Neural Quad Cortex

  • I think at this point it's best to leave others be beta testers for the platform for the coming months if not years so it can mature and be a real contender.

    It is not realistic to think that a new unit can compete with other products that have years of development and growth.

  • There are clearly a few software issues to be worked out (this is a pre-release demo, after all). The main point I took away from Paul David's video was that the captures sounded much closer in tone to his reference amp than the kemper profiles--and I think the comparison is even further in QC's favor in Paul's video than previous ones. One can argue that his profiling skills are not on par with some of the pros, even he makes this point, but it seems like the capture process is at least less finicky than profiling. And, this seems to be supported by Rhett's video (Kemper maybe didn't do a good job because of a compressor in the signal path, but the QC handled this ok) and Rabea's video (maybe if he had refined on the kemper, it would have been closer).

  • Once again, it's stuff that can be worked out in software. Button functions can change, the GUI can change. It's only the hardware that stays. As an example the Helix just came out with true preset spillover, meaning there is no delay at all when changing presets. Others will follow.

    You would hope so, but the fact that there was such a delay bringing this to market and they still couldn't fix it? That's a big leap of faith to hope they get around to it someday.

    But as it stands, it's unusable live. When will it be usable? 6 months? A year? Never?

  • You would hope so, but the fact that there was such a delay bringing this to market and they still couldn't fix it? That's a big leap of faith to hope they get around to it someday.

    But as it stands, it's unusable live. When will it be usable? 6 months? A year? Never?

    I'm a stomp guy. I sit on one preset during a song and punch in and out of effects. As long as there is no delay when punching in those I'm fine. Ordered.

  • And while we are at it..


    This kind of information I would have expected from the other promoters(I will not say reviewers anymore) first.A little bit disappointing I have to say.Maybe we will see in a next review that there is also no automatic volume leveling?

  • Once again, it's stuff that can be worked out in software. Button functions can change, the GUI can change. It's only the hardware that stays. As an example the Helix just came out with true preset spillover, meaning there is no delay at all when changing presets. Others will follow.

    I think you mean no audio gap. There certainly is delay when switching presets (delay of the latency/lag kind as well as the effect kind). The snapshots are lag free, though.

  • Serious question, ¿can we even disable the cab part in the QC? if there is anything I am impressed with the kemper, is how spot on is the algorithm for separating profiles. Also, I suppose direct captures are possible, ¿any example yet?

    I think the only way to use other cabs with Captures is to use a Capture of just an Amp. Kind of like a Direct Profile instead of a Studio Profile. If you do an Amp+Cab Capture then you're stuck with that Cab in the capture and you only have a 3 band eq and Gain. I've looked do that quite a bit because it seems like a HUGE drawback to me compared to Profiles. Not too much flexibility for the Captures so far. So I'm wondering how usable they'll be in comparison to Kemper Profiles due to the lack of tweak-ability for Captures.

  • I think you mean no audio gap. There certainly is delay when switching presets (delay of the latency/lag kind as well as the effect kind). The snapshots are lag free, though.

    No, there is a new feature in 3.0 for the Helix. all effects and amps must be on the same path and will use the same cpu if set in the global configuration. when switching to a new preset, it is absolutely immediate. THe problem is that your patches have to be simple enough and cpu friendly to run on one path and nothing is on the other path at all.


    This from Doug Castro regarding patch changes:


    "The gap between presets is an inherent limitation of having a free-form routing grid.


    With Kemper, it is not the case because the signal routing is quite pre-determined so a rig change is closer to changing a scene (updating parameters in a predetermined path) rather than unloading and loading a new one.

    Scenes and Stomps have been fixed, so they're seamless and quiet, Paul had a beta version (and probably could've done a better job at stating that in the video)."

    Edited 2 times, last by dean701 ().

  • It is an interesting contender I think.


    Small, very cool looking, great integrated editor, etc. I suspect it will get some good tones as well (I really didn't care for the spring reverb in the beginning).


    For live, it just seems a bit limiting in its capabilities. It also looks like the buttons are too close together .... but I would need to use it live. I am also wondering how the screen looks in direct sunlight, and if the buttons are rugged enough to keep working.


    For recording, while it looks to have a world-class built in editor, it really kinda pales next to the mighty Axe III FX PC UI.


    I think we have to wait for production to really start discussing tone comparisons.


    All in all, it really could be a good contender in the market. It likely not be polished at launch .... but neither was the KPA right? The KPA didn't have performance mode at launch and lacked a ton of features that have now been added. I think we should give the new guy a year and see where things stand.

  • I'm considering buying one.But, is the difference between the QC and the Kemper just processing power I wonder? If so , would a ( hopefully pending ) update and a processor update work on my beloved KPA's ?


    Anything coming from the administrators before a lot of us buy another competing product ?

  • The QC can also do “full rig” captures.

    But with full rig Captures, there's no way to use a different Cab in that Capture. With Profiles, CabDriver separates the Amp+Cab to allow Profiles to be used like Amp Models in which case you can experiment with different Cabs and IRs. With Captures, so far it looks pretty limited.


    Same with the parameters you get for Captures. Only a 3 band EQ and Gain

  • Neural seems that can capture even closer the exact sound of the amp...eventhough its in their beginning as company and updates are about to come...their capture procedure seems very successful..

  • You guys, I posted on TGP about the discrepancies mentioned in an earlier post, regarding the QCs processing power in relation to Neural DSPs research paper data.


    Doug Castro literally refused to give an explanation or any info about it to clear up how the marketing they're putting out could be possible with the QC given the processing power it has. Definitely didn't feel like a positive exchange and he was pretty defensive from the start. Kind of a bummer lol. I was hoping he'd give an explanation and prove me wrong.


    That doesn't mean it'll be a bad unit by any means. I'm still interested in what it has potential to be. But I'm starting to doubt some of what Neural DSP's marketing suggests


    And the exchange makes me feel not so positive about the company. I know that I and many Kemper users get treated right on this forum and the man Kemper himself is always communicating with us and answering our questions and it's typically positive exchanges. I really wish I felt like that kind of company-user relationship could exist with Neural DSP but I think with that, Kemper is one of a kind :)

  • All forums can be toxic and misinterpreted, the guy probably hasn't slept in days. (speculation). Not many Big Wigs chime in on threads and most of his comments have been helpful. Not like some other owners/CEOs, where they can be downright ornery and tactless. I won't mention them here. I will say it is certainly not CK.

  • You guys, I posted on TGP about the discrepancies mentioned in an earlier post, regarding the QCs processing power in relation to Neural DSPs research paper data.


    Doug Castro literally refused to give an explanation or any info about it to clear up how the marketing they're putting out could be possible with the QC given the processing power it. Definitely didn't feel like a positive exchange and he was pretty defensive from the start. Kind of a bummer lol. I was hoping he'd give an explanation and prove me wrong.


    That doesn't mean it'll be a bad unit by any means. I'm still interested in what it has potential to be. But I'm starting to doubt some of what Neural DSP's marketing suggests

    Who knows, maybe they did come up with something novel in the past year that the paper doesn't show. But another inference from their 2020 paper would be that they're running a low quality, low processor-intensive neural network, like the RNN32 from their paper which runs 8 * real-time on their PC with a perception score of just 80%. The QC Capture demos are close and impressive, but I don't really hear some unprecedented accuracy type thing listening to demos like with the Tube Screamer.

    Edited 2 times, last by yeky83 ().

  • No, there is a new feature in 3.0 for the Helix. all effects and amps must be on the same path and will use the same cpu if set in the global configuration. when switching to a new preset, it is absolutely immediate. THe problem is that your patches have to be simple enough and cpu friendly to run on one path and nothing is on the other path at all.

    Ahhhh that’s why, then. I don’t have any presets that use a single path. Now that excerpt in the manual about the Digitech processor makes sense.

  • For sure the quad cortex after some time will become a top tier modeler and will "set a new benchmark".


    The question is ofcourse..how long will it take to do so?


    Right now it seems like it is a immature product which needs a lot of time to become workable.


    Yes the KPA got similiar problems when it appeared.10 years ago.Same with the fractal stuff.


    If it takes one year or longer for the neural product to "go into full afterburner" and to be ready for meaningful live and recording work we will also talk about the possibility to have a kemper2 and a axefx4 on the horizon.Both with already built up "infrastructure".


    Right now I would not risk to jump on this prototype kind of neural product.More so that it does not seem to have really better sound results than the kemper or the fractal flag ship.Even the paid promoters can't really claim that.


    But yes..we will see in the next months how it really compares to the profiling of the kemper and the component modeling of the fractals.There is a lot to be proven.


    Indeed interesting times.