วีดีโอ

Yay science! Great video. Straight to the point. Clear demonstrations.

I'm just curious why all those youtubers how they she'll these products but don't explain what you just did, it took you 5 minutes and that's what rubs me the wrong way thanks I think I'll be watching you,

This was awesome. I always knew there was a change from what I was recording into the interface vs what the input signal in the plugin was outputting. Threshold is everything, there is a threshold for all signal input to output. Just like a compressor, you have to have enough input gain to trigger the circuit before any compression actually happens. Your example showing the noise floor with 0db vs a matched "-6" input on the interface should be a glaring point of fact. For example, you have a Marshall head, you're at less than half gain on the amp. You turn that gain all the way up and now you hear all the fiz and fuzz from the electrical signal going in and out of the cab. You're boosting the floor and killing the dynamics of a performance, then making the mix engineers job a nightmare.

Ahhhh, this is very interesting because I used to use a Behringer interface and now I use a UA Volt interface both with Neural DSP and the UA Volt sounds a lot better. Little did I know the UAD just does a better job of matching the correct gain level. I still just leave everything at zero. I'm sure the extra .3 db is pretty negligible within the context of a mix.

no

are you Icelandic ?

I really appreciate this video. You provide clear logic and language to what I have long suspected. Excellent work!

WOW.

Can somebody help me? I tried appliying this concept to a Uad Volt 1 and Neural DSP plugins, the result was way too silent and clean. I had the plugin gain at 0.3, then added 27 db gain on my Interface, and subtracted that from the plugin gain. Well, almost. The Plugin only goes to -24. And at that point, even metal amplifiers sound crystal clean.

Nope you're also wrong. We have more than two gain stages to worry about. Whatever the source is, it probably has variable loudness of some form. And it has its own dynamic range. And anything else in the signal path does too. If maximizing dynamic range without adding appreciable distortion is the goal, then this is a multivariate equation to solve, and you can order the solution algorithm by most constrained first for instance. So the rules of thumb, including yours, are all suboptimal oversimplifications. It may well be that the converter these days has so much dynamic range (and consumer converters typically output fixed point numbers, not floating point "between 0 and 1", not sure how you've made it as "a designer of DSP circuits for 15 years" sorry buddy) that not using its gain and forcing yourself to maximize dynamic range earlier in the signal path is exactly good advice for the people it was directed toward. I can't answer for anyone remotely because this is a complex optimization problem and that's before we consider the desirability of various points of nonlinearity. Or playing or setup comfort. Or the value of noise itself in the lo-fi sense. Try again.

ty for this video 😌

haven't seen TH-cam videos of this autistic high-quality in many years. Just commenting for algo boost. Bless

the amount of blatantly incorrect information posted as fact with no responses to correct them on reddit is astonishing

Really appreciate the video. I've been using the "wrong" approach for a while, and have definitely been noticing noise being a huge issue. I just assumed it was "I have a lot of gain on the amp, can't be helped"... but seems like it can be helped. What prompted me to try the "set it to zero" approach to begin with was honestly just feeling like my guitars with hotter pickups were hitting the amp sim way too hard, leaving me with a messy, difficult to control high gain tone. So, dropping the input gain on my interface definitely fixed that part (leaving the plug-in's input gain alone), but it makes total sense why the signal to noise ratio went to shit. What would honestly really help is if plugin vendors actually had proper documentation on how to set this stuff up. Believe me, I looked. Your tutorial at the end of the video is dead simple, makes total sense, but apparently NeuralDSP couldn't be bothered to write a paragraph or two about how to set your interface gain and how to adjust the plugin's input gain for best results.

Is this principle true also for recording vocals or real amp with mic? Cause it happens to hear ppl saying if you sing a soft part get closer to the mic and turn up the gain on your audio interface. If you sing a loud one do the opposite

Great Video! Nice princess bride reference.

Learned a lot. Thanks man

Ha ha.... i was conflating... thanks man... makes sense

I first reduced 6db on the software input level, according to the input gain spreadsheet. Then I put some gain in my apollo input (+24db), but when I wanted to reduce this same number to compensate in my software input level, I find I can only go down to -20db (and I should be around -30db if I take care of the spreadsheet and the gain added in my interface)

Thanks for finally clarifying this. Really helped alot. Another problem with setting the input on your interface to zero is that the audio waveforms of the recorded signal is close to flat, making it more difficult to see peaks and troughs when you wanna edit the track.

Correct, it‘s like the discussion about tonewood on electric guitars

your teeth are stunning

10/10 video. No notes.

Appreciate this information. I would propose amp sim devs add the basic process and information to their user manuals as well as include either auto-gain function or a “green light” for the setting the input gain knob in their plugin.

