The way it’s generally done is with oversampling. That way, the limiter is seeing the intersample levels and limiting that rather than just the raw samples. A lot of limiters have this option. A lot of them call it “true peak limiting”
1 Like
Ah, ok, thank you. But does that actually work? Wouldn’t it be better to have a separate plug that just turns down each true peak?
Edit: After some research it seems that although oversampling may help to tame true peaks, it has a detrimental effect on the overall sound, aside from gobbling up your resources. I still like the idea of a plug that simply turns down (potential) ISPs.
Using LUFS you would know about dBTP (dB True Peak) which looks about intersamples (or so).
Using a meter with dBTP values, it becomes a bit awkward to see that the last limiter is set to -0.5 dB (for instance) and the meter is still clipping!
Also using the “Master for iTunes” plugin who does conversion on-the-fly and tell you if the end result will clip.
For the previous example of -0.5 dB, the plugin clips as well.
Here is an example using white noise:
I hope it could make sense 
Sorry if my ignorance offended you, Mr @bozmillar .
I thought most folks around here were polite and friendly.
I suppose there have to be exceptions. I’ll revise my opinion.
Good point. But he’s one mastering engineer, not every mastering engineer.
Yeah I have already checked that out. It’s still a limiter, not a level-tuner-downer. I’m envisaging a sort of Vocal-Rider-for-ISPs type of plug.
Ah, gotcha. Yes. That would be useful. You’re suggesting something that is clearly defined to deal with the ISPs, not a catch-all. Neat idea.
Yeah exactly. I use Orban’s standalone LUFS meter which also shows the number of overs and the db of the highest over. If it knows where the overs are, and by how much they are over, it reckon it’s a relatively simple step to actually turn those overs down to 0db.
C’mon Boz, you know you want to…
1 Like
If you are hearing a quality difference once you bounce down to 16/44, your converters in your interface are weak or you are doing something wrong. If I test you blindly you will fail. Trust me.
There’s no-one I know that can hear a 24/44 or 24/44 file converted to 16/44 when it’s done correctly… It’s digital…it’s perfect and there’s no way you can hear a difference. You’ll never sell me on that. You are degrading your audio quality if you are hearing a difference or you need a new interface. By the stuff I’ve read from you, it’s easy to see why this may be happening. 0dB while shooting for commercial loudness will do it everytime. But don’t listen to me…you know better than I.
Anyone claiming to hear this…let someone else press the buttons and throw wild cards in as well, as you write down which file you think is 24/48 and which is 16/44. You THINK you can tell the difference…you’re looking at it and probably did something to make the 16/44 sound worse. I sure can’t tell a level matched 24/48 that was bounced to 16/44. Make sure when you set up the test, that it’s apples to apples and not something where there is an artifact that let’s you know which is which.
Now if I record at 16/44 and then record the same piece 24/48, yes I can tell and pass this test every time. But to record 24/48 and take that same file to 16/44…noway I can hear it. Not with the good gear I have.
You’ve not heard any reasonable argument to record at 24/48? I just told you in my last post. More headroom, better plugin processing and more clarity. That’s not a reasonable argument…that is fact. Do the research. For someone that claims to be so open minded, you sure seem to be stuck in your ways especially when you have a guy here that’s been doing this for over 35 years trying to help you. I shall not continue…do it your way.
So you talk about these numbers and how you want to compare to commercial releases yet you claim you’re not into the technical aspects and can’t understand what I’m talking about when I said try it my way? Commercial releases don’t have much by way of dynamics, yet you talk about dynamics and losing -0.3dB.Lmao! Commercial loudness sounds like shit! No wonder why you and I are not and will not see eye to eye. I really am starting to believe you want to argue with me. So I’ll stop here before it gets to that point and leave you with this…
You’re too set in your ways to see things any other way than you see them. That’s all well and good. I respect that. But you can’t say you want to learn something and not want to get into the digital technicalities. That’s being a bit closed minded in my opinion.
Next, you’ll probably never get a plugin to know how much peak to take off. It can be done using either a waves L-2 or a PSP Xenon. But it will never be able to tell how much of the peak is a peak and how much is actually a GOOD transient that shouldn’t be altered or altered as much.
Boz could probably create something custom…but I can’t see how we could stop it from overdoing it. Sometimes a transient and a peak have the same level. The transient may need a lighter decrease where a nasty peak may need additional level decrease.
This is why I do this stuff by hand. I have full control over the material at all times. The day we give up on one finger press technology is the day people actually learn something. I’m all for working smarter not harder, but we already have so much stuff that people rely on, that they aren’t even engineers at all.
Any monkey can load up ezdrummer and traxx or ozone…and then they think they are actually someone with a valid opinion in the audio field. That part has always cracked me up. If it’s all that easy and everyone knows everything…please go in business and show the world your skills.
