Originally, this was sent as an email to the developers at Positive Grid – Wed, Jun 6, 2018, 4:43 PM
So yes, you are welcome, modelling guitarists everywhere 🙂 No applause, just please subscribe to my youtube channel! https://www.youtube.com/subscription_center?add_user=pipelineaudio
*My Hopes for Bias FX 2
Hopefully, you can read this at your leisure, and I know it gets wordy, but I’m trying to keep it as focused as possible as an overall road map to what I’d love to see with the goal of live use in mind. I’ll try to put in a bit of detail between not just a software product, but hardware as well.
It seems to me that one of the absolutely revolutionary hallmark innovations of Positive Grid is that you have an overarching system where a customer can have the exact same sound in many different formats. Desktop VST, desktop standalone, iOS and hardware products can all access the same presets and get the exact same sound regardless of the device and situation they are using it in. Want your studio sound on stage? No problem! (at least for any one individual sound, which is kind of the subject of this whole verbiage). This is absolutely amazing, and really inspires confidence in what is normally a pretty high stress situation.
It is my ultimate goal here to detail a plan that could lead to not just amazing Bias FX 2 software, but be the brains for a potential Bias FX Rack (powered or not), Bias FX pedalboard (a true Helix killer), and a world breaking, paradigm shifting, Bias FX Combo amp.
For the most part, Bias Amp, Bias FX, and Bias Pedal are already perfect for the studio. Some minor quibbles about speeding up the menus, loading, and especially preset management would be nice to deal with, but during a mix or tracking, the Bias family already works extremely well.
My concerns and suggestions here will really be about these products for live use, though likely this will have many secondary benefits for studio use as well.
I’m getting a bit ahead of myself here mentioning hardware, but as you guys have always so seamlessly integrated the hardware and software side, I think it will help to make the most sense of an umbrella paradigm to detail it here in a minor way.
Whether or not you make a hardware version (please do!), these considerations will say a lot about how even the software should work, and give clues as to how customers could implement a live setup themselves with off the shelf hardware.
On the hardware side:
A dual signal path in case users are playing thru guitar amps: One signal, with no cabinet emulation, for feeding to the guitar amp, and another signal, with cabinet emulation on to feed to the PA, or for the modern guitarist, playing thru an FRFR speaker, just the second signal, but with the option always there in case you wanted to monitor through a guitar amp.
Simplest most reliable setup possible, as compact as possible. I already knew that this could be done needing only to plug the power and speaker signal lines in
A capable foot controller, hopefully wireless and rechargeable battery powered as well to really keep the simplicity, and the number of cables and potential problems down, as well as really pushing versatility up. 8 patch/stomp pedals, a tuner/tuner mute pedal, a tap tempo pedal, 2 switches for bank up/down and an expression pedal (or two)
A computer interface, possibly out to a DI to deal with any potential phantom power issues (though that is only a problem with some interfaces).
Properly written drivers so that this thing really could be used JUST as an interface, which means being able to break or mix the amount of the standard path between the input and the output so that if desired, only the processed signal from the DAW would be heard at the outputs. Line 6 has famously blown this on some of their more expensive products, while their cheaper POD series often does it perfectly well
A computer, and the trickiest part, mounting it. If at all possible to be able to avoid plugging anything in, ideally with the computer left in and usable in the same way it was transported. The tablet holder works, but it means setting that up and plugging in power and a USB cable.
Some way to carry all of this, safely and conveniently
Wireless for the guitar, but a way to plug in a cable if need be (could be done with a switching 1/4” jack with the normal going to the wireless and the jack breaking when a regular guitar cable is inserted). Not absolutely critical, but would be nice. Possibly just allowing a space for an existing off the shelf wireless around the G30/G50 size
ideally able to have everything plugged in and in place: power, signal, whatever, barring the very few things needed to interface with the rest of the world
9 volt power terminal for connecting auxiliary devices
On the software side:
Switching as quickly as possible without nasty artifacts and without giant gaps, and boy could those artifacts be nasty!
Software capable of enough routing to handle the dual path and other necessities
Tuner and tuner mute at a glance.
Tap tempo, hopefully with some visual feedback
Auto engage for relevant FX. Once you try this, its hard to go back, and people were really vocal about it when I suggested alternatives.
Ideally, in essence you could create all your favorite mixing chains from your studio recording, and be able to switch between them, that was the be all end all.
Not critical, but highly desirable:
The ability to run any sensible third party VST
Spillover FX, people really loved the automated swells between sounds, and delays ringing on after the patch was switched to another
Ability to use impulse responses anywhere in the chain and on either path.
Financial / Existing Users Considerations:
I would hate to see people who already own existing Bias heads and Bias racks feel like they’ve been obsoleted.
I think that having FX loop functions in a potential Bias FX rack, that allowed the popular 4 cable method to use Head/ Rack for both the amp sound, and then the power section would really keep the value of the older heads. Not to mention that Head/Rack would always give instant access to amp controls (though that could probably be done in a Bias FX Rack as well)
The Achilles heel of almost every product out there. Ideally, switching between patches should be as quick as possible without an audible gap, or any sort of audible glitch between them. Some examples of glitching could be a giant, short spike going well above the volume of either patch, stuck or forced buffer sounds, or massive long term volume buildup from a crossfade.
I see three main, roughly defined paradigms people use for switching right now:
- Preset switching
Being able to build a patch up from scratch if desired, setting any and every parameter as desired, with whatever routing, FX and parameter values desired, storing it and being able to recall it at will.
Advantages: Total flexibility. The ability to consider this patch in isolation without any regard for how these settings could affect other stored patches.
Disadvantages: Presets take a long, often prohibitively long, time to load. The resource cost of loading a preset can have a huge effect on stability.
It is possible (as we have done on our SWS Live switching system), to have each preset stored on a separate track, always running in parallel and simply switch which outputs we hear. However, having other patches ready to go in the background can have a significant CPU cost.
Gig Performer attempts to crossfade between presets, with some pretty extreme reliability issues, and a disturbing volume spike as it crossfades from one patch to another.
[note: the manufacturer has taken issue with this statement. The fact that the manufacturer cares that someone said it should tell you just what a good company this is! I have to stand by that this is what happened to me. I did work back and forth sharing videos or audio clips or something with one of them and with this particular paradigm it seems like the volume spike is unavoidable when switching between similar high gain sounds, but there are many people who will not have a problem with this so take it for what its worth. Also, it should be noted that there were latent plugins in the the chain when these crashes happened, including Amplitube and others that are known to be sketchy in realtime input, however, we were able to get these stable in lbx stripper. However however, there are many people, including the manufacturer who don’t have reliability problems with this software. TL;DR : Gig Performer is excellent software and I wouldn’t hesitate to recommend it, but you guys know how nitpicky I can get…take all of this for what its worth]
Still, in our scripting, we’ve been able to get decent results crossfading, and then muting at the end of the crossfade to save cpu. There is still a time cost, however, the audience won’t hear a gap.
Generally, taking a predetermined set of FX and then storing different values of their parameters, and recalling them at will.
Advantages: Usually switching FX on and off takes very little time and can potentially be gapless. While a supermassive chain of FX could be potentially daunting to CPU use, it will usually be less resource intensive than multiple patches running in parallel
Disadvantages: You really give up a lot of flexibility without being able to switch FX out totally, though this can be mitigated somewhat with long chains and various routing schemes, though again, at the cost of higher resource use compared to a single preset of each patch setup. Glitches. Turning on and off FX, switching routing and even parameter value changes can be extremely glitchy, though they don’t have to be.
- Cached banks
Having presets loaded into memory, or some other scheme, without actually running them in parallel (except perhaps for a crossfade between them), with only the active patch using any significant CPU. ReValver does this absolutely amazingly well in Gig Mode, though it can glitch a bit.
Advantages: This can be mind numbingly fast in switching time. Can be extremely low CPU use. Gives all the flexibility of presets, without incurring many of the issues of preset switching. Create the exact chain you want, and switch between them at will.
Disadvantages: Can still glitch if you aren’t careful. Like the other systems, buffering latent plugins (like Helix and Amplitube) can cause glitches and even stability issues. The number of patches able to be used in this way can be limited, for instance, ReValver limits this to 8 patches. I propose though, that you could switch banks and upon the first patch you press, the new bank would be loaded. Doing this between songs, for example, shouldn’t be much of an issue, and optimization could of course happen to limit the time bank loading takes
I would propose that Bias FX 2 looked into cached switching if possible. Of course, this does not preclude it from preset switching (and optional crossfades between presets would be very, very cool), or, with a few changes detailed later on below, snapshot switching. By externally controlling Bias FX right now, as in our LBX Stripper script, you can do very comprehensive snapshot switching, though some functions are limited or absent.
Additional DSP Models
Again, keep in mind, and perhaps I didn’t state this strongly enough, but a lot of the goal of this paper is for a completely self contained BiasFX, which would be plenty powerful on its own, though third party VST support inside it would be absolutely wonderful, I don’t think this would translate to iOS very well, so any functions it critically needs, would have to be part of the Bias FX system. Though again, third party VST hosting, especially in a combo amp, rack or pedalboard would be most welcome indeed!
I think Bias FX would benefit greatly (and maybe this could actually be two or more new families in the Bias Pedal system), from the following new types of FX blocks:
Multiband Compressor / Dynamic EQ – What’s a constant topic on the forums and TGP? Low end and mud issues. Someone gets their sound perfect on one set of speakers, but whatever they plug into at the gig woofs out on them, especially during palm mutes. The soundman goes ballistic, cats are living with dogs, its pandemonium! Even a single active band of multiband compression can make a world of difference here. After some testing, I found that Postive Grid’s own Final Touch works amazingly well for this, so you guys already have the DSP for it! It would have to be no/low latency of course.
Impulse loader anywhere in the chain. Impulse loaders can actually get you out of needing to add several linear FX. People can pop in their cabinets or EQ sounds and reverbs. I’ve had great success with using impulses for acoustic guitar simulation. This also puts certain FX in their correct place between amplifier and cabinet (though an FX loop between preamp and power amp might make some purists even happier). In addition, this could be a huge help for the dual path system, so you could easily chose which output gets the cabinet impulse for the PA system
FX Loop(s) – Starting from the Line 6 HD500 and especially well implemented in the Helix, a software movable, software switchable FX loop can be really handy for interfacing with the outside world. This can also help with the dual path issue
Harmonizer – This is a biggie, and can be quite difficult to implement, but it is one of those things that really is on the spec sheet checkbox that people quite often look for. Doing it right is even trickier. Often companies will place these after the cab modeller, resulting in some disastrous noises. Ideally, it would go before the amplifier, assuming the amplifier has a true stereo input. This works fine in Bias FX VST, with the original signal going in one side of the amplifier and the harmonized output at 100% wet going into the other amplifier input. With Bias FX’s existing dual signal path system, this could also work on two different amps, but could bring up some issues with the other important dual path situation, one with a cab emulator and one without (though are potentially tons of ways to route around that)
Again, these could be parts of some new pedal families, or parts of existing ones for Bias Pedal.
Some functionality enhancements that I think would greatly extend Bias FX’s capabilities as a live processing system.
Auto engage – This is a huge one. Once you have used it, it is really hard to go back. Moving a pedal turns on the FX block, pulling it back to zero (or some other value) bypasses the FX block. We’ve made various scripts, plugins and threshold switches to really fine tune the control of this function, such as separate and possibly overlapping on and off thresholds, dwell time, and hold times
MIDI Enhancements –
It would be really nice to assign more than one parameter to the same CC or automation in the MIDI/Automation assignment window
Much more comprehensive MIDI control parameters would be nice, such as ranges to toggle things, minimum and maximum values, critical things like polarity invert and maybe even curve types for parameter values (for instance, volume pedal control in many of these types of plugins is not so good)
Values set for on and values set for off for parameters like FX Pedal bypass. The current situation where these parameters are toggled instead can and does often lead to logic errors, leaving the particular FX block in the wrong on/off state
A master MIDI window where all assignments and control info can be viewed, modified and assigned at a glance. While some of this is existing in Bias FX currently, it seems rather divorced from some of the other MIDI and automation assignment windows. Amplitube VST actually does this part pretty well
Basic GUI control functions – Things like fine tune control (which can be really fiddly right now), direct entry of parameter values, return to preset default, return to FX Block default (both of these for both individual parameters and the entire FX block, and really importantly, some way to see the current value of a parameter without changing it (this one can be a nightmare when trying to match two different presets right now)
Wet/Dry and other utilities for all FX blocks – Some of the pedal models have wet/dry controls, some don’t, and many or all of them could certainly benefit from this function. Also many of the dynamic plugins do not have threshold controls, so a way to turn up the input would be nice. Also, some pedals won’t turn down their output past unity, and would often benefit greatly from doing so
Independent Left/Right bypass – At least for the IR block. This could kill the whole dual path issue in one fell swoop. For other FX this would turn your existing Dual Path setup into 4!
Per FX block presets – This would end so much frustration, and really give you the ability to mix and match favorite settings of individual components between presets
Right off the bat, I have to tell a little story about ReValver. Peavey was one of the first companies to come to us about licensing REAPER to run on their products, so we’ve been in pretty good contact with them for a decade or more now. Of course we flogged ReValver when it came out and actually used it on a lot of albums early on. At the time I had no interest in using this sort of thing to play live so I didn’t dive very deep into that side of it. I have spent the last year now going through the various products on the market aimed at using our DAW stuff live, and spent months on scripting and plugins before someone at Peavey told me to look at ReValver’s Gig Mode, which did about 90% of all the scripts and extensions and plugs I’ve been working on do right out of the gate!
So here I was, with a company I’m very familiar with, and a product I’d been using for ten years, and I didn’t know about a feature it had that should have been shouted from every rooftop! It blew my mind.
