View Forum Thread

> Forum Home > Feature Requests > Nuke module

	Wed, 08/Jul/2009 12:10 AM
Richard 11 Posts	Would it be possible to modify the Nuke.sx module to be able to expose the renderer switches? I'm currently attempting to make a Virtual Module so I can add the "-p" switch to turn on proxy mode, but thought this might be an easy addition to the base module. Thanks, Richard
	Wed, 08/Jul/2009 3:42 AM
Robin 1138 Posts	Yes, this should be done in the Nuke Module itself. There are some other features from other Modules that have been or are being incorporated into Nuke as well. I am taking a look at the command line parameters now to see how to implement them in the Module. Are there any switches you would want to see right on the basic info tab instead of in a "Render Overrides" tab? Any additional tabs you want to divide the parameters into? -robin
	Thu, 09/Jul/2009 7:00 PM
Richard 11 Posts	Hi Robin, I'm not a Nuke user, so hopefully someone else can contribute here. Whether the output is full res or proxy size, can be set in the job/scene script, so being able to modify this after the script has been saved out would be helpful. Perhaps a "Proxy Output" checkbox to set the "-p" switch and a "Full Res" one to set the "-f" switch, which overrides proxy res, if set in the script. A "Cache Memory" field which sets the "-c" parameter (Size in bytes, or append k, M, G or T). One switch that would also be very useful is the "-i", which allows command line rendering using a GUI, rather than render licence - Nuke has 2 distinct licences, (GUI or interactive) and render and command line rendering cannot occur on a GUI licence without "-i" set. However, in order for this to be completely flexible, this setting would have to be set as an Engine Product setting, rather than a Job setting. Other than the above the ability to add any other switches would be great. Hope this helps and let me know if you need anyy other information or clarification. Regards, Richard
	Fri, 10/Jul/2009 4:53 PM
Robin 1138 Posts	So I added the -i flag as an Engine Option, the -c, -f, -l and -p flags to a new Render Overrides tab in the Submit Job window, and I added a new optional basic info parameter for the -X flag. If you set this to the name of a write node in the file, it will add the -X flag with that name to render only the specified write node. If you leave this blank, it adds the -x flag to render every write node in the file. In the documentation, it says that the -p flag "disables proxy mode," and the -f flag "opens script at full resolution." From that wording, it sounds to me like these two flags do the same thing, but I admit that I'm not a Nuke user, so I put both of them there as options. Does this sound OK? Richard, were you doing this for Nuke users at your location? Do they have any preferences or thoughts on this? Edited by author on Fri, 10/Jul/2009 4:54 PM
	Sat, 11/Jul/2009 8:19 PM
Richard 11 Posts	Hi Robin, "In the documentation, it says that the -p flag "disables proxy mode," and the -f flag "opens script at full resolution." From that wording, it sounds to me like these two flags do the same thing..." Hmmm. Running "Nuke5.1 --help" gets the following here: Usage: nuke <switches> <script> <argv> <ranges> -a formats default to anamorphic -b start in background (fork) -c size limit cache memory usage. Size is in bytes, or append k, M, G or T -d name set X display name -f render at full size (turns off proxy; use -p to force render at proxy) -help print this help and exit -i with -x or -t use interactive, not render, license -l apply linear transfer to the file read in -m n set threads to n -n don't run postagestamps, don't open windows -p turn on proxy mode (use -f to force full size) -q quiet (don't print stuff) -s n sets the minimum stack size for each thread in bytes, this defaults to 16777216 (16MB) the smallest allowed value is 1048576 (1MB) -t terminal only (no gui) -V verbosity (print more stuff) -v nukev (rest of command line is image files to view) -view v only execute these views (comma-separated list: e.g. 'left,right') -version print version information and exit -x execute the script (rather than edit it) -X nodes only execute these nodes (comma-seperated list) -- end switches, allowing script to start with a dash or be just - to read from stdin <script>: name of a .nuke script to create, edit, or execute "-" means stdin <argv>: All words between the script name and the frame ranges can be used by [argv n] expressions to provide changing arguments to the script. Each must start with a non-digit to avoid confusion with frame ranges. <ranges>: frame numbers to execute: A single frame number A A-B all frames from A through B A-B/C every C'th frame from A to last one less or equal to B So it looks like the documentation you are referring to may old/incorrect. "Richard, were you doing this for Nuke users at your location? Do they have any preferences or thoughts on this?" Yes and my previous post was as a result of their feedback and stuff I thought might be useful. What you have now added looks like we should be all happy. Regards, Richard
	Mon, 13/Jul/2009 3:38 AM
Robin 1138 Posts	I was reading the PDF manual. I see now in the command line help that -p actually turns on proxy mode, which makes more sense. Their PDF manual (at least with 5.0, the version I have here) has a typo in it! In any case, both -p and -f are there. I'll update the help documentation in Smedge, too. One thing I have not added is the ability to access the "argv" area of the command line. This is the part after the script filename and before the frame range. Would a parameter to access this area be helpful? Thanks -robin
	Mon, 13/Jul/2009 3:22 PM
Richard 11 Posts	"Their PDF manual (at least with 5.0, the version I have here) has a typo in it!" Yes, I spent some time correcting several issues with python script examples in there as well. On the "argv" front, we aren't using this, but I'm sure if you add it, someone will appreciate it. Thanks for the responsiveness on this. Regards, Richard

Page 1 of 1