The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
Tried summing NOT Impressed
Old 19th July 2005
  #91
Lives for gear
 
gsharp's Avatar
 

Quote:
Originally Posted by RKrizman
I believe that Bob O and others are making a claim about mix levels that has to do with the actual quality of the mix, not just how you hear it through your monitors.

-R
But how you hear it is your only means of assessing the quality of the mix. If you're doing something (ie; faders too high) to adversely affect the perceieved quality of the mix, wouldn't undoing that something then positively affect the perceived quality of the mix? What if Bob's client was retrieving 6dB of lost transient info from his mix by trimming the master back? That is definitely something that would make one perceive the mix as suddenly getting better.

So to reclarify, just wondering how much of the same perceived benefit (not counting the tone of a particular box) of OTB summing would be noticed if the same guys ran more conservative levels at the 2bus. The descriptions have been awfully similar and in the case of the Folcrom, the process is awfully similar. That's all.


g
Old 19th July 2005
  #92
Lives for gear
@wallace
kind of "academic" but I'm not in the business...
just remember some facts from my student days.
now the pro's will react unpredictably, maybe stay silent if a am actually onto something, because then it is about products and innovation, patenting and justifying, ... every inch of technical progress has a rat's tail thousands of pages long, full with pro and contra and nitpicking...
kudos to all the engineers heh

as about the level thing, literature is also rich, about floating point systems that scale without a problem, and fixed point formats that clip but are more predictable..
some digital "busses" emulate saturation, and people drive them hot, but OTOH in digital mixing it is possible to stay way below and not lose anything (except a few bits out of 64), but rather catch every transient perfectly, that can be cared of while mastering.
many opinions seem to say that the "right" way is to mix with a large headroom because the noise floor is far below analog.
Old 19th July 2005
  #93
Lives for gear
 
wallace's Avatar
 

Quote:
Originally Posted by gsharp
But how you hear it is your only means of assessing the quality of the mix. If you're doing something (ie; faders too high) to adversely affect the perceieved quality of the mix, wouldn't undoing that something then positively affect the perceived quality of the mix?

I was reading some Chris Lord Alge interview in Mix Magazine and he talked about the sweet spot on his board's master fader. The same thing is probably true for digital though I don't know the best way to find that out. I think that's what wer're talking about now. Here's what he said:

"It's all about maximizing the signal strength. You hit the tape machine a certain way, the tape machine is hitting the console a certain way, you're hitting the mix bus a certain way. You're at the sweet spot. All consoles have a pretty small sweet spot where it really sounds good. Well, everybody's thinking because they're digital, they don't have to worry about a sweet spot. Well, yeah, you do. Digital craps out so quick it's disgusting. In an all-digital console, the sweet spot is tiny — like a postage stamp. But there's a small window on any console where the headroom is right and where it really sounds the best. That's where I stay, in that window. - CLA"
Old 19th July 2005
  #94
Lives for gear
>In an all-digital console, the sweet spot is tiny

I think this is a rude generalisation. the particular setup should be able to change this extremely, by the type of (sum) plugins, and the data formats and settings of the mixing software. think of saturation emulators...

...
another idea is the sweet spot as a concept of IT usability, that should help the engineeer to find a focus in his search for the desired sound.
Old 19th July 2005
  #95
There is only one
 
alphajerk's Avatar
 

sweet spot in a digital setup is the lowest sound to the loudest.... it doesnt "react" like analog components.
Old 19th July 2005
  #96
Lives for gear
 
gsharp's Avatar
 

Quote:
Originally Posted by wallace
Well, everybody's thinking because they're digital, they don't have to worry about a sweet spot. Well, yeah, you do. Digital craps out so quick it's disgusting. In an all-digital console, the sweet spot is tiny — like a postage stamp. But there's a small window on any console where the headroom is right and where it really sounds the best. That's where I stay, in that window. - CLA"
Digital mythology from a guy who dumps every project down to 16 bit before he mixes it.

Why would someone who dis's digital and is worried about digital's "sweet spot" mix off a 16 bit machine? Could it be that the "sweet spot" is actually wider than all his analog gear, only he doesn't know it? Yes, the sweet spot between maximum level before clipping and clipping in infinitely small, but why try to overdrive a linear medium?
Old 19th July 2005
  #97
Lives for gear
 
RKrizman's Avatar
 

Quote:
Originally Posted by gsharp
Have you tried this? I just did and the bounced file is clipped like a mo fo upon inspection. I don't have Jam on this system, but the resulting file is squared off anyway. Turning it down isn't going to retrieve those peaks. Now if you turn down the master before you bounce, then yes all the peaks will still be there.
So you're clipping something within your DAW, not overdriving your d/a's . That's my point--it's two different things. Of course at any given time they could both be happening. But one will be imbedded in your mix forever, and the other is just a function of your current monitoring system. Both, of course, are undesireable, and both can be solved by trimming back your faders.

-R
Old 19th July 2005
  #98
Motown legend
 
Bob Olhsson's Avatar
 

If only digital clipping were simply embedded in the mix like analog clipping would be. Once you introduce funny numbers, all bets are off about the sound when it gets processed further!
Old 21st July 2005
  #99
Lives for gear
 
ulysses's Avatar
Quote:
Originally Posted by gsharp
Have you tried this? I just did and the bounced file is clipped like a mo fo upon inspection. I don't have Jam on this system, but the resulting file is squared off anyway. Turning it down isn't going to retrieve those peaks. Now if you turn down the master before you bounce, then yes all the peaks will still be there.
Are you sure? I haven't tried it, but I wouldn't be surprised if the clipping occurs before the master fader.
Old 21st July 2005
  #100
Lives for gear
 
gsharp's Avatar
 

Quote:
Originally Posted by ulysses
Are you sure? I haven't tried it, but I wouldn't be surprised if the clipping occurs before the master fader.
The clipping is at the master fader. if you turn down the master the clipping goes away. The attenuated master rescales the mix back down into the legal sample realm.

This whole thing is all anecdotal anyway, but I am just pointing out that the described benefits achieved by those going OTB (again, 'tone' of any particular summer notwithstanding) are awfully similar to the improvements described by the guys who simply tried running more conservative levels than they had been(either by trimming every fader or simply trimming the master) into their 2 bus.

g
Old 21st July 2005
  #101
Lives for gear
 
BradM's Avatar
I think it's important in all of these discussion to make it clear whether you are using a 32-bit floating point architecture or a 24-bit fixed one. I think a lot of you guys are referring to phenomena that occur specifically within ProTools or other 24-bit fixed math digital mixers. However, there are a lot of us using Cubase/Nuendo, which does not handle audio throughout its mixer the same way since it is 32-bit float. Some of the "trimming" and headroom issues are thus misinformation and myth for those users...especially when you read ignorant comments made by big name mixer guys. In Cubase/Nuendo you can run the channel faders as high as you want. They won't clip. The only clipping occurs when the digital signal leaves the master fader and hits your D/A. In that case you simply lower the master fader until the clipping goes away. If you are a Nuendo user you know that you can easily run channels into the red no problem. I've made mixes with individual channels running in the red and the master fader pulled way down. I've also done mixes with the channels having lots of headroom and the master fader at unity. Nuendo handles both just fine and I've never noticed any difference. In my experience the sweet spot is rather large since the 32-bit floating point architecture is rather forgiving if you crank channels into the red.

However, when I owned a Panasonic DA7 digital console (which was 24-bit fixed math), I did notice improvements by not clipping individual channels. And also by putting 24-bit dither on all my output busses.

Brad
Old 21st July 2005
  #102
Lives for gear
 
gsharp's Avatar
 

ProTools TDM is actually 48 bit fixed which is truncated (and dithered should you desire) to 24 (or 16 or whatever) at the output. You are correct that the 'native guys' have a lot more leeway within the mixer when it comes to throwing things around through auxes and the like.

Either way, if your mix suddenly opened up when you trimmed all your faders or your master, I recon you were hitting the 2 bus too hard.
Old 22nd July 2005
  #103
Gear Nut
 

Ok, let me try this again:


How often do you do a mix ITB with all of the "faders" full up - meaning, with all tracks (say 24) "outputting" *24 bits full scale*, not a reduced resolution...?


What's better:

24 24bit tracks down mixed digitally to 2;

24 24 bit tracks mixed full-scale analog down to 2.


?


You guys aren't giving any context! An 8 track mix, that requires all of the tracks to be set at equal level is going to be summed ITB differently than a 24 track mix - with say half the tracks set perhaps at 33% of "max level" ITB.

Not only that, but ITB strategies for "mixing"/summing could be different from program to program.

What sounds better (24 bit math)?:



Sum 4 tracks, add result + sum of another 4 tracks, +add result etc.?

Sum 2 tracks, add result sum another 2 tracks, add result, etc.?

