Yes I just bought s1 boards from him and they will be here Friday.
Any more news on that chaining or the other mods you were onto? |
|
|
|
This may be a bit off topic, but is it possible to run 4 boards instead of 2 from the original S1 control board?
Yep it IS possible, and BITMAIN have said as much. However, the max is exactly that too, aka 4 boards. Wicked, I have 2 units, 1 won't let me log in via www anymore, but it's still hashing if power cycle it. Any links to this mod? I have not seen any mods for this, but if you look at the S1 board circuitry, you will see the unused tx and rx lines for a basic mod. If you are not an electronics person, I'd be more inclined to shoo you on the direction of sorting out the login rather than the mod, but still, someone here may have more info on the mod. |
|
|
|
This may be a bit off topic, but is it possible to run 4 boards instead of 2 from the original S1 control board?
Yep it IS possible, and BITMAIN have said as much. However, the max is exactly that too, aka 4 boards. |
|
|
|
... how would you write a script that would restart cgminer if the hash rate (which you'd get from the cgminer API) fell below a certain threshhold in an hour?
.......
... I saved the script as /sbin/hashRestart and chmod +x it ...
sleep 600 # ten minute ramp up time for cgminer?
mingh=500
lastgh=`cgminer-api lcd|grep -o '\[GHS5s\] => .*$'|cut -d' ' -f3|cut -d'.' -f1`
m=0
while [ $m -lt 3000 ]
do
m=$(( $m + 5 ))
thisgh=`cgminer-api lcd|grep -o '\[GHS5s\] => .*$'|cut -d' ' -f3|cut -d'.' -f1`
if [ $thisgh -lt $lastgh ]
then
lastgh=$thisgh
fi
echo "lastgh $lastgh"
echo "at $m seconds"
sleep 5
done
if [ $lastgh -lt $mingh ]
then
cgminer-api restart
fi
Save the script as you say, then create an entry for it in cron to run every hour echo '0 * * * * /sbin/hashRestart' >> /etc/crontabs/root
Top man! You are good at this, ey? thats done for me and documented for anyone that would like to do the same (the mingh is 500GH/s, so that is good for me as I have mine OC'ed a bit and they run a tad above that, though they tend to fall away). I'll monitor my poolside performace and report back if there are any adverse effects, else no news is good news! Thanks again (I keep saying that!) |
|
|
|
Now, I'm curious, why do you want to reboot every day? And, here's another way, put in cron or loop script to run every few minutes or hour upsecs=`cat /proc/uptime |cut -d'.' -f1`
if [ $upsecs -gt 86400 ]
then
reboot
fi
Since you are good at this sort of thing, how would you write a script that would restart cgminer if the hash rate (which you'd get from the cgminer API) fell below a certain threshhold in an hour? I'm sure this could be cleaner but here's a first go
sleep 600 # ten minute ramp up time for cgminer?
mingh=500
lastgh=`cgminer-api lcd|grep -o '\[GHS5s\] => .*$'|cut -d' ' -f3|cut -d'.' -f1`
m=0
while [ $m -lt 3000 ]
do
m=$(( $m + 5 ))
thisgh=`cgminer-api lcd|grep -o '\[GHS5s\] => .*$'|cut -d' ' -f3|cut -d'.' -f1`
if [ $thisgh -lt $lastgh ]
then
lastgh=$thisgh
fi
echo "lastgh $lastgh"
echo "at $m seconds"
sleep 5
done
if [ $lastgh -lt $mingh ]
then
cgminer-api restart
fi
You ARE good at this sort of thing, aren't you? And yes, thanks for that. Well then onto the next logical questions .... well, I think they are logical .. 1. would this be a drain on system resources to an extent that it may impact on the actual hashperformance? 2. if I saved the script as /sbin/hashRestart and chmod +x it, would it be enough to run it at startup and if not, what do I need to do? thanks again. |
|
|
|
Now, I'm curious, why do you want to reboot every day? And, here's another way, put in cron or loop script to run every few minutes or hour upsecs=`cat /proc/uptime |cut -d'.' -f1`
if [ $upsecs -gt 86400 ]
then
reboot
fi
Since you are good at this sort of thing, how would you write a script that would restart cgminer if the hash rate (which you'd get from the cgminer API) fell below a certain threshhold in an hour? |
|
|
|
Thank you for your help, but I don't want cgminer to restart, I want the whole miner to reboot. Isn't there a linux command for rebooting with a timer like "shutdown -p 300" but then for rebooting? EDIT: I got it, the command is reboot -d [SECONDS], so you could type in the command in system > startup so the command will be executed at startup and will repeat everytime after boot! Edit crontab via SSH like in the thread linked or WebUI via System -> Scheduled Tasks, and to reboot every day you'd add: 0 23 * * * /sbin/shutdown -rebootto reboot at 11 pm every day (you change the hour to suit). |
|
|
|
update - fixed the half hashing board by scratching off the pcb mask and tinning the copper and linking 2 strings of chips with wire. pics soon!
Kudos! I await the photos..... updated. its crude but it works. since i saved this board i might try my chip theory. Can not be crude if it works ...! I wonder what could have triggered the failure that it by-passes seeing the board was hashing OK. |
|
|
|
update - fixed the half hashing board by scratching off the pcb mask and tinning the copper and linking 2 strings of chips with wire. pics soon!
Kudos! I await the photos..... |
|
|
|
they wont hash. i had all 4 running. one i messed up and it only hashes at half speed. the other ones idk what happened but they just time out. i now only have 1 full hashing board.
If they do not hash but simply timeout, then cgminer debug won't have anything for you either. For the ones that don't hash: 1. When they are connected, do they flash the green light and timeout at the golden nonce or do they report an error code? 2. Were you running them without a heatsink, and if so, did you have a fan blowing over them? EDIT: Also, what setting(s) were you using? 3. Plug all of them in and check Device Manager (windows) to see if they all appear as whatever driver you chose or HID devices. If they appear as Unknown, then it may be a driver issue (long shot, but lets cover our bases). when i came home from work one was hashing at ~10ghs. i unplugged the usb converter then power cycled it and plugged it back in and no hashing. 1. no flashing, just golden nonce, no error 2. the chips have small heatsinks on the front and were placed on something for heat transfer through the back. i also had placed gridseed pods on the heatsinks on the chips with fans running 3. they come up as cp2102 usb to uart converter. side note when i plug in a uart converter with out being connected to a blade i get the golden nonce. both non working boards have .8v across the board and the first chip has 3.3v, so maybe the crystal? I can't comment on elctronics as it would just be speculation, but it seems like you have an elctronics issue. Yes, if you plug in the CP2102 into a PC without a board, cgminer will start the board initialisation process (setting the speed and testing the golden nonce) as it only checks for the USB not the board. Further, that there are no flashing lights at all is a concern, but it also looks like you have the correct drivers, which leads me to believe that it must be an elctronics issue, meaning even cgminer's debug log will not be of any help. PS. Do you have any led light on the CP2102 showing constantly? |
|
|
|
they wont hash. i had all 4 running. one i messed up and it only hashes at half speed. the other ones idk what happened but they just time out. i now only have 1 full hashing board.
If they do not hash but simply timeout, then cgminer debug won't have anything for you either. For the ones that don't hash: 1. When they are connected, do they flash the green light and timeout at the golden nonce or do they report an error code? 2. Were you running them without a heatsink, and if so, did you have a fan blowing over them? EDIT: Also, what setting(s) were you using? 3. Plug all of them in and check Device Manager (windows) to see if they all appear as whatever driver you chose or HID devices. If they appear as Unknown, then it may be a driver issue (long shot, but lets cover our bases). |
|
|
|
no config file, i try to use my other one and i get an internal error. i get nothing when i run curl.exe. if i load cgminer.exe i can only input my pool/user/pass. i can load cgminer with a .cmd with my info in it but it dont hash well with the bmsc options, if i take the options out it will load but not find my blade. i give up, i will use my cgminer 3.8 for the u1 and discard the other blades and hash on the 1. Start cgminer with just one blade, then add the others after. Also, what command line options are you using? Some of the options on this thread are for underclocking / undervolting .... I run mine with the following: cgminer.exe --bmsc-options 115200:1 -o stratum+tcp://stratum.bitcoin.cz:3333 -u pekatete -p x --bmsc-freq 4F81 --lowmemwith 3.8.5 but you should be able to set a timeout of 0.53 with that one. i am trying to figure out why 2 boards are not hashing. he wanted me to use the -d cgminerlog.txt command to see if its seeing the chips and what they are returning. it wont run with that command. so he suggested a diff cgminer where somewhere in this thread i have seen it report all the chips right in the cmd. nothing i have will return that info or will allow the commands or my configs. I know version 3.8.5 will fail with a windows error if 2 boards are plugged in while starting it without the full parameters, so I suggested to you to start cgminer with just 1 board plugged in, then plug in the other one after it has detected the first one. If that fails (and I am assuming you are using windows), then the error it shows in cgminer will be descripive enough to find out what is wrong without needing to log debug data. Just my 2-pence..... |
|
|
|
Anyone else noticed the jump in pool Hashrate ?
What's the concensus --- is this poolhoppers or new hashpower ?
That has been steady to ~10Ph/s for the last few hrs, so I'm certain to a reasonable degree they are here to stay and not hoppers! Having said that, I can not see the supposed luck they could have brought with them as we've been at this block since forever! |
|
|
|
no config file, i try to use my other one and i get an internal error. i get nothing when i run curl.exe. if i load cgminer.exe i can only input my pool/user/pass. i can load cgminer with a .cmd with my info in it but it dont hash well with the bmsc options, if i take the options out it will load but not find my blade. i give up, i will use my cgminer 3.8 for the u1 and discard the other blades and hash on the 1. Start cgminer with just one blade, then add the others after. Also, what command line options are you using? Some of the options on this thread are for underclocking / undervolting .... I run mine with the following: cgminer.exe --bmsc-options 115200:1 -o stratum+tcp://stratum.bitcoin.cz:3333 -u pekatete -p x --bmsc-freq 4F81 --lowmemwith 3.8.5 but you should be able to set a timeout of 0.53 with that one. |
|
|
|
Have you changed the IP address to DHCP?
How do you do that? I don't know much about networking. Not sure what I've done now. I clicked on DHCP and can't get in the menu at all. Once you change to DHCP the unit will get a new IP address, so you need to check your network for the new IP address and log onto that via the UI. |
|
|
|
EDIT: By-the-way, I made a mistake in my single chip anology. Even when the single chip returns early, i.e finds a result early in its range, do NOT issue new work but wait for it to finish searching its assigned range UNLESS the result that was found is the holy grail, i.e its size is 40.3G as we speak (then of-course you've solved the block!). And the wait period is ....... the timeout.
that is exactly what the wait period is about, when to issue new "orders". Ahem ...... when to issue new work, yes. orders / work same same, please remember my english is non native So did you manage to figure out how the work division is done in cgminer (for bmsc), or did you not bother to look? work devision is setup as 1 per chip/fpga (per default) but can be set to 1,2,4,8 it basically means should i send the whole work to 1 chip, or should i split it up for x number of chips (1,2,4,8) edit: correction - fpga based assumes split work in 2 - but that doesnt matter in this case Strange that ...! 1,2,4 or 8 only? The S1 board (last time I checked) has 32 chips. I am not inclined to recount them again now, but I better revisit that cgminer code .... |
|
|
|
EDIT: By-the-way, I made a mistake in my single chip anology. Even when the single chip returns early, i.e finds a result early in its range, do NOT issue new work but wait for it to finish searching its assigned range UNLESS the result that was found is the holy grail, i.e its size is 40.3G as we speak (then of-course you've solved the block!). And the wait period is ....... the timeout.
that is exactly what the wait period is about, when to issue new "orders". Ahem ...... when to issue new work, yes. orders / work same same, please remember my english is non native So did you manage to figure out how the work division is done in cgminer (for bmsc), or did you not bother to look? |
|
|
|
EDIT: By-the-way, I made a mistake in my single chip anology. Even when the single chip returns early, i.e finds a result early in its range, do NOT issue new work but wait for it to finish searching its assigned range UNLESS the result that was found is the holy grail, i.e its size is 40.3G as we speak (then of-course you've solved the block!). And the wait period is ....... the timeout.
that is exactly what the wait period is about, when to issue new "orders". Ahem ...... when to issue new work, yes. |
|
|
|
well, if your correct, then im looking forward to see what you can put together
I know I am correct, well until I am corrected! Put together something? Who said anything about that ...? Only finishing off my very own windows bitcoin miner, and as miners go not exactly re-inventing the wheel (and it won't have chrome spokes either!). But yeah, when I feel there is anything to share, you'll be in the loop! EDIT: By-the-way, I made a mistake in my single chip anology. Even when the single chip returns early, i.e finds a result early in its range, do NOT issue new work but wait for it to finish searching its assigned range UNLESS the result that was found is the holy grail, i.e its size is 40.3G as we speak (then of-course you've solved the block!). And the wait period is ....... the timeout. |
|
|
|
|
Show Posts
