Composers' Forum

Music Composers Unite!

I thought members of this site might be interested to know about SoundCloud's new Dolby based mastering system.  (Disclaimer: I have no relationship to SoundCloud except as a paying user, and I don't get any personal benefit from posting this.)

Basically, this feature allows users for a small fee per track to run any audio file they've posted to SoundCloud through their Dolby automatic remastering feature.  If you have an annually paid SoundCloud Pro Unlimited plan, you get three free masterings per month.  I've used my three free masterings for this month, and the tracks do seem to sound better, though most of my compositions use fairly limited instrumentations; a better test would probably be something like a symphony.

I'd be interested in any comments or reviews of this feature.

SoundCloud 's mastering information.

My mastered tracks:

Vella, de vos son amoros for flute, clarinet, and cello

Three Old Irish Airs

The Dream Of Lygdamis

Views: 129

Reply to This

Replies to This Discussion

 Interesting insights Dane. I have always liked vinyl.

Using a good cartridge/needle through a good system is a wonderful experience. Vinyl here in the states sort of ebbs and flows in terms of popularity. Most are using the path of least resistance which is often whatever is available. YouTube as I understand, has become quite popular for music.The audio on Youtube is one of the best. SC was a 128 mp3 uploaded. I have seen an HD designation on some tracks. Not sure what that really means.

I guess I'm not a purist in the strictest sense. I mean, a realistic mocked up orchestra is still a mocked up orchestra no matter how good it is. A real purist wouldn't accept anything mocked up.

To Jon's comment that he didn't want to be associated with electronic music. I get that. I also think though that a good representation will be more likely to bring real musicians in to play the music. Especially when there will be other demos that take great pains to be accurate who will be competing for interest. I don't claim to know exactly how that system works and maybe my comments on this are off base. In a nutshell it would appear that it works like this-

Signing of contracts and legalize>Submission to music library for consideration>Music is accepted> Composer is payed according to agreed contract.

Main questions would be, what are they looking for? What standards are they holding the material to for electronic submissions? Are they more notation centric or more audio centric?  Or is it both now?  I know what I would be looking for which might be something totally different than what they are looking for. It would need to be something playable by average well trained experienced musicians. It must be musically palatable to those with the interest. Here is where things could be very subjective. Do they like it or not? Do they look at credentials? Those with a track record? Most likely this is a consideration. Are there checks and balances? IOW how would you know if there is nepitism going on? Maybe the nephew of one of the judges is getting a lot of sales on the music library.They wouldn't want to necessarily buy a track from the local bag boy who does this on the side...or would they? One of the founders of Spitfire has no formal music education and has probably made more money than many trained composers...hmmmmm.

I'm a bit slow off the line sometimes and I missed that Jon said SC was offering three free masterings per month to SC members. Probably one of the main reasons Jon posted this. Thanks Jon for the FYI. I might try both home mastered and SC mastered to see if I can tell much of a difference. If the SC system is anything similar to others they should offer the user to select genre to determine mastering type.

I'm no authority on recording technology, but my understanding is that it's long been recognized that digital recording is more "lossy" than analogue, meaning that theoretically a vinyl LP should sound better than a digital CD, because the former gives you "all of it."  But there are two problems with vinyl.  First, the grooves wear down with use, losing sound information.  In fact, I remember reading somewhere that a significant percentage of the sound information in the grooves of a vinyl LP is lost after a single playing. Second, of course, no matter how carefully you take care of a vinyl record, it will eventually get scratches.

The ideal would be to invent a type of recording vinyl which would not scratch or wear out.  You'd think contemporary technology could do this, but for some reason the industry hasn't gone that way.  I suspect a large part of the reason for the prevalence of digital recording is just that digital can be shared over the internet, while vinyl is tied to the physical object.

Somewhat tangential but also somewhat relevant is the controversy which surrounds remastering old 78rpm recordings of classical, popular, and folk music from the past.  There are three schools of thought on this: 1) create a digital recording which sounds as much as possible like the 78 the first time it was played on a contemporary record player; 2) create a digital recording which sounds as much as possible like what the music sounded like in the studio while it was being recorded; or 3) simply transfer the LP in hand, scratches, clicks, pops and all, to a digital recording.  This last option may seem surprising, but some feel that the scratches, etc., are part of the history of the cultural object and shouldn't be discarded.

[I meant this to go as a reply to one of Dane's posts but somehow made it a reply to the discussion. ]

Yes, though I consider my software generated audio files as just demos, I try to make them good quality demos for what you might call advertising purposes.

Timothy Smith said:

[...]

