https://www.youtube.com/watch?v=iLwAjOGQcHoUnaspected wrote:Everything enroe has written here correlates with my understanding.
DAW development stopped ?
-
- Banned
- 29 posts since 31 Jul, 2018
-
- KVRAF
- 3089 posts since 4 May, 2012
Very good.marooned ufo wrote:https://www.youtube.com/watch?v=iLwAjOGQcHoUnaspected wrote:Everything enroe has written here correlates with my understanding.
-
- Banned
- 83 posts since 25 Jul, 2018
-
- KVRAF
- 2590 posts since 19 Mar, 2008 from germany
Good video, because they honestly describe the differencelearnkeys wrote:https://www.youtube.com/watch?v=CpbbljcizsA
between 32-bit and 64-bit audio processing.
And Steinberg concedes: "It doesn't have anything to do
with audio-dynamics" - because - as we all know - 32-bit
has enough dynamics already.
The only (!) reason for 64-bit audio processing is to avoid
rounding errors.
Normally the rounding errors in a pure 32-bit engine are
so tiny small that they don't matter in the practical experience.
Hypothetically there can be audible rounding aberrations
if you use huge amounts of plugins in series ... so everyone
must decide for himself if he could benefit form a 64-bit
audio-engine.
Now the 64-bit setting in cubase is only an "option": You can
choose it - or you can stay with the well proven 32-bit setting.
The option is a good thing since everybody can decide what
precision he needs.
I estimate that nearly no musician would benefit from the
64-bit setting, because they won't perceive any difference, but
their CPU would be stressed much more.
free mp3s + info: andy-enroe.de songs + weird stuff: enroe.de
-
- Banned
- 83 posts since 25 Jul, 2018
If there wasn't a benefit it wouldn't exist in several major DAWs. It's not fairy dust,enroe wrote:I estimate that nearly no musician would benefit from the
64-bit setting, because they won't perceive any difference
it's simple math. There's a difference in the sound and performance of large projects, fact.
If you have the cpu power to support it, it can increase the precision and accuracy without the need of truncation. Which simply means you'll get a more pure sound with clarity and transparency.
I don't know why it's so hard for you to admit you've been completely wrong, now let's move on, it's a waste of everyone's time to keep replying to your nonsense.
- Banned
- 11467 posts since 4 Jan, 2017 from Warsaw, Poland
That's the problem right there - if something is more accurately calculated it just means ...it's more mathematically correct. Nothing more, nothing less. It definitely doesn't mean it will sound more pleasing, more harmonious, or even clearer or transparent. Just like some music or instruments sounds better in lossy formats or processed with saturation or distortion.learnkeys wrote:If you have the cpu power to support it, it can increase the precision and accuracy without the need of truncation. Which simply means you'll get a more pure sound with clarity and transparency.
Obviously, this same argument can't be used comparing 8 vs. 16 bit or 16 vs. 24 bit, but beyond 32 bit (float) it really is the case of diminishing returns.
-
- Banned
- 83 posts since 25 Jul, 2018
The average bedroom hobbyist probably won't benefit much from it, you're right.
But, for those that work in professional, high track count projects that require pristine audio will benefit.
A lot of it is the subject matter you're dealing with. Most people making hiphop, rock, pop and other genres that don't need clarity and transparency may not benefit as much.
The same goes for 44.1k users that look at 48k, 88k, etc the same way. It probably won't make your music sound any better, or you don't have the ears to hear the difference. But, that doesn't mean people don't work on classical, scores, low level audio, sound design and other sounds and music that benefit from it.
If it doesn't appeal to you, that's fine, but to write it off for everyone is just plain nonsense.
But, for those that work in professional, high track count projects that require pristine audio will benefit.
A lot of it is the subject matter you're dealing with. Most people making hiphop, rock, pop and other genres that don't need clarity and transparency may not benefit as much.
The same goes for 44.1k users that look at 48k, 88k, etc the same way. It probably won't make your music sound any better, or you don't have the ears to hear the difference. But, that doesn't mean people don't work on classical, scores, low level audio, sound design and other sounds and music that benefit from it.
If it doesn't appeal to you, that's fine, but to write it off for everyone is just plain nonsense.
-
- KVRAF
- 2945 posts since 23 Dec, 2002
The CPU hit is very small to run 64 bit precision in Cubase. It is about 2% to 3% maximum on my I7 3930K hexa core rig. I have verified with other users at various Cubase Facebook forums and they are getting similar numbers ( up to 5% hit on slower computers). The CPU hit is very small.
I can't hear the improvement on my tracks but given that the overhead is so low perhaps someone else can hear some subtle improvements - so I have it turned on.
I can't hear the improvement on my tracks but given that the overhead is so low perhaps someone else can hear some subtle improvements - so I have it turned on.
