How to get the original response from the models - HD500/X signals routings explained

Post Reply
User avatar
hurghanico
Posts: 123
Joined: Fri Apr 06, 2018 11:23 pm

How to get the original response from the models - HD500/X signals routings explained

Post by hurghanico »

Here I explain a key detail about the HD500/X way of working,
without this knowledge you can still get some decent results with lot of tweakings, but probably never great as they could be and as easily.

The key detail of which I'm speaking about concerns its signal routings, and here it is:

Image

above here you can see the POD chain general scheme (with some effects and amp models blocks added just as an example):

going from left to right there are:
2 virtual inputs 1 and 2 assignable to any physical inputs of the device, a dual mono/stereo pre split path, the split point where the 2 signals coming from the pre split path are always duplicated, 2 parallel dual mono/stereo paths A and B, the mixer block where the 2 parallel paths are mixed down into a single dual mono/stereo signal, and the final dual mono/stereo post mixer path..

the 2 parallel paths A and B, depending on whether or not blocks are inserted in them, work respectively as full dual mono/stereo parallel chains or as simple extensions of the left and right sides of the dual mono/stereo pre-split path. in the first case, to keep the signals of the 2 paths complete (i.e. not truncated by one side) for volume and stereo separation (if any) in the post mixer path, the Pan controls of the 2 paths must be centered. if, on the other hand, the 2 paths behave simply as extensions of the left and right sides of the dual mono/stereo pre-split path, the Pan control of path A should be set to the left and the Pan control of path B should be set to the right, otherwise the pre-split signal is boosted and if it ended as stereo (ie the last FX into it is stereo) the 2 sides L and R are mixed together and the resulting signal becomes dual mono (but this could also be what you need, eg if you connect only 1 XLR output).

before going any further, please note that each mono block (amp/FX) gives 2 halves of the processed signal to the next thing/s in the chain, if the next things are the 2 parallel paths, 1 half goes toward one path and the 2nd half goes toward the other path, but first each of them must pass through the split point where as I said before the signals are always duplicated..

by default input1 is set to accept any signal coming from the Guitar In, Aux, and VDI, and input2 is set to replicate exactly what input1 does..
the only good reason of this general setting is that (in case you don't use the pre-split path) it allows to use directly both parallel paths A and B with the same instrument without having to mess with the input settings (which directly on the device unfortunately are semi-hidden between the system settings instead of being between the other patch parameters as they should be, since they are per patch parameters)..

the drawback of this default setting of the 2 virtual inputs is (see the picture) that if you put a block (amp/FX) in pre split path it will be reached by 2 same signals, and especially with humbuckers or active hot pickups the block reached by 2 same (hot) signals (instead of just one) could be overloaded, furthermore, the resulting combi filtering between the 2 parallel signals processed would worsen the tone, all this with conseguent typical issues which might be: muddy tones, excessive and barely controllable distortion, excessive compression, altered frequency response with some important mids involuntarily scooped, overwhelming bass and harshness, incapacity to clean up nicely the tone by playing more gently or by lowering the volume from the guitar, lack of definition and dynamics, exaggerate peaks when using wah and filter FXs, etc., etc..

if you followed all the steps of this post up to this point it should be clear that there are 2 points where the passing signal is duplicated: the input stage of the virtual chain (if the input settings are the defaults), and the split point..

we'll see in a moment that the duplication of the signal at the input stage is easily avoidable by simply changing accordingly the related input settings..

in regard instead to the duplication of the passing signals at the split point, before suggesting solutions it would be better first to understand the good reason why it happens, which reason simply is that when there are one or more blocks in pre-split path only one half of the processed signal coming from the last block would go towards each parallel path, here is therefore the need to duplicate at the split point each of the 2 halves in order to give full level processed signals to the blocks present in the parallel paths A and B..

and ok, the one above is a perfect functioning, but the real problem arises when instead there is nothing in pre-split path (ie when you want to keep the 2 parallel paths independent and separated) which would divide the input signals into two halves, and therefore what would be duplicated in this case at the split point are not halves but instead full input signals.. thus resulting in a duplicated input signal at the beginning of each parallel path, with the same potential consequences described previously due to the possible overloading of the models..

so, case by case, what can be done to avoid the uwanted overloading of the models?

1) when there are mono blocks (FX/amp) placed in the pre-split path:
set input2 to a dummy/unused physical input, doing so only the signal coming from input1 will reach the first mono block..

2) when you want to use the 2 parallalel paths A and B as separate chains but with the same instrument input signal, thus not sharing blocks (FX/amp) in pre split path:
also in this case make sure that input2 is set to a dummy/unused physical input, but make sure also that at least 1 mono block (even with neutral settings, like a noise gate with threshold at zero) is placed/shared at the beginning of the pre-split path,
the purpose in this case of the first mono block is to generate 2 halves of the single input signal, which halves then at the split point will be duplicated, so the blocks in path A or B will be reached by full signal (instead of doubled ones)..

3) finally, if you need to send a different instrument input signal separately in each parallel path A and B (eg guitar/voice):
the "dummy 2nd input trick" can't be used, the only thing you can do in this case is to lower by about 6 db the 2 different input signals at the beginning of the parallel paths (eg by using the Studio EQ fx). The good news are that since the Studio EQ fx is a stereo-in-stereo-out block it retains the stereo separation of the different input signals and doesn't divide them in halves, therefore there is no need to place a dedicated Studio EQ in each parallel path, but you can share just one Studio EQ in pre-split path..
--------------------------------
many have long argued that Line6 has never officially stated how the duplication of the input signal can occur as a result of the input settings.
but it's not true, Line6 did it long time ago, as you can see by reading in a certain point of the advanced guide and reported below here:

Image

but what Line6 has forgotten instead of specifying is that the duplication of the unprocessed input signals occurs also at the split point when nothing is shared in pre-split position, and this is one of the several reasons why I wrote this detailed article..


and the following is an excerpt from the old and archived official Line6 forum in which "Rowbi" a Line6 expert (who as such is in direct contact with the Line6 products developers) responds to a forum member "daenesque":

Image
Post Reply