CEP sound???

[T. Alan Smith]

Greetings,

I'm upgrading from my Roland VS studio and was just wondering how CEP sounds in comparison to Logic, or Cubase.

Thanks,

T. Alan

[carparker]

well, just an opinion this, not a statement of fact, but i have always thought that original wavs in CEP sound better than they do in logic and cubase, fuller, warmer... hmmm wishfull thinking maybe?? Anyway the point is, till CEP2 comes out logic and cubase can do more to those wavs instantly than CEP can now. But I believe CEP's effects are very fine quality, and I think alot of people here have used and loved CEP despite the attractions of those other programs.

Comparisons between audio/midi sequencers and cep are a little unfair in anycase because cep is an audio multitrack recorder/editor only, but as your queston was one of 'sound', in terms of how CEP sounds it does just fine against those 'big two' programs. But Samplitude, better than all of them imho....

Still have a big soft spot CEP though http://www.audioforums.com/forums/smile.gif

cp.

[Jazz E Bob]

I rarely flame but this is one of my pet peaves.

In terms of simple file playback - people who think that different audio applications have different "sounds" are ignorant idiots who have no concept whatsover as to how digital audio works.

You probably also think that certain DAT tapes have a "warmer" tone with a more stable stereo image.

How about getting 128 meg of more musical sounding RAM and/or hard drive space.

Ya mum,

The Jazz.

[jdechant]

Actually the software programs can cause differences in how something recorded into it will sound. what is not recorded is calculated by complex algorithms (a proccess called oversampling) of the programs.

systems such as Pro Tools, Nuendo, and Paris are highly regarded in the "professional world" as having a better "sound" that can be attributed to the algorithms they use which are much more high end and superior than other programs' alorithms.

Samplitude is also regarded as having superior alorithms over other programs, although not considered in the league of Pro Tools, Nuendo, or Paris.

[Jazz E Bob]

So what you are saying is that if I have an SBlive, for example, and I record a wav with one application it may sound better than if I recorded it with another?


[This message has been edited by Jazz E Bob (edited 12-08-2001).]

[carparker]

yeah, you may well be correct about me being an 'ignorant idiot' Bob, but I was basing my answer on my ears not on any understanding or lack of it of digital theory and I dont need to be able to explain Boyle's law to know your fart stinks worse than mine http://www.audioforums.com/forums/wink.gif

cp

[trock]

"Samplitude is also regarded as having superior alorithms over other programs, although not considered in the league
of Pro Tools, Nuendo, or Paris."

You might find the comparison below that I found on another forum interesting as ProTools and Nuendo didn't fare too well in it. I recently took some files that I had recorded in Vegas and played them back in Samplitude - what a difference! Way, way better in Samplitude. I bought EDL Convert and converted all my Vegas songs to Samplitude songs and will never record or mix in Vegas again.

"Nuendo uses the same exact mixing audio engine as Cubase 5.0. In one word,
nuendo/cubase mixing sucks. Probably the worst mixing engine in it's class of audio products. Pro Tools, believe it or not, is terrible at mixing as well, not much better than nuendo/cubase.

Samplitude is probably the best mixing software out there. Logic Audio is nearly as good, and PARIS is probably the best overall mixing product (though it uses hardware in conjunction with software). For software only, Samplitude wins hands down. I don't think anything can touch it.

I've used Cubase for 3 years now. I couldn't stand the mixes I was coming up with, so I shopped very hard. I bought a PARIS Pro system and used it for a month. Took tracks from Cubase and mixed in it. The results were outstanding and very obvious right out of the box. We're not talking about golden-eared subtle bull****, but very obvious in-your-face kind of differences. But PARIS had weird quirks that I couldn't live with (like having to steal a track
just to have a stereo reverb, and the 16-track submix limitation really screwed with my brain), so I ditched it.

Next came Logic Audio. Again, outstanding audio quality, but very different from the way I am used to working. The environment really turned me off. So I ditched it.

In all, I have extensively tested and used Cakewalk Pro Audio, Cool Edit Pro, Cubase, PARIS, Logic Audio and Vegas. Aren't too familiar with Digital Performer, though I hear it rocks, and I think Pro Tools is just a waste of money for a prestige that only the non-recordists admire.

Then I downloaded the free trial version of Samplitude 6.0 (Free downloadable trial version that is 100% functional for 90 days!). Sounded every bit as good as PARIS, kicked Cubase's ass, and had Wavelab-like editing capability in every sense, plus tons of more native stuff (FFT filter, killer EQs, surround capability, absolutely the best crossfade editor around with the exception
of Sequoia, which is really just Samplitude but with more stuff, etc.). The mixing automation has the resolution of the best SSL board in existence. I'm sold, and think that Samplitude is the industry's best kept secret. It is no accident that the really critical mixers, people doing classical and stuff, use Samplitude/Sequoia for their native solutions.

Regards,
Bravin Neff"

[Tim Z]

JAZZ E, before you call people "ignorant idiots" perhaps you should confirm your facts first. Trock has hit it right on the head. With the exception of Pro Tools and Paris, I too have tested and listened to all those programs, and to my ears Samplitude sounds best. Mind you Trock and I must have similar ears, because we both hear how great the LynxONE sounds as well http://www.audioforums.com/forums/smile.gif. Fact is different programs can, and do, sound different.

Tim

[This message has been edited by Tim Z (edited 12-09-2001).]

[Jazz E Bob]

So what you are saying is that if I have an SBlive, for example, and I record a wav with one application it may sound better than if I recorded it with another?

( And yes, my farts reek! ;-) )

[resistor man]

Oh yeah? Well, my farts sound "warmer" and "less colored" than all of yours!

[Weidel]

Just installed this Samplitude thing. Still working on understanding it, etc, but I suppose its worth it. http://www.audioforums.com/forums/smile.gif

If this Samplitude program is so amazing, why hasn't there been a forum started for Magix?

[jdechant]

Originally posted by Weidel:
why hasn't there been a forum started for Magix?

actually Samplitude was originally distributed under Sek'd. You will find us Samplitude users under that forum. Guess they just haven't got around to changing the name yet from Sek'd to Magix.

over at homerecording.com/bbs though, there isn't a Magix or Sek'd forum and it has been asked in the past and always gone unanswered.

[trock]

I started out with Cakewalk for a few years, then switched to Cool Edit and then Vegas and I also spent some time with Logic and Cubase and Pro Tools (never tried Paris except at a trade show but always heard good things about it) and was always frustrated with each of them. Then I tried the free Samplitude 6 demo and was amazed at what I'd been missing and now I'm finally happy (in love, actually <g>).

Even just playing the tracks I recorded in Vegas in Samplitude makes them sound considerably better. And that's before adding Samplitude's excellent real-time effects. I'm also able to get more tracks from Samplitude than I was with other programs. And SEK'D has the great EDL Convert program that converts the files from the formats of other programs to Samplitude format so I'm easily able to remix all the tracks that I did in other programs :)

[xanto]

Samplitude has its own great forum. Nothing so profesional to learn and share the best audio and music production with the the real audio guys. Two of the developers answer your question just daily. They are Norman and Volker. Nothing comes so close as that forum. So, I have no idea where you say that there is no forum. If you don´t know, please, first find out.

Concerning the sound. For those who call idiots to those who understand audio; Samplitude is far superior as any other software on the market today. Only it is a bit better Sequoia, due a bit more optimized. They are made by the same team of people.

