Client Differences v6 to V7

This page describes differences between the v6 and V7 clients. It is assumed the reader is familiar with the v6 client above the Novice level.

Command Line & Configuration Options

This table describes the V7 command line option equivalents from v6.

v6 Client Options V7 Client Options V7 Notes
-config --configure --config <filename> is used in the V7 core to pass in the configuration file.
-configonly --configure --configure always exits.
-help --help --help always exits.
-queueinfo --queue-info V7.0.7+
-delete # --dump <## | 'all'> V7.0.7+, the unit will not actually be removed until the WS can be notified.
-send all --send V7.0.7+, attempts to upload all completed work units to available servers
-send # Not supported
-verbosity # --verbosity # v6 supported 1-9, V7 uses 1-5, with 3 being the default, and recommended setting
-pause --finish V7.0.7+, the V7 option causes the core to quit after finishing the current WU.
-oneunit --max-units=1 V7.0.7+, V7.1.25 works with next-unit-percentage Or use the "Finish" button in FAHControl.
-forceasm --extra-core-args=-forceasm Any previous (v6) fahcore arguments can be passed with extra-core-args.
-advmethods --client-type=advanced
-bigbeta --client-type=bigbeta V7.0.7+
-bigadv --client-type=bigadv V7.0.7+
-betateam --client-type=beta
-internal --client-type=internal
-gpu <number> --gpu-index <number>
-service --service
-freeBSD --core-prep="brandelf -t Linux" V7.0.7+
-openBSD --core-prep="elf2olf -o Linux" V7.0.7+
-smp <number> --smp --cpus <number>
-threads <number> --cpus <number> The V7 client will auto detect the total number of CPU threads by default. --cpus # can set that lower, but most odd values are set to the next lower even value.
-forcegpu <type> --gpu-species <type>
-nice <number> --priority <number> nice also works as an alias for priority
-license --license
-art Not supported
-local Default setting. Not needed.
-svcstart Not supported
-d <path> Not supported
-deino Not supported
none --lspci New feature, enumerates the PCIe device list for GPUs

Most V7 command line options can also be added to the configuration or included in the configuration file, without the 2 dashes (--).

Client level options (Global - all slots) are set in FAHControl in the Expert Tab (in Expert mode). Slot level options (Local - one slot) are set in FAHControl in the Slots tab (in Advanced or Expert mode). Most options can be set either Globally or Locally (within a single slot).

Exceptions are those options which perform some operation then exit rather than running the client normally. For example --help or --configure.

New Features in V7

This table describes new features to FAH in V7. (list is still being assembled...)

V7 Option Value V7 Notes
next-unit-percentage x Download new WU while finishing current WU. Set the % when new WU downloads, x = 90%-100%, default is 99%
pause-on-start true/false Allows the client to start running automatically at system startup, but in a paused state. Does not download new work, must be started.
exit-when-done true/false close FAHControl after finishing work units, false is default
Proxy Support - Supports basic and digest authentication. NTLM, SOCKS, SSL are not supported
proxy-enable true/false true enables the proxy feature, false disables feature
proxy xxx.xxx.xxx.xxx:xxxx IP Address and Port # of proxy
proxy-user x x = proxy user name
proxy-pass x x = proxy password
Client Configurations - Allows FAH to save the last x number of configuration files after each settings change
config-rotate true/false Enables multiple copies, true is default
config-rotate-dir n n = name of the directory in which to save the config files, under the fah data folder
config-rotate-max x x = number of previous copies of the config to keep
gpu-usage x x = percentage of GPU processing cycles to use for FAH, similar to cpu-usage setting. (Functionality has been deprecated.)
cuda-index x more to follow, 0, 1, 2, etc., used to set device hardware order when mixing GPU vendors
opencl-index x more to follow, 0, 1, 2, etc., used to set device hardware order when mixing GPU vendors