HASHRA - LUNAR SERIES - ASIC SCRYPT MINERS

[crocko]

Sorry, my fault, my PSU has 6+2 pin cables, and I used all 8 of them.
Now I use 2 cables with only 6 pin and the ASIC it is started.
Thank you very much for your help !

[coinwarrior]

great! Good luck with everything! Feel free to reach out with further questions you might have, I think its great if we can share out experiences and ideas here. If you update your controla please make a full reboot - I had to cut power to the Po to get working again....

[jjj0923]

Sorry, my fault, my PSU has 6+2 pin cables, and I used all 8 of them.
Now I use 2 cables with only 6 pin and the ASIC it is started.
Thank you very much for your help !

you know - I hate to read instructions myself: LOL - but the instructions that come with the unit specificically tell you not to plug in all 8 pins...

glad you got it sorted out.

[crocko]

Sorry, my fault, my PSU has 6+2 pin cables, and I used all 8 of them.
Now I use 2 cables with only 6 pin and the ASIC it is started.
Thank you very much for your help !

you know - I hate to read instructions myself: LOL - but the instructions that come with the unit specificically tell you not to plug in all 8 pins...

glad you got it sorted out.

I didn't received any manual or any paper with instructions..
May they are posted on the Internet and I missed these.
My fault ! 
Anyway I know from the Hashra's support that the top diff for my LLW1 is 800.
I mine on wemineltc.com and on this pool you can setup the diff.
I setup the diff for my worker between 700-800 and I have 0 rejected, and only few HW errors.
BTW, I did the upgrade to my CONTROLA, nothing bad happened.

I must agree, if you don't have a cooled space, it is impossible to run several ASICs.
The SUMMER is comming !

[jjj0923]

my bad - you are correct - nothing comes with the unit. I printed them from their support page when I got mine.

[jjj0923]

does anyone besides find find their Lander incredibly slow coming up to speed on the screen?

it seems like it often takes 10 to 15 minutes for the Controla screen to display information accurately.

some times it never seems to come up to speed and I have to 'restart' (not reboot)  it multiple times to get it to report hash rates properly - it's very annoying.

[davecoin]

Sorry, my fault, my PSU has 6+2 pin cables, and I used all 8 of them.
Now I use 2 cables with only 6 pin and the ASIC it is started.
Thank you very much for your help !

you know - I hate to read instructions myself: LOL - but the instructions that come with the unit specificically tell you not to plug in all 8 pins...

glad you got it sorted out.

I didn't received any manual or any paper with instructions..
May they are posted on the Internet and I missed these.
My fault ! 
Anyway I know from the Hashra's support that the top diff for my LLW1 is 800.
I mine on wemineltc.com and on this pool you can setup the diff.
I setup the diff for my worker between 700-800 and I have 0 rejected, and only few HW errors.
BTW, I did the upgrade to my CONTROLA, nothing bad happened.

I must agree, if you don't have a cooled space, it is impossible to run several ASICs.
The SUMMER is comming !

What's your hw error % on stock speed (330)?

I'm curious because I've gone all the way down to 281 and still have between 5% and 6% errors.

[davecoin]

I have to leave to work soon, but my napkin math guess is somewhere around 350 watts.  I'll try to mess with it tomorrow and post results again.

Can't really use napkin math for this because gridseed blade power use isn't even close to linear for overclocking.  I assume the Hashra chips are the same way and they are shipped to us already in that outside of efficiency power band state.

So the lowest clock speed is 250, at which the miner pulls 398 watts at the wall.  ~11.1 mh/s.  Voltage obviously needs to be adjusted for this miner to really be efficient.

Unfortunately there's still ~6% HW errors so I can't justify running it at a lower clock speed.

[crocko]

Sorry, my fault, my PSU has 6+2 pin cables, and I used all 8 of them.
Now I use 2 cables with only 6 pin and the ASIC it is started.
Thank you very much for your help !

you know - I hate to read instructions myself: LOL - but the instructions that come with the unit specificically tell you not to plug in all 8 pins...

glad you got it sorted out.

