Topic: MixPre-3 II: Trying to use for two channel stereo

[Packgrog]

I'm finally trying out the Sound Devices MixPre-3 II in the hopes of using it for vinyl ripping. I've previously bee using an older RME Babyface Pro (non-FS) with an M1 MacBook Air. Results have been great, but I've yearned for proper 32-bit float support in hardware, and the option to record directly onto the device OR use as a USB interface.

Figuring this thing out has been daunting. I couldn't get iZotope RX 8 to work with it, then later tried setting the Audio MIDI Settings for that input and wound up capped out at 32/96 instead of my target of 32/192. When simply recording on the device itself, I eventually managed to get two channels recording and outputting to the attached headphones, but the resulting files wound up with 5 channels, two of which seemed like lower volume versions of the only two channels plugged in. Very confusing.

The documentation is a weird combination of overloading and sparse, so I could use some help. All I want is to be able to do a 2 channel stereo recording and have that for editing later. this 5 channel thing is beyond my comfort with editing.

And why is it that I can set the device itself to 32/192, but the USB interface doesn't support that sample rate? I have noticed that the higher sample rate allows for better accuracy in recording, and fully recognize that it's not useful for final listening.

Anyone have any pointers, here? This thing is well outside of my comfort zone.

[Ronmac]

32 bit float and SR192 will not create better rips from vinyl. Those settings may be useful for sound design and nature recordings.

i would suggest using "basic mode" and record only the stereo mix from the Sound Devices machine. If you find the manual confusing (many do) , perhaps viewing the short tutorials on the Sound Devices website or search for  "Curtis Judd" who also has excellent product overviews and tutorials on his web site.

Good luck with learning the functions and capabilities of your MixPre. They are powerful and great sounding devices, albeit with a bit of a learning curve.

[Packgrog]

192k helps with recording, even with vinyl. It's more about the accuracy of the capture than about capturing the ultrasonic frequencies. I tested this years ago, comparing 24/96 and 24/192 recordings of the same album side, converting both to redbook afterward. The one generated from the 24/192 source was noticeably better, had more sense of depth, etc.

I desire for the 32-bit float is to not have to keep fussing with gain with every recording. That, along with the allegedly superior noise handling of the MixPre inputs is the main reason I've yearned for the MixPre for so long.

My concern about using basic mode is the idea that 32-bit float can only be used with advanced mode. Also, basic mode supposedly keeps the limiters activated, which seems to lead to the startlingly poor measurements found at audiosciencereview.

So yeah, I'm trying to keep it somewhat simple, but I still want to maximize my recording settings for regular stereo input. Assistance with that would be appreciated.

[Packgrog]

OK, I think I got it. Following the directions here helped me get it limited down to just 2 channels. https://www.macaulaylibrary.org/resources/audio-recording-gear/mixpre3-setup/?doing_wp_cron=1721527471.2893540859222412109375#stereo

Working on doing some comparison with the Babyface Pro recordings to see how they sound.

[rocksuitcase]

I'm finally trying out the Sound Devices MixPre-3 II in the hopes of using it for vinyl ripping. I've previously bee using an older RME Babyface Pro (non-FS) with an M1 MacBook Air. Results have been great, but I've yearned for proper 32-bit float support in hardware, and the option to record directly onto the device OR use as a USB interface.

Figuring this thing out has been daunting. I couldn't get iZotope RX 8 to work with it, then later tried setting the Audio MIDI Settings for that input and wound up capped out at 32/96 instead of my target of 32/192. When simply recording on the device itself, I eventually managed to get two channels recording and outputting to the attached headphones, but the resulting files wound up with 5 channels, two of which seemed like lower volume versions of the only two channels plugged in. Very confusing.

The documentation is a weird combination of overloading and sparse, so I could use some help. All I want is to be able to do a 2 channel stereo recording and have that for editing later. this 5 channel thing is beyond my comfort with editing.

And why is it that I can set the device itself to 32/192, but the USB interface doesn't support that sample rate? I have noticed that the higher sample rate allows for better accuracy in recording, and fully recognize that it's not useful for final listening.

Anyone have any pointers, here? This thing is well outside of my comfort zone.

I am not a MIx-pre ownwr BUT, the bolded sentence implies you are recording the two mic channels, PLUS the "safety channels" PLUS the overall "mix" of the four sources together.
It sounds from your last post that you have figured out how to turn those extra three channels off.    +T
Good Luck

[Gutbucket]

Forgive me as the following has nothing to do with your question about operation of the MixPre-3 II.

192k helps with recording, even with vinyl. It's more about the accuracy of the capture than about capturing the ultrasonic frequencies. I tested this years ago, comparing 24/96 and 24/192 recordings of the same album side, converting both to redbook afterward. The one generated from the 24/192 source was noticeably better, had more sense of depth, etc.

32/192 is not "more accurate".  I'm not at all disregarding your testing which I've no doubt made apparent audible differences, only that the audible differences are more attributable to the differences in the specific ADC routines on the way in, subsequent filtering and processing, and not necessarily inherent to the difference in sample-rate and bit-format in itself.  Your testing and choosing to go that route is commendable and a clever and effective work-around to achieving best sound on the transfers you have made, but be aware that the difference is most likely tied to the equipment and routines you used when making that test, and not necessarily applicable to general statements about higher sampling rates and bit depth.  The same source material will fit quite easily within a less extreme format (when optimally implemented) in a bit-perfect way. 

Your conversion of both sources to redbook did eliminate any in differences conversion from digital back to analog during playback.

In other-words, if you change the analog-to-digital interface and/or your conversion tools, it would good idea to remake those listening tests with a blind ear to determine if they remain applicable to the new signal chain or not.

I desire for the 32-bit float is to not have to keep fussing with gain with every recording.

You can do this already.  Vinyl LP material will easily fit in 24 bits. You just need to find a comfortable input level that works for all records and leave it there.  Hot enough that the LP noise floor is well above the interface's noise floor, but don't attempt to normalize each transfer to close to 0dBfs on the way in.  Normalize afterward instead.. in the same way you would need to do with any 32-bit float transfer.

[GLouie]

The MixpreII series, when used as a USB interface, is limited to 96kHz (as shown in the specs). It might have something to do with the USB2 interface and the Mixpre10II doing up to 12 channels plus 2 out over USB. So, if you must record at 192, you'll have to do it directly into the recorder. Hopefully if doing RIAA vinyl LPs, you are using a phono preamp.

Many new users are confused by the MixPre polywav files and which channels are armed for recording. I don't think it's explained very well in the manual. The Cornell instructions are hopefully correct.

[ero3030]

Who rips vinyl???🤣🤣🤣

[Thelonius]

I rip my vinyl collection in 24/192 using the Roland R44 and I have a Mixpre 6ii that I recently acquired for live music recording so I thought I'd share some thoughts here.

It is possible to find pressings of vinyl that sound far superior, in many cases, to the redbook CDs. Note that not ALL vinyl sounds better (as quality of sources and mastering varies widely between pressings).  This is more a function of the mastering than the analog vs. digital format itself, in my opinion, but analog records can sound very good. Analogue Productions and Blue Notes Tone Poet series are good examples of recent releases that sound very good on average and certainly better than CDs I have heard of those recordings. Again, I think this is more about the mastering. I do find that capturing an all analog vinyl release at 24/192 captures a bit of that "depth" or "air" that I associate with a well recorded, and mastered, vinyl record. With respect to the question of why not just listen to vinyl, the answer is convenience and being able to take my music on the road with me.

That said, I have never felt the need to go beyond 24 bit for ripping vinyl. Unlike live recordings, my vinyl system has a limit to the volume it can produce so I can set it up at a standard gain and then normalize the last few db in post. Because the recording has already been mixed and mastered, and the physical limitations of the format, I don't have to worry about a sudden 12 db spike in levels like I do when recording a live show. Because of this possibility when recording live, I will use 32 bit.

Finally, I would suggest you consider recording your vinyl to the SD card and then using the SD card to transfer album sides to your computer. This essentially lets you split your workflow, listening to multiple albums at once if desired, and then moving the files over to the computer for processing. I use vinyl studio pro for normalizing, cutting into tracks and adding metadata. It is not free, but saves me time in my workflow.

Good luck!

[Gutbucket]

I remember doing an informal listening comp with some engineering student friends at Georgia Tech back in 1985 or so, comparing the the LP and CD release versions of the Thriller album, and specifically recall how everyone was surprised at how much better the LP version sounded. The difference was not subtle, but ear-opening.  Granted that was relatively early consumer digital, a first generation CD player, different masterings, all that.  Beyond the comparison itself, what was most interesting to me was that speculative discussion by the electrical engineering students about why the two sounded so different immediately focused on various technicalities of A-D and D-A conversion.  As that got into the weeds, I interjected to speculate about how it would probably be difficult to tease out how much of the difference was due to differences in mastering, which for vinyl record cutting were advanced and mature but for CD were then relatively recent and still immature endeavor.

~An aficionado of depth and air

[Packgrog]

I did manage to get plain stereo recording working, as stated in a prior response.

The reasons that I was interested in 32-bit recording are twofold: 1) I'm sick of having to restart recording to adjust levels, and would rather just apply gain in post without losing anything, and 2) I use iZotope RX 8 for track splitting and manual declicking, which forces everything into 32-bit anyway.

I didn't mean to start a a pros and cons of vinyl ripping discussion. I do it because for most 70's and 80's albums the mastering is superior to officially available digital versions of the same album. Also, some albums are simply not available on digital. For example, FM's Black Noise is only available on CD as a vinyl transfer done by the original artist anyway, and more modern ADC tech can provide a better capture (my recording with the Babyface Pro can be found on YouTube, which I'm sure I'm not allowed to link here). Other albums, like Buckingham Nicks, are still ONLY available on old vinyl, and I've only heard that wonderful album from other people's recordings. Mostly, though, it's an audio archeology hobby, and a desire to have versions of the album as ORIGINALLY intended, rather than the versions available now with excessive dynamic range compression.

From what I can tell so far, the noise floor of the MixPre is dramatically lower than what I was getting with the (non-FS) RME Babyface Pro. That alone makes it seem like a worthwhile exploration, as does freeing up my laptop for other things while I'm doing recordings.

[breakonthru]

I remember doing an informal listening comp with some engineering student friends at Georgia Tech back in 1985 or so, comparing the the LP and CD release versions of the Thriller album, and specifically recall how everyone was surprised at how much better the LP version sounded. The difference was not subtle, but ear-opening.  Granted that was relatively early consumer digital, a first generation CD player, different masterings, all that.  Beyond the comparison itself, what was most interesting to me was that speculative discussion by the electrical engineering students about why the two sounded so different immediately focused on various technicalities of A-D and D-A conversion.  As that got into the weeds, I interjected to speculate about how it would probably be difficult to tease out how much of the difference was due to differences in mastering, which for vinyl record cutting were advanced and mature but for CD were then relatively recent and still immature endeavor.

~An aficionado of depth and air

thats my take as well. Any perceived differences in vinyl being "superior" is the result of different mixes and mastering. I find the current trend of wanting to buy vinyl versions of new releases justified by "that pure analog sound" as misguided, as nowadays whatever gets put to vinyl is a by product of the same digital recording and mastering process.

But i also dont understand why theres a market for dozens of different variants of the same album, which is also a trend. My ears dont really care if the vinyl is clear or red or purple. I feel that more likely there is a stash of millions of sealed records somewhere that will end up being a bust like 1980s baseball cards.

[Chilly Brioschi]

Who rips vinyl???🤣🤣🤣

Once upon a time, I did at Institute for Audio Research, under the watchful tutelage of a master from Warner.
It was ... a Boy Scout project!   No merit badge, dangit!

I never attended believing that I could never make a career in the biz.
My friend did, and we lost touch, but I had heard that he was working as a bass player.
Ancient history

There is no hope in any argument that adding spinny mechanisms to save and recover audio inherently enhances it.
I believe in Nyquist's Theorem.
I have smelled the leaky start capacitor and seen the drippy motor shaft bearing.

However, if it sounds better, if you enjoy it, if it aesthetically pleases you, by all means enjoy!



But to me?

[Gutbucket]

There is no hope in any argument that adding spinny mechanisms to save and recover audio inherently enhances it.

Maybe not save and recover, although the liturgically minded could argue the C3 version of the Hammond contributed to saving souls. As for enhancement, I submit exabit A in defense of spinny tonewheels and Leslies:

[Gutbucket]

Exhibit B: