Sound Devices MixPre-6 II (second generation)

I’d never think four principal mics would handle all that plus your atmosphere mic for a client. Gives me hope that the two mics I have can give me something good from a few kids. Plus the AVX on the principal vocalist and MKH416 for atmosphere.

Once I unofficially recorded a similar symphony from the box seats with one mic. Of course I can’t ever use that audio, but I played it a lot as I was surprised how well that turned out!


Sent from my iPhone using Tapatalk

Hi Rick - how happy have you been with the AVX system? I need to buy another 2 channels here and was planning on going Lectrosonics but the AVX does appeal for solo operator doc work.

One issue putting me off is the delay - on the UMP there is already quite a delay in the headphones and half the time I’ll be going straight into camera not into a MixPre.

Cheers
Ian

What are your thoughts? Time to strike while the iron is hot? Or wait for the samples in detailed reviews based on real use?

Rick, just got a MixPre-6II a few days ago. I already have a MixPre-10T. Except for the additional features such as 32 bit float and the adjustments to the limiter the operation is pretty much identical. One thing that they added which makes a huge difference if you use a sound bag is they've taken a rubber wheel from a legos kit and slipped it over the headphone knob. It is now much easier to adjust when crammed into a sound bag with headphones plugged in.

I did get to try out the 32 bit float. It is amazing. The limiting factor is gonna be your microphone. I hope to give it a real world test this Monday on a rock band. As far as I can tell, the 24bit performance is identical to my 10T.

Both have excellent preamps and IMO well worth the money.

I got my MixPre-6 II this week and love it! One of my primary uses will be run-and-gun with my UMP. The UMP's audio circuitry is not bad, but two channels is not enough. The MixPre amps and limiters of course are superior and configurable. However I dream of what the 32bit float capability promises — no worries about clipping, even without resorting to limiting.

To try things out, I turned my Wooden Camera top handle sideways, pointing left and affixed the MixPre to the underside of the top handle, with 43Wh L-mount battery for power. Then I plugged in a Rode shotgun, an AT BP4025 stereo mic, and a Shure dynamic via a radio (not shown). The MixPre is via a wire the timecode source for the UMP. At 96/32float it sounds great.

I also did an informal power endurance test by recording four channels, 3 with phantom, until the battery died. The 43Wh battery lasted 4.5 hours.

Resolve's inability to process 32bit float audio will probably force me to record at 24bits. The gnarly problem is how to normalize the levels on the 32bit float recording, without losing the timecode. This is complicated by the fact that the MixPre's audio files split every 4GB. To normalize long takes, you then need to first concatenate the audio tracks. Meanwhile, the timecode seems to be a casualty in the software that I have. Audacity has a terrific normalization capability, but Audacity does not do timecodes, nor does it recognize and preserve most broadcast WAV metadata. Final Cut Pro can with some effort concatenate and export tracks, but the timecode disappears. WaveAgent can reinstate the timecode, but only seems to recognize files smaller than 4GB.

I was forgetting Audacity doesn’t handle Timecode. Does Reaper keep the Timecode?

Edit
I keep my MixPre on a mic stand, not on or under the camera. Of course that’s when I’m tripod bound anyway.

Sent from my iPhone using Tapatalk

Ian, I don’t mind the 19msec delay as I seldom put headphones on, looking at audio levels though. The AVX works well with the talent in line of sight. And through some thin barriers as found on a set. Today was the first day I was using the AVX and the EW500 G4 Boom. Haven’t listened to them as I was exhausted. Couldn’t get levels for either in the MixPre even though I did on my testing. Bit of a mystery but the Line Audio Design CM-4 mics seemed great.


Sent from my iPhone using Tapatalk

So i switched from F4 to Pre6 II and i think it's awesome. It is way simpler to use and just a better UI and approach. The float is fantastic as i did a side by side with the Zoom. Does what it says. Life will be a little easier now!

I need Resolve to support float properly though, i can use audition but clearly resolve is the right place for it.

As for TC, i jam sync'd off a tentacle and after an hour at 25p it's still locked. I am turning the device off as well, not leaving it on but i am powered via USB C. I will keep checking.

What i'd be tempted to do though is perhaps drive the tentacle with the TC out from the recorder and getting that master tentacle to sync the slaves on camera. That should work shouldn't it? I'm just a bit wary of jamming TC but will see how that goes... Plus i like t he idea of 2 additional tracks from the Aux In...

cheers
Paul

I repeatedly felt that jamming the Tentacle from the MixPre 3.5mm Stereo Out seemed problematic and using the Aux1 In to jam the MixPre from my master Tentacle was better. Either way should be fine, but I’ll temporarily tie up Aux1 rather than the Stereo Out.


