OLD: BFGMiner 4.10.0: GBT+Stratum, RPC, Mac/Linux/Win64, Spondoolies SP30

[tripzero]

I've got a couple BFL monarchs and using bfgminer 4.8.  At first both monarchs worked fine.  Later, bfgminer stopped and I had to restart it.  I couldn't bring up one of the monarchs after that.  Bfgminer reports:

"[2014-09-13 09:35:54] ModMiner detect: no response to version request from /dev/ttyUSB0""

The other monarch still works fine.  I've rebooted several times and even tried downgrading bfgminer to 4.7.  No luck.  Any suggestions?  Is there a "hello world" command I can toss at the device using "screen" (screen /dev/ttyUSB 9600?) or "echo" just to see if the device is alive?

[felixthecat]

Hey Luke,

Have you made any progress with the S1 drivers?  There are a few guys trying to get an S1 blade to work with BFGMiner or CGMiner; in the later pages they are either seeing errors or all 32 chips aren't running:

https://bitcointalk.org/index.php?topic=671128.140

Bryan

No, I was waiting on docs...
It's interesting that it even begins to work with that driver at all.
Any idea what the real S1 host system is using for its USB chipset? The VID/PID doesn't come up in /usr/src/linux

Thanks for the reply...

If you go to the link I believe that are using a UART adapter. The hardware guy is going to buid a prototype to control 4 or 8 blades.

The USB chipset I do believe you are looking for is http://am.renesas.com/products/soc/usb_assp/product/upd720114/index.jsp I sure hope this helps you out as I am dying to get rid of the POS bitmain version of cgminer on my S1's....

Linux® USB 2.0 Driver

Linux USB 2.0 driver is supported from kernel 2.4.19 or later. Kernel is updated regularly, and the problems in older version may be solved in the latest kernel. So it is desirable to use the latest one. The following URL is for your reference.

http://www.linux-usb.org

For Kernel 2.4.26 patch, refer to this page.

[J4bberwock]

Hey Luke,

Have you made any progress with the S1 drivers?  There are a few guys trying to get an S1 blade to work with BFGMiner or CGMiner; in the later pages they are either seeing errors or all 32 chips aren't running:

https://bitcointalk.org/index.php?topic=671128.140

Bryan

No, I was waiting on docs...
It's interesting that it even begins to work with that driver at all.
Any idea what the real S1 host system is using for its USB chipset? The VID/PID doesn't come up in /usr/src/linux

As far as I know, it's the pic32 chip that is doing the UART>USB job.
Using a third party USB>UART converter with cp2102 allows me to have it hashing (more or less) recognised as a U1/U2 since it's using the cp2102 too.
At the moment, only 8 chips are found with the default bfgminer build, and it seems I can't adjust frequency.

[oyhmxh]

Good job pal! Will have a test of this. Thanks!~

[felixthecat]

Hey Luke,

Have you made any progress with the S1 drivers?  There are a few guys trying to get an S1 blade to work with BFGMiner or CGMiner; in the later pages they are either seeing errors or all 32 chips aren't running:

https://bitcointalk.org/index.php?topic=671128.140

Bryan

No, I was waiting on docs...
It's interesting that it even begins to work with that driver at all.
Any idea what the real S1 host system is using for its USB chipset? The VID/PID doesn't come up in /usr/src/linux

As far as I know, it's the pic32 chip that is doing the UART>USB job.
Using a third party USB>UART converter with cp2102 allows me to have it hashing (more or less) recognised as a U1/U2 since it's using the cp2102 too.
At the moment, only 8 chips are found with the default bfgminer build, and it seems I can't adjust frequency.

The USB_ID is sent from the MCU via Pin 41 to the USB controller which in turn basically the UARTS are technically two USB ports just with a couple extra things as temp and fan control.
Pins 3 RX and 5 TX on the UART with Pin 1 being Fan/Power Management and Pin 4 being ground, 6 and 7 being temperature control pin 8 is ground. So with all this mind the UART plays a dual function, hosting the USB/Power and then the temperature control settings.

[mill0601]

any word on technobit integration? It's the feature i've been waiting for so i can finally mine mediterranean coin.

[WBF1]

Need some help with load-balace setup.

I have 4 pools in my config file. I have 2 of them set to quota of zero and the other two are each set at quota of 50.

All 4 pools are enabled.

