Thanks for the shout out John, you’re a total legend! I have a couple of videos on my channel showing how to set levels for various plugins, and even to measure your gear if the manufacturer doesn’t provide specs. Just a quick correction for you, 8.5dBV is actually equivalent to 10.72dBu so the amount you need to cut (if you want an accurate response) is only 1.5dB. The best advice is to look for the “Maximum input level” specs in the interface manual, usually it’s written in dBu with gain at 0 (i.e. the maximum amount of headroom the unit provides). Most spec around 12dBu because it’s equivalent to approximately 3V, which is the max you’d get out of loud hunbuckers or a 9V pedal. If you set the gain to 0, it’s likely to be close, but interfaces do vary, and some can be -3dBu, while others can be +19dBu. Neural DSP, Line 6, Nembrini and others aim for around 12dBu, so if your interface is different to that, adjust in the plugin to land in that area as John says.
@@eds4754 I mean I totally botched the name so I'm sorry!! And THANKS for the intel - seems you've been really instrumental in getting that information out there at all!!
@@johnnathancordy You’re welcome! I was hoping someone with a big audience would share this information as I’m convinced almost nobody is using the intended levels, and has no idea there’s even an issue!
Behringer says that the UMC204HD has a Max. input level Mic: -4 dBu / Line: +20 dBu / Instrument: -3 dBu If i plug directly to the audio interface, "the rule" still is to have the input LOOOW on the behringer and compensate the input of the plugin, and finally, adjust the output of the plugin to the db level i want to hear from monitoring? The dBu thing is very confusing to me :-S
Also, for the Focusrite users, make sure you still have the "INST" setting on. With it off, it's 22 dbu at minimum gain, meanwhile, it's 12.5 with INST on
For 2 years I was using it on line... I always wondered why I couldn't get good guitar sounds lol. I only realized what was going on a month ago... fml.
@@lechug43 I did when I bought it, and at the time, from what I read I got the impression I was supposed to use Line in my case and I set it that way and never thought about it for quite a while.
Offcourse, these interfaces and their makers are from a generation where plugins were only at the relative basic, from an marketing, business point of view alone, the company would have spend more budget on marketing and r&d geared towards vocals or recording tube amps with mics
Now with this in mind, I truly believe this was the case... In fact, let's go even further... plug-ins a few years ago weren't as great as they are today thanks to the advancements in tech and sound development. I'm sure that plug-ins were considered the same way as many libraries may have been considered for; to do mockups, demos, etc. not full professional sounding records when in reality it's more than possible. So, the recommendation may not come from the place of ignorance or to be actual bad advice with a bad intention and more of a: you should record with a real amp cause that's how it must be done. You can write with your plug-ins and then perfect it with your real rig. My two cents... maybe I'm overthinking it, maybe not. I only know that I've got to go to bed hahaha.
It would be nice if plugins would give you more visual feedback about where that sweet spot is supposed to be. Some do some don't, but Neural doesn't at all and it seems actually quite important.
The reason plugins can't provide what is that "sweetspot" is because every guitar pickup, tone knob, volume knob is different. The best advice is what he provided. Plug in the Hi-Z input and set the input gain to 0 then control the gain with your volume, tone and pickup selection on your guitar, or even a pedal board going into your input if that's what you would do on your normal amp. The idea is the amp input should be "dumb" just like when you plug in a guitar, some guitars have higher output levels than others. And a lot of guitar players at a live gig like to keep the input knob at half volume for normal playing so they can turn it up when they need to do a solo or something special without front of house engineer needing to adjust anything.
Man thank you for this. All the horrible fizz has gone from my distortion tones. Came for the absolutely rad playing and stay for the absolute rad playing but also top tips like this.
You can reduce the signal level AFTER you've recorded it if you're printing the clean guitar signal. The plugin won't care, as long as you haven't drastically altered the clean signal (i.e. distorted) much. Same goes the other way.
That's precisely what i'm going to do for my current song that I already started to track and don't want to re-record everything. Hope we won't notice a difference in the final mix! 😬🤞🤞
Exactly. I think this video might be making this thing a little more complex than it actually is. Depending on DAW you just have to make sure you're setting the desired level from pre-fader and obviously make sure your guitar isn't clipping. Using pre-fader on the track is easiest cause you can use it for both live and recorded. But ofc there's many ways.
I'm just flabbergasted, been grinding my gears for years trying to figure out why my amp sims always sound lackluster in comparison to every video. But now when all I did was turning my interface gain down to zero and same within the vst it sounds so good! Thanks a bunch!
@@sheshehangbecause your guitar is going to still send a signal. The gain knob is just for additional boost like a boost pedal. Which should be done within the plugin if you're even going to do that.
I discovered this early on by accident. When I was discovering the world of VST amp sims, I would purchase "presets" for helix native that sounded great on the RUclips demos. They ended up sounding very different and gainier on my laptop setup. I found turning my gain all the way down to zero on my Focusrite Scarlett it got me closest to their tone. Good to know it was the right call all along.
This advice also applies if you’re planning on reamping. I always record guitar to at least have the option of reamping it, and with my interface, if I leave the input gain at zero, and playback the recorded guitar at -10db to the amp, it sounds identical to when I plug into the amp directly.
Holy shit, i'm reevaluating my life right now. For the past 10 years everyone said the wrong thing on the internet about guitars and interface ? Even pro like misha back in the day said that you should always raise the input on the interface until it almost clip when recording guitars
Agree. I have been using guitar plugins for a number of years and found this lower input setting works better. I was never happy at first with the sound of plugins and then started playing with lower input settings and this also resolved other issues that bothered me like aliasing and HF abnormalities with clean sounds. I am now actually using plugins for gigging live. Use a midi foot controller for all the fx, patch changes, expression, etc..
I bought a high power car amp from my friend and I had an old bose am4 subwoofer, so I did some math to figure out the gain settings so I wouldn't blow the subwoofer, I had it set perfectly it sounded really good. But I kept getting greedy an adding just a little gain so it was louder and louder I think after the fourth or fifth nudge it blew
Actually there's a video on the ndsp RUclips channel where they set the input on -30 to get a breakup tone. This made me realize this an important parameter to adjust. I don't change the gain on the interface as the Noise to signal ratio still applies, I just adjust the input level by ear. For heavy tones, sometimes a high input gives a nice saturation
The biggest eye opener that I've recently discovered with some plugins and especialy with the Helix Native that some amps tend to sound a bit "wooly"is to boost some highs and cut some lows at the beggining of the signal chain. I always felt that some amps in the Helix Native sounded too "wooly", dull and sometimes boomy and not really articulate. I've just put simple EQ block as the first in the signal chain, boosted highs quite a lot ( 10db) and now I fell in love with the Helix tweed style models even though I hated them before. 2204 started to react much better and now it has this "squishy" character that I absolutely love. Thanks for the video!
The maximum gain without clipping advice comes from both maximizing the SNR, as the preamp has better noise characteristics at higher gain, and ensuring that the voltage levels on the line are maximized over what the analog to digital converter (ADC) can accept. If you don't do this, the signal gets quantized in a way where a number of bits don't actually get used. What I mean by this is that if you have a 24-bit recording, you maybe cutting out a few bits because your signal to the ADC is too low, and you're not getting the full dynamic range. Maximizing gain without clipping is all well and good for capturing a signal with the highest amount of clarity and detail as you possibly can. The problem with this is that you lose reference to the ABSOLUTE signal level. What were the actual voltage levels that your pickups are putting out and at the end of your guitar cable? That's what's getting lost here, is how do you map that absolute input signal level back to a referenced line level and how do you standardize that across interfaces and plugins? For best clarity, it would be best to use max gain without clipping and then digitally attenuate/scale the signal to the appropriate level rather than reducing the gain at the input at the preamp. A solution would be to set maximum gain without clipping, then plug in a device that provides a reference tone with known voltage amplitude. Then, in the plugin, digitally adjust the input until the level matches the reference output.
This is all correct, however the SNR isn’t really determined by the converters in this situation as the noise picked up by the pickups is going to mask it a lot. As long as the interface used isn’t exceptionally noisy, or the user isn’t recording VERY quiet (quite hard to record a DI that’s too quiet in 2024), SNR is unlikely to be as big of a problem as not knowing the input headroom. But yeah, the ideal is gear with stepped gain controls so you can set your headroom reliably to whatever you like (and can compensate accurately afterwards). Also possible to do with a sine wave as you mention, although it’s a bit cumbersome to calibrate every time you touch the gain control…..
One of my biggest frustrations from most of the channels that showcase VST amp sims is they never go over the fine details like this then people like me and most of the commenters are left dumbfounded. I really appreciate you showing this to everyone!
I've used plexi-type plugins for most of my life and I'm quite used to overdriving the plugin by having more gain on the input. I actually enjoy that sound too, but this is great advice for how to set it "right".
Well yeah a lot of where you put dials is subjective. But I think it makes a lot of sense if you're using a plugin that comes with a bunch of amp models and presets to have the front end set from the interface so that it's the "same" as plugging into the real amp. Of course if you have a different guitar then the presets probably won't sound the same, but you'll be in the right ballpark at least with a similar guitar and a preset marked 'clean' might actually sound clean etc. I think one aspect of amp modelling is the idea I could capture my amp. Then I can plug into the real amp + cab set some settings on the dials - get a tone. Now I should be able to plug into the amp model, with the same amp+cab selected, put the dials to the same settings and get an identical tone. If that happens I might decide that amp modelling has arrived, it's a joyous day and I just use my RL cab to stand beers on. Or something. Most of us at this point remember that we don't own the real life amp and cab as the beer crashed to the floor. But as we're mopping up we're pleased we don't need to buy one now anyway. Obviously there are many reasons why when I do that test I don't hear the same tone and I think "This amp model is wrong" (presumably I don't assume that real life is wrong). If I have input level gain so high that the amp model is overdriven even on the lowest volume then my 'same settings' thing won't work even if the modelling software were a perfect recreation of my amp then I'm going to say "This amp modelling software sucks" - if I'm naive and don't realise how my setup has affected things. Otherwise, there are no rules about where to put dials.
