Arguing over anyone's ears is not scientific.
If we are talking about accuracy, a Null Test will make the point perfectly.
Failing that, a blind test comparison may be a good alternative.
Arguing that someone who believes a Kemper puts out such horrific crap as to not be usable by anyone that can hear is not unreasonable with the wealth of professionals (who also have ears) that tour with the device.
Agree on accuracy. A Null Test works pretty well; however, even a Null test does not measure everything. Still, I agree.
You are criticising someone for not going into enough detail, but then claim things without going into enough detail.
I think personally this is silly.
ymmv
It's pointless beyond reason or I would be glad to post more.
The derivation of the Nyquist frequency is something I was required to do in college. I could easily look it up and possibly even explain the math, but why?
Everyone here believes that the theory works.
Nonlinear processing creates frequencies much greater than the native Nyquist frequency. For example, if your native sample rate is 48 kHz and you distort the waveform you'll create harmonics far in excess of 24 kHz.
The solution is to use what is known as oversampling where you increase the effective sample rate internally and then downsample prior to final output. If you don't oversample you'll alias all those harmonics that are above Nyquist.
Orvillain knows of what he speaks.
Yes it does. The thing is, no one but dogs can hear the frequencies above the Nyquist frequency. In your example, who will hear those 24KHz and above noise? For that matter, most microphones and certainly most speakers can't even reproduce those frequencies.
I believe Orvillain knows a great deal; however, if he does know signals and systems processing at an engineering level, he is misusing the term "Aliasing" and attributing it to things that don't cause it.
Processing any signal using an algorithm can create artifacts. Not all artifacts are Aliasing. .
This is nonsense. Not everyone who gigs plays the same gigs or has the same requirements. Someone who plays out in a 7pc wedding band 4 times a months is going to have different gear needs to someone backing a singer every day for a month is going to have different needs compared to someone who gigs 2x a month in a local original project. If I was filling in for a band like Bring me the Horizon having a Kemper makes sense (mostly because nobody is gonna even know what’s guitar once all the tracks are playing) but if it was gig with Pelican that would 100% be the wrong rig to bring.
Agree to some extent. I think every person that gigs regularly cares a great deal about weight, setup time, durability, and reliability. Most care a great deal about good tone. People that gig (at least my anecdotal evidence) don't care about how accurately a capture device captured their amp, only that it sounds good in its final form. Most people that gig at a certain level are all using IEM's anyway. It isn't like they are hearing exactly what came out the speakers.
If you go down the food chain to the hole in the wall gigs, none of those guys can afford a Kemper, Helix or Fractal. They will all be using an old tube amp and some pedals.
FWIW, I'm not an EE (or a Vodaphone, or Carrier Pidgeon), but I'm fairly sure if you do want to measure Aliasing - you can do that without being an EE, or claiming to be one. You just need to hook up to software and it'll show you what's going on - and this can all be freeware:
https://www.roomeqwizard.com/
I'd far prefer that instead of cliffs of writing that don't say anything, especially from someone who is claiming to be a professionally EE.
Fair point, except you can't determine if the modified signal is due to aliasing or something else. Note, I am not saying that a Kemper (or any other modeler) doesn't introduce unwanted output. Distortion (as an example) intentionally adds crap to the signal. It is just pleasing crap.
I am simply pointing out that:
1) Kemper sounds pretty darned good.
2) Aliasing is something you will likely only hear on satellite radio (and I am amazed at how many people actually listen to music on this crap). All modern devices have long ago sampled at higher frequencies than anyone here (especially anyone here) can detect.
I'm not a betting man, and this isn't about betting on things.
We can test it to be 100% sure.
(I would get a Kemper and do this test myself if it wasn't for all the other things that have put me off about them).
Considering you are an EE - I'm sure this wouldn't be much trouble.
I agree with you on Kemper's recent conduct. I think that the levels were particularly bad. For years they have been claiming that Profiles are as good to perfect as possible and now they are marketing more perfect profiles. And finally, it is possible that they will cripple MK1 (which has the exact same DSP chip and SRAM) to artificially keep it from sounding as good as an MK2 for business purposes. The later is just my speculation. If I am incorrect, and Kemper actually makes the "better captures" sound just as good on MK1, they I apologize for the accusation.
I think there will be about a zillion null tests performed this summer when the new algorithms in Kemper are released. Testing a device that is designed to modify a signal with engineering tools (like a frequency analyzer or scope) is useless IMO. I do have a trace on the latency of the Kemper switching from one performance slot to another (it was around 40mSec IIRC), but this is hardly that simple to quantify.
)
