1 Topic by aspender

  • aspender
  • aspender
  • Technical Author
  • Offline
  • From: Peavey Oxford
  • Registered: 2007-12-03
  • Posts: 98

Topic: 1.4.0 Release Candidate 2

1.4.0 Release Candidate 2 is available here:

http://peaveyoxford.com/NWare_1.4.0_RC2_Installer.msi

Please note that even though this is a release candidate, there is one known problem when running under Windows Vista:

To uninstall, you have to right click on the MSI file that you used to install NWare, and select the 'uninstall' option. If you try to uninstall NWare in the usual way via Control Panel, Vista will eventually tell you it can't proceed because you don't have Administrator rights - even if you're logged in as an admin.

Note also that the NWare firmware is release candidate 1. This has not changed, so we did not create a release candidate 2 for the firmware.

Here is what has changed since the last beta.  Thanks for all the suggestions so far, keep them coming!

- Windows Vista is now fully supported.

- Added accelerator keys for setting master (CTRL+SHIFT+M), slave (CTRL+SHIFT+S) and peer (CTRL+SHIFT+P) nodes for controls.

- Changed Expression Label Control Alias accelerator keys from CTRL+SHIFT+S to CTRL+SHIFT+X to avoid conflict with set control node slave (CTRL+SHIFT+S).

- Improved cab polling to reduce network load.

- Items on the Tools menu have been moved to a submenu called "Control Nodes" to improve the clarity of the user interface.

aspender's Website

2 Reply by jvalenzuela

  • jvalenzuela
  • Initech programmer
  • Offline
  • From: Orlando, FL
  • Registered: 2007-10-10
  • Posts: 223

Re: 1.4.0 Release Candidate 2

Just curious, but what exactly has changed with the CAB polling.

3 Reply by Fergy

  • Fergy
  • MM Tech Support
  • Offline
  • From: The Home Office--Meridian
  • Registered: 2007-08-15
  • Posts: 334

Re: 1.4.0 Release Candidate 2

Expression Labeling ICON on toolbar works for labels and blocks, but not for Flyoffs.  Only thing so far...

Except, I had the same question about CAB Polling.

Make it intuitive, never leave them guessing.

4 Reply by michel

  • michel
  • michel
  • PDR Engineer
  • Offline
  • From: Oxford, England
  • Registered: 2007-08-13
  • Posts: 31

Re: 1.4.0 Release Candidate 2

I changed the cab pooling in 3 ways:
1) Added subchannel mapping
2) To add #1 to the Cab4N I had to rework the way the cab4n are polled; that results in smaller packets being sent to it,
3) Since I was there, I reworked the pooling scheduler; instead of waiting 5ms between cabs, I now 'spread' the polling over time (in the same window), and added a touch of randomness to it; this will prevent two polling nions/conmen to accidentally get 'in sync' and create network spikes.

Normally, you won't notice anything. In the best case, it might make the cab4n less prone to dropping out...

5 Reply by Joe Kurta

  • Joe Kurta
  • Member
  • Offline
  • Registered: 2007-08-17
  • Posts: 1,179

Re: 1.4.0 Release Candidate 2

Michel,

Just curious, will the 'spread' of pooling affect the response time of the GPIO at all?

Thanks,

Joe

6 Reply by michel

  • michel
  • michel
  • PDR Engineer
  • Offline
  • From: Oxford, England
  • Registered: 2007-08-13
  • Posts: 31

Re: 1.4.0 Release Candidate 2

No not at all. they are 'spread' between themselves, but the pooling frequency is pretty much exactly the same as it was before.

The only difference you /could/ notice is if you play with GPIOs /across/ several cabs (IE turning relays/gpios on/off on 2 cabs 'at the same time'). In that case you might notice a difference of a few milliseconds between the two cabs responding. That would be the only way to demonstrate the change. Of course, depending on that sort of timing would be rather ill advised to start with big_smile

7 Reply by jvalenzuela

  • jvalenzuela
  • Initech programmer
  • Offline
  • From: Orlando, FL
  • Registered: 2007-10-10
  • Posts: 223

Re: 1.4.0 Release Candidate 2

I have a quick question regarding CAB polling. Why was Ethernet protocol 0x8819 chosen for this traffic? Strictly speaking, it isn't Cobranet. Is it a requirement of the Cirrus cards to only accept this for Layer 2 traffic?

Would perhaps change of state based communications be a better fit?

8 Reply by michel

  • michel
  • michel
  • PDR Engineer
  • Offline
  • From: Oxford, England
  • Registered: 2007-08-13
  • Posts: 31

Re: 1.4.0 Release Candidate 2

Cobranet allows encapsulation of packets between controlling processors; so the 8819 is used because a cab has only one ethernet port, and it's driven by a cobranet chipset; therefore you need to pass any data that is not technically cobranet audio you need to use the encapsulation mecanism. think of it as a 'tunnel'.

We have no plan to reorganize the protocol stack anytime soon big_smile