pcpaulius
Max Output Level: -90 dBFS
- Total Posts : 11
- Joined: 7/4/2009
- Status: offline
What Is Summing
Cause i found only one plain answer " Summing is the blending of audio signals (tracks) into a stereo mix" but then i heard people discussing the summing differences between different daws so i thought how can be any differences if it's just blending tracks into stereo? Can anyone explain what is involved behind the scenes of audio summing? Thanks in adv. 
|
mgh
Max Output Level: 0 dBFS
- Total Posts : 8594
- Joined: 5/10/2007
- Location: betwixt and between
- Status: offline
Re:What Is Summing
September 05, 09 1:55 PM
(permalink)
you're right, summing is just where all the audio ends up, whether it's a master buss or what, i agree that in modern daws it shouldn't matter. in hardware, different desks could be driven harder than other when summing to produce tonal variations, often going through a mastering compressor, so it would be different; i guess you can emulate that with plugs.. just don't clip in the digital domain...not nice!
|
Philip
Max Output Level: -34.5 dBFS
- Total Posts : 4062
- Joined: 3/21/2007
- Status: offline
Re:What Is Summing
September 05, 09 2:49 PM
(permalink)
Perhaps I'm off with the following: I've learned (in my mixes) that summing may produce combfiltering since skewed mics on an instrument (or vox) oft cause overlapping waveforms (5-25 msecs apart) when mono-fied. Such destructive interference is oft undesireable as much timbre can be lost ... and the summed resonance of sounds like the instrument is playing through an air conditioner (in my experience). Many producers prefer to leave things channeled in stereo ... and or bounce to stereo tracks ... instead of summing to mono. So I often resort to use a LT or RT mono signal instead ... instead of summing LT and RT down the middle. For me, its probably safe to sum: snare mic (top and bottom) samples, 50% of dimpro, hats, some distortion guitars (that are enhanced by combfiltering). Its probably not as safe (for me) to sum: EWQL, NI, Miraslov, and other meticulously stereo recorded samples Vocals, backing vocals, lead vocals. acoustic guitar (especially if there is a delay btw sides) samples that have stereo delays and reverbs orchestral, strings, section, and multi-instrument (and multi-vox) Again, JMO. Methinks an airport and/or elevator speaker sums all LT and RT channels
post edited by Philip - September 05, 09 2:51 PM
|
bitflipper
01100010 01101001 01110100 01100110 01101100 01101
- Total Posts : 26036
- Joined: 9/17/2006
- Location: Everett, WA USA
- Status: offline
Re:What Is Summing
September 05, 09 3:58 PM
(permalink)
"Summing is the blending of audio signals (tracks) into a stereo mix" I'd remove the extraneous word "stereo" from that definition. "Summing" - in a digital audio context - merely implies the algebraic sum of two or more sample streams.
 All else is in doubt, so this is the truth I cling to. My Stuff
|
AT
Max Output Level: 0 dBFS
- Total Posts : 10654
- Joined: 1/9/2004
- Location: TeXaS
- Status: offline
Re:What Is Summing
September 06, 09 0:11 PM
(permalink)
That is it for digital summing.
Analog summing means running stems or groups into an analog mixer instead of letting the computer crunch numbers. An analog summing mixer is a simple audio mixer w/ few (or none) of the extras a "real" mixer provides, the theory being that summing in the analog domain provides better sound while giving an easy means to use analog hardware during the mixdown - even if it is just some hardware for the stereo bus. Or, for example, you might digitally sum all the drums and send them through a compressor/EQ before being combined w/ all the other elements (and adding a hardware reverb to the vocal mix, etc.).
https://soundcloud.com/a-pleasure-dome
http://www.bnoir-film.com/ there came forth little children out of the city, and mocked him, and said unto him, Go up, thou bald head; go up, thou bald head. 24 And he turned back, and looked on them, and cursed them in the name of the Lord. And there came forth two she bears out of the wood, and tare forty and two children of them.
