The last microphone I ever buy?

In the last post I wrote about my recent purchase: The Neumann U 67. And it arrived today.

I just unpacked and connected it to my Sound Devices 833. The high-pass low-cut filter on the microphone is enabled and the polar pattern is set to cardiod.
The gain is set to 44dB on the recorder, with limiter and NoiseAssist enabled at -6dB. And this is how it sounds:

Sound Device MixPre-3 II Timecode Issue

In this post I would like to talk about an irritating feature, not to say a flaw, in the Sound Devices MixPre-3 II. You can also listen to the whole post on YouTube.

Let’s take it away.

One of the features of the SoundDevices MixPre-3 is Timecode.

However, as it turns out, there seems to be a major flaw in its implementation, that is only apparent when you change presets or use the File Transfer feature. This essentially makes the timecode unusable as it deviates by seconds within a short period of time, instead of microseconds over a course of 24h.

What do I mean by this?

Timecode on the MixPre-3 is received via the 3.5mm TRS Aux input. So in order to be able to jam from an external source you need to do the following:

  1. Go to Menu, Inputs, Aux In Mode and set it to “Timecode”.
  2. Go to Menu, Timecode, set the TC Mode to “Free Run” and select the “Jam” menu.
  3. Jam the timecode from the external source, by pressing “Jam TC”. Depending on your timecode generator, you need to select either “Aux In 1” or “Aux In 2” as the “Source”.

The timecode in the MixPre-3 should now be jammed from the external source. Note down the value, that is shown in the “Diff” line of the screen.

This all works as expected.

However, as soon as you load another preset or you switch to File Transfer mode, the timecode in the MixPre-3 starts to differ from the external source.

This can easily be verified by either doing on of the following:

Option A: Go to Menu, System, File Transfer, to enter File Transfer Mode; and select Exit to leave it. Do this several times. And then go to Menu, Timecode, Jam and note that the “Diff” value between the external source and the internal timecode generator of the MixPre-3 is now different from the value you noted down earlier.

Option B: Go to Menu, Presets and save the current settings to one of the internal presets by selecting “Save to Int 1, 2, 3 or 4”. Then select the previously saved preset by selecting Menu, Presets, Load Presets and selecting the number of the internal preset you saved earlier. Repeat thus serveral times. Go back to Menu, Timecode, Jam and note that the “Diff” value between the external source and the internal timecode generator of the MixPre-3 is now different from the value that you noted down earlier when you jammed from the external source.

Option C: Go to Menu, Inputs, Aux In Mode and select “Mic” as the source. Then via pressing the “Channel 3 knob” go to Channel 3, Input and select “Off” as the source and then go back to the home screen. Then go back to Channel 3 again and this time select “Mic”. Then go to Menu, Timecode, and select “Aux In 1” or “Aux In 2” as the source, even if the correct source seems to already be selected. After a moment, the screen updates and you should now see that the “Diff” value changed from the value you previously saw when jamming to the external source.

So what does this mean in reality?

First, this means, that you have to re-jam every time you want to transfer a file to your computer via the USB-C cable.

Second, even if you were to re-jam every time you used the File Transfer mode, you could only do this when you would not use Aux input for other purposes at all. This essentially eliminates the use of the TRS input as a microphone input that you would have in a headset such as the Beyerdynamic MMX 300.

Are there workarounds? Of course. But they all come with limitations. That is probably why they are called …

Workaround 1: SoundDevices Support suggested to use a USB stick to transfer files between the MixPre-3 and the computer.

This certainly works, but involves a lot of manual work, as you need to unplug from the MixPre, plug into the computer, transfer files, then unplug from the computer and re-plug into the MixPre again and again.

Workaround 2: Do not use timecode at all.

Really? But this removes the functionality of one of the main features of the device. Instead I could then also directly connect to the computer via a different device altogether.

Workaround 3: Do not use a microphone with a TRS jack. And re-jam every time you transfer a file.

Possible. But a lot of headsets actually happen to have such a jack.

Workaround 4: Use an XLR to TRS adapter if you have to use a microphone with a TRS jack. And re-jam every time you transfer a file.

