96khz 24bit profiling

  • Use what you like mate, but why come on here bashing a product you don't intend to use?


    It's like everything in life some people will love it, others will find it's not for them for some reason. No big deal.

    ESP Ltd Sir Headly, Ibanez Jem 777vbk, Jem DNA replica, Ibanez JS24p, , ESP ltd V307, Dean Zero Punk, Ibanez rg1527, ESP LTD EC1000, Chapman ML3-RC Traben Phoenix 5, Avalon L25c Redwood.


    Kemper lunchbox, Atomic CLR, Kemper Remote, Digitech Whammey DT, line 6 relay g50, Dunlop dp3, moog ep3x2, Randall KH 4x12, Carlsbro Nu Tone head.

  • Speechless... for a while, and then I got to thinking about this, so here's what I've come up with, monotone, off the top of my head (which, thankfully at my age still isn't reflective; only what resides within is so):


    In an ambient, acoustic setting, yes the difference can be heard by some. This is due to the complexity of waveforms produced by flutter echoes' adding to and cancelling each other out as they combine (collide), and is obviously dependant upon room size and construction materials (reflective-surface composition and shape). Sum and difference frequencies produced by harmonics at the source can also far exceed 20kHz, and in isolation or without a 96kHz reference, most would surely feel nothing is missing; it's subtle after all. For acoustic guitar and especially percussive and brass instruments such as trumpet, which are harmonically-rich (particularly odd-order harmonics), it can reasonably be argued that the difference is one of a sense of presence, space and air. The closer the mic'ing technique, the less ambient space is available for these interactions to occur, whether pre or post early reflections.


    Now, given what you've written, monotone, on the surface(!) it looks to me as if you either:


    1) Read about this somewhere and then, through confirmation bias, fuelled perhaps by the specs you read about the Kemper's outputs, believed you heard a compressed / squashed / smaller / less-airy sound than you might have otherwise. Note that it's acoustic, harmonically-rich (particularly odd-order harmonics such as produced by percussive hits and... trumpet) sources in ambient or distant-mic'd environments that the previous paragraph applies to.


    ... or


    2) Were affected by the Kemper's elimination of ambient bleed through the microphone/s.


    To clarify the second point, the Kemper's amp-profiled, cab-ified output is:


    1) Not going to produce frequencies above 20kHz, as is the case for any close-mic'd amp / cab combination in the real world.


    2) Not attempting to emulate distant mic'ing unless you employ onboard DSP, which is something that a discerning engineer such as yourself wouldn't dream of doing when you'd have at your disposal an arsenal of pro IR and algo-based 'verb plugs and hardware units that'd employ your favoured high sample rates, on which the extra CPU cycles are appropriate for high-end reproduction of complex acoustic reflections ('verb).


    In a nutshell (I failed here, as you'll see!), if you close-mic a real amp, thereby eliminating as much room interaction as possible, and feed that bandwidth-limited (by definition - guitar-amp cabinet) signal through your preamps to your converters, you'll be employing the standard modus operandi for recording electric guitar and bass, as you'd be well aware. This is what the Kemper replicates (including the preamp/s used during profiling if any were employed). In fact, it goes a step further in that it eliminates the room's ambience and potential coloration of the pure signal, something that countless engineers have striven to achieve but never quite accomplished; that'd take an anechoic chamber, after all. Besides, distant mics have long been employed by many engineers in preference to the use of DSP-generated 'verb, a logical consequence of many decades' worth of limited choices of artificial ambiences' being at hand. This distant-mic'ing simply isn't an option with the Kemper unless you either send a feed to a cab and distant-mic that whilst running your Profile into your DAW, or indeed close and distant-mic a cab hooked up to your Kemper, exactly how you'd do it for any other amp... provided that room's space and sound is important to you, which it may well be if it's a superb-sounding one.


    When recording the Kemper straight-up, add all the ambience you want... post tracking. Revel in the completely-dry, uncoloured (acoustically), raw tone of the Profile at hand, and thank your lucky stars all that setup work only had to be done once.


    I'm betting that this latter point has played a significant role in your judgement here. Perhaps you profiled an amp of yours and were struck by the lack of "air" compared to the original mic feed, leaving you with the feeling the profile was somehow "squashed", "smaller", and dare I say it... compressed.


    Hope this helps in some way, monotone. Your statement, amongst others, that for recording, the Kemper was "adequate" but by implication could miraculously become "good" (it's way better than good IMHO) through the use of 96kHz DACs, set all sorts of alarm bells off for me. The bandwidth-limited nature of the guitar->amp->cab->close mic signal chain renders this assumption hopelessly optimistic at best, and woefully ignorant at worst, IMHO. Something's definitely amiss here, and I can only hope that I've uncovered it.


    Cheers mate, and good luck!

  • I don't really have anything to say on topic as I don't hear any difference between 44,1 48 and 96 kHz... but many here are confusing the sampling rate with the actual frequency spectrum of a guitar amp.
    Of course a speaker only goes to about 10 kHz probably. And also nobody will hear frequencys above 20 kHz.


    But you won't be able to "recreate" a 40 kHz sine in 44.1kHz sampling rate simply because there are too little points to "match" the sine. Every halfwave will only consist of a few points.


    [Blocked Image: http://www.itwissen.info/bilder/a-d-wandlung-mit-einer-sampletiefe-von-4-bit.png]

    MJT Strats / PRS Guitars / Many DIY Guitars -- Kemper Profiler Rack / Kemper Remote / InEar

  • I don't really have anything to say on topic as I don't hear any difference between 44,1 48 and 96 kHz... but many here are confusing the sampling rate with the actual frequency spectrum of a guitar amp.
    Of course a speaker only goes to about 10 kHz probably. And also nobody will hear frequencys above 20 kHz.


    But you won't be able to "recreate" a 40 kHz sine in 44.1kHz sampling rate simply because there are too little points to "match" the sine. Every halfwave will only consist of a few points.


    [Blocked Image: http://www.itwissen.info/bilder/a-d-wandlung-mit-einer-sampletiefe-von-4-bit.png]

    You won't be able to recreate a 40kHz sine at 44.1 because the converters bandwidth filter will filter it out (anything above 20kHz, actually). This is standard anti-aliasing measures built in to all bog-standard AD/DA converters.

  • You also won't be able to HEAR 40 kHz if you're a human.


    By the way, that picture is not the best representation of the digital domain (not a criticism, just some info for anyone out there who might get the wrong idea of all this).


    The digital signal is not "stair stepped" like the graph on the right - rather it is simply represented by a number of POINTS, which are at the left side of each horizontal line.
    In reality, to recreate a sine wave perfectly, you need only TWO points (samples) for each half wave.

  • You won't be able to recreate a 40kHz sine at 44.1 because the converters bandwidth filter will filter it out (anything above 20kHz, actually). This is standard anti-aliasing measures built in to all bog-standard AD/DA converters.


    Yeah, I know - maybe my example was a little bit misleading :D


    You also won't be able to HEAR 40 kHz if you're a human.


    By the way, that picture is not the best representation of the digital domain (not a criticism, just some info for anyone out there who might get the wrong idea of all this).


    The digital signal is not "stair stepped" like the graph on the right - rather it is simply represented by a number of POINTS, which are at the left side of each horizontal line.
    In reality, to recreate a sine wave perfectly, you need only TWO points (samples) for each half wave.


    We are talking about the outputsignal of a DAC, right? It is stair stepped like that. (and so is the input signal of a ADC... printed over time!) Of course a filter will smooth it out. But in theory that's how the signal looks and one "stair" represents the value of the LSB. How do you want to create a sine wave with 4 points?


    Please correct me if I'm wrong :)

    MJT Strats / PRS Guitars / Many DIY Guitars -- Kemper Profiler Rack / Kemper Remote / InEar


  • I think this video explains everything :-)
    It also seems I was wrong - you only need two points per WHOLE wave to reproduce a perfect sine wave.


  • Thanks for the information @Michael_dk - looks like I only worked with the more simple DACs. I used to programm a functions generator some years ago using a DAC and I had visiable steps there in the analog signal :)

    MJT Strats / PRS Guitars / Many DIY Guitars -- Kemper Profiler Rack / Kemper Remote / InEar

  • Thanks for the information @Michael_dk - looks like I only worked with the more simple DACs. I used to programm a functions generator some years ago using a DAC and I had visiable steps there in the analog signal :)


    Oh, cool! :-) Yeah, I noticed that bit about older/simpler DACs while jumping around in the video to make sure I saw it contained what I thought. I'd forgotten about that bit! I assume that no current audio DACs use the simpler type, though.


    It's a very cool video, I've seen it a couple of times.

  • Yes, thank you for sharing this, Michael! I did not understand everything (as I am not an engineer and my physics classes happened to be in the 80s), but it appears this video is totally worth being watched a few times.