I stumbled upon this concept a few years ago. Almost all my amp sims sounded too saturated. My UA interface defaults to +10dB gain on the instrument input. This is as you said analogous to having a 10dB boost pedal in front of an amp. But the amp sims model the amp - not an amp plus a boost pedal. So now I simply turn down the input levels on the amp sim plugins by about 10dB, and the amp sims sound a lot nicer - particularly on “cleaner” or “edge of breakup” tones.
I've been doing this for many months now but i'm still not sure if it's the right thing to do. My Focusrite Scarlett 2i2 Gen 4 defaults to +7dB gain with the instrument input and i've been reducing plugin input by 7, everything gets amazing compared to it sitting at 0.0 but i'm worried about the thought that i might be killing my signal, maybe? It'd awesome if you could or someone else enlighten me on this!
I think the advice of setting the preamp gain on your interface as high as you can without clipping is sound advice for recording a good quality DI signsl if you will be using the DI for reamping or swapping between plugins with the same recorded track. That's why a lot of plugins come with an input level before the gain knob on the plugin. It's just good to know that some plugin companies actually list the suggested 'input' to the plugin. So, ask yourself if you need your DI signal at the right level if you will use a DI for other purposes.
I record DI via my little labs box. My instinct would be to put the line in gain as high as possible for best signal to noise ratio. But then to get the optimum gain level into the VST do you think I should use a gain plugin with a meter and set it to say 12dB ?
Just to slightly clarify dBFS vs dBV for the UR-RT2: dBFS is the level relative to where hard digital clipping happens (i.e., relative to the largest number the converters and DAW can assign to a signal amplitude). dBV is the level relative to 1 VRMS. The text in the video indicates Neural is assigning 0.707 (1/sqrt(2)) VRMS to -13 dBFS. 1 VRMS (i.e., 0 dBV) is a factor of sqrt(2) larger than 0.707 VRMS, which corresponds to 20*log10(1.414) = 3.0 dBV. So for Steinberg's UR-RT2, if for the hi-Z input 0 dBFS is +8.5 dBV (2.66 VRMS), then -13 dBFS is 0.595 VRMS, not 0.707 VRMS. For a 0.707 VRMS signal to be at -13 dBFS with the UR-RT2, you need a gain of slightly less than unity, .595/.707 = 0.84, or -1.5 dB.
I used to have my input gain way too high and was actually clipping with the gain on my interface at 0. Fixed that. Then read the Helix Native manual and realized I still had too much gain. Made a world of difference in most cases. In some cases, that crazy amount of gain actually gave me kind good tones.
Thank you both guys.........as I'm approaching to this world I realize YOU are the most prepared and sensitive person on youtube...........I LOVE your style and trend
This whole discussion has been a revelation as I had been very disappointed with my plugin (Scuffham S-gear) for all the reasons mentioned. Tried setting interface to min and immediate improvement. Only downside was minimal recorded track which did not seem good as it doesn’t use the full digital bandwidth - you would never record your modeller at that low a level. I am now setting the interface to give a good recorded signal and adding a gain stage as the first in the channel to pull signal levels going into S-gear down to where they would be with interface at zero. Any minor adjustments can now be made on the gain block or the S-gear input trimThis seems to work well and I’m now much happier with S-gear and can see where all the good reviews come from. Really grateful to all who have contributed to this - now off to wallow in lots of clean sounds.
This makes so much sense now that you say it. Applying the same gain staging methods for mics to guitars being fed into virtual rigs…of course it would be different. But I don’t know that I would’ve ever realized it. Thanks!
I was today years old when I learned this. I also watched Rebea’s video about this. I have been cranking the gain on my interface when using an amp model and it’s always sounded terrible. Yesterday I stumbled on Rebea’s video and today yours. Now my tone sounds so much better.
I kind of did this already. I turned my Steinberg MKII barely up, because putting it at zero made it feel odd. Then i reduced the input gain to level it with other recordings i have. If you want to find the sweetspot record a DI signal that's pretty quiet. Then you can loop it in the DAW and turn up the input gain until it sounds about right. Then you can either copy the Plugin Value or turn down the interface.
Looked up the value for my interface: Arturia Minifuse 2. It's maximum input level when using the instrument input is +11.5dBu. Almost right on the money. Also checked on the Minifuse 1 and 4, they have the same value.
Setting the interface input at 0 has always made sense to me, perhaps because I’m an amp user mainly and only dabble with modelling. My first modellers were line 6 pod , no way to adjust input levels… move on years later and I got a Boss GT1000 and couldn’t understand why you were supposed to alter the input gain to match guitars. So my Strat needed more input level than my Les Paul , but it made it sound like a boosted Strat. Jason Sadites goes into this in great depth with Tonex and the effect on captures, once you get it right it really opens up the pedal.
The problem with line6 "no way to adjust input levels" doesn't actually tell you where to set the input level on a plugin - and it isn't helped that they don't give any specifications for the max input level dbu for their hardware - and word is they specifically omit this information because they're scared people will think their hardware sucks. But that actually means they suck anyway. Because now we're sat using a podxt as an audio interface scratching our heads - or having to buy equipment to measure it ourselves. So, sure, you couldn't set the input gain, which means you couldn't set it too high. But that doesn't mean you're not going too hot or too low into, say, helix native, tonocracy or neural dsp plugins because each of these were created with a specific dbu value in mind (in fact for some software even different amp models would be expecting different values - although tonocracy fixes this - if you set the input gain "right" then all of the amp models, tonesnaps, presets etc should be "right" - I put right in quotes because, sure, this is subjective there is no right per se if you want to turn the input gain "too high" to create a specific noise then you can) If you have a focusrite interface it turns out that setting gain on zero is all you have to do for some of these plugins, but if you have a pod or any other interface you can't assume the same thing - and, in fact we know that for many of the interfaces you have to turn the input gain up or down to match as well as setting gain on zero. Specifically you can't even assume the lin6 hardware is giving the right input for their own helix software. When you consider their amp modelling is infamous for creating fizzy tones and for people struggling to get a good tone out of both their hardware and software perhaps this partly explains why. All we learn noting that line6 lack an input gain adjustment is that they probably really were clueless about what they were doing. Note that most of the manufacturers of hardware interfaces and plugins gave the wrong advice about how to set input gain. So don't assume they know what they're doing. Now they're creating videos saying the opposite and acting like they knew all along of course. The joke are the videos where people are saying "I never realised you were all setting your gain?!?!" as though the end user was getting it wrong - they were following the instructions given to them.
I've seen so many people crank the input levels on their interfaces and complain that plug-ins sound awful like yeah because you're clipping the hell out of your signal. The first thing that I did when I started using Helix products, about 4 years ago, was use the rack unit as an interface and see where the DI channel was metering in my DAW and use that as a baseline for all of my other interfaces. I use a Behringer UMC 18i20 in my studio setup and a UMC 204HD as my mobile interface and the difference in base levels for each is shocking considering they use the same hardware and preamps so make sure that you check the levels of each unit if you are switching between interfaces.
To be honest, for people that just want to jam this is good knowledge, but it might be a bit complex to understand. It's very simple really. Make sure your signal from guitar to interface isn't clipping and set the desired level (-13 or so db) from the channel's pre-fader in DAW. It has the exact same result as trying to find the proper gain from the interface and you don't have to go measure analog outputs. Most of all interfaces have almost exactly as clean opamps and AD circuitry so there's only minuscule differences in the sound your interface is sending to the DAW at any given reasonable gain.
I have definitely been overdriving guitar rig pro, in fact I abandoned it because I didn't understand why it had zero correspondence with a lighter touch/lighter picking. The Marshall style amp that I wanted to used was always face meltingly aggressive. Now I know what to do thank you.
Wow. I can't believe there is not a LOT more people talking about this, I was 100% for sure doing this and for a minute of there I was thinking there is a literal problem either in my software or maybe in one of my pickupups, and I'm playing a Strandberg metal NX8 which I know for a fact is in good shape being the second one I've had, so I was getting pretty confused. I did this and it literally fixed everything I was trying to get a good tone for selective picking and thumping, Animals As Leaders type stuff, and anytime I would switch to a single coil and sound on my guitar I would get so much background noise that you couldn't hear what I was playing it was getting ridiculous. Honestly Neural should be shouting this from the rooftops so people don't start thinking their stuff sucks juts because it's not being used properly, when it turns out "properly" in this case is definitely considered improper in almost every other case and goes against traditional knowledge. But I get why they designed it that way, for the consistency in response between different guitars & interfaces with different outputs, it's a way of making sure it always acts the same. You'd think they'd make it higher though, one of my interfaces can't even go that low. luckily the big one can. THANK YOU for this video!
I'd always thought that I needed a new guitar, because mine sounded muddy, darker and with some dirty frecuencies in the high range. When I set mi Focusrite Scarlett to 0 in gain and -1.5 in the input of the plugin I realize that I could save more than $1.500. I have a cheaper guitar, a G&L made in Indonesia, but now it sounds like a pro. Thanks for sharing this crucial information.
Thanks for addressing this. I have been confused about this before myself and have come to my own conclusion. This is why it's helpful, in general, to learn to dial in sounds by ear and listen to what your gear/plugins are doing. It's all a part of learning to gain stage. Sometimes audio interface preamps sound better turned up a bit. So then reduce the input directly in the plugin or with a gain control plugin if needed. Think about what you are looking to get out of the gear -> gain stage accordingly. My real amp, on clean, sounds better when pushed a bit but would sound bad if I pushed it too far. With plugins it's no different. I've used the interface input gain knob as a clean boost before when I want a little more juice out of a sim. Sounds bad if pushed to far. Everything is gain stages right down to the fingers, strings and pickups. I get that it's tricky, but learning this will help people get so much more out of the gear/plugins they already have.
