Official Avalon Technical Support Thread

[loshia]

It is a good thing we have beta testers that allow folks to wait until these are "stable builds". The Firmware seems to be improving quite a bit.

@ BitSyncom

Any change the Corsair power supply can be added to the Wiki as a viable solution?

1. The FW is making about 1-1.5 GH more
2. It still needs power off/on but it is not happening so frequently compared to the old FW
3. Results:    Elapsed:3days 08h 49m 21s (I never had such uptime - max 2 days),Difficulty Accepted:4,810,991.00, Rate based on diff1 Accepted shares 71016.56 MH/s clocked at 300

[1713983972]

1713983972

[1713983972]

1713983972

[1713983972]

1713983972

[1713983972]

1713983972

[1713983972]

1713983972

[thorvald]

2 days 19h 22m without restart latest firmware 20130419
the hardware errors are at about 8% from at about 5% up from firmware version 20130225
discarded at about 20%  nmw  824
rejected 0.8%

[loshia]

2.19h days without restart latest firmware 20130419
the hardware errors are at about 8% from at about 5% up from firmware version 20130225
discarded at about 20%  nmw  824
rejected 0.8%

The HW errors are higher because Avalon just recovers from a dead loop. And they accumulate. It works better in my unit for sure
Your rejects are bit higher. I do not know where are you located but if you like you can try bitminter

[Sitarow]

It is a good thing we have beta testers that allow folks to wait until these are "stable builds". The Firmware seems to be improving quite a bit.

@ BitSyncom

Any change the Corsair power supply can be added to the Wiki as a viable solution?

1. The FW is making about 1-1.5 GH more
2. It still needs power off/on but it is not happening so frequently compared to the old FW
3. Results:    Elapsed:3days 08h 49m 21s (I never had such uptime - max 2 days),Difficulty Accepted:4,810,991.00, Rate based on diff1 Accepted shares 71016.56 MH/s clocked at 300

Do you know what your power consumption was?

59769.72 Shares Per Hour is nice.

Unit #114 does 60416 diffA accepted shares per hour - 71795 MH/s clocked at 300
Unit #445 does 59268 diffA accepted shares per hour - 70286 MH/s clocked at 300

[loshia]

It is a good thing we have beta testers that allow folks to wait until these are "stable builds". The Firmware seems to be improving quite a bit.

@ BitSyncom

Any change the Corsair power supply can be added to the Wiki as a viable solution?

1. The FW is making about 1-1.5 GH more
2. It still needs power off/on but it is not happening so frequently compared to the old FW
3. Results:    Elapsed:3days 08h 49m 21s (I never had such uptime - max 2 days),Difficulty Accepted:4,810,991.00, Rate based on diff1 Accepted shares 71016.56 MH/s clocked at 300

Do you know what your power consumption was?

59769.72 Shares Per Hour is nice.

Unit #114 does 60416 diffA accepted shares per hour - 71795 MH/s clocked at 300
Unit #445 does 59268 diffA accepted shares per hour - 70286 MH/s clocked at 300

Nope. I have not mesured it. I am driving my avalon with stock 650W psu. However some 850W Corsair are on the way 

[cypherdoc]

Edit: or the usb communication to the controller. If this happens again I will open the unit up and unplug the usb and plug it back in. Unless that is a bad idea [/b]

Note Unit # 114 stalled after 92.5 hours of operation. Waited 4 hours to see if it would fix itself but it just kept restarting cgminer.
Like Unit # 445 it only started working after a powering the unit down.

So in this situation , do you think one of those devices that people keep talking about around here that depends on a ping response from the router is useless?

[thorvald]

do you use worker min difficulty 32 ?
mine is set to 64  on btcguild   at the utility tab I have 15.5 and the tab difa 4198161

[Sitarow]

Edit: or the usb communication to the controller. If this happens again I will open the unit up and unplug the usb and plug it back in. Unless that is a bad idea [/b]

Note Unit # 114 stalled after 92.5 hours of operation. Waited 4 hours to see if it would fix itself but it just kept restarting cgminer.
Like Unit # 445 it only started working after a powering the unit down.