My management strategy is set to load-balance.

Expected result: each of the two "quota-50" pools gets half my hashrate unless both are dead in which case we fall back into priortiy mode with failover

Actual result: whatever pool has a higher priority set gets all the hashrate

NOTE: I simplified to a 2 pool setup each with a quota of 50  and I get the same actual result.

Am I thinking about this wrong? Is there a setting I missed? or is load-balance broken?

EDIT: A little more details here... I'm seeing work slowly get queued for whatever pool is lower priority, but no shares are ever submitted. Not sure if that's useful information.

[Axiste]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

[Luke-Jr]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

It works pretty much immediately for most devices.
"2D work" devices (Stratum proxy and Avalon2/3) do not support load balancing at this time.

[Axiste]

well that shuts me down then lol

I could only report from what occurs for me with my little Antminer U1 

[WBF1]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

It works pretty much immediately for most devices.
"2D work" devices (Stratum proxy and Avalon2/3) do not support load balancing at this time.

ah okay I thought it somehow divided up the work.

Also, yes this is with stratum proxy so I guess load-balancing is completely out the window then right?

I guess in lieu of that, is there a command line I can use with cron to switch pools at certain times of the day? I assume it would be some command line rpc interface... just not familiar enough with that or where to start.

[Luke-Jr]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

It works pretty much immediately for most devices.
"2D work" devices (Stratum proxy and Avalon2/3) do not support load balancing at this time.

ah okay I thought it somehow divided up the work.

Also, yes this is with stratum proxy so I guess load-balancing is completely out the window then right?

I guess in lieu of that, is there a command line I can use with cron to switch pools at certain times of the day? I assume it would be some command line rpc interface... just not familiar enough with that or where to start.

There's essentially no way to do 2D work stuff with load balancing, as the devices create their own work.
Have you tried something like --rotate? You will probably need to ensure the pools change jobs at least as often as you switch between the pools...

[WBF1]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

It works pretty much immediately for most devices.
"2D work" devices (Stratum proxy and Avalon2/3) do not support load balancing at this time.

ah okay I thought it somehow divided up the work.

Also, yes this is with stratum proxy so I guess load-balancing is completely out the window then right?

I guess in lieu of that, is there a command line I can use with cron to switch pools at certain times of the day? I assume it would be some command line rpc interface... just not familiar enough with that or where to start.

There's essentially no way to do 2D work stuff with load balancing, as the devices create their own work.
Have you tried something like --rotate? You will probably need to ensure the pools change jobs at least as often as you switch between the pools...

Yeah what I was hoping to achieve was sending something like 95% of my work to my regular pool and 5% of my work to ck's solo mining "pool"... sort of playing the lottery with 5% of my hash power. I guess I could rotate at some interval with 20 pool entries, one of which was the solo pool and the other 19 are my regular pool? Is something like that even allowed, config wise? (i.e., can I have multiple pool entries with exactly the same url, user, password?)

Other than that, I was thinking a cron job to switch pools for 5% of the day would work.

This all hinges on the idea that sending work to a solo mining effort for a short period of time even makes sense.

[Luke-Jr]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

It works pretty much immediately for most devices.
"2D work" devices (Stratum proxy and Avalon2/3) do not support load balancing at this time.

ah okay I thought it somehow divided up the work.

Also, yes this is with stratum proxy so I guess load-balancing is completely out the window then right?

I guess in lieu of that, is there a command line I can use with cron to switch pools at certain times of the day? I assume it would be some command line rpc interface... just not familiar enough with that or where to start.

There's essentially no way to do 2D work stuff with load balancing, as the devices create their own work.
Have you tried something like --rotate? You will probably need to ensure the pools change jobs at least as often as you switch between the pools...

Yeah what I was hoping to achieve was sending something like 95% of my work to my regular pool and 5% of my work to ck's solo mining "pool"... sort of playing the lottery with 5% of my hash power. I guess I could rotate at some interval with 20 pool entries, one of which was the solo pool and the other 19 are my regular pool? Is something like that even allowed, config wise? (i.e., can I have multiple pool entries with exactly the same url, user, password?)

Other than that, I was thinking a cron job to switch pools for 5% of the day would work.

This all hinges on the idea that sending work to a solo mining effort for a short period of time even makes sense.

Everything there makes sense, except using a centralised pool for solo mining.

