I guess the somewhat less technical explanation of the point I was making is that the 4th character isn't going to change in the way you might have been thinking because the translation between fixed and floating point is not straightforward. 

Yeah, I know... I got your point

John
I think Goddard is confused about 1 bit audio recording and the rest of the audio recorders. 
 
This may clear up the confusion http://www.bhphotovideo.c...bit-better-24-bit.html
 
This also may help. http://en.wikipedia.org/wiki/Direct_Stream_Digital

You think I'm confused? Hah, that's rich.
 
Ar least you're getting warmer with your DSD/SACD links, although you still appear to be unaware of how oversampling multi-bit DSM ADCs and DACs (as equpped in almost all audio interfaces for many years now) operate, or of the effects/advantages of oversampling wrt the Nyquist frequency and filtering requirements/effectiveness (see previously linked skywired.net blog post).
 
FYI, there is a very simple reason (of which I strongly suspect that facetious scientist/music technology educator is completely unaware) why the lowly onboard sound chip in every PC/Mac for many years now (at least since Vista and Intel Macs) is capable of competently performing 24-bit (yeah, I wrote "24-bit" there, not "1-bit") @192kHz multi-channel recording and playback (and doing so simultaneously, i.e., at full duplex), and that reason comes down to Intel's "High Definition Audio" (HDA) specification (successor to the earlier "AC97") along with MS' stringent (as in, requiring use of an AP audio test rig) Windows logo cerification testing requirements, with which all those lowly little cheapo onboard sound chips must conform and comply before the millions of systems into which they are equpped ever reach market.
 
I suspect that you don't bother actually reading any of the material I've posted links for, as you've raised no objection or question in relation to anything I've linked-to, despite your earlier insistence that I cite autority for my critique of that facetious science blog, but I'll post some more links anyway just in case anyone else might genuinely be interested in this stuff:
 
http://www.intel.com/content/www/us/en/standards/high-definition-audio-specification.html
 
(I'll leave it for the curious to learn the reason for inclusion of 24-bit/192kHz in the HDA requirements)
 
http://msdn.microsoft.com...8866%28v=vs.85%29.aspx
 
http://msdn.microsoft.com/en-us/library/ff563343%28v=vs.85%29.aspx
 
http://msdn.microsoft.com/en-us/library/ff563349%28v=vs.85%29.aspx
 
http://www.realtek.com.tw/images/products/High_Definition_Audio_Codec_Selection_Guide_07182008.jpg
 
And of course, fwiw Sonar has supported 24/192 (and even 24/384) samples for some time now as well (since Sonar 4 iirc).
 
But, instead of focusing only on lowly onboard audio chips whose specs and performance are dictated by Intel and MS, let's see if I can't make more apparent the actual situation wrt more serious audio production gear which some folks hereabouts might likely be using (and let them judge for themselves whether the "problems" with 192k sampling raised in that facetious "Science of..." blog are a reality or merely non-existent pseudo-science baloney) by giving a real world example of a generally well-regarded 24-bit @192kHz-capable "audio interface". Say, one which was touted as featuring "mastering grade" converters just like those on the PTHD192 rigs used by real "pros", namely the E-MU 1212m:
 
http://www.creative.com/emu/products/product.aspx?pid=19169
 
on which, per the specs given, the following ADC and DAC chips are respectively employed:
 
http://www.akm.com/akm/en/product/datasheet1/?partno=AK5394AVS
 
(notice that AKM list "128x Oversampling" and "multi bit Architecture ADC")
 
http://www.cirrus.com/en/products/cs4398.html#
 
(note the "oversampled multibit Delta-Sigma modulator")
 
Now, just to avoid there arising any confusion, in case of any doubt as to how such oversampling multi-bit DSM converters actually operate when converting to and from 24-bit 192kHz samples, simply follow the links on the above pages to access the data sheets. And iirc, the Cirrus Logic/Crystal Semi developers published at the time about their clever stuff, so anyone interested in the real esoterica might search that out (possibly behind the IEEE's paywall though).
 
Not that any of this oversampling stuff is really all that new or even magical, even if it can be difficult to grasp at first. Drewfx alluded to it very briefly in a post early on this thread, although you may not have grasped what he was saying. Here's an article from the early DAW days ProRec webzine in which the principle of exchanging sample rate for bit-depth was given treatment (in the very last section):
 
http://web.archive.org/we...FA5EAA862566B20022F4CA
 
(sigh... I do sorely miss having Jose on this forum, those were some intelligent discussions, and CW benefitted too)
 
So, no, I'm not confused, John, because (unlike you and JohnT) I've actually known about and understood sampling and oversampling (and even floating-point math too) for a very long time, and very well know the difference between 1-bit DSD and 16/24-bit PCM (and even how to convert between the two).
 
And just to be clear, absolutely nobody has to take my word for anything which I've asserted, as I've posted links to independent references which explain and/or corroborate any technical points asserted. That is, if anyone ever actually reads what I've linked-to rather than just carrying on with hiking their post count in oblivious ignorance.
 
So by all means please don't take my word about oversampling DSM converters actually sampling in the MHz range, when you can instead read Dan Lavry's very own words saying so (to make it easier for those too lazy to follow a link, I'll just copy-in an excerpt of the relevant bit):
 
 

Dan Lavry
What rate and bits are enough for today music reproduction and recording?
 
Regarding the rate:

One has to make a distinction between the audio sample rate and the rate of a localized process:

The audio sample rate is the rate that carries the music data itself. Roughly speaking, the audio bandwidth itself is slightly less then half the sample rate. A 44.1KHz CD can contains music to about 20KHz.

At the same time, there are many cases when we use much higher “localized rates”. Such higher rates do not increase the musical content. The higher rates still offer the same original bandwidth of the sample rate. We up sample or down sample between localized rates for various technical reasons. For example, virtually all modern DA’s operate at 64-1024 times the sample rate speeds (in the many MHz range). Operating at such high rates simplifies the requirements of the anti imaging filter (an analog filter located after the DA conversion). The decision about the ideal localized rate depends on the technology and the task at hand. It is an engineering decision, not an ear based decision. As always a poor implementation may introduce Sonics, and it would be wise to refrain from the often encountered practice of far reaching false generalizations, so common in the audio community.

http://www.monoandstereo.com/2008/06/interview-with-dan-lavry-of-lavry.html
 
So, now, who exactly is confused (or uninformed/just plain ignorant) here?
 

John is very correct in saying that only 8 bits is actually required above the final playback bit depth, meaning 24 bit is all you really (ever) need. If you work with the K system you are never clipping anyway so 32 bit depths are also unnecessary.
 
Bob Katz has also said a long time ago that 50- 60 Khz is about all we need in terms of sampling rate.
 
I have read some very intersting articles too about how some hardware sounds. eg A-D and D -A converters. High sampling rates from a certain piece of hardware do not actually guarantee it will sound better. Some converters sound better at 44.1K than they do at 96K and some sound better at 96K than they do at 44.1K. When two converters are sounding good (at either sampling rate) then the differences are very very minor if at all audible. It is very hard apparently to get optimum performance at all sampling rates from the same piece of hardware. If you are determined to work at higher sampling rates eg 96K you need to do your research to actually find if the bit of gear you want to use to do it actually sounds good doing it.
 
Real World now. I have created an AB test session where a very high quality analog signal (finest turntable, vinyl,pickup, RIAA equaliser etc) was sent to one side of an AB switch. That same signal was bottle necked through 16 bit 44.1K A to D and D to A and fed to the other side of the switch. Even with expert engineers and very high quality monitoring (and environment) many had no idea what they were listening to. I did this based on this article:
 
http://mixonline.com/reco...emperors_new_sampling/
 
I wonder how well people who are getting all head up over this can mix. Probably not that great I would say. Reason I say this is because great mix engineers are usually not concerned with any of it. Remember fantastic mix at 44.1k 16 bit will do it everytime over a lousy mix at 96K 24 bit. Isn't that what is important.

John
It bothers me that I and others have been accused of giving out misinformation. This is something I have been very much fighting against ever since i have been on this forum. I have been wrong in the past. Not often, however.  When I am wrong I will broadcast that fact and try my best to correct the error. 
 
Goddard has accused me and just about all that have participated in this thread of being wrong. In fact it is he that is totally wrong. he has somehow confused 1 bit recording with 20/24 bit PCM recording. They are very different things and are interesting as a study in their differences but the technologies are very different. The one has no reason for being interjected in this thread. 
 
 

Well, it bothers me when people with "platimum" post counts who've been around this forum for years continue racking up their count by posting ignorant and uninformed drivel, when they really should know better or at least have learned some accurate info by now. Just like it bothers me when people get taken in by the baloney a facetious/pseudo scientist blogger posts (and especially when CW's CTO links to it!).
 
Are you still unable to grasp floating-point math, John? I mean, I can see from this old forum thread
 
http://forum.cakewalk.com/A-question-about-the-64bit-engine-m1724023.aspx
 
that you were uninformed 5 years ago, although I can see that being chastised then didn't matter or prevent you from posting more misinformation then so why should now be any different...
 
