Old 08-11-2011, 04:03 AM   #1
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default anyone else prefer the integer mix mode?

Although it's certainly a tough call, I think I'm actually prefering the 39bit in mode on my present material to the 64bit float. Something about the image and how solid it is. I think i'm missing some low level stuff, but who knows.

Anyone found the same? (but please let's not get into this vs that + obligatory science lessons - just how's it going for you kind of thing. No doubt in my mind that the 64bit float default is already astonishingly good and detailed).

Which makes me really want to hear what a higher resolution int mix would sound like.

Cockos, is this possible to go beyond 39bit int without an architecture re-write?
griz lee is offline   Reply With Quote
Old 08-11-2011, 04:29 AM   #2
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

I'm not preferring it yet, no. I've only been testing with a 7-track piano, cello, sax, t-bone and voice project. I've abx-ed and can't hear the difference. I've also nulled the two bounces all the way down to dither (this required no funny business at all, just a polarity inversion) ... so make of that what you will. They are the same, except for dither which is obviously not correlated on each render. I can upload the difference signal if anyone wants to hear what dither sounds like after 120dB of gain

I'm sticking with 64fp for the foreseeable future
timlloyd is offline   Reply With Quote
Old 08-11-2011, 06:11 AM   #3
vinx
Human being with feelings
 
Join Date: May 2007
Location: France
Posts: 662
Default

dither can be heard in a mix (that's why there are different ones), don't trust the 100000dB gain and noise floor and so on.
Human hearing is way more complex than "90dB+ 20 to 20000Hz"...

Interesting that cockos gives us the chance to choose between int/float.
+1 to go beyond 39bit int.
vinx is offline   Reply With Quote
Old 08-11-2011, 06:16 AM   #4
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

Quote:
Originally Posted by vinx View Post
dither can be heard in a mix (that's why there are different ones), don't trust the 100000dB gain and noise floor and so on.
Human hearing is way more complex than "90dB+ 20 to 20000Hz"...
You may have misunderstood my post if you think that this has anything to do with what the dither sounds like. In both renders the dither is the "same" - it is not relevant.
timlloyd is offline   Reply With Quote
Old 08-11-2011, 07:04 AM   #5
JasonSpatola
Human being with feelings
 
Join Date: Aug 2009
Posts: 257
Default

Quote:
Originally Posted by griz lee View Post
Something about the image and how solid it is.
Can you post an example?
JasonSpatola is offline   Reply With Quote
Old 08-11-2011, 07:32 AM   #6
airon
Human being with feelings
 
airon's Avatar
 
Join Date: Aug 2006
Location: City
Posts: 9,650
Default

Quote:
Originally Posted by griz lee View Post
Although it's certainly a tough call, I think I'm actually prefering the 39bit in mode on my present material to the 64bit float. Something about the image and how solid it is. I think i'm missing some low level stuff, but who knows.

Anyone found the same? (but please let's not get into this vs that + obligatory science lessons - just how's it going for you kind of thing. No doubt in my mind that the 64bit float default is already astonishingly good and detailed).

Which makes me really want to hear what a higher resolution int mix would sound like.

Cockos, is this possible to go beyond 39bit int without an architecture re-write?
Here's the math deal.

The 64-floating point mixing path runs on a 53-bit resolution floating point number with an 11 bit exponent.

Every time the the value changes its floating point left or right, in other words it dips below 0.1 or grows to 1.0 or more, the value loses a little over three bits of resolution on one end but gains three on the other.

The Protools HD card DSP mixer uses a 48 bit internal resolution with a 122 bit mixbus, according to the white paper. Reaper runs at this 53-bit momentary resolution at all times with no need for headroom. That sliding window is of higher resolution than any track mixer of Protools ever was. They found a very good compromise for their integer mixer, and Reaper has found the best for native CPU, which btw Protools 9 also uses if the statements on it are accurate.

The limitation if any are the plugins, not the mix paths.

A lot of plugins truncate to 32-bit float on input and send out 32-bit float results, yet somehow it makes no difference.

