Q. How can I easily match levels on MP3s?

When I listen to MP3s on a PC or Apple Mac, they are all at different levels, with the louder ones seeming twice as loud as the quietest ones.




Back in the days of tape and vinyl, you set your own recording level on the tape deck. So if you recorded three different tracks to tape you could get the levels similar; listening to a 'mix tape' was way more consistent than ripping CDs to MP3, where different tracks have different volumes.Has anyone made a piece of software that adjusts the volume level of tracks so they all match? Surely this can't be that hard? And surely I am not the only guy on earth that finds this annoying? (Please note, however, I do not mean I want a compressor or limiter that zaps the dynamic range!)



Via SOS web site



SOS Technical Editor Hugh Robjohns replies: The 'mix tape' solution is still available of course: you can transfer tracks manually, perhaps using a DAW, adjusting levels as you go. But few of us are probably prepared to invest that kind of time and care these days!



The issue of varying loudness is a serious one, and something that afflicts many different media outlets. The film industry tried to address this issue a while ago with some success, for example, and the broadcast TV industry has recently introduced a new standard that is currently being adopted worldwide and which involves a reliable and standardised way of assessing and quantifying loudness. The broadcast radio industry will follow in a year or two's time, and these initiatives will finally end the 'loudness wars'. The standard in question here is the ITU-R BS.1770, and the various local adaptations such as EBU R-128, ATSC-A/85 and the CALM Act in the USA.



The metering system that underpins these standards is already widely available and I would urge anyone involved in mixing music or involved in audio production to familiarise themselves with this system as soon as they can. There is no doubt that this is the future. While the 'loudness wars' still rage on (for the time being, at least), you can easily level out your MP3s with iTunes' surprisingly effective Sound Check feature. That's assuming you don't have the patience to adjust the levels of your favourite tracks manually using a DAW, of course!While the 'loudness wars' still rage on (for the time being, at least), you can easily level out your MP3s with iTunes' surprisingly effective Sound Check feature. That's assuming you don't have the patience to adjust the levels of your favourite tracks manually using a DAW, of course!



Returning to your question of software solutions, the easiest option, if you are using iTunes and/or iOS-based players, is to switch on the 'Sound Check' function. This doesn't quite conform to the BS.1770 standard, but it is very close and works extremely well. Essentially, it analyses the loudness of each track in the library and writes metadata into the file header which documents the playback level needed to achieve a perceived loudness equivalent to -16dBFS; this level being chosen to accommodate wide dynamic range material without clipping. Since the Sound Check function is only storing a 'level offset' instruction, the actual stored audio data isn't altered in any way, and the replay process is directly equivalent to you manually adjusting the playback level, so it is entirely non-destructive. The result is that every track ends up with a similar perceived loudness, which is exactly what you want. An alternative that I've not tried, but which I've seen recommended, is a program called Mp3Gain Pro (www.mp3gain-pro.com). This is capable of batch-processing MP3 files to establish a common loudness across all files, but I believe it does so via a destructive manipulation of the original file data.    

Published in SOS March 2013

Summer NAMM: The Moog Guitar

Q. Are wow and flutter key to that analogue tape sound?

I have come to the conclusion that wow and flutter are a lot more important in the sound of tape and analogue recordings than they are usually given credit for. Most of the discussion about tape seems to concentrate on tape compression and the effects of transformers in the signal path, for example, and the majority of plug-in treatments designed to make recordings warmer focus on this. I don't hear of many people applying wow and flutter plug-ins, or waffling on about the right type of capstan emulator. Recently I was re-reading one of those pieces Roger Nichols wrote for SOS a few years back, where he mentions that someone had invented a de-wow-and-flutter system that tracked variations in the pitch of the bias signal to correct for wow and flutter, and he said the result sounded 'just like digital'.I recently did a couple of projects where I more or less did the same thing, albeit hugely more labour-intensively: I transferred some old four-track cassette recordings to my PC. The recordings used a drum machine, which I still own, so I also made a clean new digital recording of the drum machine part. But, of course, due to wow and flutter, the old four-track recordings were out of sync with the drum machine on a couple of bars, so I ended up chopping up the four-track capture bar-by-bar, and time-stretching each bar so that the waveform of the drum machine recording on tape lined up exactly with the new, clean digital version. By the time I'd finished, the four-track did indeed sound quite different in character to what it had before. I think Nichols was right. I wonder what opinions the SOS team might have about the importance of wow and flutter on getting 'that sound'?




Via SOS web site



SOS Technical Editor Hugh Robjohns replies: I agree that the subtle (and sometimes not so subtle) speed instability of tape is an important subconscious factor in the tape sound. Any time-modulation process, including wow and flutter, creates additional frequency components, and I think the subliminal presence of these on all analogue recordings is sometimes missed from digital recordings. However, I suspect it is actually the presence of the far more complex harmonics produced by 'scrape flutter' that is the most significant element, rather than the very low and cyclical frequency modulations caused by wow and flutter. Added to which, I find wow and flutter generally quite objectionable, especially in music with sustained tones, like piano and organ recordings.



However, what you are describing here is not actually wow and flutter. You're describing speed 'drift', which is an absolute difference between the record and replay speeds. It's not unusual for two devices to run at slightly different speeds, even in digital circles. Two separate CD players might run with sample rates of at 44101Hz and 44099Hz, for example, or two analogue tape machines at 19.1cm/s and 18.9cm/s. If you start the two machines at the same time with identical recordings, they will drift in time relative to one another, just as you found with your four-track cassette — although in that case I suspect the problem was caused either by poor speed control or physical tape stretch.



Wow is a low-frequency cyclical speed variation, which is very common on vinyl records if the centre hole is punched slightly off-centre, of if the disc is badly warped. Flutter is a much faster-frequency version of the same thing, typically caused by a worn tape-machine capstan or a lumpy pinch-roller. Scrape flutter is a higher-frequency effect again, typically caused by the inherent 'stiction' or vibration of tape against the heads as it is dragged past.



Wow and flutter, being cyclical phenomena, don't usually result in a change in the average replay (or record) speed because any short-term speeding up is balanced completely by the same amount of slowing down as the cycle completes.



I'm not at all surprised that your heavily edited and time-stretched 'fixed' version of the electronic drum track sounds different from the straight digital recording, specifically because you performed so much processing on the individual sections. However, that 'fixed' version will also sound very different from the drum machine's direct analogue outputs. You're not 'fixing wow and flutter' but actually correcting for speed drift or tape stretch by time-adjusting the original material in short sections, which is naturally messing with the sonic character of the drum beats in short, unrelated sections. Though wow and flutter may once have been phenomena that we were used to and could therefore ignore, their absence in modern recording means that this is no longer the case. Celemony's Capstan is an incredibly effective tool for removing these unwanted effects, and it leaves few artifacts.Though wow and flutter may once have been phenomena that we were used to and could therefore ignore, their absence in modern recording means that this is no longer the case. Celemony's Capstan is an incredibly effective tool for removing these unwanted effects, and it leaves few artifacts.



Returning to conventional wow and flutter, though, after nearly 30 years of 'digital stability' most of us have been completely weaned off the sound of wow and flutter, and our ears have become very good once again at spotting these grossly unnatural phenomena that we were once so happy to ignore. Last year I reviewed Celemony's Capstan software, which is designed to fix both wow and flutter and speed-drift issues, and it does so extremely well and without artifacts!    

Published in SOS March 2013

Summer NAMM: Audio Technica

Q. Can you recommend a 73-key stage piano?

I seem to be the only person in the world who wants an unfussy, weighted stage piano, with — at most — 73 keys. I have little money, so can't afford to have a piano for home use and a piano for stage use, and I have no space to store them even if I could afford it. I don't play the 'dusty' ends much, so saving space by not having 88 notes suits me fine. I also have feeble arms and a small car, so it'd be great to keep the weight down too. My ideal would basically be the Casio Privia P3 with two octaves missing, as it has a great sound and lovely action. Is there really nothing out there — current or discontinued — that could do all I want? I can probably stretch to around £1500 if I had to. What might you suggest?




Lucy Weston via email



SOS contributor Robin Bigwood replies: There are actually quite a number of 73- or 76-note keyboards out there that could fit the bill. As always, you have to decide what your priorities are. For example, hammer-action keyboards are usually very heavy, so the keyboard with the action that suits you most might also be the least portable. There's also a choice to be made between a high-quality but limited piano-oriented sound set, or the 'jack of all trades' nature of a synth workstation.



I think the keyboard most worthy of your consideration is the Nord Electro. Version 4 of this well-respected and undeniably vibey keyboard was launched fairly recently, but its v3 predecessor seems to live on in Nord's range. The 73-note version comes in at around £1400, has semi-weighted keys and weighs less than 10kg. The version with a hammer action, surprisingly, weighs only 1kg more, but it'll set you back a cool £1800. Still, these are brilliant gigging instruments that are well worth the money. They can be loaded with all sounds from the Nord piano and wave libraries, and sport top-class rock organ emulations too.



Challenging Nord in this same market sector are a couple of serious players' instruments by Japanese manufacturers. The Korg SV-1-73 is £1299, offers 36 electric and acoustic piano presets, and has a decent Korg RH3 hammer action. The alternative offered by Roland is the 76-note VR700 V-Combo at about £1200. You get great organs and pianos, along with strings, synths and pads. And, with a lighter 'waterfall' keyboard, it's not too heavy. It is rather long, though, because of those extra keys and a 'bender' section to the left of the keyboard.



Next up, a couple of 76-note stage keyboard all-rounders. Cheapest of all (£599) is the Kurzweil SP4-7. There's no doubting the pedigree, but this workmanlike piano could prove a bit basic for really serious use. More flexible, though unashamedly oriented towards the synth world (the clue's in the name) is the Roland Juno Stage for £950. I spent some time with one a little while back and enjoyed playing it. Like the V-Combo it's quite long, but it has some nice live-leaning features such as audio file playback (for backing tracks and so on) from USB sticks, a click output for drummers, and a phantom-powered mic input that's routed through the internal effects. Buying a stage piano will always require some compromise, whether that means having less than you want or, in some cases, more. The excellent Nord Electro 4, for example, is easily portable, but only has semi-weighted keys, and is not cheap. However, the Korg M50, though technically a synth workstation, rather than a stage piano, is a snip at £850 but, again, only has semi-weighted keys. It's no surprise that fully weighted keys and portability do not go together!Buying a stage piano will always require some compromise, whether that means having less than you want or, in some cases, more. The excellent Nord Electro 4, for example, is easily portable, but only has semi-weighted keys, and is not cheap. However, the Korg M50, though technically a synth workstation, rather than a stage piano, is a snip at £850 but, again, only has semi-weighted keys. It's no surprise that fully weighted keys and portability do not go together!Q. Can you recommend a 73-key stage piano?



Finally we get to those synth workstations. The Korg M50-73, around £850, is a svelte 9kg and could get you safely in and out of many gigging jobs. But there's also the new Korg Krome 73 for £1000 or so, and that boasts a flagship Steinway piano sound, plus good e-pianos too: definitely one to audition. I reviewed the Kurzweil PC3LE7 for SOS a while back, and, while I thought it was a real workhorse, its pianos (in particular) are a little way off state-of-the-art. I'm sure the Yamaha S70XS at around £1600 would be nice, too, but it's a hammer-action whopper and a solid 20kg.



In essence, though, these are all rewarding, useful instruments, so choosing between them is a nice problem to have. Best of luck!  

Published in SOS March 2013

Summer NAMM: C-Thru AXiS 49

Q. Can you explain digital clocking?

Phrases like 'digital clocking', 'word clock' and 'interface jitter' are bandied around a lot in the pages of Sound On Sound. I'm not that much of a newbie, but I have to admit to being completely in the dark about this! Could you put me out of my misery and explain it to me?




Interface 'jitter', which results from clock-data degradation, can cause your waveform to be constructed with amplitude errors, seen in the diagram. These could produce noise and distortion. It's for this reason that people sometimes use a dedicated master clock, which all other devices are 'slaved' to.Interface 'jitter', which results from clock-data degradation, can cause your waveform to be constructed with amplitude errors, seen in the diagram. These could produce noise and distortion. It's for this reason that people sometimes use a dedicated master clock, which all other devices are 'slaved' to.Q. Can you explain digital clocking?



James Coxon, via email



SOS Technical Editor Hugh Robjohns replies: Digital audio is represented by a series of samples, each one denoting the amplitude of the audio waveform at a specific point in time. The digital clocking signal — known as a 'sample clock' or, more usually, a 'word clock' — defines those points in time.



When digital audio is being transferred between equipment, the receiving device needs to know when each new sample is due to arrive, and it needs to receive a word clock to do that. Most interface formats, such as AES3, S/PDIF and ADAT, carry an embedded word-clock signal within the digital data, and usually that's sufficient to allow the receiving device to 'slave' to the source device and interpret the data correctly.



Unfortunately, that embedded clock data can be degraded by the physical properties of the connecting cable, resulting in 'interface jitter', which leads to instability in the retrieved clocking information. If this jittery clock is used to construct the waveform — as it often is in simple D-A and A-D converters — it will result in amplitude errors that could potentially produce unwanted noise and distortion.



For this reason, the better converters go to great lengths to avoid the effects of interface jitter, using a variety of bespoke re-clocking and jitter-reduction systems. However, when digital audio is passed between two digital devices — from a CD player to a DAW, say — the audio isn't actually reconstructed at all. The devices are just passing and receiving one sample value after another and, provided the numbers themselves are transferred accurately, the timing isn't critical at all. In that all-digital context, interface jitter is totally irrelevant: jitter only matters when audio is being converted to or from the digital and analogue domains.



Where an embedded clock isn't available, or you want to synchronise the sample clocks of several devices together (as you must if you want to be able to mix digital signals from multiple sources), the master device's word clock must be distributed to all the slave devices, and those devices specifically configured to synchronise themselves to that incoming master clock.



An orchestra can only have one conductor if you want everyone to play in time together and, in the same way, a digital system can only have one master clock device. Everything else must slave to that clock. The master device is typically the main A-D converter in most systems, which often means the computer's audio interface, but in large and complex systems it might be a dedicated master clock device instead.



The word clock can be distributed to equipment in a variety of forms, depending on the available connectivity, but the basic format is a simple word-clock signal, which is a square wave running at the sample rate. It is traditionally carried on a 75Ω video cable equipped with BNC connectors. It can also be passed as an embedded clock on an AES3 or S/PDIF cable (often known as 'Digital Black' or the AES11 format), and in audio-video installations a video 'black and burst' signal might be used in some cases.    

Published in SOS April 2013