And yes, if you have a Sound Blaster and record with Cubase; first of all, delays will kill you(latencies) and the sound will not be the same as if you record into Samplitude using a SBlaster.

Myself has a SB Live Platinium. I make wonderful recordings using Samplitude. I mean; the best I can get with this card which I know is crapy. Recording in 48khz and Samplitude, there is almost no noise with SB live; not the same with the rest of the software. I have used probably all the software on the market. Nuendo, Logic, Digital Performer, Cubase, Vegas, Cool Edit, Wave Lab and Acid. My life changed when I discovered Samplitude some time ago. It´s a step ahead of the competition, not only in Sound, but in any other task you can think of.

xanto

[Jazz E Bob]

OK - Concentrate now....

Do people really believe that certain applications apply special transforms to the data coming off a sound card before it reaches the hard disk?

Ie - With the same sound card, a 16 bit mono wav recorded under Cool Edit won't sound as good as a 16 mono wav recorded under ProTools.

I just want to get an idea of just how gullible this audio engineering community is.

( My next test will be to gather opinions about whether a compressor should come before or after eq... )

Love this forum - but cranky since coming off my medication. ;-)

The Jazz.

[Jazz E Bob]

Originally posted by resistor man:
Oh yeah? Well, my farts sound "warmer" and "less colored" than all of yours!

That's because your colon is a high gain "all-analog" signal path with a valve on the output stage.

;-)

[This message has been edited by Jazz E Bob (edited 12-12-2001).]

[jdechant]

Originally posted by Jazz E Bob:
Do people really believe that certain applications apply special transforms to the data coming off a sound card before it reaches the hard disk?

Ie - With the same sound card, a 16 bit mono wav recorded under Cool Edit won't sound as good as a 16 mono wav recorded under ProTools.

Well I certainly would hope that people do not believe this. In fact how you might have come to this conclusion I am unsure of. No one here even implied this that I can see.

What I can see we are basically saying is that different programs will playback the audio differently. Different mixing engines in the programs and different algorithms used for oversampling (If you are doing digital recording, on playback there is an algorithm calculating an estimate for what was not recorded - i.e. the dead space between the samples - and playing that back with the recorded samples) can cause differences in sound that ARE audible. Maybe not audible to some, but in a recording setup (if its a good one) they definately should be.

Now as for the actual recorded signal, with the same settings and no sound card glitches, it should be exactly the same no matter what software was used to record it. I don't think anyone here is questioning that fact.

[Jazz E Bob]

So now we can use any old app to record a waveform! It doesn't matter which app because we agree they all record the same.

Let's hit the space bar!

The application of choice starts to stream the recorded data back to the card. Bit for bit, byte for byte, at the same rate it went in.

In what way can the application make the data sound "better"?

It simply cannot. Why?

"Oversampling" is achieved by increasing the rate of data to the card and interpolating the difference on the fly.

I imagine this CAN be done on native systems but only with a massive CPU overhead and associated latency.

(This explains why some people opt for DSP card solutions such as Ensoniqs "Paris")

It can easily be done on the sound card but that's the sound card, not the application.

How can it sound better?

The answer is it doesn't. My guess is that something about the app simply makes you "feel" better. A color scheme? The way it goes "ping" when it starts up? The salesmans enthusiasm? Your fear of consumer remorse?

I fear the emperor may be naked!

[This message has been edited by Jazz E Bob (edited 12-13-2001).]

[Jazz E Bob]

Originally posted by T. Alan Smith:
Greetings,

I'm upgrading from my Roland VS studio and was just wondering how CEP sounds in comparison to Logic, or Cubase.

Thanks,

T. Alan



Exactly the same.

[Kihoalu]

OH GREAT! Now we have a Naked Emperor whose farts go ping!! http://www.audioforums.com/forums/eek.gif !

[Jazz E Bob]

LOL!!!

[Wouter]

A very interesting thread at last - it's been a long time!
I have to admit that the remarks on software sounding "good" or "bad" seemed a lot of voodoo and psycho-acoustic nonsense to me, and I basically agreed with the views of Jazz E Bob on this subject. However, I couldn't resist trying things out for myself. And now I must conclude: yes, different apps do sound different. Don't ask me why, but CEP sounded "thinner" (or "cleaner", if you want), compared to Nuendo, which sounded "fuller" (a bit more bass). How can this be? Jazz E Bob: just try and listen, and you'll be as puzzled as I am right now.

Oops... I'm afraid I'll never show up in these forums again. It turned out that for some *&^%!@# reason the default output in CEP was set to my (low budget) Yamaha soundcard, instead of to the Gina soundcard I normally use. No wonder CEP sounded "thinner" in comparison. I redid the experiment, and heard no difference whatsoever between CEP and Nuendo. So it's bogus after all, Jazz E Bob is right I guess.

[This message has been edited by Wouter (edited 12-15-2001).]

[This message has been edited by Wouter (edited 12-15-2001).]

[Wouter]

I found the following post in Deja News. Good stuff on this subject!

"Valerio Dal Monte wrote:

> Sorry to ask it again, but the messages relative to this topic are no
> longer on the server I use. So, is it true that the quality of mixdowns
> made with nuendo sound worse than ones made with Samplitude or Sonar or
> else? Also on the Nuendo forum they talk about that, but I'd like the
> opinon of RAP. Thanks a lot

There is a post on the nuendo forum ISTR called 'The big bad buss test'
which compares the two doing a nuendo and samplitude mixdown then inverting
one mix and summing the results.

To sum it up, the results nulled.

Oh sod it, I think this is worth a repost... I'll copy it here.....

"Finally found the time to conduct a through test of the digital buss
issues that was such a hot topic here a few months back and keep popping up
from time to time. Thought some of you might find it a good read. Since
Steinberg does not allow me to post images (let alone audio samples) here,
all the nice FFT plots and such are left out. If anyone wants them (or have
a webspace where I can put them up), let me know.

The Programs:
Nuendo 1.53 Pr1, Samplitude 6 Demo, N-Track Studio 2.2. I chose Samplitude
as it seems to be one of the programs that many people think sound better
than Nuendo, and N-Track got thrown in there simply for costing less than a
weeks worth of nicotine. This test takes quite a bit of time to do
properly, and considering the result for these three programs I didn't feel
very encouraged to try other ones. I might run this test on a Sony R-100
console in the next few days, but I doubt very much the results will differ
from those given here

General Procedures:
All tests were done using the programs internal mixdown or bounce modes.
Now that automation and internal mixdowns seem to work in Nuendo, realtime
and offline mixdowns are bit identical for up to 24 bit wordlenghts, but
since no hardware supports 32 bit transfers yet the offline mode is
required to test performance at this wordlenght. Except from where noted,
each program was used with the factory default preferences and options. The
one exception was Nuendo's Pan Law Setting which was changed to 0dB, the
standard for the other two programs.]

The test files:
I had three sets of 25 44.1 1/3 octave sines starting at 20 Hz. One of 32
bit, 0dbFS peak files, one of 32 bit -36dbFS files and one of 24 bit 0dbFS
files. All files were generated by Wavelab 3

Test 1:
32-bit, -36dBFS file set. All channels, pans and master fader at unity.
This is a pure summing test, there should be absolutely no processing
performed on the audio streams whatsoever:
Nuendo vs. Samplitude delta file:
Zero. zip. Absolutely nothing but 0's for 15 seconds. The two mixes are a
perfect match.
Conclusion: The summing in these two programs are identical.
NOTE:
As I mention below, any given gain value you can set on the faders does not
pick out a unique gain coefficient in these programs. Therefore, if you
start a new project in Nuendo, import the files into channels and then do
an export the mix will null perfectly against that of Samplitude or
N-Track. However, if you first change any gain setting and then move the
fader back to zero (or even type in 0.00) the resulting file will be a few
thousandths of a dB quieter (most likely because the gain coefficient is
something like 0.99999999453 instead of 1). I would consider this a bug
that Steinberg ought to fix though. In general it does not matter that the
gain we get is off the gain we set by some thousandths of dB, but for unity
gain I should expect the signal to pass through unaltered under any
condition.