So in this situation , do you think one of those devices that people keep talking about around here that depends on a ping response from the router is useless?

For now the stall effects communication to the control board. SO if we could use a script to monitor the fan speed in this case 0 and then issue some kind of string command to reset the power then it would work.

For now I suppose the best you can hope for is your pool notify you after 10 mins that your unit is not reporting. Then use your android phone to "connect" to your network via vpn and you can issue a command reset via the browser to that unit.

[Sitarow]

do you use worker min difficulty 32 ?
mine is set to 64  on btcguild   at the utility tab I have 15.5 and the tab difa 4198161

diffa is your accepted shares and credited at x1 difficulty by the pool.

How much time has past during the shares submitted?

[thorvald]

the time is between 3 and 8 s

[Sitarow]

Here is the performance of Unit #114 before any upgrades.



Here is the performance of Unit #445 before upgrades

Unit #445 After 1 Hour Operation with out any upgrades


Unit #445 After 8 Hour 42M minutes Operation with only Power Supply Upgrades.



Note Unit #445 performance was a little better after power supply upgrade however it seemed short lived once it was upgraded to firmware 20130419.

[Sitarow]

the time is between 3 and 8 s

I meant was your Avalon total up time for those submitted shares 70.5 hours or 2 Days 22 Hours 30 Minutes?

I could estimate based on known Avalon past performance at 32 difficulty.

I use that information to help see what your average per hour share acceptance is. This way we can see if higher difficulty is better on the Avalon units.

Note that having a higher difficulty will help put less stress on the pool and help you by lowering your network bandwidth usage.

[thorvald]

from my test with difficulty min  32   +3% more shares and lower hardware errors then with difficulty set to min 64 ( test for about 2h)

[cypherdoc]

Edit: or the usb communication to the controller. If this happens again I will open the unit up and unplug the usb and plug it back in. Unless that is a bad idea [/b]

Note Unit # 114 stalled after 92.5 hours of operation. Waited 4 hours to see if it would fix itself but it just kept restarting cgminer.
Like Unit # 445 it only started working after a powering the unit down.

So in this situation , do you think one of those devices that people keep talking about around here that depends on a ping response from the router is useless?

For now the stall effects communication to the control board. SO if we could use a script to monitor the fan speed in this case 0 and then issue some kind of string command to reset the power then it would work.

For now I suppose the best you can hope for is your pool notify you after 10 mins that your unit is not reporting. Then use your android phone to "connect" to your network via vpn and you can issue a command reset via the browser to that unit.

if one is solo mining and has these in a data center what could one do to monitor them?

would it be better to connect via a VPN or SSH?

[PuertoLibre]

do you use worker min difficulty 32 ?
mine is set to 64  on btcguild   at the utility tab I have 15.5 and the tab difa 4198161

What is the utility number for?

I get ranges from 1000+ to less than 15 depending on my difficulty settings.

[ewibit]

Avalon is connected with WiFi
on router I see

Code:

Received Packets	Sent Packets
7489	                5195

and the number of Packets changes
(means for me that I am connected)
nevertheless sometimes I cannot connect to Luci
and cgminer does not automatically restart
only a cold reboot helps
any hints?
TIA

[loshia]

Avalon is connected with WiFi
on router I see

Code:

Received Packets	Sent Packets
7489	                5195

and the number of Packets changes
(means for me that I am connected)
nevertheless sometimes I cannot connect to Luci
and cgminer does not automatically restart
only a cold reboot helps
any hints?
TIA

Use cable. Rub out the antenna so you will never be able to switch to WiFi ever again

[spiccioli]

Hi,

I got mine stuck twice this morning, once on 50btc.com and once on p2pool.

Hashing at 0MH, only a full power-off/on restarts it.

spiccioli

[ewibit]

Use cable. Rub out the antenna so you will never be able to switch to WiFi ever again

unfortunately it is not possible with cable due to room layout...

[loshia]

Use cable. Rub out the antenna so you will never be able to switch to WiFi ever again

unfortunately it is not possible with cable due to room layout...

Do you know how much you are loosing?
It is your choice after all

Best