The No.1 Website for Pro Audio
PC MIDI timing - is it better on MAC?
Old 8th February 2009 | Show parent
  #151
Gear Head
 
🎧 10 years
Quote:
Originally Posted by qtuner ➡️
have you considered the fact that your mobo may be the problem? have you tested other motherboards or just yours?
I did consider it the mobo but have done no research into it and I haven't tested other mobos with my processor. I bought the PC from a daw vendor (first time) who supposedly knew what he was doing and put all my trust in that.

It seems to be running fine now I have osx running on it but I have not done a stress test yet with many midi tracks.

Audio latency has increased on osx compared with windows but not by much. The exact numbers I can't remember but osx is around 1ms input and under 2ms output latency, I can live with that. I have a mac running pro tools tdm for audio anyway. The pc is just a midi monster which is why i'm so fanatical about the midi on it.
Old 8th February 2009 | Show parent
  #152
Lives for gear
 
Timur Born's Avatar
 
🎧 10 years
Quote:
Originally Posted by gl3ny
Is there anyone who uses a mac as a dedicated daw (no internet or any software other than daw) who gets poor midi? Timur whats you bad mac?
My Mac (late 2008 Macbook Pro 2.8) isn't bad, neither is my Windows PC, just some of the Midi drivers are misbehaving. After optimizing it provides a premium Bootcamp/Windows experience as well.

NI Kore (USB) driver behave about the same on both platforms, I'd say quite well, but they don't reach full Midi bandwidth for output, but for input.

RME Fireface (FW) drivers have just been fixed for Windows and perform superb and best of the pack now. They reach full Midi bandwidth for output and input now (after the 2.94 fix). They perform better on Windows than OS X now, but that will likely be fixed soon as well.

Novation Remote SL (USB) and Kore Padkontrol (USB) performs like sh*t on Windows and only reach 1/12 of the maximum Midi bandwidth for output but full bandwidth for input. It doesn't matter whether you are using Microsoft USB Audio standard driver or the one provided by Novation and Korg. On OS X they perform flawless with full bandwidth regardless of whether you are using the OS X standard drivers or the ones provided by Novation and Korg. I did not test their latency and jitter behavior on OS X though.

Both the M-Audio 2496 (PCI) and the onboard MPU-401 (PCI) perform medium well on Windows, they reach only about half of the Midi bandwidth for output, but full bandwidth for input as far as I remember.

Midi Yoke (VIRTUAL) offer full output and input bandwidth performance. It offers upto 8 independent virtual Midi ports and thus offers 8 times the total Midi bandwidth of a single port, even on a single port as long as the total is not exceeded.

I will check the MOTU Ultralite of a friend once I have time and opportunity.

To summarize: Out of 6 different Midi devices only 1 offers full performance on Windows. Out of 4 different Midi devices 2 offer full performance on OS X. This leads me to believe that alot of Midi problems are still based on bad Midi port drivers.

PS: Yes, I can prove this, in fact you just learned for free what took me days to analyze (good enough for RME to fix their drivers). But no, I'm not gonna waste my time with Fredrik again (I already provided a guide for doing your own tests, read the thread again), I already had that in another thread, it's fruitless. No listening and learning, just lamenting and polemics. fuuck
Old 8th February 2009 | Show parent
  #153
Registered User
 
🎧 15 years
Timur .
i admire the depth of your testing/tenacity and patience with various devices.

regarding your fifth paragraph. maudio 2496.
re half bandwith.when you ran the 2496 test.
just curious....
in this test were any other pci devices operating and useing the pci bus at the same time ??

i have i guess you could call it a pci bus rule.
if useing a pci based sound device..i ensure no other pci device is operating.
thus ensuring the pci sound device has full rein over the pci bus.
same with fw or usb. too many devices on a bus can lead to put it crudely..
"traffic jams".

have you done tests with various ppqn's in the host daw software ??
upping the ppqn ?? and seeing if things improve ??
god bless.
Old 8th February 2009 | Show parent
  #154
Registered User
 
