Thank you ckolivas, I though something was wrong with the hardware.

When is considered the hardware errors rate too much ?

New release: Version 4.4.2, 17th July 2014

A stable/bugfix/driver update release for accumulated issues while I was away.


Human readable changelog:

- Fixed an issue where some types of file configuration parameters (like hfa-fan) would be lost when write config was chosen from the menu.
- Fixed a crash that could occur on trying to connect to a stratum server.
- Fixed write config file crashing when bitfury was compiled in.
- Fixed the "Failed to sem_timedwait" issue seen by some people.
- Fixed failover not working if one of the initial pools never responded.
- Fixed a corruption when solo mining (ALL SOLO MINERS URGED TO UPGRADE!)
- Rbox now accepts frequencies up to 400.
- Spondoolies driver updates including SP30 driver.
- Minion driver updates.
- Avalon2 driver updates.
- Fixed broken S1 compilation.
- Other low level fixes.


Full changelog:

- Fix processarg parameters loaded from a config file not being saveable
- We only use the jansson in our source tree so no need for special case
handling of older versions
- Upgrade jansson to 2.6
- Only clear sockbuf if it's been allocated
- Fix missing osm-led-mode support in write config
- Deal with nanosecond overflow in both directions on both addition and
subtration of timespecs
- Rename sp10 driver internally from spondoolies to sp10
- minion - add a 2nd (optional - disabled) reset test
- production stats added, reset queue added
- minion - correct led ghs2 choice
- minion - correct ghs2 display
- minion - reset the led counter when doing a chip RSTN full reset
- minion - don't reset the led counter for an SPI reset
- minion - led per chip and use all time average
- minion - report spi error counts and settings in stats
- minion - undeclared fix
- minion - chip power cycle option
- minion - record 0xff error history and reduce screen output
- minion - reset on result or fifo 0xff
- minion - clarify the 0 value of spireset
- minion - make SPI reset more configurable
- minion - make the SPI reset ms sleep a parameter and API settable
- sp10 sensors
- sp30
- minion - led+more api setting
- Avoid blocking all pool testing if one pool fails to ever init
- There is no point storing the hints addrinfo in struct pool
- minion - 'reset' SPI when getting errors
- initialise more pool values in benchmark
- minion - auto adjust freq
- merge upstream frequency changes
- icarus - timing history in its own function
- rbox - add lotsa stats, tidy up a bit more
- Fix an off-by-one.
- icarus - detect stat should be LOG_DEBUG
- icarus - tidy up rbox code, remove statics, and add rocketbox
- minion - do an early reset to clear the chip status
- minion - use descriptive names for the list types
- Avalon2: automatic adjust fan speed, using crc16 on job_id compare, turn on
the led by API, detect twice when start, remember the last stratum message
increase the hashrate, add cutoff option
- fix AntS1 breakages from AntS2 changes
- minion - disable dup nonce check
- minion - add an ioseq number for each ioctl to simplify work ordering
- minion - redo old work expiry based on txrx order
- minion - more work stats, minimise queued work, free flushed queued work
- minion - allow resetting a chip via the API
- minion - correct 'WQue Count' in stats
- minion - delay after reset, reset history also, add dups to api stats
- noncedup - give access to the internal stats
- minion - increase reset to 75%
- minion - dup checking, disable reread by default and extra ioctl debugging
- minion - always check the chip queue before queuing new work

Hello Con,

pthread_tryjoin_np was working great for last two days or so. Cgminer starts up for a sec or two:)

However you are removing it due to:
---------------------
Remove the use of the pthread_tryjoin_np which is currently unimplemented on many platforms
---------------------

If you can make it #ifdef where this is implemented will be great.

PS: i have not tested two latest commits after 4.2.2 release yet

Best

Yeah i got your point
Something like
#ifndef WIN32
 pthread_tryjoin_np
#else
...
#endif
What about it?
PS: if last two commits do same job then this is not needed of course
 

This is a bitcoin miner. All altcoin variants are forks.

Making some progress with the Terraminer. I hooked up both boards directly to my PC, installed WinUSB, and then started up version 4.3.5 using the --ps-load 80 parameter since I used to run it at power step 8.

Cgminer recognizes the boards and they get an initial hash rate of about 200g for a second on each board and the over about a minute it goes down to 0.

I get a constant output of checksum bad for each board.

The miner registers on my pool for about 2 minutes and then nothing

Is there a parameter I am missing?

Any suggestions?

Yes! I think that might have done it....thanks!

Thanks! Appreciate all the help.

Try

Instead

If you want to append the log file, use "tee -a".

Create a script with the redirect in it and call that from screen.