You guys have a lot of artists out there, a lot of good buzz, and a lot of good press. At the same time, it is extremely hard to find information about using your products to the fullest. I completely understand and would rather have you spend your time developing awesome products than writing manuals, but somehow, some way, actual use information needs to get out there more. In 2006 before I think there was even a youtube, I made some really crappy Getting Started in REAPER videos, which ended up really launching the thing in many ways. You guys have KILLER videos out there. I’d love to see more how to’s about basic and advanced functions
I know I’m not the only one to say it, but the current preset management system can be a pain. Since the menus in general in Bias are extremely sluggish, getting in and out of Tone Cloud can be really frustrating. Much has been said to hopefully enhance the Cloud as well, like search filtering, and an easy way to delete and rename your own Cloud offerings. An Import/Export Preset and Bank function would be really nice from the existing preset menu.
Off the Shelf – I have several working systems which are capable of taking to a gig and setting up very quickly, using regular products (though often heavily modified) from music and computer stores. These set up with a minimal of cables (even less than most “regular” guitar systems), and can include wireless controller pedalboards as well. However, they look (and look they they would feel like, even if they really don’t) like the proverbial tornado through an airplane junkyard. Looking at them, they take up a lot more space than they need, and have several confidence issues with mounting the computer or monitor. Internal cabling is a bit sketchy and it just feels cobbled together
Turnkey – Its quite conceivable that I could have some chasis fabricated and use some off the shelf parts to make a really clean, reliable, confidence inspiring system, using existing, readily available products, and sold as a complete turnkey product. And that this formula could be duplicated by users who wanted to do it themselves in varying degrees. However, I’d so rather not! I look at what I’ve done so far more as a proof of concept and hope that you guys could build this whole setup instead.
Bias FX Rack – Now we’re talking! I realize that much care needs to be taken not to obsolete existing Bias Heads and Bias Racks, and I think that this is very doable with the FX Loop functionality detailed in the last section. This should definitely have separate volume controls for the guitar amp optimized output (no cab) and the PA/FRFR output (with cabinet emulation). Definitely should have comprehensive MIDI control, but aside from the physical MIDI ports, this should already be handled with the MIDI functionality already existing in Bias FX and the MIDI function enhancements mentioned in the last section. The ability to handle a wireless, battery powered MIDI pedalboard would be nice, but at this price point, most of the buyers probably wouldn’t be averse to buying themselves something like the Panda MIDI wireless to put in their own MIDI control pedals. Of course, the proper combination of 1/4” and XLR outs phantom power protection on the XLRs!), and a USB port both for using as an audio interface (with again, the ability to turn off direct monitoring) and for importing and exporting presets, updates and general backup utilities. A 9v output port for powering auxiliary devices.
Bias FX Pedalboard – The Helix killer. Should have similar FX loop(s) and output considerations as Bias FX Rack detailed above, but with a built in pedalboard, including expression pedal(s)
Bias FX Combo – As the two products detailed above, but built into an FRFR speaker. Because of the onboard USB audio interface, this could conceivably be an extremely convenient way to add 3rd party FX using a laptop or tablet
Current Issues in Bias FX:
Though I’d love to see these fixed today, I put them here rather as something to consider going forward to Bias FX 2
FX block Toggle instead of on/off – I understand why toggle can often be the desired behavior, but quite often it isn’t and leads to logic errors. I would love to see actual on/off control values as well.
Some functions, such as amp on/off aren’t in the automation assignments but sometimes can be found in the other MIDI menus
Fine tune control over parameters, along with direct entry and values at a glance – I detailed this more elsewhere so I won’t dwell on it here
Issues with MIDI polling and message overload – Bias FX can respond very poorly, and even crash if it gets too much MIDI info. When polling for current state, asking for too much information from Bias FX can also freeze or dropout the audio flow.
Sluggish menus, and general stickiness and long loading times – I have detailed a little bit about how I think this can be dealt with, based on what I think the cause is, and can give more detail if needed. Right now, even launching Bias FX on a very powerful computer can take what seems a ridiculously long time. Bringing up the Amp replace menu is such a time consuming frustrating experience that I know I often settle for an amp instead of trying different ones, for instance.
Writing too much hash to the DAW’s project file – Bias FX writes more encrypted data to the project file in even one use instance than all the other plugins in a project, plus the project itself do! I have a strong suspicion that this is intrinsically related to the sluggishness above and have some ways to deal with it, which I can detail if desired.
In Summary/ TL;DR
Positive Grid already has some knowledge building MIDI pedals, and amps, and especially, the best sounding DSP on the planet, and most importantly of all, the knowledge to make this all work.
If you don’t build this hardware, but could at least get some of these changes into Bias FX 2, than its still a huge win, as I don’t have to worry about the software side of it anymore. I know you know how horrifying it is to consider all the little things that can go wrong live from some little bug or issue. Its a million times worse when you are using a DAW program as the nerve center, which wasn’t in any way meant to do this live, and connect all sorts of software that all hate each other. It would be so much easier to just run some Bias FX hardware!
Either way, this is a very exciting time for me, its as much fun, and as much stress, as when we were first making REAPER! Its like the whole ride starting up again.
I hope to hear from you guys soon!