Converting to Higher Bitrates??....we goofed.
- Thread starter dudernut
- Start date
bouldersoundguy
Well-known member
If I recall correctly the last record was done with a vintage Tele into a Super-Champ and the one before that was the same Tele into an AC30. I put a 57 on the Super-Champ in the normal fashion and a Peluso ELAM out in the room, then shifted the Peluso track to line up the direct path with the 57. In both cases a TS9 was ahead of the amp for some tunes. The amps were both on the floor but since then I have become a strong proponent of raising them to chair height or so. Upholstered piano benches work great.
miroslav
Cosmic Cowboy
I have three of my amp in line, on a small riser...like a bench...about 8" off the floor, and the other three are on top of other cabs, so they are almost hair-height.
I would think that one of the big benefits of chair height is being able to hear the amp about the same way the mic is hearing it. When it's on the floor, you get a great tone where you are standing...then you kneel down by the mic, and it's totally different.
Ethan Winer
Acoustics Expert
Time- and pitch-shifting algorithms do not work on a per-sample basis. Rather, they process chunks of audio whose lengths are in the milliseconds. When you hear gurgling from shifting the length or pitch too much, that has nothing to do with the sample rate. The degradation is caused by longer portions of audio being divided and spliced.
I see other misinformation in this thread, but I won't bother to dissect every error.
The short version is sample rates higher than 44.1 KHz offer no sonic benefit. If people want to record at 96 KHz because "why not?" is an acceptable reason, then "why not" 192 KHz or 384 KHz or 100 MHz? Where do you stop? IMO sensible people stop when there's no further benefit. It's not only a matter of disk space either. The faster the sample rate, the more work your DAW and CPU have to do, which limits your total track and plug-in count.--Ethan
Steenamaroo
...
Would the fact that chunks of audio at 96k are sampled more often than the same chunk at 44.1 not mean one could be stretched more than the other though?
Last edited:
Steenamaroo
...
Protools turns a region red when it thinks you've stretched it too far.
I just recorded a chord at 96k and bounced it as 96k and 11.025k, then imported them both to a new 96k session and stretched them to their limits.
Pro tools shows the same limit for both files.
The limit was even the same in a 44.1 session, so fair enough Ethan.
Thanks for the info.
I just recorded a chord at 96k and bounced it as 96k and 11.025k, then imported them both to a new 96k session and stretched them to their limits.
Pro tools shows the same limit for both files.
The limit was even the same in a 44.1 session, so fair enough Ethan.
Thanks for the info.
jimmys69
MOODerator
Steenamaroo
...
Yes, but that is what yoga is for. That and seeing camel toe in tight pants.
I should'a stopped at my first reply. It was pretty good.
CrowsofFritz
Flamingo!
Tim Gillett
Banned
Back to the sample rate issue, in the rarefied atmosphere of the professional audio archiving world, 96khz seems to be the default, I guess to make the digitisation of analog recordings "future proof", whatever that means. Also, the recommendation is for "24 bits or higher"...
I worked on a project archiving voice recordings all made to ordinary analog cassette. Many of these recordings were so bad that the voices were barely audible. (Cassette shoebox recorder using built in omni mic which picked up all the mechanical noises of the machine). Abysmally bad amateur recordings, many of them, and there was a total of 15,000 hours involved.
The powers that be in their wisdom agreed for us to relax the 96khz benchmark and use 48khz. Effectively the upper frequency limit was dropped from over 40khz to a "mere" 22 khz or so.
OTOH, no way would they compromise on the 24 bits. That was sacrosanct. I suspect the thinking was that maybe in years to come, denoising techniques would have improved to the point where the noise in the recordings, which was often louder than the speech, might be reduced by 100db.
This is how silly it can get.
Tim
I worked on a project archiving voice recordings all made to ordinary analog cassette. Many of these recordings were so bad that the voices were barely audible. (Cassette shoebox recorder using built in omni mic which picked up all the mechanical noises of the machine). Abysmally bad amateur recordings, many of them, and there was a total of 15,000 hours involved.