|
Jonbouy
Max Output Level: 0 dBFS
- Total Posts : 22562
- Joined: 4/14/2008
- Location: England's Sunshine South Coast
- Status: offline
Re:What Is Summing
September 06, 09 1:45 AM
(permalink)
'Summing' is one of those doing words, implied by the 'ing'... Therefore it's the process during mixing whereby the energy levels of individual audio streams, usually tracks, when combined arrive at the desired total energy output level of the project, the summed level. Nobody had yet mentioned what is being summed whether it be elephants or ice-creams, for me it is energy levels within a projects audio streams that produce the final 2-bus (in a stereo mix) totalled sum. Also see 'gain staging'... which is related somewhat as you could just sum to really high levels and just trim off the excess at the final bus by adjusting the input level which may not produce the nice result you are after. So simply and roughly put the goal for me when summing is to arrive at that desired summed output level along with a nicely balanced (relative levels of the source material) mix whilst avoiding gain levels making compensatory leaps and/or dips along the way.
"We can't do anything to change the world until capitalism crumbles. In the meantime we should all go shopping to console ourselves" - Banksy
|
pcpaulius
Max Output Level: -90 dBFS
- Total Posts : 11
- Joined: 7/4/2009
- Status: offline
Re:What Is Summing
September 06, 09 6:10 AM
(permalink)
thanks folks for answers
|
Jeff Evans
Max Output Level: -24 dBFS
- Total Posts : 5139
- Joined: 4/13/2009
- Location: Ballarat, Australia
- Status: offline
Re:What Is Summing
September 06, 09 6:28 AM
(permalink)
Here are a few things to also think about. When working in the digital domain you should be recording at 24 bit resolution. The mix buss is a difficult job to do and in order to combine two 24 bit digital signals you need at least 48 bit internal resolution to do it properly. So when you are doing a mix make sure you switch on the 64 bit double precision engine in Sonar.
Logic and ProTools are still only 32 bit applications and cannot perform the mix buss duties digitally as well as Sonar can. That is why often people use analog summing busses with Pro Tools or an analog mixer to avoid this problem. But Sonar does not need to so much.
This is one of the great features of Sonar. But even when you are doing your mixdown keep well clear of the 0db FS level. Do not attempt to master at the same time as mixing. It is unwise. You should leave time between mixing and mastering. After you have mixed import the stereo track into a mastering session and set up Sonar for that job. Then you can EQ, use Multiband Compression and limiting to get the levels up to where you want them to be.
Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
|
DaveClark
Max Output Level: -71 dBFS
- Total Posts : 956
- Joined: 10/21/2006
- Status: offline
Re:What Is Summing
September 06, 09 4:19 PM
(permalink)
Hi pcpaulius,
The most obvious difference between DAW's that *may* be heard after so-called "summing" is that different panning laws may be applied. Without knowing exactly what you are referring to by "discussing the summing differences between DAW's," it's hard to say.
Another possibility: There are many discussions out there which claim that all the audio engines in all DAW's are the same, and that this can be proven by a "null test," basically inverting the sign of all the values of a mix in one DAW and summing it with the values from another, predicting zero for all values. This is both false and fallacious reasoning. I hope you don't get mired in one of those discussions; they are absolutely pointless.
Regards,
Dave Clark
|
Kim Lajoie
Max Output Level: -88 dBFS
- Total Posts : 117
- Joined: 4/28/2009
- Status: offline
Re:What Is Summing
September 06, 09 11:10 PM
(permalink)
Actually Jeff, summing two 24-bit signals requires 25 bits in order to maintain resolution. And Protools uses a 48-bit mix engine to ensure there's enough headroom and 'footroom' for summing. They've written about it here: http://akmedia.digidesign...48_Bit_Mixer_26688.pdf -Kim.
|
Jeff Evans
Max Output Level: -24 dBFS
- Total Posts : 5139
- Joined: 4/13/2009
- Location: Ballarat, Australia
- Status: offline
Re:What Is Summing
September 07, 09 0:28 PM
(permalink)
Sorry Kim you have got it wrong. In the Pro Tools article it does not mention 25 bits anywhere. To quote from Bob Katz's Book 'Mastering Audio, the art and the secret 2nd Ed' on P57
Inside a digital mixing console (or workstation) the mix bus must be much longer than 24 bits, because adding two (or more) 24-bit samples together, then multiplying by a coefficient (the level of the master fader is one such coefficient) can result in a 48-bit (or larger) sample with every little bit significant.
So I am going to take Bob's word for it. I think he is much more knowledgeable in this area. I read an article that you also wrote somwhere on mastering and while some of it was OK, some of it was just plain incorrect.
Thanks for pointing out the Pro Tools article. I see they use a 48 bit mix engine which is certainly better. But Sonar has a 64 bit mix engine which could be argued as being even better still.
Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
|
mattplaysguitar
Max Output Level: -55.5 dBFS
- Total Posts : 1992
- Joined: 1/2/2006
- Location: Gold Coast, Australia
- Status: offline
Re:What Is Summing
September 07, 09 0:47 PM
(permalink)
Kim Lajoie
Actually Jeff, summing two 24-bit signals requires 25 bits in order to maintain resolution. And Protools uses a 48-bit mix engine to ensure there's enough headroom and 'footroom' for summing. They've written about it here:
http://akmedia.digidesign...48_Bit_Mixer_26688.pdf
-Kim.
Correct. Each additional bit doubles the resolution - thus increasing dynamic range by 6dB. (Dynamic range)/10=log10(resolution^2) ''| | as worked out from the link from Kim where, resolution = 2^(no. of bits) _| and my noise knowledge Disclaimer: The following is what makes logical sense to me, I have not actually read this anywhere: But, in 99% of case, the full 25 bits won't actually be needed. If you have a peak at 0dB on track A, and then you sum it to track B which also has a peak at 0dB at the SAME location as the peak in track A, then you will need the full 25bits of headroom. If you didn't, you would get clipping (or would you get a raised noise floor?). But in most cases, you might get a 0dB and a -6dB peak meaning not all 25 bits will be utilised - but, more than 24 bits is needed to do this - or you're gonna get clipping. Same kind of principle as having a 16 bit track, then saving it as a 24 bit track. An entire 16 bits of space goes unused. Correct me if I am wrong, I may well be.
|
Philip
Max Output Level: -34.5 dBFS
- Total Posts : 4062
- Joined: 3/21/2007
- Status: offline
Re:What Is Summing
September 07, 09 1:08 AM
(permalink)
Jeff Evans
Here are a few things to also think about. When working in the digital domain you should be recording at 24 bit resolution. The mix buss is a difficult job to do and in order to combine two 24 bit digital signals you need at least 48 bit internal resolution to do it properly. So when you are doing a mix make sure you switch on the 64 bit double precision engine in Sonar.
Logic and ProTools are still only 32 bit applications and cannot perform the mix buss duties digitally as well as Sonar can. That is why often people use analog summing busses with Pro Tools or an analog mixer to avoid this problem. But Sonar does not need to so much.
This is one of the great features of Sonar. But even when you are doing your mixdown keep well clear of the 0db FS level. Do not attempt to master at the same time as mixing. It is unwise. You should leave time between mixing and mastering. After you have mixed import the stereo track into a mastering session and set up Sonar for that job. Then you can EQ, use Multiband Compression and limiting to get the levels up to where you want them to be.
Thanks Jeff for your exceeding thoughtfullness. 1) I've never checked the 64-bit box until now (after years of mixing)! I had no idea of what it was for. Somehow you've convinced me to check it. I think I was worried about redundant CPU strain or something. 2) TBH, I don't really see your logic of mastering later (doing stuff on the master buss or mastering, later) ... when all the busses and stems are like little master busses. Perhaps you'll elaborate. Is it CPU straining or summing or something dangerous?
|
Jeff Evans
Max Output Level: -24 dBFS
- Total Posts : 5139
- Joined: 4/13/2009
- Location: Ballarat, Australia
- Status: offline
Re:What Is Summing
September 07, 09 1:18 AM
(permalink)
Mastering later is to do with giving yourself a rest from the mixing process. It has nothing to do with technical reasons. People try and do both things together and it is just a disaster when you do. After assulting your ears with a mix, mastering is the last thing you should do.
When mixing do not attempt to get levels higher or anything like that just stay well clear of 0db FS. Leave it a week after a mix. And by the way if you listen to your mix for a week you will hear everything that is wrong with it as well and you will want to change it.
Then have another week off and master. Then you just put on your mastering hat and you are hearing the music fresh like for the first time. You will be in a much better position to judge what EQ is needed, what compression and how much limiting to apply.
The reason why it is good to get a mstering engineer to do your mastering is they have never heard the music before so they hear it fresh and have a better idea as what to do. But in many cases we cannot afford to do that but my approach is the next best thing.
Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
|
Kim Lajoie
Max Output Level: -88 dBFS
- Total Posts : 117
- Joined: 4/28/2009
- Status: offline
Re:What Is Summing
September 07, 09 3:42 AM
(permalink)
Jeff Evans
Sorry Kim you have got it wrong. In the Pro Tools article it does not mention 25 bits anywhere. To quote from Bob Katz's Book 'Mastering Audio, the art and the secret 2nd Ed' on P57
Inside a digital mixing console (or workstation) the mix bus must be much longer than 24 bits, because adding two (or more) 24-bit samples together, then multiplying by a coefficient (the level of the master fader is one such coefficient) can result in a 48-bit (or larger) sample with every little bit significant. So I am going to take Bob's word for it. I think he is much more knowledgeable in this area. I read an article that you also wrote somwhere on mastering and while some of it was OK, some of it was just plain incorrect.
Thanks for pointing out the Pro Tools article. I see they use a 48 bit mix engine which is certainly better. But Sonar has a 64 bit mix engine which could be argued as being even better still.
25 bits vs 48 bits My apologies. I thought you were referring to adding two signals. Applying a 24 bit gain stage to a 24 bit signal results in a 48 bit signal, if precision is to be preserved. Mastering I'd appreciate if you could let me know where you thought my words on mastering are incorrect. I'm here to learn as much as anyone else is. Cheers, -Kim.
|
Kim Lajoie
Max Output Level: -88 dBFS
- Total Posts : 117
- Joined: 4/28/2009
- Status: offline
Re:What Is Summing
September 07, 09 3:46 AM
(permalink)
Philip
2) TBH, I don't really see your logic of mastering later (doing stuff on the master buss or mastering, later) ... when all the busses and stems are like little master busses. Perhaps you'll elaborate.
Is it CPU straining or summing or something dangerous?
I've written about this here and here. ;-) -Kim.
|
Jeff Evans
Max Output Level: -24 dBFS
- Total Posts : 5139
- Joined: 4/13/2009
- Location: Ballarat, Australia
- Status: offline
Re:What Is Summing
September 07, 09 6:24 AM
(permalink)
Hi Kim,
My apologies too. I did not realise that you had written two articles on mastering and the second article basically covers what I thought you had left out from the first article.
Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
|
Philip
Max Output Level: -34.5 dBFS
- Total Posts : 4062
- Joined: 3/21/2007
- Status: offline
Re:What Is Summing
September 07, 09 10:22 AM
(permalink)
Kim and Jeff, Thank you so much for your profound and helpful info! I realize we've hijacked this thread at this point. The Op probably needs this info. Kim, I read your blogs but they didn't seem to address my mastering question. OTOH, I have 'religious' reasons to master my own mixes, have read Bob Katz' book, and have considered it ... well ... illogical to have someone else mix the master buss (let alone arrange albums). Its like having made a cake and being afraid of adding the icing later ... to let another master your/my mixes (at any level). Plus my songs seem never finished anyway (at my current stage).
post edited by Philip - September 07, 09 10:30 AM
|
bitflipper
01100010 01101001 01110100 01100110 01101100 01101
- Total Posts : 26036
- Joined: 9/17/2006
- Location: Everett, WA USA
- Status: offline
Re:What Is Summing
September 07, 09 11:49 AM
(permalink)
I see they use a 48 bit mix engine which is certainly better. But Sonar has a 64 bit mix engine which could be argued as being even better still. Actually, Digi makes a pretty good case that their 48-bit engine is superior because it's all integer math. Instead of a floating decimal point, they slide an integer window up and down as needed. It's a clever scheme, but not as idiot-proof as SONAR's use of floating-point intermediate data.
