Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Forum Jump  
Options
Go to last post
3 Pages<123>
Posted on Wed, Oct 30 2019 14:07
by Gabriel Plalame
Joined on Thu, Feb 27 2003, Frogs eater country., Posts 153

Hi all,

About space, I agree that it is no more a problem today (including M2).

About latency, I need to study it deeper…

But let’s talk about sound quality: I did many tests about 44.1 vs 96 kHz with synth sources, and I clearly make a difference (not really with 192kHz, I’m not totally sure).

But it was (reproducible) synth sounds.

I’m not engineer so, three questions here about samples :

Does a mic source will make a perceptible difference between 96 and 44.1kHz ?

Is 96kHz better than 44.1kHz for mixing voices ? (Considering that a sample player is also a mixer.)

And what about processing ? Is is better to process an effect in 96 than 44.1kHz ?

If yes, a 96kHz library would be welcome anyway.

 

Regards

Gabriel Plalame

The French dyslexic who speaks badly English.

https://soundcloud.com/gabriel_plalame

Asus Prime Z270-K | Intel I5-7600 3,5GHz | Gskill DDR4 | Samsung SSD 960 EVO 500Go M2 2280 NVME x2 + 3 other SSDs | RME HDSP 9652 | Windows 10 pro x64 | Synchron Pianos 1.1.1413 | Samplitude X4 / X5 Suite...
Posted on Wed, Oct 30 2019 16:33
by TFIS
Joined on Tue, Dec 14 2010, Posts 106

This might answer all your questions

https://people.xiph.org/~xiphmont/demo/neil-young.html

Posted on Thu, Oct 31 2019 00:02
by mantrum
Joined on Sun, Apr 16 2017, Posts 15

It would also be interesting to consider the latency of pressing a piano key down, and the time it takes for the action of the hammer to hit the string, and compare it to this latency in using a digital keybed and computer.

I wonder how the two values would differ.....

Posted on Thu, Oct 31 2019 05:42
by littlewierdo
Joined on Sun, Apr 24 2016, Posts 236

Originally Posted by: dmidi Go to Quoted Post

Originally Posted by: littlewierdo Go to Quoted Post

Not to mention the memory footprint. A full orchestral piece requires 60 plus gigabytes of ram for me, and that is really nothing in comparison with what it could be. There is a reason why 44.1 at a 16 bit depth rate was chosen for CD audio, this is considered just above the maximum quality that anyone can hear a difference. Anything above this is really unecessary, unless you are time stretching samples (think slow motion video, the more frames you have, the smoother it will be). When it comes to orchestral, I dont think time stretching samples is something many people are apt to do.

1. No disrespect, but have you read the thread?  This is about LIVE instrument playability quality, not RECORDED/MIDI music file playback sound quality or time stretching. 

To summarize, there is a negative effect of latency on pianist's sense of hearing and motor cortex processing as impacted by low sampling frequency.  That is, if you press a key and you expect to hear the sound in 3 Milliseconds, but the sound instead comes to you in 6 Milliseconds, that will be as if the sound board has moved from 3 feet away to 6 feet away.  A bit wierd to say the least.  And that assumes no other latency elsewhere in the sound processing chain.  As there always is this latency in every component, the cumulative effect of latency might mean that the piano sounds as if the sound board is 10 feet or more away.  And that's before we start to consider the effect of jitter, which can be described as the sound board constantly shifting back and forth several feet, with this perceived movement happening from note to note or even during the resonance of a single note.  Aye...very bad indeed for your brain to be 'hearing' the grand piano vibrating back and forth.

Also note that as this jitter is likely random, good luck trying to achieve consistent playback between pieces. And that is true even for playback of MIDI notes in which instruments are closely placed.

There is a way to elilminate the 'sound shifting':  By roughly doubling the frequency from 44 to 96 Khz, the latency of the VSL library is cut in half, and that allows for a more consistent and authentic playing experience. And by doubling again from 96 Khz to 192 Khz the latency is halved again and we have more 'latency headroom' to stay under the audibility threshold.

Don't take my word for it, but do take VSL's deafening silience on this topic as being problematic.  I suggest if my explanations are not sufficient, please research and understand other trusted sources.

2. As to RAM requirements, this piano at 24 bit 44.1 Khz currently has a RAM requirement of 4-6 GB.  Doubling the sampling frequency will half the latency and only require a total of 6 GB * (96/44.1) = 13 GB.  Most laptops with any sort of multimedia claims will have 16 GB RAM or more.

