 Show Posts
|
|
Pages: [1] 2 3 »
|
|
buying 0.32485 @ 307.83
EDIT: thanks, I corrected my BTC amount (see next post)
|
|
|
|
i've made a small web interface for it. screenshot: https://www.dropbox.com/s/od0ac7djvvdyrps/gekkoweb.pngsource: https://www.dropbox.com/s/54y2crazfnyghnd/gekko.zipi am not familiar with npm or node. so you have to install a few dependencies by hand (eg npm install express). there are currently some drawbacks: - portfolio is only updated after a successfull buy or sell - web interface does work as soon as a the first advice is made. - only ema does work as the web if does show some internal data from it i use it in "advanced mode" using data from mtgox and trading on mtgox and btce (but only with small amounts) i did not test it with other configurations, so use it at your own risk (i am not responsiple for any losses etc). as this is a pure personal project i dont think i'll make other extensions, but i am always open for suggestions regards flower sorry for double post, but want to bump this thread with a question - does anyone have this package that is mentioned above? the dropbox URLs are dead. |
|
|
|
WHAT IS that method in the DEMA.js? I do not know what that code even is. It is not using a " double EMA" of any kind, and I can't seem to find any sort of indicator based on that sort of algorithm. This is how it works: ~pseudocode~
Calculate a fast and slow EMA average
A = subtract one from the other
B = add one to the other, and divide it by 2
C = 100 times A divided B
based on the value of C, assume positive values mean an uptrend, and negative values mean a downtrend
I ... don't even know what that is. It is not called "DEMA" though I don't think, I've never seen any TA indicator which does that particular combination of things to two different EMAs... I could have updated the ticket with more info after fixing the PPO code, but was tired last night... PPO has been audited and is now consistent with the new codebase, and I added the x2MACD now as well ( which worked correctly overnight) -- updated: https://github.com/kuzetsa/gekko/issues/3 maybe you still have not found it, but it is in gekko documentation: ### DEMA
This method uses `Exponential Moving Average crossovers` to determine the current trend the
market is in. Using this information it will suggest to ride the trend. Note that this is
not MACD because it just checks whether the longEMA and shortEMA are [threshold]% removed
from eachother.
This method is fairly popular in bitcoin trading due to Bitcointalk user Goomboo. Read more about this method in [his topic](https://bitcointalk.org/index.php?topic=60501.0)
You can configure these parameters for DEMA in config.js:
config.DEMA = {
// EMA weight (α)
// the higher the weight, the more smooth (and delayed) the line
short: 10,
long: 21,
// amount of candles to remember and base initial EMAs on
// the difference between the EMAs (to act as triggers)
thresholds: {
down: -0.025,
up: 0.025
}
};
- short is the short EMA that moves closer to the real market (including noise)
- long is the long EMA that lags behind the market more but is also more resistant to noise.
- the down treshold and the up treshold tell Gekko how big the difference in the lines needs to be for it to be considered a trend. If you set these to 0 each line cross would trigger new advice.
|
|
|
|
|
but I guess there is no source that can be 100% trusted? I guess blockchain.info is the most trustworthy that at least attempts to display block status?
|
|
|
|
5 blocks orphaned today  edit: or is it blockchain.info bug, because all blocks mined today show as orphaned. |
|
|
|
I kinda agree but is there any good reason to oc them?
if you are lucky and have a stable unit, then you can get better GH/WATT relation. check here for detailed info on this: https://bitcointalk.org/index.php?topic=699064.460this is cut from the thread:
#option 'freq_value' '0982' #502Gh/s 402W/ 0.800W per GH
#option 'chip_freq' '250'
#option 'timeout' '16'
#option 'freq_value' '1306' #490Gh/s 385W/ 0.786W per GH
#option 'chip_freq' '243.75'
#option 'timeout' '16'
#option 'freq_value' '0902' #478Gh/s 367W/ 0.768W per GH
#option 'chip_freq' '237.5'
#option 'timeout' '17'
#option 'freq_value' '1206' #465Gh/s 356W/ 0.766W per GH
#option 'chip_freq' '231.25'
#option 'timeout' '17'
#option 'freq_value' '0882' #452Gh/s 346W/ 0.765W per GH
#option 'chip_freq' '225'
#option 'timeout' '18'
option 'freq_value' '1106' #441Gh/s 340W Default/ 0.771W per GH
option 'chip_freq' '218.75'
option 'timeout' '18'
|
|
|
|
I have a Corsair AX1200 high end power supply having min 90% efficiency at 50% load.
