning hard left and right I am losing some of that underlying body and resonance that makes up the full signal.  A real amp feeding two cabinets would be sending the same signal to each.  I am counting on the FX blocks in the Post section to recombine the full signal at some point or trusting that my headphone/speaker will produce the full signal from the two left and right components.

 

I set up a patch to test it out using fixed Vol pedal and Brit 75 full amp model in the Pre section, reverb and stereo delay in the Post section. I copied it to another slot and set the mixer on one to 0 and the other 100 left and right.  I used my Shure 440 headphones rather than the DT to get a true stereo effect.  It is a given that the 0 setting is a bit louder than the 100 setting since 100 cuts part of the signal in the process.

 

In testing, to my ears, I definitely found the 0 mixer setting to be fuller, with more articulation and pick attack response.  As expected, the stereo separation from the stereo delay was less pronounced but seemed clearer.  The 100 L/R setting was softer and seemed to be missing certain elements like thumb harmonics, but I could definitely tell the separated delay parts in each ear.  It did not seem to have the same life as the 0 setting.

 

One should also note that if you place a mono block in the Post section for any purpose, whatever effect you were trying to accomp0lish with the mixer is lost.  All stereo separation is lost with the mono output. Even the stereo blocks in the Post section were not able to recombine the two signals well enough to recover the missing mojo...

 

So now what?  I am going to go back through my patches and see what I might have been missing just by needlessly panning in the mixer.  Unless there is a specific reason to completely separate the two Paths I may be better off with  0 setting in all cases.  Especially for single amp patches...

 

Any other ideas?  Are there other needless things we do out of habit that really do nothing to enhance our tone?

Share this post


Link to post
Share on other sites

The only thing you gain by setting the balances for A and B to 0 with the amp in the pre position is a 6dB boost from the exact same A and B signals summing, which might even clip the D/A.  The mojo is imagined.

Share this post


Link to post
Share on other sites

That "extra" that you're hearing with them together is just volume dude. The idea of the mixer being able to be fully panned means you can run two totally separate chains with no crossover between the two. If you're running a single amp, or two amps in a "layered" tone, widening the spread wouldn't help you, and it probably would sound "thin". Try building a patch with one amp in Path A to how you like it, then mute it and do the same in Path B. If you combine these two its probably gonna be a clipping, farty mess, but if you spread them full out, you get that "full" sound from both amps, but separated.

  • Upvote 1

Share this post


Link to post
Share on other sites

I expected those responses that it's only volume but I disagree...  you have to look at perperra's schematic blocks and understand how the mixer works.  If it simply panned left and right what you say might be correct but that is not what it does.  It actually functions as a balance circuit for each path before the signals mix.  By turning Path A full left you are not putting the full path A signal to left, you are cutting the right half of path A before it hits the gain section.  Same for Path B.  What you end up with is the left half of Path A and the right half of Path B.  You don't get any more separation, it just sounds more distinct because only half of each signal is being passed to the Post section for processing.  Naturally, if you leave it at 0, it is a stronger signal.  Not because it has a 6db boost, but because it has not been cut...

 

I always thought it should be a fatter signal at 100 but it really isn't.  I have some dual amp patches and they actually sound better at 0. Definitely not farty.  And in the end, what are you feeding?  Do you feed a true stereo power amp into separate cabs with stage separation?  Are you feeding the FOH/PA?  If so then folks on either side are missing part of your sound!  And if you are feeding a guitar amp/cab or a DT what difference does stereo separation make in the end?  The power section outputs in mono.  Stereo wiring in a single cab doesn't really give the full stereo effect either.

 

As I said before, unless you have a specific need for two absolutely separate paths, try the mixer at 0.  Your stereo FX blocks after the mixer will still perform their stereo functions.  For recording you can use true panning within your DAW. 

