Discussions about enhancements in future versions of Muster http://www.vvertex.com/forum/list.php?3 Wed, 08 Sep 2010 16:12:30 +0200 Phorum 5.2.11 http://www.vvertex.com/forum/read.php?3,530,530#msg-530 http://www.vvertex.com/forum/read.php?3,530,530#msg-530
here is a little list of suggestions that i have gathered from our team that uses Muster on a Daily basis.

Bugs in version 5.5.5+

Frames Check Window/Missing frames parser preferences is bugged. and has been for a couple of versions now.

It´s not possible to use the Size Limits.
When you press the radio button on Limits: the exact match is still selected and the Limits are Grayed out so you can´t specify high and low limit in file size.




Feature Requests for Muster Explorer
1. Ability to set Max node usage on Folders in the Queue list

2. Ability to change packet size without reiniting an entire job.

3. When you search for a specific job. Automatically select it in the Queue list and open the folder tree so you can se where it is. At least Highlight the entire folder path down to the job so you can find it.




Feature Requests for Management Console

1. Pools Support or some kind of sorting for easier management. Sometimes we use our regular workstations for rendering and it would be great to have some kind of pool or sorting in the Management Console and not just one long list of computers.

And as a followup. The ability to paste settings on a pool of nodes. Ex: Floor1, Floor2 have different settings.]]> skaran Enhancements Thu, 30 Apr 2009 11:00:40 +0200 http://www.vvertex.com/forum/read.php?3,506,506#msg-506 http://www.vvertex.com/forum/read.php?3,506,506#msg-506 skip rendering option. That option means Maya -rep.
And XSI and AE have same option.

When I choose "every instance", dispatcher assigned
the job to whole instance. However the destination
pool was ignored.

"single instance" considers the destination pool.

Is this behavior bug or spec? If spec, I wish same
behavior.

Masa.]]> bd_nkjm Enhancements Fri, 30 Jan 2009 08:13:01 +0100 http://www.vvertex.com/forum/read.php?3,486,486#msg-486 http://www.vvertex.com/forum/read.php?3,486,486#msg-486 gdizzle Enhancements Mon, 17 Nov 2008 17:51:45 +0100 http://www.vvertex.com/forum/read.php?3,483,483#msg-483 http://www.vvertex.com/forum/read.php?3,483,483#msg-483
the -jobparent flag is used for limiting queries.

I was hoping to write a cron job that would check for completed jobs and move them into a folder, to avoid the main list getting too clogged up with old jobs]]> Hugh Enhancements Fri, 03 Jul 2009 17:06:21 +0200 http://www.vvertex.com/forum/read.php?3,474,474#msg-474 http://www.vvertex.com/forum/read.php?3,474,474#msg-474 We have recently redesigned our production pipeline from the bottom up, and have been able to achieve a serious amount of automation of tasks. The ability to automate job submission (with dependencies and appropriate folders) to muster (maya, fusion, vue, and AE jobs) would be very easy if we had the ability to make the job ID #'s meaningful (allowing leading zero's would also be helpful).

For example: when submitting project A's (06) scene 01 shot 15 which is a Fusion 2D (02) job and is a main compositing pass (01), it's job ID # would be 0601150201, and could automatically be set as a child of the folder named "2D" who's job ID is 06011502 which is a child of the "Shot 15" folder who's ID is 060115 which is a child of the "Scene 01" folder who's ID is 0601 which is a child of the "Project A" folder who's ID is 06. Since it's a main compositing pass it could also be automatically set to dependant on the Project A Scene 01 Shot 15 "3D" folder who's ID is 06011501.

An average project of ours has 50 camera shots, and each shot has an average of 6 maya passes, 3 Vue passes, 2 Fusion passes/flows and 2 AE comps, which results in approx 650 muster jobs (not including folders to organize them). Given a generous 5 minutes to set up each job, we estimate having this ability will save approx 50 hours of work per project.

