Jump to content

New Zoom H3-VR Recorder mic combo..


Recommended Posts

2 hours ago, IronFilm said:

I bet everyone will have one. It is too cheap not to buy on a whim. 


Yes, the Rode NT-SF1 just started shipping. If the reviews are good, then I'll probably buy it. Only question is I'll buy the H3 as well to supplement it?

 

 You answered your own question. It is so cheap that everyone will have one - including us!

 

I am eyeing Ambeo for some time now, but I never had a job to excuse the 1600euros that cost here. For less than 1000euros, maybe the Rode make me bite the bullet to start experiment with, it has to be good, I am not worrying about it much, it is not that cheap and the two companies are trying to beat Ambeo, which in my opinion has the Sennheiser premium added in the price tag, which honestly, I am redundant to pay anymore!

 

We have to listen to the sound quality of the Zoom though, including everything for 350 is too low. The capsules must be of very low quality and we have to listen to self noise e.t.c

 

 

Link to post
Share on other sites
  • Replies 55
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Why Skeptical? That is about the price ratio of Zoom's product...   Zoom F8 $900 x 5 = the price of a nomad Zomm H3-Vr $349 x 5 = $1745 about the price of a Sennheiser Ambeo VR 3D

This could be a huge fishing hook for us. Looks interesting.... strange, but interesting.   The gyroscope seems to be used for when set H3VR. and it may be used for investigation o

Zoom's H2 (&H2n) had a 4 channel mode and cost very little (and sounded horrible). With more demand for VR content maybe they saw a reason to build and improve on the previous offerings (which hav

Posted Images

6 hours ago, Kisaha said:

This is for the Samsung's VR and the GoPro's and the Yi's that will come and the such, they are not competing directly with Soundfield or Sennheiser.

 

Also, as the VR, AR and 360 becomes more common, sound solutions will increase and they will trickle the prices down.

 

The new Rode will push the Ambeo for sure, but not this one.

 

 

My guess is if Zoom make another 1 of these it will be more expensive and closer to the cost of the Rode offering. Quality aside, this looks like a product that will get a lot more people interested in ambisonic not least as it has they have tried to develop a more accessible approach to the workflow. Monitoring/reference feed in Binaural/Stereo, built in decoding to 2 types of B format during OR after recording and a nice concept for the accelerometers when reviewing recordings.

3 minutes ago, daniel said:

My guess is if Zoom make another 1 of these it will be more expensive and closer to the cost of the Rode offering. Quality aside, this looks like a product that will get a lot more people interested in ambisonic not least as it has they have tried to develop a more accessible approach to the workflow. Monitoring/reference feed in Binaural/Stereo, built in decoding to 2 types of B format during OR after recording and a nice concept for the accelerometers when reviewing recordings.

Yep, I've sold myself 1. I have a trip to Sri Lanka at the end of the year and was going to put a compact ORTF rig together to record ambiance and events (on trip i'm unlikely to do again). Any idea when this is available?

Link to post
Share on other sites
On 9/15/2018 at 6:21 AM, alenK said:

This device looks fantastic, but I imagine it would be more readily received by aficionados if it was priced somewhere between $7,000 and $23,000.

 

The combination of components in a compact integrated form factor seems like a recipe for over coming the many limiting factors in traditional setups. My first experiences with ambisonics was with a corporate owned CalRec system. I have been observing the evolution of technologies, and the elimination of real life problems with ambisonic production for quite a while.

 

It does not seem hard to imagine that a mic array based on MEMS devices, a pcb with minimal length pathways, a few purpose built integrated circuits, an on board processor running optimized software, and an adjacent data storage system can easily provide superior results, when compared to the relatively complex big budget solutions, at a tiny fraction of the cost.

 

This thing seems brilliant, even if it does put us all out of business. 🙂

 

 

  

What will put us out of business is the non charging of proper rates for normal equipment packages..  not the VR mic/recorder combo...

Link to post
Share on other sites

The business of most of the people on this particular forum is the recording of humans talking, usually on location.  That work gets harder and harder for all the reasons stated on this forum over and over, very familiar to the soundies actually doing the work.  No ambisonic sort of microphone will ever be able to deliver the sort of audio demanded of pro location dialog mixers every day: we are as much (and sometimes more) about what will NOT be heard in our tracks vs what WILL be heard.  Ambisonic is about hearing everything, in a "you are there" sort of situation.  Fun, useful for some sorts of things, a value-add if post can actually use those files; not what we generally do.   Sure, AI mics and processing for all that pernicious location BG noise are incoming.  Not here yet.

Link to post
Share on other sites
45 minutes ago, Philip Perkins said:

... Fun, useful for some sorts of things, a value-add if post can actually use those files; (snip) ...

 

Yep indeed - and of course this nugget of truth counts just as much for a top quality ambisonic rig as a budget one (as indeed it could for any technique of any quality ... )

 

Not a magic formula, ambisonic tracks could be close to useless for many applications and situations, but the best solution or as good as for others. For atmos tracks useful to me as a film sound editor I'd be happier being handed a pile of ORTF tracks, Dan ...

 

But interesting, and fun - certainly

 

Jez

Link to post
Share on other sites
1 hour ago, Philip Perkins said:

The business of most of the people on this particular forum is the recording of humans talking, usually on location.  That work gets harder and harder for all the reasons stated on this forum over and over, very familiar to the soundies actually doing the work.  No ambisonic sort of microphone will ever be able to deliver the sort of audio demanded of pro location dialog mixers every day: we are as much (and sometimes more) about what will NOT be heard in our tracks vs what WILL be heard.  Ambisonic is about hearing everything, in a "you are there" sort of situation.  Fun, useful for some sorts of things, a value-add if post can actually use those files; not what we generally do.   Sure, AI mics and processing for all that pernicious location BG noise are incoming.  Not here yet.

 

Yes very good points, to add to which if ambisonic is listened to in a fairly accurate 3d surround environment, with height included, the brain is able to make more sense of it, and do it's own filtering (in a 'you are there' sort of way) using phase relationships and timing information, which the brain is very, very finely tuned to be able to discriminate and filter, as the result of life that has evolved and prospered over millions of years, from aquatic, to animals to humans are those able to derive better detailed selective information from their enviroment for both survival and food.

Listening to Ambisonic in anything other than a very good surround space is effectively stripping a lot of information which the brain uses to do it's amazing filtering all the time.
'The cocktail party effect'

Link to post
Share on other sites
1 hour ago, Philip Perkins said:

The business of most of the people on this particular forum is the recording of humans talking, usually on location.  That work gets harder and harder for all the reasons stated on this forum over and over, very familiar to the soundies actually doing the work.  No ambisonic sort of microphone will ever be able to deliver the sort of audio demanded of pro location dialog mixers every day: we are as much (and sometimes more) about what will NOT be heard in our tracks vs what WILL be heard.  Ambisonic is about hearing everything, in a "you are there" sort of situation.  Fun, useful for some sorts of things, a value-add if post can actually use those files; not what we generally do.   Sure, AI mics and processing for all that pernicious location BG noise are incoming.  Not here yet.

+1.

And in no way to contradict: I'd rather the gig delivering location dialogue tracks and mix with the capacity to provide an ambisonic atmos than get booked to delivering an ambisonic atmos with some dialogue tracks thrown in (does this make sense?).

Are the folks marketing 7 minute VR experiences that bothered on what equipment the ambisonics are recorded or more the ability to say it's 'ambisonic'. What I gather from the VR conversation thus far is when location recordists get booked to do this stuff it's because the client wants discrete dialogue and effects tracks to work into the VR environment as well as the ambisonic.

 

"as a film sound editor I'd be happier being handed a pile of ORTF tracks, Dan " so we should do a blind test - an ORTF originated recording on something of similar cost vs an ORTF derived from the H3-VR.

https://micbooster.com/primo-microphone-capsules/135-clippy-em184-xlr-matched-pair.html (I imagine the caps in the Zoom are similar to some of the primos). 

https://www.amazon.co.uk/Zoom-H4n-16GB-Card-Tripod/dp/B01DPOXS8I/ref=sr_1_3?ie=UTF8&qid=1537130046&sr=8-3&keywords=zoom+h4n

 

Obviously not expecting any of these to be comparable to a pair of 8040s and pro grade preamps etc - but very interested in something small an easy to carry.