I have another thing to chip in for this discussion. If you are using a wireless unit(trying to be lazy and not plug cables around your office), like a Lekato ws-80, you do not even need the hi-Z input engaged and probably you should also reduce your input level on the plugin. I learned this the hard way(not understanding why presets where more trembly and loud with the wireless - and in no way similar to the ones with the cable.). Also if you double click a slider on the native it defaults back to the starting value (easier to set input gain back to zero this way)
It'd be great if someone from Line 6 actually says what's the signal level they use to calibrate their Helix series so we can perfectly match our presets on Helix Native and Helix hardware because it might not be the same as the levels Neural DSP use, I actually commented this issue several times on your videos but never got a response for you as I was always asking you which exact input levels you were using on your interface to match your Helix hardware, I have an iConnectivity Audio4C and found that I actually have to put my instrument inputs on 18dB and Hi-Z mode to match the HX Stomp, on my other interface, a Behringer UMC202HD settings levels at 0dB and HI-Z is good enough
@@eds4754 Awesome!, thanks! I just subscribed to your channel too! I'm glad I finally found this answer I was looking for and strangely no famous youtuber guitarist mentioned or covered before, BTW, do you know if on iOS they use similar input levels?, I use some Nembrini amps, THU and Aurora DSP apps and it's also hard to match levels on their presents on different platforms
That’s kind of funny because on my focusrite 3rd gen I have always kept the gain knob on zero with “inst” engaged and thought I was doing it wrong but I just liked how everything sounded a lot better. As soon as I would try to set the gain at “recommended” as per your initial statement it always sounded flubby and washed out. So I guess I’m not as stupid as I thought 😂😂
So stoked to get some solid info on this! This confirmed what I thought I was hearing and the best place to start to get the best sound of of DSP. Thanks, bud!
This is exactly what happened to me when I bought Imperial Tone King plug in to use with a Focusrite Scarlet. I couldn't get a clean sound to save myself. I even tried to return it and they wouldn't. Then I worked it out and started loving it. (Mind you, they didn't explain this to me, I had to work it out myself.)
Excellent Advice. I noticed that when I compared my preset pack made on the hx stomp to the sound of helix native. Way too much gain in native! This video should be watched by everyone!
Yes indeed, now the clean sounds clean. Always wondered why they broke up so easily before. Getting much better tone all around with Neural plugins. Thanks!
I wonder whether the artists creating the presets in Neural plugins are following this method or not, because that's going to affect how the presets sound. Maybe some are and some aren't. This seems like a pretty vital thing to have not been made clear.
I've never had a problem with near clipping. As long as it doesn't clip, that's all that matters. Some plugs (TH-U) require a near clip level for accuracy (TH-U also has a humbucker and single coil sensitivity to ballpark a match). There's plugins that have a strummed autogain that will turn UP even further than near clipped in the plugin. Some will turn the level down. Just learn what the plugs want or if they don't say... Adjust to taste. You will never know if they are truly accurate anyway and who cares as long as it sounds slammin
Hey John, I have a little bit more insight on why this is. Modern amp sims that do profiling/accurate modeling and "amplify" the raw DI signal as much as 3-4x. With some Neural amp modeler profiles, my raw DI was set as low as -36dbfs to get a good tone. Older amp sims were just clipping and adding some compression so you would want as much signal as possible without clipping. Newer amp sims actually simulate the amplification so my Audio Interface signal was set as low as -40dbfs for high gain amp profiles and to as high as -20dbfs for cleaner amps and the output signal sat perfectly in the -18dbfs to -6dbfs range on my DAW for tracking. And the volume knob on my guitar at full would give me the perfect/slightly more than the perfect amount of gain for lead tones and at halfway the perfect crunch/edge of breakup tone.
I'm so happy I stumbled upon your video, always had this horrible fizz in my distorted tones and some of my guitars wouldn't get a clean tone ever. That's gone now with more clarity, needed to adjust the amp settings for the loss of gain but man, this is great. Thanks for uncovering this.
Is there a reason why this should be done already at the audio interface? Why not just turn down the input gain on the plugin? Wouldn't you then get all the benefits of a healthy, strong recording level (the waveform would be more readable visually as well) and also the clean-sounding end result. Is this just a workflow preference thing? Or have I entirely misunderstood something?
@@seanedwards94 Yeah, but what’s the problem with a gained-up DI, as long as it doesn’t clip? Wouldn’t that still be good and clean? Or is there some specific benefit?
@@ShreddingDragon I'm agreeing with your previous statement, and saying there's others that corroborate what we're saying. Nothings wrong with a strong, healthy, DI in your daw. You just trim it down going into the plugin
If you set the gain to your interface as high as it will go before clipping, and then zoom in on the actual waveform of a recording, what you will see is that the audio will be full of little micro-clipping. You can't see it unless you zoom in really closely, but it's actually much safer to always have your gain much lower than you think is necessary at the source. If you raise the gain later that is fine, but you won't have that head room unless the gain upstream in the signal chain leaves that space. Movie people have known about this for years when they do sound.
This is gold! I was 100% doing this per the old advice, using a DI and XLR into my Scarlett 2i2 and boosting for every different guitar/bass till it clipped and then backing it off. What's funny is that was varying between -18 and -12. Well, I just checked the documentation for my Scarlett 2i2 3rd gen and plugging in direct with the INST button on and the input volume down and it was a pretty darn smooth -12 with less volume variation. I agree that it is better on the way into the Neural Plugins and now I know how to check for others!! Thank you!!
Aiming for -12 to - 20 for the input signal is a well known thing, even before the digital age. This allow the necessary headroom for peaks. There is another limitation I discovered the hard way on my Behringer Uphoria that may not be present on other brand. There is a saturation that happen even before the clip led turn on. It seems internal, so I have to lower even more the input. To test that simply, set the level as you would do normally then listen the interface without any plug-in and strike the strings. If you here a saturation, lower the analog input volume is the interface, compensate the output volume in the DAW. If at almost zero input you still here the saturation it's something else, but if at some point you hear a clear sound it means you had an internal unknown saturation.
This is very VERY different than what I've been doing and explains a bit about why I never really got a fully clean tone on the Tone King from Neural DSP. Going to experiment with lower gain levels. Thanks
Same here! I got that cause of the reviews and was exited to maybe finally have an amp sim that would actually sound good. But thought I got a blues amp, cause I couldn’t get a nice clean to save my life
I started using digital VU meter plugins on every DAW channel before and AFTER vst plugins. IDK if it's proper but it has corrected my gain staging immensely.
I realised this when I bought native and noticed my presets sounded much gainer, for awhile I was manually pulling down the gain on my protools tracks until it sounded right then learned to turn the gain to zero on my scarlett and use the instrument input. Now I record a DI signal simultaneously via the stomps output 5 and that takes care of it and sounds identical to the hardware.
Yeah, makes total sense. Also, I have a separate track presets for cleans and highgain tones (any amp sim), and the highgain preset has 3 disabled pedals before the amp, so I picked the one that sounded best for an amp sim, no matter which one - NeuralDSP, Plugin Alliance or Softube or some other amp. And in the plugin I usually have to ADD somewhat around 6.0-6.5 dB of gain. I think, it's better to utilize metering plugins as well, since most of the amp sims don't show the actual level before going into the amp. Thank you, I was thinking of redoing my clean tone presets, and now I think this is a great place to start - gain staging.
in my experience, any audio plugin is like this, kinda 'tuned' to receive -13dB~-11dB. it works with any sound source... this advice to set the gain as loud as you can without clipping goes back to the beginnings of digital audio, when the noise floor was high and the bit depth was lower
I noticed back when I started using helix native I was clipping and it didn't sound right. I have a Behringer UMC404HD interface and found that setting it to instrument and adding the pad button which I believe is a 10db attenuation and gain about 1/4 way up matches what helix does. The Helix native plugin also has an input slider on the far left of the signal chain.
Ok, so tried this... When people in demo videos would say, "This Neural DSP plugin reacts like a real amp when you roll off the volume knob and actually starts to clean up...", I could never actually experience that myself. It does that thing now! I have a Scarlett 2i2 set to "INST" mode with the gain knob ALL THE WAY DOWN... now. I did notice within the app I typically have to boost the input a little bit more than before though because my volume indicator was getting above halfway up when set all the way down.
Mind blown. I had been treating my input gain on my UA Volt as a more traditional gain adjustment you would find on a mixer. I've been thinking my pickups were way too hot and fighting the idea of dropping my input gain down to 0, thinking that MUST be wrong. :D love the video! Thank you!
This is a major game changer, especially so with the Neural plugins. The SLO is now my favourite. It'sway more organic, less fizzy harshness. Focusrite 2i2 3rd gen, input gain at zero. HUGE difference in quality. It is however a problem for the Helix Native...it cannot compete with Neural really for the gainier stuff now.. Huge thanks John, this was a proper eye opener. I always set the gain to just before clipping, thinking it made sense. How wrong I was.
Whenever I see someone post anything about the internet ruining humanity, I’ll direct them to this video! Honestly, I never even knew that you can even play your plugin amps at 0 on my interface.
@@johnnathancordy Exactly, I mainly use EMG Pickups 😂 But even with lower output passive pickups I have never touched the input volume of my interace. I have an Audient ID14MKII and I reckoned the JFET input didn't need the extra boost. I haven't checked the line inputs, at the back though.
I have been pulling my hair out trying to figure out why everything I play thru plugins sounds overly gainy and distorted. Gonna give this a try later, if it fixes my issues you are a literal savior 😂
Well, that's just nuts. I was like "Leave the gain at 0 on my Focusrite 2i2?!!! Preposterous!", but I just tried it and it's perfectly fine. What. the. HELL. That little Gain Knob is really a Boost Pedal in disguise! This video's not so much a game-changer as it is a "game-un-sabotager". Thanks!
You confused me for a little bit with all the interface talk, but I think you should emphasize that the interface doesn’t really matter. What matters is the signal you are sending into the amp sim. As long as you aren’t clipping the interface input, you’ll have a clean signal getting digitally converted into the computer /daw you are fine. After that you can use the input on the amp sim to hit the sweet spot. If you can’t turn the input down low enough to reach the sweet spot you can throw a trim or utility plug-in infront of the amp to further control the gain going in to reach said sweet spot. Now, keeping that consistent, like you said say zero on your specific interface. Trim plug-in at -5 and and amp sim at -3 for a specific plug-in is the most important part so you can plug in any guitar at it will be consistent . At the end of the day keep the signal clean and control the input to the amp sim based on whatever that plug-in likes, which like you said can vary from sim to sim
10 years ago, I found turning the gain down to 0 on my interface always sounded better when using amp sims. However, I always thought the technique couldn't be the "right" way of recording. Once again, trusting your ears is fundamentally important with all things music production.
