New oversampling ASIO driver

[Scali]

24/96, as I said above.
The only reason why it goes down to 16/48 is because it's a USB1.1 interface, which doesn't have enough bandwidth for 24/96.

[microdmitry]

Originally Posted by Scali

Exactly, conclusion: dynamic range is overrated anyway.

Just because you say so doesn't make it true. See e.g. AxeFX, which can sound like a living, breathing Fender. Your Zoom can't. If you think it can, I recommend playing a real Fender for a change.

Also, the *output* has nothing to do with the 'cleaning up' phenomenon you're describing.

You have reading comprehension skills of a five year old. Where did I say that clean up has to do with output? It merely has to do with compression that's applied to the input in order to bring the signal above the noise floor of a cheap ADC. Output is just merely not exempt from this compression, so you hear it there as well.

No, you just don't have any idea what you're talking about. You're even confusing input and output, making your entire argument fall apart.

I'm not confusing anything. And just because there's a tube buffer in front doesn't mean the dynamic range is significantly reduced. It could just be a cathode follower for all you know.

Now stop polluting my thread.

I will "pollute" whichever thread I want, particularly the ones in which people say that "dynamic range is overrated" and "16 bit is enough for everybody".

[Scali]

The dynamic range is reduced because I *measured* it. You know, signal generator on one side, oscilloscope on the other.
So yes, 'for all I know' it reduces dynamic range.

Aside from that I have no desire to sound like a Fender amp... and if you think a Fender amp sounds the way it does because it has an incredible dynamic range, you are sorely mistaken. Perhaps you should take your own advice and take some measurements aswell.

Now stop trolling. This wasn't a tube vs modeling thread in the first place, and I have no desire to discuss this topic, especially not with the likes of you.

[microdmitry]

Originally Posted by Scali

The dynamic range is reduced because I *measured* it. You know, signal generator on one side, oscilloscope on the other.

Somehow, I really doubt that. :-)

[Scali]

Originally Posted by microdmitry

Somehow, I really doubt that. :-)

Oh yea, like it's so hard to do that? Pfft, some people...

[microdmitry]

See, this would require taking the thing apart and breaking out the books. Seeing that you didn't have any reason to measure anything until this thread came into being, I think you just made this up. Besides, it's not enough to have just a signal generator and oscilloscope to measure the dynamic range. At the very least, you'd need a voltmeter to measure the noise floor accurately, and preferably a spectrum analyzer as well (if you want A-weighted measurement). This voltmeter better be accurate because depending on the quality of the input circuitry we're talking mV AC voltage levels here at best. Then there's the issue of having to establish either the max THD for the upper bound of your measurement or max voltage that the guitar could possibly generate (depending on how you define your dynamic range - per AES or as just a ratio of voltages), and making sure you don't exceed that limit.

And even that is moot, since no one is arguing there's a compression of signal before ADC, whether it's done by a tube or by means of a solid state compressor. The point is, 16 bit quantization has no place in a guitar signal chain. Optimizing the rest of the chain to fit the signal in 16 bits available in "USB" ADC (which Zoom inevitably had to do to avoid clipping) is retarded.

[Scali]

Wow, guess you haven't seen an 'oscilloscope' up close in the last 10-15 years or so?
We have this sort of stuff now: http://www.tek.com/site/ps/0,,55-14869-INTRO_EN,00.html

You don't need to take it apart either, you can just use the effect loop to reach the output signal just after the first effect stages (mainly the tube and the compression effects). That is good enough to determine how the tube affects the input signal (you can just dial in more or less tube to see what happens). So yes, I can declare with certainty that I do not require a huge dynamic range (not that the number of bits says all that much about range, more about the amount of quantization noise, which is rather moot on a 'canned' 24/96 signal).

Also, you still don't seem to comprehend how the Zoom is laid out.
There's a complete 24/96 chain that is independent of the 16/48 USB unit.
The 16/48 unit just samples the output. It doesn't affect the rest of the unit, and it isn't required. You can just aswell use the analog outputs and the whole 16/48 unit is bypassed altogether.
It's not like the whole chain is optimized for 16/48, and there is no need. All they had to do was match the 16/48 unit up to the 24/96 output, which is pretty trivial at line-level with standard components (24-bit isn't necessarily 'louder' than 16-bit, if that's what you think. It's just a different kind of discrete scale, which can and is applied to the same range of signal amplitude).

For someone trying to come off smart, you say some stupid things.
It's also getting rather tiresome. Why am I explaining myself anyway? All this has nothing to do with my driver.
But anyway, you can hear my Zoom recordings in all their 16/48 glory on my Soundclick. Now let's hear some of your stuff, I'm really curious. I'm going to laugh so hard if it turns out that you can't record or play worth a damn (which is usually the case with loudmouths anyway). The proof is in the pudding.

[rvai]

wow I basically didnīt understand much lol, however the few things I did understand seem really interesting, and Iīm sure it works seeing how it has worked for scali