 

http://line6.com/support/topic/2033-pod-hd-500-500x-new-routing-schematics/

Share this post


Link to post
Share on other sites

I expected those responses that it's only volume but I disagree...  you have to look at perperra's schematic blocks and understand how the mixer works.  If it simply panned left and right what you say might be correct but that is not what it does.  It actually functions as a balance circuit for each path before the signals mix.  By turning Path A full left you are not putting the full path A signal to left, you are cutting the right half of path A before it hits the gain section.  Same for Path B.  What you end up with is the left half of Path A and the right half of Path B.  You don't get any more separation, it just sounds more distinct because only half of each signal is being passed to the Post section for processing.  Naturally, if you leave it at 0, it is a stronger signal.  Not because it has a 6db boost, but because it has not been cut...

 

I always thought it should be a fatter signal at 100 but it really isn't.  I have some dual amp patches and they actually sound better at 0. Definitely not farty.  And in the end, what are you feeding?  Do you feed a true stereo power amp into separate cabs with stage separation?  Are you feeding the FOH/PA?  If so then folks on either side are missing part of your sound!  And if you are feeding a guitar amp/cab or a DT what difference does stereo separation make in the end?  The power section outputs in mono.  Stereo wiring in a single cab doesn't really give the full stereo effect either.

 

As I said before, unless you have a specific need for two absolutely separate paths, try the mixer at 0.  Your stereo FX blocks after the mixer will still perform their stereo functions.  For recording you can use true panning within your DAW. 

 

http://line6.com/support/topic/2033-pod-hd-500-500x-new-routing-schematics/

For nearly 15 years I owned an A circuit sound company and also FOH engineer, running large Midas consoles. You are correct in what you have found, I always tried to stay away from hard panning a dual input signal just because of what you have found, panning anything hard left and right always took some of the body from  the signal. I never panned anything more than 3 and 9 oclock. This helps recover most of the body of the signal and still allows a little extra separation in the mains. It also helps retain the left and right panned signal to the left and right mains allowing all the audiance hear both signals, without listeneing to only part of the signal stage left or right.  Hope I explained this weel enough so you understand. In other words, in a live situation, no matter FOH or guitar, there is no way to achieve a totally pure stereo sound all over the venue, so why try! Experiment: Take a nice stereo recorded song and listen with stereo head phones, now take the phones off and listen with you normal speakers and walk from one side of the room to the other, unless your in a very small room, every time you move hard left or right you loose part of the stereo effect along with part of the music. In a recording environment this really doesn't apply, until the music is played back.

  • Upvote 1

Share this post


Link to post
Share on other sites

I thought we were talking strictly about the affect of centering the pans of two identical dual mono signals, rather than altering the width of a stereo signal.  Regardless of total channels, a mono signal has no stereo width.  I don't see how the energy distribution of identical mono signals in the center of a stereo field could possibly be affected unless there were some other factor canceling or boosting the center portion of signal.  For such a result, M/S processing would be required. 

Share this post


Link to post
Share on other sites

The default values for the mixer PAN A and B are set 100% L and 100% R to ensure you do not double the signal level output. These values ensure the signal level returns to what it was before the Path A and B split by keeping just one L and R signal rather than two L and R signals. From perapera routing schematic, there is a single L+R path for PRE. This is split into two L+R paths, one for Path A and one for Path B. The mixer then allows you to blend the Path A and B L+R paths as you wish. Default keeps just L of Path A and R of Path B providing a single L+R signal for the POST path section. Changing to 0 keeps both L+R of both Path A and B which is double the signal level.

 

For simple one amp scenarios the default setting is correct to maintain the signal strength. Of course if you start using two amps and using stereo effects in the path A and B then you will need to preserve both L+R from paths containing stereo signals. You will also have to work out what is happening to your signal strength and adjust subsequent gains accordingly. The mixer has a very wide range for negative values presumably to allow users to attenuate signals down to adjust for combining two paths before the POST path effects.

 

Once again we can use whatever values we like to get the tone we desire. If increasing the volume at the mixer stage by changing the PAN values to 0 is what works for you, then great, but I don't think it should be applied to every patch without first considering why you might need to do it, and what it is actually doing regarding mixing the two L+R paths from A and B, and in many cases it will just be doubling up a mono signal.

