Softube rant (NOT a deal!)

[PavlovsCat]

49 minutes ago, DeeringAmps said:

It’s the Fourier Transform Equation 

 

I hit drums. 🥁 

But I did Google it and understand the basic concept. 

[DeeringAmps]

Dave has explained this at least once. That’s why I knew it was Fourier. IIRC this equation explains why 48k (maybe 44.1) is all the bandwidth we need for audio reproduction (or not). 
I’m sure @bitflipper will chime in and enlighten us. 

t

[mibby]

1 hour ago, DeeringAmps said:

Dave has explained this at least once. That’s why I knew it was Fourier. IIRC this equation explains why 48k (maybe 44.1) is all the bandwidth we need for audio reproduction (or not). 
I’m sure @bitflipper will chime in and enlighten us. 

t

Definitely interested in hearing this!  And I thought this was just a "deals" forum. 

[Paul P]

This is also close, but not quite :

Edit :  Ah, I just discovered that fs = 1/Ts, so real close.  What remains is why w and a are different terms.

Edited June 11, 2023 by Paul P

[scook]

1 hour ago, DeeringAmps said:

Dave has explained this at least once.

yes, a couple of times onhttp://forum.cakewalk.com

this may be the first discussion on this forum

 

still think the image name is the best clue so far

[bitflipper]

Can't sneak anything past Steve!

iirc that image was likely lifted from wikipedia's article on the sampling theorem. I adopted it as my sig in response to a heated debate on sample rates (remember when those were the kinds of topics we kicked around? More innocent times, I guess.)

Once any subject has reduced down to subjective matters of opinion, it's hard to argue for or against any particular point of view. One guy loves distortion, the next guy goes out of his way to avoid it. There isn't a correct answer. It's easy to feel like there are no hard truths.

But physics and mathematics are not subjective. Harold Nyquist figured out how many samples would be necessary to accurately transmit a specified frequency range, when he was working at Bell Labs. But smart as he was, he could not come up with a mathematical proof. That challenge was ultimately conquered many years later by a fellow by the name of Claude Shannon, arguably one of the smartest math dudes to ever make the rest of us feel hopelessly dumb. Claude was the guy who originated the idea of using binary numbers and Boolean logic in computers. So none of us would be doing what we do without those guys and their big brains.

Confidently incorrect forum posts like this one (from Gearslutz) actually piss me off:

Quote

Highs over 10KHz suck with 44.1Khz becouse the Nyquist theorem its wrong.

Sure, it's silly to get angry because "somebody on the internet is wrong". But it's not fair to the novices who read nonsense like that and then perpetuate it. There's plenty of room for opinions in the digital music space, but real fundamental truths are at its foundation.

[scook]

2 minutes ago, bitflipper said:

But it's not fair to the novices those who read nonsense like that and then perpetuate it.

let's broaden that class a bit

oops I rarely pun and never on purpose

[antler]

33 minutes ago, bitflipper said:

Harold Nyquist figured out how many samples would be necessary to accurately transmit a specified frequency range, when he was working at Bell Labs. But smart as he was, he could not come up with a mathematical proof. That challenge was ultimately conquered many years later by a fellow by the name of Claude Shannon, arguably one of the smartest math dudes to ever make the rest of us feel hopelessly dumb.

Interesting. I've heard of and read about the Nyquist frequency, but this is the first time I've heard of Claude Shannon. I wonder why his name isn't part of the (Nyquist frequency) term.

As for subjective posts, I seem to remember someone used to claim that the Cubase(?) engine sounded 30-40% better than the Sonar one, despite never sharing how that measurement was qualified.

[Grem]

23 hours ago, bitflipper said:

There was an update earlier this year to my longtime go-to FET-style compressor (PSP FetPressor). Cool, I thought. Maybe they're adding some modern features such as a dry/wet mix knob. Nope. The "update" was that they were making it dependent on Pace. Big Nope. I backrevved to the previous version. If they never update it again it won't bother me at all. It's great as-is.

Yep, same thing happened to thier 2425 reverb plugin. I am still using the non-Pace version. Still works great. Excellent sounding reverb too.

[bitflipper]

3 hours ago, antler said:

As for subjective posts, I seem to remember someone used to claim that the Cubase(?) engine sounded 30-40% better than the Sonar one, despite never sharing how that measurement was qualified.

Oh, yeh. That  became a meme around here for years afterward.

For historical accuracy, I maintain a file where I collect funny forum posts, and the entry I have in there reads

          "I find the audio engine to be 20-40% better in Nuendo".