All else is in doubt, so this is the truth I cling to. My Stuff
|
Kim Lajoie
Max Output Level: -88 dBFS
- Total Posts : 117
- Joined: 4/28/2009
- Status: offline
Re:What Is Summing
September 07, 09 6:27 PM
(permalink)
|
mattplaysguitar
Max Output Level: -55.5 dBFS
- Total Posts : 1992
- Joined: 1/2/2006
- Location: Gold Coast, Australia
- Status: offline
Re:What Is Summing
September 08, 09 9:22 PM
(permalink)
My 'masterpiece', however, remains the 6000-word tome I wrote for the KVR Mix Workshop last year:
http://www.kvraudio.com/forum/viewtopic.php?p=3305233
-Kim.
Oooo I must admit I very much like that song. I really love the gritty, rough, raw distorted sound on it all. Really like that snare sound. I shall have to give the whole article a read. I'm interested in how the mixing on this was done. There was a very Muse style about the mix though. It sounds like pumping, but also doesn't. All Muse songs sound like they are compressed in a way to achieve audible pumping, but it's still not pumping as such. I'm still undecided on whether or not I like it. I always hear this in Muse and it certainly gives them their own style, but it also sometimes feels like too much flatline - when it's not actually flatline. Could you give me any insight into what I'm talking about? I still have not worked out what I think I am hearing when I hear this.
|
Kim Lajoie
Max Output Level: -88 dBFS
- Total Posts : 117
- Joined: 4/28/2009
- Status: offline
Re:What Is Summing
September 08, 09 10:34 PM
(permalink)
mattplaysguitar
My 'masterpiece', however, remains the 6000-word tome I wrote for the KVR Mix Workshop last year:
http://www.kvraudio.com/forum/viewtopic.php?p=3305233
-Kim.
Oooo I must admit I very much like that song. I really love the gritty, rough, raw distorted sound on it all. Really like that snare sound. I shall have to give the whole article a read. I'm interested in how the mixing on this was done. You can read about the whole thing here: http://www.kvraudio.com/f...iewtopic.php?p=3321417
There was a very Muse style about the mix though. It sounds like pumping, but also doesn't. All Muse songs sound like they are compressed in a way to achieve audible pumping, but it's still not pumping as such. I'm still undecided on whether or not I like it. I always hear this in Muse and it certainly gives them their own style, but it also sometimes feels like too much flatline - when it's not actually flatline. Could you give me any insight into what I'm talking about? I still have not worked out what I think I am hearing when I hear this.
The individual channels were quite compressed, and the drum bus was lightly compressed. You can read about the drum bus compression here, with before-and-after audio examples: http://www.kvraudio.com/f...iewtopic.php?p=3276197 The drum bus compression might be what you're hearing as pumping. There was no mix bus compression, and only light (inaudible) limiting using in mastering. The other thing that might be giving the some a Muse vibe is the backing vocals I sung for the final chorus. ;-) -Kim.
|
DaveClark
Max Output Level: -71 dBFS
- Total Posts : 956
- Joined: 10/21/2006
- Status: offline
Re:What Is Summing
September 10, 09 4:23 PM
(permalink)
Hi all, A very tedious discussion follows, but perhaps something a few folks will find of interest.... mattplaysguitar:
Correct. Each additional bit doubles the resolution - thus increasing dynamic range by 6dB. Not necessarily. You could indeed use the additional bit to increase resolution, but alternatively you could use it to extend the range. Consider two measuring sticks, one a meter long, and the other a kilometer long. Each has 1000 divisions. The meter-long one has better resolution, but the kilometer-long one has a longer range. Similarly, if your normalization for 24 bit representations is such that you represent values from 0 to 1 (i.e. unsigned ints representing floating point values), adding an additional bit could represent the sign and allow you to represent values in an extended range from -1 to +1. However, anything that you represented in this way (e.g. differential pressure) would still have the same resolution --- in this example, you are extending the range to include negative differential pressures. Alternatively, if you *still* wanted to represent only positive differential pressures --- for example, from 0 to 1 microPa --- then you could alternatively use the additional bit to increase the resolution. Still yet again, you could decide to represent only positive differential pressures, but from 0 to 2 microPa, in which case you are once again extending the range and not increasing the resolution. Although one could examine the use to which additional bits are put, it may be simpler just to ask whether or not the additional bit decreases the smallest measurement of whatever it is that is being measured or simply represented. This is essentially the usual meaning of resolution; that is, resolution is not really an abstract mathematical thing, but something related to real-world measurements that involve some kind of units --- pressure, length, etc. Because of this, one should think about the smallest unit being represented, not merely the number of bits, when talking about resolution. Those who don't risk becoming confused or making mistakes. As I said, a tedious discussion, but it is something that a few of us do worry about, and for good reasons. Regards, Dave Clark
|
drewfx1
Max Output Level: -9.5 dBFS
- Total Posts : 6585
- Joined: 8/4/2008
- Status: offline
Re:What Is Summing
September 10, 09 4:58 PM
(permalink)
DaveClark
Hi all,
A very tedious discussion follows, but perhaps something a few folks will find of interest....