To Jon's comment that he didn't want to be associated with electronic music. I get that. I also think though that a good representation will be more likely to bring real musicians in to play the music. Especially when there will be other demos that take great pains to be accurate who will be competing for interest. I don't claim to know exactly how that system works and maybe my comments on this are off base. In a nutshell it would appear that it works like this-

Signing of contracts and legalize>Submission to music library for consideration>Music is accepted> Composer is payed according to agreed contract.

Main questions would be, what are they looking for? What standards are they holding the material to for electronic submissions? Are they more notation centric or more audio centric?  Or is it both now?  I know what I would be looking for which might be something totally different than what they are looking for. It would need to be something playable by average well trained experienced musicians. It must be musically palatable to those with the interest. Here is where things could be very subjective. Do they like it or not? Do they look at credentials? Those with a track record? Most likely this is a consideration. Are there checks and balances? IOW how would you know if there is nepitism going on? Maybe the nephew of one of the judges is getting a lot of sales on the music library.They wouldn't want to necessarily buy a track from the local bag boy who does this on the side...or would they? One of the founders of Spitfire has no formal music education and has probably made more money than many trained composers...hmmmmm.

I'm a bit slow off the line sometimes and I missed that Jon said SC was offering three free masterings per month to SC members. Probably one of the main reasons Jon posted this. Thanks Jon for the FYI. I might try both home mastered and SC mastered to see if I can tell much of a difference. If the SC system is anything similar to others they should offer the user to select genre to determine mastering type.

Should be "simply transfer the 78 in hand ..."

Jon Corelis said:

I'm no authority on recording technology, but my understanding is that it's long been recognized that digital recording is more "lossy" than analogue, meaning that theoretically a vinyl LP should sound better than a digital CD, because the former gives you "all of it."  But there are two problems with vinyl.  First, the grooves wear down with use, losing sound information.  In fact, I remember reading somewhere that a significant percentage of the sound information in the grooves of a vinyl LP is lost after a single playing. Second, of course, no matter how carefully you take care of a vinyl record, it will eventually get scratches.

The ideal would be to invent a type of recording vinyl which would not scratch or wear out.  You'd think contemporary technology could do this, but for some reason the industry hasn't gone that way.  I suspect a large part of the reason for the prevalence of digital recording is just that digital can be shared over the internet, while vinyl is tied to the physical object.

Somewhat tangential but also somewhat relevant is the controversy which surrounds remastering old 78rpm recordings of classical, popular, and folk music from the past.  There are three schools of thought on this: 1) create a digital recording which sounds as much as possible like the 78 the first time it was played on a contemporary record player; 2) create a digital recording which sounds as much as possible like what the music sounded like in the studio while it was being recorded; or 3) simply transfer the LP in hand, scratches, clicks, pops and all, to a digital recording.  This last option may seem surprising, but some feel that the scratches, etc., are part of the history of the cultural object and shouldn't be discarded.

[I meant this to go as a reply to one of Dane's posts but somehow made it a reply to the discussion. ]

I have been on threads with both audiophiles and recording engineers who go back and forth over this very thing. It sometimes tends to be almost as predictable as the PC .vs Mac debates. Whether or not I decide to include myself in these debates depends entirely on how ornery I'm feeling that day. Am I looking to debate or to simply buy some popcorn and sit on the sidelines?

Today I'm not really feeling ornery. Maybe just more tired, but these kinds of discussions interest me greatly so sometimes I jump in knowing full well what might happen :)

Here's my .0000002 cents worth. I feel it's based on facts. The human ear is very limited compared to many animals ears. We hear somewhere in the range of 30-20,000 hz generally speaking. As we age that high end becomes less and less. Many people have trouble hearing anything above 15,000 hz and a more realistic range for older people is 5000-12000 or less on the high end.When the CD came along at the digital resolution of 16/44.1 it could be scientifically shown that this can make all of the ranges we can hear and then some. I often record in 24/48 which allows for even more more headroom or dynamics. This can also be a negative when mixing because it's a lot more to play with than some mix for and probably unnecessary for internet distribution. When CDs became common they went into all kinds of portable units and in cars. You can't bring records into those places. Many of the playback systems were smaller speakers with cheaper electronics which was a departure from the larger more pro setups used for records. As a result, I think some people faulted the format instead of looking at entirely why it didn't sound as good. Digital recording is all 1's and 0's but at such insane resolutions the human ear can't hear it. Imagine trying to see a pixel on a 4K TV screen from 10 ft. away. It's virtually impossible.

The real problem in my opinion lies in compression and the distribution now used. The common files seems to be an mp3. A higher resolution mp3 say maybe 320 actually sounds pretty good but not nearly as good as the original.