Dude, you really deserve if not a gold then a bronze monument for sure.. no irony.. I listened to advice for years, shoveled through a bunch of printed material to now find out and hear, see such an obvious thing. Thank the gods and engineers!

Nice information but I have some problem with the premise that everything and everybody on this topic is wrong. Because even you actually give some merit to the work that has been done and see how it makes sense. This whole thing was just actually a step towards the "ultimate" method you present here as you even use the spreadsheet with the measurements made by Ed S (if I'm not mistaken). And I wish you would have given more credit to it, because the idea that amp sims need to receive a certain input to be optimal isn't wrong, all the maths about said input wasn't wrong, the only thing wrong about it was the method to get said input, and there are very reasonable explanations as to why it was presented that way. It was just the easiest way to standardize thing to a wide variety of ppl because most interface don't have any value on the knobs so you don't have immediate access to the boost value and the knob course can vary from one to another. But 0 will always be the minimal gain value for any interface so it was just far simpler to use this consistency to present a method that would work for everybody and still yield way better results than having absolutely no idea what we're doing. So the problem isn't people trying to understand how things work with little knowledge they have, but actually companies who are quick to sell us a 200$ new shiny amp sim but don't give clear information on how they work and are meant to be handled (unless they do and nobody read the manual in which case, my bad). It kinda rubs me the wrong way that not a word is said about that but you were quick to jump on ppl who did the research, youtubers and redditors who saw the value in it, and call everybody out when all you did was slightly improve that method (which is still a needed step, don't get me wrong, you are right and that info needs to be out there) PS : The original method still has one advantage over yours, it can be set without the pickup input, which means the amp sim should react differently to different type of pickups like a real amp, yours kinda blurs the line by having everything coming out at the same input gain. (again not dismissing your point just keeping on with the idea you're not absolutely right while the others are absolutely wrong, it's more nuanced than that)

So, why -19.15??

Audio engineer here, this is exactly how youre supposed to do it!

That was a cool tutorial and explanation... I may need to watch it a few times to fully take it all in but the basics of gain staging makes total sense to me...

Brilliant presentation. Subscribed & liked.

Legend says that the hero nickleback sang about.

Energize!

My brain is a puddle on the floor... thanks lol

I thought this was common knowledge...

I boost my signal with a preamp and it makes the dsp plugins sound better

A result of people not reading books anymore written by specialists who have expertise, but getting their information from someone who said something somewhere...

That was a lot of talk for a super shitty guitar tone.

Scuffham S-Gear vst input varies from preset to preset 🫤

For many years Im propagating exact this approach. It is so logical and true. Finally explained, neat 😎

Of course, you give your AD the maximum input signal it can handle (with a little bit of headroom), and then work with the converted signal in the digital domain as needed (no matter which kind of digital representation or bit-depth is used in the system). As an old fart (nonpracticing EE who studied in the 80s and worked at a company who developed and produced digital audio systems in the 90s) I would've thought that this would be common sense - but it seems that nowadays many people (not only on Reddit) have no clue of how the things they use actually work … good video, even though it could've been 3 minutes long for what I needed to know … ;-)

It’s funny, I have missed this discussion, I have never researched this topic. I’ve been home-recording for about 20 years. And I have subjectively found out that maximizing the input gain to below clipping always gives the best results.... I thought this was common knowledge.

rEdDit Is BaD fOr pOlItiCs bUt HaS GrEaT aDvICe fOr tEcHniCaL PrObLemS

Makes perfect sense to me. Seems like a no brainer. I mean, you can't argue with facts like this. Thanks man. Great video!

Wait let me get this right, dumbasse.. sorry; people on the internet argued that it was smarter to reduce the noise floor by keeping the signal level low, and not increasing the signal to noise ratio? Riiiiight 😵‍💫

Saved the video and subbed, this is invaluable info here, thanks man!

I can't believe how ignorant people are of basic DSP.

This video is exactly what I love about the Internet! Surprisingly, I did everything right, just by accident or by technical instincts 😎👍

Wouldn’t it be easier to perform that exercise using a clean sound? Talking about clipping and signal distortion when using fuzzy electric guitar is really counterintuitive.

So if I have focusrite 3rd gen with neural I set the gain on my interface to 12.3 and then dial the input knob back in neural to 12.3?

Good stuff. Sometimes, I actually really like the sound of totally cranking the input gain on the interface (or the amp sim). It doesn’t work for every tone - I wouldn’t do it for cleans or generic rhythm tones, for instance. But if I’m going for that absolutely slammed high gain compressed Marshall sound, to my ear it sounds way better than a distortion pedal or something. To quote the late EVH - “If it sounds good, it is good!”