Kinda like me…I’ll sit here and tell people how I do things that bring insane money into my business and they’ll argue with me to the point of making me walk away.
Do you listen to the guy with the big mouth doing nothing or do you listen to the guy with the big mouth that is successful that can back it up? Do you listen to the guitar teacher or drum teacher that knows theory but can’t play good enough to walk the walk, or do you listen to the guy that’s kicking ass and taking names? That’s a no brainer for me. I don’t even need to be open minded for that one.
But I digress. …for I am just a grain of sand which is about what my opinion is worth in reality. 
So here is the required plugin!! 
2 Likes
You KNOW I trust you Danny. Always have, always will.
No, I can definitely tell a difference. I didn’t compare two separate files, I bounced a 48kHz project to 44.1kHz and re-imported the file back into the project so that I could A/B the rendered file against the project. (I sent the imported file to an unprocessed output of course). When I did the same, but rendering the file to 48kHz, I couldn’t hear any difference.
Further, I did it the other way around too, rendering a 44.1kHz project to 48kHz and re-importing the file. My conclusion is that when the sample rate is changed, there is a difference in the end result.
I couldn’t hear any difference when the bit depths only were changed.
So I’m just reporting what I have tested. Having re-read my original post it does sound a bit curt, so I apologise for that, but I’m not really saying anybody is right or wrong, just that I can hear a difference when a project is rendered to a different sample rate.
That’s not what I said Danny. I have already told you that I record at 24 bits. I said that I have not heard a reasonable argument to record at 48kHz. Maybe I just haven’t heard enough arguments, but as far as I know, the only reason to record at 48kHz is that it is the standard for TV and movies.
You could help most by reading my posts properly before shooting me down in flames for something I didn’t say. FYI I have been in the business as long as you[quote=“Danny_Danzi, post:30, topic:1289”]
Commercial releases don’t have much by way of dynamics, yet you talk about dynamics and losing -0.3dB.Lmao!
[/quote]It depends which commercial releases you’re talking about I guess. Not sure why you would think that making every effort to conserve dynamic range is laughable.[quote=“Danny_Danzi, post:30, topic:1289”]
I really am starting to believe you want to argue with me.
[/quote]
I don’t know why you would say that. Then again I don’t know why you would flame me for something that I didn’t say.[quote=“Danny_Danzi, post:30, topic:1289”]
but I can’t see how we could stop it from overdoing it. Sometimes a transient and a peak have the same level.
[/quote]
I’ll take your word for that seeing as there doesn’t appear to be such a plugin - but I would still like to give one a try, to see what it does, and what effect it has.
I run my own recording studio and I’ve been in the business since about 1983. Admittedly I don’t “make insane amounts of money”, but I get by.
Man, I didn’t understand it because your English wasn’t clear, not because I don’t know enough about the subject. Sure, I said it was because I’m not technically minded but that was because I didn’t want to upset you or cause a problem.
If recording at 24 bit/ 48 Khz allows more headroom and gives better plugin processing & clarity, does that mean recording at 96 Khz or 192 Khz is even better?
Yes Sir, I am the plugin! Hahaha!
That’s kind of you. Certain things, yeah, I’d bet my farm on. Other things…quite a bit can be considered subjective. 
That’s not how you do it. As I said, apples to apples or you will not be be doing a fair comparison. The other way around, you upsampled. Not the same nor is that a valid test for anything. You’re not doing it right to be able to tell any differences.
Actually, that IS what you said:
[quote=“AJ113, post:19, topic:1289”] I haven’t heard any reasonable argument to record at 48kHz.
[/quote]
And you mentioned MY English wasn’t clear later in your post? See this is the stuff that makes me think you want to argue.
Shooting you down in flames? Where did I do that? You claim to be open minded, yet you have an answer for everything I say showing me, you have your own agenda. I don’t wish to mess with that agenda. When someone listens and is open minded, they don’t handle it the way you have.
Here’s another thing to consider. You’ve been in business as long as me, but how much do you know about mastering? The time you spend in this business means nothing. You actually have been at it longer than me if you opened in 1983. I salute you for that. But what has that got to do with what we are talking about here? If you are a tracker/mix/producer type guy, that doesn’t mean you know about mastering or the manipulation of such practices You’re talking about something here that you DON’T have as many years as me doing.
That said it doesn’t mean I’m better than you or that I am the ultimate master guy. I am not. But I run a business that has never advertised and I’m booked solid into 2018. That doesn’t mean I’m good…it means other people think I’m good and God has blessed me with people that believe in me. I’ve never done a gold record or earned a Grammy. But I’ve worked with and under people who have. I have a pretty insane client base of people you have probably heard of that call on me just for my ears. None of it makes me better than anyone…but it makes me knowledgeable enough to argue and of course…on any given day, I will and can back up anything I say past or present.
