GekkoScience has a new stickminer that does 300+GH

[n0nce]

I have tried different USB cords/sticks configurations etc....it must be some kind of USB restriction?
Can you please help me with this?

I'm sorry I can't help with this; I can answer your second question.

As a side question what's the minimum USB voltage they need to run?
I know there is a voltage pod, so theoretically I can even run them at 4.3V (at the hub side) if my voltage at the chip would be ~1,45V correct?

The buck converter circuit should give you a stable output, as long as the input is in a relatively normal range. I've seen 5V USB devices running on anywhere between 4.7 to 5.1V normally, but it's possible that 4.3 could suffice. Do note that the potentiometer doesn't affect your input voltage (5V side) but just the core voltage; so it's possible that other components such as the FTDI chip will stop working even though the core is happy. 

The only problem I see is my hub shows 5.2V when NOT hashing and sticks starting up and if I adjust the screw, until I'm at 1.45V DURIUNG mining for sticks stability, it could possibly go way above the 1,45V at the INITIAL stage of mining, when sticks are just starting up and not hashing yet, would that be a problem?

I don't think 5.2V is too much out of the ordinary, but I'd be careful of having voltages on the core of 'way above 1.45V' (how much is 'way above'?).

Another side question - what file is a config file for current version of cgminer? I have used the save option during mining and now it looks like no mater what options I run the miner with it will run with the saved ones, I looked in the cgminer folder and there are tons of various config files. I assume if I delete the config file it will revert to default settings?

It takes the config file if you specify it. I.e.: run it with the -c or --config parameter and pass the config file's path.
The contents should look something like this:

Code:

{
"pools" : [
	{
		"url" : "stratum+tcp://stratum.kano.is:3333",
		"user" : "username.worker1",
		"pass" : "x"
	}
],

"failover-only" : true,
"api-listen" : true,
"api-port" : "4028",
"api-allow" : "W:192.168.1.0/24,W:127.0.0.1"
}

Just input the information that you'd otherwise pass by command line args into that file. Then you can run it simply with ./cgminer -c example.conf.

[PPICASSO]

Hi, thanks for helping me out!

As a side question what's the minimum USB voltage they need to run?
I know there is a voltage pod, so theoretically I can even run them at 4.3V (at the hub side) if my voltage at the chip would be ~1,45V correct?

The buck converter circuit should give you a stable output, as long as the input is in a relatively normal range. I've seen 5V USB devices running on anywhere between 4.7 to 5.1V normally, but it's possible that 4.3 could suffice. Do note that the potentiometer doesn't affect your input voltage (5V side) but just the core voltage; so it's possible that other components such as the FTDI chip will stop working even though the core is happy.  

Ok, got it.

The only problem I see is my hub shows 5.2V when NOT hashing and sticks starting up and if I adjust the screw, until I'm at 1.45V DURIUNG mining for sticks stability, it could possibly go way above the 1,45V at the INITIAL stage of mining, when sticks are just starting up and not hashing yet, would that be a problem?

I don't think 5.2V is too much out of the ordinary, but I'd be careful of having voltages on the core of 'way above 1.45V' (how much is 'way above'?).

Well, I don't know, haven't tried it yet:)

Another side question - what file is a config file for current version of cgminer? I have used the save option during mining and now it looks like no mater what options I run the miner with it will run with the saved ones, I looked in the cgminer folder and there are tons of various config files. I assume if I delete the config file it will revert to default settings?

It takes the config file if you specify it. I.e.: run it with the -c or --config parameter and pass the config file's path.
The contents should look something like this:

Code:

{
"pools" : [
	{
		"url" : "stratum+tcp://stratum.kano.is:3333",
		"user" : "username.worker1",
		"pass" : "x"
	}
],

"failover-only" : true,
"api-listen" : true,
"api-port" : "4028",
"api-allow" : "W:192.168.1.0/24,W:127.0.0.1"
}

Just input the information that you'd otherwise pass by command line args into that file. Then you can run it simply with ./cgminer -c example.conf.

I run the miner with these parameters -
sudo cgminer -o stratum+tcp://solo.ckpool.org:3333 -u bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv -p x --gekko-compacf-freq 400 --gekko-start-freq 375 --gekko-tune-up 85 --gekko-mine2 --gekko-tune2 60
In this case it supposed to run at 400Mhz but mine sets at 500Mhz as I believe I saved some day. Neither the start frequency or highest frequency works, it starts at 325 and goes up to 500...

