abeandund
Member
Offline
Activity: 124
Merit: 15
|
|
November 08, 2015, 05:26:39 PM |
|
nice use of scrap wood for shelves, very solid looking unit. Why not use a wire closet shelf within the frame instead of lining with boards. You can pick those up on the cheap at any box store or even home improvement. They come in different lengths, but can buy the 6 ft'ers and cut them to length. Advantage is airflow through the shelves, not to mention all the options for running cords and attaching fans.
|
|
|
|
zOU
|
|
November 08, 2015, 05:50:50 PM |
|
Thank you. I thought about wire shelves but if it provides adequate cooling there's no way to control airflow. I'll probably modify/rebuilt my U3 stand when i'll get the 3 additionnal units I've ordered. (we're very much off topic :p)
|
|
|
|
AJRGale
|
|
November 09, 2015, 02:03:49 AM |
|
Ha! I misinterpreted the use of the pot! the more it faces the USB plug, the more voltage to the ASIC, and here I have 2 of the sticks that just cannot hash it out anymore, I was tunning it to less voltage !! Right, I know what I'm doing when I get home. Oddly enough, 2 other ones I had issues with, just spitting more HW errors, I turned down the voltage (in light that I was thinking I was turning up the voltage), and 0 HW errors So, I have 2 oddities, they both start hashing at full speed, then 30 seconds later, one just drops all hashing, the other in 60 odd seconds, dwindles down to 10s of mega-hashes, I've put them to factory set power, left out clock parameters, and they still do the same thing. testing the voltage, Vcore set to 0.7v, 1.8V is 1.8, 3.3 is 3.3 etc.. yeah, I've up the Vcore voltage, still no play. Also tested both on CG and BFG, they both do the same thing.
|
|
|
|
Jake36
|
|
November 09, 2015, 02:40:55 AM |
|
Ha! I misinterpreted the use of the pot! the more it faces the USB plug, the more voltage to the ASIC, and here I have 2 of the sticks that just cannot hash it out anymore, I was tunning it to less voltage !! Right, I know what I'm doing when I get home. Oddly enough, 2 other ones I had issues with, just spitting more HW errors, I turned down the voltage (in light that I was thinking I was turning up the voltage), and 0 HW errors So, I have 2 oddities, they both start hashing at full speed, then 30 seconds later, one just drops all hashing, the other in 60 odd seconds, dwindles down to 10s of mega-hashes, I've put them to factory set power, left out clock parameters, and they still do the same thing. testing the voltage, Vcore set to 0.7v, 1.8V is 1.8, 3.3 is 3.3 etc.. yeah, I've up the Vcore voltage, still no play. Also tested both on CG and BFG, they both do the same thing. Wonder if the chip or something in the hub is going, did you try them on a different hub?
|
|
|
|
AJRGale
|
|
November 09, 2015, 03:16:09 AM |
|
Ha! I misinterpreted the use of the pot! the more it faces the USB plug, the more voltage to the ASIC, and here I have 2 of the sticks that just cannot hash it out anymore, I was tunning it to less voltage !! Right, I know what I'm doing when I get home. Oddly enough, 2 other ones I had issues with, just spitting more HW errors, I turned down the voltage (in light that I was thinking I was turning up the voltage), and 0 HW errors So, I have 2 oddities, they both start hashing at full speed, then 30 seconds later, one just drops all hashing, the other in 60 odd seconds, dwindles down to 10s of mega-hashes, I've put them to factory set power, left out clock parameters, and they still do the same thing. testing the voltage, Vcore set to 0.7v, 1.8V is 1.8, 3.3 is 3.3 etc.. yeah, I've up the Vcore voltage, still no play. Also tested both on CG and BFG, they both do the same thing. Wonder if the chip or something in the hub is going, did you try them on a different hub? yep, and in different systems too
|
|
|
|
stef942
|
|
November 09, 2015, 09:55:41 AM |
|
i have similar problem, my stick comes from bitshopper and is working fine at stock frequency and is stable at 12.47 ghs since 2 days, but as soon as i touch the frequency, by any method -- set compac:clock=200 or --set cbm:clock=200 or directly in bfgminer (M + clock speed), hashrate decrease and then stick stop hashing
|
|
|
|
|
zOU
|
|
November 09, 2015, 02:54:33 PM |
|
so I received my 4 additionnal gekkosticks today from asicpuppy. After a few hours of fiddling with them, here's the final setup, 6 compac sticks running @150Mhs/0.6V (lot's of HW errors at 0.55V and reset) on a DUB-H7 hub. I had a lot of issues running at 0.75 and 0.80 @210Mhs (my settings with 2 sticks), so I went back to stock frequency and raised the V to reach stability. (I know I have to clean the floor and put proper shoes)
|
|
|
|
chiguireitor
Legendary
Offline
Activity: 872
Merit: 1010
Coins, Games & Miners
|
|
November 09, 2015, 04:35:14 PM |
|
... (I know I have to clean the floor and put proper shoes)
Proper shoes? you mean, better no shoes?