Test 2:
32-bit, 0dbFS file set. All channel faders at -24db, pans and master fader
at 0db.
Nuendo vs. Samplitude Delta file:
Unlike Nuendo, Samplitude by default does a very quick fade in on each new
sound clip, which resulted in some garbage during the first milliseconds.
Once this was snipped off, the result was delta file with a uniform power
level of -68.65 dbFS throughout. People might expect this to have been
perfectly null, but running an FFT and amplifying the signal back up to
take a listen revealed that the delta file was in fact a perfect copy of
the main mix, only -68dB down. Amplifying the signal by that amount created
a file that was indistinguishable from the original mix, and the FFT (taken
with 64K points and a Hanning Window) shows that every sine is perfectly
represented with all distortion below -170dB.
Conclusion:
In all likelihood the reason for the non-null delta file (and consequently,
the different mixes out of Samplitude and Nuendo) is nothing other than a
difference in overall level. This makes perfect sense when you consider
that in either program you can only specify the gain to a resolution of
0.01dB, which still leaves for a numerically large range of gain
coefficients. In fact, using Nuendo's specs (Steinberg has stated that it's
faders are accurate up to the 0.01dB resolution) the difference between the
two programs gain could have been as much as 20*log10(2*0.005) = -40dbFS.
The actual difference was almost 30dB below that again, which means that
both programs gain settings are even more precise than 0.01dB.

Test 3:
24 bit, 0dbFS file set. Channel faders and pans at unity, master fader at
-24dbFS
Nuendo vs. Samplitude Delta File:
This file was much like the previous one in that it was a near perfect
replica of the mix files themselves, only way down in volume. When the
master fader is down rather than the channel faders, the file is even
quieter with a uniform power of -84.58 throughout (the same held for the
float file tests as well). There are a few things that could account for
that. The master fader resolution of these programs might be more similar
to one another than the channel faders are (in other words, given a
identical dB setting on each fader the difference between the gain
coefficients is less on the master faders than it is on the channel
faders).
In any event the main result is the same as in the previous test; nothing
but gain separate the Nuendo and Samplitude mix. Phase is uniform
throughout the frequency spectrum and any noise and distortion is below
-170dBFS.

Test 4:
32 bit, 0dbFS file set. Channel faders at -24dB, master fader at unity.
Pans, starting from the 20Hz sine and going up, as follows: -12dB Right,
-9R, -6R, -3R, 0, -3Left, -6L, -9L, -12L, -9L etc.
Nuendo vs. Samplitude Delta File:
Apart from the left and right channels being different on this test, the
delta file is exactly like it was on the previous tests. The overall power
is slightly higher at -66.42dB, but considering that the audio goes through
twice the amount of processing (one multiplication for gain and another for
pan), this is to be expected. In fact, the delta file could have an average
power of 20*log10(2*2*0.005) = -34dB if the gain settings were only
accurate to the 0.01dB. Hence again it seems that Nuendo's faders are
actually quite a bit more accurate than what Steinberg has stated.
The only other difference is that in this file you can see the slightest
hint of signal degradation due to the multiple processing steps (pan and
gain). Distortion, although still below -160dBFS throughout, is now not
quite uniform. There are slight peaks and irregularities in the distortion,
most noticeably at what appears to be even harmonics of 3.7Khz. But again,
this distortion is astronomically small, many hundreds of times smaller
than the best D/A converters can reproduce.
So again, the only difference between the two mixes is in volume

I did about 8 more tests (pan with 24 bit, Nuendo vs. N-track, N-Track vs.
Samplitude etc.) but I'm getting tired of typing and it would be a boring
read. In every test, involving any combination of pure summing, fader gain
changes and pan gain changes, the only difference in the mixes produced by
different programs was one of volume. And further since the changes in
volume between mixes was less than 0.0005dB, it is safe to say that the
audible difference between these programs is null.

The grand conclusion therefore, is that these programs produce the exact
same result from summing, gain changes and panning. It seems evident that
there is absolutely no difference in the mixing engines (again, I'm here
including summing, gain changes and panning) of these programs."

[trock]

Sounds like the kinds of arguments going around only a few years back about how there was no audible difference between 16-bit and 24-bit. I guess the earth still is flat.

[This message has been edited by trock (edited 12-15-2001).]

[Jazz E Bob]

Actually the argument is about the difference between 16 bit and 16 bit...

Trock. The technicalities and realities of this thread are clearly out of your league.

You are lucky I'm in a merciful mood today otherwise I would happily tear you a new output socket.

Here's a couple of slaps for ya anyway...

"Even just playing the tracks I recorded in Vegas in Samplitude makes them sound considerably better."

I bet you'd be the kind of guy who would own a Samplitude mouse mat, T-shirt and coffee mug...

[sk]

Ok, Jazz E Bob...
tech specs aside, how much do you think ergonomics, for lack of a better term, affect the difference between the different programs? And do you think it's more personal, rather than objective/practical, with regard to the differences between the programs?

The reason I'm asking is that I've been using Sound Forge for the past year or so, and I've gotten comfortable with what it does, and HOW it does what it does, so to speak. And since I'm not doing any multi-track recording I don't miss or need that capacity. So I'm referring mainly to the topic of two-track stereo mastering, and how the different available software programs handle it. Because I'd heard good things about CEP, I downloaded the demo. I've been trying, for a couple of days, to work with it, but have found that I just cannot get the same level of detail on the screen with CEP that I can with SF. When I zoom in with SF I can get right down to the split second of a glitch. I can't do anything even close to that with CEP. CEP also lacks the ability to just return to the beginning of a song with the click of a button; SF has a 'return to start' and 'go to end' of file button. Maybe these are just examples of minor differences, and clearly lots of people work with both programs. I was simply bringing it up because you seem to have debunked the sonic differences, and I was wondering how you would address these more program-specific differences. Thanks.

sk

[trock]

"I bet you'd be the kind of guy who would own a Samplitude mouse mat, T-shirt and coffee mug..."

Actually no. I'm the kind of guy who has been a pro engineer/producer since 1968 and who is known for his disagreeable ears (I'm the guy who first heard and was able to demonstrate to Studer that their channels had hiss when their "measurements" proved "conclusively" that they didn't. When they actually listened, they finally did hear it).

I hold very little regard for tests done with sine waves because they don't take into account the effects of the rich upper harmonics that exist in an actual music mix.

My noticing the difference between my tracks played back in Samplitude and Vegas was completely accidental. I wasn't even looking for it. I opened then in Samplitude because I wanted to use some of Samplitude's features and was quite startled to hear how different they sounded (played flat in both programs). Then I started reading up on it and found out that many others have noticed similar differences.