  • Upvote 1

Share this post


Link to post
Share on other sites

And that's why I said this is blasphemy!  I am positing something that everyone seems to take for granted yet is not actually stated anywhere.  Nowhere in the Advanced Guide or anywhere esle I looked does it say anything about default positions for the mixer, nor are there any warnings or cautions about possibly doubling the signal when set to zero.  On the contrary, the signal path is designed to provide a stereo (dual mono) signal on path A and path B to feed the mixer which mixes it down to a single stereo (dual mono) feed to the Post section.  From what I see, the mixer doesn't double any signals, it provides the correct mixed output to the Post section.  Panning hard left and right if it was real panning would give the same signal level but with wider separation.  Instead, as the manual states, it adjusts the stereo balance of Path A and B output to the Post section.  You are really cutting half of Path A and half of Path B.

 

I am not telling everyone to suddenly change everything you do, but if you are willing to experiment and trust your ears, give it a try.  Objectively.  With your live rig.  Perhaps when recording to USB it is not as evident but to me there is a real difference and it is not just volume.  I have tried it with volume compensation in the mixer too and it is still different.  I can say for a fact with my tests and my gear, feeding a DT25 head to a 2X12 cab, there certainly was no clipping induced.

 

I just think its good to know there are other options.  Just as some prefer a single input and the lower signal hitting the amp, there are other things we can do at the other end too that might lead us to use less EQ and other enhancements...

Share this post


Link to post
Share on other sites

Wait now.  First we need to delineate between dual mono and stereo in this context to avoid confusion.  When I say dual mono, I mean the signal on the L & R channel is identical.  Stereo means they are not identical.  Summed identical channel signals are simply 6dB louder.  The total signal level rises exponentially to +6dB as pans A & B are set closer together.  If  they are panned differently, there is an averaging that would be the same as using one pan, only louder depending on how closely they are set. 

 

If the signals on any of the 4 channels on paths A & B are not identical,  then you would need to set the pan for the path/s with the different signals to something other then extreme L or R to get some signal from all the channels.  At that point, it would be just a matter of using your ears to use both pans to focus the stereo energy where you want it.

 

There's no way Line 6 could have designed it to cover all scenarios without some being louder than others, or without possibly loosing unique channel content if panned a certain way.  That's why there are levels for each path. 

Share this post


Link to post
Share on other sites

I agree its good to bring this up for discussion and it is good to experiment and end up with a tone that you like. I was just trying to explain why I think Line 6 set the defaults as they did, which is to keep signal level constant and not inadvertently double it and also to maintain stereo separation for any stereo fx that may get deployed. The defaults are a good place to start and once we understand what each component does we should experiment and adjust things to get the tone we are happy with. :)

 

From peraperas schematics and my own experiments I think that the MIXER is there to mix together the dual stereo Paths A and B.

The default values for the PAN settings provide half the available signal paths to the POST path, which effectively balances the signals to what was in the PRE path (ie: one L and one R). The split into paths A and B is where the signals get doubled, because you get two L+R stereo paths. Adjusting the default PAN settings so you maintain both L+R signals from both A and B effectively means you have doubled the signal strength (ie: two stereo paths summed instead of one). The end result may well be pleasing when compared to the default and depending on our own personal tastes.

 

This is certainly something everyone should try with their Mono patches to see if they prefer the resultant tone. :)

Share this post


Link to post
Share on other sites

Every signal path is a stereo path according to the manual.  Whether or not it actually carries a stereo signal or not is the only question.  We are not feeding 4 separate mono signals to the mixer we are feeding 2 stereo signals each of which may be carrying duplicate audio rather than true left and right audio.  The mixer is intended to receive these two signals and then mix them down to a single stereo output which again may be carrying the same audio on each channel depending on the FX setup.  The mixer is not a true pan, it adjusts the balance of each path.  Leaving them centered simply allows the original stereo paths applied to the mixer to pass as received after mixdown.  Panning hard left and right cuts the signal, leaving it centered does not double the signal.

Share this post


Link to post
Share on other sites

Are you saying that the signal level is the same when you mute one path with both pans in the middle?  The signal should double when both paths are engaged and they have identical signals.  If not, the mixer attenuates when it sums, like a mono FX block does with the inputs.  I don't think the mixer block does that.

