Posts by drew_fx

    Something in your setup somewhere is different.

    These devices do not change their sound based on the wind!

    You making the effort to document and share your frustrations with the Kemper is appreciated, not out of order or anything like that. You're relating anecdotal experiences. I don't get why anyone would feel the need to invalidate your opinions, and from what I've read I don't think anyone is. Even with CK, maybe there's a bit of impatience or frustration in his tone, but ultimately I'm thinking that's because he's hoping for you to have the best results possible, not because he wants to smother a dissenting viewpoint. He's let years and years and years of misinformation about his products posted on TGP go by without a response or even a peep. Anyway, it should be enough for anyone feeling defensive (?!) that there's also a ton of anecdotal evidence, many many years of examples, where musicians are unable to discern a source amp from a profile. The Quad Cortex is a polemic device. The rollout was characterized by a lot of ugly behavior among people who hadn't even tried the device. And then there were enough discrepancies between the touted capabilities and what was delivered at release to at least question the integrity of the claims. All of that aside, you getting specifically at the tech, the capture capabilities as you measure them against the Kemper, is an interesting read. I hope you don't stop.

    Man, thank you for that.

    And I totally agree with what you say about Christophe. His composure over some of the stuff said about the Kemper by actual industry competition.... admiral.

    At the end of the day.... okay... I was born in 1984 ..... so I would've been 6 years old.... but at the end of the day.... could anyone 30 years ago expect the absolute embarrassment of riches that we as guitarists have at our disposal now??? It's astonishing. Piano players don't even have it this good!!

    "In a mix" is where 99.9% of listeners are going to hear your guitars. That's why I've always put a big emphasis on dialing in tones within a mix or at the very least double tracking (which in itself makes the guitars tonally different) and not to sound good completely isolated on their own. As I've stated if the argument is a raw QC capture is more accurate than a raw Kemper one I don't think you'll get any argument out of me or anyone else. That doesn't mean the Kemper can't be just as accurate to the real thing as the QC once you make a few adjustments, that's exactly why all the options to do so are there on the Kemper. It also doesn't mean that a capture or profile even has to be 100% accurate to sound good and honestly that's a situation that doesn't effect very man people.

    I hate to make over-arching statements based on anecdotal evidence but in my experience the majority of the people who use a Kemper and probably QC users as well do not and will not make their own profiles. They use the ones provided or buy packs from 3rd parties and at that point there's is no way to tell how accurate it is to the source because you weren't there. Sure you could make the argument you could just go based on the reputation of the algorithm but there's no way to be 100% sure. For those people the process is ultimately irrelevant because they turn the browse knob and get the sound they want.

    Would you not EQ your guitars in a mix or live? The real question is why wouldn't you that's exactly what the tools are for. It's kind of like having all the ingredients to make your favorite meal but leaving a few of them out because why should you have to add them? The meal should just taste good without them right? Maybe not the best example but hopefully one that's easy to follow. To me this is illogical, you want it to be as close as possible to the reference sound but don't want to take the appropriate steps to make it happen? You're ultimately ending up at the same place with either unit it's just one is requiring less work than the other.

    Well, now we're gonna get into the philosophy of audio engineering I suppose. To me, when I EQ a guitar in a mix or live, I'm not EQ'ing the guitar to fix inherent problems in the guitar. I wouldn't track it if there were inherent problems - I'd aim to fix the problems before tracking. Not always possible of course, but that's the ideal.

    But when I'm EQ'ing in a mix or live, I'm EQ'ing to make it fit with other instruments. This is a different thing to what we're talking about here. If I was making the claim that the Kemper can never be made to sound good, or close to the real amp, then everything you've said here would absolutely oblierate my perspective, and I should run home with my tail between my legs.

    But that isn't my claim. My claim is purely one about accuracy immediately following the completion of a capture/profile. That's it. Everything that follows after that is indeed in the realms of production, engineering, and pure subjectivity.

    If it takes extra steps for the Kemper to get closer to the sound, then this should surely indicate that advances need to be made on the profiling tech, so that the Kemper can make things easier and faster and better for their users, no???

    To me, it doesn't quite make sense to say "yes, I agree with you, but you can do XYZ to get closer" - like.... the XYZ is already a foregone conclusion. Doesn't need to be said really.

    As an aside; I try not to EQ if I don't have to. I don't want to introduce phase-shifts unless I absolutely have to.


    As I've stated if the argument is a raw QC capture is more accurate than a raw Kemper one I don't think you'll get any argument out of me or anyone else.

    To be clear - that is exactly what I'm saying. Nothing more.

    Speaking completely blue-sky a minute - What if Kemper releases Profiling 2.0 in 6 months time, which further automates the process and uses more tones during it's capture process that all of a sudden the Kemper is then way closer than the QC.... would I be out of order if I then jumped on the NeuralDSP forums and started talking about how I now prefer my Kemper for profiling?? Coz that's exactly what would happen! :D

    What I'd like to see I guess would be something like this:

    - Some clarity on what to play to solve XYZ problem during refinement. Maybe even have the Kemper prompt the user - ie: play this to make your palm mutes more accurate, play this to make your lead tone more accurate, etc.

    - Add the definition control to the pages of the refinement stage, so that it's right there and ready to go. It's probably the most important parameter going and actually if I was wanting to get closer to the amp, I'd reach for that before an EQ.

    - Some sort of option where I can tell the Kemper that XYZ set of profiles are all from the same amp channel, and to treat them in similar ways. Sometimes I notice if you profile a high-gain channel at different levels of gain, the refining stage makes them all sound quite inconsistent from one another. This isn't something you would spot unless you were right there when the profiling is occurring. The QC has the same issue btw.

    - Figure out a way to improve the "noise-gate detected" issue. Sometimes a noise-gate is detected when there isn't one.

    - Some amps like the Fryette Sig:X don't profile very well. I think it's because the Sig:X has some dynamic frequency response stuff going on at the input. So I would guess that what the Kemper hears when I'm playing my guitar is different to what it hears when it's sending all of it's test tones. I think this should be investigated; get the thing on a test bench and debug what the Kemper is doing and see if anything can be done to improve the accuracy.

    - Are there any more parameters in the amp circuit that can be exposed to the user??? I know the vision for the Kemper is to make things easy and simple to use. But if there are other parameters that we can tweak that are being hidden from us... maybe that's a path forward???

    Anyway.... I plan to do a video of each unit side by side. Not as a stupid "ZOMG! THE KEMPER IS BAD!!!" type video. But just an educational video where people can hear the differences in a post-rock/post-metal kind of context. I am a fan of the Kemper, I just would like to see some improvements.

    Some other 'in use' things...

    - The lack of decent midi on the QC is starting to annoy me now. I can't switch my pedals at the same time as switching scenes. Well... I can, if they're all setup to use the same midi PC's and what not... but that's a real ballache. And the Strymon stuff uses an additional CC to control the bypass state at the time of switching... so either way.... it doesn't quite do it for me.

    - Using it for bass is a bit limiting right now. Or at least, the gear I would want isn't there. The Helix has really spoilt me when it comes to bass!

    - I had some weirdness the other night where it wouldn't connect to my WIFI router. It hasn't done it since. Touch wood it was just a random thing.

    - The expression pedal control isn't as good as the Helix or Kemper. It's a bit annoying if you want to control multiple things all at once, and get the values right. They express (heh heh) values in percentage. So you have to program it... move the pedal.... oh shit my min and max are wrong.... tweak again.... check it again.... it's not as fluid as it could be.

    - How the banks are all arranged is starting to piss me off. And I want to delete all of the stock captures and use that space for my own ones. Grrrr.

    It's one of the main reasons I never buy something based on a youtube video or review. There's just too many subtle ways to "fool" people even if by complete accident.

    If the argument is the raw QC capture is closer to the original than a raw Kemper profile right out the gate I mean I agree 100% at least based on my tests. I would imagine the QC process is automating a lot of what a Profile maker would typically do post capture on their Kemper and baking it into the capture so you don't even see it. There are pros and cons to this of course, people who don't like tweaking or want instant gratification will obviously go for the piece of gear that gives them that. I'm guilty of this, it's one of the main reasons I dislike fractal: all the tweaking.

    I just think the point should be made that with a little tweaking you can get the Kemper just as close as a QC profile granted they will still have their own flavor. That's true in even real tube amps and cabs though, you can line up 10 5150s into Mesa OS 4x12s with the same settings and mic placement and they all will sound slightly different. Heck even just take 10 SM57s and they all sound slightly different its just the nature of electronic components. A lot of things effect sound even some things most people don't even consider like: room temperature. In this example I used a simple post EQ and in a lot of professional profiles I've bought I've noticed that quite often there is an EQ in the FX block somewhere that often has a very subtle move just like the one I made. Nothing massive just maybe a 1-2 db cut or boost to a specific frequency. Probably to achieve the same thing: to get it as close to the original as possible. You're taking a different route to the same destination.

    In the context of a mix I honestly believe I couldn't pass a blind test between the two with 100% accuracy. I did a ~2 db cut on 200hz on the refined kemper and even that might have been a hair too much but a lot of that whomp will be removed in post anyway through the use of a high pass filter and multiband compression to tame palm mutes.

    I don't disagree with any of that.

    I've never contradicted anything you've said here, and I wouldn't. It's basically spot on and is totally congruent with my opinions on QC versus Kemper.

    ckemper - again, I'm not trying to shit on it. I mean, I'm a nobody anyway. So what would be the point? I'm just offering my opinion having used the QC for some time now, right next to my Kemper. I like them both. I just prefer the QC for amp tones. There's less fighting require, and I feel less insecure about the accuracy of the tone, whether I did something wrong during refinement, etc etc.

    Whilst you're reading the thread, can you release a delay and reverb pedal?? I'd buy the hell out of that!! :D:D:D:D

    I honestly don't understand the obsession with profiles/captures having to be "100% accurate". IMO neither of them are but they're pretty close. To me all I care about is that it sounds good, when I buy a reamp zone or tone junkie profile the first thing that pops into my head isn't "I wonder if this sounds exactly like the source amp" it's "damn that sounds really good let me favorite that and turn the knob here... eh don't like that one too much delete.... oh this ones nice going to favorite that". Tone is subjective anyway what I like isn't what other people will like and vice versa if you find the piece of gear that gives you a tone you love there's no reason to go chasing another 1%.

    I did listen to Drews comparison though and immediately one thing stood out to me (and I'm not saying this was intentional so please don't get angry): The KPA files were quieter. The human ear perceives louder as better and its one of the oldest tricks in the book. I loaded the QC and refined up in reaper, hard panned them and volume matched and took a tiny dip out of 200hz and honestly I can't tell a difference between them. Yes that means I had to EQ to Kemper file but that should have been done when making the profile.

    Is a raw capture more accurate out the gate? Eh probably but having to make one less move during the process isn't worth 2 grand to me. I'll just circle back to the point where I don't see why the heck it even matters. If it sounds good: it sounds good, both the kemper and QC add their own flavor to their profiles so are you a vanilla or a chocolate kind of guy?

    EDIT: For anyone interested heres my results with Drews Files. Throw some drums/bass over the volume matched one and honestly good luck trying to figure out what is what…a1g?usp=sharing

    I'm not angry at all, I value your feedback and contribution. What I would say is what you've done there is make it impossible to hear the differences by introducing a double-tracking element.

    Your perspective is that in a mix no-one would be able to hear a difference. That isn't what I have been arguing against or disagreeing with or however you want to say it. It isn't really related to my fundamental position - which is that in it's raw post-capture state, the QC gets closer to the real amp than the Kemper. At least in a way that I don't feel I need to do any work to it in order to make it better.

    I don't need to EQ the QC. I don't need to mess around with a set of refinement stages that may or may not push the tone in the wrong direction.

    I plug in. I capture. I click okay. It's a superior process in my opinion.

    Regarding the KPA files being quieter - I perceive them as roughly the same volume. That's what my ears tell me.

    I don't understand why you would say I should EQ on the Kemper when making the profile - isn't the whole point behind any kind of profiling technique, that we want to get as close as possible to the reference sound??

    What we're comparing here is how close each unit gets to the reference tone in it's rawest state. As soon as you introduce other variables, the comparison ceases to have meaning in my opinion - I've never said you can't make the Kemper sound good. That isn't my claim.

    My subjective experience is that I often have quibbles and insecurity about what I hear from the Kemper. But with the QC I don't have that. Until I load a reverb.... and then I want to go back to the Kemper 8o8o8o8o

    I'm not trying to shit on the Kemper. I'm really not. But to me, the QC is more pleasing right off the bat. And that's what I want. Because I've got tunes to write. I don't have the time to be constantly second guessing myself.

    ckemper - We had private email conversations back in 2015 about this, iirc. From my FXpansion work email address.

    I also find it strange you'd say that about refining with palm mutes. My experience is that refining by playing palm mutes actually makes the problem worse.

    Kemper tells me some of my small amps, with valve/tube rectifiers have noise gates too. I assume it’s the voltage sag it detects when hitting it hard.

    This is with a 100-watt valve amp though. Plenty of headroom, not massively cranked loud or anything. I figured it might be some sort of compander circuit at the input of the amps causing it.

    Even when you feed the Kemper a DI'd or loadbox'd signal, therefore ruling out the need for decent mics, rooms, preamps, engineering expertise, etc etc..... the profile still more often than not comes back with some differences.

    Are they significant? Sometimes yes, sometimes no.

    Are they annoying? Sometimes yes, sometimes no.

    It's up to each individual to decide if those differences are a bother for them or not. I've never said the Kemper sounded shit. It doesn't. I've only said that for my tastes - high-gain'd palm-mutey metal and rock - that I prefer the QC, because it seems more accurate to the real amp.

    That doesn't mean the Kemper sounds bad, and you could get good tones with either unit.

    There are some weird things with the Kemper though - I own a Diezel D-Moll. When I profile it with the Kemper, the Kemper says a noise-gate was detected.... erm... no.... it's just not a very noisy amp! lol. Strangely enough I don't get that issue with my Marshall JVMJS, which DOES have a noise-gate built into it. I obviously deactivate that whenever I profile it.

    I also own a Diezel VH4, Fryette Sig X, and Orange Rockerverb. I've made over 200 custom profiles of all of these amps, and there are some really good sounds there. I'm building up my QC collection of captures now.

    Well at GuitarGuitar here in the UK, units are priced as follows:

    QC - £1599

    Helix Floor - £1230

    Kemper Toaster + Remote - £1849

    Kemper Stage - £1389

    Certainly if you care about more than just the neural capture, I can see it being considered overpriced. I don't really have a dog in the price race. I've already got my three units!

    Naaah. In an early Interview this was mentioned, as I said above, also here:…loor-board-2-0/

    (Better learn German)

    Does ir really matter? Naaah.

    That's weird. I've literally never read that on any other site. Can anyone confirm? Certainly never saw it in any of the emails from NeuralDSP over the months.

    I ordered my unit in March. I paid £1449, and received it in May. Although my friend who pre-ordered it and was part of Tier 1, he had his in February I believe. So that's how I got "early access", heh heh heh.

    He paid the same as me, which I always thought was a bit of a crappy deal for a pre-order person. I mean, from what I can tell .... all they got above and beyond what I got is an attractive looking 'making of' document, and that seems to be about it. Not too impressed with that myself.

    The first price from Neural had always been a preorder special offer. This was known. Right at the beginning, over a year ago, Neural had announced that there would be an immediate price adjustment of at least 100 Euros after the preorders, so there are no surprises here. What is kind of a surprise though, is that it took so long for the price to be "adjusted" and that 100 Euros became 250. That only can mean that there is a huge demand for this unit and cannot only be explained by inflation...:/.

    Nah, I think it's to do with component shortages. Fractal raised the price of their units too recently. Kemper hasn't yet.

    Oh, and yes. The QC is missing a lot of stuff. In my write up before I said there were many reasons that Kemper owners wouldn't like it. I also said there were some holes in the feature set - I won't go fully over it all again, but I am having to make some workarounds.

    But I bought the unit solely on the promise of the capturing technology. Because everything else I can either workaround or I can wait for updates. I wanted the best amp profiling I could get, because I am proper ADHD about my amps!!!

    QC would've been sent back to the shop if it didn't live up to the hype.

    I am sorry to tell you that your point is wrongly argumented. You implied that the quad cortex is more "accurate" than the kemper, and somehow it improves the more you capture the amp.

    I literally never said this bit. I don't even know why you would say that. I'm afraid you might have an issue with your understanding of my viewpoint.