In terms of field use, it looks like it'll be fun improvising a basket for this - maybe sugru and something like a Supershield pod ;-)

Link to post
Share on other sites
On 9/16/2018 at 1:34 PM, Philip Perkins said:

,,, Ambisonic is about hearing everything, in a "you are there" sort of situation.  Fun...

 

While this may be true for the new fangled virtual reality presentations and perhaps some surround mixes, ambisonics has traditionally been employed to provide an opportunity, either while recording live or mixing in post, to isolate specific portions of the sound field from the whole, with greater precision and latitude of choice, than is possible with other traditional microphone array geometries, when used with similar near-far or distant placement. 

 

VR surround effects have come into vogue and currently receive attention, but the introduction of new microphones and advanced dsp software also promotes the ability to practice ambisonic techniques as in the past. The craft has certainly become more accessible, and an interest in applying the directional control techniques may become more widespread.

 

ambisonics is not necessarily tied to VR, surround, or even stereo mixes. Personally, my interest in ambisonics has been motivated by an enthusiasm for mono, an ambivalence for stereo, and a disinterest in surround mixes. 

 

There is something for everybody when producing ambisonic material.

 

 

 

Link to post
Share on other sites

SoundField by RØDE Plugin is claiming -

Having the ability to change the polar pattern of the virtual microphones in post-production is a well know feature of ambisonics – allowing the user to create everything from omni through cardioid to fig-8 patterns in post-production. The new SoundField by RØDE goes beyond this, for the first-time allowing shotgun-type patterns to be created, using recent developments in beamforming technology.

Plug in download however not available yet, as of 17th September 2018

https://en.rode.com/soundfieldplugin#footer_download

https://en.rode.com/soundfieldplugin#overview

Link to post
Share on other sites
18 hours ago, pindrop said:

SoundField by RØDE Plugin is claiming -

Having the ability to change the polar pattern of the virtual microphones in post-production is a well know feature of ambisonics – allowing the user to create everything from omni through cardioid to fig-8 patterns in post-production. The new SoundField by RØDE goes beyond this, for the first-time allowing shotgun-type patterns to be created, using recent developments in beamforming technology.

Plug in download however not available yet, as of 17th September 2018

https://en.rode.com/soundfieldplugin#footer_download

https://en.rode.com/soundfieldplugin#overview

As alenK says, ambisonic not all about VR. "[] isolate specific portions of the sound field from the whole, with greater precision and latitude of

choice, than is possible with other traditional microphone array geometries, when used with similar near-far or distant placement. " (eloquently precise). If Rode have brought a new dimension to how the recording can be manipulated (and how easy it is to do this) I wonder why people would think we wont be asked to add something like this to our 'deliverables'.

Not sure if rode plugin works with ambisonic recordings from different set-ups - "Eschewing the matrices and correction filters of previous generations, it utilises state-of-the-art frequency-domain processing".

Correction filter are specific to the mic array, right?

Anyway, (perhaps) people are going to hate me for saying this, but this could change the use of the boom pole. Eg. Computational sound gives us a virtual fisher boom?

Link to post
Share on other sites
1 hour ago, daniel said:

Anyway, (perhaps) people are going to hate me for saying this, but this could change the use of the boom pole. Eg. Computational sound gives us a virtual fisher boom?


Not yet, two major problems: wrong kind of capsules being used for our line of work + the extra weight

Link to post
Share on other sites
11 hours ago, daniel said:

Not sure if rode plugin works with ambisonic recordings from different set-ups - "Eschewing the matrices and correction filters of previous generations, it utilises state-of-the-art frequency-domain processing".

Correction filter are specific to the mic array, right?

 

Correct, but the plug-in will also take as input B-Format files, so you could use an ambisonic microphone from a different brand, convert to B-Format using their respective A-to-B-format conversion plug-in (or if you have custom calibration files you can use VVEncode), and then feed that B-Format file into the Rode plug-in for further processing and manipulation.

Link to post
Share on other sites
15 hours ago, IronFilm said:


Not yet, two major problems: wrong kind of capsules being used for our line of work + the extra weight

 

Well, since the capsules could be re-encoded to directional cardioid/shotgun, as well as stereo or 5.1 wouldn't they be the excellent capsules for our work?

