u-he Satin or Slate VTM?

[recursive one]

Is it really relevant what Shy is saying? Satin is a plugin that makes virtually anything fed into it sound better. Isn't it what matters? I think, most Satin users bought it to make better sounding music rather than to fool anybody by claiming that they are using real tape machines.

I think that reasonable doubt in the extent to which Satin actually matches tape sound could be useful, but only if supported by A/B examples. IMO, pseudo scientific explanations why U-he allegedly fails don't work.

[Doug1978]

Satin is an excellent product. Is it tape? No.
Do I feel it is one of the best solutions to getting an element of 'tape sound / feel' to my tracks. Definitely.
It'll also go very nicely with my Nebula tape libraries.

Plus u-he is an excellent company in my, and many others', experience. And one that continues to develop its products long after they have been released.

How could you possible emulate the impossible ? That's impossible by it's very definition.

Impossibility, however, is only confined by failure to succeed.
Leave that behind, and impossibility becomes invalid.
Ask anyone who did something that was formerly regarded as impossible.

It's not a bad line Rob, but:
When did someone last boil water at zero degrees Celsius? I would say that was impossible.

[plastic]

Concerning Satin. (I can't try Slate)
I want to love it but I don't hear it doing anything.
I expected it to be be more like "The Glue", actually affecting the sound.
Not my thing I guess.
(I don't really know anything about real tape though)

[Burillo]

Concerning Satin. (I can't try Slate)
I want to love it but I don't hear it doing anything.
I expected it to be be more like "The Glue", actually affecting the sound.
Not my thing I guess.
(I don't really know anything about real tape though)

try putting it on every track and then bypass it

[J4R1O]

Urs, there seems to be some double presets in Satin's preset folder (29 in total). Is it ok to delete them? Example, original preset: RP A800 30ips and dobbleganger: RP_A800 30ips

Deleted them.

(no one noticed... too busy with debating )

[EvilDragon]

Concerning Satin. (I can't try Slate)
I want to love it but I don't hear it doing anything.
I expected it to be be more like "The Glue", actually affecting the sound.
Not my thing I guess.
(I don't really know anything about real tape though)

Seriously? Nothing? Even if you change the Rec/Repro EQ modes, gap width and bump? That's quite an obvious effect right there. How about bumping up Input knob with Makeup enabled? Quite obvious tape saturation happening there.

[Compyfox]

There's a very big difference between slapping together a few "basic" eqs and distortions in 2 or 3 days, or working a whole year on an analogue model that doesn't fail when its parameters are tweaked independently.

This proves Shy's assumption wrong and renders his rant silly.

...

In any case, accusing us of lying based on a lie was the dumbest thing he could do.

I definitely don't doubt that there was a lot of work put into both VTM and SATIN, maybe even UAD's creations and to a certain extend even Waves, while Nomad Factory, SKnote and ToneBoosters are a bit behind on things.

But the main point he's having, and that is the main point in all the "analog emulation" debates, is that you can't model the nuances. At least not yet. And tape machines are the kings class.



I have to admit that I haven't exclusively used tape machines in over 10 years, with the occasional use of my Grundik TK747 once in a while, or recovering an old cassette tape.

But one thing I miss with Satin so far, is the "in your face" behavor of the tape machine if I want to overdrive it.



Which brings me back to the FR of linking the VU with the saturation module's limit (that you call "headroom"). I'd like to know when that happens, I'd like to setup that especially.

My old Phillips, Pioneer, Sony, <insert name here> compact cassette tapes went to sh*t if I only ever so slightly exceeded the 0VU limit. Granted, large scale tape machines and "fast running" ones don't do that as quickly, since they have more space to align. But I can pull something similar off with VTM.


So if I have a critism, then it's that SATIN is simply not as agressive. But it's definitely the tool that is the most configurable.



The same could be said with console emulations and "crosstalk" - something I had with a user in the Klanghelm SDRR thread. It went along the lines of "this isn't crosstalk, as it doesn't influence adjacent channels/plugins". Well, so far this is not possible with any plugin. Unless it's interacting in terms of audio streams (and exchanges them as well). Then again, that's one of the negative sideeffects of a hardware piece - so why model it "exactly" unless we talk about plain purism.

Steven Slate once said it very excellently (something like):
"Imagine the Mona Lisa - what we do is recreate is as good as we can - but not with as many layers as the picture was painted.". What he meant was that his team created a plugin that "sounds" like a console, can work like a console. But cuts down drastically on internal math to pull that off. So you still get the "Mona Lisa", at a similar high resolution and detail, but the copy has not as much depth (in terms of paint layers). That doesn't mean that it's inferior sounding due to optimized code.


To some this might be bad or not suitable enough - but heck - if it sounds good and does it's job, I don't really care. If I can't afford a large scale console from the 60ies (a TG12345 for example), or if I can't afford building my dream console, then my only way is to access software.

And with the capabilities we have these days... it's possible for sometimes pocket money even. Then again, if stuff is compressed to sh*t for the radio anyway - who notices the nouances? Who even knows that this was all software?

I think all these arguments if Satin sounds like real tape or not can only be answered by ABX sound testing. Anything else is just a bunch of chat. Even then people will have their own opinions, but at least there is some science and the results are out in the open.

Oh, I waited for this.
"It's only right if we do a triple blind ABX test".

Up until this day, that's still the most stupid thing. Because you're still listening subjectively. If you expect something to be "that" (like a specific tape machine), then you intentionally rule out everything else as bad.

I've seen so many ABX tests all with results that didn't make sense. Or were absolutely biased (the most funny one was still the one studio invidation where they shot out which hosts sounds better on summing!).

If that's your thing, do it. The best ABX test I've ever heard was by Chris Hülsbeck for the now somewhat discontinued SQ8L. You clearly heard what was the software, but heck - I didn't care in a mix.





And regarding blatantly lying into the face of users. I know of a specific company (actually, several now even) that did that and still do that.

While we're at it, let's bring it up:
Crysonic is the prime example of straight marketing buzz. Undocumented software, developers that insist that they're right in what they're doing (even ignoring international metering standards), and constant sales to get even more customers.

U-HE is far from that. It's one of the rare companies that actually takes time to write a very detailed manual and also puts analysis plots in there. Something which was standard with hardware! In software form I only know of a handful of developers that do that.

[ThomasHelzle]

In the beginning the Universe was created.
This has made a lot of people very angry and has been widely regarded as a bad move.

Douglas Adams

[ObsoleteAcc99]

[DELETED]

[whyterabbyt]

Oh, I waited for this.
"It's only right if we do a triple blind ABX test".

Up until this day, that's still the most stupid thing. Because you're still listening subjectively. If you expect something to be "that" (like a specific tape machine), then you intentionally rule out everything else as bad.

I think you're failing to understand what a proper ABX test is.

[plastic]

Concerning Satin. (I can't try Slate)
I want to love it but I don't hear it doing anything.
I expected it to be be more like "The Glue", actually affecting the sound.
Not my thing I guess.
(I don't really know anything about real tape though)

Seriously? Nothing? Even if you change the Rec/Repro EQ modes, gap width and bump? That's quite an obvious effect right there. How about bumping up Input knob with Makeup enabled? Quite obvious tape saturation happening there.

OK I hear something. But it all seems very subtile, except changing the encoder/decoder blocks, but then what's happening is pretty wild and changing levels and I can't put it to musical use.
I realize that there are technical reasons for this, like emulating different tape machines, but I'd prefer a more abstract and musical approach with controls acting more like compressor/EQ controls, instead of preset modes.
Turning the input knob with makeup isn't doing what I'd expect from a saturation/compression device.

I own a UBK Fatso (ok it's a compressor) that can do crazy sound shaping and it has a "tape" mode that's not subtile at all if I drive it.

I also tried this tape hack described in this blog with an old tape deck and it's also very powerful and crazy if you drive it:

http://createdigitalmusic.com/2011/02/c ... d-adapter/

Satin is so subtile even on extreme settings that I can't really figure out what it's actually doing to dynamics and frequencies and how I can put it to use.

[Burillo]

The same could be said with console emulations and "crosstalk" - something I had with a user in the Klanghelm SDRR thread. It went along the lines of "this isn't crosstalk, as it doesn't influence adjacent channels/plugins". Well, so far this is not possible with any plugin. Unless it's interacting in terms of audio streams (and exchanges them as well). Then again, that's one of the negative sideeffects of a hardware piece - so why model it "exactly" unless we talk about plain purism.

SKNote StripBus does it

[Urs]

Ok, let's put an end to this.

Analogue tape machines were designed by electro engineers. Those engineers used a set of known formulae that were derived from scientific observation. Those formulae are of course not accurate models of what tape does, but they were accurate enough for the engineers to create those machines we're talking about. Thus, first there was a model of a machine, which was then translated into electrical/physical parts and components. Once a machine was built, it was measured against the original model to verify its fitness.

Satin was conceived the very same way. We obtained books like AGFA_Schallspeicherung.pdf and created a complete electrical and physical model of a tape machine. Unlike the olden days, we could use those formulae and graphs directly in our code, without need to build anything "real". After we implemented Satin, we could verify its fitness by testing it with real world signals, just like they did in the olden days. In other words: Satin is based on the same formulae as real world tape machines and thus it's based on the very same scientific model as those.

Now, people say that "analogue" has effects that go beyond measurement and all that, some kind of "real world randomness". But truth is, those random effects are described in those formulae too. Thus these effects such wow and flutter can be accurately applied to a virtual model, just like it's applied to a real world model by the constraints of the parts available.

Now. If anyone says that we don't have a model of tape, they're just bullshitting themselves. It's the same as saying that the people who built analogue tape machines have used the wrong formulae.

[ObsoleteAcc99]

[DELETED]

[J4R1O]

Ok, let's put an end to this.