Noise Gate

Chuck P

Chuck P

Inspired
So, the topic of the hour seems to be noise management. With the Beta thread and the Recto-and-also-noise-gate thread thoroughly derailed, I figured I would start a new thread to start a more focused discussion.

A lot of the "noise gate" conversation isn't really about noise gates, right? A noise gate opens when you're playing and closes when you're not, so there is no noise allowed through between notes. What people on the board are mostly talking about is a filter of some sort that reduces the noise while playing, effectively increasing the signal-to-noise ratio. You can do this with a notch EQ (and accept the impact on the tone), or presumably there are other "smarter" algorithms. I guess these try to analyze the noise between notes and intelligently "subtract" that from the tone, or something?

I feel like if we got a much better noise gate, it would not address a lot of the concerns being expressed.
 
What might be needed (optionally) reminds me of fiddling with the type of plugins like Waves X-Noise, X-Hum, ...
Maybe because I was not an expert at setting those parameters, there was always an obvious loss, easy to hear by pressing the Difference button.

x-noise.jpg

Guessing these techniques are 25+ years old, I hope there has been improvement. AFAIK, software like Izotope RX 10 can now do impressive things. A difference here might be that we are asking to have all these parameters under an easy knob (or two)...

When I was learning about the The Famous E (EP samples), I read that they too decided and succeeded in eliminating some remaining noise w/o hurting the true sound. So, as Jay Graydon is a perfectionist, I trust that we've indeed had some progress in this field (or it was a relatively "simple noise").
 
Last edited:
sprint said:
but a noise gate like ISP's decimator G, as described / suggested in the other thread, would address those concerns of noise behind notes played without impacting tone, - wouldn't it?
Click to expand...
The decimator does affect tone, its removing signal. Most people just dont notice it. Think about it, you cant remove bad signal without taking some of the good signal away with it. So theres a difficult balance. If you get too aggressive people will claim "tone suck" because thats the language they know.
 
Feflicker said:
The decimator does affect tone, its removing signal. Most people just dont notice it. Think about it, you cant remove bad signal without taking some of the good signal away with it. So theres a difficult balance. If you get too aggressive people will claim "tone suck" because thats the language they know.
Click to expand...
Agreed, but this thread suggests noise gate improvements would not address the "noise behind play'd notes" issue - my point is that I think noise gate improvements would address this issue as discussed in the other thread (though as you've pointed out, maybe not perfectly). Maybe it's just a matter of how such proposed improvements would be packaged - in the gate, in an EQ option...wherever (imo the a new gate option makes the most sense since the issue at hand is related to noise reduction).
 
Feflicker said:
Think about it, you cant remove bad signal without taking some of the good signal away with it. So theres a difficult balance. If you get too aggressive people will claim "tone suck" because thats the language they know.
Click to expand...
Maybe that's "old thinking" though — things progress. I think I demo'd or heard (earlier) demos of that RX 10 software once, and it seems to do great things that were not possible before(?). I would guess that "yes," you will lose something but you may not hear it anymore, as they have progressed in "restoring" the damaged part to a level you can't hear or don't mind the difference. The end result may not be true to the original, but a sound people may prefer to the noise. The old noise removal plugins seemed more obvious to me, and I was never super happy with the result.
 
Last edited:
there is no noise allowed through between notes
Click to expand...

The above is not an accurate description of a noise gate for several reasons.

1. To say "no noise" is incorrect. The ratio of the gate determines its strength, from 20:1 (extreme) to 1:1 (no effect)

2. The noise gate does not discriminate "notes" from any other sound.

Also, I recommend that you look into the difference between our "Classic" and "Intelligent" noise gates. I have yet to meet a gate problem that can't be solved with one or the other. These have included aggressive artistic djenty clamping vs. standing next to a neon sign in a club above a taxi company's transmitter.

Some good ways to reduce background noise include: use quieter pickups, shield your guitar, use good cables, eliminate external interference, reduce gain, and stand/sit in an orientation that reduces 60 cycle hum.
 
I had an ISP Decimator G String on my G System rig with 2 x Genz Benz El Diablos and then with my FX8 with 2 x Mesa Mark 5's.

The intelligent noise gate on the FX8 was better than the ISP so I sold the ISP Decimator.
 
Admin M@ said:
The above is not an accurate description of a noise gate for several reasons.

1. To say "no noise" is incorrect. The ratio of the gate determines its strength, from 20:1 (extreme) to 1:1 (no effect)

2. The noise gate does not discriminate "notes" from any other sound.

Also, I recommend that you look into the difference between our "Classic" and "Intelligent" noise gates. I have yet to meet a gate problem that can't be solved with one or the other. These have included aggressive artistic djenty clamping vs. standing next to a neon sign in a club above a taxi company's transmitter.

Some good ways to reduce background noise include: use quieter pickups, shield your guitar, use good cables, eliminate external interference, reduce gain, and stand/sit in an orientation that reduces 60 cycle hum.
Click to expand...
Thank you for this thoughtful response M@. The noise gate (really a signal gate, as you correctly point out) doesn't do anything to reduce noise when it's open, right? I think a lot of folks are looking for Fractal solutions that will remove hum/noise/buzz while playing above the gate threshold. None of the Fractal Noise Gate modes address that concern, do they?
 
Chuck P said:
Thank you for this thoughtful response M@. The noise gate (really a signal gate, as you correctly point out) doesn't do anything to reduce noise when it's open, right? I think a lot of folks are looking for Fractal solutions that will remove hum/noise/buzz while playing above the gate threshold. None of the Fractal Noise Gate modes address that concern, do they?
Click to expand...
I think that the “intelligent” one is normally done for that. But it is not “intelligent” enough compared to isp.
 
I have also done another test but different this time, using the axe fx in the fx loop of my tube head . Just using the different gates in the unit. It is the same . The noise is non filtered while playing .
Well we won’t do round 36. I think everything as been always said, asked and demonstrated. Now it is in the hands of the maker, if he can or not, if he cares or not …we will see
 
My name is mud said:
The point is the signal is not affected at all.
Click to expand...
Put it on an oscilloscope and Ill bet the house its altering the signal. Now, whether or not you perceive it as a problem, that can be very subjective. Many people can hear those differences and dont like the decimator. I personally think it does a good enough job Id use it if I needed it (havent needed it since I moved to Fractal).
 
My name is mud said:
I think that the “intelligent” one is normally done for that. But it is not “intelligent” enough compared to isp.
Click to expand...
Matt seems confident and another guy claimed his AX8 was better... With a motherload of EMFs going around, could it be that sometimes one solution is better ("more intelligent") than the other, and you're simply in a situation where the ISP would win it?
 
In theory, noise that's exactly predictable can be cancelled or maybe filtered out, but I doubt that's a real world situation.

If it varies over time, what principles would you think a gate/filter/denoiser would use to distinguish noise from music? "Sounds I don't like" doesn't seem reasonable.
 
Dave Merrill said:
If it varies over time, what principles would you think a gate/filter/denoiser would use to distinguish noise from music? "Sounds I don't like" doesn't seem reasonable.
Click to expand...
I guess you can distinguish it at least two major ways...
  • many of these interferences have typical properties you can "filter" for, but yeah, we have to deal with a huge soup of those...
  • if you wrote the code for a Fractal, you probably also have an idea what kind of noise is more likely to be caused by or needed for a certain amp sound, and try to leave that untouched or least damaged. Maybe you could even put signature data in the signal that can not be heard, but detected by your filtering code to help preserve the signal that makes it — kinda like putting a dye or a radioactive marker on things.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom