[Guide] Dogie's Comprehensive ASICMiner Cube Setup [HD]

[J_Dubbs]

...and they are all running through Slush's proxy (one for each, different ports) on an older laptop.

Why are you running separate instances of the proxy? You should only be running 1 instance of the proxy for any number of cubes. I have 3 cubes and 1 instance of the proxy and I get >32GHs on all 3 cubes.

Regarding running slush's proxy on RPi. I haven't seen anyone reporting more than 1 cube on an RPi getting full hashing rate. I tried my 3 cubes on RPi and it pegged the cpu 100% and hash rates were < 16GHs. I have not yet tried bfgminer proxy on RPi. I might get around to trying that tomorrow.

Here is a quote from the thread that convinced me to run separate proxies:

https://bitcointalk.org/index.php?topic=352658.msg4369389#msg4369389

so if you have 3 cubes connected to one single proxy they will appear as one worker to the pool

I actually run 3 instances of bfgminer proxy on my pi, each using about 10% cpu. still way less than slush's so I can use three command lines

Longpoll is currently not implemented in the bfgminer getwork proxy, so that will result in low efficiency / lots of stales.  Using slush's proxy, cubes do appear as separate workers to the pool when running a single instance of slush's proxy, however, I can't recommend that either.  While playing around with slush's proxy and the stratum-mining pool software, I discovered that if you have multiple cubes connected to one instance of slush's proxy, then the proxy will get the share difficulties confused and only submit shares that meet the highest difficulty requirements of all workers.
Based on the above findings, currently the best setup is ONE INSTANCE OF SLUSH'S PROXY PER CUBE, however much cpu time that ends up using.  I really wish longpoll would be implemented in the bfgminer getwork proxy soon...  Either that, or the cube firmware should be updated to support stratum natively.

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

[utdrmac]

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

I can only speak to my personal 1st hand experience. I run 1 instance of slush's proxy, pointed to BTCGuild, and all 3 cubes show as independent workers. Also, all 3 cubes can handle the same difficulty so whatever the OP of that page was referring to about difficulties is a moot point for me. I believe the latest version of slush's proxy has a flag for propagating difficulties so if that's of need to you (like you run cubes and other hardware that's not as powerful), you might want to test that. But if you are cube-only, you should be fine with just one instance of proxy running.

[dogie]

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

I can only speak to my personal 1st hand experience. I run 1 instance of slush's proxy, pointed to BTCGuild, and all 3 cubes show as independent workers. Also, all 3 cubes can handle the same difficulty so whatever the OP of that page was referring to about difficulties is a moot point for me. I believe the latest version of slush's proxy has a flag for propagating difficulties so if that's of need to you (like you run cubes and other hardware that's not as powerful), you might want to test that. But if you are cube-only, you should be fine with just one instance of proxy running.

This. Run ONE instance of the stratum proxy. It will identify individual clients and treat them as such. Even if you have different hash rates, set them all at the same difficulty.

[J_Dubbs]

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

I can only speak to my personal 1st hand experience. I run 1 instance of slush's proxy, pointed to BTCGuild, and all 3 cubes show as independent workers. Also, all 3 cubes can handle the same difficulty so whatever the OP of that page was referring to about difficulties is a moot point for me. I believe the latest version of slush's proxy has a flag for propagating difficulties so if that's of need to you (like you run cubes and other hardware that's not as powerful), you might want to test that. But if you are cube-only, you should be fine with just one instance of proxy running.

This. Run ONE instance of the stratum proxy. It will identify individual clients and treat them as such. Even if you have different hash rates, set them all at the same difficulty.

Cool, I'll go with one as I never liked setting up multiples anyways... Thanks!

[divinesnail01]

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

I can only speak to my personal 1st hand experience. I run 1 instance of slush's proxy, pointed to BTCGuild, and all 3 cubes show as independent workers. Also, all 3 cubes can handle the same difficulty so whatever the OP of that page was referring to about difficulties is a moot point for me. I believe the latest version of slush's proxy has a flag for propagating difficulties so if that's of need to you (like you run cubes and other hardware that's not as powerful), you might want to test that. But if you are cube-only, you should be fine with just one instance of proxy running.

This. Run ONE instance of the stratum proxy. It will identify individual clients and treat them as such. Even if you have different hash rates, set them all at the same difficulty.

That doesn't fit with my experience, and I switched back last night just to be sure before I said anything.

I'm running 2 cubes, and if I point them both at the same instance of the proxy, both run consistently in the 36 range, with my pool reporting a combined of almost 72.  If I point them at two different instances of the proxy, on the same PC, one runs consistently at about 38.8 and the other at about 38.2, with my pool reporting just over 76.  It's a small difference, and I can't say I've been particularly scientific about my testing, but nevertheless it seems to be easily replicated.

YMMV, but I'm a fan of 2 copies of the proxy.

'snail

[dogie]

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

I can only speak to my personal 1st hand experience. I run 1 instance of slush's proxy, pointed to BTCGuild, and all 3 cubes show as independent workers. Also, all 3 cubes can handle the same difficulty so whatever the OP of that page was referring to about difficulties is a moot point for me. I believe the latest version of slush's proxy has a flag for propagating difficulties so if that's of need to you (like you run cubes and other hardware that's not as powerful), you might want to test that. But if you are cube-only, you should be fine with just one instance of proxy running.

This. Run ONE instance of the stratum proxy. It will identify individual clients and treat them as such. Even if you have different hash rates, set them all at the same difficulty.

That doesn't fit with my experience, and I switched back last night just to be sure before I said anything.

I'm running 2 cubes, and if I point them both at the same instance of the proxy, both run consistently in the 36 range, with my pool reporting a combined of almost 72.  If I point them at two different instances of the proxy, on the same PC, one runs consistently at about 38.8 and the other at about 38.2, with my pool reporting just over 76.  It's a small difference, and I can't say I've been particularly scientific about my testing, but nevertheless it seems to be easily replicated.

YMMV, but I'm a fan of 2 copies of the proxy.

'snail

I dont disagree that it might work for you, but its not a limitation of the proxy. It could be something stupid as a limit on how many packets you can ram through one port with your particular setup, which could have 10 variables that create the situation. No worries.

[J_Dubbs]

I dunno, sounded right to me so I went with it... I'd love to just run one proxy though as it will simplify things... Maybe it's splitting hairs and only a small improvement?

I can only speak to my personal 1st hand experience. I run 1 instance of slush's proxy, pointed to BTCGuild, and all 3 cubes show as independent workers. Also, all 3 cubes can handle the same difficulty so whatever the OP of that page was referring to about difficulties is a moot point for me. I believe the latest version of slush's proxy has a flag for propagating difficulties so if that's of need to you (like you run cubes and other hardware that's not as powerful), you might want to test that. But if you are cube-only, you should be fine with just one instance of proxy running.

This. Run ONE instance of the stratum proxy. It will identify individual clients and treat them as such. Even if you have different hash rates, set them all at the same difficulty.

That doesn't fit with my experience, and I switched back last night just to be sure before I said anything.

I'm running 2 cubes, and if I point them both at the same instance of the proxy, both run consistently in the 36 range, with my pool reporting a combined of almost 72.  If I point them at two different instances of the proxy, on the same PC, one runs consistently at about 38.8 and the other at about 38.2, with my pool reporting just over 76.  It's a small difference, and I can't say I've been particularly scientific about my testing, but nevertheless it seems to be easily replicated.

YMMV, but I'm a fan of 2 copies of the proxy.

'snail

I can't go with high clock anyways, too unstable on my cubes for some reason, most of them at least... On low clock I see no difference in my hash rates using just the one proxy, but what I did notice is that one of my cubes took a minute to talk to the pool, which makes me think it was stuck in some type of queue behind the other cubes- I used to see small delays on my pool back when I was trying to jam 10 blades on a single slush proxy.

[IamCANADIAN013]

I'm curious. My BE Cube is running at 37200 on high.  Can I make that higher? I see some numbers running around 420, can all run that at the same time?

[Apteryx]

I'm curious. My BE Cube is running at 37200 on high.  Can I make that higher? I see some numbers running around 420, can all run that at the same time?

It's a bit of potluck with the exact speed you get on HI. All you can do is switch to Hi and make sure there are no bottlenecks in the system.

My cube runs consistently at 38,7xx with slushs proxy running on a raspberry PI

[stef942]

Hi, i have a be quiet system power PSU 700w connected to a 49 ports usb hub and im wondering if i can use it to feed a cube with the usb hub connected at the same time  
Do you think its possible?

Specs of the PSU :

[dogie]

I can't go with high clock anyways, too unstable on my cubes for some reason, most of them at least... On low clock I see no difference in my hash rates using just the one proxy, but what I did notice is that one of my cubes took a minute to talk to the pool, which makes me think it was stuck in some type of queue behind the other cubes- I used to see small delays on my pool back when I was trying to jam 10 blades on a single slush proxy.

That's just the pool update cycle being slow. If the blades really were queuing on each other, you'd see horrific efficiencies (and so hash rates).

[dogie]

I'm curious. My BE Cube is running at 37200 on high.  Can I make that higher? I see some numbers running around 420, can all run that at the same time?

~38 is the official high clock speed hash rate, you can only hit 40 with cooling mods. The difference in hash rate probably doesn't even pay for your time though.

[dogie]

Hi, i have a be quiet system power PSU 700w connected to a 49 ports usb hub and im wondering if i can use it to feed a cube with the usb hub connected at the same time  
Do you think its possible?

Specs of the PSU :

Depends on the power consumption of the hub. You'd have to use a PCI-E pin connector from TWO separate rails to power the cube.

[qwertyasdfg]

FINALLY today I received a new board via OC directly from yxt, now I've taken the old board out and put the new board in after switching over the mini blades and am trying to connect to it using 192.168.1.254:8000 but am unable to do so, I cannot even see the device anywhere in the wired devices section of my router!

Anyone got any advice on what I might be doing wrong here?? My routers subnet is 192.168.1.x

Do I have to flash this new board with some sort of firmware before I can use it to control the mini blades???

Your router isn't 192.168.1.254 is it?  That was an issue I had when initially setting mine up.

Other than that, you might have to do a factory reset... strange though that you could connect with the old cube but not with the new boards. 

Nope router is deffo not 192.168.1.254

I'm guna try a factory reset and will let you know if that works!

That seems to have done the trick however the x's have moved

ASIC_01-16: O O O O O O O O O O O O O O O O
ASIC_17-32: O O O O O O O O O O O O O O O O
ASIC_33-48: x x x x x x x x O O O O O O O O
ASIC_49-64: O O O O O O O O O O O O O O O O
ASIC_65-80: O O O O O O O O O O O O O O O O
ASIC_81-96: O O O O O O O O O O O O O O O O

So I re-arranged the mini blades once again and the x's moved again, so I must have a faulty mini blade now which is rather annoying!

ASIC_01-16: x x x x x x x x O O O O O O O O
ASIC_17-32: O O O O O O O O O O O O O O O O
ASIC_33-48: O O O O O O O O O O O O O O O O
ASIC_49-64: O O O O O O O O O O O O O O O O
ASIC_65-80: O O O O O O O O O O O O O O O O
ASIC_81-96: O O O O O O O O O O O O O O O O

I received the replacement mini blade this morning and have replaced the faulty blade in my cube and it's all up and running now with no more x's but for some strange reason I cannot seem to get it functioning properly in High Clock mode!

Every time I try and switch to overclocked mode the cube starts loosing it's hashrate and efficiency till it functions no more and I'm left with all x's at which point I have to restart the machine to get it hashing again only in low clock mode, can anyone suggest what might be going wrong here?

[gotnerves]

I cannot get mine to work for the life of me, can someone tell me what is wrong with the setup?

http://i363.photobucket.com/albums/oo76/Urlshekk/bitcoinmining_zps7fa3abe9.png

[IamCANADIAN013]

I cannot get mine to work for the life of me, can someone tell me what is wrong with the setup?

Could be a firewall issue.  I had a similar problem and it turned out my firewall was preventing the cube from connecting.

[gotnerves]

I cannot get mine to work for the life of me, can someone tell me what is wrong with the setup?

Could be a firewall issue.  I had a similar problem and it turned out my firewall was preventing the cube from connecting.

thanks for responding so quickly, i just disabled Eset to no avail, it still is doing the same thing.

[stex2009]

Does it make a difference if I run a cube with less than 6 boards in it? My cube with less blades usually disconnects itself after a few minutes, no matter what PSU I use.

[IamCANADIAN013]

I cannot get mine to work for the life of me, can someone tell me what is wrong with the setup?

Could be a firewall issue.  I had a similar problem and it turned out my firewall was preventing the cube from connecting.

thanks for responding so quickly, i just disabled Eset to no avail, it still is doing the same thing.

I'm sure someone will come along soon enough here that will be able to help.  Unfortunately my knowledge is limited when it comes to this stuff.

Good luck.

[dogie]

Does it make a difference if I run a cube with less than 6 boards in it? My cube with less blades usually disconnects itself after a few minutes, no matter what PSU I use.

Shouldn't make any difference, your problem is elsewhere.

I cannot get mine to work for the life of me, can someone tell me what is wrong with the setup?

Could be a firewall issue.  I had a similar problem and it turned out my firewall was preventing the cube from connecting.

thanks for responding so quickly, i just disabled Eset to no avail, it still is doing the same thing.

Config is right, but cube is unable to talk to the proxy (hence long pool is off). Try moving where the cube is on the network in relation to the miner.