Tuesday, December 20, 2011
Wednesday, December 14, 2011
Monday, December 12, 2011
Still monitoring on modern moving coil loudspeakers? Perhaps you need some vintage electrostatics to really hear what's on your recording?
You might think that the photo shows a pair of electric radiators. But no, they are actually some of the finest loudspeakers ever designed.
And the design dates from 1957!
I have no interest in vintage equipment for its own sake, and I'll use whatever sounds good, whatever its age. I tell you that because I don't want you to think I'm saying these speakers are good because they are old. That would be nonsense. They are good because their design is amazing. And, well-maintained of course, their age is immaterial.
The photo shows examples of the Quad ESL 57 electrostatic loudspeaker. '57' because the design dates, as I said, from 1957. If you want something more up-to-date then there's also the ESL 63 to consider :-)
Now, Question Number 1 is why are these speakers so good?
The answer to that comes from another question - why are moving coil loudspeakers so bad? (Almost all loudspeakers are of the moving coil design. You have some right there in your studio, and your living room, and your car, and tiny ones plugged into your iPod.)
The problem with moving coil loudspeakers is mainly to be found in the low-frequency drive unit, or woofer. The diaphragm has to be big to shift a lot of air. It has to be light so it can move easily. It has to be stiff so that it doesn't bend and produce distortion. And it has to satisfy all three of those conflicting requirements, which it can't.
Modern moving coil loudspeakers are amazingly good, but there is always the problem of transmitting the movement of the voice coil in the center of the drive unit, all the way to the edge of the diaphragm, without the diaphragm bending. This problem is inherent to the design.
But the diaphragm of the electrostatic loudspeaker can be driven over its entire surface area. It can be extremely light, yet still not bend. The result is a distortion-free, airy and accurate sound. The motto of Quad, the manufacturer, was and continues to be "The closest approach to the original sound". With just cause.
So now for Question Number 2 - If electrostatic loudspeakers are so good, why do we still use moving coil designs?
One answer to that is that it is cheaper to produce moving coil loudspeakers. Or to put it another way, electrostatic loudspeakers can't be produced cheaply.
Another is that electrostatics don't handle high levels and lots of bass so well as moving coils.
Another is that they are bi-directional, so you have to consider where the sound from the rear goes.
Another is that they need mains power.
Another... Well that's probably enough. They are simply not as practical as the moving coil.
Although not as accurate, many people prefer the 'speakery' sound of the moving coil design.
I am to an extent eulogizing the electrostatic loudspeaker, and it is a fact that I don't have them in my home. But I certainly do recommend you take any opportunity you can to listen to some. You will be amazed how non-speakery loudspeakers can be!
Now for some more of those lovely photos from significantsound...
Publication date: Sunday December 11, 2011
Author: David Mellor
Thursday, December 8, 2011
Tuesday, December 6, 2011
There's no such thing as too much compression, if it sounds good. But there is a point where things might get worse, not better...
Do you ever look at your compressor's gain reduction meter? Actually, it should be called a level reduction meter but 'gain reduction' is a term that has been handed down by history, so I will continue to use it.
Obviously, you should judge compression by ear, but it is helpful to have a visual indication of what's going on. If you want to hear more compression, then you need to see more segments on the gain reduction meter lighting up (or the needle going down more deeply), and faster movements.
But here's something you should look out for - while the instrument is playing (or singer singing), does the meter always show some degree of gain reduction? Even in the quietest (but still playing) sections?
If so, then you are indeed applying too much compression, through setting the threshold too low.
You will hear the effect of this in the transitions from not-playing to playing. Just after the gaps in playing in other words.
When the instrument comes in each time, the compressor suddenly has to leap into deep compression and this leap, though momentary, will probably sound ugly.
But if you set the threshold so that the gain reduction meter reads only a decibel or two in the quietest sections, the compressor never has to compress more than it needs to. The transition from not-playing to playing will be much smoother.
Of course, if you use a compressor for warmth, such as a variable-mu compressor or digitally-mimicked version, then deep compression might give you the sound you are after and you might accept a little roughness in the transitions.
Other than that however, it is a good guideline to set the threshold so that at the very quietest part of the recording, the amount of gain reduction drops almost to zero.Publication date: Tuesday December 06, 2011
Author: David Mellor
Thursday, December 1, 2011
Tuesday, November 29, 2011
How loud is loud enough? Will 'painful' do?