One thing I’ve done is use the built in pad in the helix to reduce the input and that has helped my clean tones as well if you’re going straight into the helix
Indeed I always aim for -16db to -14db at the output on average. So even lower. And using my Volt’s instrument input, gain all the way down at 0 is a perfectly fine as a clean DI signal. I’ve seen that starting-11 to -10 and higher, it may start to clip unnecessarily.
💚Man, this has been driving me insane for over a year!!! Second guessing my old gear (a id 002 rack & 2011 MacBook pro). I was literally contemplating blowing thousands on new devices…. I couldnt find my asio driver for the 002, to check the output, which is odd, but at least I made somewhat a fix in the plug ins input level. Now the next question; why is it still sounding like a hocky cocked wha???? All the presets and my own. Ive swapped everything, checked for doubling fx and all that. It just isnt as punch as everyone else online. Still, at least I got my clean back (somewhat, still muddy even with treble & presence high). These things are fun! But I hate them too!😂 Edit: I believe the digi 002 Rack is 18dbu. So if I get it correct, NDSP is 12.2, so I should probably cut the input in the ndsp by 5.8. To me it sounds less distorted by taking 8 off. AND, I just watched your buffer video. Im not completely sure, but I believe my honky cocked wha sound has to do with buffet. Low & behold, my trusty metal zone will that buffer today (turned off).😆. All I have is my ears, and they are pretty tuned. Seems like the cock wha is not so much audable, & suddenly my palm muting chugs are tight tight tight (they were loose without the buffer. Its late now, so heaphones for testing. But it seems as though I may be onto the right thing. Thanks for your videos. If theres anything else I may ave missed, please guide me along. Still confused as to why they dont include a buffer in my interface; like wtf. One would assume its somewhat plug & play with minor adjustments; not ticket science degree.
You would think the amp sim could include an optional auto input gain function where it asks you to strum hard and light and sets the input gain to the appropriate level..
The trouble is - that would bring high output humbuckers to the same level as low output single coils - which is exactly why I've ended up running into plugins way hotter than I should have been?
I can attest that the manual in STL Tonality says to get "maximum level before A/D conversion" doesn't say to reduce it afterwards. Recommends a 1MegaOhm input stage though.
just a note, if your guitar sounds weird after doing this make sure to turn off your noise gate - some of my notes started warbling and i couldn't figure it out what it was until i realized the lower input was causing the gate to chatter, so i turned the neuraldsp one off and it's all good (using ableton gate with better dialled in settings). not sure if this is a sign that i have my levels wrong and i'm too lazy to check, i just turned it all the way down on my 2i2
What I do is plug my guitar into my interface(focusrite) and play it with no plugin and check for the highest db level and then adjust the gain accordingly until it is about -14, I then add my amp sim plugin
@@Qstick333 Yes, I arm the guitar track play and in my console or track view I can see the DB level. I adjust it up or down accordingly until I get a good one
I always turn the volume knob along with evening else down while recording. I found that volume knob on my guitar @80% gives me the best results. I make sure the signal isn’t clipping then roll off the volume until I get rid of some of the high-end fizziness. I end up with a really nice tone that sounds a lot more like what I hear from a real amp
Who is the "James Santiago" of Neural DSP. I wonder if you could get an interview with him. The way James did interviews about the Lion, was an incredible resource for me in understanding how to make the best use of my UAFX pedal. I wish Neural would relate to their customers a little more, and explain how to get the best sound out of their products.
the 0dB advice was always intended for analogue tape. Digital doesn't;t need a nominal level in the same way as digital doesn't really have a noise floor (or shouldn't have if your gear is working correctly). The digital 'equivalent' of 0dB analogue is around -12dB in your DAW, otherwise this will destroy headroom when it comes to mixing. Part of my MA looked into this as I was comparing using analogue/digital console emulation against real world analogue gear and had to consider all these factors. It's so frustrating that most people aren't given the clear info about aiming for -12dB
I had a similar problem with the Quad Cortex. I knew that setting my RME Babyface Pro fs to the lowest gain setting was correct, along with using the line in (TRS) on the interface and using the line outs on the Cortex. Also, turn off any outputs you don't use on the Cortex as they are ALL connected to the output monitoring of the device. This solved my volume problem of the presets/scenes on the device. Nice and smooth now. It only took a few years to figure that out. The testing phase should of lasted a bit longer IMO.
Oh lord. Eventhough my sound was amazing before, this really dialed it in and made it all som more natural feeling. Thanks! And NDSP, u guys should add this info on the welcome screen of the plugin 😂
What an eye opener, thanks for this John, now I realise I've driving Amplitube 5 way too hard, no wonder it never quite sounded right. Just tried the theory out by setting the input level on my Focusrite Scarlett to minimum, then boosting the level to about 12 dB in AT 5 and it sounds crack on...
If you are using Scarlet, it is said in the video and in the comment by Neural DSP that you should leave Hi-z on (Inst button), gain to 0, and then set the DSP plugin input to 0. "TLDR is, if you use a UAD or (most) Focusrite interfaces, set your preamp gain to 0 and plugin input level at unity/0, and you have the perfect gain response."
Too avoid confusing things more, we should understand the different dB scales. dBu, dBv, dBFS, etc... these are all very different things. Even within a DAW dBFS (full scale) can be misleading because you can go above +0 dbFS at certain points, because most DWAs are internally running at 64 bit float point. What you are speaking of is the signal going into the plugin. This should be dBFS, and typically 0 is the highest, with all other values being negative, down to negative infinity. The signal being at -12 or -13 dBFS means you'd have 12 or 13 db of headroom. This is a good standard not just for guitars but every track you record. Most gear, analog and digital, uses this level of signal as a reference, or at least somewhat near this. Think of it like an open window and you are throwing something through it. You don't want to aim for the highest part of the opening, right? Aim for the middle just in case you accidentally overshoot. This "middle" is around -12 dBFS. For recording you are also fine aiming for -24dBFS, especially if you record at 24-bit which had a much much lower noise floor. If you can't record low enough into your interface, but you aren't clipping (over 0 dBFS), most DAWs have the ability to adjust gain prefader or pre-plugins. Simply adjusting the level at that point is okay too, or place an additional plugin to lower the volume before your amp plugin. In my templates when I start my projects I have a simple "pregain" vst plugin strapped into the first slot of my effects for every track. It's not just useful for guitar plugins, other vst plugins behave differently with different levels of signals, so it's nice to be able to control it easily like this. You can always lower the output of the track later, either with another "postgain" plugin at the end, or simply with the track's main mix fader.
I also discovered this issue based on my clean presets not being clean either. Had to dial back the input level and then testing the edge of breakup presets. Thanks for sharing.
Helix Native literally shows u not to surpass -12db on the input with the colour bar next to it. It also says its in the manual its calibrated at -12db
The colour codes are just a guide, the software has no idea what you are playing into it. You can use a HW helix to compare signal levels with as it uses the same calibration as Helix Native. If I follow their colour coding, I’d be undergained compared to using the hw Helix. If you calibrate for exactly 11.5dBu you will get identical gain levels with hx native as using the hardware.
Helix hardware HiZ input is calibrated at +11dBu=0dBFS, so if you want to have the same results in Hx Native you have to set your interface HiZ sensitivity to the same value.
Love simple advice with massive impact👊🏻 revisiting all the amp plugins again which I thought sounded like crap 😅 They sound so much better now👍 Cheers for the info no one seems to talk about !
Hey John, thanks so much for this !!! I've had this exact problem when dealing with the Neural DSP plugins. These clarifications help a ton !! One thing that wasn't quite clear: when changing guitars going into the interface why not change the input signal ....that would mean that different guitars would bring different input level seen by the plugin, which is not what we want.
But isn‘t that exactly what we want? Because when using a real amp, as has been said, the input level also varies with different guitars, pickups etc. So the amp would behave differently. And so should the plugin if it’s modeled after the real thing.
Thanks for the shout out John, you’re a total legend! I have a couple of videos on my channel showing how to set levels for various plugins, and even to measure your gear if the manufacturer doesn’t provide specs.
Just a quick correction for you, 8.5dBV is actually equivalent to 10.72dBu so the amount you need to cut (if you want an accurate response) is only 1.5dB.
The best advice is to look for the “Maximum input level” specs in the interface manual, usually it’s written in dBu with gain at 0 (i.e. the maximum amount of headroom the unit provides).
Most spec around 12dBu because it’s equivalent to approximately 3V, which is the max you’d get out of loud hunbuckers or a 9V pedal. If you set the gain to 0, it’s likely to be close, but interfaces do vary, and some can be -3dBu, while others can be +19dBu.
Neural DSP, Line 6, Nembrini and others aim for around 12dBu, so if your interface is different to that, adjust in the plugin to land in that area as John says.
oh and I pronounce it “Sock-oh-loss-key” but I’m sure Polish speaking people would say I’m saying it wrong 😂
@@eds4754 I mean I totally botched the name so I'm sorry!!
And THANKS for the intel - seems you've been really instrumental in getting that information out there at all!!
@@johnnathancordy You’re welcome! I was hoping someone with a big audience would share this information as I’m convinced almost nobody is using the intended levels, and has no idea there’s even an issue!
Behringer says that the UMC204HD has a Max. input level Mic: -4 dBu / Line: +20 dBu / Instrument: -3 dBu
If i plug directly to the audio interface, "the rule" still is to have the input LOOOW on the behringer and compensate the input of the plugin, and finally, adjust the output of the plugin to the db level i want to hear from monitoring?
The dBu thing is very confusing to me :-S
@@martiboucat -3dBu is almost no headroom. Use the pad (it’s 20dB) and you’ll be at 17dBu. Then leave input gain at 0 and boost in the plugin by 5dB
Also, for the Focusrite users, make sure you still have the "INST" setting on. With it off, it's 22 dbu at minimum gain, meanwhile, it's 12.5 with INST on
For 2 years I was using it on line... I always wondered why I couldn't get good guitar sounds lol. I only realized what was going on a month ago... fml.
@codexnecro just how? Do people not look at controls before using them
@@lechug43 I did when I bought it, and at the time, from what I read I got the impression I was supposed to use Line in my case and I set it that way and never thought about it for quite a while.
It’s not just the gain, instrument mode increases the input impedance as well. This affects the overall level and frequency response.
