Destiffmast
|
|
January 17, 2014, 12:05:41 PM |
|
What exactly are dupe shares?
I seem to have quite a few of them - 21.61 GH/s 1619976 (99.82%) 2906 / 10924 / 0
|
|
|
|
|
|
|
|
The trust scores you see are subjective; they will change depending on who you have in your trust list.
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 17, 2014, 12:17:28 PM Last edit: January 17, 2014, 12:57:07 PM by loshia |
|
What exactly are dupe shares?
I seem to have quite a few of them - 21.61 GH/s 1619976 (99.82%) 2906 / 10924 / 0
Same shares submitted to the pool twice - you should see them like rejects R you can calculate your actual hash rate as follows: 21.61 GH/s * (1-R/A) = PAID hash rate Have in mind that Rejects are caused because of stales also. for instance i mine at bitminter and my rejects are below 0.5%. It is normal to have rejects this is unavoidable To make accurate calc run your miner for a day to avoid variance and be 100% sure that there were no pool/network issues
|
|
|
|
Leahcim
Newbie
Offline
Activity: 36
Merit: 0
|
|
January 17, 2014, 12:48:37 PM |
|
Hi,
i downloaded cgminer 3.8.5 and patched it with the patch from technobit. Then i did a configure with ./configure --enable-bflsc --enable-avalon --enable-hexminerc
When its done it shows me: ------------------------------------------------------------------------ cgminer 3.8.5 ------------------------------------------------------------------------
Configuration Options Summary:
libcurl(GBT+getwork).: Enabled: -lcurl curses.TUI...........: FOUND: -lncurses
Avalon.ASICs.........: Enabled BlackArrow.ASICs.....: Disabled BFL.ASICs............: Enabled BitForce.FPGAs.......: Disabled BitFury.ASICs........: Disabled Hashfast.ASICs.......: Disabled Icarus.ASICs/FPGAs...: Disabled Klondike.ASICs.......: Disabled KnC.ASICs............: Disabled ModMiner.FPGAs.......: Disabled
Compilation............: make (or gmake) CPPFLAGS.............: CFLAGS...............: -g -O2 LDFLAGS..............: -lpthread LDADD................: -ldl -lcurl compat/jansson-2.5/src/.libs/libjansson.a -lpthread -lm compat/libusb-1.0/libusb/.libs/libusb-1.0.a -ludev -lrt
Installation...........: make install (as root if needed, with 'su' or 'sudo') prefix...............: /usr/local
I don't see a hexmierc option and the miner, after i did a "make install" does not know the hexminerc options ->
cgminer --hexminerc-voltage 900 --hexminerc-options 16:540 [2014-01-16 16:12:09] cgminer: --hexminerc-voltage: unrecognized option
can anyone help with this?
Thanks Michael
Now i managed to compile it with the exact version pulled from the git repository and autogen but is it really true, that the hexminer patch breaks the butterfly support? So i have to compile 2 versions and let them run at the same time? Bye Michael
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 17, 2014, 12:56:41 PM Last edit: January 17, 2014, 01:17:52 PM by loshia |
|
Hi,
i downloaded cgminer 3.8.5 and patched it with the patch from technobit. Then i did a configure with ./configure --enable-bflsc --enable-avalon --enable-hexminerc
When its done it shows me: ------------------------------------------------------------------------ cgminer 3.8.5 ------------------------------------------------------------------------
Configuration Options Summary:
libcurl(GBT+getwork).: Enabled: -lcurl curses.TUI...........: FOUND: -lncurses
Avalon.ASICs.........: Enabled BlackArrow.ASICs.....: Disabled BFL.ASICs............: Enabled BitForce.FPGAs.......: Disabled BitFury.ASICs........: Disabled Hashfast.ASICs.......: Disabled Icarus.ASICs/FPGAs...: Disabled Klondike.ASICs.......: Disabled KnC.ASICs............: Disabled ModMiner.FPGAs.......: Disabled
Compilation............: make (or gmake) CPPFLAGS.............: CFLAGS...............: -g -O2 LDFLAGS..............: -lpthread LDADD................: -ldl -lcurl compat/jansson-2.5/src/.libs/libjansson.a -lpthread -lm compat/libusb-1.0/libusb/.libs/libusb-1.0.a -ludev -lrt
Installation...........: make install (as root if needed, with 'su' or 'sudo') prefix...............: /usr/local
I don't see a hexmierc option and the miner, after i did a "make install" does not know the hexminerc options ->
cgminer --hexminerc-voltage 900 --hexminerc-options 16:540 [2014-01-16 16:12:09] cgminer: --hexminerc-voltage: unrecognized option
can anyone help with this?
