The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
96K really sound better? Studio Monitors
Old 22nd December 2008
  #61
Lives for gear
 

To test the difference of ONLY the sample rate and bit depth the test procedure I suggested works perfectly. As just one 24/96 take is used for both samples the equipment used for recording does not differ; if different systems are used there are also other variables besides just sampling rate/depth in the equation. In my system the same ADC and DAC is used.

Also, as the better 24/96 sample stays intact nobody can claim that the stakes are against the better sample, if anything the 16/44 version is not as perfect as possible (but just about, depending on the dithering used). But, can you tell them apart?

I have made two 10 and 14 minute long test samples using 24/96 classical music originals where the sample rate varies randomly every 30 seconds. Even though I know where the changes occur I can not hear it.
Old 22nd December 2008
  #62
Gear Head
 
Clairaudience's Avatar
 

Quote:
Originally Posted by yeahiknow View Post
Hey there,

That's only part of the story.

By the way, we do have 384kHz PCM format recording, if you like [Pyramix].

Less filtering? No filtering?

Direct sampling.


Regards.
Oh, now you've cought my interest! I can see the clear logic in Dan's scientific arguments though... I'm afraid I'm a bit confused now.
But the 384 KHz PCM format is still 1-bit, or is it 24 bit? I know Pyramix is able to process the HUGE amount of data needed for such high samplerates. I love the sound of DAD converters, which, among others, can sample at those frequencies. I also think that the impulse responds part left a bit to be explained from Dan's otherwise great insight in this technology. Plz, what's the other part of the story?

This is my setup:
I use Genelec 1039A's (Stereo, free field freq. response from 31 Hz to 20 KHz, +/- 2,5 dB), 8050A's/7071A (Stereo/Surround), K+H O 300 D's (Stereo) and a pair of Yamaha NS10M's. Monitor-wise (dampening, reverb decay time, etc.), my tuned control room is one of the best in my country. The 48-channel SSL Duality console I use has an audio bandwidth of 2 Hz - 200 KHz. I use most of the analogue outboard gear you would associate with the word "classic".
As you can see, the studio setup can theoretically cope with audio frequencies as high as 200 KHz. But I'm afraid I would not be able to hear them on my current monitor system.
So far so good, and in perfect harmony with Dan's white papers. But what about the lower impulse response time? I can see that it might have an impact soundwise... But that would still have to outweigh the "cons" of higher samplingrates (added HD space needed, limited tools for audio production, etc.).

Best regards

Last edited by Clairaudience; 22nd December 2008 at 09:11 PM.. Reason: Ohh, just remembered...
Old 24th December 2008
  #63
Lives for gear
 
Frost's Avatar
 

Quote:
Originally Posted by norman_nomad View Post
For all those 88.2 er's out there thinking that it's an easy shot to 44.1k (the computer just divides by 2)... just know that most DAWs will upsample to the lowest common multiple of your destination sample rate before converting which should give equal results across all captured sample rates.

So might as well go 96k if you're running into plug-in troubles.
Everyone re-read the above statement. I wish more people had a clue about this. Thinking that you can record at 88.2 and just divide by 2 just shows ignorance. The fact that coding engineers and marketing people even came up with this as one of the sampling frequencies is also very telling.

Frost
Old 24th December 2008
  #64
Lives for gear
 
Robert Randolph's Avatar
 

Quote:
Originally Posted by yeahiknow View Post

A 384kHz converter will not filter out a 333.333khz signal.
Yes it will. It filters everything below 192khz. Nyquist is your friend

Quote:
And we are not talking about a simple 333.333kHz tone anyway, right? It is a "click" [an impulse]. The theoretical "ideal click" is made up of all frequencies, no?
No, not really. In theory, the impulse is an infinitely sized with zero 'width', thereby it has no frequency in theoretical concern.

A signal made of 'all frequencies' is infinite bandwidth noise, which is quite nearly the opposite of a dirac impulse in practicality.
Old 24th December 2008
  #65
Gear Addict
 
BluegrassDan's Avatar
I understand the need to distinguish between sound differences in 44.1 vs 96 (or whatever in between or beyond)....

however.............

Which of these scenarios do you think poses more observable differences?

1. A vintage Martin guitar recorded through high quality converters at either 44.1 or 96?

2. The same vintage Martin guitar with fresh strings, vs three-day-old strings?

I personally do 48, simply because all the mastering houses in Nashville always ask for 24bit/48k.

Don't forget about the musician variable! We're recording music fellas!!!
Old 24th December 2008
  #66
Lives for gear
 
thenoiseflower's Avatar
 

If this were a vote....

I say yes (96 vs. less than 96), it sounds more "hi-fi" for lack of a better term, but I do not have an issue working at any sample rate.
Honestly tho, I think 88.2 is more musical. I tend to want to run sessions at that rate.

Does anyone else feel like there is TOO much openness in the top when recording at 96, for "popular" music anyway (rock/pop/hiphop/metal/ect)

and to the whoever said they'd bet no one can hear the difference in any different sample rate may wanna rethink that. Often even in something we havent worked on it can be obvious, you just have to know what you are hearing.
Old 24th December 2008
  #67
Lives for gear
 
Robert Randolph's Avatar
 

Quote:
Originally Posted by yeahiknow View Post
Hey there,

You might want to read what you wrote. You got that wrong [first part].

I probably should have said "all harmonics", though. In any event, the click / impulse being discussed is not a 333.333khz tone.

You are also assuming that there is a filter implemented in the 384kHz device. ?

Regards
A 16/24/36/etc..-bit, 384khz stream will be able to represent frequencies up to 192khz due to the nyquist theorem. We're not talking about DSD here from what I understand, so this is pretty straight forward.

Even IF there was no nyquist filter implemented, it's simply not possible to accurately represent a signal higher than 1/2 the sampling rate.

I think it's a fair assumption that any reasonable converter would have a filter implemented anyways. There is non-program material that could foldback into the audible range without it. If a nyquist filter were not required then we could 'practically' get away without one at 96khz sampling as well since microphones can not capture much hypersonics anyways. As said though, non-program material is the concern, and that still doesn't change the fact that you're limited to 1/2 the sampling frequency.

It is certainly true that a 3us pulse is not a 333.333khz signal, and if it were to be considered as such it would be of indeterminable content from the nonlinearities in the equipment involved.
Old 24th December 2008
  #68
Lives for gear
 

Quote:
Originally Posted by Robert Randolph View Post
It is certainly true that a 3us pulse is not a 333.333khz signal, and if it were to be considered as such it would be of indeterminable content from the nonlinearities in the equipment involved.
I made that argument just to show how pointless it is to use microsec pulses to compare systems not designed to take them. Many if not most people do not apparently realize that a 3 microsec pulse has no frquences lower than about 333.333 khZ which are 1) not naturally produced from any microphone 2) if they exist for some reason they are filtered out of the signal before AD 3) nobody can hear frequences even decade lower. Those graphs do not represent reality. It is like proving that a space shutle is a better car than a Toyota because it can stand zero pressure of space which Toyota can not.
Old 24th December 2008
  #69
Lives for gear
 
Tube World's Avatar
I clearly hear my top end when I record at 96, but we still have some artist who don't seem to care. Madonna's latest CD was recorded at 44. Doesn't she know that her CD would sound soooo much better if it was recorded at 96?
Old 24th December 2008
  #70
Lives for gear
 
Beyersound's Avatar
The answer is still no.................
Old 24th December 2008
  #71
Lives for gear
 
mdjice's Avatar
 

Yes it doesn't...
Old 24th December 2008
  #72
Lives for gear
 
jupiter8's Avatar
 

Quote:
Originally Posted by yeahiknow View Post
Hey there,

I am not privy to why they used a 3us pulse, but I think what they are getting at is that the impulse response is better, over the entire range, when you sample faster.

Whatever pulse they used is apparently registering in the various systems, and registering differently, unless they are literally making the whole thing up.

Without getting into a discussion about the merits of recording supersonics, it seems undeniable that gentler filters, farther away from the band of interest yield better performance in what is traditionally seen as the audio band [i.e. 20 - 20,000].

So the rationale in favor of the faster system sampling rates [aside from the numerous practical benefits previously mentioned] is more concerned with getting a better result in the 20hz - 20-khz range, than with trying to reproduce 60kHz tones or something.


Regards.
Impulse response and bandwitdh are 2 sides of the same coin. It's the same thing described in two different ways. The post you made simply tells me that you don't understand the Nyquist-Shannon theorem or digital audio in general. It's a common misconception though.
Old 25th December 2008
  #73
Lives for gear
 
DSD_Mastering's Avatar
Quote:
Originally Posted by Cato View Post
still waiting for 384.
We record/master at 384 and 5.6MHz.. do we win??


Regards,
Old 26th December 2008
  #74
Lives for gear
 
DSD_Mastering's Avatar
We try to push hi-rez formats to everyone. It's easy to distinguish RBCD and hi-rez in a good room.

I posted over in the mastering section but it applies here as well. I feel it should be mandatory for anyone that wants to get into recording/mixing/mastering to view this video.


Deep Listening: Why Audio Quality Matters

Roundtable
Participants: Steve Berkowitz, Greg Calbi (moderator), Evan Cornog, Michael Fremer, Kevin Killen, Craig Street

Regards,
Old 26th December 2008
  #75
Lives for gear
 
mobilemozart's Avatar
 

One more vote for 192!

I always bounce my masters in 192kBit mp3. Sounds far better than 96!
Old 26th December 2008
  #76
Lives for gear
 

Quote:
Originally Posted by mobilemozart View Post
I always bounce my masters in 192kBit mp3. Sounds far better than 96!
The discussion at hand is about samplerate, not bitrate. :-)

I think few people will argue against that a 192kbit/s mp3 would sounds clearly better than a 96kbit/s mp3.


/Peter
Old 26th December 2008
  #77
Lives for gear
 

I went to 88/96k ages ago when I was using a lot of plugins. Especially soft synths were improved greatly.

I seem to have an ear for hearing digital aliasing, it grates on my like nails down a blackboard.

The higher the sample rate the less the audible aliasing. I found the bass a lot tighter as well.

These days I use analog synths but still use the odd plugin etc...
Old 26th December 2008
  #78
Gear Addict
 
Greg Heimbecker's Avatar
Quote:
Originally Posted by bcgood View Post
Go 1 bit-5.8MHz DSD or go home.

heh lmao


I do a ton of classical and jazz stuff along with americana and folk and the higher rates just seem smoother, deeper and wider. Even the jump to 176.4 from 88.2 in my sennheiser 600s is not at all subtle when I'm on location. I'm loving 192 or 176.4 for classical stuff. I just wish I could reliably run 24 tracks @192 with the Pro Tools rig for jazz stuff...
Old 26th December 2008
  #79
Gear Head
 
Clairaudience's Avatar
 

@ yeahiknow...

Ohh... I suddenly see how I've been mislead...

I know that we can noticably hear and percieve frequencies under 20 Hz and above 20 KHz... I'm also aware that an improved impulse response time is better than a lower...

I realize you are right!

No need for you to elaborate further – I know that a detailed explanation on this subject is way too loooong... I understand the principles, it's just that I really "belived" Dan Lavry. I figured that his expertise must have some authority. Obviously I was wrong, not taking into account the fundamentals of capitalism...
Disregarding his "scientific arguments", my setup and my (Dan's) "arguments" is of course irrelevant...

I'd really like to thank you for the insight and taking the time to explain...

Happy holidays
Old 26th December 2008
  #80
Gear Head
 
Clairaudience's Avatar
 

Wikipedia...
"An Internet troll, or simply troll in Internet slang, is someone who posts controversial, inflammatory, irrelevant or off-topic messages in an online community, such as an online discussion forum or chat room, with the intention of provoking other users into an emotional response or to generally disrupt normal on-topic discussion."

A troll? I don't get it... I was sincere.

Plz do not respond. That would by definition make you a troll...I think...


Best regards
Old 27th December 2008
  #81
Lives for gear
 
CZ101's Avatar
 

Will higher sampling rates allow for more detailed/spectrally-retained time stretched audio? For example - a recording at 96khz will sound less aliased when time stretched (expanded) than a 48khz recording given the same algorithm?
Old 28th December 2008
  #82
Quote:
Originally Posted by DSD_Mastering View Post
We try to push hi-rez formats to everyone. It's easy to distinguish RBCD and hi-rez in a good room.

I posted over in the mastering section but it applies here as well. I feel it should be mandatory for anyone that wants to get into recording/mixing/mastering to view this video.


Deep Listening: Why Audio Quality Matters

Roundtable
Participants: Steve Berkowitz, Greg Calbi (moderator), Evan Cornog, Michael Fremer, Kevin Killen, Craig Street

Regards,
YES I agree!! THANK YOU for POSTING this in this thread as well as others!!!! YOU RULE!!

And because of you posting it some days ago in another thread, I've been thinking about this discussion non-stop for about 2 days straight. I've been doing more than the average researching/reading about PEOPLE who I think are innovators, instead of strictly, fact and theory about circuit equipment designs I lust after. Investing in people, can be much more advantageous than any one machine, but we as gearslutz know the machines are 1/2 of the equation....certainly not the whole equation. "People" are the one's who invest in stretching the limits of technology as we know it, and intern we are always exploring our own potential. We are learning things about ourselves everyday and to think that we know everything there is to know about individual hearing/listening/perception, and how it applies to what we do, we are kidding ourselves. [Well....the Pro's might have it sorted] What we need is more innovators in our field, people who will continue to push the boundaries and rules of our subjective art. You might view it from the outside in, which I am sure a lot of audio professionals do, given their experience. I do think there should be a trend in favor of simply listening, connecting with the material. For me, I intensely appreciate the free thinking designers that design tools driven by that intention, which I think is the ultimate goal of the product from A all the way to Z for everyone involved.

Gear is just like Music to me sometimes, as you can really tell when the designer puts everything he has into the end result. Gear is more dynamic than its face value I think, as its a tool for the task designed to do that task well. I think PEOPLE are responsible for pushing the current LIMITS in technology. Anyone can say what they will say, but I'm talking about the preservation of MUSIC in its highest fidelity. Most ALL audio professionals use higher quality, higher fidelity equipment to capture a PRICELESS moment in air and time, so why then should the consumer be deprived of the same experience? It makes sense; they pay a fraction of the price to get a very BIG fraction of the performance. I'm not saying CD's suck, they could sound better, as we are approaching the year 2009, that is all. I DO think we already have a HI-RES format in vinyl, but you guys don't want to hear about that; you want to continue arguing about sample rates. The CD is seemingly a stranglehold of consumerism, as the business end is evolving at a different pace than the technology. I don't claim to know sh#t about this; but that marriage [technology/science/evolution, and the business/marketing/distribution] hasn't been happy for quite some time. Its like they just stayed together for all the kids.

Music and our interesting art of recording and listening, collide and deliver the ultimate personal experience for the first and end user. Its a beautiful mystery for everyone, and it means more to me than most anything else, so I don't know how that impacts the business end, I guess it helps because I BUY the record that I BUY on ITUNES to PREVIEW. I still rock the IPOD, but I sit in front of my Solo 6's listening to records on my Denon Digital Turn Table SPDIF'd into my Rosetta 200, for hours in a trance. We all appreciate things differently and our listening/emotional experience is different. I agree that our experience is degraded nowadays with these digital convenience mediums [perhaps similar to the way gearslutz degrades the personal experience of using GEAR??] and its important that I respect the artist in my own way, and emotionally connect with the material, and the choice is an individual one, taking time to properly listen however you want in your room, by your standards. We ought to get back to truly listening to the music we are inspired by, on an emotional/personal level, even with whatever 2 dollar radio you got. Listening and learning is a intense form of expression, should you decide to harness your ability to use it, and I think no one can take anything any from our personal musical experience. The whole thing in my mind makes this thread and other "Best" Sample Rate threads nothing more a waste of time. There is no point in arguing sampling rates, it misses the point entirely to the relevance of the end user.
Old 28th December 2008
  #83
Gear Maniac
 