All you have to do is open your mind and your ears and actually listen. But then I think you're more interested in "being right" than doing any actual listening so I won't be contributing anything further to this discussion. Wouldn't want to confuse you with any facts :)

[Andreas]

I don´t know how relevant this is but let me throw in that a phase deviation in the mix files will show up as a sine wave in the delta file. This can easily lead to false conclusions. That´s one reason why I never would do such tests with sine waves.

regards

Andreas

[Andreas]

sk,
> When I zoom in with SF I can get right down to the split second of a glitch.
> I can't do anything even close to that with CEP.

Hm, When hitting the zoom in button repeatedly You´ll end up seeing the actual sample values as square shaped bullets. That is in CEP. Soundforge5 won´t show such a fine resolution (at least I could not find a way to get there).
CEP has several ways to zoom into or out of a wave form.
- Rightclick and drag on the ruler (horizontal or vertical)
- Options > Keyboard Shortcuts lets You define your own key commands for zooming
in or out. - and for almost anything else.
(default is CTRL+Arrow Left and CTRL+Arrow right I think)

> CEP also lacks the ability to just return to the beginning of a song with the click of
> a button; SF has a 'return to start' and 'go to end' of file button.

ever tried the "home" and "end" keys? ;-)

regards

Andreas

[This message has been edited by Andreas (edited 12-16-2001).]

[Jazz E Bob]

Originally posted by sk:
Ok, Jazz E Bob...
tech specs aside, how much do you think ergonomics, for lack of a better term, affect the difference between the different programs? And do you think it's more personal, rather than objective/practical, with regard to the differences between the programs?


I think it has a lot to do with which program you first use to do real work.

It takes a while to develop a rapport with any application. You gradually learn it's strengths and weaknesses for tasks specific to your job.
Eventually the application vanishes into the background and leaves you thinking in audio terms only.
This is surely the ultimate goal of good ergonomic design.

I've tried other great programs like SF but going back to CP is like putting on an old pair of jeans. This is not to say that it's better, just that "I" am better with it.

If SF was the first application I developed a rapport with the story would be different....

:-)



[This message has been edited by Jazz E Bob (edited 12-16-2001).]

[Jazz E Bob]

Originally posted by trock:

All you have to do is open your mind and your ears and actually listen. But then I think you're more interested in "being right" than doing any actual listening so I won't be contributing anything further to this discussion. Wouldn't want to confuse you with any facts http://www.audioforums.com/forums/smile.gif

Hmmm - OK. Fair enough too.

Science IS often a way that we may be absolutely confident in an answer that is absolutely wrong. ;-)

I would LOVE to be proved wrong on this one. It would truly rock my world. (It would also devestate my ego but I'm up for it!)

Which samplitude application do you recommend I test with? URL?

[Jazz E Bob]

Originally posted by Andreas:
I don´t know how relevant this is but let me throw in that a phase deviation in the mix files will show up as a sine wave in the delta file. This can easily lead to false conclusions. That´s one reason why I never would do such tests with sine waves.

regards

Andreas


Hmmm - tricky.

Let me put my lab coat on. ;-)

Thinking...

Quoting...

"a phase deviation in the mix files will show up as a sine wave in the delta file"

Thinking...

This is how the test actually works though isn't it?

The test reveals phase, frequency, distortion differences between two files.

The delta file is a waveform containing the differences between the two files.

If a sine wave were to show up in the delta file it would mean that there was a difference!

[trock]

Hi Bob - I like your attitude :)

The one I tried was the free fully-working 90 day trial version of Samplitude Studio 6 (now at 6.02). I got it at www.magix.net (http://www.magix.net)

trock

[Andreas]

> The test reveals phase, frequency,
> distortion differences between two files.

I´d rather say it tells if there is a difference but You won´t know its origin for sure. This is crucial. Differences in Phase response for example can´t be identified using the method. And very minute differences here these could have a great influx on the perception of the stereo image.


Andreas

[jdechant]

Bob,

I was led to believe that oversampling was a part of all digital recording. I could have been led astray though http://www.audioforums.com/forums/smile.gif I am just getting started and learning.

Now for which Samplitude application to test it on, I would use Samplitude Producer 2496 v6.02... available from Magix's website of coarse. (v6.0 demo + 6.02 patch)

[Jazz E Bob]

Originally posted by jdechant:
Bob,

I was led to believe that oversampling was a part of all digital recording. I could have been led astray though http://www.audioforums.com/forums/smile.gif I am just getting started and learning.



OK - going to get me Samplitude now!

Here is a link with a blurb on oversampling - gripping reading!! ;-)
http://www.earlevel.com/Digital%20Audio/Oversampling.html

[sk]

Originally posted by Andreas:
sk,
> When I zoom in with SF I can get right down to the split second of a glitch.
> I can't do anything even close to that with CEP.

Hm, When hitting the zoom in button repeatedly You´ll end up seeing the actual sample values as square shaped bullets. That is in CEP. Soundforge5 won´t show such a fine resolution (at least I could not find a way to get there).

> CEP also lacks the ability to just return to the beginning of a song with the click of
> a button; SF has a 'return to start' and 'go to end' of file button.

ever tried the "home" and "end" keys? ;-)

regards

Andreas

[This message has been edited by Andreas (edited 12-16-2001).]


Like I said, it probably depends on what you get used to. I don't know what you mean when you say the 'actual sample values as square shaped bullets' and what particular value that has. In SF you can zoom in to the practical, useable point that you can use the pencil tool to actually draw away a glitch. And with regard to the 'home' 'end' observation, I was talking controls right on the transport itself, not a short cut key. And I don't even think there's any comparison between the speed with which the arrow keys control and zoom in SF vs CE, but again it's a 'what you get used to' kind of issue.

sk

[Jazz E Bob]

Originally posted by Andreas:
> The test reveals phase, frequency,
> distortion differences between two files.

I´d rather say it tells if there is a difference but You won´t know its origin for sure. This is crucial. Differences in Phase response for example can´t be identified using the method. And very minute differences here these could have a great influx on the perception of the stereo image.


Andreas



In the case of the testing carried out so far the "origin" of the difference is not crucial because there is NO difference.

Each application renders exactly the same waveform...

[Jazz E Bob]

Originally posted by trock:
Hi Bob - I like your attitude http://www.audioforums.com/forums/smile.gif

The one I tried was the free fully-working 90 day trial version of Samplitude Studio 6 (now at 6.02). I got it at www.magix.net (http://www.magix.net)

trock

Like my attitude?
But it's such a poor attitude! ;-)

OK - going there now...

[resistor man]

Originally posted by sk:

In SF you can zoom in to the practical, useable point that you can use the pencil tool to actually draw away a glitch. sk

Ya know, I've never had any luck messing around with individual samples, it always seems to make the problem I'm trying to correct worse. Am I just a bad artist? Have you done this succesfully?

[beetlefan]

Originally posted by sk:
Ok, Jazz E Bob...
tech specs aside, how much do you think ergonomics, for lack of a better term, affect the difference between the different programs? And do you think it's more personal, rather than objective/practical, with regard to the differences between the programs?

The reason I'm asking is that I've been using Sound Forge for the past year or so, and I've gotten comfortable with what it does, and HOW it does what it does, so to speak. And since I'm not doing any multi-track recording I don't miss or need that capacity. So I'm referring mainly to the topic of two-track stereo mastering, and how the different available software programs handle it. Because I'd heard good things about CEP, I downloaded the demo. I've been trying, for a couple of days, to work with it, but have found that I just cannot get the same level of detail on the screen with CEP that I can with SF. When I zoom in with SF I can get right down to the split second of a glitch. I can't do anything even close to that with CEP. CEP also lacks the ability to just return to the beginning of a song with the click of a button; SF has a 'return to start' and 'go to end' of file button. Maybe these are just examples of minor differences, and clearly lots of people work with both programs. I was simply bringing it up because you seem to have debunked the sonic differences, and I was wondering how you would address these more program-specific differences. Thanks.

sk
SK, I don't know what you're doing but I can get down to the sample with CEP. Also, I have no trouble getting to the start with one click.

I have the exact opposite problem with SF.

I think it's all a matter of what you're used to.

Different strokes...

[Jazz E Bob]

Do different audio applications impart their own "sound" on the audio used within them?

My test compared Samplitude and Cool Edit Pro.

Method:

I downloaded samplitude producer 2496 6.0 - 90 days trial version.

I installed it and performed the following test.

I generated 2 seconds of stereo white noise using Cool Edit Pro.

I started a 4 track Samplitude session and loaded the noise file into a track.

I removed the default volume envelope that softened the head and tail of the waveform.

I checked that gain was at unity and then rendered a mixdown.

I then opened both the original file and the Samplitude mix file into a Cool Edit Pro multitrack session.

I inverted the original file.

I mixed down the two tracks.

Result:

The mixdown was null. All zeros as far as the eye can see.

Samplitude creates the same output as CEP and therefore MUST "sound" the same.

[resistor man]

Ah, Jazz E Bob... you've overlooked the effects of the Phlogiston density in the ether flux.

[carparker]

The Jazz E Bob Flower Test:

Method:

I took two different red flowers.

In my lab I performed chemical analysis on both.

I discovered from the readings that flower A had less carthamin by ten parts per billion than did flower B.

Conclussion:

The figures prove that Flower B MUST have been prettier than flower A.

http://www.audioforums.com/forums/wink.gif

[PapillonIrl]

I think that test misses the point somewhat. I understood that the debate centred around different applications interpretation of an inputed signal, rather than generated in one and opened in another.

It seems to me a better test would be have two PC's, identical sound cards,1 running samp and 1 CE. Mic an instrument, split the signal in two, and record the same signal into both using the same cable\signal chain.

Then flip the phase of one and mixdown.

Anybody in a position to try this, or would it not be accurate enough ?

pAp

[carparker]

No, Bob's test was right on the money.

Except..

What many are saying is there is a difference in how the same wavs sound when played back through different applications.

What Bob did 'wrong' if you want to call it that, is he didn't do the one thing everyone who agrees there is a difference wants him to try, namely - listen to some music and use his ears to judge.


cp

[Jazz E Bob]

It has been established that applications do not effect what they read off a sound card!

I am fast tiring of this debate.

Do ANY of you actually have any idea as to how digital audio works?

There seems to be two types of engineers out there these days.
Oldies who blindly apply their outdated valve and magneto concepts to the digital realm.
And those who simply drool over gimmickery, pseudo-science and brand loyalty.

I fart in your general direction!

[trock]

All that's been "established" is that in 2 specific applications no differences were measurable by current very limited measuring techniques.

When there exist measuring devices that can come close to perceiving what most any human being can perceive then we'll have something worth discussing.

This is probably as long a way off as having computers that can exercise the kind of judgement that (some) human beings can exercise - so-called "computers that can think". A very long way off.

In the meantime we have those who can hear and those who won't even listen in case they fall off the end of the flat earth they reside on.

[Jazz E Bob]

Limited techniques?

You just don't get it do you trock?

THE OUTPUT IS EXACTLY THE SAME.

There is no place left for your "sound" to hide.

Do you actually understand how the tests work?

If you take a DIGITAL representation of a waveform and mix it with an inverted version of itself it will null itself out.
In the case of DIGITAL audio this nulling will be PERFECT. No error - no room for God. It's just basic maths!

Look - I like to think that there is magic in the world myself. But it is simply NOT here! We forfeited magic when we changed to the digital format.

My flabber is well and truly gasted. This is like talking to a lunatic.

PS

You can take your over-romanticised "golden ears" flat earth attitude and shove it up your valve!!!

[87tsop]

As far as I see it, it doesn't matter how it sounds in an application unless you code your audio so that it would only be playable in that application by the consumer. What matters is the finished product, the cd. Whatever this thing is about "sounding" different in other programs I don't know. I personally have never noticed it. The point is that if the digital signals are exactly the same, as proven, than they are indeed the same wave. The program doesn't matter, the cd does. There will be absolutely no difference in the sound of the two cds burned the exact same way on the exact same cd player with the exact same settings. The only explanation I can think of is perhaps a missed setting internally within the program altering the out. If it does sound better on one application, throw that program away. You don't want it to make it sound better, you want it to tell you what it really sounds like.
merry christmas

[trock]

"Ah, Jazz E Bob... you've overlooked the effects of the Phlogiston density in the ether flux."

LOL! Actually what Uncle Bob has overlooked is that in his "test" he used a single stereo file (of white noise, no less) rather than testing what's been discussed in this thread - the summing of multiple music tracks using the different mixing/summing algorithms that different DAWs use in summing multiple tracks and that are quite audibly different to anyone who dares actually listen. There's none so deaf...

When I take 32 24-bit Vegas tracks and use the most recent update of EDL-Convert to convert EDL formats and then open them in Samplitude with the same volume and pan settings they sound audibly different summed in one program than in the other and neither I nor my clients care whether Uncle Bob can hear or explain the differences. He's quite welcome to his viewpoints.

I haven't tried the same thing with CEP but I will when I return home in January. I've always been happy with the sound of CEP so it will be interesting to hear if there are any audible differences between it and Samplitude or not. There may well not be. Then again...

[This message has been edited by trock (edited 12-22-2001).]

[Jazz E Bob]

Trock - I love ya' but ya' sh*t me ;-)

I feel we are getting somewhere now though.

We need to use a test that will show up differences in the summing algorithms. (Not the buss "flavour")

Hmmmm.

How about this...

4 tracks of stereo data.
Sine wave
White noise
Music
Voice

Loaded into 4 tracks of CEP and SAMPLITUDE.

MIXED the same way. eg.

Sine -6db
Noise -13db
Music -2db
Voice -4db

Master -5db

One MIXED stereo waveform from each App.
Invert one MIX... etc
Examine the Delta file for anomolies.

I will not proceed with the test until you either:

+ prove that it is not a valid test
+ make adjustments to it's parameters and approve it's validity.
+ agree that the test is gorgeous as it stands

Love,

Uncle Bob

XXXOOO

[Jazz E Bob]

87tsop

I agree with the "tell it like it is" premise.

A MIX is simply that and should not color the sound in any way.

You should be able to pass through the mixer thousands of times make no change to the data....

It seems though - we are really interested in how alogorithms choose to MIX data. eg do they simply round to the closest sample, round to the next highest sample, dither a bit, or even incorporate Phlogiston density in the ether flux???

Who knows.... (Trocks ears say they do. ;-))

The test I am proposing will not tell how they do it. It will simply tell IF they are doing it differently.

If they ARE doing it differently then I will have egg on my face. It means that there IS in fact room for Apps to have a different sound.

Stay tuned...

[windowman]

Now this is fun! Sounds like a debate between G. K. Chesterton and Robert Peel Glanville Blatchford about the existence of free will:

Blatchford: "I'm a strict determinist, all our actions are the result actions and reactions starting with the big bang (never mind that we have no idea who/what pulled the trigger on that "bang" thingamabob.)"

Chesterton: "If you don't believe the will is free, it would be unreasonable to thank someone for passing you the mustard."
**********************

Ah…but for the topic at hand…without going into any tests other than the recorded music my ears take in, I've never been able to hear the least bit of difference between the wave files I create on CE 2000 and the ones I create with Cakewalk GT Pro. I've also never been able to hear the least little difference between a 320k MP3 file and the wave from which it's existence sprung. Nor can I hear any difference between 16 bit and 24 bit recordings (as long as there was no heavy processing/re-saving of the wave files) or between 44.1 and higher sample rates.

As much as I hate to side with a determinist (sorry uncle Bob, but then, if we don't have free will then there's no point in any of us being sorry for anything :-), I must conclude that he is right. Or as they say on the Tonight Show—"You are CORRECT Sir!"

I suddenly feel like brushing my teeth. :-)

[Renfrew]

"I've also never been able to hear the least little difference between a 320k MP3 file and the wave from which it's existence sprung. Nor can I hear any difference between 16 bit and 24 bit recordings (as long as there was no heavy processing/re-saving of the wave files) or between 44.1 and higher sample rates."

Wow! I sure hope you're being satirical here! Given a good monitoring system, the differences between any of these are really quite audible. If you really can't hear them, perhaps you should look at another line of work.

Poor Bill E. Bob, if his flabber is gasted by something so simple as this discussion, I pity his flabber if he were to study these pages below!
http://www.johnvestman.com/digital_myth.htm http://stage.digitalprosound.com/2001/06_jun/features/debunk/debunk1.htm

There is also a wealth of useful information on this subject at the Digital Domain mastering pages at http://www.digido.com - especially the articles "Back to Analog", "Dither", "More Bits, Please", "Depth and Dimension in Recording, Mixing, and Mastering", "FAQ on Digital Audio--the tough questions", "The Secret Of The Mastering Engineer" and "Jitter".

Digital really is not the God Bill E. Bob thinks it is. Neither is analog but, flawed as they both are, they are definitely usable. It just pays to be aware of their limitations.

Ren

[beetlefan]

Originally posted by windowman:

Ah…but for the topic at hand…without going into any tests other than the recorded music my ears take in, I've never been able to hear the least bit of difference between the wave files I create on CE 2000 and the ones I create with Cakewalk GT Pro. I've also never been able to hear the least little difference between a 320k MP3 file and the wave from which it's existence sprung. Nor can I hear any difference between 16 bit and 24 bit recordings (as long as there was no heavy processing/re-saving of the wave files) or between 44.1 and higher sample rates.


[/B]

I suggest for you to get a serious hearing test. I can hear the differences without even trying.

[windowman]

Wow! I sure hope you're being satirical here! Given a good monitoring system, the differences between any of these are really quite audible. If you really can't hear them, perhaps you should look at another line of work.

Care to put your money where your mouth is? Now admittedly I've only gone digital about a year ago but, I've owned the Sheep Shack recording studio for almost 20 years. In that time I've had plenty of oportunity to have some fun with listening tests. There's only one I ever use and I try it on everybody that comes into my shack who's game to play. I simply have them sit down in a chair in the engineering booth with their back to the equipment. I give 2 samples of a TelLark recording of the London Philharmonic playing Vaughn Williams' "A Lark Acsending"; one from the original CD and one from a cassette tape of the CD that I recorded onto an inexpensive, 12-year-old NAD 6240 cassette deck using a TDK metal tape and encoded with nothing more than Dolby C. The cassette deck is only flat out to around 16k. I play them several short bursts of both the CD and the Cassette being careful to mix them up as well as I can. Not one person has EVER been able to consistently tell which is the CD and which is the cassette. I've played this game with everybody from musicians to other recording enthusiasts.

And you think they're going to tell the difference between 16 bit/24 bit, wave and high bitrate MP3, 44.1 and 96 etc? No way--no how.

But if you'd like to test the topic at hand, the magical ability (as Bob puts it) to distingwish between various wave files recorded on different wave editors/recorders, I'd be happy to run a few listening tests by you.

I'm willing to sit down and do a short recording of myself playing acoustic guitar (10 seconds ought to do it just to make it download fast), one on Cakewalk, one on Cool Edit, and one on Cubas (I think I have an evaluation version of this one on CD somewhere) and I'll play the exact same 10 second musical passage on all of them and upload the files to a server somewhere where you can download them and try to guess which is which. Just for fun I'll take a couple of those wave files and record them at 48 and then dither them to 44.1. That ought to change the sound right? And I'll also take a couple and convert them to a measley 128k MP3 and then back to wave again. Well that ought to be really easy to spot according to your theory eh? Game to try it?

And I resent your remarks about my apparently being Bob. Any moderator can look at my IP number and see that I'm surfing in from Illinois and then compare it to Bob's.

[Renfrew]

Well, what you say casts serious doubts on your monitoring system. On a good transparent monitoring system these differences are really not hard to hear. Plus your test was done from a 16-bit CD - already a degraded format (particularly early CDs).

P.S. I don't know what you're referring to regarding any implication that you are Bob - there was no such implication.

[windowman]

I don't know what you're referring to regarding any implication that you are Bob - there was no such implication.

Poor Bill E. Bob

Well, what you say casts serious doubts on your monitoring system. On a good transparent monitoring system these differences are really not hard to hear.

That simply is not true whatsoever; the equipment I've used over the last 20 years, most of which I still use, is not in the least inferior to anything offered today. This is especially true of monitor power amps and tube pre's. Why do you think so much of the "old" gear from bygone days is worth so much in today's market? Because it's inferior? It's worth more because it's better made, often with point to point wiring. And because it sounds better. Do you really think there's a console made today that can compete with a vintage Neve and it's tube preamps? Or a power amp that can contend with an "old" Crown tube model? (No, I don't own a Neve, but I'd give my eyeteeth for one.)

Plus your test was done from a 16-bit CD - already a degraded format (particularly early CDs).

Do you have the slightest idea as to how 16 bit was arrived at? Do you honestly think that 24 bit has only been attainable in the last couple of years? CD's could have been 24 bit almost from their inception. (Actually they started with 14 bit but later changed their minds and went with 16.) There were two reasons for going with 16:

One—any higher bit rate was considered to be of no higher benefit to the listening experience.

Two—the amount of music you could put onto a CD at 16 bit was so much more than 24 et al.

Besides, that CD is flat out to 20k compared to 16k of the cassette. That should make a much bigger difference in the sound than any differences between a 16 bit wave and a 24 bit wave or even a 320k MP3, all of which maintain a certain amount of flatness to 20k. And yet not one person has ever been able to consistently tell the difference.

If you record a two tracks, one at 16 and the other at 24, and don't make any changes to the recorded material afterwards, the finished product will sound identical. 24 bit is only better if you do a lot of processing. Obviously you loose bits with every conversion so if you make a lot of changes then it may be beneficial to record at 24. I personally think the better route is to sit down and actually think about what you're going to record before you record it so you can do it right in the first place. I often print effects as I'm laying tracks. I know exactly what they'll sound like in the final mix along side other instruments because I've been doing it for years and years. I have a log book I keep with knob settings, etc of my most used effects, (been using the same DBX compressor forever for instance.) Use the right mic, the right room, the right effect, and 16 bit is all you'll ever need. Anything more is simply overkill.