No need to thank me for taking time to explain 32-bit fp math to you-- even if you still can't grasp it hopefully someone else seeking knowledge may benefit.
 
Yeah, it can be so confusing, where those bits go, or how it is that a 32-bit DAW application can perform 64-bit floating-point math.
 
How does that old adage go? Better to keep silent and let folks guess whether one is a fool , than to open one's trap and dispel any doubts?
 
Perhaps a modern equivalent might be along the lines of better to lurk and learn than let fly with the keyboard. But then, can't work up a platinum count with years' worth of uninformed misinfo that way.
 
Anyway, folks are free to judge who is confused or misinformed or a fool.
 

Why is it always sample rate or bit threads. I swear it wasn't like this in my maths class.

John
OK before I let myself get caught up in putting some one down let it end here and now. Often threads like this can and often do result in conflict. That is not the way we should let things happen any more. Goddard was only posting what he thought was true. There is no sin in that. 
 
I like it when some one posts a correction when they believe it is needed. We need to let members feel this place will not jump on them just because of a disagreement. 
 
We need to keep this place free of intimidation or make people feel uncomfortable no matter what is posted.   

Oh, too late for that now.
 
That you or anyone else is unable to discern that I might actually know of what I write is not my fault, nor is it my responsibility to educate you/them.
 
Post misinfo if you must (or don't know any better). But be forewarned that other folks more knowledgeable than you may well undertake to point out the errors therein so that others seeking knowledge here might hopefully not become as misinformed (or remain as ignorant) as you clearly appear to be.
 
The level of knowledge and expertise around here has truly and sadly declined. Was a time when a lot more knowledgeable folks participated here and on the forerunner newsgroup which at one time was on the cutting edge in many ways concerning DAWs and digital audio. Maybe they just tired of the ignorance, er, noise level. Let's see if we can't do something about that.

CakeAlexS
Why is it always sample rate or bit threads. I swear it wasn't like this in my maths class.

Dunno. Maybe it has something to do with misinformed facetious science blog postings?
 
Btw, had inadvertently unblocked you, so thanks for not polluting this thread with lots of image postings too...

bitflipper
Gee whilickers, it's been awhile since we've had a heated multi-page technical discussion! And one with a pretty decent signal-to-noise ratio, too. (Look up similar threads on Gearslutz to see just how uninformed and rude such conversations can get.)

 
Oh, sometimes the threads on GS aren't really all that uninformed (or even rude) at all...
 
http://www.gearslutz.com/board/geekslutz-forum/771247-oversampling-what.html
 

bitflipper 
If nothing else, this has prompted folks to seek additional self-education on the subject. Way to go, CW forum.

Have a good journey on your quest for self-enlightenment!

Jeff Evans
John is very correct in saying that only 8 bits is actually required above the final playback bit depth, meaning 24 bit is all you really (ever) need. If you work with the K system you are never clipping anyway so 32 bit depths are also unnecessary.

 
Except that perhaps John might not really grasp bit-depth, or how floating-point math works inside a digital audio mixer, and if so, people following his recommendation do so at their own peril?
 

Jeff Evans
Bob Katz has also said a long time ago that 50- 60 Khz is about all we need in terms of sampling rate.

 
Well, here's something Bob Katz once said on the subject of 192k sampling, in which he admitted that he really hadn't had sufficient experience at 192k to evaluate it:
 

Bob Katz
MI: How are the 192 khz recordings? Does so much information bring us closer to the original recordings and are there more problems due to the increase of content?

BK: I have not had enough experience with 192 kHz to say. I like the results I'm getting at 96 K, and in my book I make a convincing argument that it is the converter design that counts far more than the sample rate. We have always known that a well-designed 44.1 kHz converter sounds much better than a mediocre 96 kHz model. And this has always been true. I believe that a good designer will be able to make a 96K converter that sounds as good as anything at a higher rate. But designers are getting lazy, and it is cheaper and easier to get a good sound at a higher rate because the filters are less complex and easier to design. There is nothing magic about the higher rates; it's not the higher frequencies that we're hearing, but rather, more linear performance from 20-20 kHz! Keep that in mind... We really should be labeling converters by their resolution, not by their sample rate.

http://www.monoandstereo.com/2008/02/nterview-with-bob-katz.html
 
Designers are getting lazy? Where does Bob think that "more linear performance from 20-20k" comes from? Maybe Bob should try designing his own converters with steep analog aa filters sometime...