Share this post


Link to post
Share on other sites

I just thought of a better way of explaining the reason for the default mixer pan settings. It is effectively dealing with the same situation that a mono fx block has in the PRE path. Summing two inputs and producing one output. The mono fx block is hard coded to sum and then attenuate so you get half the output signal. The mixer default settings essentially do the same. It has two paths coming in and only one going out so the default mixer settings mute half of the input signals. For a standard mono patch that is the correct logic to apply to maintain the same signal level throughout. Obviously we can adjust the balance of the two stereo paths in the mixer according to taste and what amps and fx we have deployed in the patch.

Share this post


Link to post
Share on other sites

I haven't talked about it here in quite a while, but I remember discussing this at length a few years ago. The mixer pan control isn't really a pan control at all. It's a balance control. As stated above, it's not really panning a mono signal. It's controlling the balance between the left and right sides of the tones coming from each signal path. The weird thing about the HD's routing is that with a dual tone, you actually can have two separate stereo paths coming from the mixer block, but because you only have one set of stereo outputs (the XLR and 1/4" outs receive the same stereo outs), you are potentially summing these down. If you have a dual tone with stereo effect in each tone, and you hard pan the outputs, you're actually losing stereo information. So I guess the long and short is that you are limited to what you can accomplish with a dual tone as far as stereo effects are concerned.

Share this post


Link to post
Share on other sites

I just thought of a better way of explaining the reason for the default mixer pan settings. It is effectively dealing with the same situation that a mono fx block has in the PRE path. Summing two inputs and producing one output. The mono fx block is hard coded to sum and then attenuate so you get half the output signal. The mixer default settings essentially do the same. It has two paths coming in and only one going out so the default mixer settings mute half of the input signals. For a standard mono patch that is the correct logic to apply to maintain the same signal level throughout. Obviously we can adjust the balance of the two stereo paths in the mixer according to taste and what amps and fx we have deployed in the patch.

 

 

But I'm saying the mixer appears to sum but not attenuate both channels according to perapera's block diagram (schematics include electronic component values).

 

You're not really limited regarding stereo FX in each path, because you can set the "pans" for each path any way you want.  Some settings might produce louder results than others depending on what is in each path.  Just lower the levels to compensate when needed.

Share this post


Link to post
Share on other sites

OK so after reading everything all over again, the manuals, perapera's thread, meambobbo's tone guide, I accept the consensus that the mixer sums the incoming signals just as the first mono block in the Pre section does.  Unlike the mono block, the mixer does not automatically reduce the output level for you.  So that leads to more questions.

 

1. Is there an actual need to reduce signal strength?  If you are not inducing any undesired effects on the blocks in the Post section or causing any clipping why lower the gain?  Doesn't that just lower your signal to noise ratio?

 

2. If your goal is gain reduction, why would you choose to use the Pan feature when the mixer has dedicated gain faders?  Are you potentially losing some audio details that way?

Share this post


Link to post
Share on other sites

Yes, you want to maximize the output signal without clipping.  I've already explained how you might loose L or R signal content with stereo FX on either of the dual paths.  You were correct.  If doubling a mono signal by centering both pans does not clip anything after the Mixer, then that's fine.  I depends on how high the FX outputs before it are anyway.  Just lower the Mixer levels if needed.  You got it figured out now, chief. ;) You're good to go.

 

Use external metering to make sure you never exceed 0dBfs thoughout the signal path.  Check to make sure your peaks are between ~-12 to -6dBfs.  If a few peaks go a little over -6dBfs when you strum hard up high on the neck, you will be using the full bit resolution and maximizing S/N.  I wouldn't let peaks go any higher than that though.  Any modulation FX run after amps can create wild peaks.  Either use a DAW meter through USB, or use a phone SPL app with some monitors.  Assuming it isn't clipping the input, use the direct/dry/unprocessed guitar signal as a reference.  If the direct guitar signal peaks don't ever go much over -12dBfs, that's normal.  You can set the processed signal/s up a little higher.  I'm fading, so I better be off for now.  Good luck.

Share this post


Link to post
Share on other sites

But I'm saying the mixer appears to sum but not attenuate both channels according to perapera's block diagram (schematics include electronic component values).