I didn't received any manual or any paper with instructions..
May they are posted on the Internet and I missed these.
My fault !  
Anyway I know from the Hashra's support that the top diff for my LLW1 is 800.
I mine on wemineltc.com and on this pool you can setup the diff.
I setup the diff for my worker between 700-800 and I have 0 rejected, and only few HW errors.
BTW, I did the upgrade to my CONTROLA, nothing bad happened.

I must agree, if you don't have a cooled space, it is impossible to run several ASICs.
The SUMMER is comming !

What's your hw error % on stock speed (330)?

I'm curious because I've gone all the way down to 281 and still have between 5% and 6% errors.

I don't know in percent but there was couple thousands  accepted and only ~200 HW errors, 0 Rejected.
It seems I can mine only 1.3 LTC per day with this configuration...
The chance of ROI is far, far away....

[davecoin]

If you had 2000 accepted and 200 hardware errors, that means 10% of your hashrate is totally wasted.  That's not good.  Your 14mh/s miner is effectively a 12.6mh/s miner.

[tuanvie]

I was watching it, it seems very good, I will order some if it is need

[crocko]

If you had 2000 accepted and 200 hardware errors, that means 10% of your hashrate is totally wasted.  That's not good.  Your 14mh/s miner is effectively a 12.6mh/s miner.

The clock speed is down to 310 and I have now 500 Accepted, 0 Rejected, 2 HW

[andycee]

Great thread..a few comments...


My 14M/H unit arrived promptly, direct from China pretty much when Hashra said it would be released. Comms really good all the way.

Dodgy USB PI power cable, as reported - use a spare USB phone adapter.

There was NO import tax to pay (to UK) in this instance. Unit shipped as a sample.

A few people complaining about the price $/MH for these units.
Mmmmhh....
4 x R9 280x @ 3.2M/H @ 1100W (plus PC hardware and loads of heat and noise)
 OR
1 x Lunar Launcher @ 14M/H @ 450W (plus PSU and even more noise  )

Yes I know some of these figures might not be exact, and yes R9 280xs have come down in price, but you get my drift ?

I'm a very happy customer, pleased I made the shift. GPU's all being sold off

Only issue will be if this unit dies after 90 days...

[jjj0923]

Great thread..a few comments...


My 14M/H unit arrived promptly, direct from China pretty much when Hashra said it would be released. Comms really good all the way.

Dodgy USB PI power cable, as reported - use a spare USB phone adapter.

There was NO import tax to pay (to UK) in this instance. Unit shipped as a sample.

A few people complaining about the price $/MH for these units.
Mmmmhh....
4 x R9 280x @ 3.2M/H @ 1100W (plus PC hardware and loads of heat and noise)
 OR
1 x Lunar Launcher @ 14M/H @ 450W (plus PSU and even more noise  )

Yes I know some of these figures might not be exact, and yes R9 280xs have come down in price, but you get my drift ?

I'm a very happy customer, pleased I made the shift. GPU's all being sold off

Only issue will be if this unit dies after 90 days...

I sold off all of my gpu's on craigslists a few weeks ago.

these are very fickle units - I often find myself having to restart my Landers a few times before they want to run at speed. it's as if all the chip don;t want to join the party.

they are lot of contortions going on behind the scenes with waits, timeouts, restarts etc in the controla software to get the units to cooperate - much much different than press a button and mine with the gridseeds, I'm used to. I know I've loooed at the code and the scripts in the controla software.

one thing is for sure - we need a more stable version of the firmware that runs these things...

I've got a lot of documentation showing how once these things in 'slow mode' they never recover and needs multiple restarts to get going and sometimes a reboot and I'm certainly not down on power - both of my units are being off gold 1,200 psu's.

