Sound-Devices MixPre-D

[Christoffer Glans]

Ask camera operator if he wants four cables hanging off the camera while he is walking around shooting handheld , because you all think this is so stylish looking. Or the extra weight of four channels of wireless receivers (will plug into your camera battery).

I think it's interesting that some cinematographers prefer heavier cameras because they are more "stable" when using, but feel that something like this is too heavy of bulky to use.

 

[Richard Foster]

In this case I am the DOP and operator as well as the director, by choice, I am very particular about framing and lighting, and have operated this way for over 25 years. I usually do not hand hold, but use dolly and head configs with occasional crane. What I am interested in is feedback on recording to the camera vs. to a recorder. Thanks.

 

[Tom Visser]

I have just received my MixPre-D and am very happy with its performance. The preamps are probably about 80% to 90% as good as it gets for productions sound (how can I put a percentage figure to that?) and indistinguishable by me from other SD mixers and recorders. I think your question regarding camera sound quality is totally dependent upon the camera, and more importantly, the person(s) setting it up. Do they calibrate correctly, know how to set gain levels, paying attention during rolls to make adjustments, etc...? I think that the sound quality going to camera, all other factors being equal, is going to be good enough in most circumstances. Top quality sound recorders will almost always be better, but the reason to use sound crew is not so much about the quality of the electronics and digital files, but the workflow, tradesmanship, and fact that camera perspective is not always the best "sound perspective" as the audio gear does not inherently work the same way as the optical gear. As a sound mixer myself, Marc and I may sound like a broken record preaching about hiring sound crew, so I'll spare you that tirade for now (but I'm with Marc on this one).

In this case I am the DOP and operator as well as the director, by choice, I am very particular about framing and lighting, and have operated this way for over 25 years. I usually do not hand hold, but use dolly and head configs with occasional crane. What I am interested in is feedback on recording to the camera vs. to a recorder. Thanks.

 

[Dabling Harward]

Hey Babling Dabling......MixPre is 2 channels only. And BTW, you don't have a fuzzy on your still photo camera. Good luck with your one-man-band Productions.

Hey Triangular Trailangel (whoever you are hiding behind an alias) Mix Pre D has 4 inputs. Might want to read the manual first before posting nonsense. Good luck with your whatever.

 

[Tom Visser]

I would call it 2+2 inputs, as the inputs 3 and 4 are not full featured. You cannot listen to the return monitor path and utilize the 2nd set of inputs simultaneously, probably not an issue for those of you that are camera mounting it in the first place, and listening directly to your camera output. In order to change routing or gain settings on inputs 3/4, requires one to press down on a button while simultaneously turning the multi-purpose headphone rotary encoder. It is almost possible to do this with one hand (3rd hand anyone?) but the small size of the controls and the oddly placed headphone jack which seems to partially block access to the encoder (and LEDs in some cases) makes it a dicey affair. If your a set an forget type of operator, I can't think of a better unit, but it would be difficult to call this device a 4 channel mixer in the company of other dedicated multi-channel mixers. Finally, they are not balanced connections, so could be limiting in some situations when connecting external gear.

Hey Triangular Trailangel (whoever you are hiding behind an alias) Mix Pre D has 4 inputs. Might want to read the manual first before posting nonsense. Good luck with your whatever.

 

[Dabling Harward]

I would call it 2+2 inputs, as the inputs 3 and 4 are not full featured. You cannot listen to the return monitor path and utilize the 2nd set of inputs simultaneously, probably not an issue for those of you that are camera mounting it in the first place, and listening directly to your camera output. In order to change routing or gain settings on inputs 3/4, requires one to press down on a button while simultaneously turning the multi-purpose headphone rotary encoder. It is almost possible to do this with one hand (3rd hand anyone?) but the small size of the controls and the oddly placed headphone jack which seems to partially block access to the encoder (and LEDs in some cases) makes it a dicey affair. If your a set an forget type of operator, I can't think of a better unit, but it would be difficult to call this device a 4 channel mixer in the company of other dedicated multi-channel mixers. Finally, they are not balanced connections, so could be limiting in some situations when connecting external gear.

When any channel of the Tape Return input is routed to the main output bus, the RTN Monitor Source is disabled. That's only the return monitor. You'll now hear all four inputs in your headphones.

And I don't believe I ever called this a 4 channel mixer. I've always said it has 4 inputs.

 

[Dabling Harward]

Granted, this isn't a 744 or a 788 but it's a great little mixer that I'm sure many people will find very useful on those tiny shoots.

The lane that is box is in, isn't for big productions that need a sound department...

It's for those productions that simply is just that…simple.

SoundDevices made a great little box to fill a niche and I think they've done a great job with this one.

 

[Tom Visser]

Granted, this isn't a 744 or a 788 but it's a great little mixer that I'm sure many people will find very useful on those tiny shoots.

The lane that is box is in, isn't for big productions that need a sound department...

It's for those productions that simply is just that…simple.

SoundDevices made a great little box to fill a niche and I think they've done a great job with this one.

I agree, it is a great box, quite a fantastic form factor. I hope that they apply the same philosophy on the -D upgrade to the 302, which is by far my favorite small mixer. The MixPre-D will primarily be used by me to reduce my weight on these smaller simpler jobs, or to give to my boom op for larger ones.

 

[Marc Wielage]

Not trying to hijack this thread here, but ask a question related to it. With a good mixer like being discussed here would a sound professional feel comfortable recording location sound for a feature directly into a RED or would he still go double system? And if double system do you then tether camera time code to the recorder?

I would vote for double system, with a mono scratch track going to camera. I would also have a Denecke SB-T or an Ambient Lockit timecode generator on the camera (Red or otherwise), so that camera timecode and sound timecode are identical. The advantage of going with a separate recorder: better sound quality, more total audio channels, more flexibility in mixing, iso tracks, mix tracks, and the benefit of having an actual sound mixer doing the work. A wireless feed to the camera would give them more flexibility in terms of mounts and not having to worry about tripping over cables going back to the sound mixer.

The main issue I see is that the camera operator rarely is in a position where he or she can constantly monitor the sound. If you have an actual sound mixer on the set, working separately from the camera crew, the sound mixer can constantly optimize the sound, change microphone position, add wireless mikes (if necessary), and make all the other decisions designed to help maximize the quality of the dialog in your film.

 

[Christoffer Glans]

I see this as great for recording sounds for sound design, not primary for work together with a camera for sync. In those cases it's best to use something with timecode.
But I suppose that it's very good for a sound designer if you are on a budget.

 

[Tom Visser]

I also vote for double system when dealing with most any camera, and RED does have some unique quirks too. When I accidentally roll on something that peaks a little "hot" say -3dBfs with my limiters kicking in on my recorder, I can be assured that the recorded sound is still intact and sounds good. When asked to make the call, I can put my reputation on the line and say it is good to go. The specific problem I've had with RED cams is that the inputs are not necessarily "linear" or more correctly, correlated logarithmic scales. If I calibrate for -20dBfs = 0dBu on the mixer and the camera, I find that as the levels increase, the RED scale is more sensitive and clips earlier than my mixer. (could vary with output / input impedance combination differences as a possible explanation) The sound quality at the upper extremes of the RED are not quite as progressive as a decent audio circuit either, more prone to distortion ealier. What this means is that I cannot be as confident and have to be much more conservative when mixing to the camera. The equivalent I can think of in terms of camera speak would be using a camera with less dynamic range. Because I'm running lower levels, it means the RMS audio signal is that much closer to the noise floor of the electronics, thus a reduced audible dynamic range.

Don't mean to imply that the only way I can get good results is to do double. Sometimes for whatever reason I must record to RED. I feel that I can get professional results, but when asked, I'm going to be honest and say that sound quality has been sacrificed if double was ever a legitimate option. For live broadcast, performance, ENG, or reality, I wouldn't really think of endorsing recording to camera at all other than for scratch, but these are not the typical applications for RED in the first place, either. In a controlled environment with expert operation, I would even go so far as to say the results would be "almost identical" assuming you didn't need the ISO tracks. Fix it in post or have it already fixed in production, there is more than one way to skin a cat, but it (sound quality) must be addressed at some point.

 

[Gabriel Winebrenner]

Has anyone tried tape out (sound devices TA3 cable) into the camera and usb out into a mac/pc at the same time, in essence doing a dual recording? Wondering if the usb and analog as well as digital outputs can all be used at once?

 

[Lauri Kettunen]

What I am interested in is feedback on recording to the camera vs. to a recorder. Thanks.

As a wildlife shooter in most cases don't even have any other option than recording to the camera. Although not a professional sound person, during my studies I worked several years as a recording engineering in a studio and live in the understanding I'm still more or less able to recognize differences in quality. After the upgrade of R1 soundboard I've been recording most of the time directly to the camera through Sound Devices amplifier, and have been pleased with the outcome. The noise level is low and enables one to add quite a lot of gain in post if needed. In my experience there's no significant difference between to-the-camera and external Sound Devices recorders. If I listen my sound tracks and don't have my notes with me, I'm not able to say which one is to-the-camera and which to-the-recorder. But should add, outdoors other factors, such as wind and wind shielding easily overwhelm technical issues making it difficult to take sides on the fine details.

 

[Richard Foster]

Thank you Tom, Mark and Laurie for your replies. This is the kind of feedback I was looking for.

To make it clear I will be using a professional sound mixer and boom man and shooting the feature on the Red One MX. Since it is very seldom that more than 4 tracks would ever be needed with typical drama recording it seemed simpler to record directly to the camera through a good mixer with good preamps.