Thanks Michael
Now i managed to compile it with the exact version pulled from the git repository and autogen but is it really true, that the hexminer patch breaks the butterfly support? So i have to compile 2 versions and let them run at the same time? Bye Michael Yes it does for sure written in Changelog If you have nanos use --enable-hexmineru (make sure stock bitfury is disabled though) works great otherwise compile and run two instances with appropriate devs enabled only Or better for other devs just compile stock version
|
|
|
|
Destiffmast
|
|
January 17, 2014, 01:46:21 PM |
|
What exactly are dupe shares?
I seem to have quite a few of them - 21.61 GH/s 1619976 (99.82%) 2906 / 10924 / 0
Same shares submitted to the pool twice - you should see them like rejects R you can calculate your actual hash rate as follows: 21.61 GH/s * (1-R/A) = PAID hash rate Have in mind that Rejects are caused because of stales also. for instance i mine at bitminter and my rejects are below 0.5%. It is normal to have rejects this is unavoidable To make accurate calc run your miner for a day to avoid variance and be 100% sure that there were no pool/network issues Thanks for the calculation. Miner been running for 3 days, so pretty accurate. I don't believe I have power issues either, so maybe a problem with the pool rather than the miner itself.
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 17, 2014, 01:49:23 PM Last edit: January 17, 2014, 02:09:21 PM by loshia |
|
What exactly are dupe shares?
I seem to have quite a few of them - 21.61 GH/s 1619976 (99.82%) 2906 / 10924 / 0
Same shares submitted to the pool twice - you should see them like rejects R you can calculate your actual hash rate as follows: 21.61 GH/s * (1-R/A) = PAID hash rate Have in mind that Rejects are caused because of stales also. for instance i mine at bitminter and my rejects are below 0.5%. It is normal to have rejects this is unavoidable To make accurate calc run your miner for a day to avoid variance and be 100% sure that there were no pool/network issues Thanks for the calculation. Miner been running for 3 days, so pretty accurate. I don't believe I have power issues either, so maybe a problem with the pool rather than the miner itself. The problem with dupes is that they afect your hash rate and you can get false feeling that it works better when do not take in account rejects. Anyway watch technobit site for update fixing this it should be released soon Once again rejects will be always there even dupe issue is fixed that is the way it works with other boards/vendors also. There is no workaround for this And finally precise formula as suggested by kano Paid shares cgminer-api devs [ASC1] => ( [ASC] => 1 [Name] => HEXc [ID] => 0 [Enabled] => Y [Status] => Alive [Temperature] => 0.00 [MHS av] => 22030.36 [MHS 5s] => 19677.28 [Accepted] => 179 [Rejected] => 7 [Hardware Errors] => 1245 [Utility] => 0.57 [Last Share Pool] => 0 [Last Share Time] => 1389967128 [Total MH] => 414125041.5510 [Diff1 Work] => 96422 [Difficulty Accepted] => 104448.00000000 [Difficulty Rejected] => 524.00000000 [Last Share Difficulty] => 512.00000000 [No Device] => false [Last Valid Work] => 1389967232 [Device Hardware%] => 1.2747 [Device Rejected%] => 0.5434 [Device Elapsed] => 18798 To be more precise as ugested by kano 2^32/[Device Elapsed]*[Difficulty Accepted]/10^9 My score is lucky though i rebooted cgminer but it shal give you almost same result as the 21.61 GH/s * (1-R/A)
|
|
|
|
Gator-hex
|
|
January 17, 2014, 02:20:06 PM |
|
If your PSU has multiple rails one of the rails will be on the PCI-E 6/8 pin connector.
What I do is chop a molex off on an overloaded rails and wire it up to the PCI-E.
|
|
|
|
infinitecoin1993
Member
Offline
Activity: 107
Merit: 10
|
|
January 17, 2014, 11:06:51 PM |
|
you should connect the 24pin atx connector and put in a paperclip or a little piece of power cable in the green and black wire like in the picture , i`don`t know if your fan is going to spin i think it will i have corsair cx 750m and when i put in a paperclip in my green and black wire the fan spins without even connecting any device to the psu.. so do this: disconnect the psu from the wall socket connect 24 pin atx put paperclip in green and black wire put some electrical tape on top of the paperclip (24 pin connector) make shure every thing is connected to the HEXc power on psu happy hashing This fixed my problem after hourrrssss of trying!!!!!!!!!!!
|
|
|
|
gusto88
Newbie
Offline
Activity: 21
Merit: 0
|
|
January 17, 2014, 11:13:42 PM |
|
you should connect the 24pin atx connector and put in a paperclip or a little piece of power cable in the green and black wire like in the picture , i`don`t know if your fan is going to spin i think it will i have corsair cx 750m and when i put in a paperclip in my green and black wire the fan spins without even connecting any device to the psu.. http://s8.postimg.org/z174nra3l/download.jpgso do this: disconnect the psu from the wall socket connect 24 pin atx put paperclip in green and black wire put some electrical tape on top of the paperclip (24 pin connector) make shure every thing is connected to the HEXc power on psu happy hashing This fixed my problem after hourrrssss of trying!!!!!!!!!!! You can always buy one of those - often used with water cooling. Also look more professional.
|
|
|
|
Zich
Legendary
Offline
Activity: 1190
Merit: 1000
|
|
January 18, 2014, 04:56:23 AM |
|
.................. Lack of steady 12V to the board explains why it was disappearing BUT there is a bug which submits dup shares which is fixed and under testing. So watch technobit site for updates dup shares were caused because of voltage problems New patch for cgminer 3.10.0 + openwrt image v 0.1.7 Download
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 18, 2014, 07:37:56 AM |
|
.................. Lack of steady 12V to the board explains why it was disappearing BUT there is a bug which submits dup shares which is fixed and under testing. So watch technobit site for updates dup shares were caused because of voltage problems New patch for cgminer 3.10.0 + openwrt image v 0.1.7 Download Yeah no more dupes and back to max nano performance
|
|
|
|
Zich
Legendary
Offline
Activity: 1190
Merit: 1000
|
|
January 19, 2014, 09:43:53 AM |
|
... Most probably because of the USB hub. Is it a tt one? Lsusb -v?
Indeed the problem from usb hub You are really great loshia I just buy another cheap usb hub that use FE2.1 mtt chip AXP807I ask them about the chip before buy it Work great on TPLINK TL-MR3020
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 19, 2014, 11:16:18 AM Last edit: January 19, 2014, 11:35:47 AM by loshia |
|
... Most probably because of the USB hub. Is it a tt one? Lsusb -v?
Indeed the problem from usb hub You are really great loshia I just buy another cheap usb hub that use FE2.1 mtt chip AXP807I ask them about the chip before buy it Work great on TPLINK TL-MR3020 Super, Try latest version no more dupes on hex16c and high CPU usage because of the nanos, but you will be fine just do not pay attention to it I can guarantee 5% performance increase for nanos
|
|
|
|
Zich
Legendary
Offline
Activity: 1190
Merit: 1000
|
|
January 19, 2014, 11:46:18 AM |
|
Super, Try latest version no more dupes on hex16c and high CPU usage because of the nanos, but you will be fine just do not pay attention to it I can guarantee 5% performance increase for nanos Wow 5% is great, going to try it tomorrow
|
|
|
|
iFA88
|
|
January 19, 2014, 01:40:41 PM |
|
|
|
|
|
gusto88
Newbie
Offline
Activity: 21
Merit: 0
|
|
January 19, 2014, 02:18:50 PM |
|
You have six miners, but cgminer only shows five. Something is wrong. My guess is, that a raspberry pi can only handle up to three hex16a2 or ~1200 WU/m I have 4 hex16a2 on one raspberry pi and similiar GH/s.
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 19, 2014, 02:31:10 PM |
|
Cgminer-API stats Cgminer-API debs No caps pls ? Lsusb -v Dmesg also
|
|
|
|
iFA88
|
|
January 19, 2014, 02:32:13 PM |
|
Oh sorry, on the image is all connected, but today i have one unplugged. There is only 5 board connected to the RPI. When i plug the sixth works too. RPI can handle much more boards.
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
January 19, 2014, 03:08:03 PM |
|
Oh sorry, on the image is all connected, but today i have one unplugged. There is only 5 board connected to the RPI. When i plug the sixth works too. RPI can handle much more boards.
Chimner API output?
|
|
|
|
|
|