“Give me some headroom” was the title right?
I gave you that in one sentence actually…“record in 24/48”. But…there’s an argument for that too. So where do I go from here, AJ? It’s cool to disagree. I’m fine with that. I respect that. But when someone tries to add something that can make a difference for the better and you sort of give an answer with an answer, it doesn’t really help your cause any.
As I mentioned in honesty…some things are justified as hype and bullshit. Other things, we do because the big guys in the industry say we should. Other things…make sense because it’s just the right way to do them. Like I said 4 posts ago…do what you think is right. If you have great results and a happy client base, you are right where you want to be.
Quick example: I do something in my mastering chain that Bob Katz hates…but he can’t tell me where I’ve failed. I save myself a step in the process. He wants to argue with me, and will try his best…but at the end of the day, he can’t justify an audible change enough to prove me wrong. In that instance…I stick to my guns even though he is one of the high end mastering guys and truly knows his stuff. It’s like anything else. When you learn from someone, you take what they taught you and sort of do your own thing with it. Whatever works, just works.
It;s laughable because everything you are talking about goes against the ways you are going about them. You mentioned dynamics, but you believe in mastering to 0dB. You are not leaving yourself room for anything. But because your meters aren’t clipping and you have a volume boost that stops you from having to turn up your volume knob, you think you have the answers. That’s how you come off.
Listen hard to those commercial masters. You do not hear all instruments at all times. They take turns taking back seats because of side chain compression because the engineers don’t know how to deal with frequency masking or choose not to. Then listen to their snare drums. The crack is gone…every hit sounds the same…it’s pure garbage. It’s laughable that you would mention -0.3dB making a difference in your world instead of giving yourself enough processing space to stop your volume cap from happening.
I said it because you have an answer for everything. No matter what I say, you have your way and that’s just the way it is. As I said, I respect that…but to me, you’re a bit too stuck in your ways to see reason. Where did I flame you?
See? Prime example. I’m sitting here telling you how to improve your audio and that no such plug will do what you can do by hand (also, the by hand method will get your volume as loud or louder than St. Anger, with commercial loudness if that is your wish, You will obliterate it) and you still want to try a plug that doesn’t exist?
quote=“AJ113, post:33, topic:1289”]
It’s not closed minded, it’s just not an area that I’m interested in. I need to know enough in order to provide a service to my customers but it’s not something I get excited about. Everybody’s different.
[/quote]
Not interested…why post this thread then? See, now you’re challenged by someone that is interested and is living this that runs a pretty killer business. This stuff isn’t stuff I get excited about either. But if clients are to be taken care of, it is my job to know MORE than just a little to get by, I have to know enough to justify taking their money and giving them the advice and info they need.
I explained it just fine. I talked about your final mix coming out at -3dB. Then, processing in your mastering chain after so that the final level is -0.3dB. If you do it right and then compare to one of your 0dB masters, MY way should be louder in volume as you turn up the knob. The 0dB master is already capped out. The louder you go, the more it will distort. How was that unclear? If you don’t know enough about the subject, you have to sort of be careful what you say, don’t you think?
You’ve already caught yourself being “curt” can’t you see how I can take some of this stuff as you wanting to be argumentative? I’m trying to help you but you call it flaming. Flaming to me, is name calling or personal attacks. I’ve said I felt you were too stuck in your ways to learn anything from me, and seem to be a bit argumentative. That’s not flaming…I’d never do that unless you attacked me first.
People seem to think so. I’ll give you my take. For rock, metal, current rock country, R&B or anything where we have sonic sounds, there is no reason to go past 48 in my opinion. Now, the headroom and the plugins…that remains true. The more tracks you have, the more the higher sample rate can help you. In one of my posts to AJ, I mentioned how recording at 16/44 sounds restricted to me as well as mixes possibly having frequency masking plague them more easily. Add more tracks in, and you better believe this will happen.
With higher sample rates, you gain good stuff from them if you have loads of tracks and most of all, mic’d instruments. Most of us these days are using direct recording as much as possible because we don’t have the money for an extensive mic locker, or live in an apartment etc. So, in situations like that, the higher sample rates are NOT going to make a huge audible difference to people in those realms. I’ll give you a scenario in which this can really shine…
I recorded a school orchestra one time. I was recording them at 24/48 for the entire session. They wanted to try a piece they were working on just to see how it was coming along. They knew they would tank eventually. So we recorded it and they stopped about 2 minutes in due to a screw up. They wanted to do it again. I decided to switch my clock to 96.