Even if I run it with sudo cgminer -o stratum+tcp://solo.ckpool.org:3333 -u bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv -p x  it will still start at 325 and target 500Mhz

I don't remember what I did with settings as last time I fired up the sticks was back in Feb....

I just re-installed cgminer completely (removed the folder and redownloaded and made/installed it) and still are getting the same thing, so the config file was not in that folder.
The reason I know it shouldn't target for 500Mhz is it wasn't like this in the beginning. I think it was T at ~300Mhz, then I did something (saved the config?) and it starts at 500 now, but when I run it - I don't specify the config file - is there one that runs by default, not located in the cgminer folder? For some reason parameters dont work, like I said it always starts with 500 now, no matter what I put into --gekko-compacf-freq

[n0nce]

I run the miner with these parameters -
sudo cgminer -o stratum+tcp://solo.ckpool.org:3333 -u bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv -p x --gekko-compacf-freq 400 --gekko-start-freq 375 --gekko-tune-up 85 --gekko-mine2 --gekko-tune2 60
In this case it supposed to run at 400Mhz but mine sets at 500Mhz as I believe I saved some day. Neither the start frequency or highest frequency works, it starts at 325 and goes up to 500...

Even if I run it with sudo cgminer -o stratum+tcp://solo.ckpool.org:3333 -u bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv -p x  it will still start at 325 and target 500Mhz

That's odd; the sticks have no memory and this command doesn't load a configuration file. There is no reason it should go up to 500. You could try removing --gekko-compacf-freq 400 --gekko-start-freq 375 and it should only go up to 200MHz (default frequency). If that doesn't work, there's something really wrong (such as having setup an alias that calls cgminer with other options). Do you remember having done anything like that?

alias -p should show you whether you've set up an odd alias, by the way.

[PPICASSO]

I run the miner with these parameters -
sudo cgminer -o stratum+tcp://solo.ckpool.org:3333 -u bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv -p x --gekko-compacf-freq 400 --gekko-start-freq 375 --gekko-tune-up 85 --gekko-mine2 --gekko-tune2 60
In this case it supposed to run at 400Mhz but mine sets at 500Mhz as I believe I saved some day. Neither the start frequency or highest frequency works, it starts at 325 and goes up to 500...

Even if I run it with sudo cgminer -o stratum+tcp://solo.ckpool.org:3333 -u bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv -p x  it will still start at 325 and target 500Mhz

That's odd; the sticks have no memory and this command doesn't load a configuration file. There is no reason it should go up to 500. You could try removing --gekko-compacf-freq 400 --gekko-start-freq 375 and it should only go up to 200MHz (default frequency). If that doesn't work, there's something really wrong (such as having setup an alias that calls cgminer with other options). Do you remember having done anything like that?

alias -p should show you whether you've set up an odd alias, by the way.

Even if I run it with no frequency setting options (see above) it still runs at 500Mhz.

I remember setting something similar to alias, is there another way to do this other then alias command?
I always ran cgminer with no ./prior to it (as above) from anywhere in the filesystem (so I guess I register it as a command? but alias -p never showed it there?)
Now I found I had another copy in different folder (per kanos troubleshoot instructions), deleted it, reinstalled it and now have to only run it with ./ and only from its directory (I followed some guide when I was setting this up initially)

We have been out of state for a while, so sticks were collecting dust, came back and fired the system back up and it fired up right at 500Mhz and it never seems more then 8-11, that is my main problem.

How do I insert screenshots here btw?

Here is an example
Error /Timeout / Failure / Zombie messages appeared after I turned on the power for second hub, after first one initialized and started working(each has 9 sticks in it currently) at the end it only left 2 sticks running.....

Code:

 cgminer version 4.12.0 - Started: [2022-10-20 23:35:09.808]
--------------------------------------------------------------------------------
 (1s):2.000 (1m):1.338T (5m):832.2G (15m):341.5G (avg):2.114Th/s
 A:60000  R:0  HW:48  WU:29470.6/m
 Connected to solo.ckpool.org diff 10K with stratum as user bc1qgkdg0quzzpt5rcah7elrg06u5lvxdc34nkxwxv
 Block: 7d400d75...  Diff:35.6T  Started: [23:35:09.822]  Best share: 39.9K
--------------------------------------------------------------------------------
 [U]SB management [P]ool management [S]ettings [D]isplay options [Q]uit
 1: GSF 10051054: BM1397:01+ 500.00MHz T:500 P:497 (25:13) |  100% WU:100% | 1.625K / 227.4Gh/s WU:3176.4/m
 2: GSF 10051031: BM1397:01+ 500.00MHz T:500 P:499 (25:13) |  0.0% WU:100% | 2.227K / 229.1Gh/s WU:3200.1/m
 3: GSF 10051044: BM1397:01+ 500.00MHz T:500 P:496 (25:13) |  100% WU:100% | 2.522K / 220.2Gh/s WU:3075.7/m
 4: GSF 10051092: BM1397:01+ 500.00MHz T:500 P:497 (25:13) |  100% WU:100% | 2.728K / 221.4Gh/s WU:3093.4/m
 5: GSF 10053236: BM1397:01+ 500.00MHz T:500 P:0   (25:13) |  100% WU:100% | 2.347K / 230.3Gh/s WU:3217.9/m
 6: GSF 10051041: BM1397:01+ 500.00MHz T:500 P:497 (25:13) |  100% WU:100% | 3.226K / 221.9Gh/s WU:3099.4/m
 8: GSF 10053233: BM1397:01+ 500.00MHz T:500 P:500 (25:13) |  0.0% WU:100% | 2.607K / 216.8Gh/s WU:3028.2/m
 0: GSF 10051074: BM1397:01+ 500.00MHz T:500 P:499 (25:13) |  100% WU:100% | ZOMBIE
 7: GSF 10051045: BM1397:01+ 500.00MHz T:500 P:500 (25:13) |  100% WU:100% | ZOMBIE
 9: GSF 10051076: BM1397:01+ 500.00MHz T:500 P:482 (25:13) |  100% WU:100% | ZOMBIE
--------------------------------------------------------------------------------
 [2022-10-20 23:36:02.603] 1: GSF 1 - new frequency 495.00MHz -> 500.00MHz
 [2022-10-20 23:36:03.054] 7: GSF 7 - new frequency 495.00MHz -> 500.00MHz
 [2022-10-20 23:36:03.155] 8: GSF 8 - new frequency 485.00MHz -> 495.00MHz
 [2022-10-20 23:36:05.380] 8: GSF 8 - new frequency 495.00MHz -> 500.00MHz
 [2022-10-20 23:36:17.588] GSF 9 failure, disabling!
 [2022-10-20 23:36:27.450] Accepted 01a4632b Diff 39.9K/10000 GSF 7 pool 0
 [2022-10-20 23:36:45.540] Accepted 01eb8e86 Diff 34.1K/10000 GSF 8 pool 0
 [2022-10-20 23:36:52.473] Accepted 05e33c59 Diff 11.1K/10000 GSF 4 pool 0
 [2022-10-20 23:37:05.566] Accepted 024db1d3 Diff 28.4K/10000 GSF 4 pool 0
 [2022-10-20 23:37:14.266] GSF 7 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:24.541] GSF 7 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND
 [2022-10-20 23:37:24.541] GSF 0 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:24.543] GSF 3 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:24.542] GSF 7 failure, disabling!
 [2022-10-20 23:37:24.542] GSF 5 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:24.541] GSF 1 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:34.197] GSF 0 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND
 [2022-10-20 23:37:34.197] GSF 6 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:34.197] GSF 4 SendWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
 [2022-10-20 23:37:34.197] 7: GSF 7 - usb failure (-5)
 [2022-10-20 23:37:34.198] GSF 0 failure, disabling!
 [2022-10-20 23:37:37.652] GSF 3 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND
 [2022-10-20 23:37:40.954] GSF 5 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND
 [2022-10-20 23:37:44.271] GSF 1 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND

[n0nce]

Even if I run it with no frequency setting options (see above) it still runs at 500Mhz.

I remember setting something similar to alias, is there another way to do this other then alias command?
I always ran cgminer with no ./prior to it (as above) from anywhere in the filesystem (so I guess I register it as a command? but alias -p never showed it there?)

If you could run it without ./, it means you installed it; i.e. copied the binary to a location on your path.
It's possible that the binary on the path with that name, was actually a shell script though, that started cgminer with some specific parameters.
You could check that with which cgminer, which gets you the location of the script and cat /whatever/location/you/got/from/last/command to check the contents.

Then modify that file or delete it altogether.

Now I found I had another copy in different folder (per kanos troubleshoot instructions), deleted it, reinstalled it and now have to only run it with ./ and only from its directory (I followed some guide when I was setting this up initially)

If you download the software from https://github.com/kanoi/cgminer, build it and run it from that folder using ./cgminer, I see no reason why it would target a frequency other than the default. Maybe someone else has more ideas than me.

How do I insert screenshots here btw?

Upload somewhere (like imgur or postimg) and embed with [img] tags.

GUIDELINE ON POSTING IMAGES (WITH SIZE ADJUSTMENTS), HYPERLINKS

[...]

[Twister315]

I can run 9 sticks stable but as soon as I try to run 10 then one fails to start and the others start dropping into zombie. I'm using a pi 3 and gekko hubs. Cpu usage is high cgminer is using 85% with 9 sticks. Any ideas?

[PPICASSO]

I can run 9 sticks stable but as soon as I try to run 10 then one fails to start and the others start dropping into zombie. I'm using a pi 3 and gekko hubs. Cpu usage is high cgminer is using 85% with 9 sticks. Any ideas?

Are you running multiple hubs?
If I use them on single hub, I can usually run up to full hub, which is 15 on mine and won't be surprised it would run all 30 for few mins until voltage will get too low and they will start rebooting (30 port industrial hub with some spacing in between, every other port)
In order to stably run more then 9 I would need to dial in the pod on each, voltage gets too low on my hub...
My PI4 CPU actually stays around 20%, But if I plug in the second hub and/or start splitting sticks in between, THEN it goes nuts, like 80-100% and sticks start to zombie out, the ones that been hashing for a while, new ones don't appear,
Interesting enough - all sticks are flashing the white diodes (like they are hashing), even the ones that system does not see...

The problem is multiple hubs.....

[PPICASSO]

If you could run it without ./, it means you installed it; i.e. copied the binary to a location on your path.
It's possible that the binary on the path with that name, was actually a shell script though, that started cgminer with some specific parameters.
You could check that with which cgminer, which gets you the location of the script and cat /whatever/location/you/got/from/last/command to check the contents.

Ok, will check it out, thanks!

If you download the software from https://github.com/kanoi/cgminer, build it and run it from that folder using ./cgminer, I see no reason why it would target a frequency other than the default. Maybe someone else has more ideas than me.

Did that and oddly enough it still started at 500Mhz

Upload somewhere (like imgur or postimg) and embed with [img] tags.

Ok, thanks!

The issue is multiple hubs, looks like we got another victim here too...

I can run 9 sticks stable but as soon as I try to run 10 then one fails to start and the others start dropping into zombie. I'm using a pi 3 and gekko hubs. Cpu usage is high cgminer is using 85% with 9 sticks. Any ideas?

[Twister315]

I can run 9 sticks stable but as soon as I try to run 10 then one fails to start and the others start dropping into zombie. I'm using a pi 3 and gekko hubs. Cpu usage is high cgminer is using 85% with 9 sticks. Any ideas?

Are you running multiple hubs?
If I use them on single hub, I can usually run up to full hub, which is 15 on mine and won't be surprised it would run all 30 for few mins until voltage will get too low and they will start rebooting (30 port industrial hub with some spacing in between, every other port)
In order to stably run more then 9 I would need to dial in the pod on each, voltage gets too low on my hub...
My PI4 CPU actually stays around 20%, But if I plug in the second hub and/or start splitting sticks in between, THEN it goes nuts, like 80-100% and sticks start to zombie out, the ones that been hashing for a while, new ones don't appear,
Interesting enough - all sticks are flashing the white diodes (like they are hashing), even the ones that system does not see...

The problem is multiple hubs.....

I'm running 3 hubs. I can run 8 on 2 hubs or 9 on 3 hubs but as soon as I add one more stick the problems start.

[PPICASSO]

I'm running 3 hubs. I can run 8 on 2 hubs or 9 on 3 hubs but as soon as I add one more stick the problems start.

Check this out - https://forums.raspberrypi.com/viewtopic.php?f=28&t=53832

I finally figured it out. I'm running 16 sticks now @500Mhz on 2 hubs. Probably can even run on single hub, but will need voltmeter and pod adjusting...

My problem was that both of my 30 port hubs are constructed with 8 usb hub chips each. Each chip controls 4 ports, so its like 8 separate 4 port hubs connected.
For my two hubs system actually sees 16(!) 4 port hubs. It just cant support that many hubs.