🎧 15 years
Quote:
Originally Posted by Timur ➡️
PS: Yes, I can prove this, in fact you just learned for free what took me days to analyze (good enough for RME to fix their drivers). But no, I'm not gonna waste my time with Fredrik again (I already provided a guide for doing your own tests, read the thread again), I already had that in another thread, it's fruitless. No listening and learning, just lamenting and polemics. fuuck
Wow, you really take this personal? Now you are discussing drivers, you didn´t before. I understand that there are differences between performance among different drivers, I never said anything to the contrary and that wasn´t what we discussed either. My point is that doing tests on the output and input of the system, can give you performace figures, but it wont tell you enogh to pinpoint exactly where the problem stems from and certainly not enough to make general statements on either MME or coremidi.
Old 8th February 2009 | Show parent
  #155
Gear Head
 
🎧 10 years
Quote:
Originally Posted by manning1 ➡️
Timur .
i admire the depth of your testing/tenacity and patience with various devices.
I also admire your depth of testing too. You have certainly said a lot of uselful things people can try.

The problem is this is not controlled testing and everyone is doing there own thing to get it to work. Someone mentioned 'control' earlier in a thread and that is key. That really is key if everyone is going to definatively come up with truely comparable results. Unfortunately people are coming together with all different experience levels and trying different things to test. There is a whole spectrum of experience levels and when this all comes together it can get extremely frustrating.

There are a few people in this thread that should realize this and work together to come up with a procudure that can be used with everyones available hardware. That would be a good thing. I think its more important for people to find out what works rather than argue over why it does or doesn't work. Hardware can be changed, os can be change (in so many ways) and sometimes drivers can be changed.

The fact is that for one person a platform or os change can make a difference and that may well be because the drivers are written better for one os than another for that particular hardware. For another person a change of midi interface might fix the problem and the list goes on.

My point is that if someone was to post saying that their system showed good results because they had followed the same test everyone was following then there wouldn't be any arguments. People would be able to find very useful information to build their systems instead of going through all the frustration we all do everytime we build a DAW. Example start for windows set up procedure

Step 1. Detailed hardware list (TBD)
Step 2. No cards installed on motherboard other than graphics
Step 3. In Bios disable onboard audio
Step 4. Clean install of MS XP Pro
Step 5. Follow instructions to optimize XP for Audio at www.XXXXXXXXX.XXX
Step 6. System>Device Manager>disable any graphics card audio



...so on and so on. Anyone want to argue to creating something usefull?

Last edited by gl3ny; 8th February 2009 at 06:50 PM.. Reason: typo
Old 14th February 2009 | Show parent
  #156
Lives for gear
 
Timur Born's Avatar
 
🎧 10 years
Thanks for the flower Gl3ny and Manning!

The M-Audio Midi Bandwidth limitation is a driver thing, PCI shouldn't be a problem on my system.

Curiously M-Audio changed the Midi behavior of their drivers when they introduced the new GUI. Midi latencies are higher now. M-Audio claims that they will be better with lots of audio happening concurrently, but I liked the Midi behavior of the old drivers better. I don't need them anymore anyway.
Old 14th February 2009 | Show parent
  #157
Gear Head
 
🎧 10 years
Quote:
Originally Posted by Timur ➡️
Thanks for the flower Gl3ny and Manning!
It made it there just in time then

Sorry a bit off topic but since my hardware midi problem are fixed I moved on to software synths. Some are very tight <10 samples, most tight <100samples jitter.

Reason using rewire in cubase is very sloppy on my my system > 8ms jitter. What a waste of money reason was. I knew cracked software had a use.
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 2 views: 1124
Avatar for IGWright
IGWright 17th February 2007
replies: 4624 views: 651152
Avatar for mxeryus
mxeryus 3 weeks ago
replies: 50 views: 14994
Avatar for LesC
LesC 8th January 2015
replies: 443 views: 60005
Avatar for Neil Martin
Neil Martin 14th February 2021
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump