PlanetZ forums - scopeusers.com

I happened to enter a few web sites.

http://www.digitalprosound.com/Htm/Soap ... Apogee.htm

At the end of this article is a paragraph about 96k:
http://www.prorec.com/prorec/articles.n ... 88000FBE08

Are these just plain bullshit? Marketing hype?

Every statement in the Apogee post are true, and there's even more reasons to record, and especially mix, in 96kHz and above.
Only problem is off course that you need twice as much dsp power and hd space...

Kim.

Yeah and then you need to press an SACD or DVD-A for anyone else to hear it

<font size=-1>[ This Message was edited by: tonyr on 2002-12-07 16:47 ]</font>

On 2002-12-07 09:35, TonyR wrote:
Yeah and then you need tp press an SACD or DVD-A for anyone else to hear it

No you don't. That's a common misunderstanding.
No matter what the target medium is, you'll get better results with higher sample rates.
(I did notice your Tony, just clarifying things...)

Kim.

So, I recorded the last project at 44.1 with tripleDat and Pulsar II (not SFP yet)... It was a 2 CD set at 27 tracks... took up almost 20 gigs HDD... How much more space would I need to record at 96?

First you have 16->24bits:
20*1.5=30GB
Then 44.1-96kHz:
2.18*30=65.4GB !!!

Enjoy, and remember to get a good backup system before you begin

Kim.

Hi kim,
Yes I'm aware that the higher sample-rate can, with the right kit (usually expensive Apogee converters), somehow 'add' or reintegrate something to 16bit 44.1kHz recordings. Sony have a system called Super Bit Mapping which even appeared on some of their DATs for a while, I never really thought SBM added that much though (I think it was more to do with bit depth than sample-rate, but you get my drift).

Anyway I think my point is that for the outlay in terms of upgrading all your equipment yet again, buying more storage etc, it seems obvious that someone is always trying to outdate our kit quicker and quicker and make us all feel inadequate. Sometimes you have to just stop and take check. Already there is talk of 192kHz for fuck's sake. If we keep going along upgrading our equipment constantly then the various industries will happily keep increasing the sample rates until bloody dogs and bats will be sent crazy by our recordings. I'm sticking to 24bit 44.1kHz unless the technology becomes really cheap really quickly. Life's too short and I wonder if we shouldn't just get on with the music and stop tinkering. Jeez, The Beatles didn't hang around when they only had 4-track analogue shit to work with for the next big thing (thank god).

On 2002-12-07 17:04, TonyR wrote:
Jeez, The Beatles didn't hang around when they only had 4-track analogue shit to work with for the next big thing (thank god).

I wish they did

Hello there. We have talked about it too many times already, perhaps 12 or more already in this forum. It is unavoidable to believe the higher the better, but it’s not completely true. At least, not for kHz…

You can’t trust everything you read about, there is, in fact, a high marketing culture going on just to sell more expensive equipment, but the results are almost unperceivable.

BIT RATE it’s something quite different… There, the difference it’s real huge and patent, perfectly audible and indeed, it’s advisable if you can handle it. I actually am recording at 24bits and the difference to 16bits it is grand. At the beginning, I was rather against 24bits cos I though it was like wasting too much power for an almost unperceivable difference in quality. I was wrong!

Lately, I interested myself seriously into this and did a practical al theoretical study of the market as well as of my own system and possibilities, the result was to convince myself of using as high a bit rate as possible… but kHz is a completely different matter.

I have read much about what all of those famous engineers who produce the big names like Hollywood Productions, the group Pink Floyd, etc., have to say, and they clearly have said it is a commercial matter, rather than a real issue.

I’m sure that, if you have a couple of high end motors costing you $4000, you’ll be able to hear it in your own studio… yes… but who the hell is going to hear it with current systems? The real thing is that originals sounds great, but once you reproduce it, the effect is lost. I would be doing something like 32bit floating and 192kHz, if I were to produce a DVD for special FXs or something like that otherwise, it’s useless.

Conclusion for at least a couple of years to come: I’m going to keep working with 24 bit, if I can afford it, 32 floating, and 44.1, that is a great standard and it sounds already great!

imho the high sample rates are unavoidable.
Certain digital processing (like filtering) is much easier at high rates and afaik some circuitry is even cheaper to produce.
You don't have to exchange your complete signal chain to benefit from this.
That's what the guy in the review meant with the example of processing 44 khz material.
The quality on the front end is mostly influenced by the fidelity of converters and that isn't related automatically to the bit- or samplerate. I found it quite amazing that the famous Lexicon 480 has 18-bit converters (or was that a typo ?).
Reverb, compressor, eq and other fx are a significant amount of processing and will all be more detailed with less aliasing at the higher sample rates. But of course it be nonsense if the data gets stuck on it's way to the disk

cheers, Tom

I think you´re right about it, but the 480s do vary concerning the bit-depth of converters. But they had a 18-bit fixed point processer built-in, at least that´s what the manual stated. The Lex-200 had a 10-khz converter and sounded "warm" and "dense", also interesting. It must have something to do whith natural reverbs are cutted-of at 8khz anyway....