I just bought the latest CD by fingerpicker, Don Ross, "Huron Street", because it's been touted by magazine promoters as "superior" in sound because of it's 24/96 bit/res. I believe it's more of an attempt to promote the sales of the Genex Digital recorders it was made on. I'm telling you, I've got recordings of acoustic guitar made at 16/48 that are way better than this. Part of it may be the mic's they used…dunno. I only know that I've yet to hear a 24 recording that sounded any better than my 16. They also said that DVD's would sound better because of their 48 sample rate. Nope.

But like I say, we can always do a listening test….

[Sonic Valley]

I say we go for it. Bring on the test. I'd like to see if I'm a sucker or not. I recently switched my system over to 24 bit and I can hear the difference. Let's see if my ears are right. Anyone else wanna experiment?

[Renfrew]

"Poor Bill E. Bob"

That section was about Bob, not you - sheesh.

Reading what you say is like stepping back into a time warp and coming out about 5 years ago when such "arguments" were prevalent. I'm not going to bother to reply to them because this stuff has been debated ad nauseam over the web for years with the tide finally having swung far in favor of the audible benefits of 24-bit (due in part to the nature of upper harmonics and how much they contribute to the sounds of musical instruments - part of why so many companies make amps that have frequency responses in the 100-300K range) but it's pretty amazing to see somebody still spouting this kind of stuff in 2001. And actually a major part of what made early CDs sound so bad was mainly the very poor quality converters in use back then that produced high-mid harshness rather than real highs.

Still curious to know what you're using to monitor with (speakers), though.

Ren

[windowman]

"Poor Bill E. Bob"

What part of Bob's name has the word Bill in it?

Speakers? Sony for small desktop but believe it or not I use an old pair of American Acoustics for high-fi comparisons (8"woofer-1"tweeter-2-way.) They've been re-coned once.

I say the proof's in the hearing--not scope tests. Either you can hear the difference or you cannot. I'd be willing to bet that I can record a CD track in real-time on both Cool Edit and Cakewalk and then convert one of them to a 320k mp3 and back again and no one would be able to tell me which of the three was converted. No way...no how....

Yes CD's were grainy when they first came out because of a lack of upper harmonics. But NOT of the 300k variety. Anyone who believes that is just fooling themselves IMHO.

The Proof's in the listening tests.

I'll load some files tomorrow. Should be fun! I won't ask for anyone to post their name. We'll just get a good amount of testers and compare overall results.

[windowman]

Ren;

Sorry about the Bill--Bob remark. I forgot that my name doesn't show up in the profile. It's Bill. My apologies.

[Heavens to Betsy 2]

Nor can I hear any difference ... between 44.1 and higher sample rates.

Three weeks ago, I would have grudgingly agreed with you. But--three weeks ago, of course--I happened to reload some older mixes for a band's album that I'm finishing. I've been working with 48kHz for the last six months, and these were done in 96kHz.

It was not my intention at all to try an A/B comparison--or even to notice any difference at all. As fate would have it, though, I listened to a track of drum overheads in 96kHz. I nearly fell out of my chair from the sound! Most prominent was the clarity and smoothness of the higher frequencies--cymbals, high-hat, drum attacks. And this was recorded with lowly SM57s, to boot. I finally A/B'd this track to a recent track of overheads at 48kHz, same set and drummer; and a notch or two higher in mic quality, using a KSM32 and a KSM44.

Yet still a difference! I A/B'd tambourines recorded with the KSM32, 96kHz-vs-48kHz. Again, a clear, treble lightness as opposed to a grainier bite.

I don't know how, what with the Nyquist theory employed to debunk 96kHz, but I trust my ears.

The best analogy I can make would be 96kHz is to 48kHz as vaseline is to wet sand. I have officially been cured of my skepticism. For the sake of expedient downsampling, I'll use 88.2kHz on future projects--but it'll be close enough to 96.

[windowman]

For the sake of expedient downsampling, I'll use 88.2kHz on future projects--but it'll be close enough to 96.

Yeah, definately go with 88.2 if you're gonna record that high.

On the aside, I had a big workday. It's 7:30 and I just got in so I won't be putting up any samples tonight. I'll try to get at them tomorrow. Also, you never know, I just may fool myself. http://www.audioforums.com/forums/wink.gif I'm certainly open to doing whatever it takes to get the best recordings I can and if the majority here can correctly guess the samples I put up then I'll no doubt re-think the whole sample/resolution thing. I am open-minded!

~W~

[Bajoran]

If these tests are to have any merit, please publish the full details/specs etc. of any equipment - hardware and software - used in the recording chain. Reason is that not all 24-bit equipment - sound cards/converters etc. - are created equal and not all are capable of capturing or reproducing the nuances that 24-bit is capable of.

Also, ideally your sound source would be a full-range mix that included well-recorded drums, cymbals and reverb tails etc. If you use single sound sources, instruments with rich high harmonics - such as 12-string guitar, tambourines etc. would be recommended. Sometimes people even use jangling keys in such tests.

[This message has been edited by Bajoran (edited 12-29-2001).]

[windowman]

Actually, I decided to do something very different. In fact your comments about the musical content are pretty well in line with what I've done. It's the 16/24 thing I've abandoned though. Why?

Because first and foremost I don't believe hardly anyone (if anyone) can even tell the difference between musical content well within the bounds of human hearing, let alone upper harmonics above 20k or just more redundant information between 20--20k either which is all going to 24/96 is gonna do in my opinion. It's like taking 16 tracks recorded on 2" tape and saying that because 16 on 2 inches sounds better than 16 on 1 inch then 16 on 2 foot wide tape should be better still. Then why not go for 16 on 2 yard wide tape or even bigger. Well even if a machine could be made to accomodate that request there's simply a point you get to where any further improvement in sound won't be detectable by humans. I'm convinced we've easilly reached that point with 16/44.1 and that manufacturers are simply giving you a load of manure to try and entice you into buying new gear and software that you don't really need.

Matter of factly this whole notion of humans being able to hear anything from 20 to 20k is not entirely correct. Actually, it's only children who are capable of that for the most part. After the age of 15 or so the numbers start to change dramatically. Few adults can hear anything above 16k.

For this reason I've decided to test people on their ability to destinguish between recorded sounds within the normal 16/44.1 wave file frequency rage and lower. I think it's safe to say that if a person can't hear changes that are made to files within the already established ranges of human hearing that they certainly won't hear any differences within files that are any better and it would be silly to think otherwise. I mean, if we can't tell the difference between a cassette tape and a CD then there's not much point in testing if we can hear differences in sound files that are supposed to be even better. That would be fairly absurd.

So here's the test:
http://www.geocities.com/billseper/index.html

It contains just an egg shaker, triangle and batas. That should cover the frequency spectrum more than well enough. If there's any skewing in the high end, a triangle certainly ought to be telling of it. Likewise, if there's any bass bump distortion in the low end the batas would give it away I'm sure.

Good luck all; and I've sent an email to our moderator here offering to give him the answers as to which file is which as soon as he writes me back so you'll have someone to keep an eye on me and make sure I'm not trying to pull anything.

[This message has been edited by windowman (edited 12-29-2001).]

[guitar ed]

Boy, I love this s*#t !!!!!!!!! I swear I'm gonna dust off my PortaOne and say to hell with it! 8 tracks on a cassette is looking a whole lot less confusing to me! Keep it up, though, cuz I'm really diggin' this one!
Ed

[Steve H]

I was once involved in a double blind listening test to evaluate the merits of Monster cable versus #8 Romex for speaker wire. The results: If there were perceivable artifacts caused by either wire it wasn't humans who could tell the difference. Yet, sales hype and advertisements still claim that there is a significant difference, ..... but that's what sells speaker wire.