I loaded it with an overclocked S3 (s) running at 250Mhz and 500Ghs+.
The power consumption at Wall is around 400W at 220VAC.
I then connected it to a 12V DC Source directly omitting the Power Supply.
Now the power consumption is around 260Watts to 300Watts, the DC-DC converter's efficiency is near 90%.
So, I should assume from my findings that before PSU the S3 consumes <300Watts and at all 430W.
The ammeter used in both case where same and did testing multiple times with same results.
Antminer S3 <- 24VDC-12VDC Converter <- 24VDC Solar Charge Controller <- <300W
Antminer S3 <- Power Supply <- 24VDC Invertered to AC <- 24VDC Solar Charge Controller <- 430W
do I understand correctly: Antminer S3 <- 24VDC-12VDC Converter <- 24VDC Solar Charge Controller <- <300W Antminer S3 <- Power Supply <- 220V power outlet <- 340W ? Maybe then: Antminer S3 <- 24VDC-12VDC Converter <- 220V power outlet <- <300W ? |
|
|
|
|
now it is impossible to tell what happened, but you can make arrangements so that next time if this happens you would know more.
you could modify original monitoring script that runs on your Antminer every 3 minutes:
/usr/bin/cgminer-monitor
cp /usr/bin/cgminer-monitor /usr/bin/cgminer-monitor_BACKUP
add new line before or after every:
/etc/init.d/cgminer restart
the lines you want to add are like this:
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') no process running >> /usr/bin/avg.log
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because no valid work being done >> /usr/bin/avg.log
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because of miner count >> /usr/bin/avg.log
if the issue repeats itself, then you can check by command:
tail /usr/bin/avg.log
and at least see which of the cases had occurred.
if for example - no process running, then your cgminer crashed
if no work being done, then probably your or pools network problem.
you can have additional checks as well, by adding:
ping -c 4 POOLIP >> /usr/bin/avg.log
ping -c 4 RANDOMIP_THAT_IS_ALWAYS_REACHABLE_LIKE_GOOGLE.COM >> /usr/bin/avg.log
next to the echo commands and you will see if this was due to your or pools network connectivity - it will try to ping 4 times.
|
|
|
|
M's Ant Monitor can reboot Ants when they drop below a certain hashrate as well. See my sig.
M
If I would have known about this application couple of days ago then I would probably not have written the script. But as I have written it, then I will keep running it, advantage of the script is that - it will not require a computer to be on 24/7 for the monitoring to work (because the script runs on miner itself) as opposed to the application. Thank you for the comment, I wish I had known this already  |
|
|
|
any news about starting to ship batch 8 ?
I paid 0.58 BTC x2 , for something that will never ROI
At least they should send it faster !
at least you are losing 0.58 unlike me, I ordered batch 6 and payed 1.32 BTC, I only made 0.30 BTC back (not counting electricity) so I am down 1 BTC + electricity. |
|
|
|
S1-1 - 218 - 5hr - 437gh - 1HR Restarts
S1-2 - 250Mhz - 5hr - 466gh - 1/2hr restarts.
This is an increase of over 40GH on both miners for me!
I'm going to clock-up S1-1 and see if I see increases at 1/2hr intervals.
I encourage those experiencing degrading performance over time to do something similar. You lose less than 10 seconds of mining per restart.