[iikun]

@Hashra, is the firmware okay to upgrade yet?  I don't want to upgrade & have any issues

[jjj0923]

I did upgrade one of my two units last night and it's been running fine but I did have to reboot it after upgrading

[iikun]

I did upgrade one of my two units last night and it's been running fine but I did have to reboot it after upgrading

ok, cool.  was there any performance or stability boost do you think?

[jjj0923]

I did upgrade one of my two units last night and it's been running fine but I did have to reboot it after upgrading

ok, cool.  was there any performance or stability boost do you think?

my only observation(and this has not been proven with multiple tests) is that the unit did not need to be restarted multiple times after changing pools, like I've experienced with the stock firmware.

oh and even though it reboots itself after the update - I neede to reboot it a second time because after it's own reboot all I say was this  in the syslog file:

Jun 12 14:18:54 hashracontroller cgminer: Started cgminer 3.1.1
Jun 12 14:18:59 hashracontroller single_ltc_monitor[2201]: Started LTC PRocess waiting 10 seconds to boot cgminer
Jun 12 14:18:59 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:18:59 hashracontroller cgminer: Icarus Detect: Test failed at /dev/ttyUSB0: get 00000000, should: 00038d26
Jun 12 14:18:59 hashracontroller cgminer: All devices disabled, cannot mine!
Jun 12 14:19:04 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:09 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:14 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:19 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:24 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:29 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:34 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:39 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:44 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:49 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null

Jeff

[maxzilla]

I did upgrade one of my two units last night and it's been running fine but I did have to reboot it after upgrading

ok, cool.  was there any performance or stability boost do you think?

my only observation(and this has not been proven with multiple tests) is that the unit did not need to be restarted multiple times after changing pools, like I've experienced with the stock firmware.

oh and even though it reboots itself after the update - I neede to reboot it a second time because after it's own reboot all I say was this  in the syslog file:

Jun 12 14:18:54 hashracontroller cgminer: Started cgminer 3.1.1
Jun 12 14:18:59 hashracontroller single_ltc_monitor[2201]: Started LTC PRocess waiting 10 seconds to boot cgminer
Jun 12 14:18:59 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:18:59 hashracontroller cgminer: Icarus Detect: Test failed at /dev/ttyUSB0: get 00000000, should: 00038d26
Jun 12 14:18:59 hashracontroller cgminer: All devices disabled, cannot mine!
Jun 12 14:19:04 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:09 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:14 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:19 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:24 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:29 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:34 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:39 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:44 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:49 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null

Jeff

The trick for rebooting is, first reboot the miner, and then reboot the pi. The miner has to be reset completely and alive, before connecting the pi to it.

[iikun]

I did upgrade one of my two units last night and it's been running fine but I did have to reboot it after upgrading

ok, cool.  was there any performance or stability boost do you think?

my only observation(and this has not been proven with multiple tests) is that the unit did not need to be restarted multiple times after changing pools, like I've experienced with the stock firmware.

oh and even though it reboots itself after the update - I neede to reboot it a second time because after it's own reboot all I say was this  in the syslog file:

Jun 12 14:18:54 hashracontroller cgminer: Started cgminer 3.1.1
Jun 12 14:18:59 hashracontroller single_ltc_monitor[2201]: Started LTC PRocess waiting 10 seconds to boot cgminer
Jun 12 14:18:59 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:18:59 hashracontroller cgminer: Icarus Detect: Test failed at /dev/ttyUSB0: get 00000000, should: 00038d26
Jun 12 14:18:59 hashracontroller cgminer: All devices disabled, cannot mine!
Jun 12 14:19:04 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:09 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:14 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:19 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:24 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:29 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:34 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:39 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:44 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null
Jun 12 14:19:49 hashracontroller single_ltc_monitor[2201]: socket connect(127.0.0.1,4001) failed returning null

Jeff

right, ok thanks.
I don't seem to have a big issue with restarts on the original firmware so I might just leave it. I took the clock down to 301 & using 50w less power for minimal loss of hash. Now if I could only do something about that infernal noise it makes :/  Though I suspect its as people say & its more the heatsink airflow rather than the fan which is the cause of the issue