rt – rtsound

April 10, 2023April 10, 2023

32bit float – Is there anything to Gain?

High dynamic range analogue inputs

We’re getting to a stage where most professional recorders (and a number of ‘prosumer’ ones) operate with analogue inputs capable of capturing very high dymamic range. In fact, some have little or no amplification in front. For example, the Sonosax R4+ and AD8+ which I’ve talked about before in more detail. Some specs quote 135dB+ of dynamic range- that’s equivalent of being able to capture a pin drop to a jet taking off.

This is usually done with the signal being split between more than one analogue to digital converter chip, with each set to different sensitivities; converting the voltage level to a digital value. It’s like recording the same source to 2 inputs with different gain settings for safety- but it’s combined as the same signal.

24 bit converters (even multiples) can’t actually reproduce 24bits accurately though. That’s 144dB. Or in plain numbers 2^24 = 16777216 different values. We’re currently maxing out around 23bits (138dB) .

Analogue garbage = Digital garbage

So we have these wonderful converters- what about our wonderful analogue sources? They can’t manage as much, now- we’re limited by physics with microphones. Smaller diaphragms have more uniform frequency response but higher noise. DPA have a good overview of this on their website. Also condenser microphones always have an equivalent input noise level, which is their ‘base level’. If the sound you’re trying to record is around the level of the input noise- you’re recording a lot of the input noise compared to the source. You’re not going to be able to get around this with amplification or conversion as the issue is at the microphone. Even if your microphone can manage over 120dB dynamic range- it has to be at the right level, if the signal is down in the bottom 20dB or so then it’s going to be fighting the noise. Adding to this is the amplifier noise- it’s lower but will add to the noise floor of the microphone.

What is Gain any more?

In a lot of cases, the gain knobs in our machines aren’t controlling a mic preamp any more. It’s controlling digital gain- just multiplying the signal by a number, which is exactly the same case in post production where you have a waveform and want to make it louder or quieter and add or remove gain in the editing software. It doesn’t change what’s coming in, just the level of what gets sent to a track.

What’s the floating point?

Once the audio is in a DSP system (such a digital mixer), any changes made are with mathematical operations- multiplication, division, addition, subtraction. This means that the numbers in these operations can get significantly bigger. Once you’ve already got a 24 bit number, and multiply it by something- even if it gets divided again it needs space for more digits. With fixed point calculations, that’s exactly what happens- some systems use 40bit or 56bit numbers for calculations.

Another way of doing this is to ‘move the decimal point’. To make a simpler example, say your biggest number in a system is 99. Using fixed point arithmetic it can’t change, however if you could add a decimal point, the 99 could be expressed as 0.99, 9.90 or 99. What was 100 different numerical levels is now 10000.

However, if you make a calculation which results in fraction or irrational number- you end up with a load of digits. If the decimal point moves again, you’ll need to round the number to the nearest digit, so some precision can be lost with these calculations.

Both higher level fixed point and floating point allow levels higher than 0dBFS to exist within the system.

32bits = 24bits (or 23)?

Something floating point numbers require, along with the value is to record where the decimal point is. This needs another number. It also needs to record if it’s a positive or negative value.

Here’s a breakdown of a 32bit float representation base on IEEE754 format (standard for a lot of computer hardware systems) – (image licensed under CC BY-SA 3.0)

The sign bit sets whether the number is positive or negative, it’s -1 to the power of either 0 or 1. So, (-1)^0= 1 and (-1)^1= -1

The 8 exponent bits are to set where the decimal point is. If we were in base 10, to move a decimal point we’d multiply by a multiple of 10. So multiplying by 10^1 =10 would move the decimal point one point to the right. To undo that you’d multiply by 10^(-1) = 0.1. On the example, as we’re in base 2, we’d multiply by 2^124. However we also need to be able to go down as well as up. 8 bits allows 256 values, so IEEE174 allows the top 128 as positive and bottom 128 as negative, so 127 is taken away from that number. 2^(124-127)= 2^(-3)

The 23bits of fraction is the actual value we’re processing, however 24 bits of precision is possible with this method. Because the exponent dictates where the first part of the number will be, it’s always going to be 1 if there is an exponent- so we get a ‘free’ bit.

Output

So, what comes out at the other end? For us to hear it, it needs to feed digital to analogue converter. These can have similar performance to high end ADCs (i.e. 22bits+ of dynamic range), but even with the (theoretical) amplifiers and speakers to keep up with it, do we want to or can our brains deal with pindrop to jet engine? A lot of people are finding cinema dynamics too much- having to strain to hear dialogue then be deafened by an explosion. Some 32bit DACs may be designed to be fed floating point information, but the performance output won’t be any better than 23bits.

What is it good for? [TL;DR]

Given that we can’t get more than 22bits or so of useful information in or out of any system (and I’m not sure we need to), it doesn’t make a difference to inputs or outputs. 32bit float is very useful inside machines for making calculations and some of our gain controls are now mathematical operations done in DSP, rather than controlling analogue amplifiers. If you’re using a 32bit float recorder, your gain knobs are very likely to be controlling digital processes and nothing in the analogue domain. If the files stay digital as 32bit floating point files, the maths can be undone (try exporting a 32bit file in a DAW and re-importing it). If they’re ‘baked in’ at 24 bit with an overload, then there’s missing information in the file as it’s a number bigger than the file can use.

However, if you don’t do any maths to the input and leave it as is, then you’ll also get what goes in and the maths can be done later in post production. In some systems this may mean it’s difficult to monitor, though.

What’s it bad for?

Not all programs are designed to read 32bit float audio files. A lot can, but some picture editing programs can’t or may not. A lot of the time we don’t always know the post workflow, there may be ingest programs used or older software kept on to stick with a specific workflow.

If a program expecting a 24bit .wav file gets a 32bit float one, it’ll see it as a 24 bit file and not see any of the information above 0dB, so anything clipped will not be recoverable.

If you’re going to use them, ensure you do a workflow test with audio running over 0dB or if your recorder only records in 32bit floating point- don’t clip it!

March 18, 2023March 19, 2023

Making a start in Production Sound

This post has been a long time coming. Over years I keep being asked “How do you get into this?” as it seems like almost everyone has a different route in, however I’m going to try and point some in the direction of some of the more established paths. From talking to a number of people it seems to be a combination of these which has led them to becoming established freelancers. There have also been clear ways in previously, however they now seem to have disappeared – BBC staff careers for example. There aren’t many staff jobs available either- almost all of the work is freelance. This is also all quite UK specific and may be different in other regions.

What is Production Sound?

While making any kind of film or video production, there are 3 stages. I’m going to reduce them to a very basic level:

Pre-production = Planning
Production = Shooting
Post-Production = Editing

Production Sound is the sound recorded during production (it’s self explanatory). This is almost always what’s happening in front of camera. Most of it is dialogue.

Post-Production Sound is sound edited and recorded after shooting. This includes all the sound editing and mixing process and any extra recording required to complete the soundtrack.

Education

There are no formal educational requirements to doing the job. In fact, it’s basically down to whether you’re trusted to be able to do it or not. Education can really help with practical elements, though. Knowing your way around audio equipment; physics, electronics, acoustics, music technology are all useful in getting to know the principles of how it all works. Business skills and Psychology are also very useful- you’ll be running a business and we need to work with other people. There’s negotiation involved in a lot of what we do.

There is one specialist Location Sound Recording for Film and TV diploma course in the UK at the National Film and Television School. It’s a 15 month full time in person course and aimed at postgraduate level. Entry requirements are based more on interest and aptitude and bursaries are available. A number of people have gone through there as mature students having previously had a different career. Some advantages of this are you can learn in a non-pressurised environment and have access to professional equipment. The tutors are also working professionals. You’ll also be working with other students in other disciplines, so you can learn how to work with them and relationships can continue after the course has finished and well into later careers. They also do a week long short course, which is at least a solid introduction.

Non specific film courses can also help, especially in seeing how to work with other departments on set, but there won’t be as much sound specific teaching.

Trainee Schemes

Trainee schemes are probably the most formal route of getting into drama work. Either (“High End”) TV or Feature Films. It’s well worth being a trainee first- even if you’ve got years of audio experience or degrees, film sets are quite a specific thing and this is where you’ll learn etiquette. You’ll learn how the sound dept fits in with everything else.

Screenskills run two different trainee schemes. One for ‘High End TV’ and one for ‘Feature Films’. The reasons for this is that it’s from two different sources of money, the jobs aren’t technically that different on a day to day basis and some sound teams may go between feature films and high end TV. However, you can only apply for one – if you’re not sure, toss a coin. If you’re successful, you’re not guaranteed a placement through screenskills directly- however screenskills will pay a portion of your wage while on the scheme for any trainee work within the category you’ve applied for. It’s likely you’ll have to ask sound teams directly to get something out of this.

Sara Putt Associates is an agency, but they run a trainee scheme. In this case it’s not about getting work directly, but building up skills and a network in order to get more work and progress.

In both cases it’s a bit ‘Catch 22’, you need some work or track record to get on the scheme, but not enough that you’re already seen as established. Bear in mind that you are applying as a trainee, if you’re putting forward a CV saying “I know everything” it’s unlikely you’ll succeed.

You do not have to be on these schemes to be a trainee. I have started to see some sound trainee positions advertised now, in the past this hasn’t been the case.

Facilities Companies

The ‘traditional’ route into factual or factual entertainment TV is via facilities companies. TV productions often want a ‘one stop shop’ where they get a quote for all the equipment and crew for a job and this is where they get it. They need people to look after the hire equipment and it’s can be an entry level job (although having a degree in something audio or media related can certainly help). You’ll get familiar with a lot of different equipment and soon end up troubleshooting issues. If the facility does hire out staff, then they may send you out on jobs with them- you’ll typically be doing more work than a freelancer (as you’re on the payroll already) and quickly build up experience and contacts.

I’d suggest only looking at facilities with a dedicated sound department- most are camera focused, but you do need more experienced staff to learn from.

They’re not just useful if working directly for them. If you regularly hire out or buy equipment from them, or even come in to hire things out, they might have jobs come in and ask you to work freelance, or pass your details on. If you’ve got the time, do turn up in person, you might bump into someone useful even.

Networking

Personally, I’m really not keen on using this word when it isn’t used in conjunction with IP Addresses and RJ45 connectors [actually, learn that too]. However, this whole industry is based on recommendations- meet people, get involved, talk to us about batteries. We all have other people we pass work on to if we can’t do it. If one of us trusts you to be able to do something then they can suggest you for work. It doesn’t always happen.

There are organisations and communities- AMPS (Association of Motion Picture Sound), IPS (Institute of Professional Sound) and a number of communities on facebook (search for the ‘Film & TV Sound UK’ group). Some of the criteria seem a bit like you need to know people already, but they’re much more open to newcomers and students. If in doubt, apply anyway.

Come to stuff! I’m also much more receptive to people I’ve met, rather than a cold email or call (also I don’t get much work where I can hire a trainee).

Social media can also very useful for making and maintaining relationships, although it can work both ways. If you’re interesting, informative or amusing it helps. Asking good questions can go down well – things like “how would you deal with this situation?”, rather than “what’s the best microphone?”. Good answers are usually not a specific piece of equipment.

Microbudget World

Microbudget projects- things like short films and other personal projects often don’t make money but directors, producers and actors can use them for showreel and begin getting a track record. I’ve already got quite a detailed piece on them.

They’re a bit of a tricky thing for a newcomer, in some ways they’re a way of meeting more people involved in making film or tv and to get a bit of experience. However you’re working with other people who need experience and there’s more and more pressure to have a single person sound department on these jobs (as there’s no money), which means there’s no one to learn from. The sound recordist as head of department becomes an entry level position.

There isn’t much in it for more established production sound mixers in doing things like this and to have you as an assistant, except maybe trying out a new rig. If something like this comes up, that’s a great opportunity. However if you’re on your own, there’s not much you can learn from except your mistakes.

You can meet people in other departments who might recommend you for something else, though.

I’d say don’t buy equipment to do these jobs, but sometimes a very barebones kit might be useful for last minute requests. It’s often not stuff you’ll use later. Make sure you get money to hire it (even if you’re not paid), it’s not free to buy or maintain. Try and get to know other sound people who may also be able to rent equipment.

Conclusions/Confusions

Yes it is confusing, there isn’t a clear way to get in and make a career in production sound.

You’re looking for multiple jobs all the time, most of which aren’t advertised (and those which are are usually advertised due to to them wanting to pay less than they should be). It’s difficult to get going and sustain yourself financially in the beginning. You are reliant on a network of other people in both sound department and other departments.

The more you’re out on jobs, the more you’ll learn and you’ll work with more people.

Lots of this is being at the right place at the right time, try and get yourself in the right places as much as possible (and this can be online).

I expect this is especially difficult looking in as someone who isn’t middle class/white/male. Most of us are middle class, white and male, but it’s beginning to change and I’d like to be working with more people with different perspectives. Do apply for stuff (even if it seems financially impossible- there is help), come and say hello and things might happen.

It’s very unlikely you’re going to be able to make a living from this immediately. Work will be slow to pick up while you grow networks. Make backup plans- have a look at flexible unrelated work- delivery driving- for example, which can be moved if you get a job. If you’re put in a situation where you’re taking sound work out of financial necessity, it puts you in a weak position with negotiations and may damage both your future prospects and that of your peers.

January 30, 2022February 8, 2022

Network Switches for location

You can’t have your cake and eat it

Location requirements

I’ve had a number of people ask me about network switches over the last week or so. More mixers seem to be looking at Dante on set in various configurations, however most of the switch manufacturers don’t exactly have us in mind when designing products, so we’ve got to make compromises.

Things we need/want:
No disruption using AoIP protocols, like Dante, AES67 etc
Silent (usually fanless)
12V DC powered
Fast boot up
PoE distribution

I’m not aware of any switch which necessarily ticks all the boxes, but there’s quite a few which will tick some.

Managed and Unmanaged Switches

As far as best practice goes in putting together a Dante network, a lot of the choices made can only be done using managed switches, especially if anything which isn’t Dante is on that network.

One of the key functions of managed switches is Quality of Service. This prioritises certain traffic (with the most important being the clock). It’s recommended by Audinate to implement this if there are any slower 100Mbps devices (like the small ones which use an Ultimo chipset) on the network.

Another is IGMP Snooping- this only comes into play if multicast sends are being used, as it limits the number of destinations the data is sent to and reduced the overhead on the switch.

If there are only a few gigabit devices (not small ultimo devices!), without much traffic then an unmanaged switch should be fine. The main thing to look out for is if it has ‘Energy Efficient Ethernet’- this can cause a Dante network to fall over, so you don’t want this feature and if it’s on a managed switch make sure it can be turned off.

Audinate and Shure both publish lists of switches which are not recommended.

A drawback of managed switches can also be boot time. The more switches seem to do, the more complicated the operating system. Some can take a few minutes to boot up, which really isn’t useful if you’ve just had a power outage.

Power over Ethernet (PoE)

Power over Ethernet isn’t quite as straightforward as you might think. There are multiple standards, which use different pins and some have handshake methods with the other end before turning it on. You need to check the standards required of the boxes you’re plugging in, they won’t work with every switch and different switches have different capacities

The basic 3 standards are:

PoE – this can be very vague, from a power source being plugged into some of the wires on a CAT5 cable. The base standard is 802.3af, which uses 48V to carry up to 15.4W of power on one network cable

PoE+ or 802.3at. This allows up to 30W and runs at 52V

PoE++ or 802.3bt This is the big one, either 60W or 90W. Also needs a juicy power source

Cisco also do a proprietary UPOE+ method which appeared before 802.3bt was standardised.

It all needs a high voltage and in some cases can carry quite a lot of power. This means that if equipment is running off 12V; it needs to be stepped up somewhere with a DC-DC converter.

PoE doesn’t have to be implemented at the switch, however. It can also be added with a PoE injector. This is useful if there’s only one or a couple of devices which need it.
I’ve sourced some cheap 10W injectors which will step up from 12V: https://rtsound.com/collections/power-over-ethernet/products/48v-10w-poe-injector-with-voltage-step-up-from-10-24v

Planet make some industrial 802.3bt injectors which will step up from 12V

Some switches can be powered by PoE and some can then pass that on. They’ll never be able to pass on all of the power; don’t take either as given. “PoE PD” (PoE Powered Device) is often a term used.

Some Commonly used Switches

Cisco SG350-10

This is the ‘industry standard’ managed Dante switch. Audinate commonly use them and multiple manufacturers make guides to setting them up.

They’ll do everything you need management-wise.

They can output PoE+ and proprietary 60W UPOE+ and it’s possible to power a second switch with the first and pass through PoE+

They have SFP optical ports, which can be used for very long cable runs.

However- they need 54V. I’ve made a solution for this with a step up converter. This can also be adapted to other switches and the voltage tweaked fairly easily:
https://rtsound.com/collections/power-over-ethernet/products/dc-power-supply-for-cisco-sg350-and-sg250-poe-switches

It takes around 90 seconds to boot up, which doesn’t seem to be able to be got around. If there are 2 switches being powered by each other, that’s almost 3 minutes to get the remote switch up. And it’s been discontinued- it’s replacement, the CBS350 apparently takes even longer!

Netgear GS105Ev2

This is a bit of an unusual choice, as Netgear GS – E switches appear in the Shure Disqualified list. It is, however quite a commonly used switch as it’s small and can be powered by 12V.

Personally I’d be a bit wary of using this without knowing exactly why it’s on a disqualified list. It’s only the v2 version where Energy Efficient ethernet can be turned off (I believe it is off as a factory setting).

I have seen recommendations to turn off IGMP Snooping too as it apparently doesn’t work properly with Dante. This may also be the reason for its disqualification.

Trendnet TI-PG62B

This is an industrial unmanaged switch, which has the holy grail of PoE+ and the ability to run from a 12V supply.

It has 4 PoE+ ports and 2 SFP ports.