mattplaysguitar:
Correct. Each additional bit doubles the resolution - thus increasing dynamic range by 6dB.
Not necessarily. You could indeed use the additional bit to increase resolution, but alternatively you could use it to extend the range.
Consider two measuring sticks, one a meter long, and the other a kilometer long. Each has 1000 divisions. The meter-long one has better resolution, but the kilometer-long one has a longer range.
Similarly, if your normalization for 24 bit representations is such that you represent values from 0 to 1 (i.e. unsigned ints representing floating point values), adding an additional bit could represent the sign and allow you to represent values in an extended range from -1 to +1. However, anything that you represented in this way (e.g. differential pressure) would still have the same resolution --- in this example, you are extending the range to include negative differential pressures.
Alternatively, if you *still* wanted to represent only positive differential pressures --- for example, from 0 to 1 microPa --- then you could alternatively use the additional bit to increase the resolution. Still yet again, you could decide to represent only positive differential pressures, but from 0 to 2 microPa, in which case you are once again extending the range and not increasing the resolution.
Although one could examine the use to which additional bits are put, it may be simpler just to ask whether or not the additional bit decreases the smallest measurement of whatever it is that is being measured or simply represented. This is essentially the usual meaning of resolution; that is, resolution is not really an abstract mathematical thing, but something related to real-world measurements that involve some kind of units --- pressure, length, etc. Because of this, one should think about the smallest unit being represented, not merely the number of bits, when talking about resolution. Those who don't risk becoming confused or making mistakes.
As I said, a tedious discussion, but it is something that a few of us do worry about, and for good reasons.
Regards,
Dave Clark
Technically a db is a relative unit. For instance, 100db has no meaning on its own, only relative to another db figure. It doesn't have any fixed value, though audio has certain defined standards it's measured against (dBu, dBV, etc). drewfx
|
foxwolfen
Max Output Level: 0 dBFS
- Total Posts : 8256
- Joined: 3/29/2008
- Status: offline
Re:What Is Summing
September 10, 09 5:35 PM
(permalink)
Simplistically:
What is Analog Audio summing? Amplitude Modulation.
What is Digital Audio Summing? Error correction.
A scientist knows more & more about less & less till he knows everything about nothing, while a philosopher knows less & less about more & more till he knows nothing about everything. Composers Forum
|
DaveClark
Max Output Level: -71 dBFS
- Total Posts : 956
- Joined: 10/21/2006
- Status: offline
Re:What Is Summing
September 10, 09 7:30 PM
(permalink)
Hi Drew, Technically a db is a relative unit. For instance, 100db has no meaning on its own, only relative to another db figure. It doesn't have any fixed value, though audio has certain defined standards it's measured against (dBu, dBV, etc). Yes, that is correct. What does this "relative unit" have to do with what I posted? I hope you are not under the impression that I'm claiming that every measurement unit needs to have physical units associated with it. If you are posting something as a counterexample, you have quite seriously misunderstood what and why I posted --- and you are rashly and wrongly assuming that I am somehow technically grossly ignorant. Several folks here would be happy to convince you otherwise. The context of the discussion was Kim's post about adding a bit to "maintain resolution" and mattplaysguitar's comment about how to measure what the resolution is by simply counting the number of bits. Although one does "maintain resolution," adding the bit(s) was not done for that purpose. The purpose is to handle overflow. (One could just ignore it, and in some cases, this is fine.) This is a programming-oriented thing; I'm advising folks interested in representations to think about what they are measuring when discussing resolution, because I've seen a lack of information about this topic. My post also applies to resolution of db, but perhaps you didn't consider that? 0 to +1 dB, -1 to +1 dB, 0 to 2 dB. Regards, Dave Clark
|
drewfx1
Max Output Level: -9.5 dBFS
- Total Posts : 6585
- Joined: 8/4/2008
- Status: offline
Re:What Is Summing
September 10, 09 8:45 PM
(permalink)
Hi Dave,
I actually was agreeing with you (sorry if you misunderstand; I should have made that clear). I was just trying to give the "technical" definition of a db relative to what you explained.
It's easy for people to get confused about this, as they are used to always seeing db's talked about in the context of dBu, dBV or SPL, where 0db is defined to equal a fixed voltage or sound pressure level. In that case, you can calculate what voltage (or SPL) any +/-db value is (because it's relative to a fixed 0db voltage).
The point is, without 0db being defined, +/-10db could be a huge or tiny difference, which I think is exactly what you explained. It's essentially a ratio, not a fixed value.
Sorry for any misunderstanding I might have caused.
drewfx
|
Jeff Evans
Max Output Level: -24 dBFS
- Total Posts : 5139
- Joined: 4/13/2009
- Location: Ballarat, Australia
- Status: offline
Re:What Is Summing
September 10, 09 10:20 PM
(permalink)
I can sort of see what Shad is saying. It is a form of amplitude modulation but also Amplitude Modulation is something that goes inside AM radio transmitters, where an audio signal modulates a high frequency carrier wave. Analog summing could be described as voltages that are instantly adding and subtracting at any point in time to create a final sum voltage. I think digital summing is about more than error correction but also adding of two digital words at any point in time. Check out this article and you might have an idea. (and by the way dont feel bad if you find it hard going, it's pretty full on) This is a link that was mentioned earlier and a great article on it. http://akmedia.digidesign...48_Bit_Mixer_26688.pdf And digital summing is just as effective as analog summing if its done properly. Sonar gets off to a great start with its double precision engine. I believe the secret is in choosing 24 bit as the recording resolution and keeping things well clear of 0db FS even in your final mix. People make the mistake of attempting mastering for volume all at the same time as the mix. It is a separate process. By the way I have just finaished Bob Kats's book on Mastering Audio and another very interesting thing came out of the last few chapters. 96KHz sampling rate is also a bit of a waste of time and data too. 48KHz sampling rates are about as high as you need to go. They did some extreme listening tests and yes no one could pick 96K either. But 24 bit is very important though.
post edited by Jeff Evans - September 10, 09 11:01 PM
Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
|
drewfx1
Max Output Level: -9.5 dBFS
- Total Posts : 6585
- Joined: 8/4/2008
- Status: offline
Re:What Is Summing
September 10, 09 11:03 PM
(permalink)
Jeff Evans
That is the problem with these forums (and the Internet in general). There is plenty of great info but also bad or wrong info. I pity the poor person who is actually trying to find something out. You have to do a lot of research to verify what people are saying to be true or not.
Actually Jeff, I think the important thing is for "bad or wrong info" to be at least questioned by people who have knowledge and doubt, or who don't understand quite what the poster was getting at (as you did here). Hopefully through polite, respectful debate, the truth can become clear, facts can be distiguished from subjective opinions, and questions that have not yet been resolved can be recognized as such (I know, I'm being an idealist here). Personally, I always try (occaisionally creating some controversy) to correct what I believe are incorrect or misleading statements when I see them. I believe the worst thing we can do is to let what we believe to be a false or misleading statement stand unchallenged. Of course it goes without saying that I should always be called on my own BS as well. drewfx
|