Jump to content
IronFilm

NoiseAssist, $600 Plugin for the Sound Devices 8 Series

Recommended Posts

1 hour ago, daniel said:

In a way you make the case for the 8 series NR quite well. By this I mean, spend 3k on Cedar DNS, may be it's tempting to use it for more than just previewing what post can do by way of a clean up (if you've managed to get production to cough up some extra for it being in the bag). At $600 though, NA is easier to incorporate into the workflow and if you owned an 8 series why wouldn't you? In 1 context the sound you doing is going to get no more finessing than the picture editor will find time for, and the other context the tracks will be worked by experienced post sound people who work the magic perhaps only someone like yourself (post and PSM) can have so much of an informed idea of. $600 to remove the frown from my face while I consider how much of a problem a shitty BG will be in post sounds worth it to me. No one wants to hire the guy who stops for every BG sound. In short, at the very least it is the audio equivalent of the LUTs used on set to monitor log footage.

I think it is possible that you are not familiar with how NR is done in post.  As I said, for your mix, rock on.  For the isos, don't tie my hands.  Audio NR is in no way equivalent to a monitor LUT for picture.

Share this post


Link to post
Share on other sites

Geez $600 dollars for a plugin could we at least get a free Nvidia graphic card for that price...

Share this post


Link to post
Share on other sites

I‘m really surprised we are basically having pretty much exactly the same discussion again which we had when Cedar DNS 2 was announced. It’s the same thing. We can’t do this again and again. We already know to leave the isos alone, but on the mix it may be ok.

Although I have a story (which I also told before) that taught me to leave this whole NR thing to post altogether. Listen to it on set, check it out, relax your ears. But do not record it anywhere. 

Share this post


Link to post
Share on other sites
33 minutes ago, Constantin said:

I‘m really surprised we are basically having pretty much exactly the same discussion again which we had when Cedar DNS 2 was announced. It’s the same thing. We can’t do this again and again. We already know to leave the isos alone, but on the mix it may be ok.

Although I have a story (which I also told before) that taught me to leave this whole NR thing to post altogether. Listen to it on set, check it out, relax your ears. But do not record it anywhere. 

I'd love a link to your story please!

Share this post


Link to post
Share on other sites

If you can record a mix with and without NR you are good. If I’m doing post to something and the NR has already been done well, I might just use that track but I don’t have to.

Share this post


Link to post
Share on other sites

This can be a good tool or a loaded gun like it’s been said. Just remember that if you opt to use it, probably best to only use it on your mix track and leave those ISOs alone.

 

Certainly useful in live scenarios, to check to see what can be done in post, and also WHEN the need for a clean mix track/camera scratch track, or clearer IFBs are necessary (although for the latter, I feel like allowing the director/client to hear exactly how noisy their location or camera might be, can actually help your case for going about production WITH should in mind and with less of a “fix it in post” attitude). 

 

And also remember that we are in this for the business, not to give things away. I know many non scrupulous folks out there will throw this in like they’ve already been doing for so many things without charging appropriately, thinking that this will give them another leg up. But really you’ll just be giving away the goat and giving production another thing to start expecting from everybody for free. 

 

So so let’s just try to remember that how you run your business will affect the industry for everyone else. 

Share this post


Link to post
Share on other sites

A small price comparison:  Izotope RX, a very versatile audio fix tool (VST AU AAX) with many useful modules is $400 for the standard version.  In a DAW you can use it on as many channels as you have DSP for.  

Share this post


Link to post
Share on other sites
9 hours ago, BAB414 said:

Is it safe to assume that whatever this algorithm is capable of can be matched in post?

It is safe to assume you can likely exceed it in post, because for at least a couple of reasons (there would be more too):
1) you've got many more variables you can tweak
2) you can rewind time and take a second go at it if you make a mistake

Share this post


Link to post
Share on other sites
14 hours ago, Constantin said:

Cedar DNS 2, which many of us use all the time on the cart, is learning permanently, too. That’s a good thing. 

 

Not necessarily - it is up to the user - you can turn it off and it will retain the last noise signature it grabbed. 

Share this post


Link to post
Share on other sites
9 hours ago, Philip Perkins said:

I think it is possible that you are not familiar with how NR is done in post. 

True.

 

Quote

As I said, for your mix, rock on.  For the isos, don't tie my hands. 

Hard to imagine a when I would apply this to an ISO in the production process.

 

Quote

Audio NR is in no way equivalent to a monitor LUT for picture.

I meant only in the context of a production preview as neither the NA or the picture LUT used for preview purposes on set will necessarily make it through to be heard or seen in the release.