-
- KVRAF
- 2590 posts since 19 Mar, 2008 from germany
Considering your understanding of audio in this thread I have the clearlearnkeys wrote:The average bedroom hobbyist probably won't benefit much from it, you're right.
impression that you are one of the "average bedroom hobbyists". But
that's not a bad thing, because mostly hobbyists have a lot of fun in
what they are doing ...
It's just the opposite:learnkeys wrote: But, for those that work in professional, high track count projects that require pristine audio will benefit.
Those who work professional - with high track counts - will run into
CPU-overload very soon - and thus it is crucial to have a very perfomant
system. So they'll almost never do the 64-bit audio switch.
free mp3s + info: andy-enroe.de songs + weird stuff: enroe.de
-
- KVRAF
- 2590 posts since 19 Mar, 2008 from germany
Poooh, that is pretty high! That are 50% more CPU-stress:Scotty wrote:The CPU hit is very small to run 64 bit precision in Cubase. It is about 2% to 3% maximum on my I7 3930K hexa core rig.
From 2% to 3% are 50% more!
Guess you have a big project with 200 tracks - and your
CPU hit is at 70%.
On your system the switch to 64-bit audio would increase
the CPU-load up to 105%, which means overload - and your
audio output would just stop.
-------------------------------------------------------------------------
I guess on more modern systems (kaby-lake or coffee-lake)
with 64-bit audio there'll be only a slight increase in CPU-
load. But sometimes you have big projects - and even
a small increase in CPU-load is too much.
free mp3s + info: andy-enroe.de songs + weird stuff: enroe.de
- KVRAF
- 23471 posts since 12 Jul, 2003 from West Caprazumia
You don't have much of a clue what you are talking about I'd say. The amount of rounding errors doesn't have much to do with the size of the project but rather with the number of subsequent renders on the same file(s). The more often you destructively edit the same file, the more rounding errors will start to creep in. If all your editing is done non-destructively, 64bit will yield no advantage in regards to rounding errors, regardless of the size of your project.learnkeys wrote:If there wasn't a benefit it wouldn't exist in several major DAWs. It's not fairy dust,enroe wrote:I estimate that nearly no musician would benefit from the
64-bit setting, because they won't perceive any difference
it's simple math. There's a difference in the sound and performance of large projects, fact.
"Preamps have literally one job: when you turn up the gain, it gets louder." Jamcat, talking about presmp-emulation plugins.
-
- KVRAF
- 2945 posts since 23 Dec, 2002
That 2% was on a project that was pushing ASIO to 85% ... the ASIO on my machine would jump to 87% to 89%.... just test it. It is like running one or two additional plugins on the project that is all... the CPU drain is minimal. IF you are pushing your computer to the limits everything matters... I understand that point. But this is never going to up your ASIO load from 70% to 105% - not in my testing or those of others with whom I corresponded with when this was made available. There is no need to speculate... load a big project... turn it on and see what happens to your ASIO load.... the answer will be not much.
enroe wrote:Poooh, that is pretty high! That are 50% more CPU-stress:Scotty wrote:The CPU hit is very small to run 64 bit precision in Cubase. It is about 2% to 3% maximum on my I7 3930K hexa core rig.
From 2% to 3% are 50% more!
Guess you have a big project with 200 tracks - and your
CPU hit is at 70%.
On your system the switch to 64-bit audio would increase
the CPU-load up to 105%, which means overload - and your
audio output would just stop.
-------------------------------------------------------------------------
I guess on more modern systems (kaby-lake or coffee-lake)
with 64-bit audio there'll be only a slight increase in CPU-
load. But sometimes you have big projects - and even
a small increase in CPU-load is too much.
-
- Banned
- 83 posts since 25 Jul, 2018
enroe wrote:Considering your understanding of audio in this thread I have the clear
impression that you are one of the "average bedroom hobbyists".
I'll let you do the talking from now on.enroe wrote:Advantage:
1. For plugins with 64-bit-depth dynamic the
conversions can be left out.
-
- Banned
- 83 posts since 25 Jul, 2018
I think you're confusing the information, like most of the uneducated opinions displayed here.jens wrote:You don't have much of a clue what you are talking about I'd say. The amount of rounding errors doesn't have much to do with the size of the project but rather with the number of subsequent renders on the same file(s). The more often you destructively edit the same file, the more rounding errors will start to creep in. If all your editing is done non-destructively, 64bit will yield no advantage in regards to rounding errors, regardless of the size of your project.
The amount of truncations, calculations and including the up/down sampling between multiple plugins that are 32 bit and 64 bit, used on inserts and busses, etc goes down when everything is at 64 bit precision. It's simple math, let's not try to complicate things with your need to disprove anything here.