Topic: Avalon ASIC users thread - page 200. (Read 438378 times)

He paid about BTC 114 for the machine at the time (last week of September 2012)

The price in $USD is $USD 1359 ($USD 1299 + $USD 60 shipping), so 5 or 6 days at the current exchange the purchase breaks even in $USD if he mines about BTC 8 / day

The price in BTC is BTC 114, so unless difficulty skyrockets dramatically, a Batch #1 machine can break even in BTC in about 15 days.

Now..

for Batch #2, wave 0, a machine cost $USD 1500 or BTC 76.2 at the time
for Batch #2, wave 1, a machine cost $USD 1500 or BTC 56.45 at the time

Time will tell when these are delivered and what the $USD/BTC exchange is and more important difficulty. I hope they arrive before BFL starts shipping ;-)

Congrats! You won the asic miner lottery!

4-5 days i guess, he is doing about 8 btc a day

How long did that take?

and......the avalon has paid for itself in USD (paying for itself in BTC will take a lot longer)

*high fives* avalon.

Have you tried the BFGMiner Avalon firmware? This should work with both Avalon and BitForce devices.

The current Avalon code will ONLY work with an Avalon (not anything else) ... as I said when I first saw the code.
Yes you need to run two separate cgminers.
One with the Avalon only code (that doesn't support other devices ... unlike what Avalon said it would do ... not surprising at all) and one that is based on the current proper cgminer code.

I've no idea what the USB identification of an Avalon is ... I don't have one nor will I ever
If it overlaps with the BFL or MMQ device (same Vendor/Product as one of them) then yes there will be problems.
Unfortunate ...

ttyUSb is different from Avalon one right?
and binary name is different?
Maybe BFL cgminer is probing somehow USb com ports and fucks avalon sending it crap to see if this is BFL?

1. Disable hotplug in your cgminer BFL config
2. Kill avalon cgminer
3. Run BFL miner
4. Run Avalon Miner
But all that is damn stupid just buy another router
PS: You need just to ask Kano or someone which knows cgminer better and be 100% that when you run your BFL instance it is not sending crap to avalon to probe if it is BFL that is the problem as you are describing it

That IS what i'm doing. I've compiled a NEW version of cgminer for the 703n. And I'm trying to run a separate instance of cgminer. The avalon is still using the original version of cgminer.



Generally, I'm not shutting down my avalon. I compile cgminer on a separate linux machine, upload the binary and test. If the avalon cgminer hangs, it's a matter of seconds to restart the original cgminer process.

Just buy one Wr-842ND and do whatever you want. In current situation every minute downtime counts.

at the moment you need another cgminer instance. currently you can't reach your goal...

I'm trying to run a BFL single from the avalon. I'm not messing with the avalon miner itself.

kaerf, what on earth are you trying to do? Your stats look fine. They report 67 Ghash/s. 3439 getworks indicate cgminer works on shares of an average difficulty of 428. There is no need to recompile cgminer.

I know the field is currently limited, but has anyone else tried to do some stupid shit like me and attempted to get a custom version of cgminer running on the avalon?

I compiled a new version of cgminer without avalon support but with bitforce support (since the avalon version of cgminer is incompatible with other hardware). Everything compiles, but when run it I get "BFL0: invalid nonce - HW error"

The post at https://forums.butterflylabs.com/fpga-single-minirig-support/266-bitforce-single-fails-2.html seems to point at some bad/incompatible usb issues. My avalon miner also tends to crash after I plug a bitforce into the usb hub.


Edit: this post is separate from my previous post about stats. i'm not trying to tweak the avalon with a new cgminer. i'm trying to use the avalon as a host for a bfl single.

Initially, I think my avalon was restarting every now and then, but I upgraded the firmware.

Here are some stats after running for a day.



Not quite sure what the expected stats should be. Anyone know?

You can use a root password for ssh, so you could actually bypass the priv/pub in your first go.

you will have to use telnet / http first to edit passwd. at the first ssh login, the priv/pub generation would take ~10s for 703n, have fun waiting XD

http://www.ipnetcontrol.net/en/item-products-en/p-4r1t1a-bg

No English version you can use google translate from Bulgarian

There are two options
1. power device can be controlled via snmp and power off / on each of the four power lines independently. It has web interface also which can be used for power off/on manually. The process of monitoring and restarting four independent power lines can be automated via shell sell script

2. Automatic way that is what i am going to use first when more Avalons come one of them will monitor the rest and control power lines via snmp accordingly
How it will be done Power Device has a built in function to monitor every minute via ping  given IP address and when ping fails it can restart (power off/on) one of the configured power lines where avalon unit is connected. So there is built in script inside avalon which does soft restart when accepted shares are not changing or when cg miner is dead


#!/bin/sh
# This file is for cron job

C=`pidof cgminer | wc -w`
if [ "$C" != "1" ]; then
/etc/init.d/cgminer stop
/etc/init.d/cgminer start
exit 0;
fi

A=`cat /tmp/cm.log`
B=`cgminer-api | grep "^ \[Accepted\]"`
cgminer-api | grep "^ \[Accepted\]" > /tmp/cm.log
if [ "$A" == "$B" ]; then
/etc/init.d/cgminer stop
/etc/init.d/cgminer start
exit 0;
fi
https://github.com/BitSyncom/cgminer-openwrt-packages/blob/master/cgminer/files/cgminer-monitor

The quick and dirty hack to this script is changing
/etc/init.d/cgminer stop
/etc/init.d/cgminer start
to ip a f dev br-lan or whatever interface name is
then power device will not be able to ping avalon unit and within a minute it will be rebooted powerd off and on:)

Nice easy and clean:)

Can you explain how you are going to accomplish this?

Kano,

There are some rumors here that not every time soft restart solves the problem. There might be something wrong with controller itself. I have not heard of any soft way to restart the controller yet. I have asked this many times...so to be on a safe side we have to use power plug:) having said all this it actually does not matter how you are going to soft restart it when it will not work again. I took care of this already and when i got my Avalon it will be powered off and on automatically if any problem occurs. As long it does not have HD like PC it is not hurting at all:)

About the default API settings - most of 99% the guys here do not care at all. And 99% of the cases unit will be behind NAT - soho router modem or whatever this is by design i think. Having that in mind i can say that here is 0% problem with default settings. For those of us who know what we want and how we want it we will do it. change files use iptables, vpn, miner.php and all fancy stuff that i have in mind