I have same problem with one of my antminers, that randomly it will start under performing, like for example when I first turn it on it will hash fine for random amount of hours averaging 440 GH/s, but then will go to 400-410 GH/s AVG for days unless I will reboot it, sometimes needs couple of reboots. you suggested restarting every hour or more often, but I wrote a script that should only restart when it is really required. This is for advanced users only that know how to use vim or other ways of editing text file in Unix shell. And only do this with the unit that has problems with performance, do not do this with good units! first create a file: /usr/bin/cgminer-avg-monitor for example by issuing following command: vim /usr/bin/cgminer-avg-monitor #!/bin/sh
# This file is for cron job
# This if for make sure the start cgminer when there is network problem
nowtime=$(date +%s)
avglog=$(date +%s -r /usr/bin/avg.log)
age=$(( nowtime - avglog ))
if [ "$age" -lt 600 ]; then
exit 0;
else
set -x
date
avg=`/usr/bin/cgminer-api -o summary | sed -n "s/.*GHS av=\([^']*\)...,F.*/\1/p"`
if [ "$avg" -lt 439 ]; then
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because avg hashrate: $avg >> /usr/bin/avg.log
exit 0;
fi
fi
change file permission by: chmod 700 /usr/bin/cgminer-avg-monitor then back up original monitor script: cp /usr/bin/cgminer-monitor /usr/bin/cgminer-monitor_BACKUP and modify (or replace) it so it looks like this: nowtime=$(date +%s)
avglog=$(date +%s -r /usr/bin/avg.log)
age=$(( nowtime - avglog ))
if [ "$age" -lt 600 ]; then
exit 0;
else
set -x
date
C=`pidof cgminer | wc -w`
if [ "$C" != "1" ]; then
echo "Case C"
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') no process running >> /usr/bin/avg.log
exit 0;
fi
A=`cat /tmp/cm.log`
B=`cgminer-api devs | grep "^ \[Last Valid Work\]"`
echo "$B" > /tmp/cm.log
if [ "$B" != "" -a "$A" == "$B" ]; then
echo "Case B"
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because no valid work being done >> /usr/bin/avg.log
exit 0;
fi
D=`cgminer-api stats | grep "^ \[miner_count\]"`
if [ "$D" == "" ]; then
echo "Case D"
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because of miner count >> /usr/bin/avg.log
exit 0;
fi
fi
your crontab should look like this: root@antMiner:/usr/bin# crontab -l
*/3 * * * * /usr/bin/cgminer-monitor
59 19 * * 0 reboot
0,30 * * * * /usr/bin/cgminer-avg-monitor
for safety create the log file, that is mentioned in the scripts by command: echo START >> /usr/bin/avg.log what will happen now: 1) as usual there will be /usr/bin/cgminer-monitor executed every 3 minutes, but will not do anything if something has already been done during last 10 minutes (600 seconds) - like a restart. 2) /usr/bin/cgminer-avg-monitor will run every 30 minutes, its purpose is to check if AVG hashrate of the miner is below 439 GH/s, if it is, then it will restart cgminer (only if it has not been done within last 10 minutes) 3) reboot antminer every sunday at 19:59... just to reboot once in a while. UNINSTALLING: rm /usr/bin/cgminer-monitor
mv /usr/bin/cgminer-monitor_BACKUP /usr/bin/cgminer-monitor
rm /usr/bin/cgminer-avg-monitor
rm /usr/bin/avg.log edit crontab to its original version: */3 * * * * /usr/bin/cgminer-monitor Hope someone finds this useful. EDIT: I did this for all (2) of my units and am forgetting they exist even if internet goes offline, or losing electricity, the units will go back and will make sure they hash at 439+ GH/s even if couple of restarts are required. I found out that even the good unit will start under performing if internet goes down for a while (I know what I am talking about, I am taking into consideration only pool side 3 hour AVG after the internet is back), I know that after internet goes down in miner status it will show less AVG until rebooted and it does not show actual AVG, so I am looking at poolside - I want it to be ~880+-5% GH/s poolside and it was not hapenning before this script, I had to manually restart after downtime. |
|
|
|
The green and the yellow lights are on.
the browser cant connect with the Miner ip.
Should i get a new router?