Each hub has 2 usb cables, one for each side, so each usb cable supports 15 ports or half the hub. I originally had all 4 cables plugged in into the PI.

What I did - I only connected 2 cables to the pi, each cable goes to a half of each hub and loaded them with sticks (with single port spacing)
This achieved voltage/amperage needed to run the sticks (both sides share the power line) without overloading the system with amount of hubs connected.
I can only run 16 sticks this way, it will do for now.

Hope this makes sense.

[n0nce]

I'm running 3 hubs. I can run 8 on 2 hubs or 9 on 3 hubs but as soon as I add one more stick the problems start.

Check this out - https://forums.raspberrypi.com/viewtopic.php?f=28&t=53832

I finally figured it out. I'm running 16 sticks now @500Mhz on 2 hubs. Probably can even run on single hub, but will need voltmeter and pod adjusting...

I'm aware of Raspberry Pi (and generally on computers) USB limitations in terms of hubs and limited number of total connected devices.
Funnily enough, stick miners are one of the main scenarios where I've seen those happen; personally I prefer 'pod miners' or any other term that doesn't require a dedicated USB controller and complete power circuit per individual ASIC chip. Besides advantages like cost savings and compactness, it also prevents such issues.

But I don't get the connection to the frequency setting. That issue still persists, right? Without any frequency specified, they still go right up to 500?

[Twister315]

I'm running 3 hubs. I can run 8 on 2 hubs or 9 on 3 hubs but as soon as I add one more stick the problems start.

Check this out - https://forums.raspberrypi.com/viewtopic.php?f=28&t=53832

I finally figured it out. I'm running 16 sticks now @500Mhz on 2 hubs. Probably can even run on single hub, but will need voltmeter and pod adjusting...

My problem was that both of my 30 port hubs are constructed with 8 usb hub chips each. Each chip controls 4 ports, so its like 8 separate 4 port hubs connected.
For my two hubs system actually sees 16(!) 4 port hubs. It just cant support that many hubs.

Each hub has 2 usb cables, one for each side, so each usb cable supports 15 ports or half the hub. I originally had all 4 cables plugged in into the PI.

What I did - I only connected 2 cables to the pi, each cable goes to a half of each hub and loaded them with sticks (with single port spacing)
This achieved voltage/amperage needed to run the sticks (both sides share the power line) without overloading the system with amount of hubs connected.
I can only run 16 sticks this way, it will do for now.

Hope this makes sense.

I was able to remove a hub that my keyboard and mouse were using, so now I can run 10 sticks. Starts doing the same thing again when I try to go to 11 now.

[PPICASSO]

But I don't get the connection to the frequency setting. That issue still persists, right? Without any frequency specified, they still go right up to 500?

I dropped it all together, reinstalled fresh version of RPI and cgminer. Now it listens to  me!
I remember messing with system commands, somehow I set cgminer as a system command and set it with parameters. Last thing I found - I could just run

Code:

sudo cgminer

and it will run with all parameter (including my wallet address and all) So you were right, I just don't remember how I did it, and really wondering now.
I usually start to learn linux/unix only when I need it and usually go deep into it, then don't use it for a while and loose the knowledge.
Thanks for helping me out with suggestions!

I'm aware of Raspberry Pi (and generally on computers) USB limitations in terms of hubs and limited number of total connected devices.
Funnily enough, stick miners are one of the main scenarios where I've seen those happen; personally I prefer 'pod miners' or any other term that doesn't require a dedicated USB controller and complete power circuit per individual ASIC chip. Besides advantages like cost savings and compactness, it also prevents such issues.

Well, I got really good deal on these, they are powerful and compact. It took me some time to find the right hub (6 hubs later) and cooling setup for it. I'm using this gig just for lottery mining. I'm glad (now) I didn't buy S19+ farm when had a chance, Bitcoin is very unpredictable...

We are traveling a lot and I'm not mining then, I've seen a guy setting up a Tesla trunk farm, could be my next project.

[PPICASSO]

I was able to remove a hub that my keyboard and mouse were using, so now I can run 10 sticks. Starts doing the same thing again when I try to go to 11 now.

Is is it doing the same thing or different? Meaning do sticks just reboot from lower to higher frequency or they drop out completely? What's the CPU usage?
Mine is at 20% now, running 16 sticks at 500Mhz

The way to see that your USB bus is overloaded is your CPU usage, it will skyrocket. It only happened when I plugged in another hub WHILE cgminer RUNNING. When it happens, just stop cgminer (ctrl-c) and run it again, to save your PI from extra heat, it should zoom at around 20% normally.
If I plug in and turn all hubs on BEFORE i run it, it would just find x amount of sticks (not all) and just run normally at low CPU load.
So, it the cgminer that gets confused by all the hub stuff and goes nuts.

Btw, I'm only using USB 2.0 ports (white inside), don't use USB 3.0 ports (blue inside)
Try only running 2 USB wires to PI (to USB 2.0 ports!), only from 2 hubs, rest of the hubs just interconnect, meaning connect the wire from hub 1 and 2 to PI, as for the other hub - just plug in the USB wire to hub 1 or 2. Let me know if it works. You can plug in keyboard/mouse directly into your other 2 ports on PI.

[Twister315]

I was able to remove a hub that my keyboard and mouse were using, so now I can run 10 sticks. Starts doing the same thing again when I try to go to 11 now.

Is is it doing the same thing or different? Meaning do sticks just reboot from lower to higher frequency or they drop out completely? What's the CPU usage?
Mine is at 20% now, running 16 sticks at 500Mhz

The way to see that your USB bus is overloaded is your CPU usage, it will skyrocket. It only happened when I plugged in another hub WHILE cgminer RUNNING. When it happens, just stop cgminer (ctrl-c) and run it again, to save your PI from extra heat, it should zoom at around 20% normally.
If I plug in and turn all hubs on BEFORE i run it, it would just find x amount of sticks (not all) and just run normally at low CPU load.
So, it the cgminer that gets confused by all the hub stuff and goes nuts.

Btw, I'm only using USB 2.0 ports (white inside), don't use USB 3.0 ports (blue inside)
Try only running 2 USB wires to PI (to USB 2.0 ports!), only from 2 hubs, rest of the hubs just interconnect, meaning connect the wire from hub 1 and 2 to PI, as for the other hub - just plug in the USB wire to hub 1 or 2. Let me know if it works. You can plug in keyboard/mouse directly into your other 2 ports on PI.

Sticks just start dropping out into zombie. Cpu is at 85%. I'm only using usb 2.0 and tried running the third hub off of one of the other two and that didn't help. Still can't go over 10 sticks.

[PPICASSO]

Sticks just start dropping out into zombie. Cpu is at 85%. I'm only using usb 2.0 and tried running the third hub off of one of the other two and that didn't help. Still can't go over 10 sticks.

That's the same exact issue I have had. I remedied it by using only half the hub, but my hubs are 30 slots each. Read up on that link I sent above, pull up USB tree and see what chips are used in the hub and how many hubs system sees there. I would also try to completely reinstall the Raspbian and Cgminer (30 min) and start over from clean setup.

[kano]

But I don't get the connection to the frequency setting. That issue still persists, right? Without any frequency specified, they still go right up to 500?

I dropped it all together, reinstalled fresh version of RPI and cgminer. Now it listens to  me!
I remember messing with system commands, somehow I set cgminer as a system command and set it with parameters.

If it was loading the config file you saved:
If you save a config file, using the default it tells you, but then don't want to use it, you gotta find it and delete it.

It will be saved in the directory of the user who ran cgminer.
If you use sudo then the user is root, not the user that typed the command.
That's what 'sudo' means.

When you pressed S+W it told you where it saved it.

Something like /root/.cgminer/cgminer.conf or /home/yourusername/.cgminer/cgminer.conf

[citb0in]

Can anyone run

Code:

cgminer --benchmark

without getting an error message? I tried several variations and combinations with or without --benchfile and this work file (genesis block)

Code:

1,0e3e2357e806b6cdb1f70b54c3a3a17b6714ee1f0e68bebb44a74b1efd512098,000000000019d6689c085ae165831e934ff763ae46a2a6c172b3f1b60a8ce26f,1d00ffff,1231469665

but whatever I tried, when running in benchmark mode I'm always getting a memory error and abnormal program termination. "cgminer -n" shows correct and expected output and mining itself works.

Is benchmark deprecated from the latest version?

[cygan]

today an usb hub pair (4-5) on one of my gekko hubs died and the two ports have no voltage anymore
on each of the two ports 1 gekkosience compac f (580mhz) stick and the corresponding fan were connected

[citb0in]

hopefully your Compac-F didn't get damaged ?