It has been said as a joke but it's actually true that today's online music works like this-

Mega star records at a multi million dollar recording studio using state of the art equipment and experienced engineers>Mastering is done at a top notch facility>High end master is produced  often 32/ 96khz or higher.>music is distributed > reduced to 16/44.1 wav file for upload to streaming site>16/44.1 file is reduced to a little ole' 128 mp3 file. 

The issue is they haven't found a way to replicate the quality of a master online as streamed mainly because of the file size and hardware limitations over the web. iTunes has made a higher resolution compressed file. It's still compressed.  Analog .vs digital isn't as much the issue IMHO. Options are limited so  compression online continues to occur.

 I see the dilemma as-

  • audio equipment designs made less capable due to cost constraints and what the consumer is willing to pay.
  • distribution methods that while effective at distribution trade off some quality.

Truth is that all recording removes elements and reshapes the sound. Nothing is as good as our ears. Recording to tape introduces distortion and cuts and boosts frequencies. Old studio equipment does the same. 

Recording is still a creative process. What you are aiming for is to keep the feel of the performance. Even a classical recording with the decca tree is nowhere near what it is like to be sitting in the hall listening.

Every step of the signal chain can add something or remove something - you need to know what each is doing to make it work.

Jon Corelis said:

I'm no authority on recording technology, but my understanding is that it's long been recognized that digital recording is more "lossy" than analogue, meaning that theoretically a vinyl LP should sound better than a digital CD, because the former gives you "all of it."  But there are two problems with vinyl.  First, the grooves wear down with use, losing sound information.  In fact, I remember reading somewhere that a significant percentage of the sound information in the grooves of a vinyl LP is lost after a single playing. Second, of course, no matter how carefully you take care of a vinyl record, it will eventually get scratches.

The ideal would be to invent a type of recording vinyl which would not scratch or wear out.  You'd think contemporary technology could do this, but for some reason the industry hasn't gone that way.  I suspect a large part of the reason for the prevalence of digital recording is just that digital can be shared over the internet, while vinyl is tied to the physical object.

Somewhat tangential but also somewhat relevant is the controversy which surrounds remastering old 78rpm recordings of classical, popular, and folk music from the past.  There are three schools of thought on this: 1) create a digital recording which sounds as much as possible like the 78 the first time it was played on a contemporary record player; 2) create a digital recording which sounds as much as possible like what the music sounded like in the studio while it was being recorded; or 3) simply transfer the LP in hand, scratches, clicks, pops and all, to a digital recording.  This last option may seem surprising, but some feel that the scratches, etc., are part of the history of the cultural object and shouldn't be discarded.

[I meant this to go as a reply to one of Dane's posts but somehow made it a reply to the discussion. ]

That's interesting Dane, can you tell us more about that? Is this a hobby thing or are you in the electronics business?  Sounds like you are seriously into vinyl.

Dane Aubrun said:

My interest in vinyl started with a commission to build an RIAA preamp circuit (designed by J Linsley Hood, the pickiest of audio electronicists)! Then I built one for myself. 

Hi, Jon,

I'm no authority either but I'm more familiar with recording technicalities (and electronics) than I am, midi! And I'd be way behind the jargon of a modern studio. I know very little about studio technology though I've built mixer strips and get the gist of it. 

But please excuse me discussing the endurance of vinyl. It simply isn't true that a "significant" amount of info is lost even with just one playing, if played using a halfway decent cartridge and stylus properly balanced, weighted and biased it'll take many plays before a significant loss is noted. It is also possible to cut some pretty high frequencies in vinyl - at least to 50kHz. What use is that you may ask?

It came down to a quadraphonic system known as CD-4 (Compatible Discrete 4 Channel - not this matrix stuff and all its decoding problems). As with other quad systems it had to be stereo compatible. The way it worked was that front left + back left' and front-right + back right info were cut into the groove as normal. A carrier frequency of 30kHz was also cut into the groove and the front L minus back L, and front R minus back R were frequency modulated about this carrier covering a band of about 18kHz to 42kHz, IIRC - correct me if I'm wrong. The signals were matrixed together to extract the discrete channels. I think the cut-off for the front+back normal groove was about 15kHz.

For this you needed a decoder and a cartridge capable of picking up these frequencies. It wasn't too difficult to cut the frequencies as engineers had already learned to cut masters at half speed. But it did lead to a great research into vinyl. Some very fine stuff came from it (taken up by the audiophile bunch and still used today on the 180gm records you can buy).

If there was a chance of this vinyl spoiling after just one play it would never have got anywhere. Sure, you were better off with a Shibata-tipped stylus, which is spear-shaped rather than rounded, and still elliptical along the smaller axis. It sat deeper with more contact in the groove so a slightly higher weight was permissible. I still use one now - Ortofon. But it took these developments to show exactly of what vinyl was capable. It could compete with CD, able to cope with frequencies down to 6Hz (as the famous Telarc 1812 Overture recording proved). CDs can't go below 20Hz, the limit is something about decoder ambiguity unable to determine if it's hit a DC signal. a no-no!

The modern vinyl will last a long time if cared for properly. It can still be scratched. 

== 

Yes, there's debate about remastering old 78s and it's usually done via intermediate digital or analogue. Some companies like Pearl believe they should give the listener exactly what came off the 78 (which leaves the treatment to you!! If you're good at constructing audio filters, fine, if not...ah well.... Others like Dutton apply quite a lot of processing. But as we've said before, start noise removal and you're sure to remove some of the music. I suppose Dutton have found a compromise or use a multi stage thing that attempts to distinguish between noise and music. Nimbus has yet another system to enhance and attempt to reproduce the sound that was being recorded in the pre-electrical era - and they've done pretty well with opera singers who were the mainstay of early acoustic recording.

It's quite an interesting subject but I'm just a dilettante, not an engineer.

Aside: My granddad gave me some ancient 78s from the early 1900s. Among them are a few by Adelina Patti who was born in 1843 and first sung in Opera in 1859. Listening to those discs kind of freaks me listening to someone born more than 150 years ago! recorded more than 100 years ago. EMI was called the Gramophone and Typewriter Company!

It was an era when true 78s were rare. Mastering in those days meant tinkering with the lathe cutting speed, reducing it a shade so the singer could sing a semitone lower key and things! 


Jon Corelis said:

I'm no authority on recording technology, but my understanding is that it's long been recognized that digital recording is more "lossy" than analogue, meaning that theoretically a vinyl LP should sound better than a digital CD, because the former gives you "all of it."  But there are two problems with vinyl.  First, the grooves wear down with use, losing sound information.  In fact, I remember reading somewhere that a significant percentage of the sound information in the grooves of a vinyl LP is lost after a single playing. Second, of course, no matter how carefully you take care of a vinyl record, it will eventually get scratches.

The ideal would be to invent a type of recording vinyl which would not scratch or wear out.  You'd think contemporary technology could do this, but for some reason the industry hasn't gone that way.  I suspect a large part of the reason for the prevalence of digital recording is just that digital can be shared over the internet, while vinyl is tied to the physical object.

Somewhat tangential but also somewhat relevant is the controversy which surrounds remastering old 78rpm recordings of classical, popular, and folk music from the past.  There are three schools of thought on this: 1) create a digital recording which sounds as much as possible like the 78 the first time it was played on a contemporary record player; 2) create a digital recording which sounds as much as possible like what the music sounded like in the studio while it was being recorded; or 3) simply transfer the LP in hand, scratches, clicks, pops and all, to a digital recording.  This last option may seem surprising, but some feel that the scratches, etc., are part of the history of the cultural object and shouldn't be discarded.

[I meant this to go as a reply to one of Dane's posts but somehow made it a reply to the discussion. ]

Thanks for the further interesting information.

I can't remember now where I read the claim that information is lost from a vinyl LP with the first playing.  It's possible that the loss they were talking about is only apparent if you compare before and after on some kind of oscilloscope, but isn't apparent to the human ear.

There exist recordings, which can be found on the internet, of readings by Tennyson, Browning, and (very brief) of Walt Whitman.  This last was supposedly recorded by Thomas Edison.  Its authenticity is disputed, but I tend to think it is real.



Dane Aubrun said:

Hi, Jon,

I'm no authority either but I'm more familiar with recording technicalities (and electronics) than I am, midi! And I'd be way behind the jargon of a modern studio. I know very little about studio technology though I've built mixer strips and get the gist of it. 

But please excuse me discussing the endurance of vinyl. It simply isn't true that a "significant" amount of info is lost even with just one playing, if played using a halfway decent cartridge and stylus properly balanced, weighted and biased it'll take many plays before a significant loss is noted. It is also possible to cut some pretty high frequencies in vinyl - at least to 50kHz. What use is that you may ask?

It came down to a quadraphonic system known as CD-4 (Compatible Discrete 4 Channel - not this matrix stuff and all its decoding problems). As with other quad systems it had to be stereo compatible. The way it worked was that front left + back left' and front-right + back right info were cut into the groove as normal. A carrier frequency of 30kHz was also cut into the groove and the front L minus back L, and front R minus back R were frequency modulated about this carrier covering a band of about 18kHz to 42kHz, IIRC - correct me if I'm wrong. The signals were matrixed together to extract the discrete channels. I think the cut-off for the front+back normal groove was about 15kHz.