And weight would be about the same as a Sennheiser 416 or Røde NTG3, as the zoom unit is about 120 grams without batteries. Two AA batteries weigh about 46 grams.

There could be plenty of other challenges though, such as transmitting the ambisonics wirelessly for mixing or monitoring. I like the idea though.

Link to post
Share on other sites
4 hours ago, Jacob Gustavsson said:

 

Well, since the capsules could be re-encoded to directional cardioid/shotgun, as well as stereo or 5.1 wouldn't they be the excellent capsules for our work?

And weight would be about the same as a Sennheiser 416 or Røde NTG3, as the zoom unit is about 120 grams without batteries. Two AA batteries weigh about 46 grams.

There could be plenty of other challenges though, such as transmitting the ambisonics wirelessly for mixing or monitoring. I like the idea though.

I have no idea how hard it is for a human to find the sweet spot ("isolate the specific portions") in an Ambisonic sound field but I imagine at some point it is not implausible a computer will be small and fast enough to do this, if not in real time but efficiently enough after the event for a review to be available on set. Perhaps it starts out a bit like 'steady shot' for boom, micro adjustments to the mic pattern or direction to level out a derived mono (cardioid) signal (without riding gain) or smoothly cue to 'on mic' sources in a multiple speaker, Ad Lib situation. Perhaps a zoomable mic, from omni through to hyper cardioid and shotgun. Or model the mic you left at home :-).

Am I right in thinking if 1 could successfully model the polar pattern of an MKH50 from an ambisonic recording (dataset?), at this point in time, the noise floor would be greater than an MKH50 because to the 4 caps and preamps used to create ambisonic?

Link to post
Share on other sites

There are two problems with creating mono virtual microphones with directional patterns from a soundfield:

 

1- The first one is that First Order Ambisonics has a relatively low spatial resolution, so localisation of sound sources is not as accurate as it is with Higher Order Ambisonics, like 2OA or 3OA for example. So while beamforming may help create a highly directional pattern, you are still using a low spatial representation to extrapolate the sound source from. Maybe once we start working with HOA this will become much better.

 

2- The second, which was alluded to already, is that you're using four capsules with equivalent self noise of 15 dBA each (in the case of the Rode NTSF-1), to create a virtual microphone that would theoretically have a lower self noise. In the case of an MKH50, the self noise is 12dBA, so you can see the math doesn't really work, your virtual microphone would have a higher self noise than the real thing.

 

These are not as apparent with stereo or surround, but definitely with mono from my experience. So while yes, you can create a virtual microphone pattern from the soundfield, you do have trade offs to consider.

Link to post
Share on other sites

I was asked about this unit by some in-house emerging media producers recently so I'm following this discussion with some interest. First off I have to ask (maybe I'm ignorant) about the B-format channel order displayed in Zoom's promotional videos, I have never heard of a channel order WYXZ, can someone enlighten me?

 

I believe this unit should be regarded as a replacement for the H2n, geared towards VR/360 shooters who work without audio support. As such I can see it being an upgrade.  In close proximity to the sound source it'll probably gets you something that "moves" with head rotation and you get an actual track 3 waveform,  bingo, mission accomplished.

If you want things like timecode, better capsules , proper windprotection, additional tracks for Boom and lavs so you can actually record dialog, hire an audio guy.

One thing I'd find useful here (and in the F8) would be binaural monitoring via Bluetooth.

Looking forward to some reports from the field.

 

Link to post
Share on other sites
1 hour ago, Werner Althaus said:

First off I have to ask (maybe I'm ignorant) about the B-format channel order displayed in Zoom's promotional videos, I have never heard of a channel order WYXZ, can someone enlighten me?

 

 

Maybe a typo? AmbiX ACN channel order is WYZX.

 

2 hours ago, IronFilm said:

These trade offs though could become smaller and smaller as technology improves, maybe ambisonic mics will become the boom mic of the future? But I strongly suspect that future is easily a decade away or more, probably much longer.

 

I personally doubt it. A directional mic in the hands of an experienced operator will yield better results and save more time and money in post. An ambisonic mic as a boom mic will require processing (both in production for the production mix and in post for the final mix), and just adds more complexity to a workflow that doesn't really need it.

 

For sound effects and ambiences recordings for post, however, I think Ambisonics is definitely the way to go. The flexibility it affords for all the different discrete channel outputs is unparalleled, and in my experience the trade-offs are not as apparent.

Link to post
Share on other sites
41 minutes ago, Werner Althaus said:

A typo in the display? I wonder if the channel order in the 4channel file matches Ambix ACN or the display.

WYXZ.jpg

It could be a typo in the ad itself? Or maybe a software bug if it really is a photo of the live product? Can't speak on their behalf, but it's definitely not a standard channel order for Ambisonics.

Link to post
Share on other sites
2 hours ago, Jose Frias said:

It could be a typo in the ad itself? Or maybe a software bug if it really is a photo of the live product? Can't speak on their behalf, but it's definitely not a standard channel order for Ambisonics.

I just grabbed a quick screenshot of their live  video with meters moving, so a bug maybe? Doesn't really matter, I was just unsure whether this is a legit B-format channel order that I hadn't heard of. Thanks.

Link to post
Share on other sites

WXYZ was standard Soundfield order (and why would it be other than alphabetical) for as long as I know. Autocorrect???

 

Jose, re- "ambisonics the way in post" and Dan, "successfully model the polar pattern" - Jose, I know you use Ambisonics regularly and have far more experience than me - but I think it's to an end (VR from where I've followed your posts) where the technique works. I'm going to throw out the gauntlet once again and say in many areas it can be problematic. And theatrical film, under Dolby Digital for instance, is one: both for the mixing side of the signal and the presentation side of the theatre. I'm pretty certain that with the expansion of VR and renaissance of ambisonics with it that cinema, in tech and presentation, will probably finally move to incorporate it (and maybe this is already easier with Atmos and Auro3D for instance but not enough to mix with older formats in mind). Please - if any theatrical mixers are reading please chip in here.

 

What really has to be understood by all is that Ambisonics is not a magic formula. It has been around a long time and has been useful and not-so-useful many times over that time. Any capsule is going to be receptive of the higher frequencies only on its axis - diffuse field applications might accentuate or compensate to achieve otherwise. Ignoring the magical tetrahedral capsule head Ambisonics is a simple extension of MS, except that the M is not fixed but a potential derived from an (hopefully decent enough) omni signal and however many side signals as one has. So the first point to bear in mind is that the mic really has to be in the 'right' place (it is a single point mic) to be good - or often of any use whatsoever. Since Soundfield was a music mic for two decades before its first comeback in stereo tv this was something understood in its placement. In VR (and the design of the Zoom) it is obviously in the right place because it's on axis with camera. The second point is that in the maths of Ambisonics the directional characteristic comes from an in/out phase possibility of summation from omni (W) to figure of eight (any of the others) and no more. Even with HOA and a great deal of further development this is hardly going to replace the age-old "put it in the right place and point it in the right direction."

 

I AM an advocate of recording ambisonically in addition to the technique chosen as best decided if enough channels are available. I just know that, currently, it's going to be less use than other techniques I might choose (for my work or end use). I have to stress that I don't advocate using any fixed technique nor position (such as ORTF or binaural for instance) above choosing techniques/positions for each setup - and crucially the format the recording is heading for. But I do understand that it is not obvious (in a frankly difficult technical area) where recording techniques might present problems down the line. The solution is in communication which we're hopefully doing here.

 

Jez

Link to post
Share on other sites
12 hours ago, Jose Frias said:

Maybe a typo? AmbiX ACN channel order is WYZX.

 


People have pointed this error out before in Zoom's marketing materials. 
Guessing it is just a (surprising!) mistake by Zoom, and they'll fix it before it ships. 

 

12 hours ago, Jose Frias said:

I personally doubt it. A directional mic in the hands of an experienced operator will yield better results and save more time and money in post. An ambisonic mic as a boom mic will require processing (both in production for the production mix and in post for the final mix), and just adds more complexity to a workflow that doesn't really need it.

 


I also personally doubt it, that it will happen any time soon that is. 

However the marching forward progress of technology is relentless, especially the pace of computing power. 

So I couldn't rule out the possibility of this being a reasonable choice a few decades from now into the future. 

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...