Sent from my iPhone using Tapatalk

One big difference between sound devices and other manufacturers is RF noise rejection. On sets their is more RF noise than ever. I have been having increasing noise issues with the zoom F8 due to RF noise.

Sound Devices has great design for RF noise rejection.

I just got my Sound Devices MixPre6 II from Adorama after a wait of 10 days on backorder.

I had done some ADR for a class assignment using the school's Zoom F8 recorder (same preamp as the F6) and an Azden shotgun Mic at 12 inches away in a college computer lab. When I tried using it at a workstation that was located five feet from fluorescent lights, all I got was a loud hum and could barely hear my voice, no matter how much I moved the xlr cable. I then moved to another workstation located seven feet from the overhead fluorescent lights and was able to get a recording, Occasionally I would get a lighter hum that went away when I twisted the school's xlr cable.

I re-recorded the ADR using the new MixPre6 II at 24 bit, with my new Sennheiser MKE-600 supercardiod Mic at 12 inches with a lightly used Kopul 5000 xlr cable. I did it at the same location that was five feet away from the lights that produced the loud hum before. To my surprise it sounded quite nice, and more natural than the F8, with the airy sound that you get from a preamp with real headroom. It was usable in spite of the harsh RF interference.

Some photos of MixPre-3 (first generation) internal components from this source https://fccid.io/2AKLX-739M3/Internal-P ... os-3297643

Resolve's inability to process 32bit float audio will probably force me to record at 24bits.

I've been checking in periodically to see if there is any sign of progress on this. Has Blackmagic acknowledged that they are aware of the 32 bit float problem, or made any statement that they are working to address it?

I've been checking in periodically to see if there is any sign of progress on this. Has Blackmagic acknowledged that they are aware of the 32 bit float problem, or made any statement that they are working to address it?

Not yet, at least not publicly as far as I've seen.

Has anyone confirmed that Reaper will preserve Timecode values on their output? I’m going to try hopefully this week but working on some other issues as a top priority. Hope it’s in Resolve 16.1 GM.


Sent from my iPhone using Tapatalk

Looking at photos of MixPre-3 internals i can't understand where are all those well known analogue transformers that amplify microphone signal? Are they change something and amplification is done in completely different way now? Maybe that is a reason why some people noticed during testing that new models sounds different with less character?
Here is image of older "Classic" MixPre to compare. You can clearly see two large transformers.

Looks like metal XLR sockets replaced with plastic too.

Those Lundahls are specific for microphones.
They convert balanced in to unbalanced. They galvanically separate its input side from the output making sure no signals travel through ground.
As they are transformers they are very hard to overload and saturate. They handle +12dBU quite ok.

On the question of power to the MixPre-6 II:
The DC Power supplies 5V. Okay.

The recorder can run from a four battery sled. The batteries can be 1.5V. Used serially, the batteries are 6V one time use or about 5V rechargeable. I have a Li-ion.

I bought larger Sony L-series batteries so I wouldn’t have to think about batteries on a very long day. The salesman over the phone suggested the Sound Devices XL-B3 and I ordered two with charger. The online documentation for the MixPre says you need to mount two of the large batteries.

Today I got them and each battery is 7.4V 7800mAh. It looks like it will take forever and a day to charge one battery in the puny single battery charger.

Does anyone know if both batteries are required to be mounted in the MixPre sled for proper operation?

If both are required, just how many hours will these batteries last between charges, 600 hours!?

Thanks!


Sent from my iPhone using Tapatalk

No Rick,

I always use one on the top slot so that the unit can sit flush on a surface.

Watson makes an expensive dual charger that shows levels ... works great and will charge both in
a couple of hours.

https://www.bhphotovideo.com/c/product/ ... r_for.html

The battery plates are changeable ... for Sony, Nikon, Canon etc.

The 7.4v li-ion packs you can mount are hot swappable.
You can run a single is what ever mount you like and add the second one before you remove the first if it’s depleted. Or you run them both at the same time.

When I asked about the first series the battery mount accepts 4-8.4v input.
So the third party sleds which offers 10-30v Hirose input simply run a dc-dc inside them to reduce that down to 8V.