For this you needed a decoder and a cartridge capable of picking up these frequencies. It wasn't too difficult to cut the frequencies as engineers had already learned to cut masters at half speed. But it did lead to a great research into vinyl. Some very fine stuff came from it (taken up by the audiophile bunch and still used today on the 180gm records you can buy).

If there was a chance of this vinyl spoiling after just one play it would never have got anywhere. Sure, you were better off with a Shibata-tipped stylus, which is spear-shaped rather than rounded, and still elliptical along the smaller axis. It sat deeper with more contact in the groove so a slightly higher weight was permissible. I still use one now - Ortofon. But it took these developments to show exactly of what vinyl was capable. It could compete with CD, able to cope with frequencies down to 6Hz (as the famous Telarc 1812 Overture recording proved). CDs can't go below 20Hz, the limit is something about decoder ambiguity unable to determine if it's hit a DC signal. a no-no!

The modern vinyl will last a long time if cared for properly. It can still be scratched. 

== 

Yes, there's debate about remastering old 78s and it's usually done via intermediate digital or analogue. Some companies like Pearl believe they should give the listener exactly what came off the 78 (which leaves the treatment to you!! If you're good at constructing audio filters, fine, if not...ah well.... Others like Dutton apply quite a lot of processing. But as we've said before, start noise removal and you're sure to remove some of the music. I suppose Dutton have found a compromise or use a multi stage thing that attempts to distinguish between noise and music. Nimbus has yet another system to enhance and attempt to reproduce the sound that was being recorded in the pre-electrical era - and they've done pretty well with opera singers who were the mainstay of early acoustic recording.

It's quite an interesting subject but I'm just a dilettante, not an engineer.

Aside: My granddad gave me some ancient 78s from the early 1900s. Among them are a few by Adelina Patti who was born in 1843 and first sung in Opera in 1859. Listening to those discs kind of freaks me listening to someone born more than 150 years ago! recorded more than 100 years ago. EMI was called the Gramophone and Typewriter Company!

It was an era when true 78s were rare. Mastering in those days meant tinkering with the lathe cutting speed, reducing it a shade so the singer could sing a semitone lower key and things! 

Hi Ingo,

Apologies not replying last night. We're going through a heatwave here. I was just flaked out. Our son who usually sleeps upstairs (we live in a chalet bungalow) had to sleep in the living room and he flaked out too so we just let him get his head down. No amount of fans is helping! 

But to reply, there's nothing special really. I've always been avidly interested in audio and electronics as hobbies. I occasionally design something but audio electronics have been so explored already that it's pointless thinking I could come up with innovations! J Linsley Hood (alas passed away now) had an enviable reputation as an audio engineer. His 80watt Mosfet Amplifier truly did live up to its reputation of being better than anything you could get commercially. I built many of his "projects". Then there was the Project 80 Synthesizer that used Curtiss chips saving masses of effort and testing.

Vinyl started with that RIAA commission. The recipient tried it out comparing the commercial pre-amp's RIAA with the Linsley Hood one simply by connecting the latter between his cartridge terminations and an Aux input on the amp. The difference was nigh unbelievable. The module used the best components I could get - telecoms capacitors, super low noise transistors and things. So I suppose I had to have a go myself! 

The other factor was my maybe misguided belief that interpretations by conductors who knew the composers probably met approval and so many of these works were not being transferred to CD. Things like Jensen conducting Nielsen and Barbirolli, Vaughan-Williams. Such LPs were available from places like discogs and with that kind of music were usually well kept. I've yet to buy a bad one! (Keeps fingers crossed!) 

I have a small collection of both vinyl and CDs. The technology behind CDs and digital recording is a bit different. Another story and it's easy to see why some post-production work is necessary to ensure the music/content fits the red book standards. Subjects of great interest to me that probably bore the pants off everyone else.

Not much of a story to tell otherwise, Ingo.

So...let's hope the heat subsides a little today so I can get back to composing and things!

à bientôt,

Dane


Ingo Lee said:

That's interesting Dane, can you tell us more about that? Is this a hobby thing or are you in the electronics business?  Sounds like you are seriously into vinyl.

Dane Aubrun said:

My interest in vinyl started with a commission to build an RIAA preamp circuit (designed by J Linsley Hood, the pickiest of audio electronicists)! Then I built one for myself. 

Reply to Discussion

RSS

Sign up info

Read before you sign up to find out what the requirements are!

Store

© 2020   Created by Gav Brown.   Powered by

Badges  |  Report an Issue  |  Terms of Service