Regards,
~Yuris Mangolds
Director
C2i Studios
www.c-2iinc.com]]> ymangolds Enhancements Fri, 03 Jul 2009 17:27:34 +0200 http://www.vvertex.com/forum/read.php?3,471,471#msg-471 http://www.vvertex.com/forum/read.php?3,471,471#msg-471 It would be nice if there was an option to disable all confirmation dialogs in the Muster Explorer window and sub menus. It can get rather tedious telling muster that I'm positive that that's what I'd like it to do (for example requering missing frames or chunks that messed up, or realizeing that someone forgot to set the "number by" option and having to edit all the jobs for an entire project).]]> ymangolds Enhancements Fri, 03 Jul 2009 17:28:34 +0200 http://www.vvertex.com/forum/read.php?3,465,465#msg-465 http://www.vvertex.com/forum/read.php?3,465,465#msg-465 jpspgd Enhancements Fri, 03 Jul 2009 18:23:29 +0200 http://www.vvertex.com/forum/read.php?3,463,463#msg-463 http://www.vvertex.com/forum/read.php?3,463,463#msg-463
We have a number of shared pools and i'd like aritsts to be able to use some a but not others .

Unfortunatly limiting them to just one pool is too limiting. Prehaps you could select poolsin a similar way to the system for selecting allowed job templates?]]> doogstar Enhancements Fri, 03 Jul 2009 18:15:54 +0200 http://www.vvertex.com/forum/read.php?3,458,458#msg-458 http://www.vvertex.com/forum/read.php?3,458,458#msg-458
say i have a machine called RENDER_ONE and it is in the following pools:

project_A
Project_B
Entire_farm

I want to render all jobs for Project_A in preference to Jobs in Project_B pool and then with least priority jobs in the entire_farm pool.

currently the only priority is job priority so a job submitted to entire farm at 100% would render on RENDER_1 before jobs submitted to projec_A and Project_B pools at 60 % priority.

If we could assign a weighting or priority to a given pool then we could ensure that Jobs submitted to Project_A and Project_B always rendered on our machine RENDER_1 before jobs submitted to the entire farm whatever the job priority of the entire farm job.

There is another benefit to this, if I split the entire farm into 2 pools Project_A and Project_B and then submitted a job to Project_A and a Job to Project_B both jobs render in parallel but if the job for project_A finishes before the job from Project_B the second job will only render on half the farm while the project_A pool sits Idle. If we had pool Priority running and submitted the project_B job to the entire_farm pool,then when the project_A job finished the project_B job runing on the entire farm would pick up the available Project_A processors this gives better utilization of resource and faster render times.

Some of our pools are set up by job type say shake or after effects so we would need a switch to say this pool is exclusive never render jobs submitted to other pools on this pool.


with both pool and job priority we can be much more granular about which jobs get rendered in which order and have much better control of the farm and its throughput.]]> doogstar Enhancements Wed, 17 Sep 2008 17:02:11 +0200 http://www.vvertex.com/forum/read.php?3,364,364#msg-364 http://www.vvertex.com/forum/read.php?3,364,364#msg-364 A feature that I would really like to see would be the ability to set CPU affinity for render nodes. For example: using a Dual-Quad Core machines, and setting the renderclients to use CPU 2,3,4,5,6, and 7, leaving CPU 0 and 1 free for other use. It is easy to do manually in winxp64 by opening the Task Manager, going to processes tab, selecting the mayabatch.exe, right clicking it and choosing "set Affinity..." then unchecking processor 0 and 1. Problem is; since a chunk is 4 frames long, a new mayabatch is used for each chunk, so the only reasonable way to do it would be to have muster set affinity as it opens the mayabatch.

Thank You,
~Yuris Mangolds
C2I
Director]]> ymangolds Enhancements Mon, 26 May 2008 12:52:22 +0200 http://www.vvertex.com/forum/read.php?3,324,324#msg-324 http://www.vvertex.com/forum/read.php?3,324,324#msg-324 example:
\\host\share\scenes\!part1\scene1.mb is converted to /host/share/scenes/!part1/scene1.mb
but for it to work i think it should be:
"/host/share/scenes/\!part1/scene1.mb"

i've tried to add "! to \!" path substitution, but muster changes it to "/!".
any chance for it to be fixed?]]> tomekp Enhancements Tue, 01 Apr 2008 19:36:20 +0200 http://www.vvertex.com/forum/read.php?3,314,314#msg-314 http://www.vvertex.com/forum/read.php?3,314,314#msg-314 dsliter Enhancements Tue, 01 Apr 2008 19:34:05 +0200 http://www.vvertex.com/forum/read.php?3,280,280#msg-280 http://www.vvertex.com/forum/read.php?3,280,280#msg-280
You have put an incredibly needed feature into Muster 5.31, but I honestly think you've implemented it in a terrible way.