Thanks for the feedback. I have been charging a single battery for over eight hours! Still a red light.


Sent from my iPhone using Tapatalk

Rick, the Atomos Fast Charger for the NP-F batteries charges a 5200mAh in 3 hours, which is 1733mAh per hour. For your 7800mAh it would need 4.5 hours. So I guess with your slow charger it will take 13.5 hours for your battery.

The review on B&H on the Sound Devices charger says it all:
https://www.bhphotovideo.com/c/product/ ... ml/reviews

So i recover some info from PCB photos and found this:

AKM AK5558VN A/D chip https://www.akm.com/global/en/products/ ... /ak5558vn/

ZYNQ 7000 SoC FPGA chip https://www.xilinx.com/products/silicon ... -7000.html

Looks like metal XLR sockets replaced with plastic too.

Less weight. Also those Neutrik XLRs are made of Polyamide (Nylon). It is very rugged heat resistant industrial material. https://www.neutrik.com/en/product/ncj6fa-h

To be honest i never liked metal XLR connectors for field recording work. Earlier i done a lot of modifications to XLRs shells (original Neutriks) and it appears that they are made of very soft and heavy alloy.

Robert, thanks. Both batteries took just over nine hours for their first charge. I’ll be using them Saturday on a shoot perhaps two hours untethered. The rest of the time the MixPre will be running with the DC Power supply.


Sent from my iPhone using Tapatalk

The mixpre first gens uses the AK557-series ADC.

The 3 is running a single AK5576 in a 6 to 3 summing setup.
The 6 uses the AK5578 in a 8 to 4 setup.
The 10 uses two AK5578 in 8 to 4 summing. One for each side of xlr inputs.

my mistake, but looks more like AK5558VN

By the way, Tascam DR-100MKIII also use dual AKM chips to extend dynamic range, but it records only to 24bit. So formally this technology is not extremely new.

my mistake, but looks more like AK5558VN

Yes I know them from the FCC photos.
Those where most likely from the development stage. Probably due to the 557 series were not yet released or early in production.

The AK557x series was confirmed by Paul Isaacs himself.
https://www.gearslutz.com/board/showpos ... tcount=383

Has anyone confirmed that Reaper will preserve Timecode values on their output? I’m going to try hopefully this week but working on some other issues as a top priority. Hope it’s in Resolve 16.1 GM.

I haven't had a chance to try it yet... argh. Too much to do, not enough time...

For now I'm planning on just staying with 24-bit recording, and hoping that Resolve gains 32-bit audio support soon.

Even if Reaper works, it would still make for a clunky workflow.

I have played around with Reaper a bit (I normally use it just for recording voice overs, so I'm no expert with Reaper).

For my tests I have used the Sound Devices 32_bit_float.WAV sample they provide on their website.

Make sure you set the project settings to the same frame rate like your footage.

Reaper can read the embedded timecode.

If you import your clips, you can right click and from "item processing" select "move items to source preferred position (BWF)".
That will place each clip to the correct timecode position of the track timeline.

Then I used "Normalize Items"

For exporting individual files:
Select them all. Go to "File -> Render".
In the "Render to File" dialog box set:

Source = Selected media items
... all selected items will be exported

[x] Tail = 0 ms
otherwise you will get a tail added

File name = $item
... that will restore the individual clip name when exporting, found under the button "Wildcards"

I was successfully able to render the 32-bit clip into a 24-bit normalized file, re-import it and by using "move items to source preferred position (BWF)" it got aligned correctly.

Any info how 32Bit Float will work with compressors and software limiters in real life?
If your source have a lot of really loud splashes and you bring back loudest part to visible range in one click, other parts of track gain became too low. Simple normalize tool for selected regions only or manual gain adjust with curve is ok for simple situations, but for instant random loud splaches you need some kind of automatic limiter. So will those software plugins see that hidden dynamic range in 32Bit mode? Is it possible that software limiters in 32Bit mode will produce same result as analogue limiters or even better?

So I downloaded samples from SoundDevices website again and play with very simple AU limiter plugin in Cocos Reaper. Works great. All you need do is set attack/release time. Artifacts-free real time recovery!

So Cocos Reaper is the AU plugin for Resolve?


Sent from my iPhone using Tapatalk

Cockos Reaper is standalone audio editor https://www.reaper.fm/

Here are processed examples:
https://www.dropbox.com/sh/rm6c9sdgimeo ... gKG7a?dl=0

Thanks, Dmitry. I was pretty tired this morning and after the shoot today, I’m dead now so hardly more functional. Maybe tomorrow. Your Cocos post makes perfect sense now but threw me this morning!