The powers that be in their wisdom agreed for us to relax the 96khz benchmark and use 48khz. Effectively the upper frequency limit was dropped from over 40khz to a "mere" 22 khz or so.
OTOH, no way would they compromise on the 24 bits. That was sacrosanct. I suspect the thinking was that maybe in years to come, denoising techniques would have improved to the point where the noise in the recordings, which was often louder than the speech, might be reduced by 100db.
This is how silly it can get.
Tim
bouldersoundguy
Well-known member
The amps are miced and that's fed to the player, so he's hearing what is getting recorded. The Vox was tilted back but the Fender was actually aimed 90° from the player. The main benefit I see is reduced reflection from the floor.
And we recorded at 48/24 (to keep it on topic).
Bobbsy
Boring Old Git
Totally OT aside about the hazards of technology...
Many years ago, in the early 1970s, my first job in TV was a local station in Calgary, Canada. Besides the "proper" studio and mobile gear, we also had a little two camera black and white setup in a van that was used mainly to record the weekly City Council meetings and bi-weekly School Board meetings. As you can imagine, it was a boring job delegated to the juniors like me.
An ongoing problem at the City Council meetings was the low light levels--and they wouldn't let us add extra lighting. After tons of complaints, the chief engineer came up with a solution and changed the vidicons in the cameras to ones that were sensitive into the infra red range. Great idea! The first meeting with the new tubes suddenly looked bright and clear and we were all happy...until, that is, one of the councillors lit a cigarette. (This was back in the day when smoking in public was still the norm.) The heat from his lighter then cigarette end bloomed like mad! It was like one of the space ships out of Close Encounters!
So, the moral of the story is that technical changes made for the best of intentions can sometimes have unexpected consequences!
Ethan Winer
Acoustics Expert
As I expected. I admit I never actually did that test, but just knowing how the process works I was sure that the sample rate wouldn't matter. So thanks for going to the trouble to test it and post the result.
--Ethan
Ethan Winer
Acoustics Expert
LOL, indeed.
I do all my home recording at 44.1 / 16, though I might use 24 bits for a remote classical orchestra concert. But that's just for safety, not better quality. Many people don't understand how incredibly clean 16 bits really is. I recently did a test with four different music clips recording ten generations each through my Focusrite Scarlett at 16 bits. Here are the test files, and I'd love to hear from anyone who emails me which files they think are which:
Converter Loop-Back Tests
--Ethan
bouldersoundguy
Well-known member
And many people don't understand that 16 bit has exactly the same resolution (dB per bit) as 24 bit but a higher noise floor. With careful level setting it's indistinguishable. I use 24 bit because I don't always have time or attention to devote to precise level setting.
Ethan Winer
Acoustics Expert
Exactly. The notion that 24-bit audio is somehow cleaner or has more resolution than 16 bits because the "steps" are smaller is simply wrong.
I guess you never had to cope with the limited dynamic range of analog tape.
--Ethan
bouldersoundguy
Well-known member
I guess you never had to cope with the limited dynamic range of analog tape.
At this very moment I'm transferring some old live recordings I did on cassette in the 90s. I seem to have done an okay job of setting levels while also mixing the shows. I tweak the head alignment to get the cleanest possible HF output without adding any noise. It involves balancing any L-R differences with the interface gains, turning off any NR, summing L-R, cranking the playback HF and adjusting the head alignment for maximum clarity and minimum phase interaction. Doing that improves the inherent S/N and makes Dolby track better. Of course I set the playback back to normal (stereo, no HF boost) and the NR to whatever the tape says for the transfer.
We (son and I ) record everything at 44.1kHz and 24bits (unless for DVD vid).
I have followed many higher sample rate arguments over the last 6 years or so and most of the pros I read about use 44.1/48 as well.
Some made the argument that 96k halves the latency but a worker in the AI field (TAFKAT, Sound on Sound forum) has stated that computers can take longer to process 96kHz files and so the benefit is largrly cancelled out. I am just an old amp bottle jockey, I only go by what I read!
I am also a BOF like Bobbsy. I came up thru valves, transistors, AM, FM, FM stereo vinyl and cassette and open reel. The first CD I ever heard was Bat out of Hell on a 90W quad system with Castle speakers (bigger versions of LS3/5As) and was blown away by this blast of sound from velvety silence. I could not afford a CD player for years after (never mind the hi fi!) but black disc never did it for me ever again!
We have "un progressed" with sound quality. MP3 and over here we have a thing called BAB. B awful I am told (I can't get a smell at my 10/20 and I am f...d if I am putting up an aerial! And do not knock FM! A live Prom broadcast recieved on one of the top Revox or Sony tuners was THE best sound quality you could get next to being in the van!
Dave
I have followed many higher sample rate arguments over the last 6 years or so and most of the pros I read about use 44.1/48 as well.
Some made the argument that 96k halves the latency but a worker in the AI field (TAFKAT, Sound on Sound forum) has stated that computers can take longer to process 96kHz files and so the benefit is largrly cancelled out. I am just an old amp bottle jockey, I only go by what I read!
I am also a BOF like Bobbsy. I came up thru valves, transistors, AM, FM, FM stereo vinyl and cassette and open reel. The first CD I ever heard was Bat out of Hell on a 90W quad system with Castle speakers (bigger versions of LS3/5As) and was blown away by this blast of sound from velvety silence. I could not afford a CD player for years after (never mind the hi fi!) but black disc never did it for me ever again!
We have "un progressed" with sound quality. MP3 and over here we have a thing called BAB. B awful I am told (I can't get a smell at my 10/20 and I am f...d if I am putting up an aerial! And do not knock FM! A live Prom broadcast recieved on one of the top Revox or Sony tuners was THE best sound quality you could get next to being in the van!
Dave
Farview
Well-known member
I know I'm a week late with the answer....
You need to change the sample rate of the session back to 48k. When Nuendo asks if you want to convert the files, say "no".
Once that is done, hit play to see if everything is playing at the right speed.
If it is, change the session back to 96k. This time, when it asks if you want to convert the files, say yes. It will convert the files to 96k and everything will work.
The other way to do this is to start another session at 96k and import your 48k files into it and make sure that you tell it to convert the files to the session sample rate.
Not necessarily wanting to get into a flame war, but I know very few professional engineers that record at 96k. Even when they do, it's normally at the request of someone who doesn't know any better. Unless you are recording dog wistles, there isn't much point in going any higher than 48k. I record at 44.1k.
You need to change the sample rate of the session back to 48k. When Nuendo asks if you want to convert the files, say "no".
Once that is done, hit play to see if everything is playing at the right speed.
If it is, change the session back to 96k. This time, when it asks if you want to convert the files, say yes. It will convert the files to 96k and everything will work.
The other way to do this is to start another session at 96k and import your 48k files into it and make sure that you tell it to convert the files to the session sample rate.
Not necessarily wanting to get into a flame war, but I know very few professional engineers that record at 96k. Even when they do, it's normally at the request of someone who doesn't know any better. Unless you are recording dog wistles, there isn't much point in going any higher than 48k. I record at 44.1k.
Hi Jay,
Can I also add that if you delve into the specc's of many AIs they do NOT double the bandwidth for a 96kHz rate instead the claimed upper frequency response is something like 25kHz!
Then, there are very few mics that even GET to 20kHz leave alone surpass it.
Dave.
Can I also add that if you delve into the specc's of many AIs they do NOT double the bandwidth for a 96kHz rate instead the claimed upper frequency response is something like 25kHz!
Then, there are very few mics that even GET to 20kHz leave alone surpass it.
Dave.