Let me quote myself here. Only double-blind ABX tests have even a shred of credibility.
__________________
Dialogue/FX Editor & Re-Recording Mixer
Using Latch Preview
"My ego comes pre-shrunk" - Randy Thom
airon is offline   Reply With Quote
Old 08-11-2011, 07:51 AM   #7
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

Quote:
Originally Posted by airon View Post
Only double-blind ABX tests have even a shred of credibility.
Amen Airon. I've only done abx personally.
timlloyd is offline   Reply With Quote
Old 08-11-2011, 08:06 AM   #8
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Yes indeed. But I'm not debating that 64bit float is a good choice, or higher resolution. It is. Totally with you on that.

Thing is, there is something about the actual sound of the 39bit int that I quite like.

I have no idea what that is, or even why I like it, but it's working for me.

Thing is, I might stay with FP64 for another project, or even another track in this project. It's lovely to have a choice. I'm just curious what a higher int resolution would bring.

But alternatively, not so curious I'm going to lose any sleep over it.
griz lee is offline   Reply With Quote
Old 08-11-2011, 08:22 AM   #9
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

Quote:
Originally Posted by griz lee View Post
I'm just curious what a higher int resolution would bring.
Simply less quantisation distortion. The thing is, there is a point at which higher bit-depth will stop correlating with a perceived audible difference between it and lower bit-depths. I suggest doing some more abx listening between 64fp and 39int mode. Like I said, with the source material I used, I could not differentiate between them at all, and they also nulled. Just from this experience, I feel compelled to ask how you did your listening tests? It's very relevant.

That's my experience so far, and if I have time I'll test some more. It's painfully important though, that these test be done properly as Airon mentions if they are to be valid.
timlloyd is offline   Reply With Quote
Old 08-11-2011, 08:30 AM   #10
Xenakios
Human being with feelings
 
Xenakios's Avatar
 
Join Date: Feb 2007
Location: Oulu, Finland
Posts: 6,562
Default

Quote:
Originally Posted by vinx View Post
Human hearing is way more complex than "90dB+ 20 to 20000Hz"...
Yes it is. For example in a way that you hear what you believe you hear.
__________________
For info on SWS Reaper extension plugin (including Xenakios' previous extension/actions) :
http://www.sws-extension.org/
https://github.com/Jeff0S/sws
--
Xenakios blog (about HourGlass and λ) :
http://xenakios.wordpress.com/
Xenakios is online now   Reply With Quote
Old 08-11-2011, 08:32 AM   #11
airon
Human being with feelings
 
airon's Avatar
 
Join Date: Aug 2006
Location: City
Posts: 9,650
Default

Expectations are a powerful influence.
__________________
Dialogue/FX Editor & Re-Recording Mixer
Using Latch Preview
"My ego comes pre-shrunk" - Randy Thom
airon is offline   Reply With Quote
Old 08-11-2011, 08:34 AM   #12
DuraMorte
Human being with feelings
 
Join Date: Jun 2010
Location: In your compressor, making coffee.
Posts: 1,130
Default

Quote:
Originally Posted by vinx View Post
Human hearing is way more complex than "90dB+ 20 to 20000Hz".
That's not true... Human hearing is exactly that.
However, human perception is far more complex than that.
Which is why double-blind testing will always be the great equalizer; it removes bias, and allows for accurate comparisons to be made.
__________________
To a man with a hammer, every problem looks like a nail. - yep
There are various ways to skin a cat :D - EvilDragon
DuraMorte is offline   Reply With Quote
Old 08-11-2011, 08:46 AM   #13
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Quote:
Originally Posted by timlloyd View Post
Simply less quantisation distortion. The thing is, there is a point at which higher bit-depth will stop correlating with a perceived audible difference between it and lower bit-depths. I suggest doing some more abx listening between 64fp and 39int mode.
I suspect that might be the case as you climb higher in bit resolutions. However, I'm not doing any 'tests', and I'm not going to do any abx whatsoever. I'm just flipping the slider and having a listen.

Before purists bake their noodles with me, I'll first need to explain that I've spent quite a few years trying to convince various firms to make plugins which actually process doubles for all the reasons you point out. I've given up doing this, as many don't understand why 32-bit fp isnt enough when you want a 'real' 24 bit result out the end.

I would be highly worried about all this if we didn't have the 64bit FP option on the mix bus. But, we do, and I'm past worrying if I'm psychologically deluding myself.