What if you have a Radial J48 hooked up to the Focusrite? It’s at 10 plus maximum db?
The plug-in sounds so much better at the proper levels. Thank you so much for sharing this it’s always been a question I’ve had.
It makes me wonder if the advice we have been given for input levels has been aimed at microphones rather than guitars.
it has
Offcourse, these interfaces and their makers are from a generation where plugins were only at the relative basic, from an marketing, business point of view alone, the company would have spend more budget on marketing and r&d geared towards vocals or recording tube amps with mics
Now with this in mind, I truly believe this was the case...
In fact, let's go even further... plug-ins a few years ago weren't as great as they are today thanks to the advancements in tech and sound development.
I'm sure that plug-ins were considered the same way as many libraries may have been considered for; to do mockups, demos, etc. not full professional sounding records when in reality it's more than possible.
So, the recommendation may not come from the place of ignorance or to be actual bad advice with a bad intention and more of a: you should record with a real amp cause that's how it must be done.
You can write with your plug-ins and then perfect it with your real rig.
My two cents... maybe I'm overthinking it, maybe not.
I only know that I've got to go to bed hahaha.
It would be nice if plugins would give you more visual feedback about where that sweet spot is supposed to be. Some do some don't, but Neural doesn't at all and it seems actually quite important.
The reason plugins can't provide what is that "sweetspot" is because every guitar pickup, tone knob, volume knob is different. The best advice is what he provided. Plug in the Hi-Z input and set the input gain to 0 then control the gain with your volume, tone and pickup selection on your guitar, or even a pedal board going into your input if that's what you would do on your normal amp. The idea is the amp input should be "dumb" just like when you plug in a guitar, some guitars have higher output levels than others. And a lot of guitar players at a live gig like to keep the input knob at half volume for normal playing so they can turn it up when they need to do a solo or something special without front of house engineer needing to adjust anything.
Sweet spot is relatively...
@@adam2178 Shh....that's the wrong advice.
Man thank you for this. All the horrible fizz has gone from my distortion tones.
Came for the absolutely rad playing and stay for the absolute rad playing but also top tips like this.
You can reduce the signal level AFTER you've recorded it if you're printing the clean guitar signal. The plugin won't care, as long as you haven't drastically altered the clean signal (i.e. distorted) much. Same goes the other way.
That's precisely what i'm going to do for my current song that I already started to track and don't want to re-record everything. Hope we won't notice a difference in the final mix! 😬🤞🤞
So what You mean is turn down the channel know or rather the recorded item itself?
@@aremenius no. The signal level, pre-fader.
Exactly. I think this video might be making this thing a little more complex than it actually is. Depending on DAW you just have to make sure you're setting the desired level from pre-fader and obviously make sure your guitar isn't clipping. Using pre-fader on the track is easiest cause you can use it for both live and recorded. But ofc there's many ways.
@@eliteextremophile8895 that said, keeping it as close to ideal going in is always the best move
I'm just flabbergasted, been grinding my gears for years trying to figure out why my amp sims always sound lackluster in comparison to every video. But now when all I did was turning my interface gain down to zero and same within the vst it sounds so good! Thanks a bunch!
I'm flabbergasted you wouldn't have discovered this sooner.
How do you get a sound when gain is 0?
@@sheshehangbecause your guitar is going to still send a signal. The gain knob is just for additional boost like a boost pedal. Which should be done within the plugin if you're even going to do that.
@@adam2178Im flabbergasted you felt the need to write this comment
I'm flabbergasted at you feplying just to type flabbergasted one more time
O MAN, thank you so much!!!!!!!!
Immediately improved everything.
That’s what I thought it should sound like. How did most of us not know this?
I discovered this early on by accident. When I was discovering the world of VST amp sims, I would purchase "presets" for helix native that sounded great on the RUclips demos. They ended up sounding very different and gainier on my laptop setup. I found turning my gain all the way down to zero on my Focusrite Scarlett it got me closest to their tone. Good to know it was the right call all along.
This advice also applies if you’re planning on reamping. I always record guitar to at least have the option of reamping it, and with my interface, if I leave the input gain at zero, and playback the recorded guitar at -10db to the amp, it sounds identical to when I plug into the amp directly.
Holy shit, i'm reevaluating my life right now. For the past 10 years everyone said the wrong thing on the internet about guitars and interface ? Even pro like misha back in the day said that you should always raise the input on the interface until it almost clip when recording guitars
Agree. I have been using guitar plugins for a number of years and found this lower input setting works better. I was never happy at first with the sound of plugins and then started playing with lower input settings and this also resolved other issues that bothered me like aliasing and HF abnormalities with clean sounds. I am now actually using plugins for gigging live. Use a midi foot controller for all the fx, patch changes, expression, etc..
home and car stereo guy here. I have been waiting for the guitar community to figure this out!
I bought a high power car amp from my friend and I had an old bose am4 subwoofer, so I did some math to figure out the gain settings so I wouldn't blow the subwoofer, I had it set perfectly it sounded really good. But I kept getting greedy an adding just a little gain so it was louder and louder I think after the fourth or fifth nudge it blew
Actually there's a video on the ndsp RUclips channel where they set the input on -30 to get a breakup tone. This made me realize this an important parameter to adjust.
I don't change the gain on the interface as the Noise to signal ratio still applies, I just adjust the input level by ear. For heavy tones, sometimes a high input gives a nice saturation
The biggest eye opener that I've recently discovered with some plugins and especialy with the Helix Native that some amps tend to sound a bit "wooly"is to boost some highs and cut some lows at the beggining of the signal chain.
I always felt that some amps in the Helix Native sounded too "wooly", dull and sometimes boomy and not really articulate. I've just put simple EQ block as the first in the signal chain, boosted highs quite a lot ( 10db) and now I fell in love with the Helix tweed style models even though I hated them before. 2204 started to react much better and now it has this "squishy" character that I absolutely love.
Thanks for the video!
The maximum gain without clipping advice comes from both maximizing the SNR, as the preamp has better noise characteristics at higher gain, and ensuring that the voltage levels on the line are maximized over what the analog to digital converter (ADC) can accept. If you don't do this, the signal gets quantized in a way where a number of bits don't actually get used. What I mean by this is that if you have a 24-bit recording, you maybe cutting out a few bits because your signal to the ADC is too low, and you're not getting the full dynamic range.
Maximizing gain without clipping is all well and good for capturing a signal with the highest amount of clarity and detail as you possibly can. The problem with this is that you lose reference to the ABSOLUTE signal level. What were the actual voltage levels that your pickups are putting out and at the end of your guitar cable?
That's what's getting lost here, is how do you map that absolute input signal level back to a referenced line level and how do you standardize that across interfaces and plugins? For best clarity, it would be best to use max gain without clipping and then digitally attenuate/scale the signal to the appropriate level rather than reducing the gain at the input at the preamp.
A solution would be to set maximum gain without clipping, then plug in a device that provides a reference tone with known voltage amplitude. Then, in the plugin, digitally adjust the input until the level matches the reference output.
This is all correct, however the SNR isn’t really determined by the converters in this situation as the noise picked up by the pickups is going to mask it a lot. As long as the interface used isn’t exceptionally noisy, or the user isn’t recording VERY quiet (quite hard to record a DI that’s too quiet in 2024), SNR is unlikely to be as big of a problem as not knowing the input headroom.
But yeah, the ideal is gear with stepped gain controls so you can set your headroom reliably to whatever you like (and can compensate accurately afterwards). Also possible to do with a sine wave as you mention, although it’s a bit cumbersome to calibrate every time you touch the gain control…..
@@eds4754 Agree with that!
Great playing, I just bought Neural Morgan Suites ........ UNBELIEVABLE ! Best amp sim I've ever heard !!!
One of my biggest frustrations from most of the channels that showcase VST amp sims is they never go over the fine details like this then people like me and most of the commenters are left dumbfounded. I really appreciate you showing this to everyone!
I've used plexi-type plugins for most of my life and I'm quite used to overdriving the plugin by having more gain on the input. I actually enjoy that sound too, but this is great advice for how to set it "right".
Well yeah a lot of where you put dials is subjective.
But I think it makes a lot of sense if you're using a plugin that comes with a bunch of amp models and presets to have the front end set from the interface so that it's the "same" as plugging into the real amp. Of course if you have a different guitar then the presets probably won't sound the same, but you'll be in the right ballpark at least with a similar guitar and a preset marked 'clean' might actually sound clean etc.
I think one aspect of amp modelling is the idea I could capture my amp. Then I can plug into the real amp + cab set some settings on the dials - get a tone. Now I should be able to plug into the amp model, with the same amp+cab selected, put the dials to the same settings and get an identical tone. If that happens I might decide that amp modelling has arrived, it's a joyous day and I just use my RL cab to stand beers on. Or something. Most of us at this point remember that we don't own the real life amp and cab as the beer crashed to the floor. But as we're mopping up we're pleased we don't need to buy one now anyway.
Obviously there are many reasons why when I do that test I don't hear the same tone and I think "This amp model is wrong" (presumably I don't assume that real life is wrong). If I have input level gain so high that the amp model is overdriven even on the lowest volume then my 'same settings' thing won't work even if the modelling software were a perfect recreation of my amp then I'm going to say "This amp modelling software sucks" - if I'm naive and don't realise how my setup has affected things.
Otherwise, there are no rules about where to put dials.
You solved a huge problem I had since I started using plugins. Huge respect mate and thank you.
I stumbled upon this concept a few years ago. Almost all my amp sims sounded too saturated.
My UA interface defaults to +10dB gain on the instrument input. This is as you said analogous to having a 10dB boost pedal in front of an amp. But the amp sims model the amp - not an amp plus a boost pedal.
So now I simply turn down the input levels on the amp sim plugins by about 10dB, and the amp sims sound a lot nicer - particularly on “cleaner” or “edge of breakup” tones.
I've been doing this for many months now but i'm still not sure if it's the right thing to do. My Focusrite Scarlett 2i2 Gen 4 defaults to +7dB gain with the instrument input and i've been reducing plugin input by 7, everything gets amazing compared to it sitting at 0.0 but i'm worried about the thought that i might be killing my signal, maybe? It'd awesome if you could or someone else enlighten me on this!
