racebyu
|
 |
December 15, 2017, 02:28:32 AM |
|
>> Does this new software just address the restart issue?
From the release notes: "This update fixes a few bugs and adds better support for multiple sticks on a hub. The update is recommended for everyone since it adds much needed low level restart logic."
>> I have no problem running the older version when a higher current supply is used on the hub.
Me neither, until the odd moment one MLD stops, or, for some reason, they all stop. Which has happened to me once every few days. I cannot babysit these units all the time, so being able to run them unsupervised for a long time without is very very useful.
>> Getting around 5Mh per x 5 and running at 876 in BFG miner bat file, extra usb fan for cooling.
Same here. Note that hashrate = 5.66 x clockfrequency, so 876x5.66= 5Mh/s (if all cores are working). I had to trim the core voltage a bit more compared to my previous 832Mhz. I tried first 856, then I got a bit more HW errors, then 876 and a lot more HW errors (>> 1%), that gave me an idea how much to change the core voltage pot. Now it is steady at just below 1% error rate. I have got two small but powerfull USB fans blowing on them, keeps them cool. I have 6 units, and 5 more on their way (thanks to jstefanop for finding time last monday to ship them over!!!).
With ~11 MLD2's one reaches (imho) the limit of what makes sense before it is time to move up the big step to an Antminer L3+, these are about 3x more costefficient (MH/unit_cost). With the expected increases in hashing power over the next 6mo, returns will fade fast (unless the price of LTC goes to the sky).
Great thanks for the info, do you actually measure your core and memory voltage on the board? I just turned the Core voltage pots up a bit to approx .82 as per the photos at begging of the thread and the Mem left as we got them. I would like to know the settings for 5.7Mhs, Core and Memory settings and BFGminer Freq?
|
|
|
|
|
|
|
|
|
"There should not be any signed int. If you've found a signed int
somewhere, please tell me (within the next 25 years please) and I'll
change it to unsigned int." -- Satoshi
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
RdStrcklnd
Jr. Member
Offline
Activity: 70
Merit: 1
|
 |
December 15, 2017, 02:37:12 AM |
|
With ~11 MLD2's one reaches (imho) the limit of what makes sense before it is time to move up the big step to an Antminer L3+, these are about 3x more costefficient (MH/unit_cost). With the expected increases in hashing power over the next 6mo, returns will fade fast (unless the price of LTC goes to the sky).
How do you figure that 3x number? The L3+ is over 100x faster and costs a bit less than 25x as much by my calculations.
|
|
|
|
pokotylo
Newbie
Offline
Activity: 1
Merit: 0
|
 |
December 15, 2017, 05:01:49 AM |
|
I'm considering getting a USB hub for (currently, always the potential to grow..) my 2 FutureBit Moonlander 2's, any amperage in particular that would be optimal? If it helps, core voltage (for both) is at .72 and memory voltage is .79. Also, dumb question, will lower voltage lead to less heat? Thanks in advance for the help 
|
|
|
|
hercthx
Newbie
Offline
Activity: 52
Merit: 0
|
 |
December 15, 2017, 06:13:57 AM |
|
>> Getting around 5Mh per x 5 and running at 876 in BFG miner bat file, extra usb fan for cooling.
Same here. Note that hashrate = 5.66 x clockfrequency, so 876x5.66= 5Mh/s (if all cores are working). I had to trim the core voltage a bit more compared to my previous 832Mhz. I tried first 856, then I got a bit more HW errors, then 876 and a lot more HW errors (>> 1%), that gave me an idea how much to change the core voltage pot. Now it is steady at just below 1% error rate.
Hey lenny, just curios what core v you're running for 832Mhz? would like to give that a try.
|
|
|
|
lennyNO
Newbie
Offline
Activity: 76
Merit: 0
|
 |
December 15, 2017, 10:05:53 AM |
|
With ~11 MLD2's one reaches (imho) the limit of what makes sense before it is time to move up the big step to an Antminer L3+, these are about 3x more costefficient (MH/unit_cost). With the expected increases in hashing power over the next 6mo, returns will fade fast (unless the price of LTC goes to the sky).