 

You're not really limited regarding stereo FX in each path, because you can set the "pans" for each path any way you want.  Some settings might produce louder results than others depending on what is in each path.  Just lower the levels to compensate when needed.

Correct. It does not attenuate on the default settings but it only provides half the available signals, ie: just L from path A and just R from path B. cantering the pan controls will give L+R from both path A and B, which is twice the amount. The mixer can be used to attenuate the signal strength by using the volume controls for path A and B, so for a dual amp path a hot mixed signal can be adjusted down if necessary.

Share this post


Link to post
Share on other sites

OK so after reading everything all over again, the manuals, perapera's thread, meambobbo's tone guide, I accept the consensus that the mixer sums the incoming signals just as the first mono block in the Pre section does.  Unlike the mono block, the mixer does not automatically reduce the output level for you.  So that leads to more questions.

 

1. Is there an actual need to reduce signal strength?  If you are not inducing any undesired effects on the blocks in the Post section or causing any clipping why lower the gain?  Doesn't that just lower your signal to noise ratio?

 

2. If your goal is gain reduction, why would you choose to use the Pan feature when the mixer has dedicated gain faders?  Are you potentially losing some audio details that way?

1. In a mono amp patch where only one path is needed, the provision of two available paths doubles the available signal which is then summed at the mixer block. To avoid unintentional signal boosting in that scenario you have to counter the doubling effect of splitting the single path into two separate paths, and the simplest way of doing that for a mono patch but still maintain the ability to have a stereo output to two amps or a mixer, is by muting the duplicated L+R signals introduced by the path split. This emulates what would happen in a real world single amp set up where you would not have a splitter giving two paths from the amp and your post fx.

However, there is no reason why you cannot choose to centre the pan controls and utilise the boosted signal output if you prefer the tone produced.

 

2. If you just want to reduce gain to reduce clipping or to even out volumes across patches then sure you can use the dedicated mixer gain faders. In fact for a simple mono patch, I guess cantering the pans and setting gain faders to -6 db would produce the same effect as the default mixer setting in terms of overall volume. However, why bother when for a mono patch you won't be losing any audio details as each L+R channel should be identical, BUT if you use stereo fx in the signal path THEN the default should still be ok if the fx is in the PRE path but if it is placed in path A or B then you WILL have to centre the Pan for that path to get both parts of the stereo signal if it is not followed by a mono fx block.

 

I am sure this will be easy enough to test out if you want to be sure.

Share this post


Link to post
Share on other sites

1. If doubling a mono signal by centering both pans does not clip anything after the Mixer, then that's fine. 

 

2. Either use a DAW meter through USB, or use a phone SPL app with some monitors.  

1. This is also one reason I use one signal path and put everything in it. That way the mixer comes last and I can use it to boost the signal without the danger of clipping something coming after it.

 

2. When you use a DAW meter through USB I presume the driver should not have th +18db checked and the "record send level" at full?

Share this post


Link to post
Share on other sites

 

So now what?  I am going to go back through my patches and see what I might have been missing just by needlessly panning in the mixer.  Unless there is a specific reason to completely separate the two Paths I may be better off with  0 setting in all cases.  Especially for single amp patches...

 

Any other ideas?  Are there other needless things we do out of habit that really do nothing to enhance our tone?

You haven't been missing anything! The two signal paths are the same as  two stereo tracks in a real world mixer or a DAW with unity gain at Zero. If both are centered and each is outputting zero decibels, there will be clipping. If you had a vocal track and guitar track, panning is used to achieve an overall sound by listening and it's no different when mixing two guitar tracks. Once you're done mixing two guitar tracks to get a sound you're after, you don't go back and say let me go and see how each one sound separately to see if it would be sound better in the center or if you're losing something.  Another example would be humbucker pickups, the best humbucker pickups don't necessarily have good sounding single coil when split into single coil, as a matter of fact they rarely ever do.

 

To simplify matters, in many of my patches, I mute one of the mixers (path B) and use Path A dead centered (I think this would have been a better default setting) but whenever I use dual path I make sure the output are connected to an audio interface or external mixer to monitor the signal to avoid clipping. 