Bernhard/

<font size=-1>[ This Message was edited by: Sunshine on 2002-12-08 08:23 ]</font>

I personally gave up on the whole bit-depth and sample-rate thang.. I mean, sure, the results are definitely better, the more hertz and bits you use.. It's just a matter of informatics.. the more material you use to record something, the more accurate it gets. But heck, to me, the little details were of so little benefit, considering the amount of money that has to be spent.

So yeah, if you do have the system that can handle it, use it by all means.. Me though, since I'm working in a less than pro-quality situation, I'd still stick with lower bitrates and samperates. And in the mean while concentrate on other aspects of music making. Most of the times, a well thought out gimmick is worth much more than a ultra-hi-fi recording.

yes,and the quality of each of those "frames"(the things that happen at specific intervals or "sample rates")is part of the equation(the real-world hardware,you know, garbge i/o).also,bad things happen in the downsampling process,computer programs are not infallable ,it seems...

ultimately,the thing that people want is the music,not the recording process.the music is definitely the priority.sometimes the technical end gets extra attention when the music would be a better investment.(sometimes not )

Moving to 24bit recording isn't just a thing of changing presets; I'd have to change my hardware, too.
24bit is very fine because of the bigger headroom during recording with mics. And because of many other things. But then I could not further use my Fostex VC-8, an only 16bit ADDA-changer, which I connected with 3 expanders. If I wanted to change to 24bit, I'd have to buy a better ADDA.
Everyone has to decide himself, to make such a step and to answer the question, is such an improvement of quality really necessary or not.
Most of my works I have to record on mini discs and I believe, in this case a 24bit improvement is not necessary to me.
But a real quality freak may decide in an other direction.

It all depends on what you have to change and the way you do your music. For somebody like me, who does everything into the computer itself, there is no doubt, 24 bits it’s the way. I do not use any converter or digital gear. Everything is done with Pulsar. BTW, this is one of the reasons I’ve chosen Pulsar, even if I first was not convinced to work with 24 bits, a friend told me: “You’d better get a card 24 bits capable, or you’re going to be upgrading a couple of years time”.

If we are EXTREMELY careful when processing our audio material, and use very little well chosen and extremely few plug-ins to reach the final stage, 24 bits so it’s a must, cos the distance in quality between 16 to 24 is absolutely great and noticeable. The problem sometimes is that people keep trying too many plug-ins into audio, without thinking that this experiments are indeed, lowering the quality, despite you are working with 24 bits.

In fact, whenever I have to do a professional work, I try to use as few plug-ins as possible, so I can reach the maximum. What I do? I think well and try out different possibilities, to reach the same with less plug-ins in a shorter way. If you keep it thin and simple, 24 bits IS a great improvement in your final work.

Nestor, if you use Pulsar synths only, why don't you work with 32bits?

About sample rates, at least 96khz gives smaller latency (Why's that btw, can somebody explain?)

And a question, is it possible to mix different sample rates and bit depths in one project? Sure it is possible, but will there be any problems?
We have recorded drums with only 16b@44k and I though would it be better to record vocals using more bits (and maybe more samples too)

the latency thing is very simple:
sound 'happens' twice as fast at 96 khz
The actual 'unit' of latency is the number of samples in a device buffer. If those samples are played back at a higher rate, then less time is needed.
Imho you should convert your various data rate and bit deepth sources to one common format, either up- or downsampled, whichever fits your needs best.
Otherwise it might get very confusing and in some cases it's even impossible to mix up different formats (see above). Keep the original files at a safe location and you'll have a backup if something runs terribly wrong.

cheers, Tom

Spoimala, as far as I know, you can’t work 32 bits from Pulsar, Pulsar only reaches 24 bits. How could I work 32 bits? I don’t think it’s possible otherwise than recording audio to a program supporting 32 bits. Pulsar only supports 24. Anyway, my wave editor only does 24 bits for the moment.

Yes, you can record at different bit depth different parts, as long as your sequencer supports it. I’m not so sure about sample rate, there I don’t know.

Nestor, Pulsar's internal processing is 32 bit and with VDAT you may record this at up to 96 khz samplerate. ....wanna buy new harddisk, dude ?
That 32-bit is really appealing. It's symetric. 2x16 is more aestethic than 16+8.
And certainly better to transmit.

cheers, Tom

VDAT support #@bits, Cubase SX support 32bits, Pulsar drivers also, look at the Asio2-64 modules... They are modules that support up to 64 channels of 32bits audio streams...

Recording in 32bits put less strain on cpu, cuz no padding of data is needed (pci bus for example is 32bits, not 24bits!) for transport, but it take huge amount of disk space!

There is also one other thing to think about... If you record from analog input of pulsar (or a16, a16u & luna io box, and even adat) they all have 24bits transport, or ad/da... So, there is actually no use for any of this...

If you are recording a samplers track, if your programs are not 32bits, no use either.

If you record pulsar synth, since they are intern to Pulsar, and Pulsar is 32bits all the way, you can benefit 'a bit' in adding some 'bits' , but i don't know to what extent...

Also, like you said Nestor, i don't know of any wav editor that edit 32bits waves now, except in SX or samplitude. But those can hardly be called 'wave editor'... A bit heavy just for editing small (maybe not that small if recorded in 32bits) wavs...