I think the advice of setting the preamp gain on your interface as high as you can without clipping is sound advice for recording a good quality DI signsl if you will be using the DI for reamping or swapping between plugins with the same recorded track.
That's why a lot of plugins come with an input level before the gain knob on the plugin. It's just good to know that some plugin companies actually list the suggested 'input' to the plugin.
So, ask yourself if you need your DI signal at the right level if you will use a DI for other purposes.
I record DI via my little labs box. My instinct would be to put the line in gain as high as possible for best signal to noise ratio. But then to get the optimum gain level into the VST do you think I should use a gain plugin with a meter and set it to say 12dB ?
Just to slightly clarify dBFS vs dBV for the UR-RT2: dBFS is the level relative to where hard digital clipping happens (i.e., relative to the largest number the converters and DAW can assign to a signal amplitude). dBV is the level relative to 1 VRMS. The text in the video indicates Neural is assigning 0.707 (1/sqrt(2)) VRMS to -13 dBFS. 1 VRMS (i.e., 0 dBV) is a factor of sqrt(2) larger than 0.707 VRMS, which corresponds to 20*log10(1.414) = 3.0 dBV. So for Steinberg's UR-RT2, if for the hi-Z input 0 dBFS is +8.5 dBV (2.66 VRMS), then -13 dBFS is 0.595 VRMS, not 0.707 VRMS. For a 0.707 VRMS signal to be at -13 dBFS with the UR-RT2, you need a gain of slightly less than unity, .595/.707 = 0.84, or -1.5 dB.
Yep - Ed clarified that too - but it's still closer than me chucking a +13db boost at the start like when I follow the advice of the plugin makers!
I used to have my input gain way too high and was actually clipping with the gain on my interface at 0. Fixed that. Then read the Helix Native manual and realized I still had too much gain. Made a world of difference in most cases. In some cases, that crazy amount of gain actually gave me kind good tones.
Thank you both guys.........as I'm approaching to this world I realize YOU are the most prepared and sensitive person on youtube...........I LOVE your style and trend
That intro was absolutely beautiful John! And the content really helpful!
This whole discussion has been a revelation as I had been very disappointed with my plugin (Scuffham S-gear) for all the reasons mentioned. Tried setting interface to min and immediate improvement. Only downside was minimal recorded track which did not seem good as it doesn’t use the full digital bandwidth - you would never record your modeller at that low a level. I am now setting the interface to give a good recorded signal and adding a gain stage as the first in the channel to pull signal levels going into S-gear down to where they would be with interface at zero. Any minor adjustments can now be made on the gain block or the S-gear input trimThis seems to work well and I’m now much happier with S-gear and can see where all the good reviews come from. Really grateful to all who have contributed to this - now off to wallow in lots of clean sounds.
This makes so much sense now that you say it. Applying the same gain staging methods for mics to guitars being fed into virtual rigs…of course it would be different. But I don’t know that I would’ve ever realized it. Thanks!
I was today years old when I learned this. I also watched Rebea’s video about this. I have been cranking the gain on my interface when using an amp model and it’s always sounded terrible. Yesterday I stumbled on Rebea’s video and today yours. Now my tone sounds so much better.
PS. Neural Amps are amazing amp models. Best I’ve heard.
I kind of did this already. I turned my Steinberg MKII barely up, because putting it at zero made it feel odd. Then i reduced the input gain to level it with other recordings i have. If you want to find the sweetspot record a DI signal that's pretty quiet. Then you can loop it in the DAW and turn up the input gain until it sounds about right.
Then you can either copy the Plugin Value or turn down the interface.
Looked up the value for my interface: Arturia Minifuse 2. It's maximum input level when using the instrument input is +11.5dBu. Almost right on the money. Also checked on the Minifuse 1 and 4, they have the same value.
Setting the interface input at 0 has always made sense to me, perhaps because I’m an amp user mainly and only dabble with modelling.
My first modellers were line 6 pod , no way to adjust input levels… move on years later and I got a Boss GT1000 and couldn’t understand why you were supposed to alter the input gain to match guitars. So my Strat needed more input level than my Les Paul , but it made it sound like a boosted Strat.
Jason Sadites goes into this in great depth with Tonex and the effect on captures, once you get it right it really opens up the pedal.
The problem with line6 "no way to adjust input levels" doesn't actually tell you where to set the input level on a plugin - and it isn't helped that they don't give any specifications for the max input level dbu for their hardware - and word is they specifically omit this information because they're scared people will think their hardware sucks. But that actually means they suck anyway. Because now we're sat using a podxt as an audio interface scratching our heads - or having to buy equipment to measure it ourselves.
So, sure, you couldn't set the input gain, which means you couldn't set it too high. But that doesn't mean you're not going too hot or too low into, say, helix native, tonocracy or neural dsp plugins because each of these were created with a specific dbu value in mind (in fact for some software even different amp models would be expecting different values - although tonocracy fixes this - if you set the input gain "right" then all of the amp models, tonesnaps, presets etc should be "right" - I put right in quotes because, sure, this is subjective there is no right per se if you want to turn the input gain "too high" to create a specific noise then you can)
If you have a focusrite interface it turns out that setting gain on zero is all you have to do for some of these plugins, but if you have a pod or any other interface you can't assume the same thing - and, in fact we know that for many of the interfaces you have to turn the input gain up or down to match as well as setting gain on zero. Specifically you can't even assume the lin6 hardware is giving the right input for their own helix software. When you consider their amp modelling is infamous for creating fizzy tones and for people struggling to get a good tone out of both their hardware and software perhaps this partly explains why.
All we learn noting that line6 lack an input gain adjustment is that they probably really were clueless about what they were doing. Note that most of the manufacturers of hardware interfaces and plugins gave the wrong advice about how to set input gain. So don't assume they know what they're doing. Now they're creating videos saying the opposite and acting like they knew all along of course. The joke are the videos where people are saying "I never realised you were all setting your gain?!?!" as though the end user was getting it wrong - they were following the instructions given to them.
I've seen so many people crank the input levels on their interfaces and complain that plug-ins sound awful like yeah because you're clipping the hell out of your signal. The first thing that I did when I started using Helix products, about 4 years ago, was use the rack unit as an interface and see where the DI channel was metering in my DAW and use that as a baseline for all of my other interfaces. I use a Behringer UMC 18i20 in my studio setup and a UMC 204HD as my mobile interface and the difference in base levels for each is shocking considering they use the same hardware and preamps so make sure that you check the levels of each unit if you are switching between interfaces.
Wow! Thanks a bunch. No wonder, why my guitar tracks sound muddy. They were pretty oversaturrated.
To be honest, for people that just want to jam this is good knowledge, but it might be a bit complex to understand. It's very simple really. Make sure your signal from guitar to interface isn't clipping and set the desired level (-13 or so db) from the channel's pre-fader in DAW. It has the exact same result as trying to find the proper gain from the interface and you don't have to go measure analog outputs. Most of all interfaces have almost exactly as clean opamps and AD circuitry so there's only minuscule differences in the sound your interface is sending to the DAW at any given reasonable gain.
I have definitely been overdriving guitar rig pro, in fact I abandoned it because I didn't understand why it had zero correspondence with a lighter touch/lighter picking. The Marshall style amp that I wanted to used was always face meltingly aggressive. Now I know what to do thank you.
Wow. I can't believe there is not a LOT more people talking about this, I was 100% for sure doing this and for a minute of there I was thinking there is a literal problem either in my software or maybe in one of my pickupups, and I'm playing a Strandberg metal NX8 which I know for a fact is in good shape being the second one I've had, so I was getting pretty confused. I did this and it literally fixed everything I was trying to get a good tone for selective picking and thumping, Animals As Leaders type stuff, and anytime I would switch to a single coil and sound on my guitar I would get so much background noise that you couldn't hear what I was playing it was getting ridiculous. Honestly Neural should be shouting this from the rooftops so people don't start thinking their stuff sucks juts because it's not being used properly, when it turns out "properly" in this case is definitely considered improper in almost every other case and goes against traditional knowledge. But I get why they designed it that way, for the consistency in response between different guitars & interfaces with different outputs, it's a way of making sure it always acts the same. You'd think they'd make it higher though, one of my interfaces can't even go that low. luckily the big one can. THANK YOU for this video!
I'd always thought that I needed a new guitar, because mine sounded muddy, darker and with some dirty frecuencies in the high range. When I set mi Focusrite Scarlett to 0 in gain and -1.5 in the input of the plugin I realize that I could save more than $1.500. I have a cheaper guitar, a G&L made in Indonesia, but now it sounds like a pro.
Thanks for sharing this crucial information.
Thanks for addressing this. I have been confused about this before myself and have come to my own conclusion.
This is why it's helpful, in general, to learn to dial in sounds by ear and listen to what your gear/plugins are doing. It's all a part of learning to gain stage.
Sometimes audio interface preamps sound better turned up a bit. So then reduce the input directly in the plugin or with a gain control plugin if needed.
Think about what you are looking to get out of the gear -> gain stage accordingly. My real amp, on clean, sounds better when pushed a bit but would sound bad if I pushed it too far.
With plugins it's no different. I've used the interface input gain knob as a clean boost before when I want a little more juice out of a sim. Sounds bad if pushed to far.
Everything is gain stages right down to the fingers, strings and pickups. I get that it's tricky, but learning this will help people get so much more out of the gear/plugins they already have.
I have another thing to chip in for this discussion.
If you are using a wireless unit(trying to be lazy and not plug cables around your office), like a Lekato ws-80, you do not even need the hi-Z input engaged and probably you should also reduce your input level on the plugin. I learned this the hard way(not understanding why presets where more trembly and loud with the wireless - and in no way similar to the ones with the cable.).
Also if you double click a slider on the native it defaults back to the starting value (easier to set input gain back to zero this way)
It'd be great if someone from Line 6 actually says what's the signal level they use to calibrate their Helix series so we can perfectly match our presets on Helix Native and Helix hardware because it might not be the same as the levels Neural DSP use, I actually commented this issue several times on your videos but never got a response for you as I was always asking you which exact input levels you were using on your interface to match your Helix hardware, I have an iConnectivity Audio4C and found that I actually have to put my instrument inputs on 18dB and Hi-Z mode to match the HX Stomp, on my other interface, a Behringer UMC202HD settings levels at 0dB and HI-Z is good enough
It’s 11.5dBu. You can measure the HW, and compare to Helix Native. This is actually how people discovered Helix’s value.