Also possible. But then you need to carry one more adapter with you. And this changed the form factor of the MixPre-3 considerably.

Workaround 5: Go to Menu, System, USB-C and change the setting to “Power Only”.

When doing this, you cannot use your MixPre-3 when you want to play back sound from your computer. This somehow defeats the workflow to record sound on the MixPre-3 with headphones on, transfer the recording to the computer and edit it and then listen back to it with the same headphones you recorded earlier with.

With most of the workarounds we need to re-jam manually every time we transfer a file. In combination of an “auto jam” mode like on the 8-series recorders this might even be a acceptable workaround.

Sound Devices Support told me, that the out-of-sync behaviour is expected when using File Transfer mode as the device would have to sync to the computer clock. This sounds plausible. But there are two things, that do not seem to fit into the picture.

Regarding the out-of-sync behaviour there has not been any explanation at all.

First, this does not explain the behaviour that the timecode also changes when selecting presets and changing Channel settings.

Second, the Sound Device 8-series does not seem to show the out-of-sync behaviour when using File Transfer mode.

I am a little bit disappointed that depsite the richness of features of the Sound Device MixPre-3 – and the price tag that comes along with it – the device cannot deliver its features at the same time.

Hopefully, Sound Devices addresses this in a future firmware update.

Regarding firmware: the behaviour can be reproduced on a Sound Devices MixPre-3 mark II with firmware version 8.0 with build number 5136. The external timecode generator is a Tentacle Sync E connected via a 3.5mm TRS cable to the mixer. The computer used runs on Windows 11 but has shown this behaviour also with Windows 10.

Conclusion:

You cannot use the MixPre-3 reliably with timecode and as a playback device and a TRS microphone at the same time when using the File Transfer mode.

Speech synthesis with the Neumann TLM 67

As a follow to my previous post, here is the first audio from the new Descript Overdub voice recorded via the Neumann TLM 67 along with a comparison to the previous overdub voices I recorded.

Descript speech synthesis 44.1kHz @ 16bit

Here is the uncompressed audio, in case you want to find out if you can hear a difference:

Descript speech synthesis 48kHz @ 24bit

In addition, I recorded a section for Librivox (in german), in case you want to hear more audio:

Wilhelm Busch, Volksmärchen, Die Zwei Brüder, 44.1kHZ @ 16bit 128kbp CBR MP3

A new addition to my microphone locker – Neumann TLM 67

Having to stay in a sound-wise medieval croft for the next couple of months could make me think to invest in proper acoustic room treatment – or: to get a new microphone. And this is what I did. I ordered the Neumann TLM 67.

Neumann TLM 67 ordered from Thomann

Originally, I was up for a Neumann U 67 Reissue. However, price and portability really made me hesitate.

So, when it arrived yesterday, I quickly tested it with a short recording from my kitchen desk (speaking directly against a window).

For the recording, I connected the microphone to my Sound Devices MixPre-3 II. The microphone was set to cardoid and had its internal high-pass filter enabled.

As I did not have a fitting microphone shock mount, I had to hold the microphone in my hand during the recording. So sorry for any unwanted noise.

As a side note: I temporarily mounted the TLM with cable ties to my Rode SMR shock mount. It works remarkebly well, so I will leave it for a while. At least as long, as have something better, such as a Rycote InVision USM.

The mixer was running NoiseAssist at -6dB and also had the high-pass filter enabled (as 120 Hz w/ 18dB/octave). The latter should not have a significant effect to it, as the built-in low-cut filter of the microphone covers a much wider range, as you can see in the diagram below:

Neumann TLM 67 Cardoid graph, taken from https://en-de.neumann.com/tlm-67#technical-data
Neumann TLM 67 Cardoid graph, taken from https://en-de.neumann.com/tlm-67#technical-data

The only change I did in post was to level the Loudness to EBU128 (-23 LUFS).

Below are two samples: the first is a short recording with freely spoken text from Instagram. The latter is the “Planet Earth” Descript sample training script from coda.io (a 30min version on YouTube and a 5min version of the raw 48kHz/24bit WAV file).