How do you figure that 3x number? The L3+ is over 100x faster and costs a bit less than 25x as much by my calculations. L3+ is 2000$ for 500Mh/s = 4$/Mh MLD2 is 65$ for 5Mh/s = 13$/Mh = 3x more expensive (preorder price) Both need an external power (usb hub for MLD2), so that sort of evens out. The same for shipping and import duties At the current $/LTC price, power overhead is marginal. The noise level/Mh/s is probably comparable. One MLD2 is very quiet, but once you get to 10+ with external fans, it also starts to make a bit of sound. There is a point where scaling up doesn't make sense unless you can get them dead cheap 2nd hand maybe. That being said, they are massive fun. The L3+ uses 288 chips divided over 4 boards to get to 500Mh. We only need 100 MLD2 chips on 1-2 boards. Here is my proposal for the MLD3 : a 500Mh/s 100 chip unit on a single board, at halve the price of a L3+
|
|
|
|
lennyNO
Newbie
Offline
Activity: 76
Merit: 0
|
 |
December 15, 2017, 10:55:38 AM |
|
>> Great thanks for the info, do you actually measure your core and memory voltage on the board? No, I don't want to fry it. I am not too concerned about power or heat (have 2 fans) and tweaking right now.
>> I just turned the Core voltage pots up a bit to approx .82 as per the photos at begging of the thread I had ~ 0.82V core setting up to 832Mhz, but at higher clocks I start to get HW errors, so tweaked a bit more, about midway between 1/8th and 1/4th of a turn, 0.85V. I use a magnifying glass to actually look at the potmeter setting, too tiny for me without.
>> and the Mem left as we got them. I have increased that as well by 1/8th turn, to .9V setting. I was getting dropouts (mld stopping hassing). Then it disappeared (as per page 1 of this thread).
>> I would like to know the settings for 5.7Mhs, Core and Memory settings and BFGminer Freq? 5.7Mhs?? The max these things can do is ~5.3Mh/s, at highest clock (clock * 5.66). Then you need to cranck up the voltage!! I would just try one frequency step at a time, see how the HW rate increases, that should give a clue. A couple of % HW error is a tiny twist, 10% is a larger twist, 100% is a quarter twist, something like that...
I am 5Mh/s now at 875Mhz, and I think I leave it at that. Unit lifetime is also important, these things need to run >> 1yr 24x7 without burining out. If one dies, the few percent speed gain is lost big time, it's not worth it.
|
|
|
|
mrgoldy
Member

Offline
Activity: 82
Merit: 10
|
 |
December 15, 2017, 12:45:21 PM Last edit: December 15, 2017, 02:39:56 PM by mrgoldy |
|
New driver working well on Win10  working great for me also on rasberry pi with binary it's switching pools and restarting. When I start up, my HW's are .2-.5% If I switch pools, or change settings thru a proxy, my HW's jump to 3-4%, if I switch again, they jump to ~6%, switch again they jumpp to 8-9% it's like a cascading effect So there's an initial wave of HWs when pools or config switches, but then that % is held each time If I SBY, they restart at 0 and stay on the .2-.5% until I change pool/config settings. Are the HW's just visual, or is there an actual problem? In your pic above, you have 50% HW, but it's only showing a few % EDIT: And now that I've let mine run for a long time, I'm getting around 100HW to 1000Accepted, or 10%, but it's only displaying as 1%, similar to the pic above
|
|
|
|
Nicey2
Newbie
Offline
Activity: 6
Merit: 0
|
 |
December 15, 2017, 12:53:19 PM |
|
Hi, I'm using a RPi with ML2 and bfgminer 5.4.2-7. I can see /dev/ttyUSB0 but when I start bfgminer it says "NO DEVICES FOUND". I have pressed M+ to add to no avail. Any pointers on where to look? Thanks in advance. Steve
|
|
|
|
mrgoldy
Member

Offline
Activity: 82
Merit: 10
|
 |
December 15, 2017, 12:58:56 PM |
|
Hi, I'm using a RPi with ML2 and bfgminer 5.4.2-7. I can see /dev/ttyUSB0 but when I start bfgminer it says "NO DEVICES FOUND". I have pressed M+ to add to no avail. Any pointers on where to look? Thanks in advance. Steve
open a terminal, change to the dev directory then list you should be able to see what USB number is assigned Sorry, bitcointalk wont let me type it out and post it here
|
|
|
|
dem0nlord
Newbie
Offline
Activity: 41
Merit: 0
|
 |
December 15, 2017, 01:24:20 PM |
|
Hi, I'm using a RPi with ML2 and bfgminer 5.4.2-7. I can see /dev/ttyUSB0 but when I start bfgminer it says "NO DEVICES FOUND". I have pressed M+ to add to no avail. Any pointers on where to look? Thanks in advance. Steve
Version 5.4.2-7, well where did you get that? That's not the version number that has moonlander 2 support.
|
|
|
|
Christmaskkid
Newbie
Offline
Activity: 14
Merit: 0
|
 |
December 15, 2017, 01:28:13 PM |
|
Hey guys..I got a couple questions here..so I am currently running MACOS high Sierra, and apon following the walk around to download the driver, everything works great!(thanks to the team to tackle that obstacle) so my miner was currently running at stock core and memory voltages, 756mhz, 4.04/4.17/4.28h A:1362 R:3 hw 50/.14% (estimated numbers from my last run in a 4:45 hour timeframe) Question 1. Do those numbers look acceptable to you guys? It doesn’t seem bad to me but I want to shoot for the 5-5.5 mh/s so I might jump into tuning slightly today. Question 2. So my miner was running one night for 16 hours straight(unfortunately don’t have numbers from that) then it randomly stopped hashing and basically went dormant..*Is this normal? I want it to get to a point where I can set it, and forget it, maybe check on it a couple times a day but not babysit and worry about it stopping. Question 3. Is there an update for the driver so when I go to install the new one(as of 12/14/17) I have to do the walk around again just to update? Question 4. So every time I open “run moonlander 2” terminal(for Mac) I constantly have three pools showing but only one of them is the one I’m actually using, and every time I go and delete the other unneeded ones they keep constantly showing back up even after I close/save session and start the program back up again. Any reason for this?? Thanks everyone!! 
|
|
|
|
dem0nlord
Newbie
Offline
Activity: 41
Merit: 0
|
 |
December 15, 2017, 01:34:20 PM |
|
Hey guys..I got a couple questions here..so I am currently running MACOS high Sierra, and apon following the walk around to download the driver, everything works great!(thanks to the team to tackle that obstacle) so my miner was currently running at stock core and memory voltages, 756mhz, 4.04/4.17/4.28h A:1362 R:3 hw 50/.14% (estimated numbers from my last run in a 4:45 hour timeframe) Question 1. Do those numbers look acceptable to you guys? It doesn’t seem bad to me but I want to shoot for the 5-5.5 mh/s so I might jump into tuning slightly today. Question 2. So my miner was running one night for 16 hours straight(unfortunately don’t have numbers from that) then it randomly stopped hashing and basically went dormant..*Is this normal? I want it to get to a point where I can set it, and forget it, maybe check on it a couple times a day but not babysit and worry about it stopping. Question 3. Is there an update for the driver so when I go to install the new one(as of 12/14/17) I have to do the walk around again just to update? Question 4. So every time I open “run moonlander 2” terminal(for Mac) I constantly have three pools showing but only one of them is the one I’m actually using, and every time I go and delete the other unneeded ones they keep constantly showing back up even after I close/save session and start the program back up again. Any reason for this?? Thanks everyone!!  To get the speed you're wanting you're going to be at 900mhz or more and need external cooling. What driver did you use? There was new binary release yesterday that addressed restarts.
|
|
|
|
Christmaskkid
Newbie
Offline
Activity: 14
Merit: 0
|
 |
December 15, 2017, 02:15:28 PM |
|
Hey guys..I got a couple questions here..so I am currently running MACOS high Sierra, and apon following the walk around to download the driver, everything works great!(thanks to the team to tackle that obstacle) so my miner was currently running at stock core and memory voltages, 756mhz, 4.04/4.17/4.28h A:1362 R:3 hw 50/.14% (estimated numbers from my last run in a 4:45 hour timeframe) Question 1. Do those numbers look acceptable to you guys? It doesn’t seem bad to me but I want to shoot for the 5-5.5 mh/s so I might jump into tuning slightly today. Question 2. So my miner was running one night for 16 hours straight(unfortunately don’t have numbers from that) then it randomly stopped hashing and basically went dormant..*Is this normal? I want it to get to a point where I can set it, and forget it, maybe check on it a couple times a day but not babysit and worry about it stopping. Question 3. Is there an update for the driver so when I go to install the new one(as of 12/14/17) I have to do the walk around again just to update? Question 4. So every time I open “run moonlander 2” terminal(for Mac) I constantly have three pools showing but only one of them is the one I’m actually using, and every time I go and delete the other unneeded ones they keep constantly showing back up even after I close/save session and start the program back up again. Any reason for this?? Thanks everyone!!  To get the speed you're wanting you're going to be at 900mhz or more and need external cooling. What driver did you use? There was new binary release yesterday that addressed restarts. I have a nice 14” fan sending some cool air my way:) and I’m using the original released driver and haven’t had a chance to update to the new one yet..
|
|
|
|
jstefanop (OP)
Legendary
Offline
Activity: 2043
Merit: 1364
|
 |
December 15, 2017, 03:52:10 PM |
|
Hi, I'm using a RPi with ML2 and bfgminer 5.4.2-7. I can see /dev/ttyUSB0 but when I start bfgminer it says "NO DEVICES FOUND". I have pressed M+ to add to no avail. Any pointers on where to look? Thanks in advance. Steve
My current bfgminer version is 5.4.2-38, sounds like you either compiled on my main branch(instead of the futurebit2-driver branch) or luke-jr's upstream repo? Just download my pre-compiled binaries.
|
|
|
|
jstefanop (OP)
Legendary
Offline
Activity: 2043
Merit: 1364
|
 |
December 15, 2017, 03:57:03 PM |
|
Hey guys..I got a couple questions here..so I am currently running MACOS high Sierra, and apon following the walk around to download the driver, everything works great!(thanks to the team to tackle that obstacle) so my miner was currently running at stock core and memory voltages, 756mhz, 4.04/4.17/4.28h A:1362 R:3 hw 50/.14% (estimated numbers from my last run in a 4:45 hour timeframe) Question 1. Do those numbers look acceptable to you guys? It doesn’t seem bad to me but I want to shoot for the 5-5.5 mh/s so I might jump into tuning slightly today. Question 2. So my miner was running one night for 16 hours straight(unfortunately don’t have numbers from that) then it randomly stopped hashing and basically went dormant..*Is this normal? I want it to get to a point where I can set it, and forget it, maybe check on it a couple times a day but not babysit and worry about it stopping. Question 3. Is there an update for the driver so when I go to install the new one(as of 12/14/17) I have to do the walk around again just to update? Question 4. So every time I open “run moonlander 2” terminal(for Mac) I constantly have three pools showing but only one of them is the one I’m actually using, and every time I go and delete the other unneeded ones they keep constantly showing back up even after I close/save session and start the program back up again. Any reason for this?? Thanks everyone!!  1)Numbers look good 2)Download the new bfgminer version that addresses this issue 3)Your confusing the silicon labs UART driver with my bfgminer mining software(that includes the driver for the Moonlander). You don't need to reinstall the UART driver and do the workaround again, only thing you need is to just download the updated binary and run that instead of the old one. 4)Do you have a config.txt file that your running? This picks up settings as well, also make sure your not running off the desktop image when you download the driver, and save the bfgminer folder somewhere on your hard drive, otherwise changes you do to the start-moonlander file wont save.
|
|
|
|
jstefanop (OP)
Legendary
Offline
Activity: 2043
Merit: 1364
|
 |
December 15, 2017, 04:05:06 PM |
|
New driver working well on Win10  working great for me also on rasberry pi with binary it's switching pools and restarting. When I start up, my HW's are .2-.5% If I switch pools, or change settings thru a proxy, my HW's jump to 3-4%, if I switch again, they jump to ~6%, switch again they jumpp to 8-9% it's like a cascading effect So there's an initial wave of HWs when pools or config switches, but then that % is held each time If I SBY, they restart at 0 and stay on the .2-.5% until I change pool/config settings. Are the HW's just visual, or is there an actual problem? In your pic above, you have 50% HW, but it's only showing a few % EDIT: And now that I've let mine run for a long time, I'm getting around 100HW to 1000Accepted, or 10%, but it's only displaying as 1%, similar to the pic above HW errors are not correlated with accepted shares. Its an internal count that has nothing to do with accepted shares, only number that really matters is the %. HW errors on pool switches is a know bug with how bfgminer handles HW errors, not much you can do about it other than using the "Zero all stats" option under Display options until I find a fix.
|
|
|
|
hercthx
Newbie
Offline
Activity: 52
Merit: 0
|
 |
December 15, 2017, 04:29:07 PM |
|
Another good report jstefanop. Had another power outage this morning, same as yesterday. BIG difference with the updated driver, this time on power restore from UPS backup, all my MLDs remained monitoring, and continued operating once internet came back online!!!! Yesterday with the prev driver all MLDs hung!!!! I'll monitor the next hour to see that the MLDs continue to operate. But on initial work restore, all went well! I see my nBTC/h score is lower by @40, but I think that is just due to the MLDs hashing restart. I'm expecting that score to increase over the next few hours, will report back on status after a few hours. Great update!!!!!
|
|
|
|
Christmaskkid
Newbie
Offline
Activity: 14
Merit: 0
|
 |
December 15, 2017, 04:48:11 PM |
|
[/quote] 1)Numbers look good 2)Download the new bfgminer version that addresses this issue 3)Your confusing the silicon labs UART driver with my bfgminer mining software(that includes the driver for the Moonlander). You don't need to reinstall the UART driver and do the workaround again, only thing you need is to just download the updated binary and run that instead of the old one. 4)Do you have a config.txt file that your running? This picks up settings as well, also make sure your not running off the desktop image when you download the driver, and save the bfgminer folder somewhere on your hard drive, otherwise changes you do to the start-moonlander file wont save.
Awesome thank you so much for the very well explained answers, I appreciate it so much, your doing great at everything you do..keep up the awesome job  Response to #4.. I believe I’m running a txt file, the macs use an app called text editor witch allows me to change things like user and MHZ rate, etc. I think I forgot to actually save the start-moonlander file to my computer as I only copied it and moved the copy to my desktop..so I’ll try that tonight along with some tuning stuff and see if it corrects it. Thanks!
|
|
|
|
asiago
Newbie
Offline
Activity: 4
Merit: 0
|
 |
December 15, 2017, 08:06:53 PM |
|
How tough would it be to change the UI to display LTC/hr on the income rather than BTC?
|
|
|
|
xorxfon
Newbie
Offline
Activity: 8
Merit: 0
|
 |
December 15, 2017, 08:14:14 PM |
|
Is this only able to scrypt mine?, or could settings be tweaked to mine other hash algorithms? .... Asking ....for a friend. 
|
|
|
|
|