miBoolean mi_job_set_nthreads( int n)
Set the number of rendering threads for the next render. This allows to control the number of compute intensive rendering threads in mental ray. To achieve optimal rendering performance on a multi-processor machine this should be set to the number of CPU cores, which is the default behavior of the mental ray standalone, for example. For integrations, compute power sometimes needs to be balanced between the parent application and the rendering engine, especially if both run expensive operations in parallel.
miBoolean mi_job_set_slaves_active( miBoolean on)
Enable or disable slave host participation for the next render. Sometimes previews are faster without network rendering due to the network transmission latencies. This function allows to disable slaves temporarily.
miBoolean mi_job_set_slaves_only( miBoolean on)
Enable or disable rendering on the master if slaves are enabled, for the next render. This allows to render on slave hosts only and keep the master host free from compute intensive operations. Note, that mental ray schedules some tasks to be executed on the master only. Also, all network communication and final image tile delivery is managed by the master host.
typedef enum { /* type of progress, for callback */ miJOBP_FINALGATHER, /* finalgather preproc */ miJOBP_FASTLOOKUP, /* fg fastlookup preproc */ miJOBP_PHOTONEMISSION, /* photon emission */ miJOBP_RENDER, /* standard rendering */ miJOBP_PHOTONREEMISSION, /* photon re-emission */ miJOBP_IMPORTONEMISSION, /* importon emission */ miJOBP_IMPORTONREEMISSION, /* importon re-emission */ miJOBP_AOPOINTEMISSION, /* amb occl points emission */ miJOBP_AOPOINTREEMISSION, /* amb occl points re-emission */ miJOBP_IPADAPTPROCESS, /* irr particles geom variab comp */ miJOBP_IPADAPTREPROCESS, /* irr particles geom variab re-comp */ miJOBP_IPINDPROCESS, /* irr particles ind pass comp */ miJOBP_IPINDREPROCESS, /* irr particles ind pass re-comp */ miJOBP_IPFINALPROCESS, /* irr particles final irr pass comp */ miJOBP_IPFINALREPROCESS, /* irr particles final irr pass re-comp */ miJOBP_IPTRACE, /* irr particles tracing */ miJOBP_IPRETRACE, /* irr particles re-tracing */ miJOBP_N /* number of legal values */ } miJob_progress; typedef void (*miJob_progresscb)( miScalar percent, /* percent of task completed */ miJob_progress current_task); /* type of current task */ void mi_job_percent_register_cb( miJob_progresscb callback);
This function registers a callback function that will be called by mental ray for progress events in the various rendering phases. The function will be called regardless of the current verbosity setting. This allows applications to generate feedback and progress information as needed, like drive user interface progress bars or print messages at any desirable frequency.
Copyright © 1986, 2015 NVIDIA ARC GmbH. All rights reserved.