I am of course referring to electronically processed sound. There isn't much in the way of natural sound that I ever find too loud. Only the irritating tick of the clock in my Rolls Royce. Kidding!
I went to an amateur dance show in a local town last weekend expecting a lot of enthusiasm, an impressive amount of talent, the occasional slip-up, and a good time being had by all. Oh, and a raffle. Everything was in the aid of a hospice care charity.
I also expected the sound to come entirely from playback. The days of dancing to real instruments, or even a piano, seem to be largely in the past, unfortunately.
The biggest drawback to dancing to playback is that the dancers have to follow the music rigidly, whereas with live music there can be a degree of give and take. A conversation I had with a top-notch ballet conductor some years ago confirmed this for me.
But the other drawback can be the quality of the sound. Looking at the loudspeaker installation in the small, publicly-funded theatre, I didn't hope for much. Oh well, I know that sound is always well down the list when it comes to allocating budgets.
What I didn't expect however was how ear-blastingly loud the sound would be, even from a seat much closer to the back of the room than the front.
It wasn't that the decibel level was too high in itself, but the combination of a high level and the very harsh sound from the speakers made the experience extremely unpleasant.
I am of course aware that as one ages, one's tolerance for harshness in sound diminishes. So I took care to canvass a much younger person's opinion. She disliked the sound so much that although invited to attend the second performance, she decided not to, because of the sound.
Why did this happen?
My feeling about how this happened is not that the sound operator (probably amateur) intended to distress his audience. But he assessed the quality of the sound he was reproducing from his listening position - right at the back of the auditorium. So what sounded loud enough to him was louder for everyone in the audience, and very much louder for those at the front.
One of the most important rules of live sound is to 'walk the room'. A wise sound operator will listen during rehearsals and sound checks from every possible audience location in the room. They will make sure that everyone gets a satisfactory - and not painful - experience.
Apart from having equipment that works properly, I can't think of anything that will make a bigger difference.
So in rehearsal or sound check, get the sound approximately right from the operating position, then walk the room. Come back and make adjustments and walk the room again. Your audience will not thank you for it, but they would do so heartily if they knew what the alternative might have sounded like!Publication date: Tuesday November 29, 2011
Author: David Mellor
Monday, November 28, 2011
Saturday, November 26, 2011
Wednesday, November 23, 2011
If your studio doesn't have acoustic treatment, then you're not making great recordings. But what is the first thing you need to know?
I recently came across an ad from an acoustic treatment company. Prominently featured was a list of "What you need to know..."
I can see that it is a problem for a company to sell products to a market that doesn't correctly understand the need for them. All the points on their list were important, although not all of them well-explained.
But their Number 2 point, which I would have put at Number 1, or the 'First Rule of Acoustic Treatment' was this...
"Acoustic treatment is NOT sound proofing."
The reason I put this point at the top of my own list is that I have had to explain this very many times to very many people. And when I have explained it in full and precise detail, they go away not believing me and still cling to the idea that somehow acoustic treatment and soundproofing are the same.
(By the way, some people like to refer to soundproofing as sound isolation or insulation. That's fine, I'm happy with any of these terms. I'm not so happy with 'acoustic insulation' as it sounds far too much like acoustic treatment for comfort.)
To try and make things crystal clear I'll put two important facts in bold...
Soundproofing reduces the amount of sound coming into or going out of your studio.
Acoustic treatment makes your studio sound better.
But why are they separate things? Couldn't you kill two birds with the proverbial one stone? (As I once won two coconuts with a single throw at my local fair!)
The problem is that it is impossible to provide any reasonable degree of insulation against sound using absorbing materials. Even the best sound absorbing materials, thickly applied, can only absorb a relatively small amount of sound energy at mid and low frequencies.
It is much more practical to block sound by reflecting it. Build a heavy wall from plasterboard (gypsum board), bricks or concrete and a decent degree of soundproofing will quite easily be achieved. (But note the word 'heavy' in that sentence. For truly effective soundproofing, you need a lot of mass.)
A soundproof studio will therefore require heavy walls, floor and ceiling, windows made from thick, multiple panes of glass, and a thick, heavy door with good seals around the edges.
At this point however, since most of the sound energy produced in the room will be reflected back and forth many times, the room will be VERY echoey.
So you need acoustic treatment to reduce the amount of reverberation, and in particular control it carefully across the full range of frequencies. For this you will use absorbent materials and structures (and some irregular hard surfaces for diffusion).
Absorbent materials and structures can soak up sound, but only over multiple passes, which is what happens in a soundproofed and acoustically treated room.
You can buy materials for soundproofing and acoustic treatment at a builders' merchant. But for acoustic treatment it's quicker and easier to go to a specialist such as www.universal-acoustics.com whose ad inspired this article.Publication date: Wednesday November 23, 2011
Author: David Mellor
Tuesday, November 22, 2011
Monday, November 21, 2011
Do you only have a 32-bit computer or 32-bit DAW? Is that what is holding you back from success?
Here is an interesting question received by RecordProducer.com...
"I have worked with PC computers which were built from the ground up and all have been Windows XP and Vista Ultimate. I have used Pro Tools 003 with very little problems. Written over 300 songs. After recording over 15 to 24 tracks my plug-ins are coming in slow. However, When I use the plug-in Omnisphere I run out of memory.
I only have 4 gigs and Fryers computer store says I can only go up to 3 gigs on my 32 bit board. I have a 250 gig hard drive and a backup one and a half terabit with an Intel dual core processor. I've had and still do good recording with minor problems. However, I want better performance. To prove my point go to Reverbnation and search Raleigh Simmons and take a listen to the music I have recorded with Protools on a rebuilt Windows Vista Ultimate system. Please let me know what you think about the recording. I need your opinions and help to do a better job.
Well it's time to build another Windows PC. I've priced everything. $787. out the door. Up to 32 gigs of ram. AMD Phenom 2 Quad processor, 500 watts power supply and a good sound card. The problem now is 003 can not run on a 64 bit board. You tell me...What do you do next ?"
Among the correspondence we receive both at Audio Masterclass and RecordProducer.com, a common theme is that somehow it is the writer's equipment or software that is holding them back from success.
You don't have to search the Internet for more than a few minutes to find that 64-bit computing is considered a big issue in audio. "64-bit Pro Tools 10 is coming. This is HUGE!" is the title of a popular thread on one forum.
But does this mean that anyone who is still in the land of 32-bit computing cannot succeed in music or audio?
The answer is ABSOLUTELY NOT. You don't need 64 bits to succeed. You need a good song, well-performed, produced, recorded, mixed and mastered. Oh, and well-marketed.
You can do this on a 10-year old computer. The endless quest for having the latest equipment and software is much more likely to result in wasted time and effort than a hit record.
Having said that, improvements in hardware and software are always welcome. The correspondent above is having problems using sufficient plug-ins for his requirements, and with one of his software instruments. These are exactly the problems that 64-bit computing will solve. It has to be said though that on listening to his music, there is nothing so musically or technically demanding that it couldn't be done on a 32-bit system.
In my view, 64-bit computing hasn't quite 'arrived' yet. Pro Tools 10 isn't 64-bit for example, although it will work with a 64-bit computer and 64-bit OS. In a year's time things will have settled down and 64-bit will be the norm.
So in my view, if you are making great music with a 32-bit system, then you can quite easily stick with it for a while. If you run out of steam plug-in-wise, then print effects to disk to free up processing power.
The place to be in the lead is in the quality of your music and audio. No-one who is likely to buy your music cares in the slightest how many bits you used to make it.
By the way, Avid's Pro Tools 10 FAQ seems to indicate that the 003 interface is still supported. It is worth bearing in mind however that Pro Tools 8 and 9 are still perfectly capable of turning out work of fully professional quality.Publication date: Sunday November 20, 2011
Author: David Mellor
Wednesday, November 16, 2011
Could anyone check out your work without being told who it is by and know within seconds that it is yours?
Fingerprints. Yes I mean literally fingerprints. The kind that Sherlock Holmes inspects under his magnifying glass to identify the villain of the story.
At Audio Masterclass we receive CD and DVD showreel discs regularly. Since we are here on the Internet you might think that physical discs are old-fashioned, but I can tell you that there is a whole world that exists outside of the Internet, and - believe it or not - it is actually bigger than the Internet! In the audio and musical regions of that world, discs are still a very common currency.
But discs are not without their problems. CDs are robust and DVDs even more so. You can scrape and scratch them and probably get away with it. But CD and DVD players have very little tolerance for fingerprints. A scratch causes some data loss that is in most cases compensated for by digital error correction and concealment. But a fingerprint scatters the light of the laser pickup over such a wide area that the data loss is too much and the discs skips, or playback grinds to a halt.