Ron Kuper then offered to test his test files, saying he'd done null tests against Cubase/Nuendo and found no differences. The poster sent him two MP3s. When others suggested that they needed to be lossless files, he came back with this retort:

          "i dont see the point in you coming back with a lot of useless scientific information about the difference in extraction of two wavefiles.Surely its what we hear that matters. Any wavefile will sound much the same even if played through windows media player,".  IOW, a null test means nothing if *I* perceive a difference.

[Grem]

15 minutes ago, bitflipper said:

For historical accuracy, I maintain a file where I collect funny forum posts,

Great idea.

[Nick Blanc]

11 hours ago, bitflipper said:

Oh, yeh. That  became a meme around here for years afterward.

For historical accuracy, I maintain a file where I collect funny forum posts, and the entry I have in there reads

          "I find the audio engine to be 20-40% better in Nuendo".

Ron Kuper then offered to test his test files, saying he'd done null tests against Cubase/Nuendo and found no differences. The poster sent him two MP3s. When others suggested that they needed to be lossless files, he came back with this retort:

          "i dont see the point in you coming back with a lot of useless scientific information about the difference in extraction of two wavefiles.Surely its what we hear that matters. Any wavefile will sound much the same even if played through windows media player,".  IOW, a null test means nothing if *I* perceive a difference.

This reminds of a discussion I was never part of and have no use for remembering, but here we are. In the Beatmaker community, Fruityloops (FL Studio) is basically king. And a lot of (well known, multi platinum etc.) producers use a particular older version which "sounds better", harder, snappier, etc. Turned out that version had a limiter on the master under the hood. 

One of the issues I have with Ableton (besides the current erratic CPU behaviour), is that sample previews are much louder and I believe compressed. Oh the continuous dissapointment when I drag a great sounding kick in my project to have it flop away somewhere in the distance of my lowend. When all I have to do is turn it up a bit.

The 'same' goes for old MPC's, which have 'that mojo', but a lot of that magic was that they worked on 12 bit. You can watch hours of beatmaking tutorials but rarely do you see someone say "just use a bitreducer on the drumbus" as if it's some kind of secret. Clip your kick to mush, bitreduce the transient back to life and filter the high-end (or use a good tape emulation)...done. It sounds like doodoo on its own, but in a mix it's perceived as much louder and harder than a clean kick. 

[Starship Krupa]

2 hours ago, Nick Blanc said:

Turned out that version had a limiter on the master under the hood.

Which kinda validates a point that I have brought up: while theoretically all DAW's should sound the same, we don't know what any of them is truly doing under the hood.

Yes, Nyquist, Fourier, et al, but then we have Image Line....

Like Helmuth von Moltke's contention that "no plan survives contact with the enemy," I say that design principles are often compromised during actual implementation of them. Engineer: "it presents the recorded audio with no resampling, added jitter, or coloration of any kind." Marketing person: "great, is there anything we can do to make our program sound better than the competition?"

[Paul P]

5 hours ago, Starship Krupa said:

Engineer: "it presents the recorded audio with no resampling, added jitter, or coloration of any kind."
Marketing person: "great, is there anything we can do to make our program sound better than the competition?"

That's up to marketing !

[Fleer]

The Three Marketeers!

[scook]

18 hours ago, Starship Krupa said:

Like Helmuth von Moltke's contention that "no plan survives contact with the enemy,"

Mike Tyson said it better "Everyone has a plan until they get punched in the mouth."

At least this resonates with me more (I wonder who really came up w/ it).

[Nick Blanc]

12 hours ago, scook said:

Mike Tyson said it better "Everyone has a plan until they get punched in the mouth."

At least this resonates with me more (I wonder who really came up w/ it).

Another Tyson gem and my personal favorite:

"Social media made y'all way too comfortable with disrespecting people and not getting punched in the face for it."

The stage was set, but I have the feeling that since covid  this holds an insane amount of truth.

[e-cue]

Also a thread related Tyson quote from a press conference:

 

you know what... I don't wanna get banned here....  nevermind... 

[iNate]

On 6/10/2023 at 12:39 PM, mibby said:

WHY would I continue to use a plugin in new projects that is no longer going to be supported??

Speaking for myself.  I don't.  I uninstall and move on.  There are material risks in continuing to use it in projects.  This is especially true if you don't have a good system for stemming things out and archiving the audio.  If you rely primarily on your ability to access Archived DAW Project Files, then it's a good idea to simply cease using out of development products.