    My viewpoint is that the QC captures are closer to the real amplifier than the Kemper. It's to do with accuracy when compared to "the real thing" - That's all I have said.

    This has been true for me across all of my comparisons for months now. I've posted some clips here, but even those do not encapsulate the totality of my experience. I've spent months with both units side by side now.

    I have had plenty of time and plenty of variation in the profiling and capturing processes, enough to know that to my ears, the QC gets closer.

    I even said it doesn't get 100% of the way, but that the differences do not annoy me as much as the differences between the amp and the Kemper. So there is no blind fanboy-ism going on here.

    I'm currently running the QC in the FX loop of the Kemper, and I'm turning off the stack section on the Kemper and using the QC in the X slot - so basically, Kemper for effects and QC for the amp. It works very well. The QC can even control the Kemper via midi in this configuration too.


    From a tech level..... and I don't expect to fully understand everything involved, but in the last decade we've seen improvements in machine learning techniques and music information retrevial techniques. Things like self supervised variational auto-encoders, anomaly detection, and deep learning techniques that were not production ready back in 2011.

    It really shouldn't be a surprise that the QC is able to get closer to the real amp, much in the same way that the Kemper and Helix and AxeFXIII and QC all sound better than a Line 6 PodXT. There wasn't the knowledge or the computational power back when the PodXT was launched to get as close to a real amplifier experience as there is today.

    it’s kind of a strange concern of some. Aside from commercial profile makers, how often does one profile or capture anyway; even pros only get there hands on just so many amps? And when doing so, how much time is first spent setting everything up, tweaking, moving the mic or mics around, listening to the result in another room on monitors to the mic’d sound for perfection. These time consuming steps prior to starting the simple profiling or capturing processes make whatever 15 seconds is spent on refining meaningless. It’s kind of like when my wife spends an hour rearranging a hard to reach shelf in the kitchen in order to save 10 seconds accessing a plate she uses once every year or two.

    It's just about accuracy. Nothing to do with our wives, nothing to do with how often we do or don't make profiles, and nothing to do with time consumption.

    Just accuracy.

    And that isn't a strange concern. It's the entire point of the technology.

    We usually deal a lot with musicians that do not see Refining as a burden, but in stead as a chance to control and polish the profiling outcome. As a good reference, there are the Guido Bungenstock videos that are spot on.

    I have the assumption that you are not interested in finding reasons for your Profiles having too much bass.

    How do you honestly expect me to react to this? Have I not demonstrated open-mindedness in several direct conversations with you over the years? I'm pretty sure I have.

    Also, you're making the fallacy of assuming new technology must necessarily yield better results, which might be often the case, but need not be (independently of my own personal preference for the KPA)

    Not true. I never said that new technology always results in better results. I said that if it does, it shouldn't come as a shock or a surprise given the maturation of the underlying mathematical frameworks that these profiling/capturing technologies use.

    If the QC sounded not as good to me, then I would say so. I have no allegiances or brand loyalty. I'm not a hyped up Youtuber, I'm not a gun-for-hire, and I'm not a partisan. These things are just tools to me.

    When comparing the Kemper to the QC for capturing, and having the reference amp still setup for comparison between all three, time and time again the QC gets closer to the real amp.

    This isn't to be unexpected or a surprise. The Profiler has tech from a decade ago. Machine learning and AI assisted systems have moved on a lot since then.

    No-one gets defensive or passionate when the next model of phone comes out and it has a better camera built into it. It's just the way technology evolves.

    It feels kinda strange to not have too many complaints about the QC. I'm just getting on with it, and I'm getting the best "in the box" tones I've ever gotten. I see people's complaints and none of it rings true for me. Funny how our experiences can be so different. And my needs aren't simple blues rock stuff. I play in a post-rock/prog-metal band, and my patches are complex.

    I'm more than happy with the capturing tech. It sounds excellent. I bought a Suhr Loadbox IR so I could make captures. Previously I was using a DI box in between the amp and cab, but now I can do it in a more silent fashion.

    I've decided I'll keep my Kemper and Helix for the time being though. As there are loads of great things about all three units, and I'm not in a rush to sell, I can afford to be a gearnerd for a bit longer!