What Tom said about possibly crowding the headroom on the Red's audio does concern me, although I had the opposite experience (but was not using a mixer) on a piece I did with the mic taking phantom power from the Red and directly in and had trouble getting enough gain even when I had the Red's gain all the way up.

I guess the best is for us to run tests with the Red audio recording vs. Separate and see what we think. I guess I also thought editorial would be simplified since it would save the syncing step. I come from a 35mm film background and this will be my first Red feature.

 

[Mike P.]

Hey Tom, aren't the mixers' limiters (MixPre-D's in this case) supposed to prevent the recorder (in this example, the camera) from clipping highs? Or are you just noting that, even though the mixer might not be running hot, the camera could be recording it hot (in which case, you'd have to set the mixer's limiters according to how hot the camera is recording the audio)?

 

[Lauri Kettunen]

.... aren't the mixers' limiters (MixPre-D's in this case) supposed to prevent the recorder (in this example, the camera) from clipping highs? ... the camera could be recording it hot (in which case, you'd have to set the mixer's limiters according to how hot the camera is recording the audio)?

Not trying to answer in behalf of Tom but this is indeed the purpose of the limiters. A practical problem is, one is not always necessarily able to see distortion from the led meters or from the camera metering. I've noticed a flute is an instrument creating some high mids that easily get distorted. Among the birds the same happens with whopper swans when they sing. The only way to avoid this is to listen the signal with head phones while recording and retaining a marginal between the peaks and what the meters indicate the maximum level.

 

[Tom Visser]

In theory, yes, this is what the limiters are for. IF the limiters are even engaged at all, you have to make some serious evaluations on whether or not your audio is compromised. As Lauri pointed out, it's not simply a matter of reading the LEDs. Different sounds are weighted differently. The SD are PPM and are much faster than VU meters, but are still not completely indicative of true dBfs levels (standard practice is to calibrate 0dBU on the mixer to -20dBfs on the recorder.... so if the mixer has 20dB of usable headroom, then +19dBU of output is theoretically at -1dBfs or still in the safe zone). My experience has been that I cannot trust 0dBU/-20dBfs calibration on the Red. In the past, I have actually had to do something like calibrate to +15dBU = -6dBfs just to make sure I knew where the clipping point actually was. I don't actually own a Red, so those of you who do and have a mixer should do a little experiment. turn on the internal tone generator, calibrated to 0dBU @ 1KHz. Adjust the input on the RED so that this level hits your -20dBfs mark on the camera. You could roll on this tone and confirm that your levels are at -20dBfs in a NLE. Then use an external tone generator, I use a little XLR plug device that operated off of phantom, or you could feed it 1KHz tone from a computer or other mixer, into the mic/line inputs of the mixer. Now adjust levels on the mixer so that your now hitting the +16dBU mark on your mixer's meters. The RED should be now seeing -4dBfs, can also confirm by rolling and checking in an NLE. My experience is that this is actually clipping or very close to it. I recall a 4 to 6dB shortfall in calibrated dynamic range due to an inaccuracy in the metering. It could have been a bad input board on a camera I was working with. (I tend to work with the same couple of cameras repeatedly) I certainly don't want to spread misinformation on the internet, but in my own work experience, I am very cautious about levels when sending to RED. I would be curious to hear from actual owners if their experience matches my own. I also feel, as wonderful of a camera that the RED is, that the audio circuity is compromised at hot levels. This does not make it unprofessional, but you have to know the limitations of the gear you are working with. I know that I can record at near clipping levels on a professional recorder (this is of course not appropriate to do regularly, but it happens, hand clapping, screaming dialogue, vehicles, etc...), and it will still sound clear, but I have experienced audible distortion much before the digital clipping point (estimate in the -4 to -6dBfs range) with the RED compared to dedicated recorders.

 

[Lauri Kettunen]

have experienced audible distortion much before the digital clipping point (estimate in the -4 to -6dBfs range) with the RED compared to dedicated recorders.

Haven't tested this in controlled situation, but this is the margin I've adopted for my audio chain by experience. Here's a sample http://www.wildcine.fi/Sample.wav of swans recorded to the camera and where peaks are in the range of -4 to -3.8dBFS below clipping point. For me it sounds as if there is distortion, but not able to say what in the audio chain caused it.

 

[Dabling Harward]

Same here, but when using Varta batteries it runs two to three times longer.

Was going to buy some Lithiums when I saw the Vartas you mentioned. Thanks for the tip. Vartas last so much longer than the Procells. Went all day with them, periodically turning the unit off. Never could've done that with the Duracells.

 

[Christoffer Glans]

So I got my MixPre-D and it works great with my macbook air as a recording device. But I really would like some kind of field option so that I could go out and collect sounds. Are there any cheap 24bit 96Khz recorders with XLR inputs? Almost everyone I looked at has built in microphones but I really just need a XLR input and because I put most money on the mixpre-d I would like some cheap option on the recorder side.

Any tip?