Share this post


Link to post
Share on other sites

You haven't been missing anything! The two signal paths are the same as two stereo tracks in a real world mixer or a DAW with unity gain at Zero. If both are centered and each is outputting zero decibels, there will be clipping. ...

I keep my Pod HD 500x mixer block set panned center and set at 0. Does this mean all my presets are clipping? I didn't think they were based on what I hear but I'm no expert.

 

I use the XLR outputs into a Tascam USB mixer for direct recording and I have to turn up the Input Trim on the two channels the Pod's XLR's are plugged into quite a bit just to get the meters on the Tascam mixer to rise up to 50%. (Not sure if the meters that far 'downstream' are telling me anything about clipping in the Pod's signal chain...)

Share this post


Link to post
Share on other sites

Those meters will not tell you about internal clipping but your ears should.  If you are not hearing clipping or unwanted distortion, it probably isn't there...  Clipping can be introduced at any stage including right from the guitar input and once there will be carried through to each successive stage.  Setting the mixer to center is hotter than panned left and right but it also depends on how hot the signal is coming in to the mixer.  Try to keep it as hot as possible, factoring in balance levels between clean and dirty patches, without any clipping or unwanted distortion.  Keep your POD master up full feeding the Tascam mixer. 

Share this post


Link to post
Share on other sites

Sounds correct.  It should only be 6dB hotter if both paths have identical signals and are panned center.  Some dual amp configurations will be louder than others when center panned. 

 

BTW, the Tascam mixer might just be 16bit through USB.  Pod HD's are 24bit, so you'd be loosing some resolution recording with the Tascam. 

 

I can't say whether engaging +18 in the Pod HD USB driver record level actually matches the internal Pod HD signal level or not.  My guess is:  with the USB driver record level up all the way and the +18 off,  if the dry/direct fully side-panned guitar signal going through the Pod HD produces peaks in your DAW up to anywhere from roughly -18 to -6dB when strumming really hard, then that is probably the same level of internal headroom inside the Pod HD.  I'd use that dry guitar level as a reference for peaks for your FX presets.  If those dry signal guitar peaks are down near -18dB (pro digital standard recording level), then you should be able to push the Pod HD internal signal up another ~12dB with ~6dB headroom before internal clipping, and you certainly wouldn't need the USB driver +18dB boost.  That's my guess, and it's the best answer I can give because Line 6 is not specific about the internal levels.

Share this post


Link to post
Share on other sites

I keep my Pod HD 500x mixer block set panned center and set at 0. Does this mean all my presets are clipping? I didn't think they were based on what I hear but I'm no expert.

 

I use the XLR outputs into a Tascam USB mixer for direct recording and I have to turn up the Input Trim on the two channels the Pod's XLR's are plugged into quite a bit just to get the meters on the Tascam mixer to rise up to 50%. (Not sure if the meters that far 'downstream' are telling me anything about clipping in the Pod's signal chain...)

In my HD500 (not the X), the xlr outputs are mic level meanwhile  the 1/4" unbalanced are line level (there's a switch to select line level or amp) so the Tascam meters are right. I doubt that they changed this in 500X. You're better off using the SPDIF if the TASCAM has it, if not,  the unbalanced outs at line level and the best option would be using the POD's USB recording for guitar tracks. The only reason I would dismiss the XLR for recording because at MIC level, the sound quality will depend  a lot on the quality of the MIC pre in the tascam that will be doing all the heavy lifting. I'm not saying it's bad, but if the digital signal remains digital using the SPDIF or USB recording, you will undoubtedly get the cleanest recording. Regarding recording, my advice is  (in your DAW or TASCAM ) not to use the Zero DB as a reference for clipping -12db to -6db range is more than adequate and will insure no clipping. Even going down as low as -16db digital would be adequate,

Share this post


Link to post
Share on other sites

 

... If those dry signal guitar peaks are down near -18dB, then you should be able to push the Pod HD internal signal up another ~12dB with ~6dB headroom before internal clipping, and you definitely wouldn't need the USB driver +18dB boost.  That's my guess, and it's the best answer I can give because Line 6 is not specific about the internal levels.