To editorialize a bit, VSL is doing an enormous dis-service to themselves and their customers by attempting to ignore the issue.  Unfortunately, an uneducated customer at some point becomes a frustrated customer...and that's a net negative for musical creativity and classical music as a whole (LONG LIVE POP MUSIC?).

I have, and its chalk full of misinformation and jumps back and forth between audio quality and latency, I was speaking to the audio quality side of the discussion (my response back to you, did YOU read the thread?).

Latency is a huge, convoluted mess of a problem that cannot be nailed down to just one thing, as some have tried to do in this thread. Audio interfaces do not "perform" better with certain resolutions over others, and yes, I confirmed this with a friend of mine who designs audio interface circuitry for ASUS.

Also conveniently missing from this conversation is the significant increase in processing power (and thus delay) required to play back audio with effects at higher sample rates. RAM access times and storage to RAM access times are another big factor, and then, probably the biggest elephant in the room not even discussed is, where the majority of the latency comes from, the MIDI / USB / Firewire interface.

I work with sounds from all sorts of sources, ranging from Orchestral Tools crazy 24 bit samples to, by comparison, EastWests lighter sounds and latency is of ZERO concern to me. It is within expected tolerances of 15-20 MS, all of which is easily compensated for by Kontakt, Play, and Vienna's engines, which have settings that work very well for this. Very rarely do I have to go back and correct timing, unless I am working with something that needs to be REALLY tight (big percussion sections or many staccato notes that need to be really tight).

Finally, most DAWs, on the playback side of things, have a track delay setting that virtually eliminates this problem. My guess is, you havent set a predelay setting in your DAW, look into it, this conversation becomes a moot point once you discover this setting. Then you run into a different problem, calculating the predelay, or determining what the predelay setting should be, because it is different for each manufacturer of soundfonts, and can even be different from library to library (the general default numbers to set in your DAW are 30, 60, or 120 ms). From there, you can add additional time depending on your standard latency, so I usually add about 15 ms and everything I record and playback is spot on the beat.

Posted on Fri, Nov 01 2019 14:20
by robinlb1900
Joined on Sat, Jun 22 2019, Posts 21
In fact, hope VSL can put up a section of audio recorded with 96K audio source for comparison.
I think high quality DAC is also helpful for playing high rate audio source.
Posted on Mon, Nov 04 2019 08:12
by Gabriel Plalame
Joined on Thu, Feb 27 2003, Frogs eater country., Posts 153

Originally Posted by: TFIS Go to Quoted Post

This might answer all your questions

https://people.xiph.org/~xiphmont/demo/neil-young.html

Hi, TFIS, thanks.

Very good link, the ultrasonic part is enlightening.

If I understand well, high resolution is good for mixing, not delivering. Well, can we imagine a more polyphonal instrument than a piano ? Specially a so powerful multi mics one like Syncron pianos… I very easily get 600 voices with my every day preset with only 3 mics…

Plus, if the VSL team chose the 96kHz resolution to record… it can’t be just for fun…

 

Regards

Gabriel Plalame

The French dyslexic who speaks badly English.

https://soundcloud.com/gabriel_plalame

Asus Prime Z270-K | Intel I5-7600 3,5GHz | Gskill DDR4 | Samsung SSD 960 EVO 500Go M2 2280 NVME x2 + 3 other SSDs | RME HDSP 9652 | Windows 10 pro x64 | Synchron Pianos 1.1.1413 | Samplitude X4 / X5 Suite...
Posted on Tue, Nov 05 2019 06:07
by robinlb1900
Joined on Sat, Jun 22 2019, Posts 21
Originally Posted by: Crystal Go to Quoted Post

Plus, if the VSL team chose the 96kHz resolution to record...it can be just for fun.

Regards
Gabriel Plalame


From this thread, wasn't synchron piano recorded in 96K or higher?
Only release version is 44.1K/24bit.
Posted on Tue, Nov 05 2019 07:41
by Gabriel Plalame
Joined on Thu, Feb 27 2003, Frogs eater country., Posts 153

Originally Posted by: robinlb1900 Go to Quoted Post
From this thread, wasn't synchron piano recorded in 96K or higher? Only release version is 44.1K/24bit.

Yes, that's what I mean. "Vienna Instruments are recorded in 96 kHz" (Dietz)

The question is : why ?

If it's better to record in 96 kHz, why reduce it to 44,1 ?

Or, to say it differently : if 44,1 is enough, why recording in 96 ?

Again, SSDs size is not an argument any-more today.

 

Gabriel Plalame

The French dyslexic who speaks badly English.

https://soundcloud.com/gabriel_plalame

Asus Prime Z270-K | Intel I5-7600 3,5GHz | Gskill DDR4 | Samsung SSD 960 EVO 500Go M2 2280 NVME x2 + 3 other SSDs | RME HDSP 9652 | Windows 10 pro x64 | Synchron Pianos 1.1.1413 | Samplitude X4 / X5 Suite...
Posted on Tue, Nov 05 2019 08:28
by robinlb1900
Joined on Sat, Jun 22 2019, Posts 21
Originally Posted by: Crystal Go to Quoted Post
Originally Posted by: robinlb1900 Go to Quoted Post
From this thread, wasn't synchron piano recorded in 96K or higher? Only release version is 44.1K/24bit.

Yes, that's what I mean. "Vienna Instruments are recorded in 96 kHz" (Dietz)
The question is : why ?
If it's better to record in 96 kHz, why reduce it to 44,1 ?
Or, to say it differently : if 44,1 is enough, why recording in 96 ?
Again, SSDs size is not an argument any-more today.

Gabriel Plalame


For comparation, concert grand from production voices provides 44.1/16bit, 44.1/24bit files and will be 96/24bit.
They said different rate for different usage and hardware configuration. Lower rate means wider adaptability and compatibility.
I can understand why VSL use 44.1/24 for products distribution at present.
Actually, I remember synchron piano is recorded in 192/32bit originally.
That's why I hope VSL team can provide high rates format for users to choose in the future.
Posted on Wed, Nov 06 2019 11:28
by Gabriel Plalame
Joined on Thu, Feb 27 2003, Frogs eater country., Posts 153

I just find this from a french site (https://deveniringeson.com/frequence-echantillonnage/) :

“(...)les pilotes de votre carte son (surtout les convertisseurs professionnel) seront optimisés pour une fréquence d’échantillonnage donnée. En général, les pilotes ASIO de vos convertisseurs sont optimisés pour la fréquence d’échantillonnage maximum qu’elle propose : 96 000 Hz et 192 000 Hz dans la majorité des cas. Ca peut être surprenant, mais vous aurez moins de latence et plus de soulagement pour le micro-processeur avec une fréquence d’échantillonnage plus grande.”

Reverso translation :

“drivers on your sound card (especially professional converters) will be optimized for a given sampling frequency. In general, the ASIO drivers of your converters are optimized for the maximum sampling frequency it offers: 96,000 Hz and 192,000 Hz in the majority of cases. This may be surprising, but you will have less latency and more relief for the microprocessor with a higher sampling rate. »

It could be an element for the discution about latency.

 

Gabiel Plalame

The French dyslexic who speaks badly English.

https://soundcloud.com/gabriel_plalame

Asus Prime Z270-K | Intel I5-7600 3,5GHz | Gskill DDR4 | Samsung SSD 960 EVO 500Go M2 2280 NVME x2 + 3 other SSDs | RME HDSP 9652 | Windows 10 pro x64 | Synchron Pianos 1.1.1413 | Samplitude X4 / X5 Suite...
Posted on Sun, Nov 10 2019 00:54
by littlewierdo
Joined on Sun, Apr 24 2016, Posts 236

Originally Posted by: Crystal Go to Quoted Post

I just find this from a french site (https://deveniringeson.com/frequence-echantillonnage/) :

“(...)les pilotes de votre carte son (surtout les convertisseurs professionnel) seront optimisés pour une fréquence d’échantillonnage donnée. En général, les pilotes ASIO de vos convertisseurs sont optimisés pour la fréquence d’échantillonnage maximum qu’elle propose : 96 000 Hz et 192 000 Hz dans la majorité des cas. Ca peut être surprenant, mais vous aurez moins de latence et plus de soulagement pour le micro-processeur avec une fréquence d’échantillonnage plus grande.”

Reverso translation :

“drivers on your sound card (especially professional converters) will be optimized for a given sampling frequency. In general, the ASIO drivers of your converters are optimized for the maximum sampling frequency it offers: 96,000 Hz and 192,000 Hz in the majority of cases. This may be surprising, but you will have less latency and more relief for the microprocessor with a higher sampling rate. »

It could be an element for the discution about latency.

 

Gabiel Plalame

Right, but we arent just talking about playback, we are also talking about adding FX. The reverbs and other FX we add to our sounds becomes more complex (thus, significantly higher processor usage) with higher sample rates. Its easy enough to test, if you have a library that contains multiple sample rate options, both with and without FX.

