Recording At 24 Bit, 96 KHZ

  • Thread starter Thread starter Village Idiot
  • Start date Start date
Village Idiot

Village Idiot

The Love Butler
This is your resident Village Idiot...

I am recording my band's demo now..
My assistant told me to record the tracks at 24-bit, 96 KHZ instead of the regular 44.1 I have been doing it at.

So far, so good, great sound.
Does this mean there are clear advantages in usuing the 96 KHZ?

Now...This is sucking up more of my CPU it looks like, correct?
And will it automatically all mix down to 44.1 KHZ when I burn it to CD, being 44.1 is the expected standard?
Or do I have to get it down somehow to 44.1 myself?

I am recording using my Audiophile 2496 card and N-Track 24 bit.

Any opinions would help!

VI
 
Gidge?

Have you tried recording with your Audiophile at 24/96 yet?

:D
VI
 
Your assistant tell you what to do? Are you sure you're not his assistant?

Look, I've argued this a million times, and no one ever listens to me, but if your music's final destination is going to be a CD (44.1 kHz/16-bit) then you are better off recording at 88.2 kHz/24-bit.

The only situations where I would ever record at 96k instead of 88.2 are:
1) the destination format is film, DVD, or DVD-Audio
2) the project is going to be mastered in the analog realm by someone who has great analog gear and excellent D/A and A/D converters (in other words, by a mastering professional).

Just trust me on this one.
 
Ok Then...

Thanks for your response, Charger.

Why 88.2?

Does this take up less CPU?

And what exactly am I listening for sonic-wise that makes recording at the higher KHZ's a better thing?

Charger...Do you know if burning the tune to CD will automatically default to 44.1, or do I have to do that within my software program?

Where are you, Gidge? (!)

VI
 
88.2 is a lot easier for the computer to toss the data away when you convert back down to 44.1 (it just averages two of them and then tosses every other sample).

The questions you have to ask yourself:

1. Can I hear a difference???

2. Is it worth the extra storage space??

3. Is it worth the reduced track count??

4. Can my poor PC keep up with processing that much extra data with plug-ins or will it roll over and die??

5. How much extra time will it take to resample down to 16/44.1 for CD output and more importantly, will there be any difference in my sound when I do????

I'm sticking with 24-bit/44.1kHz right now and when I have a 10Ghz PC with a terrabyte of storage MAYBE I'll reconsider, but that's my personal preference....
 
Thanks for your input, Mr. O'Brien:

My computer is in no way a hot rod.
It's a Compaq Presaro with an AMD K-2 processor,
I think (!) 533, with 250 megs of RAM.

I was thinking the plug-ins will kill this thing running at 96 KHZ.

But you are saying I am better off stability-wise running it at 88.2, though ultimate stablilty would be 44.1, but I may lose some crucial sonics?

VI
 
Simply put, recording at higher bit depth and frequency response affords you a buffer for signal degradation during "cook down" to CDs 16/44.

Stability will be a function of your computers muscle. If it plays it back without hickuping, you're fine.

Make sense? :D
 
Why thank you, Grippy ol' boy...

But another question, and I suppose it is a subjective one...

If recording in the digital domain means no "degradation" of the recorded signal, then what am I missing recording at 44.1?
Will I be missing certain sonics that the average ear would otherwise hear on a 88.2 or a 96 KHZ recording?

I have recorded the first 4 Drum tracks to our Demo, and my CPU resources at the Bottom of my N-Tracks shows about 53% usage.

Since we are recording all of the Drum tracks first, would it be ok for me to dump all the tracks so far recorded at 24/96 KHZ to DAT or Direct CD, and continue to do so when CPU resources hit 50% or more?

Or am I better off recording the remaining tracks at a lower KHZ rate?

I'm giving myself a headache...

:D
VI
 
Question for yah..If you had a camera and were gonna take a picture of a once in a lifetime event..something that you would always treasure..and you had a choice to capture it at 300 pixels or 600 pixels..which would you choose?

If you knew you were gettting ready to nail your all time best guitar solo..and have a choice to capture it 44.1k 16bit or at the highest resolution available to you which is 96k 24 bit..??

Remember, the initial capture of the sound at the higher resolution gives you the flexibility to downsample if you like and as far as you like,and from my own expieriences the downsampled version is sounding better than the same material that was done at 44k 16..this is gonna be particularly true in the multitracking environment when the track count increases..it also gives you the best archive of the recording for future needs..

Here are some ways to maximize the effectiveness of the 96k recording. This will be based on being able to playback 12 tracks at 96k..

When tracking ,use the minimum amount of tracks needed for playback to lay down each sucessive track..
Archive each of the individual tracks to a CDR as wave files so that they can be imported back into the puter at a later date if needed...there will really be no need to leave these wave files on your hard drive in the puter after the project is done. This takes care of the storage problem.
Use submixes during the mixing stage..mixdown the rythem section( drums, bass, rythem guitars..etc) into stereo tracks. This would be probably be a good idea regardless to sample rate..and can be done without the lead instruments,backing vocals and other fills being played at the same time. Again, disable or archive the tracks after you have working submixes.
When you have the project mixed down to 2 tracks...downsample to 44.1k before you do any processing to the main buss..this will allow you to make adjustments if the downsampling created any sonic degraditon surprises..which it shouldnt.
Use software for the conversion..the hardware SRC.s seem to be the ones that have the problem with conversion done in realtime and do better with 88.2 conversions rates..
Sound Forge has done a good job with the conversions so far for me..and a 96k to 44.1k is a piece of cake for software..those puters just love the simple math like that :)
Don,t dither down to 16 bits until the absolute last step..
Backup everything you do at every step if youre not certain where things are heading..another fabulous feature of puter recording :)

Best of luck into your venture into high resolution recording!!
 
Elbenj:

Thank you soooo much for your wisdom and insight.

All comments have been a help to me so far, yours kinda put it into a great perspective for me to chew on.

I will post one of the demo songs at the MP3 mixing clinic next week.

It will be the best stuff I have ever recorded, so I hope I pass the test with youse guys!

:D
VI
 
Village Idiot said:
Thanks for your input, Mr. O'Brien:

My computer is in no way a hot rod.
It's a Compaq Presaro with an AMD K-2 processor,
I think (!) 533, with 250 megs of RAM.

I was thinking the plug-ins will kill this thing running at 96 KHZ.

But you are saying I am better off stability-wise running it at 88.2, though ultimate stablilty would be 44.1, but I may lose some crucial sonics?

VI
Click to expand...
I have nothing insightfull to add.It's just nice to know I'm not the only one desperately in need of upgrading!
 
With your PC I wouldn't even consider doing 96khz. The very, very slight added sonic advantage would be overcome by the endless headaches caused by the strain on your CPU.
 
Look, I've argued this a million times, and no one ever listens to me, but if your music's final destination is going to be a CD (44.1 kHz/16-bit) then you are better off recording at 88.2 kHz/24-bit.
Click to expand...


I totally agree and I can't understand why no one listens. Finally someone who agrees with me. Thank you!


/fim
 
But what about just recording in 44.1 if it's going to be for a 16/44.1 CD anyway?

I've done my own experimenting with 16/44 compared to dithered 24/44 and nondithered 24/44, and I do think dithered 24/44 sounds best and so can see choosing to record in the higher bit resolution. But I'm wondering, is there really a meaningful (ie, has anyone here HEARD) diff in tracks recording at the higher vs. lower sample rates once at 44.1 and beyond? (Sorry if I have my terminology wrong. I hope it's clear what I mean. If not, let me know.)

Thanks,

gg
 
But I'm wondering, is there really a meaningful (ie, has anyone here HEARD) diff in tracks recording at the higher vs. lower sample rates once at 44.1 and beyond?
Click to expand...
No. I think 9 out of 10 people will get the same results using 44.1 or 88.2. And that 10th person will be able to outspend you by hundreds of thousands of dollars.
 
charger said:

No. I think 9 out of 10 people will get the same results using 44.1 or 88.2. And that 10th person will be able to outspend you by hundreds of thousands of dollars.
Click to expand...

:D

Yeah. that's kinda what engineers I've talked said too. One guy is going to the Pro Tools HD system but it's driven his work with video, not deficient sound quality in his current system.
 
There is definitly a sound quality increase from 44.1 to 96

Can most people hear the difference? Not unless you have trained ears.
 
And not unless you're a dog or a bat. The difference between 44.1 and 96, to me, are not noticeable. Like most people, I can't tell a difference. I do, however, hear a distinct difference between the various bit rates. 24-bit is just noticeably superior.

What good is it if your converters can capture frequencies that go beyond the range of human hearing, and more importantly, beyond what your mics (or the listener's speakers) are capable of reproducing accurately?

Using elbenj's snapshot analagy, a higher bitrate will give you the best possible resolution, as more bits are used for each "snapshot." The sampling rate just has to do with how many of these snapshots are used in a given second to provide a coherent illusion of the passage of time (as with movie frames). It also has to do with the frequency ranges it is able to capture accurately. And in both departments, it seems like we're just fine at 44.1. I hear the passage of time quite well. I've also maxed out in terms of the frequencies I can perceive beyond that point. And it's detereorating the older I get.

Geekgirl: I heard a really good explanation / answer to your question recently - "If it all is mixed down to 44.1 / 16-bit anyway, then what's the point?"

And my friend (who happens to be an apprentice for a very well-recognized recording engineer by the way), explained it this way:

"20 years ago, most analog mediums were ultimately recording down to 1/8" cassette tape as it's final destination, but does that mean there was no benefit to tracking with 2" tape? It's the same principle."
 
I havent seen an article about a sound engineer using 44.1

They seem to stick with the highest resolution they can get their hands on. Are they fanatics or is every drop counting in the recording industry?

I added that so I wouldnt have a biased opinion
 
You must log in or register to reply here.
Back
Top