@@eds4754 Awesome!, thanks! I just subscribed to your channel too! I'm glad I finally found this answer I was looking for and strangely no famous youtuber guitarist mentioned or covered before, BTW, do you know if on iOS they use similar input levels?, I use some Nembrini amps, THU and Aurora DSP apps and it's also hard to match levels on their presents on different platforms
That’s kind of funny because on my focusrite 3rd gen I have always kept the gain knob on zero with “inst” engaged and thought I was doing it wrong but I just liked how everything sounded a lot better. As soon as I would try to set the gain at “recommended” as per your initial statement it always sounded flubby and washed out. So I guess I’m not as stupid as I thought 😂😂
Got some Marco Sfogli vibes in that intro, chef's kiss.
This was eye opening. I have always had problems getting helix native and s gear to clean up.
Thanks for the great work as always!
I go -18 on mine
So stoked to get some solid info on this! This confirmed what I thought I was hearing and the best place to start to get the best sound of of DSP. Thanks, bud!
This is exactly what happened to me when I bought Imperial Tone King plug in to use with a Focusrite Scarlet. I couldn't get a clean sound to save myself. I even tried to return it and they wouldn't. Then I worked it out and started loving it. (Mind you, they didn't explain this to me, I had to work it out myself.)
Excellent Advice. I noticed that when I compared my preset pack made on the hx stomp to the sound of helix native. Way too much gain in native! This video should be watched by everyone!
Yes indeed, now the clean sounds clean. Always wondered why they broke up so easily before. Getting much better tone all around with Neural plugins. Thanks!
I wonder whether the artists creating the presets in Neural plugins are following this method or not, because that's going to affect how the presets sound. Maybe some are and some aren't. This seems like a pretty vital thing to have not been made clear.
What you record in the DAW is just the raw input signal, it will be amplified by the plugin. So when you print it out, it's gonna be a thick waveform.
@@phanhuybach92 Ah yeah that makes sense, thank you
That’s one of the most useful piece of advice I got from RUclips guitar gearwise! Thanks for that! 🙏
I've never had a problem with near clipping. As long as it doesn't clip, that's all that matters. Some plugs (TH-U) require a near clip level for accuracy (TH-U also has a humbucker and single coil sensitivity to ballpark a match). There's plugins that have a strummed autogain that will turn UP even further than near clipped in the plugin. Some will turn the level down. Just learn what the plugs want or if they don't say... Adjust to taste. You will never know if they are truly accurate anyway and who cares as long as it sounds slammin
Good points.
Hey John, I have a little bit more insight on why this is. Modern amp sims that do profiling/accurate modeling and "amplify" the raw DI signal as much as 3-4x. With some Neural amp modeler profiles, my raw DI was set as low as -36dbfs to get a good tone. Older amp sims were just clipping and adding some compression so you would want as much signal as possible without clipping. Newer amp sims actually simulate the amplification so my Audio Interface signal was set as low as -40dbfs for high gain amp profiles and to as high as -20dbfs for cleaner amps and the output signal sat perfectly in the -18dbfs to -6dbfs range on my DAW for tracking. And the volume knob on my guitar at full would give me the perfect/slightly more than the perfect amount of gain for lead tones and at halfway the perfect crunch/edge of breakup tone.
I TESTED THIS - IT’S RIGHT!! ⭐️⭐️
Nothing more to say!
I'm so happy I stumbled upon your video, always had this horrible fizz in my distorted tones and some of my guitars wouldn't get a clean tone ever. That's gone now with more clarity, needed to adjust the amp settings for the loss of gain but man, this is great. Thanks for uncovering this.
Is there a reason why this should be done already at the audio interface? Why not just turn down the input gain on the plugin?
Wouldn't you then get all the benefits of a healthy, strong recording level (the waveform would be more readable visually as well) and also the clean-sounding end result. Is this just a workflow preference thing? Or have I entirely misunderstood something?
Someone else mentioned this. Getting a good clean DI for any/all uses, and then using the input knob to feed the plugin properly.
@@seanedwards94 Yeah, but what’s the problem with a gained-up DI, as long as it doesn’t clip? Wouldn’t that still be good and clean? Or is there some specific benefit?
@@ShreddingDragon I'm agreeing with your previous statement, and saying there's others that corroborate what we're saying. Nothings wrong with a strong, healthy, DI in your daw. You just trim it down going into the plugin
@@seanedwards94 Ah, gotcha. Sorry about the confusion, my bad!
If you set the gain to your interface as high as it will go before clipping, and then zoom in on the actual waveform of a recording, what you will see is that the audio will be full of little micro-clipping. You can't see it unless you zoom in really closely, but it's actually much safer to always have your gain much lower than you think is necessary at the source. If you raise the gain later that is fine, but you won't have that head room unless the gain upstream in the signal chain leaves that space.
Movie people have known about this for years when they do sound.
This is gold! I was 100% doing this per the old advice, using a DI and XLR into my Scarlett 2i2 and boosting for every different guitar/bass till it clipped and then backing it off. What's funny is that was varying between -18 and -12. Well, I just checked the documentation for my Scarlett 2i2 3rd gen and plugging in direct with the INST button on and the input volume down and it was a pretty darn smooth -12 with less volume variation. I agree that it is better on the way into the Neural Plugins and now I know how to check for others!! Thank you!!
Aiming for -12 to - 20 for the input signal is a well known thing, even before the digital age. This allow the necessary headroom for peaks.
There is another limitation I discovered the hard way on my Behringer Uphoria that may not be present on other brand.
There is a saturation that happen even before the clip led turn on. It seems internal, so I have to lower even more the input.
To test that simply, set the level as you would do normally then listen the interface without any plug-in and strike the strings. If you here a saturation, lower the analog input volume is the interface, compensate the output volume in the DAW. If at almost zero input you still here the saturation it's something else, but if at some point you hear a clear sound it means you had an internal unknown saturation.
This is very VERY different than what I've been doing and explains a bit about why I never really got a fully clean tone on the Tone King from Neural DSP. Going to experiment with lower gain levels. Thanks
Same here! I got that cause of the reviews and was exited to maybe finally have an amp sim that would actually sound good. But thought I got a blues amp, cause I couldn’t get a nice clean to save my life
I started using digital VU meter plugins on every DAW channel before and AFTER vst plugins. IDK if it's proper but it has corrected my gain staging immensely.
I realised this when I bought native and noticed my presets sounded much gainer, for awhile I was manually pulling down the gain on my protools tracks until it sounded right then learned to turn the gain to zero on my scarlett and use the instrument input. Now I record a DI signal simultaneously via the stomps output 5 and that takes care of it and sounds identical to the hardware.
Yep. basically the same experience here.
Yeah, makes total sense. Also, I have a separate track presets for cleans and highgain tones (any amp sim), and the highgain preset has 3 disabled pedals before the amp, so I picked the one that sounded best for an amp sim, no matter which one - NeuralDSP, Plugin Alliance or Softube or some other amp.
And in the plugin I usually have to ADD somewhat around 6.0-6.5 dB of gain.
I think, it's better to utilize metering plugins as well, since most of the amp sims don't show the actual level before going into the amp.
Thank you, I was thinking of redoing my clean tone presets, and now I think this is a great place to start - gain staging.
Time to go re-record 4 years of songwriting
in my experience, any audio plugin is like this, kinda 'tuned' to receive -13dB~-11dB. it works with any sound source...
this advice to set the gain as loud as you can without clipping goes back to the beginnings of digital audio, when the noise floor was high and the bit depth was lower
I noticed back when I started using helix native I was clipping and it didn't sound right. I have a Behringer UMC404HD interface and found that setting it to instrument and adding the pad button which I believe is a 10db attenuation and gain about 1/4 way up matches what helix does. The Helix native plugin also has an input slider on the far left of the signal chain.
I found it more acceptable without the PAD button and gain level at Cero i tried it with the PAD button sounded duller to my ears
Ok, so tried this... When people in demo videos would say, "This Neural DSP plugin reacts like a real amp when you roll off the volume knob and actually starts to clean up...", I could never actually experience that myself. It does that thing now! I have a Scarlett 2i2 set to "INST" mode with the gain knob ALL THE WAY DOWN... now. I did notice within the app I typically have to boost the input a little bit more than before though because my volume indicator was getting above halfway up when set all the way down.
Mind blown. I had been treating my input gain on my UA Volt as a more traditional gain adjustment you would find on a mixer. I've been thinking my pickups were way too hot and fighting the idea of dropping my input gain down to 0, thinking that MUST be wrong. :D love the video! Thank you!
This is a major game changer, especially so with the Neural plugins. The SLO is now my favourite. It'sway more organic, less fizzy harshness. Focusrite 2i2 3rd gen, input gain at zero. HUGE difference in quality.
It is however a problem for the Helix Native...it cannot compete with Neural really for the gainier stuff now..
Huge thanks John, this was a proper eye opener. I always set the gain to just before clipping, thinking it made sense. How wrong I was.
When you say “input gain at zero” what exactly does that mean?
Whenever I see someone post anything about the internet ruining humanity, I’ll direct them to this video! Honestly, I never even knew that you can even play your plugin amps at 0 on my interface.
Today is literally the first time I heard about this being an issue. I have always set my interface to zero, as I thought it was logic to do so.
That's good - I imagine (from your username?) that you're using some higher output pickups in general? Which would totally make it logical yeh!
@@johnnathancordy Exactly, I mainly use EMG Pickups 😂 But even with lower output passive pickups I have never touched the input volume of my interace. I have an Audient ID14MKII and I reckoned the JFET input didn't need the extra boost. I haven't checked the line inputs, at the back though.
I use a Behringer UMC22 and the instrument level is +2dBu and i still can't figure out how much i need to set my input level on plugin. Please help.
I've been doing this for years apparently. I can't wait to test this out. I could never get a decent clean sound!