People complained so much about lower level when recording so Line 6 finally gave up and  gave them the +18db boost. Most didn't realize that line 6 had it right in the beginning and there's no real need for the added decibels;  but most like to live dangerously  and watch for the zero digital mark. since most meters aren't that accurate when it gets to zero DB many will clip the recordings if running this close.  If -18db digital is 0db analog, even recording to -18db digital is adequate and in the mixing stage normalizing can bring the level up as desired without issues.

Share this post


Link to post
Share on other sites

Absolutely, Deano.  -18dB is the pro digital standard recording level.  it still leaves 21 bits out of 24 for a theoretical noise floor of 126dB -- more than ample resolution for accurate low level low frequency signals.  I've run tests with sine waves recorded in 16bits and 24bits.  Generated frequencies right below -70dB in the ~300hz - 1kHz range all become quite deformed, creating harmonics and weird modulations -- while the same affect doesn't occur in 24bits until well below -100dB.  Most pop music is  around -12dB RMS, so the low frequency wave deformations are then ~60dB below average.  That's pretty low compared to the average level, but I think we can hear the affect with dynamic acoustic music as slightly muddled upper bass/mid harmonics during quite passages.  We've been listening to CDs like this for decades.

Share this post


Link to post
Share on other sites

Well, once the input signal hits the amp/s it won't really affect the output.  As to whether it may clip FX before the amps is a matter of debate.  My guess is that some of the older Pod FX are 20 to 16bit resolution and may clip more easily.  It's kind of lame for them not to have documented this stuff so we'd know for sure.

Share this post


Link to post
Share on other sites

Amen to that!  I think they presently underestimate the value of customer interaction and feedback.  Perhaps they should derive philosophical analogies to customer relations from the functioning of the equipment they model?

Share this post


Link to post
Share on other sites

Use external metering to make sure you never exceed 0dBfs thoughout the signal path.  Check to make sure your peaks are between ~-12 to -6dBfs.  If a few peaks go a little over -6dBfs when you strum hard up high on the neck, you will be using the full bit resolution and maximizing S/N.  I wouldn't let peaks go any higher than that though.  Any modulation FX run after amps can create wild peaks.  Either use a DAW meter through USB, or use a phone SPL app with some monitors.  Assuming it isn't clipping the input, use the direct/dry/unprocessed guitar signal as a reference.  If the direct guitar signal peaks don't ever go much over -12dBfs, that's normal.  You can set the processed signal/s up a little higher.  I'm fading, so I better be off for now.  Good luck.

 

I believe the guidance above is very important to keep in mind and I would in fact recommend a more conservative position. Based on my own measuring and sound checks I believe no peak signal level between HD models should ever exceed -12dBFS unless you are intentionally looking for some soft compression for those peaks that exceed the -12dBFS level. I can't be sure that this is true for all models but all the ones I measure seem to behave like that (compressors seem to be a bit more forgiving and additional compression kicks in for peaks above -8dBFS). The only exception I see so far seems to be the mixer and possibly the volume pedal (I did not measure the latter, it just sounds like it), i.e., if you put the mixer at the very end of the model chain you can push the peaks all the way close to 0dBFS w/o getting compression or distortion (at least not when measured via USB).

In regard to the quoted post's advised peaks I think you're really ok as long as you can live with a little soft compression/clipping and I suspect the effect of any extra 3 dB peak won't be noticable by ear and only show up in high resolution measurements. Violating the guidance won't immediately result in digital clipping noise you can immediately hear unless you are really pushing into the -3dBFS range. But why live on the edge and get unintended soft compression/clipping that does not come from the modelled gear but internal HD signal level limits?

Share this post


Link to post
Share on other sites

Don't be surprised no-one says anything.... these guys seem to think the launch codes are in their "sauce" or something..... just know that it makes guitar noises, cause that's all you're getting out L6.

Share this post


Link to post
Share on other sites

Don't be surprised no-one says anything.... these guys seem to think the launch codes are in their "sauce" or something..... just know that it makes guitar noises, cause that's all you're getting out L6.

