Avid Pro Audio Community

Avid Pro Audio Community

How to Join & Post  •  Community Terms of Use  •  Help Us Help You

Knowledge Base Search  •  Community Search  •  Learn & Support


Avid Home Page

Go Back   Avid Pro Audio Community > Pro Tools Software > Pro Tools

Reply
 
Thread Tools Search this Thread Display Modes
  #31  
Old 03-30-2023, 04:01 AM
RobertDorn RobertDorn is offline
Member
 
Join Date: Oct 2008
Location: Netherlands
Posts: 405
Default Re: Tracking with Buffersize 32?

Quote:
Originally Posted by LDS View Post
The devil is really in the detail though. The problem with DSP in 2023 is that most DAWs now offer dual sample buffers for recording and playback as standard. If you use as 64 sample buffer natively in Pro Tools, that doesn't actually mean your playback channels are using a 64 sample buffer too. That is the benefit of playback channels! You can buffer them extensively before playback actually starts.

When it comes to latency, round trip latency is the only thing that actually matters. The native sample buffer is only one aspect of round trip latency. Some food for thought:

- the difference between 32 and 64 sample native buffers is 64 samples of latency.
- HD I/O have an AD/DA calculated latency of 50 samples.
- Lynx Auroras are 25 samples.
- RME are 14 samples.
- A HD Native card as a digital I/O interface has a latency of 82 samples.
- HDX + HD I/O interfaces has a latency of 76 samples before plugins.
- third party AAX-DSP plugins add between 30 and 75 samples per instance.
- many UAD DSP plugins add 55 samples per instance.

There is really nothing more wasteful than talking about latency as something that only exists with sample buffers. This is particularly applicable in an era of DAD/MTRX interfaces which don't adhere to the calculated digilink latency of HD I/O interfaces. Like most things digital, converter latency has improved leaps and bounds compared to the HD I/O latency of 2012 that most digilink interfaces adhere too.
Allright. Perhaps I have to check again now that I'm on an M2 max cpu, actually curious how it works in practice. Last time I compared how it worked in real world practice was when I was on an HD-Native thunderbolt + HD I/O system, tracking at the lowest possible playback buffer in pro tools and working with mix downs in a fresh session. Till I plugged in a guitar into an UAD Arrow (solo nowadays) and thought 'wowww, this feels like just playing through analog gear' . I then swapped my HD native system for some Apollo X racks. But due to the lack of connectivity swapped that later on to HDX with a MTRX , so let's say since 2017 I've only recorded with DSP based systems.

Curious to check out how it'd be nowadays, when tracking just completely natively, especially when overdubbing on big sessions, because I can do that without thinking about the native playback buffer now with DSP, even if the session runs on a native playback buffer of 1024 samples.
__________________
Apple MacBook Pro M2 Max, 96GB ram | Pro Tools HDX | Avid MTRX | Pro Tools Ultimate 2023.12 | macOS 13.6.3
Reply With Quote
  #32  
Old 03-30-2023, 05:43 AM
JFreak's Avatar
JFreak JFreak is offline
Moderator
 
Join Date: Jan 2003
Location: Tampere, Finland
Posts: 24,849
Default Re: Tracking with Buffersize 32?

Quote:
Originally Posted by midnightrambler View Post
Actually we used to get clients in who did complain, back when we got our first ProTools rig (late 90s). Not a huge number, admittedly, but there were definitely some who said it made them uncomfortable.
Was that with or without plugins? I started with MIX3 888|24 in 1998 (rented at the time). It had roundtrip latency of 71 samples at 48k. Without plugins nobody ever complained about that. But the thing with TDM is every plugin adds some amount of latency so the real-world latency builds up sometimes really fast.

God only knows how the artists managed to sing with DA/AD to digital reverb, totalling +142 samples of latency (plus being the internal latency of the reverb).

I still hold my opinion that nobody should have a problem with 128 playback buffer with native systems. And for the few who may always complain, 64 should drop it. 32 setting is just simply unnecessary, unless you want to convince your boss to buy you more powerful computer.
__________________
Janne
What we do in life, echoes in eternity.
Reply With Quote
  #33  
Old 03-30-2023, 05:50 AM
Sardi Sardi is offline
Member
 
Join Date: Aug 2002
Location: Perth
Posts: 2,986
Default Re: Tracking with Buffersize 32?

Quote:
Originally Posted by JFreak View Post
God only knows how the artists managed to sing with DA/AD to digital reverb, totalling +142 samples of latency (plus being the internal latency of the reverb).
Are we STILL doing this?

Now you’re talking about something that is absolutely irrelevant to the discussion. How is latency of an effect like reverb a problem when tracking?

I seriously have no words.


Sent from my iPhone using Tapatalk
Reply With Quote
  #34  
Old 03-30-2023, 06:03 AM
JFreak's Avatar
JFreak JFreak is offline
Moderator
 
Join Date: Jan 2003
Location: Tampere, Finland
Posts: 24,849
Default Re: Tracking with Buffersize 32?

Are we still arguing? Time and time again everything says that the roundtrip latency of superfast TDM systems were higher than that without plugins? And no problems tracking with plugins either.

If the artists were eating magic mushrooms back then, maybe they didn't notice that those MIX/HD systems were crap. But a lot of good recordings were made with those.

If the systems were okay, maybe it is psychological when someone cannot perform with 64 latency. Go figure.
__________________
Janne
What we do in life, echoes in eternity.
Reply With Quote
  #35  
Old 03-30-2023, 06:06 AM
Darryl Ramm Darryl Ramm is offline
Member
 
Join Date: Nov 2010
Location: USA
Posts: 19,488
Default Re: Tracking with Buffersize 32?

Reply With Quote
  #36  
Old 03-30-2023, 06:18 AM
JFreak's Avatar
JFreak JFreak is offline
Moderator
 
Join Date: Jan 2003
Location: Tampere, Finland
Posts: 24,849
Default Re: Tracking with Buffersize 32?

Darryl, just let me know if you run out of popcorn. USA makes a lot of it, we can perhaps arrange a food delivery.

Anyway.

For most musicians acceptable latency is 12ms, and first digital pianos had latency of 10ms from keypress to output. Playing guitar with cabinet 1m away means 3ms latency.

In 48k session, these numbers would be 580 samples (too high for me), 480 samples (doesn't bother, but not analog piano feel), 145 samples (perfectly okay).

I know some for example classical singers have not used to cans and want to leave one ear open, but it is not the cans to blame.

But saying 64 samples is too slow means the problem is somewhere else than the playback buffer.
__________________
Janne
What we do in life, echoes in eternity.
Reply With Quote
  #37  
Old 03-30-2023, 06:35 AM
Sardi Sardi is offline
Member
 
Join Date: Aug 2002
Location: Perth
Posts: 2,986
Default Re: Tracking with Buffersize 32?




Sent from my iPhone using Tapatalk
Reply With Quote
  #38  
Old 03-30-2023, 06:44 AM
midnightrambler midnightrambler is offline
Member
 
Join Date: Oct 2008
Location: London, UK
Posts: 2,862
Default Re: Tracking with Buffersize 32?

Quote:
Originally Posted by JFreak View Post
Was that with or without plugins? I started with MIX3 888|24 in 1998 (rented at the time). It had roundtrip latency of 71 samples at 48k. Without plugins nobody ever complained about that. But the thing with TDM is every plugin adds some amount of latency so the real-world latency builds up sometimes really fast.
Can't really remember, just have the memory of having to come up with some elaborate routing patches on the board so they wouldn't get upset. It was a ball ache.

Quote:
Originally Posted by JFreak View Post
God only knows how the artists managed to sing with DA/AD to digital reverb, totalling +142 samples of latency (plus being the internal latency of the reverb).
If you're talking external verbs such as the 480 etc then to be fair they'd have also been monitoring the clean signal without any latency so it's not really an applicable example.

Quote:
Originally Posted by JFreak View Post
I still hold my opinion that nobody should have a problem with 128 playback buffer with native systems. And for the few who may always complain, 64 should drop it. 32 setting is just simply unnecessary, unless you want to convince your boss to buy you more powerful computer.
I'm not really disagreeing. I have no problem tracking myself at 128. But I have a singer who can always, always tell the difference between LLM and 128. Much to my annoyance. So you know, horses for courses and all that.
Reply With Quote
  #39  
Old 03-30-2023, 07:03 AM
JFreak's Avatar
JFreak JFreak is offline
Moderator
 
Join Date: Jan 2003
Location: Tampere, Finland
Posts: 24,849
Default Re: Tracking with Buffersize 32?

^ yeah.

Whenever I have disagreement I just tell the singer to drink water while I fix this, then continue without doing anything. Usually no complaints afterwards.

If there really is a problem, and I know some are more sensitive, usually best solution is to leave one ear open. Singer with a real singing technique cannot live with two ears closed no matter what.

Still, unless using some super crazy latency introducing plugin, 256 buffer is a good starting point. I have absolutely no symphaties for "64 is too slow" because the problem is somewhere else. MIX/HD had bigger roundtrip latency with 0 buffer (no plugins).
__________________
Janne
What we do in life, echoes in eternity.
Reply With Quote
  #40  
Old 03-30-2023, 09:31 AM
dominicperry dominicperry is offline
Member
 
Join Date: Apr 2019
Location: UK
Posts: 898
Default Re: Tracking with Buffersize 32?

Quote:
Originally Posted by JFreak View Post
^ yeah.

Whenever I have disagreement I just tell the singer to drink water while I fix this, then continue without doing anything. Usually no complaints afterwards.
Yes, you've been criticised for this dishonesty before, and I will agree again with the criticism.

Has it occured to you that all of these years that you've been lying to people, they may have just concluded it's bad equipment, a bad studio, or a bad engineer?

Dominic
__________________
MacBookPro M1Max 16" 10/32 64GB 2TB, Ventura 13.6.3, Pro Tools 2023.12.1, Carbon.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
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

Similar Threads
Thread Thread Starter Forum Replies Last Post
How to get the low H/W buffersize of 32 Dutchmountain 003, Mbox 2, Digi 002, original Mbox, Digi 001 (Win) 7 12-18-2009 10:34 AM
How do i change the I/O Buffersize? One-i 003, Mbox 2, Digi 002, original Mbox, Digi 001 (Win) 1 01-15-2006 09:26 AM
Buffersize vs RAM soebx 003, Mbox 2, Digi 002, original Mbox, Digi 001 (Mac) 0 06-01-2005 02:27 PM
H/W Buffersize and Rewire am.syn 003, Mbox 2, Digi 002, original Mbox, Digi 001 (Mac) 2 09-23-2004 03:03 AM
buffersize PT 6.4 ?? hoijandee 003, Mbox 2, Digi 002, original Mbox, Digi 001 (Mac) 0 05-05-2004 01:01 AM


All times are GMT -7. The time now is 07:41 PM.


Powered by: vBulletin, Copyright ©2000 - 2008, Jelsoft Enterprises Limited. Forum Hosted By: URLJet.com