Well said Adam
Man that video really shifted my whole paradigm of thinking. That's why it is so important for me to listen to people who have the experience and credibility that those cats at the round table had.
Some questions can simply be answered by trying out your question.
Thats enough I have to go finish watching that. Its my new drug
Old 28th December 2008
  #84
Lives for gear
 
s12512's Avatar
most of everything sounds great at 44. i havent moved up to 48 yet. im not a full-time engineer but im a full-time producer. so i guess engineering sometimes falls under that!thumbsup if u want 96 to sound better ur mind will tell u it sounds better. if u do a blind test and its hard to tell the 2 apart then isnt it kinda pointless to use anything over 44/48?? most or all of the hits that u hear are recorded at 44/48. r they not smooth/wide/detailed enuff?? if those dudes dont use it, then wuts the point of putting more stress on ur computer/drives/interfaces/etc??
Old 28th December 2008
  #85
Gear Addict
 
jwleeman's Avatar
 

what i gather from bob katz' articles / books, recording at higher resolution & bit depth allows the music to be more effectively mastered, so if you can do it, i would.
Old 28th December 2008
  #86
Lives for gear
 
Froombosch's Avatar
 

Quote:
Originally Posted by DSD_Mastering View Post
We try to push hi-rez formats to everyone. It's easy to distinguish RBCD and hi-rez in a good room.

I posted over in the mastering section but it applies here as well. I feel it should be mandatory for anyone that wants to get into recording/mixing/mastering to view this video.


Deep Listening: Why Audio Quality Matters

Roundtable
Participants: Steve Berkowitz, Greg Calbi (moderator), Evan Cornog, Michael Fremer, Kevin Killen, Craig Street

Regards,


AMAZING!! Thank you for the link!
Old 28th December 2008
  #87
Old 7th January 2009
  #88
Lives for gear
 

I had a golden eared guest (local Linn etc. dealer) over at my house to take a listen and played him some of my 24/96 masters with SD722 and ADAM S3A monitors. Of course the sound was muffled and the notes squashed, no surprice there. Unbeknownst to him some of those files were my test files where the resolution randomly varies between the original 24/96 and RBCD. He did not notice anything.

Three possible conclusion: 1) my system really is so bad that the difference can not be heard. 2) He claims he can hear different formats easily (and power cords...) but he can not. 3) Nobody can hear it.

I think #3 is most likely, which means #2 is also true.
Old 7th January 2009
  #89
I am amazed that there are so many opinions on this idea of 44 vrs 96. To each his own I guess, but there are three things that I wanted to point out.

1. Many plug ins are optimized at 96 including many of the newer Waves plug ins.
2. It is very important to have a very good down sampling software. I know Bob Katz likes the Weiss Saracon, and the R8BrainPro though not as good is pretty decent. But if you don't have a good one, all the higher end clarity that many hear at 88/96 bascially you loose when you down sample to 44 with only average down sampling software. (This is different than dithering which is going from 24 bit to 16). SRC Comparisons
3. You have more headroom when you work with the higher sample rates. I have spoken to Hutch (formerly from Manley Labs and now works with Rupert Neve) and he even said he likes to upsample a 44 song to 96 when mastering just to have the extra dynamic range.

For me, I hear an small improvement in the high end, but of course if people listen to the music on a MP3, I don't think it would make a difference, but on a CD yes. I think though better mic placement, and better engineering on tracks make more of a difference.
Old 7th January 2009
  #90
Lives for gear
 

Headroom is about dynamic range, not bandwith.


/Peter
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
nexxoussone / So much gear, so little time
3
jax72 / Low End Theory
8
not_so_new / Music Computers
8

Forum Jump
Forum Jump