Planet also make an 8+2 port 12V industrial switch.

Other Choices

Having had a look around, Cisco do make managed industrial switches. They’re really not cheap and some are still limited to 100Mbps and require a separate DC PSU to step up.

Another expensive remote option could be Cisco’s Catalyst Micro Switch series, which are small, managed layer 2 switches.

Luminex make switches especially for live sound and video with neutrik ethercon connectors and come pre-configured, however they only have AC powering options.

Netgear make their M4250 series switches specifically for Audio and Video and can come pre-configured for Dante. Again, AC power only…

February 20, 2021February 8, 2022

Interior/exterior windshield solutions for short shotgun microphones

What’s the problem?

I’ve been using Sennheiser 8060 microphones for a number of years. Initially they were a revelation; it was possible to fit a directional microphone into a small windshield (Rycote WS1 MZL), massively reducing the size and weight of an exterior setup. However they had a drawback: most of the microphone’s length is now interference tube, there’s not much to suspend it by. It’s not easy to quickly change from a ‘full blimp’ to foam or lighter weight windshield as it needs suspension on the interference tube, so I left it in a ‘full blimp’ almost all the time. Some people keep the foam on the mic and make a hole in it for the suspension, although it won’t take repeated removal. If the foam’s left on in the ‘blimp’ it’ll reduce the air volume around the mic and the blimp’s effectiveness, whilst also dampening higher frequencies.

What’s changed?

At the time, the 8060 was the only solution, however there are now more very short shotgun mics, the DPA 4017C and Schoeps MiniCMIT. There are also some other windshield solutions- the Cinela COSI is a excellent compromise between an interior and exterior mount. Another change was the availability of low profile right angle XLR connectors; these could be used instead of the expensive MZL connector which needed to be screwed on to the microphone capsule. Rycote have also made the INV-Lite mount, which is designed for these microphones, but has the drawback of sitting too high in the blimp to be practical.

What have you done?

It’s actually not particularly complicated, although I’ve ended up using quite a few older Rycote parts I’ve shoehorned in. I’ve put an Duo-Lyre suspension at the back of the microphone and low profile XLR. Here’s an image which tells you what the bits of the suspension are called. The modular bar is an older one, on backwards. It only has one screw hole through the filler strip so I’m using an older metal modular bracket as has a position in the middle. It could possibly do with a second to stop it going wobbly when changing between interior and exterior. The other hole needed to be filled- I’ve used some Joe’s Sticky Stuff for that. This brings back the whole ‘hinge’ mechanism and helps with balance.

Microphone suspension without windshield

Low Cut Filtering

A lot of microphones do require some filtering to help get rid of handling noise and mechanical motion in the suspension and pole- the 8060 has some in it but more is often required. I did make a previous version of this with one of the special order 3rd order 70Hz MZF8000. Unfortunately I found out at the end of last year this had been discontinued and the stock version is only designed to counteract proximity effect. I’ve got some of Rycote’s discontinued Tac!t cables for this, which have an active 3rd order 60Hz filter in.

Windshielding

This now allows different levels of windshielding to be added to the mic. Here it is with a Rycote Softie-lite 19 as a heavier interior windshield

Suspension with Softie-lite 19 windshield

and with a full WS1 blimp

April 28, 2020February 8, 2022

‘Half rack’ compatible equipment

Standards

Just to begin, I’m sure most people looking at this post will be familiar with the 19″ rack standard, it’s a very common way of mounting professional audio and computer equipment. ‘Half rack’ is effectively mounting equipment which is half the width of that (or is it?).

Basically half rack isn’t a real standard, it’s what happens when you get equipment where 2 units are designed to fit in one 19″ unit side by side, however it just so happens that you put rack ears on one of them, they seem to be the right size and a few manufacturers have made racks for this equipment.

19 divided by 2 is…

So, how wide is this equipment then? The original 19″ rack width includes the rack ears. Here, the rack ears are the same width, so it’s going to end up being slightly wider than 9.5″- in fact 10.5″ seems to be a standard.

What fits?

As it’s not a proper standard, I thought I’d start compiling a list on here. Please feel free to add stuff if you find it. Also as it’s not a proper standard, they may be some specific combinations which don’t work (especially if there are holes for screws rather than cage nuts)

* items I’ve not personally tried, however the measurements look correct

Racks

Soundcart Production Sound Cart and MiniCart
Swan Flight 10.5″ Half Rack Cases
Odyssey 2U Half Rack Case*
Thomann Flyht Pro 9.5″ Racks*
Gator G-Tour Half rack cases*

Rack Parts

Allmetalparts 10.5″ AV Rack components
Odyssey Half Rack Mountable 3U Drawer*
Thomann Flyht Pro 9.5″ accessories*

Equipment

Sennheiser Evolution half rack transmitters and receivers with GA3 kit
RME half rack units do fit, although I had to ask Synthax about specific rack ears
MOTU half rack units in an ‘ultralite‘ style box. Rack ears come supplied

Quite a lot of items will also screw into 10.5″ rack trays, including Sound Devices 970/270i rack recorders

April 24, 2020February 8, 2022

Adapting a network switch for location use

What do you need one of those for?

We’re starting to see more useful products using audio over IP, all of which need switches to connect multiple devices together. This is effectively a patchbay for these devices, however the routing is determined in the devices themselves, while the switch is effectively the patch cables sending the data to and from the right destinations. Here’s an overview of what a switch does and how it works: https://www.oreilly.com/library/view/ethernet-switches/9781449367299/ch01.html

Specifications

I’m mainly going to be using this one using Audiante Dante, which is a proprietary audio over IP system. This will work using most off the shelf switches, however they do recommend using a managed switch, especially if there’s going to be other data on the network. It need to be able to carry gigabit ethernet, supply power over ethernet, have over 5 locking ports and run off 12V DC power. And not make any noise.