Amazingly, many of the discs we receive have fingerprints on the playing surface. Admittedly not as many as a typical Blockbuster DVD, but even one tiny print is one too many when you need playback to be perfect, first time without having to clean the disc.
And by the way, if you ever make a master CD for manufacture in bulk, it has to be ABSOLUTELY pristine!
P.S. If there is a fingerprint on your disc, Sherlock Holmes would know who made it!Publication date: Wednesday November 16, 2011
Author: David Mellor
Tuesday, November 15, 2011
Many engineers and producers separate recording and mixing into two distinct processes. But there are good reasons why you should mix as you record.
We take a lot of our current operating procedures from the traditional days of recording. Before DAWs, it was difficult or near-impossible to store and recall a mix. So during recording you would use a relatively crude 'monitor mix' and save mixing for real until after every note of music had been laid down.
(The alternative was to book a studio on 'lockout' so that the settings of the mixer could be kept intact overnight.)
Also, in the modern world of music, it is common that mixing will be carried out by a specialist mix engineer, so again there is a separation between recording and mixing.
But when you think about it, as you add new musical lines to a recording, shouldn't you be adding them in the best possible way, that contributes most to the quality of the production?
Why would you ever add a new musical line in a half-hearted way and not consider fully how it will sound in the finished piece of work?
It makes sense therefore when you add, say, a rhythm guitar part to optimize the sound from the guitar, amp and microphone. And then in your DAW's mixer set the EQ, pan and level that sounds exactly right next to your previously recorded tracks.
If you do this all the way from the beginning of a new session to the end, there need be no mixing process because you have built up the perfect mix for your song as you have gone along.
Of course, there will always be minor tweaking. But as a way of working, mixing as you go is a tremendously valuable technique.Publication date: Tuesday November 15, 2011
Author: David Mellor
Monday, November 14, 2011
You have to start your song somewhere. But where is the best place to start?
It is interesting to read or watch interviews with people who have been massively successful in music. Take some seriously big hitters from an earlier age of music...
Paul McCartney for example, whose song Yesterday apparently started as a melody accompanying the words 'scrambled eggs'. Take Andrew Lloyd Webber (who is actually higher in the Sunday Times Rich List than Sir Paul) who is on record as composing the music first, then having a lyricist fit words to it.
So the music is the best place to start then?
Well not for everyone. A television documentary on Elton John showed him rummaging through pages and pages of lyrics by his co-writer Bernie Taupin, before finding one that he liked, to set to music.
So you can also start with the lyrics. Gold records and multi-million dollar success seem to confirm that.
But what about starting with the production?
Well once again, the musical masters of the past have lessons for us.
U2 for instance are known to start from the germ of a musical idea, go away separately to add to that idea, start the recording process, and only then does Bono begin to put lyrics together. This is indeed starting with production, where the process of musical creation starts with sounds and textures rather than entire melodies.
And then there are the Rolling Stones. It has been documented many times that they used to go into the studio devoid of ideas, start the tape machines rolling, then jam. Eventually ideas would be pulled together into a song. Once again, this is starting with production.
So the moral here is that you can start anywhere in the song creation process. Pick the method that suits you best. Or try them all.
P.S. There's a rumor that the Stones used to start a song by jamming over Beatles tracks. Surely that can't be true!Publication date: Monday November 14, 2011
Author: David Mellor
Friday, November 11, 2011
Perhaps your microphone came with a shock mount. Perhaps it was an expensive accessory. But why won't a simple clip or stand adapter do the job?
The function of a shock mount is to prevent vibration traveling up the mic stand from getting to the diaphragm of the mic. If you have ever recorded with mics set on stands on a portable staging system, then you will know how bad things can get. (Worse still with older designs where the space under each stage element is enclosed and resonates.)
Some microphones are naturally more prone to this kind of noise than others, and need a shock mount to guard against even a small amount of vibration. Other mics are less sensitive and rarely need any special treatment.
It has to be said though that a shock mount can be an expensive accessory, if not supplied with the mic. They can be fiddly to use and not nearly so handy as a simple stand adaptor or clip.
In a home recording studio, it is actually quite unlikely that a shock mount will provide any benefit at all. There is absolutely no reason why you should not record without one, with any mic, as long as there is no stand-transmitted vibration.