Share this post


Link to post
Share on other sites

LUTs are kinda comparable in another aspect too: in that sometimes LUTs *do* get burned in, when they're going live to air, or with very quick turnaround or otherwise with minimal / no color grading in post (or if the DoP secretly doesn't trust the editor/director/client... and wants to burn in "their look" rather than risking it getting "messed up" later on down the road. It does happen).

Share this post


Link to post
Share on other sites
4 hours ago, soundtrane said:

 

Not necessarily - it is up to the user - you can turn it off and it will retain the last noise signature it grabbed. 

 

Yes, you are absolutely right, of course. But for our on-set scenarios it's usually best to keep on the learning mode. That's also the recommendation Cedar gave. 

Share this post


Link to post
Share on other sites
11 hours ago, BAB414 said:

I'd love a link to your story please!

 

It's a lot less exciting than it may seem. I think I recounted it here somewhere, or maybe on Facebook. Can't immediately find it. 

Anyway, I recorded the Cedar on my mix tracks in an extremely loud location. I had previously discussed this with the re-recording mixer, but another guy worked on the actual job and he wasn't aware. In the post-production meeting he listened to my Cedar'd mix and thought this doesn't sound too bad. He normally assumes for himself that he can improve the sound by about 50%, and so he thought ADR was not needed on this. When he actually worked on the material it turned out he could not improve the sound that much anymore, only by maybe 10%, because Cedar had already done a good job on it, but it still didn't sound good enough, so he called for ADR after all, which he found very embarrassing, because they had already made their calculations without it. So he was very friendly to me about it, especially after he heard that I had discussed this with predecessor and it was mostly a communication error. Still, to be on the safe side he asked me to not use Cedar anymore. I decided to agree to that, at least when it comes to recording it. Tells me a lot about who reads my sound reports, too. 

Share this post


Link to post
Share on other sites

that's an interesting aspect I wouldn't have thought of, thanks for the story.

 

39 minutes ago, Constantin said:

Tells me a lot about who reads my sound reports, too. 

 

true, seems sometimes like nobody reads them.

to be fair it might be that in this case the guy just got handed some sound samples to evaluate and didn't get to see the sound reports when making his calls.

 

 

Share this post


Link to post
Share on other sites

Pardon my ignorance, this is just a thought experiment. My experience with Lectro wireless is somewhat limited. I tried the various noise algorithms that they offer and think I prefer going with the 'Normal' mode, even if it is destructive NR so to say. Are most off you going with normal or without noise reduction in your Lectro systems? How would it differ to use no noise reduction on the lectro RX, but instead apply lets say 2db noise reduction on an 8 series?

Share this post


Link to post
Share on other sites
3 hours ago, Mattias Larsen said:

Pardon my ignorance, this is just a thought experiment. My experience with Lectro wireless is somewhat limited. I tried the various noise algorithms that they offer and think I prefer going with the 'Normal' mode, even if it is destructive NR so to say. Are most off you going with normal or without noise reduction in your Lectro systems? How would it differ to use no noise reduction on the lectro RX, but instead apply lets say 2db noise reduction on an 8 series?


From a Lectrosonics manual:

 

Quote
Smart Noise Reduction (SmartNR)
The wide dynamic range of digital hybrid technology, combined with flat response to 20 kHz, makes it possible to hear the –120 dBV noise floor in the mic preamp, or the (usually) greater noise from the microphone itself.
(To put this in perspective, the noise generated by the recommended 4 k Ohm bias resistor of many electret lavaliere mics is –119 dBV and the noise level of the microphone's electronics is much higher.) In order to reduce this noise and thus increase the effective dynamic range of the system, the R400A is equipped with a Smart Noise Reduction algorithm, which removes hiss without sacrificing high frequency response.
The Smart Noise Reduction algorithm works by attenuating only those portions of the audio signal that fit a statistical profile for randomness or "electronic hiss."

SmartNR offers significantly increased transparency over the sophisticated variable low pass filters used in previous designs. Desired high frequency signals having some coherence such as speech sibilance and tones are not affected.
 
The Smart Noise Reduction algorithm has three modes, selectable from a user setup screen: Off, Normal and
Full.
 
OFF - No noise reduction is performed and complete transparency is preserved. All signals presented to the transmitter's analog front end, in-
cluding any faint microphone hiss, will be faithfully reproduced at the receiver.

NORMAL (factory default) - Enough noise reduction is applied to remove most of the hiss from the mic preamp and some of the hiss from lavaliere microphones. The noise reduction benefit is dramatic in this position, yet the degree of transparency maintained is exceptional.
 
FULL - Enough noise reduction is applied to remove most of the hiss from nearly any signal source of reasonable quality, assuming levels are
set properly at the transmitter. This additional noise reduction comes at the cost of some transparency for low-level room noise, yet the algorithm remains undetectable under most circumstances


So I'd recommend leaving it usually on the default setting. 