Posted on Mon, Nov 11 2019 12:53
by Gabriel Plalame
Joined on Thu, Feb 27 2003, Frogs eater country., Posts 153

Hi, littelwierdo,

I agree that a 96kHz project needs more resources in general, of course, but my quote from the french site really surprise me, and is not totally clear about this : does the Asio optimization for the biggest sample rate concern only the A/D conversion (audio capture) or also the D/A restitution (playback) ? If the playback is also affected, then, logically, VSTIs are or could be affected. Plus, is it the case for all sound cards and all manufacturers ? And how to test this before buying ?

We also have to keep in mind that, as it was said, when the library and project sample-rates don’t match, the real-time re-sampling charge is a problem, specially for a piano library and its outrageous and ferocious polyphonic appetite.

Considering that some engineers make an audible difference between 44.1 and 96kHz at live recording sessions (file by file and/or for the mix ?),

Considering that (in my experience) VSTIs synth software really sound better in a 96kHz project than 44,1kHz, and that when we use that magnificent VSL library, many of us also add synth softwares for sound tracks,

And considering there is no more hardware limitations for 96kHz projects today with a decent computer, in any case not for such a library like Syncron pianos,

So, to use a 96kHz library in a 96kHz project may be the best sound quality/resources compromise for reasonable soundtrack project.

Plus, the big advantage of Syncron pianos is to give us a wonderful natural concert venue for zero CPU charge...

The jitters effects that dmidi mention, and that I didn’t know, has to be taken in consideration too.

And considering that, as Dietz mention, it is recorded in 96kHz, if the VSL team sells a 96kHz upgrade of the Steinway D, I think I’d buy it. They should study the market with current customers.

I’m very curious about the sound quality gain (a really more controversy topic, I think, not sure to launch it here !..)

 

Regards,

Gabriel Plalame

The French dyslexic who speaks badly English.

https://soundcloud.com/gabriel_plalame

Asus Prime Z270-K | Intel I5-7600 3,5GHz | Gskill DDR4 | Samsung SSD 960 EVO 500Go M2 2280 NVME x2 + 3 other SSDs | RME HDSP 9652 | Windows 10 pro x64 | Synchron Pianos 1.1.1413 | Samplitude X4 / X5 Suite...
Posted on Tue, Nov 12 2019 01:15
by robinlb1900
Joined on Sat, Jun 22 2019, Posts 21
Originally Posted by: Crystal Go to Quoted Post

if the VSL team sells a 96kHz upgrade of the Steinway D, I think I’d buy it. They should study the market with current customers.


Me too.
Releasing different rates libraries including 96K to be choosen by users for different usage is smart solution.
As I said above, ProductionVoices has already done so, even though their 96K hasnt been sold yet.
Posted on Tue, Nov 12 2019 09:35
by TFIS
Joined on Tue, Dec 14 2010, Posts 106

Originally Posted by: robinlb1900 Go to Quoted Post
Originally Posted by: Crystal Go to Quoted Post
if the VSL team sells a 96kHz upgrade of the Steinway D, I think I’d buy it. They should study the market with current customers.
Me too. Releasing different rates libraries including 96K to be choosen by users for different usage is smart solution. As I said above, ProductionVoices has already done so, even though their 96K hasnt been sold yet.

 

Maybe before that, the other keys should be sampled, which haven't been.

Posted on Tue, Nov 12 2019 10:48
by PaoloT
Joined on Tue, Dec 27 2016, Posts 751

I can't understand how people can bear 96kHz nowadays. I can't even listen to things under 192kHz. I can feel all that aliasing! So, VSL, please, before going for low numbers, plan to offer us 192kHz samples, multi-channel and sampled in surround!

Paolo

Posted on Tue, Nov 12 2019 11:39
by Gabriel Plalame
Joined on Thu, Feb 27 2003, Frogs eater country., Posts 153

(When I said it was controversy…)

Paolo,

They already record in 96kHz.

Why ?

 

Gabriel Plalame

The French dyslexic who speaks badly English.

https://soundcloud.com/gabriel_plalame

Asus Prime Z270-K | Intel I5-7600 3,5GHz | Gskill DDR4 | Samsung SSD 960 EVO 500Go M2 2280 NVME x2 + 3 other SSDs | RME HDSP 9652 | Windows 10 pro x64 | Synchron Pianos 1.1.1413 | Samplitude X4 / X5 Suite...
Posted on Tue, Nov 12 2019 14:54
by robinlb1900
Joined on Sat, Jun 22 2019, Posts 21
Originally Posted by: Crystal Go to Quoted Post
(When I said it was controversy&hellip;)Paolo,They already record in 96kHz.Why ?

