...if our only destination is always Blu-Ray and DVD? We're using a Zoom H4n with the built-in mics to record stage shows and concerts. Sound is actually pretty decent, given the limitations of the hardware.

Premiere defaults to 48/24 as an editing preset, and our cams (Z5Us, from which we use some audio as well) record in 48/16. Isn't 96/24 overkill? Will we hear any difference, when all is said and done, given the aforementioned limitations of the hardware?

Thanks, audio experts!

Most people are still delivering SD... isn't shooting HD overkill? By the same logic?

I shoot 16/48 in camera and 24/48 on the external recorder. I have shot 24/96 in-studio. Probably overkill for what I was doing, but didn't cause me any undue problems, and worked out just fine.

There's probably no benefit in recording 96/24 on your Zoom, as your device won't deliver that kind of real world quality. The professional standard for video and film is 48/24, but only when using high-end equipment. I'm told that 96/24 is used for professional audio recordings of symphonies and string instruments and things like that, but again only with high-end equipment.

I tend to record 96/24 on my zoom. The 96k may be a bit optimistic but the 24 bit has come in real handy. The more bits you get to work with the better if you have to do a bit of audio tweakage. I can see no reason to NOT record 96/24.

I would say a 24 bit depth can be useful as it will give you more dynamic range(range of loudest to softest signal) but 96k sampling rate in my experience is overkill most of the time. I'll put it this way, almost all of the record studious I have worked in Nashville typically record at 24 bit but only a few of them were typically recording at 96k.

Burk, supposedly recording at 24 bits on inexpensive recorders doesn't really give you more information to work with. Sure, you are recording those extra bits, but since your recorder doesn't have a high enough signal to noise ratio to justify it, you aren't really getting the dynamic range that you are looking for. It's like those cheap cameras with lots of megapixels but with lenses that can't resolve all that detail. They just record a really big fuzzy image. The same idea applies here.

Most people are still delivering SD... isn't shooting HD overkill? By the same logic?No, of course not, because a) I'm making Blu-Rays, and b) there are good, demonstrable and well-established reasons to shoot and edit in HD even if you are delivering in SD, because HD is significantly better than SD at the acquisition phase. I guess I'm asking if 96kHz is better than 48kHz in the same way that HD is better than SD, given that the Zoom is far from high-end, and the use of the onboard mics would likely negate any advantages that recording at 96kHz would bring.

My understanding is that it isn't, at least when limited by the inherent characteristics of the Zoom, but I wanted to get others' opinions. So I'll put you down as a yes?

Adam, as far as the 96kHz question goes, there's no reason to do it. Productions with budgets of millions of dollars don't do it, if that helps. 48kHz is standard in the professional world, with the exception of some music recording.

+1 Adam. 24/96 works for classical music if you have the right rig and handle it correctly in post, but unfortunately a lot of people don't.

For everything else it satisfies the producer, looks good printed on the label, but the average *viewer* doesn't notice it, couldn't tell the difference and probably baulked at the price.

Cheers.

Thanks guys. Good and useful advice from all. As it looks like even Perrone shoots at 48kHz on his recorder, that's what I'll start doing. Up till now have been recording at 96, assuming the SD/HD analogy was valid, but as Premiere wants to re-render the audio every time you nest a sequence that contains elements that don't match the sequence defaults, guess I'll settle for 48.

Thanks again.

Just another vote for 24/96 - but then I use Sound Devices/Schoeps and hardly ever do anything other than classical. And CF cards are cheap, so no economic rationale for doing otherwise. Would I use it on a Zoom? ????????? I think I'd listen and see if I thought it made any difference. Classical piano is hard to get right and I'll take any help (real or imagined) I can get from higher sample rates and I think I'd be crazy to drop back to 16 bit.

But in the end, I think it all depends on whether YOU hear any difference. I do, or at least THINK I do, my hardware is capable, and it costs next to nothing extra. So I do it. YMMV.

Peace!

Jim .. I thought of you while posting :) and your fortunate clients.

O/T our Convair 440 has slowed to a crawl at Pima, sponsor has gone belly up AFAIK.

Cheers.

There is no reason to record dialog at 96 kHz, unless you plan to slow it down for sound design effects. In fact, the resampling filters can potentially make the end result sound worse than recording in 48 kHz and avoiding resampling.

Regarding 24-bits, it's probably not going to buy you anything on the H4n. I've tested it. It falls a bit short of 16-bit signal to noise. That said, unless you are starved for storage space, it doesn't hurt to record in 24-bits. It's just a bit optimistic.

If you were recording through a clean preamp and good A/D converter, DEFINITELY record in 24-bits. But I'd still stick with 48 kHz for dialog to avoid resampling filters.

There are high-end recording engineers who absolutely avoid digital resampling filters. They will output the 96 kHz signal through a high-end D/A and resample the analog signal with a high-end A/D at the target sample rate in order to keep things organic.

Hi Allan. Too bad about the project - would have made a nice video taking off. So here's the question - would a vintage aircraft engine sound better in 24/96 or 24/48! (:-)

The professional standard for video and film is 48/24, but only when using high-end equipment. I'm told that 96/24 is used for professional audio recordings of symphonies and string instruments and things like that, but again only with high-end equipment.

The last time I was at Abbey Road Studios they recorded at 24/44,1 !

In my audio studio I tend to record everything in 24 bit 96K simply because I can, and that's the default resolution unless you change the preferences. With processor power and storage not remotely a problem, then tell me one good reason NOT to record to the highest quality you can? I'm not remotely suggesting I can hear much above 14KHz as due to my age, above that is lost forever - but I do know that when I get mid 90s 16/48 DAT and ADAT tapes off the shelf and mix them with newer material I can hear a difference.

If I understand what is being said here, it's good to shoot HD even if you only need SD but there's no need to take advantage of better audio, because 16/48 is good enough?

I can assure you that pro studios world wide ARE recording in 24/96 - and a quick look at pro audio (recording and live) equipment shows the format is well established.

With cat 5 and fibre cabling systems, 24/96 is the standard audio transfer protocol now.

In fact - even the home and educational audio world is happily producing on 24/96 - so I have no idea where the notion that they are not has come from?

Hm, so what I'm gathering is that while there may not be a compelling reason to record at 96kHz, neither is there any real reason not to, and you might as well get your original source files in as high a quality as possible because you never know what you might want to do with them at some point. So the minor workflow issues in Premiere are just something you deal with and no reason to compromise your initial recording quality, even if you're just using a Zoom with non-optimal mics.

As someone who's always advised others to shoot in HD even if delivering SD, I feel hypocritical to not follow that same advice with audio. I've been recording 24/96 up until now, so I suppose I may as well stick with it.

All recording at 96kHz does is to extend the top frequency response up to about 40kHz.

Seeing as most people cannot hear above 18kHz anyway, and others much less than this.....

The only advantage 96kHz has is that you can use gentle filters instead of brick-wall ones and minimise the risk of things happening in the audible spectrum.

Yeah, CDs deliver at 44.1 kHz, so music production should be done at that rate. Film is 48 kHz, so sample dialog at 48.

An interesting factoid is that music samplers are used in almost all films today. It's the primary music source in low-budget film. In TV, you generally get samples with a few live soloists. For big budget films, you'll record an orchestra and they will still mix in samples. All of the most expensive, large sample libraries are made for film, rather than for CD music.

And what rate do the samplers work at? 44.1kHz! So much for purity... ;)

Hmmm. I've been scratching my head about where 96kb entered the picture and then recalled that 96/24 (Actually up to 192/24) was supported by Audio DVD which never seems to have really happened in any big way SACD may also have been a reason for going to higher sampling rates - also not a big seller, although I think there are a few thousand titles on the market.

There have been a raft of studies done that seem to indicate that nobody can reliably tell the difference between the higher sample rates.

So the more I think about it, maybe I'm crazy to think 96k yields any real benefits - but I also don't see any reason not to stick with what I'm happy with.

Edit: And by the way, "something" about CD sound annoys me and drives my pianist wife nuts. She's always yelling that digital sound is "cold". So I got a CD player with a tube final stage and now I don't hear any complaints. The problem with all this stuff is that we're dealing with perception - until there's an ISO standard brain/ear combination I don't see that changing.

I've got a DVD player that plays DVD Audio. I bought Yes, Close to the Edge in both CD and DVD-A formats.

I can definitely hear the difference - when I'm paying attention. The main difference is the clear-as-a-bell soundstage. The image is muddier on CD. That said, for casual listening, I wouldn't care, which is why I didn't run out to buy a bunch of DVD-A discs. I don't tend to listen to music with my head clamped in the sweet spot - unless I'm composing/mixing/mastering my own stuff.

The thing about film is that it is delivered at 48 kHz. You could sample at 192 kHz, but you need to deliver at the lower rate. So why introduce a resampling filter into your chain? Why not just record at 48 kHz and avoid an unnecessary level of digital processing?

"Is there ever any reason to record at 96/24?"

Only unless you have a source which in itself is extremely rich in harmonics and so has sidebands outside the "hearing" spectrum thus minimising the folded sideband and resultant artefacts. But only the real golden eared individual could truly differentiate between the two. But in most blind tests, few could, can, might. Although the "muddied" colouration effect is often the first identifier.

Broadcast spec is 48Khz sampling at 24 bits. But yet again gear "specmanship" rears its head through the marketing message promoting 96 is "better" than 48.

In addition, if you are involved in recording such sources and use a long post chain, then there is some justification. I believe MTV Nashville at the MTV Music Awards did a 96Khz recording using the Lawo MC (squared) console @96Khz sampling, but their end product in addition to "to-air" was targetted at DVD sales and involved post sweetening and production.

Some mid-range console manufacturers a while back (and I'll leave you to identify them) wanted to isolate their competitors by offering this concept in the broadcast side of the business - but it was always particularly prevailent and appealing to the music boys and gals. Problem is, 96K sampling eats processing resources, when 48K is more than adequate in the majority of situations.

Finally, and for most product delivered by the many who frequent these boards, customers will always give critical analysis of the images and the video long before they turn their attention to critiquing the audio.

Therefore, is there any real benefit and value in 96K delivery - does the customer really benefit, or does it simply satifisy a golden eared ego?

Thanks Jon and thanks to you also Claire

This is all very interesting. Personally, I do a lot more audio only than audio with video. Classical music is not usually very visually exciting - sort of a bunch of people dressed like penguins just sitting there wiggling their fingers and puckering, etc. And sometimes the backside of a conductor wiggling his or her - well you know.

By the way - sort of OT but I've been watching the London Symphony masterclasses on You Tube lately and have been struck by the atrocious sound. Of course, some of the atrocity is thanks to You Tube itself, but I'm talking about things like hearing the instrument clearly but struggling to hear the almost inaudible and muddy voice - seems like they've set the mic for the instrument but nothing for the performer him/herself. You'd certainly think an outfit like the LSO could get it right!

Oh well, as we enter the age where everybody's idea of audio is a pair of $11.95 computer speakers competing with the system fans and probably a TV running at the same time maybe none of this matters much anymore

To be honest, I didn't expect John to side with the lowest denominator approach. I certainly don't subscribe to the quite insane hi-fi/audiofool understandings but why are we not applying the 'record as high quality as possible' approach to audio? My hearing at 50+ is lacking harmonics now, and a 17KHz tone that drives teenagers wild is totally inaudible to me, but surely the more gentle filtering and other small benefits of increased bit rate have a beneficial effect on what we can hear? Why measure a microphone's top limit, if it's above what we can hear? Would we get away with the Rolls Royce approach - Top limit? Higher than you can hear, sir!

All I can say is that my recordings today sound better than those made in 1995. I don't know if it's the result of the higher bit rate, the extra bit depth from 16 to 24, or just better kit all round. For me, the extra headroom with more than 16 bit recording does help with level 'pushing' not being quite so critical. It only seems to even raise it's head when I'm editing in Sound Forge - and copy an old file into an empty new one - when I get the message warning me. If I need a different format to drop into a video, it's no great shakes to produce one. Virtually all my video projects end up in SD or on DVD, but shooting HD is a small amount of futureproofing.

Ethan Winer has a good presentation on Audio Myths and human perception. It's an hour long, but it's really good info to consider when thinking about these highest quality recording parameters. 24/44.1 is good for most audio things, and bump that to 48k for video. 24 bit is important in that it allows you to record at lower levels without dipping into the noise floor. But recorded correctly (with a good strong signal), 16 bit can sound just as good. 96k (double 48k) can be dithered down to 48k nicely at the final process of the mastering stage. I don't like that many recorders dropped 88.2 (double 44.1) which dithers down more gracefully for CD output. So I and many more just keep the sample rates at the setting of final delivery, 48 or 44.1, though starting out at 24bit, then dithering to 16 after all mastering is done. Classical music has so many wonderful overtones and harmonics happening that the higher sampling rate may just grab some more of that info cleanly enough for us to appreciate - given that one's mics are placed well and your gear is nice enough to move sound through without molesting the signal.

For the most part, we hear what we want to hear when attempting to judge these higher sampling / bit rates.

YouTube - Audio Myths Workshop (http://www.youtube.com/watch?v=BYTlN6wjcvQ)

Given a preamp capable of more than 16-bit S/N ratio, and a final delivery at higher than normal sampling rates, I'd be all for recording at 24/192 or whatever. And, if I'm going to re-pitch the audio - especially when turning a kitten's meow into a dinosaur roar - I recommend recording at the higher sampling rate.

That said, the original post was about using an H4n, which has a front end capable of about 14-bits of S/N. And the audio wasn't going to be re-pitched. 16/48 would likely sound just as good as 24/48, though I'd probably do 24-bits anyway if memory weren't a concern. And recording at 96 kHz is unlikely to sound better when delivering at 48 kHz and might actually sound (very slightly) worse, depending on the downsampling filters.

Imagine that somebody recorded some audio and asked you to critique it. You might hear clipping, distortion, a high noise floor, room echo, comb filtering, bad EQ, a bad mix, and so on. You might hear some zipper noise, if they used 7-bit MIDI controls to do a fade out. But you'd never say, "you recorded this at 48 kHz rather than 96, didn't you?"

Then again, you might sense that the recording is dull and lacks that "it" factor. This could be due to a so-so mic, an amateur player or actor, a low-end preamp, a poor A/D, clock jitter, poor plug ins, and/or poor monitors and room.

Maybe with a specific model recorder, you can get a hint of that "it" factor by recording at 24-bits vs 16 (even if the preamp noise is high), or by recording at 192 vs. 48 kHz. But now we're out of the realm of theory and into the sound of a specific piece of gear. At this point the only way to know the difference would be to record some tests at the various settings and to do some blind listening sessions.

Maybe the H4n's clock is more stable when recording at 96 kHz. Maybe it's anti-aliasing filter is better for 96. Then again, maybe at that setting, you get more crosstalk from the electronics. The difference could even be peculiar to your specific unit.

So, while we can talk about theory, if you really want to know the best approach, you need to test with your own gear. And, if you can't tell the difference, don't sweat it. There are lots of things (mic, mic placement, levels, EQ, effects, mix, mastering) that you can hear and that really do matter.

Thanks for all the great info and viewpoints. I'm enjoying the hell out of this debate, even if I am just as confused as before I posted. The great thing about the audio experts (which I obviously am not) is that you can ask what you think is a simple yes/no question and get passionate positions around four different answers -- yes, no, sometimes and maybe.

I'm just going to sit back and try to absorb some of the collected wisdom on display here.

Now if you guys could weigh in on my mic recommendation issue posted here, I'd love that.

Thanks everyone.

Adam I would say that the mic built in on the H4N are pretty chintzy. Nice for song idea capture, but for a real show you should invest in real mics. If you can't afford to buy mics then there's no reason to ask the question. But since you asked, I would say a pair of Rode NT5 mics would do the job for decent stereo at a decent price. More handy is the NT4 stereo mic. Same capsules built into one X/Y stereo configuration. About 100.00 more, but also runs on 9v, battery, so no need for phantom. You have more placemet options with the matched pair of NT5s. Also I would recommend 4 mics across the stage to get full coverage, but it sounds like you want stereo.

24bit 44.1 (audio only) or 48k (for video) will suffice.

To be honest, I didn't expect John to side with the lowest denominator approach.

LOL

Actually I normally record at 24/96.

But I do know that many classical recordings are done at 24/44.1 with the thinking that it is better to record at 44.1 than to sample-rate convert down from 96 or 192.

Hmm. This is all really quite interesting. I wonder just how bad the sample rate conversion really is with today's systems.

Also occurs to me that the 24/96 recording strategy is best if you may want to re-purpose the material, ie burn a few CD's at 44.1, lay the sound track in with video at 48 and maybe even put it on DVD Audio or SACD or whatever else comes along some day. Maybe just distribute on SD cards or something or download to a system that's connected to your regular stereo system.

You could probably make the same argument for 24/192, but I haven't felt much urge to go there yet.

But back to the first point - how do they actually do the re-sample - if it's the logical equivalent of D (at 96) to A to D (at 44.1) I'm not sure I would expect any noticeable degradation - if they were using high end hardware.

Anyone for 32/384 or 64/768????

Resampling software isn't "bad", though it is an additional step of digital processing.

Going from 96 to 48 is quite simple. Apply a filter with a pass band to 20 kHz and a stop band at 24 kHz, on all the data, then throw away every other sample. Or, more accurately, do the calculation only for every other sample. Going from 88.2 to 48 or 96 to 44.1 is more complicated. You apply the appropriate filter at constantly shifting phases.

The thing is, you've recorded frequencies that you cannot reproduce at your delivery sample rate. Either you can get rid of them when you initially capture the audio with an analog filter, or you can sample higher frequencies and throw them out digitally. Either way, those frequencies get thrown out - even if you apply the standard processing of compression, EQ and reverb.

The exception, in my book, is when you will re-pitch your sounds. In that case, those higher harmonics can make it into the audible spectrum. So, if you're recording for sound design, definitely record at higher rates. You might play your broken glass sample straight in today's project, but you might pitch it down for a slow motion explosion in tomorrow's film.

I guess you could record everything at 96 kHz or 192 kHz and do your whole project at that rate, sampling down only in the final step. That would give you a high-frequency master. Personally, I wouldn't bother if you're using mid-fi gear. The Hollywood pros with their $1,500 mics and Sound Design mixers and recorders don't do this for film, so why bother when using an Rode NT2 into a Zoom H4n? At 192 kHz, you'll cut your real-time plug in count in four. Ouch!

I like to think about audio like this: there are primary issues and secondary issues. Primary issues are things you can hear and you can identify the source of the problem, like distortion, the wrong mic for the situation, or a high noise floor. Spend your money and time eliminating these primary issues. Secondary issues are the lack of polish on the apple. These issues might be improved with things like gold plated, oxygen free cables, super-stable word clocks, premium A/Ds, and 192 kHz sampling. Don't sweat these things until you've got the primary issues sorted.

If you own a studio and are marketing yourself as the best-of-the-best, then by all means develop a 192 kHz workflow with all the esoteric tricks. That attention to detail may or may not result in a better sound, but you can probably bill for it. In other words, it offers a tangible benefit - higher hourly rates.

Unless you have top equipment, top monitors, and a top room - and aren't re-pitching - I question the tangible benefit of higher sampling rates for audio with video.

But back to my earlier point. I'm talking theory. If motivated, do the tests with your own equipment and ears. Do blind tests. If the higher sampling rate offers a tangible benefit, go for it! :)

I agree .. but everyone should try 24/96 at least once.

Cheers.

But wouldn't 64/768 be so much better?????????????

Seriously, I think what I'm curious about is whether going from 96k to 48k actually throws away every other sample or does something more sophisticated like averaging the two samples or interpolating between them. Could be!

And I know enough about sampling to know that my math is too rusty to really understand it anymore - I've been in meetings with the guys who design the signal processing algorithms for disk read write channels and I understand enough to know what I don't understand.

If you were to simply throw away every other sample, you would get aliasing. It's similar to throwing away two of three lines of an image to create DSLR video. It works great on low frequency stuff, but gets ugly with high and out of band frequencies.

From 96 to 48, you low pass filter by passing everything below 20 kHz and cutting everything above 24 kHz. You can then throw away every other sample without penalty. (Or, better yet, only apply the filter calculations for as many samples as you need in the end result.)

Great thread!

My take is that a way to look at this is aquisition, post production and delivery.

In my opinion, 24/96 would most benefit the post production arena as the more information an EQ or effect has to work with the more impact it will have. So the way I see it is beef up the aquisition specs to feed the processors more information so they can create the most natural sound in a digital environment.

Then let the output ditherers cull the large amount of info down to the delivery specs.

But one can deliver 24/48 on DVD & BD so you are really only lowering the sample rate.

Do a test to see if you can hear a difference in post and in delivery.

All A>D converters use oversampling. The required low pass filtering is done digitally. It also seems the use of digital gain is quite common in cheap equipment.
High end audio recording is done using Direct Stream Digital (DSD), a bear of a thing to work with. Arguably only of benefit if delivery SACD which is surprisingly enough quite alive and kicking in Japan. I read the Sydney Opera House has become one of the world's premium recording venues for classical music since Vladimir Ashkenazy took over as conductor of the SSO.

As many have said none of this is very relevant to the H4n.

Is there ever any real reason to record at 96/24?

Yes, if you can hear a difference on your equipment. I can hear a difference on Edirol R-44s, even when editing the material on my laptop monitoring via the headphone output.

I have not done any A/B comparison to see if it "trickles down" to lower bit-rate exports of the audio. But since my time is spent working with the original files, I'm happier when it sounds better.