If you are in doubt, set up your microphone. Set your preamp to the amount of gain you would typically use for vocals. Start recording, then walk around the microphone. If you can hear vibrations on playback, then you need a shockmount. If you cannot hear vibrations, and there are no subsonic frequencies visible in the waveform display, then you don't.
Sometimes I think that shock mounts are used because they somehow look more 'professional'. Actually this might not be a bad thing. If you invite a vocalist into your studio, they will sing better if they are impressed with what they perceive as the professionalism of your setup.
But if you're working on your own and do not suffer from stand-transmitted vibration, you don't need a shock mount. So you can save yourself some expense and unnecessary messing about with elastic and still end up with perfect recordings.
And if you're working on your own, your studio doesn't have to look professional, it just has to sound it!Publication date: Friday November 11, 2011
Author: David Mellor
Thursday, November 10, 2011
A RecordProducer.com reader finds himself always with the left meter showing higher levels than the right. Why is this?
I was once in a mastering studio with a very experienced mastering engineer. It wasn't my session and I was just observing. I kept quiet so that the expert could do his job, but I couldn't help noticing that, one track after another, he always kept the left meter a couple of dB higher in level than the right.
At the end of the session I couldn't resist asking him why this was so.
"I just like it that way." was his reply.
I can think of all sorts of reasons why this isn't the right thing to do. But I don't make my living day-in-day-out from mastering, with a stream of high-paying clients coming steadily through the door.
But suppose you find yourself doing this, automatically and almost without thinking, like the reader who asked the question. Why should this be so?
Well indeed, you might just like it. If you do, and you are sure of what you are doing, there is nothing holding you back other than the needs of your clients and your potential market.
But if it is just happening for no particular reason, it indicates one of two possibilities...
The first is that there could be something wrong with your equipment or monitoring surroundings. These days, it is very unlikely that anything that comes before the monitor outputs of your interface could be causing the problem.
In the 'olden days' of audio, it could be a hundred things, including inaccurate meters. But that wouldn't happen now.
No, the problem is likely to be downstream of your monitor outputs - your power amplifier or your speakers.
And it could be your acoustics. Something about the shape of your room is emphasizing one channel over the other. It is always best that your mixing room is symmetrical about a line projecting out from halfway between your speakers.
The simple way to check your equipment is to swap everything around from left to right. If the problem changes channel, then you can quickly home in on the cause.
But suppose your equipment is perfect? Then it could be your hearing. Probably no-one has perfectly symmetrical hearing and it could indeed be the case that one of your ears is a couple of dB down on the other, and you are setting your faders and pans to compensate.
Once again there is a solution. If you make a switch that can instantly swap the two monitor channels, then if you set your faders and pans so that the relative 'weight' of sound is pretty much the same either way, then you can be confident your stereo balance is fine.
That could be the end of my article, but I have one more point...
A lot of people would simply look at the stereo output meters and assume that if they were balanced, then everything is fine. But the correct way to judge a mix is on what you hear, not on what you see. Meters are useful, but for anything other than clip indications they can never be more than a guide.Publication date: Thursday November 10, 2011
Author: David Mellor
Wednesday, November 9, 2011
From a grunt to a scream, can a compressor even out all the varying levels of a vocal?
By David Mellor
David Mellor, Course Director of Audio Masterclass
"When I was recording a CD with my band, vocals day, I asked the producer what about the different levels of the vocal regions. The vocalist during the recording was moving a little, and there were some parts where the vocals were not at the same level. He told me, "When mixing I'll put in a compressor and all the vocals will be at the same level". My question is how I can do this? I have some metal vocals (screams and guttural vocals) and I want them all to be at the same level. Is this method correct?"
The simple answer is, "No - a compressor is not the right tool for this particular job."
Suppose you have a song where some parts are sung in a low key, in a dark and moody style. Then during the choruses the singer screams at a high pitch. This scenario is not uncommon in the metal genre. There is going to be a huge difference in level.
Technically, a compressor could be used to bring down the screams to the level of the quieter sections. Now however, the screams would sound very compressed, while the quieter sections would hardly be compressed at all. This is unlikely to give a pleasing result overall.
One easy solution is to separate out the sections onto different tracks. You can then process the sections according to their own individual requirements and set the faders so that the vocals blend correctly with the instruments all the way through the song.
Some songs might benefit from breaking up the vocals onto three tracks.
There are other ways of evening out a vocal that varies in level, but this is a quick and easy solution that works very well.
Publication date: Monday November 07, 2011
Author: David Mellor