Gabriel Plalame


In this thread, Dietz said "Vienna Instruments are recorded in 96 kHz with 24 bit resolution", that doesn't mean Synchron Pianos are also recorded in 96kHz. I remember they are recorded in 192/32 format.

Of course, reduced to 96/24 is enough for me :)
Posted on Tue, Nov 12 2019 15:01
by robinlb1900
Joined on Sat, Jun 22 2019, Posts 21
Originally Posted by: TFIS Go to Quoted Post


Maybe before that, the other keys should be sampled, which haven't been.


Maybe for ProductionVoices, but for VSL it's already recorded originally I guess.
Posted on Tue, Nov 12 2019 15:16
by TFIS
Joined on Tue, Dec 14 2010, Posts 106

Originally Posted by: robinlb1900 Go to Quoted Post
Originally Posted by: TFIS Go to Quoted Post
Maybe before that, the other keys should be sampled, which haven't been.
Maybe for ProductionVoices, but for VSL it's already recorded originally I guess.

Lower and higher keys have been sampled in whole tone steps, not chromatically.

Posted on Wed, Nov 13 2019 00:21
by littlewierdo
Joined on Sun, Apr 24 2016, Posts 236

My concern is, resources are still an issue for me. I write and compose using a single PC, an Intel core i7 8700k with 64 GB of ram. A full orchestral piece consisting of about 60 tracks in which I use every instrument that is loaded, will eat every bit of that ram up in a heartbeat. I cannot upgrade my RAM any more than that, 64 is my motherboard limit, so my next option would be to start freezing tracks or converting tracks to audio, which would severely hamper my creativity (waiting minutes to unfreeze and reload a track is a creativity killer).

Factor in MIR Pro and perhaps a light compression / delay VST (I use the FX builtin to Ensemble 7 as the FX are quite good), and the video editing software I use (Magix Vegas Studio) which I like to have open with the music writing process so that I can write music and time / render it correctly in time with the video and things start to go quite nuts on resource usage. While I know there are options for reducing resource usage, and I certainly could do what I do without having everything loaded all at once, It would slow down the final bit of the process for me, when I am fine tuning the work I do.

It is also worth noting, probably 70% of what I work with is already what I would consider a light library (Special Edition libraries - 1,1+,2,2+,3,4,5,6,&7). I shudder to think how bad it would be if I were using the complete libraries. I mean, in all honestly, with exception to the solo instruments, luckily, I dont see a big enough value in the complete libraries (for me) to use them or I would have to consider replacing my motherboard and doubling my ram, a prospect that would run me about $1000.

As much as I absolutely love Orchestral Tools libraries and some of Spitfire libraries, the main reason I dont use them is the size. They are unbelievably ginormous and unnecessarily so.

I also, as a side note, do not want to introduce complexity in my workflow. Running Ensemble over a network, while neat, adds unecessary complexity that I just dont need or want. As the famous line goes, the more they overtake the plumbing, the easier it is to stop up the drain.

Synchron pianos is already one of the biggest piano libraries out there, surpassed only by another piano library that touts a terabyte and a half as I recall. I am glad that VSL decided to release the special edition synchron variant, as it serves well enough to my ear to use for orchestral work, and hell, I think the special edition variant is STILL larger in size than most piano libraries on the market.

Finally, I dont think it is beneficial to anyone to suggest that performance is better with higher sample rates. What we are doing is vastly different than what many articles are writing about, which is simple playback. We are adding FX and we are not merely simply playing wav audio. The audio format is in a proprietary format, not native to what soundcards are acustomed to playing, which adds overhead, increasing processor usage.

All this having been said, I am not against having an optional download for those that want higher quality, although, if this option is provided, I dont understand why they should need to sell it as a "give us more money" choice, because they already have the samples in this format. It would not or should not require much more effort on their end to make this option available, and it should be a free choice, as it is with every other library on the market that offers quality choices (Sonokinetic for example, it is a free optional download).

Personally, no offense meant, but I find it highly suspicious that anyone can tell the difference between 48 and 96 sample rates. Now, between 16 and 24 bit, in very specific circumstances, that I can tell the difference, but the circumstances where I can tell the difference are very specific, having to do with dithering with certain frequencies and lower volume levels. However, who am I to get into a debate over whether someone can hear a difference?

Anyway, I have Physics homework to get back to, make the option available free of charge, choice is never a bad thing, I merely wanted to address the notion that performance was better with higher sample rates.

3 Pages<123>
You cannot post new threads in this forum.
You cannot reply to threads in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.