And what SmartNR is aiming to do (removing the hiss) is quite different from what the purpose of NoiseAssist on the 8 Series is. (which has a broader goal to achieve)

Share this post


Link to post
Share on other sites

So does that mean SD and other manufacturers are on the way to no longer provide significant new feature in free upgrade but charge for "plug-in" rather?

Share this post


Link to post
Share on other sites
Just now, Fred Salles said:

So does that mean SD and other manufacturers are on the way to no longer provide significant new feature in free upgrade but charge for "plug-in" rather?

From what I was gleaming from the Sound Summit a few weeks back, it seems Sound Devices sees the 8 series as a long term platform with a lot of potential in terms of future capabilities down the road.  The literature on their site touts how there's a ton more horsepower under the hood in terms of processing power vs the 6 series.

Share this post


Link to post
Share on other sites
5 minutes ago, codyman said:

...it seems Sound Devices sees the 8 series as a long term platform with a lot of potential in terms of future capabilities down the road.  The literature on their site touts how there's a ton more horsepower under the hood in terms of processing power vs the 6 series.

Yes I hear that, and that is a selling point for them and for us to buy a new recorder, but this point drops down to zero if one has to buy every single new feature they come up with as "plug-in".

I mean I am dreaming of buying the Scorpio, but the idea that what they claim to be "a potential in terms of future capabilities" actually means a potential for even more costs does make me bitter somehow rather than cheer me up.

Share this post


Link to post
Share on other sites
Just now, Fred Salles said:

Yes I hear that, and that is a selling point for them and for us to buy a new recorder, but this point drops down to zero if one has to buy every single new feature they come up with as "plug-in".

I mean I am dreaming of buying the Scorpio, but the idea that what they claim to be "a potential in terms of future capabilities" actually means a potential for even more costs does make me bitter somehow rather than cheer me up.

I mean, it is so feature packed out of the box not sure what else you'd really want from a recorder at this point in time?  I like the idea of the 8 series in how as-is it is more than enough powerful / has all the features I'd ever need BUT down the road, if something new pops up in terms of "must haves", having to pay a pittance vs buying a whole new recorder sounds good to me.  I bought my 633 a few months after it came out and it has paid for itself who knows how many times over at this point and still keeps on trucking (and the surprise Dugan update they handed out a few years later was an awesome surprise too).

Share this post


Link to post
Share on other sites

The good thing with optional, payable, add-ons is that they are optional. You don’t have to pay for them. Let’s say down the road there’s going to be 10 plugins available for the 8-series. You can kit out your recorder in any combination you like and pay the price accordingly, or you can choose to opt out and pay the entry price for that recorder. 
The alternative would be to start off with a recorder which had a much higher entry price point, but all the later add-ons (which you paid for, but may not need) will be free. 
I would prefer the first model. And you never know if SD isn’t going to add another new feature free at some later point. 
But SD never promised this feature and I really don’t think you can feel treated unfairly for getting this feature at a good price. Many of us bought the Cedar DNS 2 for 4 times as much, but for only half the possibilities. 
 

I was well impressed when they added digital limiters to the 788T a while back, which it didn’t have originally. I don’t remember if they had previously advertised for that feature, but I would say that’s the kind of feature I would expect them to provide free, as that really completes the digital inputs. And that’s what they did then. But with the NoiseAssist that’s clearly a different situation 

Share this post


Link to post
Share on other sites

For me, it's not a question of fairness. They can do whatever they want with their products. It just changes how I see them as a company. Doesn't feel like the same old SD to me. Feels like the kind of thing other companies would do. But I get it. We live in a digital world. Makes perfect sense.

Share this post


Link to post
Share on other sites

If you want a new hardware add-on like the SL2 or the CL16, everyone will find normal to pay for it.

But in the case of software, one would want to get it for free ?

In both cases, engineers have to work, and be paid for it to design these plug ins or devices.  
If you look at Aaton, they just released a new version of the Cantar firmware with optional features : 
AatonMix, 600€ excl. VATDante+, 300€ excl. VATAuxiliary mix buses, 300€ ecl. VAT

DAW manufacturers use this model for decades and that’s not unfair.
The good thing is that SD opens his 8 series platform to other manufacturers. 
With the release of the V5.0, the Sonosax LC8+ is now compatible with my 833 and this is great !

Imagine in the future having 3rd party Plugins like Ambisonics or Schoeps double MS, or the ability to control Tentacle boxes or Audio ltd or Wisycom transmitters from the recorder via the Bluetooth Interface of the 8 series. 
I think we are just at the beginning....

 

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...