What makes analog so analog?

[griels]

Here's something I did just for the kicks:

http://www.koti.org/basstation/

I compared the real thing to the vst-i version.

The mp3s sound very different... The 'real' mp3 sounds a lot smoother.

Everybody can draw their own conclusions, I know where I stand

Where is that exactly?

Are you sure you used exact the same parameters as for both? Maybe the 'real' BassStation uses a higher sampling rate - 96kHz maybe? Was it recorded using a digital connection (SP/DIF or ADAT) or did it go via DACs/ADCs, in which case the sound has been coloured?

[tony tony chopper]

And from several pages before...

Quote:
While people care about dithering & other BS like that

& I maintain that dithering to 16bit is complete bullshit. You wanna fight about it?

[HanafiH]

What about CD players then, do they all sound like shit too? They're 44.1khz

As a matter of fact, I do think that most CD players sound a bit harsh, especially in the way they reproduce the higher frequencies. Playing 96 kHz audio from my computer definitely sounds smoother to me.

My understanding is that this is more to do with the requirement for a lowpass filter to be placed at the cutoff frequency inside the CD Player than the theoretic nature of 44.1k sound.

The harshness detected in CD recordings arises because of the failure during manufacture to remove frequencies from the source material higher than the Nyquist frequency (22.05kHz and 44.1 sampling) to prevent aliasing. The red book specification requires a single pole filter at about 20.5kHz, this is also known as 55/45 filter, as it filters out 55% of the required high frequencies but allows 45% to pass through.

It was thought when the red book was defined in the 80’s that these residual higher frequencies would be too weak to be heard. Which is true, but when aliased down, they are loud enough to intermodulate the primary acoustic signal causing audible high frequency distortion. As there is no such thing as a brick wall filter, many 44.1kHz sampling systems are prone to some extent or another to audible aliasing.

96kHz sampling places the nyquist frequency above the threshold for human hearing, consequently any intermodulation arising from aliasing is also inaudible.

[Sami]

Where is that exactly?

I like analog synthesizers better in sound than their digital counterparts.

Are you sure you used exact the same parameters as for both? Maybe the 'real' BassStation uses a higher sampling rate - 96kHz maybe? Was it recorded using a digital connection (SP/DIF or ADAT) or did it go via DACs/ADCs, in which case the sound has been coloured?

Yep, those were presets from the vst-i version which I tried to create as accurate as possible on the hardware version. Mind you that the hardware bassStation has analog filters. That might explain the smoothness you hear. Oscillators are DCO as far as I know but I'm not 100% sure... As for SP/DIF, my soundcard doesn't have one. And there's no digital output on the hardware bassStation either I used the shortest possible cables straight to my M-audio soundcard. I think there might be then some coloration but it shouldn't be that evident. Of course if you have those famous golden ears you might hear that

Anyway, there's smoothness in the sound of the real bassStation as you noticed though they did pretty good job on the vst-i version too I think...

[valley]

vsti's never have anywhere near the same energy at the same volume.

For one sound two have more energy than another suggests that the former sound has been compressed or saturated in some form, no?

There is no way that two identical sine waves from two different sources can impart differing levels of SPL.

It's one thing being a scientist about it....being a musician/engineer is another

If the physics doesn't stack up then what you are hearing is an illusion, or your test is broken.

[valley]

I just wake up and this thread got boring. Lots of BS, but not funny ones. Filters in CD players?

Are you saying you don't believe your CD Player has a low pass filter built into it?

[griels]

Where is that exactly?

hardware version. Mind you that the hardware bassStation has analog filters.

Ah, that makes more sense. I thought the BassStation was purely VA.

Re: DACs etc... It was a long shot - I was just wondering why a native version of a DSP algorithm should sound significantly different (given high enough accuracy calculations). As the filters were in fact genuine analogue, I was barking up the wrong tree.

I have no beef with people who prefer genuine analogue to software emulations - the sound is quite plausibly different, and the choice as to which is 'better' is purely subjective. It's those who assume that the use of digital siganl processors makes the same algorithms sound better who give me the hump... Anyway, think I'll bow out of this mellee once again

[foosnark]

Here's something I did just for the kicks:

http://www.koti.org/basstation/

I compared the real thing to the vst-i version. Everybody can draw their own conclusions, I know where I stand

The EQ is a bit different on the hardware. As far as value judgements though, I don't think either of them sounds "better" than the other.

Just curious though, is the Bass Station truly analog? I've read description of it as "quasi-analog" and Novation apparently called it "Analog Sound Modeling (ASM)"

[valley]

The harshness detected in CD recordings arises because of the failure during manufacture to remove frequencies from the source material higher than the Nyquist frequency (22.05kHz and 44.1 sampling) to prevent aliasing. The red book specification requires a single pole filter at about 20.5kHz, this is also known as 55/45 filter, as it filters out 55% of the required high frequencies but allows 45% to pass through.

I'm surprised actually. I'd have thought that given the increased recognition in the mastering industry of how easily people can here alias distortion, that more care would be taken in eliminating hamromics beyond nyqist. Wouldn't it be possible to use a steep slope filter at say 21K to remove a higher percentage of unwanted harmonics?

[jmh]

Gol

Thanks for taking it lightheartedly - I guess I'm improving with my output as I wasn't misunderstood

Fight for it? Naah... it's just that along with several other 'puzzling' digital audio concepts, the effects of dithering are a phenomenon that can be heard and thus validated as ... pardon the expression ... audible.

For me, a lot of this has been a learning experience. A while ago I was still rather suspicious about the audible effect of jittering. All it took was a simple demonstration with a M-audio Audiophile first playing a tune of mine using its internal clock, followed by using an external clock source. Granted, the monitoring was done on not-perhaps-your-average-monitors, the new Genelec 8xxx something, but I was amazed at the difference it made. Likewise hearing Waves plugins RTAS and TDM versions compared (in a good studio with rather expensive gear, which affected the results to some degree, I admit), I was able to hear a clear difference - and I was suffering from a severe cold, resulting in both ears being 'locked' (I so hope I knew the right term for this in english) and my right ear hearing high frequencies worse than the left ear.

Anyways... as for analogue sounding analogue, that's something I wouldn't touch with a ten foot pole But for the sweeping generalisation about 'it won't be heard in a mix' and especially 'the audience doesn't give a damn'... It's not about the audience, the subject is about us musicians, in the studio. And there it does make a difference - as long as there is a difference, that is

For me, the convenience of software outweighs the claimed superiority of analogue, but I still hear differences between the two. I still haven't been able to identify it, but replacing 909 samples with the real thing makes a marked difference for me. If I knew why, I'd be a whole lot happier And as long as I'm perceiving a difference, I'll kling on to the few bits of hardware I still own (including a marvellous Roland JX-3P, dunno what it is about it but oh boy, does it sound wonderful).

Regards,

JMH

[Sami]

The EQ is a bit different on the hardware. As far as value judgements though, I don't think either of them sounds "better" than the other.

Yes, it's entirely subjective of course. I prefer the sound of hardware version myself. Especially in high resonance settings, the h/w bassStation definitely sounds more pleasing to me though vst-i version has quite nice filter too! I don't know about the eq thing though...

Just curious though, is the Bass Station truly analog? I've read description of it as "quasi-analog" and Novation apparently called it "Analog Sound Modeling (ASM)"

At least the filters are truly analog, not modelled. I've read somewhere that it's digital but that's not the case. Novation drumboxes for example use ASM but not bassStation. That might explain the confusion... Oscillators might be digital but as I said I'm not 100% sure. Have to check it out...

And JMH, I have jx-3p too! I it!

[tony tony chopper]

Are you saying you don't believe your CD Player has a low pass filter built into it?

since I don't see any reason for them to, I don't believe it, no.

actually it'd be easy to test.. just put some sine above the freq of the filter you pretend to be in, and see if the player still plays it. I can't imagine any reason for a CD player to filter anything..

it's just that along with several other 'puzzling' digital audio concepts, the effects of dithering are a phenomenon that can be heard and thus validated as ... pardon the expression ... audible.

audible? no, not with 16bit.

I dare you to:
-browse to one of those web sites that tell you the amazing benefits of dithering
-check one of their audio examples (sure, they will probably have very convincing 8bit audio examples), pick a 16bit one, raise your listening level as they tell you
-hear the 'benefits of dithering'
-now, WITHOUT SETTING YOUR level back, listen to another piece of music
-come back and tell me your ears didn't break

[jmh]

audible? no, not with 16bit.

I dare you to:
-browse to one of those web sites that tell you the amazing benefits of dithering
-check one of their audio examples (sure, they will probably have very convincing 8bit audio examples), pick a 16bit one, raise your listening level as they tell you
-hear the 'benefits of dithering'
-now, WITHOUT SETTING YOUR level back, listen to another piece of music
-come back and tell me your ears didn't break

But why hearing the difference on a lower level volume is nullified by suddenly getting tormented by a high level volume signal afterwards? I fail to see the relevance, honestly

I would rather dare you back: get yourself in a situation where you're mixing something at 24bit, let's for the sake of argument say you're still on 44.1kHz. Now, do a mix and downmix to 16bit without any dithering. Then, do a few more mixes, utilizing preferably different dithering algorithms so you have more than two pieces to compare (and it goes without saying that applying the dithering should be done according to common sense and/or instructions ). Listen to them at the volume you're listening to when doing mixing. And most importantly, do this in a proper studio with decent gear (which doesn't imply millions invested in it, just a room that's been treated for the task and proper monitors). Now tell me you don't hear any differences? I'm sorry, but if you don't... well, I do. And I'm certainly not alone with this amazing skill.

I don't see dithering as a magical tool that makes music sound so much better, gets me all the chicks, a porsche or two and makes me penis five inches longer from its normal 12"... but it is an audible phenomenon with benefits (especially apparent in those loud listening situations as even you implied).

You don't hear the difference listening to the examples you referred to? Or does the occuring terrible volume explosion distract from it?

Just in case, we're hopefully still taking this lightheartedly

Regards,

JMH

* edited some incomprehensible parts and mistakes from two sentences *

[tony tony chopper]

But why hearing the difference on a lower level volume is nullified by suddenly getting tormented by a high level volume signal afterwards?

it's not, I was talking about your listening level. Ever heard a song using only the first 4 bits of the 16bit range? Are you able to hear any of those 16bit dithering demo's at your normal listening level?

You don't hear the difference listening to the examples you referred to?

I don't

personally, I only start hearing noise starting at 14bit (and that's still without dithering).

Btw why do anything in 24bit? What if someone on hearth could hear its quantization noise? It's 128bit or nothing

[jmh]

Damn, this would be faster if done on irc

Yes, I'm able to hear a difference on my normal listening level - using my trusty AKG K-301 headphones I've learned to love most of the time (as my current hellhole of an apartment makes it impossible to have monitors positioned properly - not to mention my home setup consists of shit speakers ). And compared to the superior hearing abilities some friends of mine like bTm and Kingston possess that bats would be jealous of, my ears are clogged

(Sorry guys for mentioning names, I'm just stunned about your skills in this area)

128bits, blergh. Only if the host has at least double the bit depth in the summing stage!!!

Regards,

JMH