Loving Satin!...but

[pdxindy]

I havent touched it, I guess I should read what the asperity control does

Turning it all the way down will also lower the noise floor. If you happen to have it turned up some, then you would be able to lower the noise further.

[bungle]

I bought it for the Dolby A emulation

I dont know who's yelling

I don't think many people realize what the Dolby A Emulation is capable of, it's a pity you can't get that on it's own.

[Ch00rD]

My own tests confirm that only very large multi-channel uses of Satin could create a hiss problem for users. I had multiple instances running last night, then tried all the stuff that used to create hiss symphonies in the old days, and as long as I turned down the hiss knob, there just wasn't any audible issue. There were, however, splendid sound old-school sounds. For me it's not an issue as I, like most users I suspect, will only be using a handful of instances per project.

I would agree to a large extent, but not completely. In my experience, noise can be easily avoided in the vast majority of cases. However, even when using only a single instance of Satin, it is still quite possible to have very audible noise levels, even with the controls for Hiss, Asperity (and Crosstalk, W&F) all the way down to their minimum values, in at least three different cases (which can overlap, though, making the noise louder when they do):

(1) With high Feedback settings in Delay mode (where noise can build up quickly);

(2) When lowering the Input level with Makeup enabled (Output kept at unity gain);

(3) When adding a significant amount of gain afterwards (for example, I often prefer to use a delay effect before an overdrive effect, to make the decreasing volume of a delayed signal with feedback hit the distortion at different levels).

For all three cases, the possibility to further decrease or even fully eliminate the noise that Satin produces would (marginally) increase the amount of creatively useful things Satin can do for me, which is why I'd welcome such a possibility. As a "nice to have" rather than "must have", though. I would definitely not call this a "problem", that's much too strong.

Btw, users may also want to get rid of noise for other reasons than it being audible at some point. For example, when using lossless compression formats such as Wavpack, absolute silence compresses much better than (very) low level noise.

PS: Nowadays, there is much more stuff we can do to create hiss symphonies than in the old days.

[Compyfox]

Being that so many of our modern plugins do respond to peak signals to do their decision making, I don't have the luxury of ignoring something just because its RMS level might be (arguably) "low"

But you can't define a definite noise level by a peak meter. It's the same way as you can't really level in a kick drum with a peak meter in direct comparision to a snare drum.

If the snare drum has a dBFS (peak) value of -9dBFS, and the Kick does so as well, then the snare is ultimately lower in RMS (loudness) than the kick. Of course, nowadays people give a funk about it. But that doesn't mean that it's the correct way to do so.


Learn how to use and interpret your tools first before being ignorant and saying "I don't care what level it is, it's there, it has to go".

How often do the developers and certain users need to repeat, that you can EASILY control the level of the noise, and tame it even futher with noise gates. Don't like it (the concept of the developers), don't use it - which sums up what zerocrossing said.

It's ultimately the decision of the developers if they give us the capability that certain users demand (in this particular discussion - more control over the noise), or not. And we have to accept that.

I hope I don't need to explain why I don't need to respond to the Argument from Antiquity parts of your post

You don't have to agree with me, you don't have to even argue with me. But you can't compare a RMS value of studio R2R with all 16 or even 24 signals summed, to a stereo consumer compact cassette without NR modes.

That's like comparing Apples with Oranges.

I havent touched it, I guess I should read what the asperity control does

The manual states:

Imperfections in the surface of the tape cause a roughening / blurring of the recorded frequencies. Although they also cause some constant mid-range noise, asperity noise is mainly perceived as an enharmonic distortion of the recorded signal. How noticeable this turns out to be depends on the frequency content of the original signal: if the signal contains very few high frequencies, you should notice an accompanying noise 'tail' when you turn the parameter up. Like 'hiss', asperity noise is also stereo. Strangely, asperity noise can make monaural recordings sound as if they were stereo!

Or in short:
Adds imperfections of a tape (imagine vintage tape compared to modern tape).

[Sampleconstruct]

We finally have a new War: The Noise War (lasted until the first update of Satin where a switch was introduced, the noise-defenders then all sold their Satins in the Marketplace and used white noise generators instead).

[Compyfox]

I think you might be on to something.


*though I'm still debating for educational reasons, and showing that there are always several sides to a medal*

[pdxindy]

For all three cases, the possibility to further decrease or even fully eliminate the noise that Satin produces would (marginally) increase the amount of creatively useful things Satin can do for me, which is why I'd welcome such a possibility. As a "nice to have" rather than "must have", though. I would definitely not call this a "problem", that's much too strong.

+1

[hibidy]

Maybe you got the next version, so the issue is hisstory.

Hiss, you've hissed your last hiss!

[Urs]

Well, those who think they need to come up with a "tone" can happily go and stick to some tape emulation that doesn't sound like tape at all. Reminds me of the argument when Diva came out and people missed the omg-so-important oscillator reset. Which we then built in without noticable increase in sales. It was just nodded off by some, but the majority of people barely noticed. It'll be the same here. We'll add a button, and it'll create a bit more "silence" - quite literally -, but the majority of people who want a realistic tape simulation simply won't care. At all.

As for the question about a solo NR - click "Bypass Tape" and there's no hiss at all. Just EQ and compander. Which still at the price of Satin is a steal, compared to what the vintage hardware goes for on eBay.

(I guess you can see that I'm tired of the useless bickering after we already announced a "fix" in about every Satin thread known to man)

[hibidy]

We finally have a new War: The Noise War (lasted until the first update of Satin where a switch was introduced, the noise-defenders then all sold their Satins in the Marketplace and used white noise generators instead).

"And let it be known that hisstory will show the "noise war" started by Uhe satin"

[Gonga]

My own tests confirm that only very large multi-channel uses of Satin could create a hiss problem for users. I had multiple instances running last night, then tried all the stuff that used to create hiss symphonies in the old days, and as long as I turned down the hiss knob, there just wasn't any audible issue. There were, however, splendid sound old-school sounds. For me it's not an issue as I, like most users I suspect, will only be using a handful of instances per project.

At a hiss level of -51dBFS for only 16 tracks with an easily "normal" project setup (compressors with 6db of makeup) we are just under a cassette tape's level of noise. I don't know about you, but I'd call that a problem

Is it accurate? Maybe, maybe even probably

Do I want the hiss? No

Fair enough. Maybe I should have left off the "very large." My impression, which may be incorrect, is that this is a plug that is meant to be used on busses, and perhaps a bit here and there for tracking too, like on particular instruments or voice, or for flanging. So 16 tracks is still a lot. But point well-taken.

That said, I was happily surprised that in cases where I used to hear lots of hiss, like doing heavy echo bouncing at high levels (feedback), Satin displayed extremely low levels of hiss.

I would conclude that it's a valid issue but not a big problem with the software - a very minor problem at worst for the majority of users.

[Pipelineaudio]

As for the question about a solo NR - click "Bypass Tape" and there's no hiss at all. Just EQ and compander. Which still at the price of Satin is a steal, compared to what the vintage hardware goes for on eBay.

The vintage hardware sits in a corner of my studio, unused. If I wanted what it did, I'd be using it. The plugin is much more convenient and far more cost effective from a cost per chargeable hour standpoint.

Is your point that you'd like me to pay you to make a version of the plug where you don't *ADD* the hiss? I think I'd be happy to, name your price and lets see what we can do

[Pipelineaudio]

Fair enough. Maybe I should have left off the "very large." My impression, which may be incorrect, is that this is a plug that is meant to be used on busses, and perhaps a bit here and there for tracking too, like on particular instruments or voice, or for flanging. So 16 tracks is still a lot. But point well-taken.

In my case, it wasn't a case of looking for trouble. I made a preset with the hiss control all the way down (blissfully unaware of any hiss when it was on the first track) and started adding it to other tracks, adjusting the mix control of the compander as necessary.

Pretty soon, there was a very noticeable hiss that was bothering the artist and we tracked it down to satin. I'm not trying to have an academic complaint, this is a problem in real world use. Yes, if this were 1990, it would be a non-issue. So would minor, human timing variations in playing and tuning. So would some slightly out of key vocals. But today we live in the age of the "million dollar demo". For my needs, the hiss is the problem. It may not be a problem for others, but that doesn't change that it is for me.

I'm not saying that everyone should lose their hiss, I am just requesting that there is an option not to add it.

I'm not taking anything away from anyone, but I won't accept (not that you are saying this) that it isn't an issue for me. I won't accept that its not measurable, perceptible and that it doesn't cause some undersired effects among levels, processing and overall output

[Compyfox]

Actually, the concept of a multi track tape machine (4 track, 8 track, 16 track, 24 track) is to use the tape machine plugins on individual channels rather than groups or the summing bus. (this part is directed at Gonga btw, but also to others that read along).

And in this case, a noise build up is bound(!) to happen. And totally normal behavior.


I can only emphasize it once more:
What's written in the manual of the original hardware, doesn't necessarily reflect the usage of all channels, but only a selected few. If not just one ir a stereo group only! So if there was only one channel modeled, which has exactly the same specs as written in the manual, using all 16/24 channels of the machine will inevitably result in a higher noise floor on the summing bus.

Period.


As example with one of the best mid-class R2R tape machines (IMO), the Grundig TK747. This machine is a 4 track, 1/4" inch R2R with three IPS (while 7,5ips being the fastest).

It has a noise floor of about -62dB (RMS) at 7,5ips.
According to the succesor (the TK847), the crosstalk for stereo tracks is 45dB, for dual mode (2x stereo) it's about 60dB, measured with a 0dB sine @ 1kHz

Now, the manual didn't tell me how exactly the device was measured. It (the TK747) only has a stereo output, so I assume this port was measured.

In case of large scale R2R's, you can connect the individual channels and the summing bus (depending on whether or not it's a stereo or multi-track device). So I assume the individual channels were measured - not the sum.


And, there is another thing to consider:
Companies lie to us. They have to, else they won't sell a thing. So if (i.e.) Studer tells us "the tape machine has a noise floor of -75something dB and barely any crosstalk - we believe them. Because we sure as hell don't think initially what will add up signal wise, or if they measured one channel module only. We only read "low values, check - bought!".

This is the same thing that's happening here.
Barely anyone makes a connection from marketing, to specs and measurements (one channel), to actual usage.

It'll be the same here. We'll add a button, and it'll create a bit more "silence" - quite literally -, but the majority of people who want a realistic tape simulation simply won't care. At all.

...

(I guess you can see that I'm tired of the useless bickering after we already announced a "fix" in about every Satin thread known to man)

Fair enough.

And yes, it will tone down the bickering. But sadly it won't fix the knowledge holes with some people.



Whatever you do, you'll do it right. And unless I get a Studer/Revox/Uher/Otari/Ampex tech on the phone to ask what's really up with the specs of their machine (measured the whole machine, or just one channel) - I assume they measured one channel module and be done with it.

And your preset providers have probably done the same.


Slate wouldn't have released his tape sim either if we wouldn't have been sure that it's indistinguishable from the real deal. And my measurements clearly showed that the machine noise is building up if you use several instances in parallel - however each channel stays within specs.


So... all is fine.
End of story for me.

See you next week with the 1.03 update (if you're that fast), Urs.

[Pipelineaudio]

And people used to die from smallpox, but then we invented vaccines. Are peoples' lives diminished now that they have the option of not dying from smallpox?

I'll take the best of the old with the best of the new. This hiss is entirely orthogonal to the desired behaviour of this plugin

You will notice that not ONCE did I say it should be taken away from those who want it

Apologetics aimed at forcing me to desire it are another matter