Sent from my iPhone using Tapatalk

I preferred MixPre-3II over 6II because less cost, smaller form factor and more free space around rotary headphones button. MixPre-3 body also looks less ugly than MixPre-6. I can understand that lack of combo XLRs is due smaller size, but seems there is one more very odd artificial limitation of MixPre-3 series.

RECORDING: MixPre-6ii = 8 Tracks: stereo mix + 6 ISOs
RECORDING: MixPre-3ii = 5 Tracks: stereo mix + 3 ISO’s

My idea was to keep classic MixPre preamp, and use ultra compact MixPre-3II for everyday stereo field recording. And in some rare situations when i need more mic inputs i was thinking to plug MixPre to MixPre-3II line in and have 5 mic inputs total. But seems it will not work like this with MixPre-3II. It can record only 3 separate inputs.

Also if someone interested to read, here is Sound Devices patent application "High dynamic range analog-to-digital conversion with selective regression based data repair " https://patents.google.com/patent/US9654134B2/en

I wonder this hole fuss about the 32 bit float. Do not understand why float is needed or why 32 bits is needed. 24-bit digital audio affords 144 dB dynamic range and that is more than needed, unless recording whisper and gunshot with one microphone, but anyway more than the microphone can handle. True though that the current 24 bit systems do not archive the 144 db, but maybe that is the aspect that should be improved.

That said though, it is easy trick to feed mono microphone to camera left and right channels on different levels and always be able to recover clipping. To me it seems the 32 bit float using 3 AD, is just bad implementation of what it should be.

I do not know why I feel so negative about it now, the advantages are clear, I guess I am just doubtful of the possible side effects. Probably will get MixPre, but not sure at this time what model and if it is the original or second generation.

I wonder this hole fuss about the 32 bit float. Do not understand why float is needed or why 32 bits is needed.

This might help: https://www.sounddevices.com/32-bit-flo ... explained/

It’s early days and in time it will emerge as invaluable or over-hyped. My acoustic music video (last shoot Monday) certainly doesn’t stretch any limits of 24bit audio even without limiters! But I used 32bit to get used to the new MixPre-6 II capabilities and operational characteristics as well as the post-processing in Resolve or Reaper.

But why bother? Since I don’t have a dedicated sound guy, I set up sound levels before takes, but once recording I can only monitor the MixPre output display to see all tracks are responding—I can’t adjust faders etc and leave it to the 32bit wizardry to be able to handle the situation. That’s my main attraction.

When previously recording in the BMD camera, I fairly often would experience clipping. It’s a comfort to know those days are in the past. But the greater experience and capabilities of others recording with 24bit audio may not need to switch to 32bit as the MixPre 24bit limiters may be completely sufficient. Perhaps after I have more experience, I’ll give 24bit a try as well.

I do feel I may be guilty of jumping on 32bit prematurely given I may not be able to manage 32bit completely from the Resolve Media page through the Deliver page! In the next week I hope to see if that limitation affects me, but I do expect Resolve to include full 32bit management end-to-end this year. Since Fairlight handles it, Resolve will follow, no?

But why bother? Since I don’t have a dedicated sound guy, I set up sound levels before takes, but once recording I can only monitor the MixPre output display to see all tracks are responding—I can’t adjust faders etc and leave it to the 32bit wizardry to be able to handle the situation. That’s my main attraction.

Indeed, I think that's the main attraction. It's going to be especially useful in documentary work or recording anything that's either unrehearsed or where the person doing the recording is unfamiliar with (and thus unprepared for) the full dynamic range of what's about to unfold. In a way it's the audio equivalent of recording raw video, although it's even better than that because you don't have to worry about clipping even if an unexpectedly loud sound occurs.

Brad, looks like the marketing department missed an opportunity to promote 32bit, which is a bit stuffy and too many math implications, as “Audio raw” and not even talk about 32bit at all.


Sent from my iPhone using Tapatalk

Brad, looks like the marketing department missed an opportunity to promote 32bit, which is a bit stuffy and too many math implications, as “Audio raw” and not even talk about 32bit at all.

Ah, SDRAW and BRAW, a match made in heaven.

We need sensor with something like 5x gain circuits combined internally to something like 30 stops of dynamic range HDR BRAW to match current 32 bit HDR audio flexibility.