I haven't thought it through yet, but maybe I could have quotas act as multipliers in rotate mode... so quota 2 with rotate of 5 minutes results in a 10 minute rotation time for that pool. Thoughts?

[Taugeran]

seems logical

[WBF1]

How long are you waiting to see the load balance results?

It wont mine to both pools at the same time, it will switch between the two for an equal amount in a 24 hour period.

So it will give 100% hashrate to pool A for up to 12 hours, then 100% to pool B for the other 12 hours.

It works pretty much immediately for most devices.
"2D work" devices (Stratum proxy and Avalon2/3) do not support load balancing at this time.

ah okay I thought it somehow divided up the work.

Also, yes this is with stratum proxy so I guess load-balancing is completely out the window then right?

I guess in lieu of that, is there a command line I can use with cron to switch pools at certain times of the day? I assume it would be some command line rpc interface... just not familiar enough with that or where to start.

There's essentially no way to do 2D work stuff with load balancing, as the devices create their own work.
Have you tried something like --rotate? You will probably need to ensure the pools change jobs at least as often as you switch between the pools...

Yeah what I was hoping to achieve was sending something like 95% of my work to my regular pool and 5% of my work to ck's solo mining "pool"... sort of playing the lottery with 5% of my hash power. I guess I could rotate at some interval with 20 pool entries, one of which was the solo pool and the other 19 are my regular pool? Is something like that even allowed, config wise? (i.e., can I have multiple pool entries with exactly the same url, user, password?)

Other than that, I was thinking a cron job to switch pools for 5% of the day would work.

This all hinges on the idea that sending work to a solo mining effort for a short period of time even makes sense.

Everything there makes sense, except using a centralised pool for solo mining.

I haven't thought it through yet, but maybe I could have quotas act as multipliers in rotate mode... so quota 2 with rotate of 5 minutes results in a 10 minute rotation time for that pool. Thoughts?

That'd make sense. If you commit some changes towards that sort of setup, I can pull some code from git, compile, and test.

re: using a pool for solo mining... it just saves the hassle of setting up all that is required for solo mining on my end. Also, it's probably hosted on a faster connection etc etc etc.

Also I posted this elsewhere, but I'm still trying to determine theorhetically if say 1/8 of my hash power for 24 hrs is better or worse than 8/8 of my hash power for 3 hrs, in terms of solo mining, everything else being held equal.

[Luke-Jr]

re: using a pool for solo mining... it just saves the hassle of setting up all that is required for solo mining on my end.

All you have to do is run a bitcoind. Not much of a hassle. :|

Also I posted this elsewhere, but I'm still trying to determine theorhetically if say 1/8 of my hash power for 24 hrs is better or worse than 8/8 of my hash power for 3 hrs, in terms of solo mining, everything else being held equal.

Theoretically, it's all the same... only difference is variance.

Side note: I've also been working on a miner-oriented Bitcoin Core based on 0.9.x that I hope to release after 0.9.3 is out. In the meantime, the current build is at http://luke.dashjr.org/programs/bitcoin/files/bitcoind/luke-jr/0.9.x/0.9.3rc2.ljr20140913/

[WBF1]

re: using a pool for solo mining... it just saves the hassle of setting up all that is required for solo mining on my end.

All you have to do is run a bitcoind. Not much of a hassle. :|

Also I posted this elsewhere, but I'm still trying to determine theorhetically if say 1/8 of my hash power for 24 hrs is better or worse than 8/8 of my hash power for 3 hrs, in terms of solo mining, everything else being held equal.

Theoretically, it's all the same... only difference is variance.

Side note: I've also been working on a miner-oriented Bitcoin Core based on 0.9.x that I hope to release after 0.9.3 is out. In the meantime, the current build is at http://luke.dashjr.org/programs/bitcoin/files/bitcoind/luke-jr/0.9.x/0.9.3rc2.ljr20140913/

I know but bitcoind will take more space than I'd like to devote to it on my proxy-dedicated old eeepc 701, although I do have an external HD sitting around... maybe down the line sometime.

But yeah will bfgminer freak out if i have 19 identical pool entries?

[Luke-Jr]

But yeah will bfgminer freak out if i have 19 identical pool entries?

It shouldn't...

[WBF1]

But yeah will bfgminer freak out if i have 19 identical pool entries?

It shouldn't...

okay then i'll go that route for now to get my automatic lottery entries up and running.