I've been laboring to come up with a way to monitor CPU usage across render nodes, but in 5.31, you include CPU usage as part of the node status, which is an absolutely excellent development. You really, really need this.

But then, in order to save a couple of bytes of bandwidth out of the terrabytes of data that pass through the network, you only update CPU usage when the node's status changes, which to me is bizarre. You literally have to go to each render client an click Refresh to see updated CPU usage status. What??? I mean, what??? This is the same as putting the feature in and then taking it back out!

Even with 10,000 render nodes, updating the CPU status is such a minor thing, like a couple of bytes maybe every 10 seconds. Why on Earth have you put in this feature and then decide to cripple it? It doesn't make sense at all. Updating CPU status is not a drain on bandwidth, especially if you stagger the requests. Or to put in another way, if checking the CPU usage is a drain, then there's something very very wrong with the code you're using to request it.

Apparently this is not a bug, as you've described it in the manual. But there is no technical basis for this limitation, none whatsoever. If you simply poll one render client every second, we're talking about maybe 6 bytes of request data and 40 bytes TCP overhead. Any gigabit network has capacity for maybe 80,000,000-100,000,000 bytes per second. What's 46 bytes more?

Simply scan a new render client every second. People who have only 5 render clients get updated fast. People with 5,000 render clients simply get updated slower.

But don't cripple the feature. The CPU Usage feature is, currently, completely useless. But you really need it.

Per]]> perholmes Enhancements Fri, 11 Jan 2008 01:26:55 +0100 http://www.vvertex.com/forum/read.php?3,279,279#msg-279 http://www.vvertex.com/forum/read.php?3,279,279#msg-279
In earlier versions of Muster, there used to be the ability for people whose machines were used as Render Clients, to open the Render Client manager and quickly do a "Kill And Pause" or "Resume".

In 5.31, this ability has disappeared, as all machines are managed centrally. I understand the design philosophy of managing render clients centrally, and that's a good thing.

But you are also misunderstanding the user demographic, as it is quite common to use every PC in the building as a render client and leave them on 24/7. During the day, these workstations are used for regular business, at night they render.

It's neither safe or appropriate to start giving a receptionist or an accountant access to Muster Explorer, so that they can turn their own computer on or off as a render node.

I think that you need to write a tiny and dedicated tray application that is idiot-proof, where people can quickly select their computer's status, such as Active, Pause, Kill And Pause. You could possibly also make an availability detector that reenables a render client after a preset amount of inactivity (i.e. 20 minutes or 30 minutes). This could then open a floating window in the middle of the screen alerting the user that the machine is now being used as a render node, giving people the option to Kill And Pause.

The important part is that on a big job, everyone's computer will be used for rendering, but many of those people don't even know what rendering is, let alone know what to do inside Muster Explorer. I think you need to write an interface allowing people to control their computer's status without requiring them to be a network administrator as well.

Someone who is a power user would simply opt to not have this tray application installed on his computer. Power users can control their computer through Muster Explorer. Everyone else can control their computer through a friendly and simple tray application.

Thanks,

Per]]> perholmes Enhancements Fri, 11 Jan 2008 01:30:55 +0100 http://www.vvertex.com/forum/read.php?3,270,270#msg-270 http://www.vvertex.com/forum/read.php?3,270,270#msg-270
or we can try the "deadline" way, they give a screensaver version of their renderclient, when screensaver starts, render client starts]]> oglop Enhancements Tue, 24 Jun 2008 15:50:10 +0200 http://www.vvertex.com/forum/read.php?3,269,269#msg-269 http://www.vvertex.com/forum/read.php?3,269,269#msg-269 oglop Enhancements Mon, 17 Dec 2007 19:40:38 +0100 http://www.vvertex.com/forum/read.php?3,250,250#msg-250 http://www.vvertex.com/forum/read.php?3,250,250#msg-250
Thanks,
Jeff]]> jeffm365 Enhancements Thu, 22 Nov 2007 02:51:58 +0100 http://www.vvertex.com/forum/read.php?3,244,244#msg-244 http://www.vvertex.com/forum/read.php?3,244,244#msg-244
one very important feature that we would like to see is the possibility to have time rules for the slaves that incorporate the number of used processors.

in our company we have access to a huge amount of computers that are used by regular computer users (word, excel, mail, web...). these machines all have dualCore CPUs.
so during daytime we want to use 1 processor of these machines and both processors during night time.

that would be very nice to have

cheers

alex]]> alexx Enhancements Tue, 27 Nov 2007 01:55:31 +0100 http://www.vvertex.com/forum/read.php?3,221,221#msg-221 http://www.vvertex.com/forum/read.php?3,221,221#msg-221
I think it'll be neat to have different actions if different events occur,It's better to give the user the ability to define search strings in the render output, not only just "Warning" and "Error", and muster should take different actions when it encounter different situations, say if the render output "contains an uncached hair system", it should kill and pause all render and inform the submitter to check the cache files, or something like this.

i've tried frantic films deadline, and a neat feature of that is like maya ,they have pre frame script/ post frame script / pre chunk script /post chunk script, they can use vbscript in these scripts, hope we can run shell script, or python script after rendering is complete...


hope to see these new features in coming version 5.4]]> oglop Enhancements Sat, 06 Oct 2007 11:54:56 +0200 http://www.vvertex.com/forum/read.php?3,220,220#msg-220 http://www.vvertex.com/forum/read.php?3,220,220#msg-220 oglop Enhancements Tue, 01 Jan 2008 02:43:28 +0100 http://www.vvertex.com/forum/read.php?3,142,142#msg-142 http://www.vvertex.com/forum/read.php?3,142,142#msg-142
Could you add skip error check box to maya multiframe connector? When I submit job through multiframe connector, I want the job to be process immediately without error red text coming up in the explorer's log pane. few more request are also adding frame check option to connector's interface.It doesn't have to be full frame check option. make it lite version for frame check. Giving just image's name should be quite good enough. Also put the bullet iconic instead of text in the password field would be sweet.
Or there are some other way around to transfer these option from connector to the submission panel I've been missed looking.Please let me know.

]]> nooknik Enhancements Fri, 25 May 2007 20:38:54 +0200 http://www.vvertex.com/forum/read.php?3,115,115#msg-115 http://www.vvertex.com/forum/read.php?3,115,115#msg-115 oglop Enhancements Tue, 22 May 2007 01:51:24 +0200 http://www.vvertex.com/forum/read.php?3,75,75#msg-75 http://www.vvertex.com/forum/read.php?3,75,75#msg-75
help your loyal costumers.

R.Andrade]]> rick3d Enhancements Sat, 17 Mar 2007 21:08:21 +0100 http://www.vvertex.com/forum/read.php?3,24,24#msg-24 http://www.vvertex.com/forum/read.php?3,24,24#msg-24 i have to propogate it to all hosts ?
or only hosts that i submit jobs from ?
or those and the dispatcher ?

wouldn't it be better if you could just tell all nodes to read the templates from a central location so editing and testing would be easy

- jeremy]]> jeremy Enhancements Sun, 14 Jan 2007 14:05:14 +0100 http://www.vvertex.com/forum/read.php?3,14,14#msg-14 http://www.vvertex.com/forum/read.php?3,14,14#msg-14
and they have changed the network render system.

"Fusion 5.1 introduces the ConsoleSlave, a new command line style renderer that greatly simplifies integration with these third party render managers."


More information here
[www.vfxpedia.com]

Could you take a look at it for Muster 5?]]> skaran Enhancements Thu, 31 May 2007 12:34:38 +0200 http://www.vvertex.com/forum/read.php?3,10,10#msg-10 http://www.vvertex.com/forum/read.php?3,10,10#msg-10
In muster 5 it´s named only after Job ID instead :/ the que can be lost u newer know, and i have no idea what render that was in 6 months or so if i need to backtrack and find a specific version of a scene.


Old format example: "cam02_wheel_bg_02.lws_591_12.TXT"
New format example: 2.7.log :/ what job was that?

Can u put back the old naming?

Or the ideal would be a little option under logs, so we could specify on our own what kind of output logs we get out.

%Q.C%.J%.log
(QueID.ChunkID.JobName.log)


Keep up the good work guys but please take a look at this.]]> skaran Enhancements Sun, 07 Jan 2007 01:49:52 +0100 http://www.vvertex.com/forum/read.php?3,3,3#msg-3 http://www.vvertex.com/forum/read.php?3,3,3#msg-3
Welcome to the enhancements requests section.
In this forum channel, we welcome any kind of user that wanna share its idea about possible enhancements to the Muster software.

Any suggestion we think may be included in a future release of Muster, will be accepted and made sticky. You can find a list of accepted features at the beginning of the forum itself.]]> admin Enhancements Wed, 15 Nov 2006 13:04:55 +0100