Pro Tools 10HD How does it deal with AAX

[JuanPC]

Originally Posted by wolfman_rob

I was curious how the AAX versions of plug ins compare to RTAS counterparts

wonder if there is any difference in latency and cpu usage

I guess if i was on 10 at least i would be able to use all of my plugins and have aax also so all would be good for me if 10 is good and stable

RTAS sounds like it has a compressor, lacks dynamic range, specially in the highs, more audible in PT9...
Plugins like Focusrite D2/D3 AAX sound more dynamic, "better." in PT10.
of course you need a transparent enough DA converter & amplifiers/loudspeakers to hear that.

PT9 & PT10 are both 32-Bit app.
most AAX plugins are more CPU Optimized for some reason... specially the 64-bit version PT11&12.
can insert more.

for example: WaveArts Tube Saturator v1.33 AAX32 vs. RTAS is "similar"
same plugin count... 44x stereo channels PTHD at 44.1Khz 1024 buffer in i7 970/x5675 at 3.2GHz HT-On. 1-Thread reserved for system.
download the Installer, and run in demo mode.

but the new TubeSaturator v2.02 AAX64 sounds better, has more features, and can insert a lot more plugins, over 128 stereo channels, 1 stereo channel empty for monitor input or Signal Generator Tone/Noise.
to hear there is no artifacts in the audio.


so i guess it depends on the plugin manufacturer...
but AAX has the potential, to sound better and have more plugins.

if you want to do the test your self...
download the demo of PTHD10/11
http://connect.avid.com/Pro-Tools-Trial.html
http://connect.avid.com/Audio-Plug-i...uginTrialName=

If PT10 detects both AAX32 & RTAS installed of the same plugin, when you start PT, PT will prefer AAX32 over RTAS version.

you must manually move to other/unused folder the AAX32 version, if you want to compare AAX vs. RTAS, Quit first, move & start PT again each time.
to be absolutely sure is recommended to also move the RTAS to unused folder, if want to compare the AAX.
you only have to open the same .ptf or .ptx session.

folder path are different...

if Windows 32-bit:
C:\Program Files\Common Files\Digidesign\DAE\Plug-Ins
C:\Program Files\Common Files\Digidesign\DAE\Plug-Ins (Unused)
C:\Program Files\Common Files\Avid\Audio\Plug-Ins
C:\Program Files\Common Files\Avid\Audio\Plug-Ins (Unused)


if Windows 64-bit: because PT10 is 32-bit, AAX 32-bit path is different.

C:\Program Files (x86)\Common Files\Digidesign\DAE\Plug-Ins
C:\Program Files (x86)\Common Files\Digidesign\DAE\Plug-Ins (Unused)
C:\Program Files (x86)\Common Files\Avid\Audio\Plug-Ins
C:\Program Files (x86)\Common Files\Avid\Audio\Plug-Ins (Unused)


also mixing in PT9 is a lot more difficult because signal path is limited to 24-bits "clips/distorts over 0dB" even it has 48-bit oversampling mixer audio path still is 24-bit,
32-bit & 64-bit file and signal path/mixer does not clip internally if goes over 0dB.
but some plugins are designed to soft-saturate if goes neat 0dB, like real analog, for example Softube Valley People Dyna-myte.


if Windows 64-bit & PT 64-Bit & AAX64:
C:\Program Files\Common Files\Avid\Audio\Plug-Ins
C:\Program Files\Common Files\Avid\Audio\Plug-Ins (Unused)

[WernerF]

AAX sounding better than RTAS should be taken with a quite large grain of salt. This is not at all my experience BTW.

[JuanPC]

Originally Posted by WernerF

AAX sounding better than RTAS should be taken with a quite large grain of salt. This is not at all my experience BTW.

#1.
im talking about Playback and monitor input...only, no bounce.
PT10 bounce is inferior/different than PT9,11or12.