When I was totally done with them and got the tracks back, 24/96 was instantly noticeable. The reason? Mic’d instruments that were NOT sonic in nature. When you mic orchestra instruments per section and capture the room, it’s a totally different environment. Remove anything distorted/degraded and your mix opens up in a way that can’t be described. Distorted guitars, driven keys etc are degraded signals…no matter how good they may sound. Unless you are mic’ing the cab and have mic’s in the room going on to increase the sound size of the instrument while making it a little more organic, higher sample rates are just a waste of hard drive.
The other thing to consider is…a good interface will not show you in black and white that the difference is grand. Good interfaces are good because their converters will not let you down no matter what you use. So, there are some yes and no results with higher sample rates. 192…I think is the biggest load of hype and marketing ever. The pro tools guys are hell bent on recording like that or by using 96. It’s simply not needed unless you are faced with some of the stuff I have mentioned. And even there…24/48 handles all that stuff quite well.
So to recap…use loads of mics, have loads of tracks (over 60 playing simultaneously) use organic instrumentation that ISN’T loaded with distortion, and the bigger sample rates will make a BIG difference. The other difference you may hear…interfaces with converters that are consumer grade. All my colleagues with killer interfaces cannot tell any differences other than 16/44. Some swear they can…but they never pass a blind test of someone else pushing the buttons. But cheaper interfaces…yeah, they will make you think you are hearing things better because their lower sample rates are actually lower in quality. As soon as you pump them up, that difference you are hearing is actually the interface performing a little more like it should. :).
.
Interesting stuff Danni.
Thanks for explaining !
I’m going to assume that you’ve made an error, because otherwise it looks like you’re deliberately looking for a fight.
You said
It’s not what I said. I didn’t say anything about 24 bit. YOU added the 24 bit part.
I know about headroom. I record at 24 bits like you. I already told you that. Three times now.
I said I have not heard of a good reason to record at 48kHz. You even quoted me saying that!
So, to clarify, I record at a bit depth of 24 because of the headroom, and a sample rate of 44.1kHz, because I almost always render to 44.1kHz, and because I haven’t heard a good reason yet to record at 48kHz. OK, you say that some plugs work better at 48kHz, that may be a reason to consider 48kHz. I’ve checked that before and my research was inconclusive, but I 'll check it again.
Agreed, hence the thread, hence the questions.
No, it’s “Give Me Some Head…room.” It’s a play on words. You’re a man of the world Danny, I’m sure you know what’s going on in that title. It’s not accurate as a title. The thread is actually about trying to understand exactly why people master below 0db.
It’s not unclear, it’s clear, because you used different phrasing, as I requested. I’m very very sorry for not understanding the previous version.
Because I still need to learn regardless! I aready said that!
I’m very disappointed that I come across to you like that. The exact opposite is true. I’m inquisitive, I’m asking questions. I’m always looking to improve my methods, my workflow, take it up a notch. I have no preconceptions, if I see a better way of doing things, then I change - no problem.
My logic seems fine to me, but I’m sure you can point out where I’m going wrong:
I want the rendered file to sound exactly the same as the project. The only way I can think to test that accurately is to import the rendered file back into the project and do an A/B comparison. If the rendered file doesn’t sound exactly the same as the project, then something has happened in the rendering process. That ‘something’ has proven to be when sample rates are converted as part of the process.
I won’t argue with you. I didn’t flame you and I.wasn’t discourteous. I’m arguing? Brian asked me to give my opinion in this thread and so.I did to the best of my ability. You see things a different way and I.said 4 posts ago that I accepted and respected that.
I showed you the areas where I thought you were wrong, and you proceeded to have an answer for everything I’ve had to say while disagreeing and repeatedly accusing me of flaming you.
You asked why we use lower levels…I told you. Your research and personal observations tell you different. But you asked a question…I answered it in my first post. You not accepting my answer or wanting to believe it is something for you to decide on. I’ve provided a lot of information in this thread. Hopefully some of it will help others.
2 Likes
Ooh ee. Pots and kettles, I fear. Calm down man. This is meant to be a friendly forum.
Been reading back through your stuff and you do seem to try and bait people by being argumentative.
IMHO
I don’t have the time or inclination for this playground stuff, I certainly don’t need to be told to calm down.
I posted a comment by Stephen Massey who effectively (to me at least) dismissed the practice of mastering to -0.3db. I was hoping to get a convincing counter claim. So far that hasn’t transpired. Danny says drop to-0.3db because otherwise you are ‘pushing things too far’. For me, that’s not enough reason to change. I want to know what that means. It’s not a reasonable explanation to me. I want to know why he (and other people) only drop to -0.3db when it’s clear that maximum ISPs frequently go to 2db, 3db and even above, and you’re almost certainly not going to hear any clipping created by a 0.3db ISP.
These things vex me. I want answers. If you think I am argumentative I agree. Argument is good. Debate is good. It helps us dig out the answers. Throwing your toys out the pram because you perceive that someone doesn’t agree with you is not good. It doesn’t resolve anything.