I have been pulling my hair out trying to figure out why everything I play thru plugins sounds overly gainy and distorted. Gonna give this a try later, if it fixes my issues you are a literal savior 😂
Well, that's just nuts. I was like "Leave the gain at 0 on my Focusrite 2i2?!!! Preposterous!", but I just tried it and it's perfectly fine. What. the. HELL. That little Gain Knob is really a Boost Pedal in disguise!
This video's not so much a game-changer as it is a "game-un-sabotager". Thanks!
You confused me for a little bit with all the interface talk, but I think you should emphasize that the interface doesn’t really matter. What matters is the signal you are sending into the amp sim. As long as you aren’t clipping the interface input, you’ll have a clean signal getting digitally converted into the computer /daw you are fine. After that you can use the input on the amp sim to hit the sweet spot. If you can’t turn the input down low enough to reach the sweet spot you can throw a trim or utility plug-in infront of the amp to further control the gain going in to reach said sweet spot. Now, keeping that consistent, like you said say zero on your specific interface. Trim plug-in at -5 and and amp sim at -3 for a specific plug-in is the most important part so you can plug in any guitar at it will be consistent . At the end of the day keep the signal clean and control the input to the amp sim based on whatever that plug-in likes, which like you said can vary from sim to sim
10 years ago, I found turning the gain down to 0 on my interface always sounded better when using amp sims. However, I always thought the technique couldn't be the "right" way of recording. Once again, trusting your ears is fundamentally important with all things music production.
One thing I’ve done is use the built in pad in the helix to reduce the input and that has helped my clean tones as well if you’re going straight into the helix
Indeed I always aim for -16db to -14db at the output on average. So even lower. And using my Volt’s instrument input, gain all the way down at 0 is a perfectly fine as a clean DI signal. I’ve seen that starting-11 to -10 and higher, it may start to clip unnecessarily.
what do you mean with output, the signal you see coming in in uad console? i see it at -25dbfs now, so thats good?
💚Man, this has been driving me insane for over a year!!! Second guessing my old gear (a id 002 rack & 2011 MacBook pro). I was literally contemplating blowing thousands on new devices…. I couldnt find my asio driver for the 002, to check the output, which is odd, but at least I made somewhat a fix in the plug ins input level. Now the next question; why is it still sounding like a hocky cocked wha???? All the presets and my own. Ive swapped everything, checked for doubling fx and all that. It just isnt as punch as everyone else online. Still, at least I got my clean back (somewhat, still muddy even with treble & presence high). These things are fun! But I hate them too!😂
Edit: I believe the digi 002 Rack is 18dbu. So if I get it correct, NDSP is 12.2, so I should probably cut the input in the ndsp by 5.8. To me it sounds less distorted by taking 8 off.
AND, I just watched your buffer video. Im not completely sure, but I believe my honky cocked wha sound has to do with buffet. Low & behold, my trusty metal zone will that buffer today (turned off).😆. All I have is my ears, and they are pretty tuned. Seems like the cock wha is not so much audable, & suddenly my palm muting chugs are tight tight tight (they were loose without the buffer. Its late now, so heaphones for testing. But it seems as though I may be onto the right thing. Thanks for your videos. If theres anything else I may ave missed, please guide me along. Still confused as to why they dont include a buffer in my interface; like wtf. One would assume its somewhat plug & play with minor adjustments; not ticket science degree.
You would think the amp sim could include an optional auto input gain function where it asks you to strum hard and light and sets the input gain to the appropriate level..
The trouble is - that would bring high output humbuckers to the same level as low output single coils - which is exactly why I've ended up running into plugins way hotter than I should have been?
I can attest that the manual in STL Tonality says to get "maximum level before A/D conversion" doesn't say to reduce it afterwards. Recommends a 1MegaOhm input stage though.
just a note, if your guitar sounds weird after doing this make sure to turn off your noise gate - some of my notes started warbling and i couldn't figure it out what it was until i realized the lower input was causing the gate to chatter, so i turned the neuraldsp one off and it's all good (using ableton gate with better dialled in settings). not sure if this is a sign that i have my levels wrong and i'm too lazy to check, i just turned it all the way down on my 2i2
Yes! Thank you! I hear so many guitar channels that are clipping the input stage. All I hear is frying ozone.
What I do is plug my guitar into my interface(focusrite) and play it with no plugin and check for the highest db level and then adjust the gain accordingly until it is about -14, I then add my amp sim plugin
So easy. Why didn't I think of that!? Thanks for the tip!
@@Pab7osanYou are welcome
Where do you check the level, on your DAW?
@@Qstick333 Yes, I arm the guitar track play and in my console or track view I can see the DB level. I adjust it up or down accordingly until I get a good one
But that just sets the level according to one guitar. Other guitars will be different, and you won't know which one is the "true” one"?
I hoped this was the reason why I was the only person on Earth who couldn't fully enjoy Neural DSP mid/hi-gain tones, but the mistery still continues
Makes sense. Otherwise your interface is acting like a boost pedal in front of your amp.
OMG Amplitube 5 just came alive, sounds, dynamic it just got 1000% better i can't believe it. Kudos to this Jonny guy!
I always turn the volume knob along with evening else down while recording. I found that volume knob on my guitar @80% gives me the best results. I make sure the signal isn’t clipping then roll off the volume until I get rid of some of the high-end fizziness. I end up with a really nice tone that sounds a lot more like what I hear from a real amp
Who is the "James Santiago" of Neural DSP. I wonder if you could get an interview with him. The way James did interviews about the Lion, was an incredible resource for me in understanding how to make the best use of my UAFX pedal. I wish Neural would relate to their customers a little more, and explain how to get the best sound out of their products.
Exactly. Not only Neural but most of the plugin makers don't mention this
the 0dB advice was always intended for analogue tape. Digital doesn't;t need a nominal level in the same way as digital doesn't really have a noise floor (or shouldn't have if your gear is working correctly). The digital 'equivalent' of 0dB analogue is around -12dB in your DAW, otherwise this will destroy headroom when it comes to mixing. Part of my MA looked into this as I was comparing using analogue/digital console emulation against real world analogue gear and had to consider all these factors. It's so frustrating that most people aren't given the clear info about aiming for -12dB
I had a similar problem with the Quad Cortex. I knew that setting my RME Babyface Pro fs to the lowest gain setting was correct, along with using the line in (TRS) on the interface and using the line outs on the Cortex. Also, turn off any outputs you don't use on the Cortex as they are ALL connected to the output monitoring of the device. This solved my volume problem of the presets/scenes on the device. Nice and smooth now.
It only took a few years to figure that out. The testing phase should of lasted a bit longer IMO.
Oh lord. Eventhough my sound was amazing before, this really dialed it in and made it all som more natural feeling. Thanks! And NDSP, u guys should add this info on the welcome screen of the plugin 😂
What an eye opener, thanks for this John, now I realise I've driving Amplitube 5 way too hard, no wonder it never quite sounded right. Just tried the theory out by setting the input level on my Focusrite Scarlett to minimum, then boosting the level to about 12 dB in AT 5 and it sounds crack on...
If you are using Scarlet, it is said in the video and in the comment by Neural DSP that you should leave Hi-z on (Inst button), gain to 0, and then set the DSP plugin input to 0.
"TLDR is, if you use a UAD or (most) Focusrite interfaces, set your preamp gain to 0 and plugin input level at unity/0, and you have the perfect gain response."
Too avoid confusing things more, we should understand the different dB scales. dBu, dBv, dBFS, etc... these are all very different things. Even within a DAW dBFS (full scale) can be misleading because you can go above +0 dbFS at certain points, because most DWAs are internally running at 64 bit float point.
What you are speaking of is the signal going into the plugin. This should be dBFS, and typically 0 is the highest, with all other values being negative, down to negative infinity. The signal being at -12 or -13 dBFS means you'd have 12 or 13 db of headroom. This is a good standard not just for guitars but every track you record. Most gear, analog and digital, uses this level of signal as a reference, or at least somewhat near this. Think of it like an open window and you are throwing something through it. You don't want to aim for the highest part of the opening, right? Aim for the middle just in case you accidentally overshoot. This "middle" is around -12 dBFS. For recording you are also fine aiming for -24dBFS, especially if you record at 24-bit which had a much much lower noise floor.
If you can't record low enough into your interface, but you aren't clipping (over 0 dBFS), most DAWs have the ability to adjust gain prefader or pre-plugins. Simply adjusting the level at that point is okay too, or place an additional plugin to lower the volume before your amp plugin.
In my templates when I start my projects I have a simple "pregain" vst plugin strapped into the first slot of my effects for every track. It's not just useful for guitar plugins, other vst plugins behave differently with different levels of signals, so it's nice to be able to control it easily like this. You can always lower the output of the track later, either with another "postgain" plugin at the end, or simply with the track's main mix fader.
A much needed video. This makes so much sense now.
I also discovered this issue based on my clean presets not being clean either. Had to dial back the input level and then testing the edge of breakup presets. Thanks for sharing.
Helix Native literally shows u not to surpass -12db on the input with the colour bar next to it. It also says its in the manual its calibrated at -12db
The colour codes are just a guide, the software has no idea what you are playing into it. You can use a HW helix to compare signal levels with as it uses the same calibration as Helix Native. If I follow their colour coding, I’d be undergained compared to using the hw Helix. If you calibrate for exactly 11.5dBu you will get identical gain levels with hx native as using the hardware.
The top of the green colour is -12db so yeah, close enough. @@eds4754
Helix hardware HiZ input is calibrated at +11dBu=0dBFS, so if you want to have the same results in Hx Native you have to set your interface HiZ sensitivity to the same value.
Glad to hear I've been doing it right for the past 20 years :)
Love simple advice with massive impact👊🏻 revisiting all the amp plugins again which I thought sounded like crap 😅 They sound so much better now👍 Cheers for the info no one seems to talk about !
Yea my amp sims suddenly have that tube warmth now 😂
Hey John, thanks so much for this !!! I've had this exact problem when dealing with the Neural DSP plugins. These clarifications help a ton !! One thing that wasn't quite clear: when changing guitars going into the interface why not change the input signal ....that would mean that different guitars would bring different input level seen by the plugin, which is not what we want.
But isn‘t that exactly what we want? Because when using a real amp, as has been said, the input level also varies with different guitars, pickups etc. So the amp would behave differently. And so should the plugin if it’s modeled after the real thing.