Unfortunately I couldn’t find any switches on the market that fulfilled all those specs, so I’ll have to adapt one

Starting points

To begin with, I need a switch. One that ticked most of the boxes was the Cisco SG250-08HP. The Cisco SG250 series seem to be one of the more frequently recommended switches used for Dante networks. This is the smallest one in the series which can supply PoE. However, it comes with a (rather hefty) 52V PSU. It also doesn’t come with ethercon connectors, they seem to only be on some more specialised switches for the AV industry (all of which seem to have mains supplies)

Power

In order to get the switch to run off 12V, it needs the voltage to be increased with a DC-DC Converter to something approaching the 54V specified on the input of the switch. I say approaching, because I had a 48V DC-DC converter which I’d previously been using to power another PoE switch. It also already had the Kycon KPPX-4P connector on and had the same wiring (which is fortunate as that took some finding…).

I’m not sure exactly what the power draw of the switch is, however it can supply up to 45W of PoE. The mains power supply which comes with the switch is just short of 60W (54V at 1.1A).

The DC-DC converter I’ve used is only 40W at 48V- the switch seems to run fine on it though, however there isn’t enough power to run the full capacity of PoE devices. There is a bigger 60W model, however it is bigger and may be difficult fitting it into the case.

Casing

My intended use for this is to mount it on a Soundcart MiniCart. This has 10.5″ ‘half rack’ mounting points, which seems to be a ‘half-standard’, as there’s another 9.5″ half rack spacing. This seems to work with most boxes which are designed to be mounted side by side in a rack by putting both rack rears on the same box, although they’re not really designed for this.

I’ve used some off the shelf parts from a metal fabrication company, however there were a few issues getting things to fit- all 8 ethercon ports won’t fit on the panel and still get in the enclosure- and the 4 pin power connector just fits, with some filing down of the sides! The case itself is steel, so has a fair bit of weight to it.

Connectors and Cables

I opted for the ‘feedthrough’ ethercon connector as otherwise there was a lot of soldering involved. I also realised that regular moulded RJ45 cables wouldn’t fit in the chassis, so had to swap them out for low profile flat ones. This put the build of this back a week. These are probably less hardwearing, however won’t be moving about in the chassis.

I did consider some of the more weatherproof connectors, and even bought some flip down panels– however they were fouling each other when packed together that densely. Given there are holes in the chassis for airflow (which the switch and PSU will need), it seemed a bit much fully weatherproofing the connectors, which would have added further cost

Layout

It ended up all being a bit tighter than expected- I originally imagined the switch ports behind the ethercon connectors. This way, however does allow you to look through the side vents to see LEDs

I want one!

I’ve had a few expressions of interest in people buying these- there might need to be a bit of refinement of the metalwork, including a little more space and an extra output for the 8th port. This is currently a prototype but let me know if you’re interested in one- it’ll need a few refinements, but I could potentially get a run of these put together. Back of an envelope costs are around £300+vat for the whole thing

April 19, 2020February 8, 2022

‘One cable’ audio networking connections for analogue devices using PoE

Why?

I’m looking into running out a single cable to connect a box to an audio over IP network. For example, a transmitter or receiver to cover a separate area. There are a number of different AOIP standards, which include Audinate Dante, Ravenna and AVB- this is mainly a demonstration of physical connections, but different hardware may be required for different networking standards. This method can be scaled up to devices with more analogue/digital inputs and outputs as long as they can be powered.

Ingredients

You need the following:
Target device with a DC power input (in this case a Sennheiser SR300 G2 transmitter)
Network switch with power over Ethernet (PoE) (Cisco SG250-08HP)
PoE or DC powered AoIP to analogue endpoint (Dante AVIO)
Passive PoE switch (Linovision mini passive switch)
12V PoE Splitter (I used this)
CAT5e cables (or better)

Setup

Here we have the main PoE switch connected to the passive PoE switch, which powers that, which in turn powers both the Dante AVIO and Ethernet splitter. The Dante AVIO runs to 2 XLR outs which plug into the Sennheiser transmitter and the PoE Splitter feeds 12V to the transmitter. It powers up as soon as the main switch has booted up. The PoE Splitter also has an RJ45 connector on. If this was a G3 or later Sennheiser transmitter, this could connect to the data port and the transmitter’s settings could be accessed over the network.

Things to bear in mind

Everything in the network needs a certain amount of power and the PoE switch can only supply a certain amount (usually at 48V). Check the specs for power consumption of devices. Ohm’s law is useful for calculating current at different voltages

Look at the required voltage and current for the device you want to power. Voltage has to be within the specified range, while the supplied current can be more than specified (devices only draw as much current as they require). Check that you can find a PoE splitter that fits these specifications.

There are different PoE standards, the current ones are 802.11af or 802.11at which supply different amounts of power per port. Some things which just say ‘PoE’ without a standard may not play nicely or even damage equipment.

There will be some loss of power through the cable. The reason the voltage is reasonably high (48V) is to reduce the current and associated loss due to heating the cable- the longer and thinner the cable, the greater the loss.

The passive switch used here is only 10/100Mbps, this is fine as the Dante interface is also this speed, but if more channels need to be passed though a 10/100/1000 switch may be required

Your DC powered device may have a different connector to the splitter- it’s possible to chop off the moulded connector and solder on a different one

What’s the passive switch for?

In this case the PoE Splitter converts all the power on the port, so it doesn’t pass through to the RJ45 connector, so 2 network ports are required. This also shows how this could work with a box with a separate network port for control. Bigger passive PoE switches are available, so multiple interfaces could be run from one long cable. All this stuff can go in the back of a rack case with a more robust ethercon connector dealing with the connection with the main switch on a long cable.

September 18, 2019February 8, 2022

Sonosax SX-M2D2

Sonosax have offered me a pre-release SX-M2D2 for beta testing, I’ve had a few days to poke around it and try to find where it falls over, so this doesn’t happen in the production version.

SX-M2D2 connected to 2x Sennheiser MKH8040 and Wisycom MCR42 receiver

What is it?

On the Sonosax website, it’s placed under ‘preamplifiers’. It is a 2 channel preamp, using the current R4+ / AD8+ style preamps with dual ADC, which give you a lot of clean gain. Personally, I really like Sonosax preamp design- they’re very clean and seem to add less noise than almost any other design I’ve heard. For the analogue purists, the dual ADC is an integral part of the preamplifier design, so it has to go into the digital domain. The SX-M32 or SX-M2 may be a good alternatives if you want to stay analogue and just want a preamp.

