SAMPLITUDE RULES (DAW Summing)

[defjamm]

okay you are right samplitude does add soemthing to the sound.

I just did another test.

I didn`t even try to sum anything (lol we are talking about audio software).

So I just loaded one single wav file into Cubase SX and Samplitude and rendered this one to a wav file. If I mix these two ( one flipped in phase), there still is some sound at -83/-82 db...

okay than I did mix the original wav file with the Cubase render. and it perfectly cancels

then I did mix the original wav file with the one generated/rendered with Samplitude....
and finally there is some rest of sound ( noise and something from the original sound....
I did normalize this to hear it with my own ears....

So Samplitude does alter the sound or there is a bug.... there is also some noise.

So why do I and some other guys think to hear that Samplitude sounds better?

Perhaps the noise does give the analog feeling and help not sounding like in vakuum? or is this some kind of inbuild mastering fx?

PLZ only serious answers!!


PS: I can upload the files to my private webspaces, if you like.... maybe later, have to do some work now...

i don't know what you are doing, but they null perfectly here(i took a 24bit-drum-file).

[stefancrs]

If so, how's the quality of for instance Podium? It has _no_ bundled effects. Does this mean that it sounds better or worse than Cubase? If Samplitude sounds better than Cubase because it has better bundled effects I mean...

No actually it's the other way round. In 2003 the 3D audio bulleting board did a fairly definitive test of a great many DAWs and hardware mixing desk. As a result of this test it was discovered that Samplitude 7.1 had a summing error at -133dbFS. As a consequence this error was removed in S7.11. The version of Samplitude used in my comparison with Cubase SX 1 is just about the only version of Samplitude in which it is certain summing is absolutely accurate.

http://www.3daudioinc.com/3db/archive/i ... /f-15.html

It's worth visiting the above web site, as the tests conducted then are in all probability definitve summing tests.

Oh, I misunderstood your post then. Sorry. I've been to that site before, and I didn't then nor now see any published results. It seems to me, when briefly looking around that is, that their test is based upon listening to, not measuring, the output from different hosts. Also I'm always suspicous towards truncating the master to a lower bitrate (because then the truncation will introduce an unknown to the equation) instead of rendering to direct 32bit float. But that's maybe a personal preference from my side.

However, it seems that Samplitude 7.11 sounds like all other hosts so far. Haven't seen anything pointing towards anything different yet. Guess I maybe should get the demo and try some stuff out on my own

[stefancrs]

So why do I and some other guys think to hear that Samplitude sounds better?

It COULD sound different, but I seriously doubt it. If it's of any consolation, people in the audio industry hear things that aren't there all of the time So it could be any of the two, really.

[inzeenir]

My tests show little or no difference between logic, samplitude and eXT but that was just summing 4 16bit wavs(compared with inversed phase). The cool edit mix was a tad louder though. So I'm comfortable with eXT regarding sound quality.

[hifiboom]

@hifiboom
You did'nt have dither on? That would account for the noise.If you think about it there is no reason for the added noise. And sines are'nt so good for testing 'cause if they are just 1 sample off timewise you will have a really low signal when trying to cancel out..If you use noise instead you would have full volume. So it is easier to hear if you made a mistake somewhere.

okay just checked this and I have dithering set to on....

so this was the problem.

now the files cancel out perfectly!

one question, does active dithering involve every channel in a sequencer, or just to the final output(master output channel)?

[stefancrs]

Final output. Dithering is used when lowering the bit depth, which you usually do when you make a mixdown. Internally, the master is 32bit (and in some hosts, 64), and if you make a 24 or 16bit mixdown, dithering can, and should, be applied.

[stefancrs]

Reaper Beta 0.62 (not officially released), the last sample of the render is off, so I cut that one out. Also it seems to have some issues with fader precision:
All faders at 0dB cancels out with a peak amplitude at -156.53dB against Podium.
The faded one does a bit worse, -91.29dB peak amplitude after cancellation against Podium.

[l33troll]

Funny, I've read recently that PT has objectively sounded very bad for a long time, but that didn't prevent pros from using it because sound was only part of the equation for pros. They did sacrifice sound for ease of use/cost/editing, and did produce the usual pop with the early PT systems that sounded worse than any current DAW.
The conclusion was that differences in 'sound' are not a legend but who cares?

[jupiter8]

Funny, I've read recently that PT has objectively sounded very bad for a long time, but that didn't prevent pros from using it because sound was only part of the equation for pros. They did sacrifice sound for ease of use/cost/editing, and did produce the usual pop with the early PT systems that sounded worse than any current DAW.
The conclusion was that differences in 'sound' are not a legend but who cares?

Here's a tip for you: Don't believe everything you read.

[jackson]

Funny, I've read recently that PT has objectively sounded very bad for a long time, but that didn't prevent pros from using it because sound was only part of the equation for pros. They did sacrifice sound for ease of use/cost/editing, and did produce the usual pop with the early PT systems that sounded worse than any current DAW.
The conclusion was that differences in 'sound' are not a legend but who cares?

Most pro's who said they didn't like the "ProTools Sound" had been using analogue gear previous to ProTools. It DID sound different to that.

[kilroy]

If it's of any consolation, people in the audio industry hear things that aren't there all of the time

To true...alot of them are convinced they are hearing good songs.

[sboullion]

To true...alot of them are convinced they are hearing good songs.

Now THAT'S funny! And sadly, too true.

[JonasNorberg]

This is the best thread ever! It took hours to read through and It cracked me up so many times

My god, it's so funny when people first claim to be experts and then make the stupidest comment/question ever ... aaahhh ... internet

Hey, steffo, I admire your will to do some constructive comparisions and I'd love to help out with the tests. I'm pretty sure that my host (Cubase LE 1.0.7 Build 98 Built on April 8 2005) would have the exact same sound as Cubase SX3.

And to all placebo kids I say: Keep trolling!

[Sascha Franck]

To true...alot of them are convinced they are hearing good songs.

[stefancrs]

This is the best thread ever! It took hours to read through and It cracked me up so many times

My god, it's so funny when people first claim to be experts and then make the stupidest comment/question ever ... aaahhh ... internet

Hey, steffo, I admire your will to do some constructive comparisions and I'd love to help out with the tests. I'm pretty sure that my host (Cubase LE 1.0.7 Build 98 Built on April 8 2005) would have the exact same sound as Cubase SX3.

And to all placebo kids I say: Keep trolling!

Unless there's any bugs in it, it will most likely cancel out somewhere in the -150 to -160dB range, yah. But wtf, what's the point? There's only a point if someone proves us wrong _without_ the difference coming from a temporary bug, but from a "different audio engine design" type of thing.