|
|
|
|
hedgy73
Legendary
Offline
Activity: 1414
Merit: 1077
|
|
November 09, 2015, 04:52:53 PM |
|
... (I know I have to clean the floor and put proper shoes)
Proper shoes? you mean, better no shoes? My wife's got slippers like that .
|
|
|
|
zOU
|
|
November 09, 2015, 04:53:46 PM Last edit: November 09, 2015, 05:07:00 PM by zOU |
|
God these things are fiddly !!
I just spent 2h finely tuning Viltage on one of these to get it stable at 150Mhs.
between the 6, V goes from 0.59 to 0.63 to get them to be stable at stock frequency !
(and 2 of them have quite a lot of HW errors: 27% and 81%. Ah well.... that's a job for tomorrow :p
|
|
|
|
abeandund
Member
Offline
Activity: 124
Merit: 15
|
|
November 09, 2015, 05:04:04 PM |
|
Very Nice back @zOU. Glad to notice you pulled the compacts over the edge rather than resting the heat sinks directly on the shelf, and upon second look I saw the fan in the picture. You gave me some new ideas.
|
|
|
|
zOU
|
|
November 09, 2015, 05:08:38 PM |
|
Very Nice back @zOU. Glad to notice you pulled the compacts over the edge rather than resting the heat sinks directly on the shelf, and upon second look I saw the fan in the picture. You gave me some new ideas.
I've change it since I took the picture, as I wasn't satisfied, the hub is now vertical so airflow on the heatsink is a lot better (in fact, I tried a dozen possibilities before finding one that satisfied me )
|
|
|
|
Mikestang
Legendary
Offline
Activity: 1274
Merit: 1000
|
|
November 09, 2015, 06:39:19 PM |
|
God these things are fiddly !!
I just spent 2h finely tuning Viltage on one of these to get it stable at 150Mhs.
between the 6, V goes from 0.59 to 0.63 to get them to be stable at stock frequency !
(and 2 of them have quite a lot of HW errors: 27% and 81%. Ah well.... that's a job for tomorrow :p
I don't think that fine tuning is required. If you see HW errors just turn the knob up until the errors stop. Could it be that you are asking too much of that usb hub?
|
|
|
|
zOU
|
|
November 09, 2015, 06:45:56 PM |
|
The thing is: It's always the 2 same sticks have HW errors whatever the slot I put them in. To adjust the knob I have t take them out and use a spare hub to check V. I do not want to overload the hub, I've tried running all 6 at 0.75 but that was a disaster. So I set them all at 0.6 and I'm working my way up by 0.1 or 0.05 increments. One of the 2 culprits works better at 0.59 than 0.64 V. I found out when I tried lowering V rather than overloading the hub After all, they're all supposed to work fine at 150Mhs/0.55V.... But if I set them like this, cgminer (gekko build) reports them as zombie...
|
|
|
|
RichBC
|
|
November 09, 2015, 06:52:42 PM |
|
After all, they're all supposed to work fine at 150Mhs/0.55V.... But if I set them like this, cgminer (gekko build) reports them as zombie...
0.55V at 150MHz is very impressive, best I have reliably done with an R1 is 0.61V. Rich
|
|
|
|
zOU
|
|
November 09, 2015, 07:35:41 PM |
|
OK, plan for tomorrow. Run cgminer Plug 1 compac Adjust until stable Add 1 compac Adjust until stable Repeat until all 6 are stable
|
|
|
|
buckrogers
Legendary
Offline
Activity: 2736
Merit: 1181
|
|
November 10, 2015, 06:02:55 AM |
|
Hi, I'm running BFGMiner version 5.4.0 64bit on win 7 64bit ultimate. I am trying to run the GekkoScience Compac UBS's in a powered USB hub. I have 5 of the USB miners. I'm using an Anker 10 port Powered Hub: http://www.amazon.com/Upgraded-Anker-SuperSpeed-Including-Charging/dp/B005NGQWL2 mor from the manufacturer: http://www.ianker.com/product/68ANHUB-B10AI keep getting this error: CBM 0: Comms error (werr=1) CBM 1: Comms error (werr=1) I had 1 USB miner in the hub and got those errors then it started to go away. The first one was hashing at stock about 8 GH/s and the second the same I ran a .bat file as follows: bfgminer -o stratum+tcp://stratum.antpool.com:3333 -u xxx.x -p x --set compac:clock=0x0b83 I also tried: bfgminer -o stratum+tcp://stratum.antpool.com:3333 -u xxx.x -p x --set compac:clock=150 I am not sure why im getting this error. I did also try the cgminer version and ran zadig, but that didn't work for me, so i removed the zadig driver. I just removed the second compac and im still getting the errors. [2015-11-10 00:44:50] CBM 0: Comms error (werr=1) [2015-11-10 00:45:07] CBM 0: Comms error (werr=1) [2015-11-10 00:45:15] CBM 0: Comms error (werr=1) [2015-11-10 00:45:18] CBM 0: Comms error (werr=1) [2015-11-10 00:45:21] CBM 0: Comms error (werr=1) [2015-11-10 00:45:28] CBM 0: Comms error (werr=1) [2015-11-10 00:45:33] CBM 0: Comms error (werr=1) [2015-11-10 00:45:34] CBM 0: Comms error (werr=1) [2015-11-10 00:49:01] Accepted 0053084d CBM 0 Diff 789/512 [2015-11-10 00:49:26] CBM 0: Comms error (werr=1) [2015-11-10 00:49:55] Accepted 000b5ac7 CBM 0 Diff 5.77k/512 [2015-11-10 00:50:09] CBM 0: Comms error (werr=1) it seems as if it accepts once in a while. Does anyone have a fix for this? thanks!
|
Well I'm dr. spock I'm here to rock y'all
|
|
|
zOU
|
|
November 10, 2015, 06:40:07 AM |
|
Comms error (werr=1): seems to be a write error.
Have you tried the special cgminer version in the first post ?
What is the V setting on your sticks ?
|
|
|
|
Luke-Jr
Legendary
Offline
Activity: 2576
Merit: 1186
|
|
November 10, 2015, 07:16:21 AM |
|
Hi, I'm running BFGMiner version 5.4.0 64bit on win 7 64bit ultimate. I am trying to run the GekkoScience Compac UBS's in a powered USB hub. I have 5 of the USB miners. I'm using an Anker 10 port Powered Hub: http://www.amazon.com/Upgraded-Anker-SuperSpeed-Including-Charging/dp/B005NGQWL2 mor from the manufacturer: http://www.ianker.com/product/68ANHUB-B10AI keep getting this error: CBM 0: Comms error (werr=1) CBM 1: Comms error (werr=1) I had 1 USB miner in the hub and got those errors then it started to go away. The first one was hashing at stock about 8 GH/s and the second the same I ran a .bat file as follows: bfgminer -o stratum+tcp://stratum.antpool.com:3333 -u xxx.x -p x --set compac:clock=0x0b83 I also tried: bfgminer -o stratum+tcp://stratum.antpool.com:3333 -u xxx.x -p x --set compac:clock=150 I am not sure why im getting this error. I did also try the cgminer version and ran zadig, but that didn't work for me, so i removed the zadig driver. I just removed the second compac and im still getting the errors. [2015-11-10 00:44:50] CBM 0: Comms error (werr=1) [2015-11-10 00:45:07] CBM 0: Comms error (werr=1) [2015-11-10 00:45:15] CBM 0: Comms error (werr=1) [2015-11-10 00:45:18] CBM 0: Comms error (werr=1) [2015-11-10 00:45:21] CBM 0: Comms error (werr=1) [2015-11-10 00:45:28] CBM 0: Comms error (werr=1) [2015-11-10 00:45:33] CBM 0: Comms error (werr=1) [2015-11-10 00:45:34] CBM 0: Comms error (werr=1) [2015-11-10 00:49:01] Accepted 0053084d CBM 0 Diff 789/512 [2015-11-10 00:49:26] CBM 0: Comms error (werr=1) [2015-11-10 00:49:55] Accepted 000b5ac7 CBM 0 Diff 5.77k/512 [2015-11-10 00:50:09] CBM 0: Comms error (werr=1) it seems as if it accepts once in a while. Does anyone have a fix for this? thanks! Can you tell me what that werr code is with this custom build? http://luke.dashjr.org/tmp/code/webisect/webisect.php?dobuild=tmp3933
|
|
|
|
|