I bet "trock" uses Monster cable, ..... I however use Romex (it's cheaper).

Steve H

[Jazz E Bob]

Trock

Get this...

I made up the mix as specified and I CANNOT get the mixdowns to null.

The Delta file seems to be a tiny version of the original mixes though....

This must mean that -6dB on one app isn't the same as -6dB on another...

Hope you had a great Xmas - I'll keep you up to date with my findings...

[Jazz E Bob]

OK - the test's I've done produced the same results as the tests that Wouter published earlier in this thread.

At best a Delta file is produced that is identical to the source material but at an extremely low level. (The audio is represented by one bit!!!)

On the bright side your golden ears are intact - there IS a difference in volume level AND samplitude puts a little fade in and fade out at the beginning and end of waveforms - this could also improve poorly truncated waveforms... (less clicks?)

I have learned a lot out of this process. Samplitude looks brilliant!!! Very refined and professional.

:-) All the best!

[jdechant]

Jazz E Bob,

glad to see this thread got back on track. I dropped out cause for the longest time, each side was debating something completely different...

glad to see it got back to the original question of mixing alroythms...

nice test by the way...

also, Cool Edit Pro and Samplitude are very similar progs... an interesting test would be to do the same test with "high end" prog such as Nuendo and compare its mixing alorythm to Cool Edit Pro or Samplitude... I have copies of all 3 progs if anyone would like I do a mixing test on a "high end" vs. CEP or Samp.

Josh

P.S.. might just do it for my own enjoyment http://www.audioforums.com/forums/smile.gif

[Robert D]

Is it too late in the song to cut in on this dance? Just finished reading through this thread, I'm exhausted!
I'm glad, after much flailing around and gasted flabbers (got a good laugh out of that line http://www.audioforums.com/forums/smile.gif) that the discussion and testing finally got around to the relavent topic. No, there should be no difference between any of these programs when recording and then playing back a wave file. The program isn't doing anything in this case, just passing the numbers on to the disk drive, then passing them back to the sound card. The difference's begin to show up with simple gain mixing, but only in a very subtle way, probably measurable but audible only to the golden ear set, of which I won't pretend to be a member. The difference here is in how accurately the programs do their math, and how they truncate the results. Then we'll throw in simple fades, and then crossfades. Now there's more math going on, and we're starting to use some algorithems. The difference gets more audible.
Now start doing some EQ'ing and the real differences between applications come into play. Filter design is filter design, be it analog or an algorithem describing an analog model. If you give Hartly Peavy and Rupert Neve the same components to work with, do you think the EQ they design are going to sound the same? NO!!! (Sorry Hartly)
On to Native effects......well, you get the point. If Wolfgang Puck and I both make toast, it's gonna be bread in / bread out. It's when we really start cooking that it's going to become apparent who's the talented chef.
RD

[oki]

ok ive heard enuogh.

absulutely, there's a differnce in multitrack
quality features etc.if u think otherwise than u need you'r ears checked out, while you'r at it have em check you'r brains too.

i dont know who trock is but well said trock.
i too tried almost every multitrack u can think of,in my book samplitude at this moment pisses on all of em hands down .in march they are coming out with version 7 i am only excited to see,they say wait till u see version 7.

in samplitude not only they are good software engineers it seems like they know something about pro recording,thats what the software industry needs people who have had hands on experience or knowledge in the real world of pro recording.believe me i guess i am falling in love with samplitude too it is unbelievable i felt at home with the software the minute i downloaded it and its the demo full working,and it reminded me that i am supposed to record.believe me i am experimenting on 400mhz pc never said ouch or crashed or any.bottomline samplitude at this moment rocks.

and jazzbob,by calling people idiots you are only embarrising yourself my friend.open up
your ears and eyes you will see the differnce in beauty,comes from simplicity.

in this case it is samplitude to some us.

[jecahn]

I'd like to chime in, here with something that really isn't relative to this discourse. (And doubly so because I'm picking up the "differences in bit rate / sample frequency ball and running in the opposite direction with it.) I'm gonna say it anyway... I don't get involved (for the most part) with the ongoing debate relative to bit depth and sample rate, etc. (although, I will comment that it does tend to be one of the most civilized arguments on the 'net) However, I will say that if you can't hear the difference, there isn't one. Now, that doesn't mean that I won't play the same file pair for Bob or Vance or Trock and expect them, too, not to hear a difference.

So, first and foremost, we KNOW that everyone's hearing range and manner is different: we all hear things differently. We can (I think) agree upon that. I can hear the difference between and MP3 and a first generation WAV or even a CDA track. I can't necessarily tell you which is which, but I can tell you that I can hear that there is a difference. I think that most people can hear the difference, even if they think that they can't. Take any CD out of your collection, rip it to as high a quality MP3 as you like, convert it to WAV files and burn it back onto a CDR. For the next year, when you want to listen to that CD, listen to the ripped, converted version. After the year is up, go back to the original version and I guarantee you'll hear the difference!

Now, as for straight sample rate / bit depth differences, I didn't think that I could hear the differences either until I upped those numbers on projects that I was working on. That's not to say that someone else working on the same files would have a similar opinion.

However, for the ultimate disclaimer.... My #1 rule above all else when it comes to all of this is as follows: "Either you start with crap or you don't. If you start with crap, you end up with crap." Like Beavis says, "You can't polish a turd, now pass me the mustard!" I have (previously referred to) Magic Sam bootlegs that were made live to 1 track on a crap tape deck. The recordings are unlistenable from an engineering stand point. However, the material that was recorded is so unbelivable that the quality of the sound matters less than it would ordinarily.

So, I think, not only does sound quality not matter so much if the material is there, but in some instances, cruddy sound quality adds to the character and feeling of the recording. In these instances, cruddy sound quality is necessary. I like to think of sound quality as a potential paint brush in my box of colors. Nylon string, solo guitar? I'd work on that one in total digital in as high a quality as I could. Punk band that practices in the basement of an "adult movie" store? You'd be doing them a dis-service not to record them onto either a cassette 4-track or even live, right to 2 track reel to reel. So what's my point? Some folks like Fords some folks like Chevy's and never the two shall meet. It's interesting to watch the tide of conversation, though. Have fun!

[Sonic Valley]

Hey Jecahn. The whole 24bit thing...You can't really hear the difference ...UNTIL...you start processing, that's where having the higher resolution of 24bit comes in handy. Remember...everytime you process a file, real time or not...it's recalculating the file, thus degrading it.

[jecahn]

Actually, I've heard a difference even before any processing. I find that it's more difficult to get a 24 bit file to clip than a 16 bit file. I have to laugh at myself, though because while I can hear a difference between 24 and 16, I secretly think that anything above 24 probably wouldn't make that big a difference. Like there's a brick wall of sound improvement that stops at the 24 bit mark, right?

[oki]

bottumline is,theres is a huge difference in software to software,if anybody dosn,t see that,than download a free software and get a radioshack mic and run a pro studio. and save the rest of the mony for you,r future
downpayment on a house.how doe,s that sound
for a start?

btw,those who bring in electronics engineering fine,but we r in the recording business we could care less about the inner structure of an electronics,as long as it delievers quality,stability,etc,etc. yes we care about the external specs of what a software,soundcard ,dynamics,converters,etc.
etc.

we sit here and shoot all we want,meanwhile
in real life,in pro studios u need product with quality ,stability,ease of use,features
etc.those who make such software i believe they will win the industry period.