The current Router is a wifi one Not a network lan router
I am using WIFI router also, but both - laptop and antminer ar connected to router by wire, but I can see you already have that. can you show me your IP for laptop or you do not know how to? Us the tutorial here: http://www.wikihow.com/Find-out-Your-IP-Address(the ipconfig version). |
|
|
|
I have this problem with one of my antminers, that randomly it will start under performing, like for example when I first turn it on it will hash fine for random amount of hours averaging 440 GH/s, but then will go to 400-410 GH/s AVG for days unless I will reboot it, sometimes needs couple of reboots. see example here: https://bitcointalk.org/index.php?topic=671189.4980https://bitcointalk.org/index.php?topic=671189.2820https://bitcointalk.org/index.php?topic=671189.3680user suggested restarting every hour or more often, but I wrote a script that should only restart when it is really required. This is for advanced users only that know how to use vim or other ways of editing text file in Unix shell. And only do this with the unit that has problems with performance, do not do this with good units! first create a file: /usr/bin/cgminer-avg-monitor for example by issuing following command: vim /usr/bin/cgminer-avg-monitor #!/bin/sh
# This file is for cron job
# This if for make sure the start cgminer when there is network problem
nowtime=$(date +%s)
avglog=$(date +%s -r /usr/bin/avg.log)
age=$(( nowtime - avglog ))
if [ "$age" -lt 600 ]; then
exit 0;
else
set -x
date
avg=`/usr/bin/cgminer-api -o summary | sed -n "s/.*GHS av=\([^']*\)...,F.*/\1/p"`
if [ "$avg" -lt 439 ]; then
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because avg hashrate: $avg >> /usr/bin/avg.log
exit 0;
fi
fi
change file permission by: chmod 700 /usr/bin/cgminer-avg-monitor then back up original monitor script: cp /usr/bin/cgminer-monitor /usr/bin/cgminer-monitor_BACKUP and modify (or replace) it so it looks like this: nowtime=$(date +%s)
avglog=$(date +%s -r /usr/bin/avg.log)
age=$(( nowtime - avglog ))
if [ "$age" -lt 600 ]; then
exit 0;
else
set -x
date
C=`pidof cgminer | wc -w`
if [ "$C" != "1" ]; then
echo "Case C"
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') no process running >> /usr/bin/avg.log
exit 0;
fi
A=`cat /tmp/cm.log`
B=`cgminer-api devs | grep "^ \[Last Valid Work\]"`
echo "$B" > /tmp/cm.log
if [ "$B" != "" -a "$A" == "$B" ]; then
echo "Case B"
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because no valid work being done >> /usr/bin/avg.log
exit 0;
fi
D=`cgminer-api stats | grep "^ \[miner_count\]"`
if [ "$D" == "" ]; then
echo "Case D"
killall -s 9 cgminer
sleep 1
/etc/init.d/cgminer restart
echo restarted at $(date '+%Y-%m-%d-%H-%M-%S') because of miner count >> /usr/bin/avg.log
exit 0;
fi
fi
your crontab should look like this: root@antMiner:/usr/bin# crontab -l
*/3 * * * * /usr/bin/cgminer-monitor
59 19 * * 0 reboot
0,30 * * * * /usr/bin/cgminer-avg-monitor
for safety create the log file, that is mentioned in the scripts by command: echo START >> /usr/bin/avg.log what will happen now: 1) as usual there will be /usr/bin/cgminer-monitor executed every 3 minutes, but will not do anything if something has already been done during last 10 minutes (600 seconds) - like a restart. 2) /usr/bin/cgminer-avg-monitor will run every 30 minutes, its purpose is to check if AVG hashrate of the miner is below 439 GH/s, if it is, then it will restart cgminer (only if it has not been done within last 10 minutes) 3) reboot antminer every sunday at 19:59... just to reboot once in a while. UNINSTALLING: rm /usr/bin/cgminer-monitor
mv /usr/bin/cgminer-monitor_BACKUP /usr/bin/cgminer-monitor
rm /usr/bin/cgminer-avg-monitor
rm /usr/bin/avg.log edit crontab to its original version: */3 * * * * /usr/bin/cgminer-monitor Hope someone finds this useful. EDIT: I did this for all (2) of my units and am forgetting they exist even if internet goes offline, or losing electricity, the units will go back and will make sure they hash at 439+ GH/s even if couple of restarts are required. I found out that even the good unit will start under performing if internet goes down for a while (I know what I am talking about, I am taking into consideration only pool side 3 hour AVG after the internet is back), I know that after internet goes down in miner status it will show less AVG until rebooted and it does not show actual AVG, so I am looking at poolside - I want it to be ~880+-5% GH/s poolside and it was not hapenning before this script, I had to manually restart after downtime. |
|
|
|
is your antminer on (lights blinking)? if so, then what is you laptop IP? to find it out, on windows press WINDOWS+R buttons in "Run.." windows that popped up, type cmd and press OK. In CMD promt, type ipconfig /all and find your "IPv4 Address" you should know that by default antminers IP is: 192.168.1.99, so in your web browser open: http://192.168.1.99if it does not open, what does the error message say? |
|
|
|
Well the dust on the heatsinks is probably just left over from manufacturing. Bitmain had to have the sinks made someplace then shipped to them to assemble.