#2.
i´ve tested TDM vs. RTAS vs. AAX,
OSX vs. Windows,
PT9 vs. 10 vs. 11 vs. 12
the same plugin, with same settings, sounds different.

some examples:
Sonnox EQ for TDM was awesome, for RTAS crap in OSX.
Focusrite Forté Suite TDM vs. RTAS was strange, sometimes i liked more the RTAS.
Focusrite D2/D3, OSX TDM was better than RTAS for OSX, but... RTAS for Windows was as good as OSX TDM.
McDSP Channel G for TDM was awesome. CompressorBank also.
PspAudioware MasterQ, VST wrapped as RTAS vs. RTAS sounds different.
the VST´s wrapped with Fxpansion sounded more transparent, less vintage.

#3. maybe you like the compressed sound of RTAS, with AAX some eq plugins need more compression or limiting to sound as loud as RTAS..

changes must be made, same happens when moving from 44.1KHz to 96Khz.
same EQ settings do not work. 44.11KHz is brighter.

#4. same PTHD9.0.6 Windows7 vs. XP sp3... W7 was better.
Win7 32-bit vs. W7x64bit there was a small difference recording files.

the bit bridge that converts the 64-bit audio drivers to 32/24-bit app/software did change the sound a bit, Lynx audio drivers had some kind of noise shaping.. difference was the smallest.

[WernerF]

I hate to say this but I would submit that there might be a lot of expectation bias going on here.

[wolfman_rob]

Thanks Guys

[crizdee]

Hi,

I've been using 100's of plugs for over a decade that are both TDM and RTAS and I've never found a difference in sound when levels are set correctly! Only if you push a level too hot into TDM it will clip before the RTAS version which is 32bit floating point, the TDM being 48 bit fixed. Only fools run their mixing levels that hot

I've found only a few plugs where the AAX version will be slightly different when switched between TDM and AAX32. Not better or worse sounding!!! Just slightly different when you switch between the 2 types.

If anyone is hearing a difference then it's a null test or it doesn't happen.


Chris

[WernerF]

Yup, a null test for sure. Totally agree.

[JuanPC]

a null test to -infinite.
anything else means it´s different...Test Needs a Meter Plugin that allows to see all the 24-Bits/ -144dB range.

for example Table Salt "NaCL", usually comes in 1gr. envelopes.
1gr. for 500gr. of Beans+Rise.
is 1/500 and most tongues can detect that difference like a BIG Difference.
some people is more or less sensible than others, in sugar, or spice department.

other example:
some people when they eat spicy food, they see better, everything becomes brighter.
that does not happen to me, but when i eat protein i can smell better. if i eat sugar does not work, if i eat enough Milo(TM) i can smell 100x better, smelling that much becomes annoying.
if i don´t eat for a long time +/-20hours, i become smell deaf.


Also some people eating certain medications become Ear Deaf.
diuretics, anti-inflammatory drugs, aspirin, ibuprofin, naproxin, celecoxib, diclofenac, acetaminophen, quinine, antibiotics, some erectile dysfunction drugs, etc...
https://en.wikipedia.org/wiki/Hearing_loss

But...
a Null Test could not reveal all the differences if done ITB, unless is done with 2 different PC´s ...
at same time synched with MTC & Wordclock.
a True Null Test.

Adat or Toslink to be fair.
s/pdif cable could affect result. has too low voltage 0.5v needs to be triple-shielded.

¿Why a Null Test could not reveal all the differences if done ITB,?
when i tested Logic using a plugin called WormHole2, to send each channel through LAN 1Gbps, to ProToolsHD9.0.6 with FXpansion RTAS Adapter.
Logic Mixer sounded different...vs. same Logic synths playing through ProTools mixer, no Logic mixer involved.
PTHD9.0.6 Mixer/engine for me was better/different.
The audio of each channel was interrupted before the Logic Mixer & sent to the ProTools Mixer directly.
Mixer engine algorithms can minimize or null the differences,
because it´s a reduction algorithm.
http://www.vttoth.com/CMS/index.php/technical-notes/68
http://www.voegler.eu/pub/audio/digi...ion.html#idx_8
http://docs.oracle.com/cd/E19957-01/..._goldberg.html
https://www.reddit.com/r/audioengine...ing_algorithm/
http://forum.breakbeat.co.uk/tm.aspx...1&key=�
https://software.intel.com/en-us/art...ewalk-sonar-x1
...
Pro Tools white paper describing the 48 bit mixing engine used since Pro Tools 7 upto 8 maybe 9.
http://akmedia.digidesign.com/suppor...ixer_26688.pdf
...
Also in older PT Bounce vs. Record.
https://www.gearslutz.com/board/gear...rd-better.html
...
https://code.google.com/archive/p/wormhole2/
https://code.google.com/archive/p/wormhole2/downloads
...
http://plasq.com/2007/12/wormhole2-open-sourced/