I know having them divulge all the details would be helpful to those of you who can make good use of the data/science that they could provide....but...I also have empathy for the company that knows too well that releasing the recipe will help only a relative few but spawn a whole feces storm of deeper confusion and mostly undeserved bad press for them from the multitudes who won't understand what they read but still use it to "prove L6 sucks"....

 

Not releasing only stirs up a small storm from the few....

Share this post


Link to post
Share on other sites

MartinDorr, incorporating a soft-clipping feature would hardly be worth keeping a secret.  Nothing new there.  I doubt there is one, though.  However, it's not uncommon to calibrate the D/A converter for 6dB analog headroom before 0dB, which could mean analog gain compression starts at -6dB.  I used to accidentally drive my old EMU 0404 soundcard to soft-clipping.  I've not seen it ever calibrated to -12dB though. 

 

The meters in Reaper would register any clipping down to 1 sample.  You also said that some models don't seem to have any gain compression even right up to 0dBFs, but then warned against driving the internal signal that high to avoid any gain compression "...that does not come from the modeled gear but internal HD signal level limits".  That could not happen if some models don't compress up to 0dBFS. 

 

If the Pod HD USB driver is calibrated so when set up all the way without the 18dB boost engaged to be the same as the Pod HD D/A, and your software meters are set the same, then you will have a reference to judge clipping by.  I suspect the USB driver is calibrated as such, so you might test if internally driving the Pod HD USB output past 0dBfs in your software causes clipping through your analog outputs.  Again, those outputs might have 6dB or more headroom for good measure, but it would be worth getting straight.

Share this post


Link to post
Share on other sites

Good advice on a couple things to double check, but I can't get to that for a while.

 

Here is my experience to get you a bit more background (not intended to prove anything):

 

In a nut shell I measured the output level behavior to study Drive and Channel Volume interaction for all preamp models, the Mid Focus amd Studio EQ, and a couple compressors and over drive blocks.

What I think I found is that for all those blocks the input to output attenuation for any given setting is very consistent for any input and output peak levels as long as those (both) stay below -12dBFS. For example, if a given preamp setting causes a 10dB attenuation, the preamp model will do this independent from whether the input signal is -28dB or -22dB. BUT, it will do less and less if say the input level is -20, -18, or -16dB, and so on because the resulting output exceeds -12dBFS.

If any clean output increasing setting is engaged so that the output peaks exceeds the -12dBFS limit the resulting peak output signal attenuation slowly starts to get reduced (relative to the input peaks) and RMS level increases (I'd guess due do extra harmonics of compression and later distortion). For the first couple dBs the effect is very slight in numbers but usually around -6dBFS there is a clear effect on the sound that most people will be able to hear (trained or not).

I am not arguing that I have the full picture or that I am sure my meaurements are truly accurate. I used a looping sound sample from a Variax and measured via USB, i.e., no DA conversion in the HD involved. All dials as far as I can tell are set to 0dB attenuation in drivers and DAW. I used an Izotope Ozone 4 plugin and it reports peak and RMS dB in 0.1 increments down to -30dBFS and counts every single event of clipping, but these effects occur way before any clipping is reported.

As I said in my post, the only model that seemed to break from that behavior was the mixer. I did not detect any compression or clipping untill its attenuation setting actually cause the output signal to clip.

 

Appreciate your constructive feedback. It never hurts to review these kind of observations. There are just too many things that can go wrong or stay unnoticed. Hope I can reproduce and document a repeatable test case for anyone interested to try.

 

Martin

Share this post


Link to post
Share on other sites

I think it should be made clear that it's the models you are using that cause the compression, and not something applied to the internal signal path regardless of the model.  Otherwise, the same thing would occur when just using the mixer block.  Yes, I imagine any model with a gain or drive knob would compress the input gain.  That should come as no surprise.

 

In experimenting with on-line A/V meetings using a Pod HD as the A/I in a Mac system, I did notice that the A/V meeting app caused the Pod USB input to auto adjust as I spoke.  That's because the A/V meeting app has built in auto-leveling.  If you want to avoid that situation, make sure no other active app is sharing the Pod USB driver when using the Pod through USB.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...