Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 44. (Read 5805649 times)

There is no API command to switch the strategy ... currently.
https://github.com/ckolivas/cgminer/blob/master/API-README

There is no API command to switch the strategy ... currently.
https://github.com/ckolivas/cgminer/blob/master/API-README

There is no S5 code in the official cgminer nor will you get there any support for altcoins on this thread so unfortunately you'll have to seek help from whoever maintains your unofficial fork of cgminer.

so im using a profit switch sha256 pool (nomp) every time they switch coins they send out the "disconnect" command. then my 2 antminer S5's crash in hashrate and take 1-5 minutes to go full speed again nutil the nest coin switch (which is usually every 10 minutes) is there any way to shorten the time between disconnect and reconnect? im using cgminer 4.9 on the nicehash firmware. I know nicehash uses extranonce.subscribe  . but most pools dont support this feature (wish they did since my hashrate never crashes on nicehash).

it will run on the following

antminer U1
antminer U2
antminer U3
block erupter
block erupter sapphire

yea, works, thank you

I'll update them once I finally can properly make an S3 firmware.
The current S1/S2 has all the changes up to "Ant voltage isn't S1", just before the S3 code was added, so it wont make a difference unless your S1/S2 has a high --queue value

The code for S3 is all there, but I need to make an S3 firmware since the bitmain firmware has all the API and cgminer option changes and I'm not going to change all that in a cgminer branch to deal with the silly things they do.
Working on it ...

It's working great!!!
I even have an increase in performance from my even more dated hardware!!

 Thank you!!

Excellent, look forward to playing with it.  Thank you!

Will the Antminer firmwares that have already been released (4.9.1c) be updated to 4.9.2 as well?

New release: Version 4.9.2 - 12th June 2015

Long term stability updates, clean ups and driver updates.


Human readable changelog:

- Various drivers that failed to initialise (notably Antminer U3s) have been fixed by addressing various low level detection issues with other drivers.
- Blockerupter tube driver has been disabled by default in binary builds since it's broken, and the configuration advises against it.
- Fixes for various BFL devices not being properly recognised.
- Getwork support has been completely REMOVED. Only local work generation protocols are now supported (stratum, GBT and solo mining). No pool should be providing getwork support in the modern mining era anyway.
- Queue option has been REMOVED. The option is now silently ignored to not break existing configurations but no more will any antminer binaries be able to set it to 8000 or something stupid. Any attempt to tune this value is pointless and dates back to getwork.
- Expiry and scan-time have also been REMOVED. They are irrelevant with local work generation and asic mining speeds and no getwork.
- failover-only option has been REMOVED. As per above.
- Pool failover has been cleaned up without getwork
- Default Antminer U3 voltage has been changed to 775 since 750 is rarely enough to get 2nd generation ones running at full speed.
- U3 will USB reset on no shares for 2 seconds instead of 1.
- Major antminer S1/S2/S3 driver updates courtesy of kano. Use his provided firmware links.
- Systemd support courtesy of Gabe Evans
- Update to latest libjansson with custom fixes imported from ckpool
- Support for building with system libjansson (dunno why you'd want to do this so it's advised against in the config option but it seems people want useless stuff and it's harmless) - Gabe Evans
- SP30 uses applog instead of printf -  Gabe Evans
- Make json rpc requests valid -  Gabe Evans
- Syslog limits to logging - Gabe Evans
- Build configuration fixes for building from git with newer systems.
- Support mining.reconnect with port as a string or value.
- Numerous clean ups and fixes to minimise risk of invalid memory operations in the code.
- Fix for a rare crash on pool switching during failover.


Full changelog:

- Only blacklist devices on iManufacturer if they don't have an iProduct
- Simplify the pool work choice switching now that only local work generation is
supported
- Check for absense of stratum notify on stratum pools to see if they're usable
- Remove getwork support
- Remove test for localgen since all work will be
- Deprecate failover only option in anticipation of removing getwork
- Deprecate expiry option and fix it to 10 minutes, replacing all deprecated
config calls in the API with a deprecated message
- Deprecate scan-time which is irrelevant with ASIC speed mining
- Deprecate opt_queue since all mining is now local work generation which is
rapid and it serves no useful purpose to generate work that is potentially
stale, increasing CPU usage to check it before using it
- Change default au3 voltage to 775 to more reliably run at default frequency
- Set correct timeout for AU3 in icarus_detect_one though it won't affect
functioning
- Blacklist LIX device which doesn't exist
- Round up fail time to next highest integer on icarus
- AntS3 - set default work delay
- add pool block quality to api
- AntS3 - redo work send/stats and add a tuning option - not yet tuned
- Ant - ignore useless options in case they are passed
- AntS3 - initial S3 merge basic support
- Ant voltage isn't S1
- API - restore IPv4 address .0 padding and update copyright dates
- ants2 - separate freq from options
- ants2 - make it compile :p
- ants2 - add voltage
- ants2 - remove libusb/udev dependency since it doesn't use USB
- AntS1 ensure correct endian and correct wid
- libsystemd: Notify watchdog
- libsystemd: Notify run state and status line
- Catch SIGABRT signal and exit gracefully
- Add libsystemd configure option (default disabled)
- Document --with-system-jansson configure option
- SP30: Fix logging to use applog over printf calls
- Allow building with system jansson
- Set request ID in validateaddress JSON-RPC request.
- Cope with reconnect being sent a port number as an integer or string
- Added syslog mask to limit amount of output
- regressed with "BAL" little-jalapenos
- Add yet more whitelisting of BMA devices, removing the case insensitive match
for manufacturer
- Fix detection of no mining device configuration.
- Remove LT_INIT from configure which breaks ltmain.sh detection for some
automagic reason
- Add missing hashtable_seed file
- Add missing file
- Speed up shutdown by decreasing usb poll time to 100ms and using no timeout on
exit
- Cope with attempted reuse of strings in input_pool
- Manage failed URL entry at startup more gracefully, adding stratum+tcp://
automatically if http:// isn't specified
- Fix various symbol size change warnings
- Import jansson updates from ckpool
- Check for attempts to copy to/from null in cg_memcpy
- Use cg_memcpy throughtout bitfury driver
- Use cg_memcpy throughout cgminer.c and util.c
- Use alloc helper in miner.h
- Use alloc helpers in bitfury driver
- Use alloc helpers in spondoolies drivers
- Use alloc helpers in bflsc driver
- Align_len in all alloc helper calls
- Use align_len in all alloc helpers
- Use alloc helpers in usbutils
- Use alloc helpers in icarus driver
- Use alloc helpers in avalon4 driver
- Use alloc helpers in api.c
- Use the alloc helpers in util.c
- Convert all users of alloc to use the helpers in cgminer.c
- Add helper functions to various alloc functions that automatically check for
failure
- Fix rare dereference error on pool stratum queue

It's this line that may (or may not) identify the issue:

[2015-06-03 21:13:16] BitForceSC detect (5:1) Warning unknown firmware ' 1.4.50x00' using Ver2

Oh, Joy.

It does work, kind of.  80 or 90 Ghs, 120Ghs sometimes briefly.  Also the temp goes all the way up to 50 or so as if it was mining full speed.

Thanks,
Sam

Thanks. It's recognising it fine and running it as a monarch so if it's not working then they changed the protocol and, well, there's no sign of any updated protocol documentation and I do believe all communication channels between the cgminer development team and BFL have effectively been severed so... fuck.

Here's the portion that was relevant last time.  Let me know if you need the entire dump.
Thanks,
Sam

My guess is yet another USB identity to confuse cgminer further, possibly even intentionally from their end. Do the usual and get the dump of adding "-D -T 2>log.txt" to your cgminer commands on startup and dump the contents of log.txt here to see what cgminer makes of it.

My old monarch died quite a while ago - well it stopped working each time I tried it for a few days and I gave up on it.
Try a few older versions of cgminer and see if any of the changes to the BFL drivers caused a problem that needs to be fixed.
Otherwise I guess they could have changed the hardware ... and, well, that means bye-bye to having the monarch work with cgminer.

That's not cgminer.
There never has been and never will be --bmsc options in cgminer.
It's some manufacturers version you were running before.

If you ran it on the command line it will tell you that there's a problem with your options.
Something like:
[2015-06-07 13:06:24] ./cgminer: --bmsc-options: unrecognized option

With the official master cgminer 4.9.1 you most probably want to be using the icarus driver options if that is a bitmain U1/U2/U3
However, as with all of our drivers we write, you don't have to supply any device options to make them work in default settings.

For the U3, the options (if you wish to change the default working options) are --au3-freq, --au3-volt and --icarus-options
See the ASIC-README
https://github.com/ckolivas/cgminer/blob/master/ASIC-README

I had a Monarch die so I had to send it in.  It came back from repair and now it only hashes around 70 to 80 Ghs with CGMiner 4.9.1 with the replacement .exe.  If I use the "special" version of BFGMiner on a 64 bit machine it hash's fine.  It used to hash fine with CGMiner too.  Any idea's?
Thanks,
Sam

When he wrote and saved the original bat file, it was saved in the 4.8* folder and most likely has that path defined in the hidden part of the titled path. Try deleting the bat file in 4.9*. Navigate to 4.8*,  open the old bat file in edit, then left click and drag over its text, select copy. Close that session. Open the 4.9* CGminer folder and create a new text file in it. Open that file with notebook/edit and right click to Paste your text/only from the original bat file. Now save it in 4.9* as a bat and see if it works.

Quote- So the only thing i did was taking the .bat file from 4.8.0 and place it in the 4.9 and 4.9.1 ( tested both ) but cgminer wont open, yeah well just for 0.5 sec and it close directly.

That's what you do 2nd.  First you run your original command line in a DOS window and see what you get.

like: cgminer.exe -D --verbose --bmsc-options 115200:0.57 -o stratum+tcp://hashgoal.com:3332 -u 2XdsyV7ThMVPp65PxSmWmjpoL1ctW1rJDk -p x --bmsc-voltage 0800  --bmsc-freq 1286
?
because of i do this, i get nothing to see.