even ProToolsHD12.3.1 Mixer engine is not 100% bit transparent... but is in the TOP2. Sonar 8.5 64bit engine was the most transparent i´ve heard. i have not tested Harrison Mix Bus, but since they claim to be Analog Emulation, i guess is not transparent.
PT12.3.1 seems more transparent than PTHD9.0.6 but never could make a direct A/B comparison.
i remember PTHD9.0.6 engine was nicer sounding than VSTHost software, the highs were different.
and VSTHost was VERY Transparent with double precision activated at the Input&Output. http://www.hermannseib.com/english/vsthost.htm

when i test A/B, audio coming to the soundcard Input-->DSP Mixer-->DA, vs. same audio through the ASIO driver --> ProToolsHD12.3.1--->ASIO-->DSP Mixer-->DA.

PT does not allow 32 samples of buffer size in Windows @96KHz, and Lynx DSP mixer does not allow to Time Align, i cannot make a true Null test.
DSP "FPGA" Mixers are FIFO "First In First Out", 1or2 samples of delay.

[guitardom]

Originally Posted by JuanPC

a null test to -infinite.
anything else means it´s different...Test Needs a Meter Plugin that allows to see all the 24-Bits/ -144dB range.

for example Table Salt "NaCL", usually comes in 1gr. envelopes.
1gr. for 500gr. of Beans+Rise.
is 1/500 and most tongues can detect that difference like a BIG Difference.

But...
a Null Test could not reveal all the differences if done ITB, unless is done with 2 different PC´s ...
at same time synched with MTC & Wordclock.
a True Null Test.

Adat or Toslink to be fair.
s/pdif cable could affect result. has too low voltage 0.5v needs to be triple-shielded.

¿Why a Null Test could not reveal all the differences if done ITB,?
when i tested Logic using a plugin called WormHole2, to send each channel through LAN 1Gbps, to ProToolsHD9.0.6 with FXpansion RTAS Adapter.
Logic Mixer sounded different... Logic synths playing back through ProTools, no Logic Mixer involved.
PT9HD.06 Mixer/engine was better.
Audio for each channel was interrupted before the Logic Mixer.
the mixer engine algorithm can minimize or null the differences.
https://code.google.com/archive/p/wormhole2/
https://code.google.com/archive/p/wormhole2/downloads
...
http://plasq.com/2007/12/wormhole2-open-sourced/

even ProToolsHD12.3.1 Mixer engine is not 100% bit transparent... but is in the TOP2. Sonar 8.5 64bit engine was the most transparent i´ve heard. i have not tested Harrison Mix Bus, but since they claim to be Analog Emulation, i guess is not transparent.
PT12.3.1 seems more transparent than PTHD9.0.6 but never could make a direct A/B comparison.
i remember PTHD9.0.6 engine was nicer sounding than VSTHost software, the highs were different.
and VSTHost was VERY Transparent with double precision activated at the Input&Output. http://www.hermannseib.com/english/vsthost.htm

when i test A/B, audio coming to the soundcard Input-->DSP Mixer-->DA, and same audio through the ASIO driver --> ProToolsHD12.3.1--->ASIO-->DSP Mixer-->DA.

but since PT does not allow 32 samples of buffer size in Windows, and DSP mixer does not allow to Time Align, i cannot make a Null test.
DSP "FPGA" Mixers are FIFO "First In First Out", 1or2 samples of delay.

The gibberish and bad information you post in threads would be amusing if I didn't think you were serious.

[WernerF]

Some people just have a bunch of time on their hands and the imagination goes wild. I agree though, when it's meant to be serious it's also no longer funny.