Thanks, I have seen that, and it probably is the main reason I feel negative about the 32 bit float. They completely fail to explain why 32 bit and why float. Well maybe the reasons are there, but one has to look very carefully.

They say "The dynamic range that can be represented by a 32-bit (floating point) file is 1528 dB. Since the greatest difference in sound pressure on Earth can be about 210 dB, from anechoic chamber to massive shockwave, 1528 dB is far beyond what will ever be required to represent acoustical sound amplitude in a computer file. "

So what is the point of having more than 210 db dynamic range. Well the 24 bit is only 144 db, but is there any microphone that can handle 144 db?

They say" There is one other aspect of 32-bit float files which is not immediately obvious. Files recorded with 32-bit float record sound where 0 dBFS of the 32-bit file lines up with 0 dBFS of the 24- or 16-bit file. "

Not sure, but maybe this is why float is useful, but not sure, if 16 bit and 24 bit integers line up, so why would 32 bit not? There is just as many different levels on 32 float and integer.

I guess the true advantage is that one can go and scale up very low level from 32 bit recording to 16 bit and maintain all the details that would have been there if it was properly recorded at 16 bit. And maybe on that the float helps, but not sure why it would.

I am engineer so I do care how things work and do get some what irritated when SoundDevices gives "32-Bit Float Files Explained" with plenty of technical details, but completely loosing what they where about to explain.

I wonder this hole fuss about the 32 bit float. Do not understand why float is needed or why 32 bits is needed.

This might help: https://www.sounddevices.com/32-bit-flo ... explained/

Who knows, maybe in future they will add 24bit float.
At least we have similar example for images - 32 float and 16 bit float EXR files which are also ok. You can try to post this idea to SD or Zoom support.

Yeah, or sound LUT and only 10 bit

Who knows, maybe in future they will add 24bit float.
At least we have similar example for images - 32 float and 16 bit float EXR files which are also ok. You can try to post this idea to SD or Zoom support.

Who knows, maybe in future they will add 24bit float.
At least we have similar example for images - 32 float and 16 bit float EXR files which are also ok. You can try to post this idea to SD or Zoom support.

24 bit float would be 16-but audio and 8-bit mantissa (resolution)
That would not really be that a step forward.

32-bit float for audio is a very good way for a lot of us.
Especially for us who does work alone, those recording volatile sound sources or simple just want a rec and forget kind of recorder.

Thanks, I have seen that, and it probably is the main reason I feel negative about the 32 bit float. They completely fail to explain why 32 bit and why float.

I'm guessing they picked 32-bit float because many DAWs operate at that resolution internally. If you're using Pro Tools or Nuendo they'll both (unless set differently) default to 32-bit float even if you're importing and playing back 24-bit fixed audio files. The conversion is in the background so we don't see it, but it's there.

Also like the previous poster wrote the actual range of the signal that is described is about 23 or 24 bits fixed in a 32-bit float file, if memory serves me. So that's maybe what they were referring to when they wrote 0dBFS float = 0dBFS fixed (16 and 24 bit).

It really seems to me that this functionality is mostly a matter of convenience.

My only worry is that it'll eventually translate to inconvenience on the audio-post side of things with plenty of recordings going above 0dBFS that then need to get significantly trimmed down. For sure better than clipping or excessive noise, but still less than ideal.

By the way, my 18650 Battery Enclosure Module can be mounted directly at the top of MixPre and combined with DC Power to USB Type-C Adapter Converter can power MixPre all day. It may be not too slim form factor, but at least feels like rugged single brick stack.

https://lavky.com/handmade/home-and-hob ... t=UW590888
https://lavky.com/handmade/home-and-hob ... t=AM429494

When I made my recent recordings using Timecode, I left the ubits all zeros. Now I’m thinking I should use them to record CC:YY:MM:DD such as 20191011 to complement the Timecode HH:MM:SS:FF. Is that of any value in metadata to sort and sync clips by Timecode recorded over a period of time?

The Timecode Mode has a downward arrow and upwards arrow to select a Mode. It defaults to U:U:U:U which seems the only appropriate mode for a date CC:YY:MM:DD. When I select Mode D:M:Y:U, it does rearrange the values automatically but that’s not the smartest sort sequence.

I’m trying to determine how best to handle audio recorded potentially spanning a period of months or years for one project. That audio will need to be synced to video.


Sent from my iPhone using Tapatalk