Neumann TLM 67: 1min
Neumann TLM 67: “Planet Earth” Descript Training Script 30min
Neumann TLM 67: “Planet Earth” Descript Training Script 5min 48kHz / 24bit

So far I am incredibly enthusiastic about this microphone. The sound (at least for my voice and in that environment) is incredible (and incredibly better than the other microphones I tested here). Finally, a microphone that sounds really well out of the box – even in more unforgiving environments.

As soon as the OverDub voice is ready, I will do some tests and post them so we can compare the difference in speech synthesis based on a high(er) quality microphone.

And maybe, I still go for a Neumann U 67 to see how the difference between the two mics is. We’ll see …

Shure MV7 vs Rode PodMic

So finally, today’s the day.

I bought myself a USB microphone. The Shure MV7. Actually, a dual USB / XLR microphone. I did a quick comparison with the Rode PodMic, which you can listen to in this short video:

Here is the transcript of the video:

So finally, today’s the day.

I bought myself a USB microphone. The Shure MV7. Actually, a dual USB / XLR microphone, but what you are now listening to, is the XLR version of that microphone or the XLR output of this microphone, which is connected to a Sound Devices MixPre-3 mk II.

Gain is set to 67 dB, high-pass filter of 100 Hz applied.

NoiseAssist plugin is enabled at -6 dB, because I’m recording here in my living room with actually no sound treatment at all.

So that the whole thing will not become too dry. I will be comparing it to another mic.

The Rode PodMic, which you’re listening to right now.

Gain is set at 65 dB, high-pass filter is also set at 100 Hz. NoiseAssist is enabled at -6 dB.

And, maybe the main difference:

The Rode PodMic has an integrated pop screen, which is only sort of effective;

and the Shure MV7 has the wind foam on top of it, which of course was installed.

So, what do you think? Any differences in sound? What microphone does sound better to you?

Microphone comparison in an untreated room

In the next couple of months I will be in a sound-wise unfriendly environment where recordings are very likely to suffer. As we will have to produce “live” there will be no time for post processing. But as I still want to achieve some kind of intelligible sound, I conducted some tests for find a more forgiving microphone for that environment.

I used a 1 minute sample text from Voices (according to the website free to use) to compare the different microphones and their sound (with windows and doors were open).

All microphones have been recorded with the Sound Device MixPre-3 II with HighPass filter (at 80Hz) and NoiseAssist enabled (at -6dB). For the ribbon microphones Rode NTR and Beyerdynamic M160 the HighPass filter was set to 120Hz. All samples have been levelled to -23LUFS.

Microphones tested:

  1. Beyerdynamic M160 (ribbon)
  2. Neumann TLM103 (large condenser)
  3. Rode NT-1 (large condenser)
  4. Rode NTR (ribbon)
  5. StamAudio SA-47 (large condenser, tube)
  6. StamAudio SA-47FET (large condenser)
  7. WarmAudio WA-87 (large condenser)

In the video you see the visual representation of the audio samples, for which I used WaveLab Pro 10.

If I had to pick my favourite mic, I would go for the Beyerdynamic M160 (when listening on the headphones). However, when listening to the samples on a TV, I would go for the Rode NT-1 or the TLM103.

What is your favourite mic and why?

Sound Devices 8-series Firmware v8.90 bug leads to disabled Channel Input after Recording

The other day I finally got my new sound Devices A20 Mini and connected it to my Scorpio via A10-RX / SL-2. After a quick test recording I came across a small firmware error (that actually has nothing to do with the A20).

[Update 2020-05-21] Sound Devices confirmed to me that they are aware of this and seem to have no intentions to address this due to its rare occurrence. Something I can understand and live with it. [Update End]

Normally, after PowerOn the 8-series asks you if you wanted to create a new (daily) folder. Something which I normally confirm. In addition, if you leave the recorder running after midnight, upon stopping the next recording, the 8-series asks you again, if you wanted to create a new folder (or keep recording to the same folder). So far, so good.

So now to the error: whenever we record something on the 8-series and select any Channel the actual (physical) input to that channel is greyed out (as it makes no sense to change any of its settings. As soon, as the recording has finished, the input menu becomes active againg available for modification.

However, when the 8-series asks for a new recording folder, after a recording has finished, that input menu stays disabled. After leaving the Channel screen and re-entering it, the input is active as normal.

And this is all to it. Only a minor nuisance. But maybe Sound Devices will fix it in a future firmware update.

See the following video for a reproduction of the error:

Sound Devices 8 series v8.90: Channel input disabled after recording when selecting a new folder

Powering the Sound Devices MixPre-10 II via a MyVolts Hirose DC Adapter and a USB-C PowerBank

The Sound Devices MixPre-10 II unfortunately does not support power via USB, but either gives you power options via the battery sled or its Hirose 4-pin adapter. As normal NiMH batteries last to something like zero and the original AC/DC adapter is a bit bulky, I looked for an alternative, which I found at MyVolts with their Hirose DC adapter.

This adapter provides a 2.1mm x 5.1mm socket, so you can plug in virtually any AC/DC adapter as long as it supplies a voltage between 9 V and 18V (and of course a bit of current). For this to work, you have to set the “Ext Power” option in the “Power” menu to either “Full Range” or “12V Ext DC”.

I tested this with a Blackmagic Design 12V adapter which supplies a current of 1 A. With that I could switch all 8 inputs to 48V Phantom Power and have the NoiseAssist plugin enabled. The power meter on the home screen shows a more than 1/2 empty “green battery” when using “Full Range” and 2/3 full with text displaying “EXT”. So far so good.

USB-C Trigger Board

Taking this setup one step further, I connected a USB-C trigger board, set that fixed to 12V output voltage (check YT for a couple of videos on how to do this) and then connected a power bank with a USB-C port. Now I can use my MixPre-10 II on the go with an ordinary USB-C power bank without having to resort to a SmartBattery along with its costly adapter.

This is it for today. Happy recording.

Sound Devices Scorpio with Hirose 10-pin Output

The Sound Devices Scorpio comes with 2 Hirose 10-pin outputs that – according to the block diagram – can be configured for L, R or bus B1 to B10. Sound Devices sells this cable under the product name XL-10. However, this item is no longer sold. And my supplier in Switzerland told me, he would have nothing left in stock, but: maybe Ambient in Germany could still get hold of some cables. One call later, and I found out that these guys neither had them for order, but they manufactured and distributed an exact clone of that cable under the catchy name of HBS10Y10-35W. Once ordered, and waiting for the shipment completed, I could finally make use of that cable giving me an extra pair of full size XLR outputs on my Scorpio.

Sound Devices XL-10 Ambient HBS10Y10-35W
Sound Devices XL-10 Ambient HBS10Y10-35W

In the package, you actually get two cables:

  1. Hirose 10-pin coiled extension
  2. Hirose 10-pin male to 1x 3.5mm TRS jack and 2x full size XLR male jacks

Here is what the cables look like when packaged:

Sound Devices XL-10 from Audient HBS10Y10-35W
Sound Devices XL-10 from Audient HBS10Y10-35W

Mixing a Concert with a Sound Devices Scorpio

In this post I tell the story how it came that I bought myself a Behringer X32 as a “replacement” for my Sound Devices Scorpio. Ok, not really a replacement. I am totally happy with my Scorpio. Maybe more of an addition …

And here the story goes: some days ago, a colleague via a friend of mine asked me, if I had “two or three microphones”. – “Sure. What for?”, I replied. My friend could not tell. So I got in touch with the colleague who turned out to be an aspiring musician having agreed to playing a gig in a location nearby. Some questions later it surfaced, that the location nearby was a bar with “only basic sound equipment”. “In that case, a good idea to bring a mic.”, I thought. But maybe an even better idea was, to check out the bar, which turned out to have virtually no equipment except for some speakers (and a Mackie mixer no-one could operate). And besides that, we found out that neither the musicians nor the bar tender would have a sound technician who would set up the whole stuff. I roughly told them, what I would think, that they needed for playing the gig and finished my beer. End of story, I thought …

But as no good deed gets unpunished I found myself in the situation of having become *the* sound technician the band was in need for. Splendid. The only problem being, I virtually had no gear for live mixing and my Neumann Voice-Over condenser microphones were probably not the first choice for playing live in a small crowded place, where “accoustic treatment” was a foreign word. Of cource, I had a mixer. It is from Sound Devices and could easily handle the track count of the small band consisting of three members. But is Sound Devices really the go-for gear when mixing live concerts? Another question to myself left unanswered was: can live mixing be so different from recording voice-over work in a studio? Something I would have to find out.