Sum a group of 8 as above; sum a separate group of 8, then sum the two subgroups?


..because I bet, at low-levels different things get lost in the bit bucket when the math is done to reduce the numbers back down to a single 24 bit number. I would think from a waveform standpoint the second example would be like additive synthesis, the last more like FM, since in essence you're allowing 1 waveform to alter another, versus 1 being superimposed on a group and then that group being superimposed on another.... different strategies that's going to all work differently based on track counts and how those track's levels are set.

So I guess what I'm saying is I think you guys are discussing things both without a relative context to each other, as well as oversimplying because we not only don't know the internal mix programming strategies, you're also not specifying what different DAW's you're using.
Old 22nd July 2005
  #104
Lives for gear
 
gsharp's Avatar
 

I can only speak to ProTools since that's what I use. This concept of losing info because you pulled a fader down is simply incorrect up to a point, and that point is an extreme one. That's the idea of their 48 bit mixer. It allows all 24 bits to be retained while scaling the fader up or down within the context of a mix. You would have to pull a fader WAAAAYYY down bfore you start losing data. -96 to be exact. So, this concept of a mix with faders at unity being somehow better than a mix with faders down 20 or 30dB is simply erroneous. This linearity within such a broad range just might be the only 'beauty' of the digital world.
Old 24th July 2005
  #105
Gear Addict
 
Zooey's Avatar
 

Quote:
Originally Posted by gsharp
The clipping is at the master fader. if you turn down the master the clipping goes away. The attenuated master rescales the mix back down into the legal sample realm.
Forgive me if this is misstating what you're saying, but to address the topic further up the thread, it seems that when the clip indicator lights up on the master fader meter, you are indeed clipping the output, and not the mix buss.
Old 24th July 2005
  #106
Lives for gear
 
gsharp's Avatar
 

You are clipping the output of the mixbus. And usually your d/a as well.
Old 24th July 2005
  #107
Gear Addict
 
Ianneve's Avatar
 

I had the same feeling when I tried my first mix through a D2 buss and 8 ch of RME AdI 8 pro. It sounded like ****. No low end and just bad. So I realized it was all about the Convertors and bought 16 Ch of Mytek. Now,,, I'm in heaven and there is no way my ITB mixes could even touch my D2 buss/Mytek set up. It's killer...... But Samplitude (best sounding) and 3 UAD 1 cards with all the fixins doesn't hurt. :>) The Fairchild plug sounds better to me then any outboard compressor I've ever heard.. including my Tubetech CL1b. I just wish I could track with it. Anyway, you have to spend the $money$ on converters to mix outside the box. It's the only way....... Otherwise it's a waist.
Old 24th July 2005
  #108
Lives for gear
 
DirkB's Avatar
Quote:
Originally Posted by Ianneve
The Fairchild plug sounds better to me then any outboard compressor I've ever heard.. including my Tubetech CL1b. I just wish I could track with it.
My experiences are so the opposite that I'm just amazed. I have tried a lot of the plug ins (including the UAD cards) and nothing comes close to a Distressor or CL1B. But that's just another ones experience.
On the other hand, I never hear much difference between DAW platforms, so perhaps my ears are just not good enough ...

I'll by the CL1B from you for the price of another UAD card though!

Greetings,
Dirk
Old 24th July 2005
  #109
Gear Addict
 
Ianneve's Avatar
 

LOL. Very funny. No I LOVE the CL1b tracking vocals. But when I try it in the mix the Fairchild is better. I wish I could track with the Fairchild plug. That would be an interesting test. And there is a difference between Nuendo and Samplitude. That is why I switched. Samplitude being better. :>) To my ears. Which last time I checked were the only ones I had... And I know you could tell the difference between RME and Mytek converters. :>)
Old 25th July 2005
  #110
Lives for gear
 
Sofa King's Avatar
 

Quote:
Originally Posted by jbuntz
I think the extra conversion steps outweighs the benefits of analog summing alone.
Thats always been my concern.
Ive always treated conversionsas evil, and tried to avoid them as much as possible.

best
Sean
Old 25th July 2005
  #111
Gear Nut
 

Quote:
Originally Posted by gsharp
It allows all 24 bits to be retained while scaling the fader up or down within the context of a mix. You would have to pull a fader WAAAAYYY down bfore you start losing data.
How does 24 tracks of full-scale 24 bit data retain their full resolution when reduced to a 48 bit number?
Old 25th July 2005
  #112
There is only one
 
alphajerk's Avatar
 

24 tracks of 24bit full scale info is just noise.... ideas should probably be kept to realities...