However, there are also other connections: an AES input and output, supporting both AES3 and AES42 and USB-C in and out (as a 2×2 interface) and it’ll run up to 192kHz, for fx recordists, audiophiles and bats.

There is also a 2 channel unbalanced output (on TA3) and a headphone amp, which sounds very clean and is happy driving 250ohm headphones

It’s a very small device, but feels quite weighty and sturdily built. I did have thoughts of putting it on a boompole, but it’s a bit too substantial for that.

Digital bits

This is where it gets interesting and makes it far more than just a preamp. Software-wise there’s actually quite a lot going on, there’s a full routing matrix and mixer. Knobs are assignable to a number of functions with turns, short presses and long presses. Similar thinking to how routing works in the R4+ is applied here, with inputs and outputs routed in pairs and with 4 assignable presets. Different routings are possible on all inputs and outputs, either going via the mixer or direct from inputs.

Despite the complex routing possibilities, the menus are clear and reasonably ‘flat’, only running 3 layers deep for some of the more advanced settings. They’re generally navigated with the knobs with ‘forward’ and ‘back’ with short presses and the entire screen and knobs can be flipped.

What’s it useful for?

Initially it seemed like an add on for the R4+ ecosystem, simultaneously adding 2 more preamps and analogue outs in a small box, running in and out.

It seems like a very good way for production sound mixers using other manufacturers’ equipment to get access to Sonosax sound for cabled booms, where 2 preamps is usually enough. It also allows them to get AES42 as an input- it’s also capable of adding lots of digital gain to these signals, which users of some other equipment have found lacking.

It’s a 2×2 audio interface with AES3 in and out, so allows direct digital interfacing with a computer (or even a smartphone!) for playback, processing or recording direct to the device. It can even be a very expensive headphone jack adapter, for those who demand high quality music playback on the move.

There are a few small hardware recorders which can interface over AES3 and record timecode stamped files, making a very compact additional recorder.

Phone recording

Having a bit of a look at various bits of phone recording software, I’ve found that Apogee Metarecorder for iOS is possibly the most suitable for production sound. It’ll talk to Timecode Systems’ Ultrasync Blue over bluetooth and receive accurate timecode. Cable wise, I’ve found the neater apple lightning to USB-C cable doesn’t work, but the ‘camera connection kit’ does, I think it requires a USB host port.

Powering

We have another battery system here. It’s got a single 18650 Li-Ion cell- you may not have heard of these, but they’re some of the most common batteries in the world. They’re used in NP1 and other Li-Ion battery packs, such as the 2054 format smart batteries. They’re also in a lot of e-cigarettes, so you can buy them from ‘vape shops’, which seem to be everywhere now. It’ll charge the battery over both a USB-C input or 9-18VDC in on the hirose input, so can work with both consumer phone batteries or professional sound bag distribution systems.

Field mixer?

Given the input and routing possibilities, along with being able to pair it with a recording device- I’m wondering if it could effectively be used like a 4 channel field mixer with recorder, running a pair of radio mics into the AES3 in, mixed to the right channel on the knobs and a boom on the left out. At the moment the software doesn’t quite allow this, but possibly could with a few tweaks. It would be possible to balance the unbalanced outs, and run a mono camera return to the second analogue in and route that to headphones.

July 24, 2019February 8, 2022

Radio Mic Range Test

Aim

To put together a comparative real world test for different digital and analogue radio mic systems, whilst keeping as many variables as possible constant. With thanks to Audio Dept for providing equipment and a place to test it.

Method

A recorder (Sound Devices 688) was placed in a static position and connected to various radio microphone receivers using balanced analogue connections. Antennas were attached to a Lectrosonics D2 receiver and RF loop out cables with 50ohm BNC to SMA connectors to any other receivers in turn.

All receivers and transmitters were tuned to 607.500MHz (this was a clean frequency on the D2 scan, and covered under the shared license) and only turned on for the duration of their walk test. The Lectrosonics D2 remained powered up throughout for antenna distribution as the antennas remained attached.

For each test the transmitter was put in my left pocket and a bare sanken cos11 microphone was attached to the transmitter and a comparable gain level was set (different on all transmitters). The microphone was held in front of me at chest level (not clipped on). All transmitters were set to 50mW.

I walked out of the door (the recorder was next to it), past some vans in the car park to the gate, turning right by a brick wall and getting to a junction and then walked back along the same route.

Walk test gate to corner. The house on the left is really quite something.

While running the test I would count out paces, 40 to the gate and 100 to the corner and count on the return journey.

The Wisycom MCR42 was set to 0dBV squelch, none of the other systems had an option to adjust this.

If there were any instances of multiple receivers being able to receive the same signal, then I would use them simultaneously (in the case of Lectrosonics D2 and Lectrosonics LR receivers with an LT transmitter).

I did originally intend on using a Lectrosonics SMDB transmitter and UCR411a receiver alongside the D2 in digital hybrid mode, but unfortunately couldn’t get it to receive either SMDB or SMB transmitters in block 606.

Results

It’s a bit difficult to put these together objectively from listening to the files – especially comparing both digital and analogue systems with different settings. Different systems break up in different ways. I’m going to attempt to describe what happened on each test.

Lectrosonics LT with D2 receiver

Some pops at 26 paces, shash at 43 (behind the wall). ‘Shash’ and major dropouts to 70m when the signal dropped. Returned at 68 paces on the way back relatively clear until it got worse at 55 paces. Reasonably good again from 43paces back, but a couple of instances of ‘raised noise floor’ and low level pops.

Lectrosonics LT with LR receiver

This test is done on the same run as the LT with the D2 receiver. Single pop at 35paces, clean til 43 when shash comes in. Clicky and shashy all the way to 100paces (end of test) and loses signal when turning around, however intelligibility remains. Signal comes back at 84, but is fairly unusable until 65-55 where it’s good with a few pops- then interference comes back until 44 where it’s clean until the end

Lectrosonics DBu with D2 Receiver

This is Lectrosonics’ new digital system, dropouts on this seem to have a ‘thump’ quality to them, in that they have a low frequency element to them when they drop off, rather than the ‘blip’ from Zaxcom and Sony digital systems and may be easier to work with in post. In fact, a single dropout seemed to keep dialogue intact over the top.
In the test we got our first dropouts at 36 and 38 paces. Signal was relatively clean until 43 paces (round the wall) and then became unintelligible from dropouts apart from the odd number. Signal comes back at 70 paces on the way back and remains relatively intelligible with some dropouts until the gate at 40 when it’s clean again til returning

Zaxcom Mono-XR

Zaxcom have a number of different digital modulations which will work on their transmitters and receivers. The first I’m testing is ‘Mono-XR’, their ‘long range’ modulation. Got to 45 paces before a blip, then unintelligible until 55-60 which is intelligible, then it’s basically gone. On the way back 74-68 is intelligible, then unintelligible until 41 and it’s then clean all the way back

Zaxcom ZHD96

This is one of Zaxcom’s modulations designed for lower bandwidth and squeezing more channels into a smaller space. They also recommend this for using in more reflective environments. First dropout was at 46 and 50-55 is lost, but remained surprisingly intelligible all the way up to 80. On the way back, the signal’s properly picked up at 78 and remains intelligible until 46 where it’s clean all the way to the end

Audio Ltd A10

I’m not completely sure we had a fair test with this one, although we had vehicles coming in and out of the car park throughout the day- there was a van pulled up in the gate entrance by the time I did this test, and I’m sure it was putting out a fair bit of RF.
On this run it was clean until 25 paces when some lower level dropouts started. Made it to the gate at 40, but unintelligible by 41. Coming back 69-60 was intelligible, bigger dropouts from 59-41, then clean from 40 til the end.

Sennheiser SK5212 and Wisycom MCR42

Here the Wisycom MCR42 was put in SEN emulation mode to receive the Sennheiser SK5212 transmitter. Heading out had a single splat at 31 paces, then clean all the way to 49 paces. At 53 paces there’s quite a lot of shash on the signal, peaking at 100 and subsiding to some minor pops and shash around 85 on the way back, gradually getting less frequent until it stops at 50 on the way back to be clean the rest of the way.

Wisycom ENR

Wisycom use two of their own compander modes- ENR describes as ‘noise optimised’ and ENC described as ‘voice optimised’. This is with an MTP40S transmitter and MCR42 receiver. Single splat at 30 paces and a couple of very minor ones before 40. More pronounced interference at 49 paces, but intermittent and a good amount of useable audio all the way up to 80 paces. Again, mostly good audio with some interference from 89 paces onwards on the way back, but only being totally clean at 40 paces til the end

Wisycom ENC

This is the other Wisycom compander, which seems to work better in situations with more high frequency transients. Here we had a couple of big splats at 29 but generally minor until 49 paces. Was then bad shash until 82 on the way back and then clean again at 40 paces

Graph

I made a graph! It took far too long

I’ve roughly divided the audio into what I consider clean (green), Minor dropouts (over 95% good audio), Major dropouts/shash (intelligible but not useable audio) and red for either totally unintelligible or no audio. Distance along the bottom is how far along the route, in paces (sorry, not an SI unit) with 100 paces being the furthest point.

Conclusions

I think there are a number of ways you can look at these results. Basically orange is only useful for comms. Yellow could be described as ‘borderline’ range- you don’t want to be there, but you may get what you need. Green is the only truly “in range” area.

It’s also worth noticing that the digital systems don’t really have any orange, and the analogue systems don’t really have red. The analogue systems keep transmitting (noisy and/or distorted) audio, while the digital systems just fall over. All three digital systems seem to fall over in a different way, though- Audio Ltd seem to have a short fade out, Lectrosonics add some low frequency, while Zaxcoms make a ‘blip’ noise. The Lectrosonics method seems to psychoacoustically cover the gap a bit – it’s harder to distinguish it between something like cable movement and a dropout, though.

Range-wise it seems like as far as picking up useable audio is concerned that the Wisycom ENR came up top, however the Wisycom in SEN combined with the Sennheiser transmitter had the highest ‘green’ percentage. The Lectrosonics LR/LT also had a healthy ‘green’ percentage, and although there’s a fair bit of orange on there, it’s mainly from consistently raised noise floor and audio was lost because squelch can’t be disabled. Would be interested in how a SMDB/411 combo works in comparison to this.

On the digital side of things, the biggest surprise for me was the ZHD96 performance. longest run from the start before a dropout and seemed to keep fairly consistent audio up to 80 paces. I actually expected the other zaxcom modulation to do better.

Solid performance from the Lectrosonics digital, especially on the way back. It seems the the way it deals with the odd dropout is good, but once you get quite a few it’s mush.

It’s also interesting how the D2 dealt with hybrid and digital signals differently, it didn’t do as well as the LR, but it seemed to work in some places where the digital receiver didn’t.

I’m also not sure the A10 got a fair run out here, I think this test may not have worked to its strengths and I’ve heard there are significant performance increases with a recent firmware update, which wasn’t installed.

Further tests

Don’t take this as the be-all and end-all. It’s just one test and probably do with being repeated and may have different results even in the same place on a different day. It wasn’t in a stable RF environment and with the amount of obstacles and metal around was very challenging. It also only tested one transmitter, things could get very different with a bunch of them out, for both analogue and digital systems.

This also doesn’t take into account sound or features, which may both be more important to some users.

If anyone really wants to listen to me counting in a car park, I can send over the recordings (and hide track IDs for blind comparison)

July 7, 2019February 8, 2022

Radio Microphones 2019 update

I’m now having another look at radio microphones, and a few things have changed since the 2011 comparison. There haven’t been a lot of changes to the base technology, although a number of companies have ‘gone digital’ and there have been a number of refinements made. I’m going to comment a bit more about newer features and how useful they are in the ‘real world’.

Analogue and Digital

Audio Limited have gone full digital now with their A10 system and Sennheiser have their 6000 and 9000 series. Lectrosonics have re-launched a new digital system, the D squared (which I’m yet to test). The Sennheiser EK6042 receiver is backwards compatible with Sennheiser analogue systems, however they perform similarly to the digital transmitters. The Lectrosonics D2 can also do this (with Lectrosonics digital hybrid transmitters) and I’ve not been able to compare performance with an analogue receiver.

Digital does have some advantages and disadvantages compared to analogue systems, for more details see Analogue vs Digital Wireless.

Some of the more established digital manufacturers (Sony and Zaxcom) now offer choices of modulation, some of which can offer better range or ability to squeeze more transmitters into an even tighter frequency range, however this is at the cost of higher latency.

Recording Transmitters

There are some jobs where this is the only option and back in 2011, the only option for this was Zaxcom. Zaxcom also have a US patent on this which has stifled competition in the US. Audio Limited now have recorders on the A10 system (enabled outside the US only) and some of the newer Lectrosonics transmitters can be enabled to either record or transmit.

There is also an advantage to the Zaxcom system, in that they simultaneously be timecode jammed and record enabled over Zaxcom’s 2.4GHz ‘Zaxnet’ control frequency. This can save a lot of time over individually jamming a number of transmitters manually, as is currently the case with other manufacturers’ systems.

Small Transmitters

For some time, the limitation in how small transmitters could be made was down to the size of the batteries that could be put in them. We’ve now got access to smaller Li-Ion batteries with higher energy density and a number of manufacturers have incorporated these into wireless (especially given the higher power draw on digital systems). There is a trade-off, however: smaller batteries=shorter run time. Some of the smaller transmitters are also more limited with their output power- I expect as a decision to retain battery life.

The Lectrosoncics SSM and Zaxcom ZMT3 both use the ‘Fujifilm NP50’ standard battery, while Sony DWT-B03 uses their own NP-BX1 camera battery. Both of these are available from consumer camera shops, however there are a lot of ‘fakes’ around which usually don’t perform as well.

Sennheiser use their own proprietary batteries, which are less readily available, however they have the advantage of providing accurate runtime telemetry.

Another company that’s worth mentioning is Q5x, and although they do make their own receivers, really specialise in transmitters. They tend to be very small and flexible so can be fallen on without injury, for example in sports or during stunts. They use analogue transmission and can be received by wisycom and lectrosonics. However a drawback is that the batteries are built in to the transmitters, so they can’t be swapped out in a shooting day, they only solution they have for this is a secondary battery which can be plugged in.

Bandwidth

This is something which seems to have increased across the board. It was only really Wisycom and Audio Wireless providing proper wideband systems before, however now most manufacturers have at least 75MHz to play with, with some of the Wisycom receivers now going from 470MHz right up to 1.1GHz ‘Air band’.

This can allow greater flexibility, both with larger jobs in the UK where site specific licenses are required, or jobs in other countries where the clearest spectrum may be somewhere else

Close frequency Co-ordination

Something which has been said about digital systems is that frequency co-ordination isn’t something you need to thing about any more (I’m not entirely convinced about this), however there have been significant improvements in analogue systems too. Sennheiser introduced an intermodulation suppression mode in the SK5212-II and Wisycom seem to have taken this further in their newer ‘linear’ transmitters.

The ZHD modulations seem to have allowed zaxcom to squeeze even more frequencies in on digital, however this is at the cost of much higher latency and can only do this with 1 channel per receiver using ZHD48

Remote Control

This actually hasn’t changed much, but can be something which makes a big difference, especially with drama when there are costumes which are awkward to get at. There seem to be a few different ways of doing this:

Lectrosonics use the mic itself to play a ‘dweedle tone’ down. It a modulated audio frequency carrier (like timecode), which tells whichever transmitters that can ‘hear’ it to change a parameter (e.g. frequency, gain etc). Although it’s a bit clunky in some ways, it works rather well and the actors actually realise you’re doing something.

Sony and Zaxcom both use a 2.4GHz signal to remote control their transmitters. It’s possible to monitor and remote control a large number of transmitters using Sony’s rack receiver and a computer. A 3rd party program was made to do the same thing with Zaxcom, but it’s now been discontinued- however Zaxcom did show off something at this year’s NAB.

Audio Ltd also use a 2.4GHz communication system involving bluetooth and a phone app.

Serial Communication

We’re now getting some different recorders and radio mics talking to one another. Zaxcom have been doing this for a while, however it’s only been between their own products.

A few years ago, Sound Devices launched ‘Super Slot‘ as a standard. However, it’s only a connection standard:
“The protocol of serial communication is outside the scope of this electromechanical
specification. In addition to the sample commands listed below, Sound Devices will work to accommodate manufacturers’ existing command sets and protocols.”

So, although there is a mechanical standard, it requires the sound devices firmware to be able to interpret whatever wireless manufacturers output or can receive from their devices. At the moment, communication is possible between the Sound Devices SL-6 and the Audio ltd A10, Lectrosonics SRb and SRc, Sennheiser EK6042 and Wisycom MCR42.

Aaton have also made a serial connection system they call ‘Hydra’ which will allow the Cantar X3 and mini to talk to Audio ltd A10, Lectrosonics SRb and SRc, Sennheiser EK6042, Sony DWR-S02 and Wisycom MCR42.

The Mac OS program Wavetool will also talk to a number of (mostly rackmount) receivers in a similar way and can stream audio.

Comparison

I’ve put together a small table comparing features of different systems, as a bit of a round-up. Digital doesn’t necessarily mean good and all the different systems have their advantages and disadvantages. Also (apart from whether they’re digital or not), this doesn’t really have any bearing on sound or performance, it can be quite subjective. Just because a system has all the bells and whistles does not necessarily mean it’ll perform as well as another.

	Digital	Recording TX	Small TX	100MHZ+ Bandwidth	Remote Control	Serial Comms*
Audio Ltd A10	x	x			x	x
Audio Wireless				x
Lectrosonics Digital Hybrid		Either/Or	x		x	x
Sennheiser 6000/9000	x		x	x		x
Sony DWX	x		x	x	x	x
Wisycom				x		x
Zaxcom	x	x	x	x	x

*with other manufacturers’ equipment