Lastly, I'm also not intending to tell anyone what bit depth to mix at, as ultimately, this is a matter of total personal choice - just like dither. I used a less revealing dither on a record which had some shocking playing on it. I felt it did some good. Powr-3 just made thing way worse as the acoustic responded in the wrong way, accentuating some problems. I also have put cds out without any dither what so ever. Nobody wrote and complained - quite the opposite in fact! Better engingeering isn't always the answer, though it is essential.
griz lee is offline   Reply With Quote
Old 08-11-2011, 09:09 AM   #14
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

Quote:
Originally Posted by griz lee View Post
However, I'm not doing any 'tests', and I'm not going to do any abx whatsoever. I'm just flipping the slider and having a listen.
Then with all due respect, because of what I know about psychophysics I cannot logically interpret your preferences as anything more than potentially inaccurate and probably biased anecdote.

I really am sorry if this offends you. I have said the same sort of thing to other people and will continue to do so.

This isn't about baking noodles, it's about adherence to the scientific method in light of certain realities of sensory perception.
timlloyd is offline   Reply With Quote
Old 08-11-2011, 09:20 AM   #15
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Quote:
Originally Posted by timlloyd View Post
Then with all due respect, because of what I know about psychophysics I cannot logically interpret your preferences as anything more than potentially inaccurate and probably biased anecdote.

I really am sorry if this offends you. I have said the same sort of thing to other people and will continue to do so.

This isn't about baking noodles, it's about adherence to the scientific method in light of certain realities of sensory perception.
No, it certainly doesnt offend me! Particularly, as I've moved onto the next track of the afternoon, and am prefering 64bit FP on this one, due probably to the baggage of inaccurate science and personal anecdote that is my auditory perception.

I'm just simply curious about different options for different material. I agree with you about doing the science right.
griz lee is offline   Reply With Quote
Old 08-11-2011, 09:23 AM   #16
JasonSpatola
Human being with feelings
 
Join Date: Aug 2009
Posts: 257
Default

Griz - I'd still like some files of what you're hearing, and the project file with a portion of same if you have the time.
JasonSpatola is offline   Reply With Quote
Old 08-11-2011, 09:42 AM   #17
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

It would be interesting if Zee Wavesurfer joined in this discussion, as iirc he/she was one of the people requesting the modes in the first place.

Quote:
Originally Posted by griz lee View Post
No, it certainly doesnt offend me!
I'm glad, that's very refreshing
timlloyd is offline   Reply With Quote
Old 08-11-2011, 09:46 AM   #18
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Jason, sorry mate, I can't give out this material. But there again, I'm not really trying to prove a particuar point as to what is 'best'. I'm quite happy to be sitting in front of a daw that actually gives me some kind of choice, and would simply like to see that choice cover more options if it's a simple thing to impliament. I'm also willing to accept if we did get a 56bit int mode or a 122 or whatever bit mode I'd not be able to hear it. But I don't know that now.

Having a higher spec on the ints may just end all these arguments once and for all, as people could simply do what they felt was right for the project.

No other daw offers that.
griz lee is offline   Reply With Quote
Old 08-11-2011, 09:52 AM   #19
JasonSpatola
Human being with feelings
 
Join Date: Aug 2009
Posts: 257
Default

Quote:
Originally Posted by griz lee View Post
Jason, sorry mate, I can't give out this material. But there again, I'm not really trying to prove a particuar point as to what is 'best'.
Not trying to harp, but is it the only material you have that exhibits this particular effect? Just trying to find something to sink my ears into to observe/understand/recreate this phenomenon.
JasonSpatola is offline   Reply With Quote
Old 08-11-2011, 10:24 AM   #20
buckman
Human being with feelings
 
buckman's Avatar
 
Join Date: Jul 2008
Posts: 1,536
Default

Sorry but i am desperately trying to find this float point mix engine thing, and think i have seen a screenshot on here, but i've looked in all the usual places in v4.01 but can't find it?

Anyone?
__________________
==
Reaper 4 needs Area Selection!
http://forum.cockos.com/project.php?issueid=122
buckman is offline   Reply With Quote
Old 08-11-2011, 10:56 AM   #21
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

It's in 4.02pre2 under the Advanced tab in the new Project Settings dialog.
timlloyd is offline   Reply With Quote
Old 08-11-2011, 11:00 AM   #22
buckman
Human being with feelings
 
buckman's Avatar
 
Join Date: Jul 2008
Posts: 1,536
Default

That will be why i can't find it then! do'h hehe
__________________
==
Reaper 4 needs Area Selection!
http://forum.cockos.com/project.php?issueid=122
buckman is offline   Reply With Quote
Old 08-11-2011, 05:23 PM   #23
musicbynumbers
Human being with feelings
 
musicbynumbers's Avatar
 
Join Date: Jun 2009
Location: brighton, uk
Posts: 12,290
Default

If something nulls (aka silence) to >120dB and you can hear the difference over computer fans and/or even just general room ambience I'm amazed and would love to see you do this on a double blind test.

Not only for your own benefit but frankly to aid in educating people (and not in a subjective way) properly on a topic which my subjective opinion definitely tells me needs "more nails in it's coffin".
musicbynumbers is online now   Reply With Quote
Old 08-11-2011, 09:01 PM   #24
pj geerlings
Human being with feelings
 
pj geerlings's Avatar
 
Join Date: Aug 2009
Location: SoCal
Posts: 40
Default

FWIW, I've always kind'a thought 32-bit *fixed point* summing is plenty

If you let 2^26 equal 1 volt ...

32-bits can represent (nearly) a +-32 volt peak signal
-- which is hotter than virtually any analog gear out there

and have s/n of better than 156 dB
-- which is better than *literally* any analog gear out there

peace y'all
pj

Last edited by pj geerlings; 08-11-2011 at 10:19 PM. Reason: left out the "fixed point"
pj geerlings is offline   Reply With Quote
Old 08-11-2011, 09:37 PM   #25
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 18,829
Default

Quote:
Originally Posted by airon View Post
Let me quote myself here. Only double-blind ABX tests have even a shred of credibility.
I might be willing to poke one of my eyes out to test but not both.
__________________
Your whole life people will tell you what you can't do. Getting past them is the first step to actually getting things done.
karbomusic is online now   Reply With Quote
Old 08-11-2011, 09:59 PM   #26
alanofoz
Human being with feelings
 
alanofoz's Avatar
 
Join Date: Sep 2009
Location: Oz - Blue Mountains NSW, formerly Geelong
Posts: 612
Default

Actually, if you take the trouble to really learn about 32 bit floating point you'll find that it's capable of representing a dynamic range in excess of 1500 dB.

I'm on the road at the moment with only an iPad, so I won't be in a position to back that up for a week or so, but I've created such a file in the past to prove the point. Of course, that's orders of magnitude more than any practical signal requires. (Do we really need 64 bit?)

If you like, google for IEEE floating point format. I was dealing with this in the early '80s, long before the format was incorporated into the wave file specification.
alanofoz is online now   Reply With Quote
Old 08-11-2011, 11:17 PM   #27
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,434
Default

Quote:
Originally Posted by griz lee View Post
I'm past worrying if I'm psychologically deluding myself.
But we still now you are indeed.

(which in my humble opinion we can't stress enough, in order to stop you from also deluding some beginners who otherwise might fall for the nonsense you are posting here)



ABX or STFU!
jens is offline   Reply With Quote
Old 08-11-2011, 11:23 PM   #28
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,434
Default

Now onto something more interesting than some random subjective interpretation of sound-quality:

I was always desiring 32bit FP, so that Reaper doesn't uselessly burn precious CP-cycles, but now that it's suddenly here I fail to notice any significant difference - same with 49bit integer b.t.w. .
jens is offline   Reply With Quote
Old 08-11-2011, 11:47 PM   #29
robo
Human being with feelings
 
robo's Avatar
 
Join Date: May 2006
Location: On A Mountain Top
Posts: 342
Default

I really can't tell between 16bit int and up through 64bit fp.

12bit int is hissy but that's ~72db noise floor anyway.

I say Hong Kong Fooey. I made renders of 24, 39 and 64fp and while they didn't null and it's not a real ABX test, still didn't notice anything. I really wanted to and I'm worrying about suddenly becoming so honest with myself.

Huddling in a corner,
robo
robo is offline   Reply With Quote
Old 08-12-2011, 12:03 AM   #30
hihat
Human being with feelings
 
hihat's Avatar
 
Join Date: Aug 2010
Posts: 119
Default

nobody knows what original signal is. since no way to translate it 100% original. the narrowest place is still monitors even if they are thousands usd dynadam or jblec. woofers and tweeters are not 100%synchronized also..
as i understand that modes like integer or float only presentation of audio,how data is stored. all plugin internal processing the same.
i would even go 24int for final 16bit and 32fp for 24. but unfortunately that modes doesnt bring system performance boost probably because reaper was optimized for long time for 64fp.
hihat is offline   Reply With Quote
Old 08-12-2011, 12:30 AM   #31
Xenakios
Human being with feelings
 
Xenakios's Avatar
 
Join Date: Feb 2007
Location: Oulu, Finland
Posts: 6,562
Default

Quote:
Originally Posted by jens View Post
I was always desiring 32bit FP, so that Reaper doesn't uselessly burn precious CP-cycles, but now that it's suddenly here I fail to notice any significant difference - same with 49bit integer b.t.w. .
If I've understood this new mixing engine thing correctly, the new calculation methods only happen for the summing point of the mixer. Everything else is still 64 bit floats. (Mixing a bunch of audio buffers together is a trivial operation and won't account for much CPU in any case.)

The Reaper devs have also said(*) that they've benchmarked a build using 32 bit floats everywhere against the public build which uses 64 bit floats, and found no performance difference in favor of 32 bit floats. If you are pressed for performance, maybe look into some new computer hardware...?

(*) Not sure if that was said anywhere here on the forum, could have been at IRC chat.
__________________
For info on SWS Reaper extension plugin (including Xenakios' previous extension/actions) :
http://www.sws-extension.org/
https://github.com/Jeff0S/sws
--
Xenakios blog (about HourGlass and λ) :
http://xenakios.wordpress.com/
Xenakios is online now   Reply With Quote
Old 08-12-2011, 12:58 AM   #32
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Wierdly, I didn't ask for a science lesson in my original post. I wondered simply if anyone prefered the sound of int on their material. So far no comments on sound!

It's never nonsense to take a call on what you think makes a subjective better product. And how you feel about a mix in progress is important. I find I make better decisions using some equipment not simply because it's sonically 'better'.

But it's also true that to make objective sound quality comparisons you need to abx.

The two are related, but not for the same reaons.

This has been a great thread, but perhaps mods one that should perhaps continue in the lounge if anyone wants to take it further. Personally I'll stfu and get my (lab) coat.
griz lee is offline   Reply With Quote
Old 08-12-2011, 02:15 AM   #33
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,434
Default

Quote:
Originally Posted by griz lee View Post
It's never nonsense to take a call on what you think makes a subjective better product. And how you feel about a mix in progress is important. I find I make better decisions using some equipment not simply because it's sonically 'better'.

Yes and no - if a gorgeous GUI inspires you, then what you 'feel' might be an influential factor. But if you 'feel' a certain mathematical formula sounds better because of certain associations and memories the name of this formula triggers in your brain, then no, what you 'feel' should be irrelevant.
Or in what way do you think it might influence your desicions regarding the mixing task?
jens is offline   Reply With Quote
Old 08-12-2011, 02:18 AM   #34
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,434
Default

Quote:
Originally Posted by Xenakios View Post
If I've understood this new mixing engine thing correctly, the new calculation methods only happen for the summing point of the mixer. Everything else is still 64 bit floats. (Mixing a bunch of audio buffers together is a trivial operation and won't account for much CPU in any case.)

The Reaper devs have also said(*) that they've benchmarked a build using 32 bit floats everywhere against the public build which uses 64 bit floats, and found no performance difference in favor of 32 bit floats. If you are pressed for performance, maybe look into some new computer hardware...?

(*) Not sure if that was said anywhere here on the forum, could have been at IRC chat.

Thanks for that bit of info.

Looking into some new hardware is not so simple in my case, since I'm on laptop. I'll eventually buy a new one, but until then It'll more or less stay the way it is. However as it's not exactly old crap I'm relatively well sorted.
jens is offline   Reply With Quote
Old 08-12-2011, 04:53 AM   #35
musicbynumbers
Human being with feelings
 
musicbynumbers's Avatar
 
Join Date: Jun 2009
Location: brighton, uk
Posts: 12,290
Default

I often do a simple A/B comparison with my students where I show them two compressors both set up with same parameters from two different programmers, one looking "all analogue" and the other looking a bit shitty and ask them which sounds better as we switch between them.

They always say the "one with the big shiny GUI knobs"

I then let them know that I in fact did not even switch between them.

Now this is a little dirty and not scientific but an important part of mixing is that the person listening to your mix at home has NO visual context at all so believing that a mix sounds better simply because all the channels have a big fat neve eq/comp on them is actually much easier to do then people like to admit.

It really really does effect the way we perceive the sound of our mix.

I have noticed this when doing music video editing too, even changing the video output from full colour to black and white changes perception and so does lowering the resolution to an extent (highly subjective though.)

The mckirk effect http://en.wikipedia.org/wiki/McGurk_effect shows simply the link between visual perception and sound.
musicbynumbers is online now   Reply With Quote
Old 08-12-2011, 05:04 AM   #36
timlloyd
Human being with feelings
 
Join Date: Mar 2010
Posts: 4,713
Default

McGurk ... the McKirk effect is what used to happen to women when Shatner walked into a room ...
timlloyd is offline   Reply With Quote
Old 08-12-2011, 05:51 AM   #37
vinx
Human being with feelings
 
Join Date: May 2007
Location: France
Posts: 662
Default

Quote:
Originally Posted by DuraMorte View Post
That's not true... Human hearing is exactly that.
However, human perception is far more complex than that.
Which is why double-blind testing will always be the great equalizer; it removes bias, and allows for accurate comparisons to be made.
That's not true, human hearing is not exactly that : "enter any numbers"
Ears can not be so easily computed as you say.
But ok for double-blind testing as the great equalizer!

And if "perception" artifacts can sometimes fool you, as a counterpart what seems to be at first sight insignificant differences can truely be heard.

Last edited by vinx; 08-12-2011 at 05:57 AM.
vinx is offline   Reply With Quote
Old 08-12-2011, 06:49 AM   #38
Subz
Human being with feelings
 
Subz's Avatar
 
Join Date: Jun 2006
Location: UK
Posts: 2,919
Default

Quote:
Originally Posted by DuraMorte View Post
That's not true... Human hearing is exactly that.
However, human perception is far more complex than that.
Which is why double-blind testing will always be the great equalizer; it removes bias, and allows for accurate comparisons to be made.

i don't only use my ears to listen!

but i agree "Perception" plays a massive factor in the inconsistencies of personal opinion,

this thread has me curious! even though i'm very happy with my current default setting!

Subz

Last edited by Subz; 08-13-2011 at 03:58 AM.
Subz is offline   Reply With Quote
Old 08-12-2011, 07:27 AM   #39
XITE-1/4LIVE
Human being with feelings
 
XITE-1/4LIVE's Avatar
 
Join Date: Nov 2008
Location: Somewhere Between 120 and 150 BPM
Posts: 5,511
Default

I have been using an ancient HDD MultiTrack recorder from Scope DSP platform, now in x64.
I have always noticed it's 32bit Integer sound was better than the quality I got from Cubase or Reaper, but, once transferred from the project lost it's magical quality.
But I would choose this for sync'd background vocals or segues for live work as I noticed the sound quality and I always trust my ears.
But in the last few years that ancient app has lost it's edge, and more recently Reaper AB'd next to VDAT ( 32bit Integer ) has no difference.
Made a nice Recording using no time based effects, just some Panning for seperation and Reaper has now replaced my beloved VDAT.

I was using Reaper for stable MIDI playback sequencer only, but more and more have begun to incorporate it as the entire audio/MIDI solution.

Looks like VDAT, Alesis HD24/96 and Yamaha hardware sequencer can finally be retired.

Here's a simple pre recorded MIDI drum track, no FX. And Solaris hardware synth playing BitCrusher Bass, and Waldorf Wave emulation.
THe quality is excellent.

http://soundcloud.com/jimmyvee/wormhole

ooops, wrong link....scroll to Distorted Bass/Dual Ribbon controller

Sorry, I used the 64bit float thingamujig..
__________________
iPad with killer synths

Last edited by XITE-1/4LIVE; 08-12-2011 at 07:33 AM. Reason: Need Sleep
XITE-1/4LIVE is offline   Reply With Quote
Old 08-12-2011, 05:58 PM   #40
robo
Human being with feelings
 
robo's Avatar
 
Join Date: May 2006
Location: On A Mountain Top
Posts: 342
Default

In the vein of XITE's observations I too swear there was a huge sound difference between cubase 3x and vegas in the old days with asio vs. dx drivers.

Maybe I'm just deaf now but I still can't hear anything (or maybe refuse to) with reaper's summing bit depths besides 8 bits and 12 bits.

-robo
robo is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 08:36 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2017, vBulletin Solutions Inc.