I'm sure each heatsink wasnt cleaned prior to assy.
As for the design, maybe they had different batches made by different suppliers.
As long as the circuit boards are the same and say S3 on them then they are legit.
The heatsinks are not causing your problems.
You said you were going to let them run un-molested for a while.
Hows that going ?
This is my last offtopic post about this here, if anyone is interested in finding out more, lets take it to PMs. My bad unit looked almost like this when I first opened it, I had less dust on it (like 70% of what you can see on following pic): https://bitcointalk.org/index.php?topic=330665.2280The dust only was on air intake side of the unit, the other side was fresh clean. Moreover the intake fan wings were also covered in thick layer of dust, I hope I do not need to go further proving that unit was running for a very long time before I got it. I asked bitmain about different heatsinks on same batch and dirty heatsinks - they asked me the order number and then held back to answer, but answered that there is nothing they can do unless the unit is faulty. The lastest update from me is - good news: 1) on the pool I am on now - they do not penalize hoppers so I can restart the unit whenever I feel it starts to underperform 2) I ran the units without restart for 24 hours hoping to run them until end of the week, performance wise they ran as expected for 4-5 hours, then the bad unit took a plunge for 20 hours non stop underperforming and no sign of recovering (400 GH/s) 3) seeing that last night I decided not to wait until end of the week as I promised and took the good unit apart and by popular demand took pics to show different heatsink design on same batch units and to find that it is almost perfectly clean inside the good unit, just a little dust on it which most probably collected while running at my place, I cleaned that off, tightened heatsink screws, reversed both fans which caused the good unit run a lot cooler, too - fans ran 1900-2100 RPM before the procedure, now they run 1600-1700 RPMs, combined with the bad unit already cleaned and reversed fans now the units run at pleasant noise level compared to what they were running at upon receiving them. as the units generate less heat now, the room they are in is little cooler than it was. After procedure I started both units to find them running 880 AVG for 12 hours, then there was an internet connection problem for 1 minute at my place after which the bad unit started underperforming again, after restart the units now run good for 8 hours. to conclude - the problem is still there, keeping room temperature low seems to fix it. |
|
|
|
Not sure how to take the trolling thing lol. No i wasn't saying you were making it up. Just odd they would do that. i just ordered 4 batch 8's hope they don't show up like that.
JT
by trolling I mean - I sent 2 pms but you claimed you never received them, which seemed strange to me. but then I understood - maybe I am censored and it will take some time for PMs to arrive for you. Anyway, if anyone interested, I have taken pics to compare 2 antminers s3 batch 6 that I received from https://bitmaintech.com/product.htmgood unit: http://postimg.org/gallery/34jlrdgao/8c5bf8e8/only today I took off fans on this unit and I could see only a little dust on the heatsinks, which might be collected only while running at my place. The bad one: http://postimg.org/gallery/2r88uer0q/f6b4681a/As you can see - completely different design, and despite my best effort to clean the heatsink - they are still dirty as hell. |
|
|
|
Not today anyway.
JT
I feel like being trolled by you Here is proof of one of pms sent: http://postimg.org/image/3zmo8h3qx/edit: maybe my PMs are being censored because I am junior member.. |
|
|
|
I find it hard to believe that Bitmain would send you a defective dusty unit. On top of that i find it hard to believe that they would send you an upgraded S1 instead of an S3 like you ordered something is not right there. I would contact them and ask whats up.
JT
you should not have hard time to believe that - it is what it is, I sent you full explanation in pm as this thread is not bitmain related and this conversation is offtopic. |
|
|
|
|