the difference between 16bit and 24bit is 48db... quite a large window to be operating in w/o loss.

i think CLA's comment might be due to the fact he is running a 16bit system where the window is much smaller.
Old 25th July 2005
  #113
Lives for gear
Quote:
Originally Posted by hakkamacher
How does 24 tracks of full-scale 24 bit data retain their full resolution when reduced to a 48 bit number?
a simple engineering question...

sum up some 32 tracks of the same 0dB sine wave and you have used 5 bits of overhead. 2^5 = 32 as we should know.
you would use 29 bits then.
you mention only 24, and the tracks are not correlated. so you expect even much less amplitude. 48 bit is overkill and should always work perfectly.
Old 31st July 2005
  #114
Gear Nut
 

Quote:
Originally Posted by NeoVXR
sum up some 32 tracks of the same 0dB sine wave and you have used 5 bits of overhead. 2^5 = 32 as we should know.
I'm afraid I get what you're saying, but... I don't believe that's how it works...

Quote:
you would use 29 bits then.
you mention only 24, and the tracks are not correlated. so you expect even much less amplitude. 48 bit is overkill and should always work perfectly.
You're saying 24 24bit numbers added together is a smaller number than the largest 48 bit number...?
Old 31st July 2005
  #115
Lives for gear
no mistake to attend a digital class... heh
Old 1st August 2005
  #116
Lives for gear
 

Quote:
Originally Posted by Ianneve
LOL. Very funny. No I LOVE the CL1b tracking vocals. But when I try it in the mix the Fairchild is better. I wish I could track with the Fairchild plug. That would be an interesting test. And there is a difference between Nuendo and Samplitude. That is why I switched. Samplitude being better. :>) To my ears. Which last time I checked were the only ones I had... And I know you could tell the difference between RME and Mytek converters. :>)
Ianneve,

I am curious because I did tests a year or so ago and also found Samplitude much better. However I never did tests on individual tracks but rather just a stereo mix.

I know the summing is superior but in your opinion do the individual tracks have that same quality?

I did not like the interface at all and at the time I found the automation flaky so I kept Sonar (because my associate owned it already) as it had a passable interface.

I use Sonar to automate the faders and track & mix on analog console mostly.

Jim
Old 1st August 2005
  #117
Gear Head
 

Quote:
Originally Posted by Brad McGowan
However, when I owned a Panasonic DA7 digital console (which was 24-bit fixed math), I did notice improvements by not clipping individual channels. And also by putting 24-bit dither on all my output busses.
The Ramsa DA7 has a 32bit fixed engine..
Old 1st August 2005
  #118
Gear Addict
 
Ianneve's Avatar
 

Yes there is a difference period. And I loved the interface from day one and still do. I hated Nuendo from the first to last day I used it. Never tried Sonar probably never will. I'm very happy with Samplitude and no reason to change. The editing is incredable and the sound is the best so why use anything else.. :>)
Old 1st August 2005
  #119
Lives for gear
 

I guess I'll have to try the new version and compare to Sonar 4 as it seems some say it sounds better than v3.

Interesting that you like the interface. I only liked the mixer interface but then again I found the automation was kind of flaky. Maybe things have inproved interface wise.
I am kind of fussy with user interface. Protools has got the best with Digital Performer second. I haven't tried Nuendo though and many say it is a great DAW. But I don't really use DAWs like most. I record to HD24XR and use DAW for a little editing, volume/mute automation as well as some plugins (I don't have a lot of outboard yet). Its basically a fancy tape recorder for me.
Old 1st August 2005
  #120
Gear Addict
 
Ianneve's Avatar
 

Even more reason to use Samplitude. It sounds the best and has the best sounding plug ins included. The reverb is the best I've heard and the delay is nice as are the compressors. I have 3 UAD 1 cards so I don't use the compresssors much but Samplitude 8 has these new analog compressors that are close to as good as the UAD 1 stuff. I don't automate though. The editing in Samplitude is so good that I just edit my mixes. Works great and saves CPU. And it has none destructive mode so you don't damage your wav files. I just wish Samplitude had millions to spend on marketing like Steinberg did with Nuendo so people would know about it and use it. Once I heard it I never went back to Nuendo.. Plus the Steinberg people are mean. The Samplitude folks try harder and are really nice. And did I mention it sounds the best? :>) It's the best tape machine in the box you can buy....
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump