Jump to content

Recording 88.2kHz and above ...


miker71

Recommended Posts

I'm relatively new to recording digital at such high sampling rates. Whereas I can fully grasp and understand why 24bit is advantageous, I've yet to be fully convinced that recording over 48kHz yields any better recording, especially if I'm using mics that top out at 20kHz and am recording voice. So what are the reasons to choose 88.2, 96 or 192 sampling rates?

Link to comment
Share on other sites

So what are the reasons to choose 88.2, 96 or 192 sampling rates?

I have not really followed any detailed discussion of higher sampling rates, those above the standard 48K we generally employ, but I have heard some discussion amongst sound effects and sound design people what the possible advantages may be. In these cases, real world sounds often go through, by design, a lot of digital signal processing and sample rate manipulation to create certain sorts of sounds (obviously artifical and synthetic sounds that do not exist in the real world but are used in sound design, often to create the feeling of a certain "reality"). So, the higher sampling rates are useful, not as one might assume to make "a better recording" of the dialog, for example, but to provide greater flexibility in manipulating real world wounds for creative effect.

Regards,  Jeff Wexler

Link to comment
Share on other sites

Most of the guys in movie sound who use high SR are SFX guys who want to be able to slow down the sounds (probably the most used treatment for turning a sound into "sound design") while still keeping a lot of detail without a lot of digital artifacts.  Higher sample rates are common in certain kinds of music recording, but as you might expect controversy rages about whether or not it is worth it.  24 bit pretty much everyone agrees is a good idea, when possible.

Philip Perkins

Link to comment
Share on other sites

Yep, I can totally understand the rationale of recording high sampling rate for post effects. Elsewhere people have mentioned to me that A/D convertors just tend to sound better when they are working harder, no matter what the actual frequency range that you are actually capturing. Also they say, that for mastering, it makes sense to record at twice the intended delivery sample rate (so for a CD for example, acquire at 88.2). Just on the basis that you never know when you might want to stretch time, it could be worth using the higher rates when they (and the storage) are available.

Link to comment
Share on other sites

Hello everyone,

I also wanted to mention that Dolby TrueHD (http://en.wikipedia.org/wiki/Dolby_TrueHD) format itself is 24bit/96kHz, this is the format as how it is now presented at home on a HD DVD player. This format also uses a lossless compression, so that you will not get compression artifacts like with mp3 or dolby digital.

If the destination format is 24bit/96kHz would you not want to record in this format as well?

Cheers,

    Take

Link to comment
Share on other sites

Thats really interesting Take... I didnt know that, but that is quite a major development. Here in SA HD hasnt really hit the ground yet (quite a lot of shooting HD, but no HDTV, and products for HD only starting to appear), so its real news to me. Id be very interested to hear the responses of Jeff and the other folks here...

Link to comment
Share on other sites

Id be very interested to hear the responses of Jeff and the other folks here...

One must read the spec for Dolby TrueHD (I hate the name but it is typical of Dolby to adopt a name which implies that everything else is most probably inferior) carefully, as it states that it "supports" 24-bit/96K rates. There are other proposed formats (and some actually in use for quite some time) that also support 96K sample rate --- but it has been rarely used. DTS, the digital release print format that utilizes a double system approach (whereas other release print formats have the soundtrack ON the film) supports 96K but to my knowlege it has never been used. I know there have been some titles announced (and possibly even released) using either one or both of the proposed HD DVD formats (HD-DVD and Blu-Ray), but there is no assurance that these disks will use 96K sample rate. A case in point: when digital television was first proposed in all of its various permutations, most of the specifications provide for exceptionally high quality, resolution, bit rates and data throughput for both the image and sound. The content delivery mega-corporations (networks, studios, etc.) decided for the most part that they would provide most content at only slightly better rates from SD, but the increased bandwidth would allow them to put a lot MORE of the same crappy image and sound out there.

I don't pretend to be an expert on all of this but I know that any discussion of sample rates and quality must take into account what actually happens in the real world.

Regards,  Jeff Wexler

Link to comment
Share on other sites

Ultimately the question that most probably needs to be asked is whether 96kHz presentation systems are merely marketing hype - its highly debatable as to whether anything is even gained by doubling SR to 96, unlike the massive improvements gained in HD video.

So it would appear that while the money spinners can market and push this all they want, we will all still know that its not really going to make a difference. Lets be honest, in any case everyone is still recording dialogue at 48 (or 44.1), so if someone released something at 96k, I'd be highly sceptical of whether it was not simply upsampled digitally after the fact.

Link to comment
Share on other sites

  • 3 years later...

One of the concepts to wrap your mind around regarding higher sampling rates is that while Dr. Nyquist says that higher sampling rates allow for resolution of higher frequencies, that  higher frequency concept is not really very useful for audio frequencies (regardless of what some audio snobs might say!)

The advantage of higher sampling rates is the greater resolution in the reproduction of the original wave.  Dr. Nyquist was a mathematician, not a sound person!  This is a Calculus problem; we are guessing what the connection from one sample to the next is, and by having more samples of the same time period, when reconnecting the samples to redraw the wave, the more samples, the more accurate the reproduction of the original wave. If we could take an infinite number of samples, (and had an infinite sample size!) we would have an exact reproduction of the original wave, and thus 

"analog".

Link to comment
Share on other sites

You gotta love the math.. !

I record at 24bit 48khz on set, I asked some guys in post if the 88.2 range botherd them for dialogue. The said yes cause of syncing problems with visual. I have not researched the reason why yet, but I'm happy with 48Khz so far.

With being a tracking engineer I believe in recording the best master to the highest possible spec for the job. This approach I believe gives more options in post if needed.

Link to comment
Share on other sites

Greg,

You are more likely to have success in higher sample rates if you go all the way to 96k rather then the 88.2.

Its eaiser for a computer to handle doubling or halving rather then the more complex equation of chaning to the 44.1 group of sample rates.

Link to comment
Share on other sites

Thanks Michael

I kinda understand the 'odds' and 'evens' in the math. But I'm happy at the moment with 24 / 48 on Commercials and films. One day I will explore new possibilities in higher sampling rates, but I think even 3D films will settle with the current sound formats from set. Do you know if we are to expect and changes on set sound because of the 3D / HD stuff in the near future ?

Regards

Greg

Link to comment
Share on other sites

Do you know if we are to expect and changes on set sound because of the 3D / HD stuff in the near future ?

Given the limitations of track counts in Pro Tools, I think it's unrealistic and impractical to record at anything beyond 48K/24-bit. I recently visited a mix session where they had three Pro Tools systems locked together to get over 400 tracks simultaneously playing back for one reel; I can't imagine doing this with files twice the size of 48K files.

I'm also skeptical about the real-world limitations of microphone preamps and microphones to capture frequencies higher than 20kHz and a dynamic range over 120dB, at least for dialogue. I can understand a need for 96K/24-bit for recording orchestras or sound effects, but it seems to be total overkill for dialogue to me.

--Marc W.

Link to comment
Share on other sites

I my experience anything slow motion needs 96khz khz or more.

Also I have been gathering and editing sound FX in 96khz

Software plugins tend to work better in higher resolution

1

Given the limitations of track counts in Pro Tools, I think it's unrealistic and impractical to record at anything beyond 48K/24-bit. I recently visited a mix session where they had three Pro Tools systems locked together to get over 400 tracks simultaneously playing back for one reel; I can't imagine doing this with files twice the size of 48K files.

I'm also skeptical about the real-world limitations of microphone preamps and microphones to capture frequencies higher than 20kHz and a dynamic range over 120dB, at least for dialogue. I can understand a need for 96K/24-bit for recording orchestras or sound effects, but it seems to be total overkill for dialogue to me.

--Marc W.

Link to comment
Share on other sites

I have recorded 192 but only ever for FX. I compleatly agree that for dialogue purposes anything over 48 is overkill. The question has to be asked why bother having the improved sample rate support in the Dolby digital systems if no-one is recording the key elements in anything other then 48.

Link to comment
Share on other sites

"I didn't think so...In fact, I could not tell the difference amongst the three rates..."

As I said when tweeking in post EQ and such the difference is more of an issue.

From my experience the difference for playback is very small depending of course on your A/D D/A chain and playback equipment.

" 88.2 is for musik and stereo...  "

"44.1 is for CD audio, and 88.2 is double that rate... "

Correct me if Im wrong... Im sure u will :) ...Thus stereo produktion

//Christian Holm

Link to comment
Share on other sites

For SFX and sound design you can record at any SR you want, but what you deliver to me for a mix better be at 48k.  Recording dialog at anything other than 48k is going to make trouble for a lot of people down the line from you: again, if you acquire at 96k you'd better deliver a 48k downconvert too.  All the edit sessions, Avid, FCP, PT, etc etc will assume everything is at 48k and the deliverables we are contracted to give to our clients are at 48k too.  I understand recording certain kinds of audio at high SRs, but have never found an appreciable difference in how plug ins or mix engines in DAWs work at that rate, other than SLOWER with LOWER TRACK COUNTS.  In other words, 96k is not helping me at all.

Philip Perkins

Link to comment
Share on other sites

I agree 100% with Mr. Perkins. It's fine to originally record SFX at higher sampling frequencies for later processing, but the files submitted for the final mix (or sub-mix) have really got to be 48K. The concept of mixing an entire film at 96K, let alone 192K, boggles the mind.

BTW: my old pal E. Brad Meyer of the Boston Audio Society wrote a terrific AES paper a few years ago on the audibility of 192kHz sampling vs. 44.1kHz sampling. The paper was titled "Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback," and stirred up some controversy in the audiophile community. You can look it up in the September 2007 issue of the AES Journal. Meyer's conclusion was: none of the test engineers could consistently tell the difference, not even with music. That's my reason for believing it won't make an audible difference with dialogue.

--Marc W.

Link to comment
Share on other sites

VST and software engines work better at higher resolution.

I agree 100% with Mr. Perkins. It's fine to originally record SFX at higher sampling frequencies for later processing, but the files submitted for the final mix (or sub-mix) have really got to be 48K. The concept of mixing an entire film at 96K, let alone 192K, boggles the mind.

BTW: my old pal E. Brad Meyer of the Boston Audio Society wrote a terrific AES paper a few years ago on the audibility of 192kHz sampling vs. 44.1kHz sampling. The paper was titled "Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback," and stirred up some controversy in the audiophile community. You can look it up in the September 2007 issue of the AES Journal. Meyer's conclusion was: none of the test engineers could consistently tell the difference, not even with music. That's my reason for believing it won't make an audible difference with dialogue.

--Marc W.

Link to comment
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...