So I checked with the requirements of the band and agreed to be there the next day to set everything up. As a preparation, I came up with a small wiring plan which largely looked like this:

  1. Keys L / Keys R via TS from Line-Out to Scorpio as Line-In to Ch01/Ch02
  2. Shure BETA 58 A for Vocals to Scorpio as Mic to Ch03
  3. Shure SM 58 A for Drums (just recording) to Scorpio as Mic to Ch04
    (Drums probably needed no amplification as the room was so small, but I placed a microphone next to it for the purpose of recording it.)
  4. Bass from amplifier Line-Out to Scorpio as Line-In to Ch05

On that very day: arriving at the scene on time I quickly found myself with a large beer, but without a band. “Ok, not my gig'”, I thought. On the other, having them showing up at the very last moment *could* raise the stress level on both sides, if I had to explain them, they would have to play “unplugged”, because of lack of time. But wait – no band starts playing at the announced time! So no stress doing the cabling and sound check in front of the audience, right? Eventually, the band manifested itself into the venue and we could start.

So easy, so good. However, inside the bar there were some “challenges” with the area where the musicians would perform (noticed, how I avoided using the word “stage”?). They were supposed to play directly in front of a large mirror mounted on a drywall in front of the toilets. Curtains to the rescue! But only *after* we would have moved a couple of tables and other stuff. A carpet being available with almost the size needed, trying to guarantee for a virtually echo-proof environment, completed the picture. Stage ready. And then the speakers … the main speakers were mounted directly where the musicicans were playing. “Oh great, we can use these speakers at the same time as monitors, so no separate monitors needed”, I thought, being an optimistic sound technician. (Sneak preview: after I connected them to the X1/X2 output of the Scorpio, and setting up the vocal microphone, at the start I constantly got feedback from the main speakers, as they were mounted *behind* the singer.) Following some *careful* hyper-cardiod Shure BETA 58 A mic placement, I therefore resorted to my trusty Sennheiser G4 wireless IEM with some Voice Technologies VT-600 headsets, and configured a separate bus without drums to get the mix to the to the band, as a separate monitor was really not a possibility here. A third speaker was placed next to the bar, maybe 10-12m into the room, giving me the opportunity to make use of my output delay option (around 2.92ms/m it is, right?) on the Scorpio’s X3 TA3F output.

After connecting all the instruments, I persuaded the band, that I would *really* appreciate, if they could do some sound check *together* instead of only playing *individually*. Good that I constantly upgraded the firmware of my Scorpio, so I could not only smooth out things a bit with an EQ, but also do some good on bass and drums with a compressor (starting with firmware version v7.0). Ready to play? Not really, as the bass guy revealed being the proud owner of a guitar, he intended to play on a song during the gig. As easy as updating my configuration, and placing the guitar on Ch06 and into the mix. Problem solved. Except we could not find a line out on the guitar amp. “Not my amp”, the bass guy explained. Strange enough, there really seemed to be none, except for some unlabelled “aux send”. After the actual connection was made, I was happy to be able to get rid of some hum and making use of my NoiseAssist plugin to attenuate some strange noise out of the amp.

Eventually, all seemed well *and* we were on time, meaning: sound was ok, Scorpio armed for recording, Remote Audio headset ready, and a fresh beer to cool down the heat. Let’s roll! What I did not calculate into the equation: no band starts on time. So grabbing another beer was the next logical step, and waiting …

In the end, it played out all surprisingly well. The band played two gigs, with the second gig being much better, than the first one. And interestingly, not only much better, but also much faster and *louder*, having me to readjust all the gain settings. Part of my new life as a live sound technician. And then for the teardown: armed with a beer offered by the bartender, I quickly put my stuff into boxes, gave some microphone technique advisory and was off to go. And this is, where the idea about buying a Behringer X32 or some other mixer comes into play.

Why would that be? I really have a bad feeling lending out my Scorpio for a future event to some band or a technician at some venue (even with insurance). Having to double-check everything on return does not feel that great either. Conclusion: I should have some gear that is not so delicate and most importantly sufficiently cheap, that an accident not only not breaks anyone’s heart, but also not the bank.

So I collected the requirements I would like to see in such a setup:

  • Minimum amount of cables required (yes, that’s fuzzy)
  • Least amount of setup and configuration needed (yep, also fuzzy)
  • Front of House must be separate from Stage
  • Personal mix should be adjustable by the artist
  • All configuration settings must be savable/restorable
  • The gear must be controllable via configuration app and/or external control surface
  • Dante for interop with other audio gear
  • Minimum of 16 channels (extendable to 32 channels)
  • Optional Hi-Z / DI inputs
  • Rack transportable
  • Less than 5k CHF (including cables)

Note ahead: I am in no way affiliated nor financially supported by any brand nor do I get any financial benefit from talking about any of the gear. In addition, every item I mention here I bought my with own money.

After some research I found that Behringer offers something that closely matched the requirements. So in the end, I opted for the following gear:

  • Behringer X32 Rack as the front of house digital mixer (with a single Ultranet port and 2 AES50 ports)
  • Behringer X-Dante extension card (swapping out the pre-shipped USB extension card)
  • Behringer X-touch control surface
  • Zoom H6 Handy Recorder to be able to record up to 6 tracks from the mixer (this device I just already happened to have, and had no use for it, so I decided to give it a “new life”)
  • Adam Hall power conditioner 230V@10A with 8 outputs
  • charger with 4 USB 3.0 ports
  • 8 port Netgear GS108PP-100EUS PoE+ switch
  • TP-Link nano router for providing IP addresses to the control surface, X32 remote port and Dante card and the optional PC for the remote app
  • 2U drawer for cables and other stuff
  • all this packed into a 8U case

For the stage part I opted for the following components:

  • Behringer SD16 stage box with 16 pre-amped (with 2 Hi-Z) inputs and 8 line outputs and with an additional 4 Ultranet ports with power distribution. Plus, the stage box has 2 AES50 ports, with port A being connected to port A of the X32 Rack via a 50m Cat5e cable drum with locking RJ45 connectors
  • 4 Behringer P-16M personal mixer connected via Ultranet to the SD16 (with no additional power cord needed), which give each artist the possibility to mix up to 16 tracks into their own mix, as channels 01-16 on the SD16 (mapped to channels 17-33 on the X32) are routed to the Ultranet channels 01-16
  • 4 Behringer B105D monitor speakers with 50W which can be connected to the P16-M via 2 TS 1/4″ in case you do not want to use the headphone output of the P16-M (also 1/4″)
  • Adam Hall power conditioner 230V@10A with 8 outputs
  • charger with 4 USB 3.0 ports
  • 1U shelf for holding the 4 P16-M mixers
  • 2U drawer for cables and other stuff
  • all this packed into a 6U case

After playing around with the X32 a little bit and coming from the Sound Devices 8-series world, I really had to “readjust” my understanding on how flexible the mixer effectively was and how it could be configured. But one thing really impressed me: how easy the overall configuration of the digital links actually was. Really plug’n’play. This also hold true for the Ultranet connections. Once I found the place in the X32, I configured the channels for Ultranet and I instantly got signal on the channels of the P16-M! Configuring the effects, I found somehow counter-intuitive: routing the signal over an FX bus, instead of just activating the desired effect on the channel itself was not something I had expected. Also the compressor on an actual channel seemed to replicate its settings over to the other inputs as well. But maybe, I have not yet figured out, how this is supposed to work. So be prepared, to read more about it in a later post.

A strange thing I experienced with the X-touch was, that every time I start up the control surface, I have to press the “Scan” encoder knob to actually connect to the X32 mixer. No automatic reconnect.

And this is the end of the story how it came I got myself an X32 mixer in addition to my highly appreciated Scorpio …