ROCKMINER ASIC miner official thread

[ThePhwner]

Hey Rockminer,

Thanks to Eyeboot, I got a new heatsink for that R3. However I think you'll notice something wrong in the picture below. Same unit. Not even 30 days in. Using a pretty nice 80plus gold PSU to run it at stock speed. I don't expect much given how the overheating issue was handled but thought others might like to see this. Again, I got 2 and the 1 running right next to it has been fine. I've tried to RMA this twice now, and people just quit talking to us.

https://i.imgur.com/oOlQNGU.jpg

Can't see in the picture - is either of the burned pins (in the red PCI-E plug or in the white socket) pushed in? Typically this happens when there is poor contact and that can be caused by slightly misaligned pins. They don't go into each other, you force the plug, and one of the pins gets pushed in or bent resulting in bad connection, heat, sparks, and melted plastic.

I'll take a closer look but tbh, this unit has been a dud since the beginning, see my earlier post. I've connected enough cables to enough miners to know not to force and make bad connections, and really RockMiner needs to replace this unit for me.

I'm sorry, I'm not sure I understand. You mentioned one of the blades had a different heatsink and was overheating. You got a replacement heatsink - did that fix the problem? If it did and now the issue is the burned cable it would be counterproductive for you to lose mining time waiting for a replacement. Check the connector on the blade and the blade itself if there aren't any other visible problems with it. Clean the scene of the accident. Connect another PCI-E cable. Check if it doesn't get hot. That's what I would do anyway. Unless there are other issues with the miner of course.

The heatsink did not fix the issue, in fact. One blade still runs much hotter than all others causing the unit to reboot, and now the power connection issue. At this point, any mining time I would lose, is balance by the amount of my own time I have spend babysitting, adding additional fans, replacing the SD card in the Rpi, re-doing thermal compound twice, and now running on 3/4 capacity because of this fried connection.

Given that the R3 sitting right next to it is 100% fine, this one seems defective.

If you think this is acceptable, you have a much lower standard for ASICs than I do.

[suchmoon]

If you think this is acceptable, you have a much lower standard for ASICs than I do.

I was just trying to help, please try to not read between the lines when there is nothing there.

[ManeBjorn]

My review of the T1 went live this morning.
I really enjoyed working with the miner it works quite well and is a great evolution of the Rockminer series.
http://www.cryptocoinsnews.com/news/review-rockminer-t1-800-ghs-bitcoin-asic-miner/2014/08/25

Rockminer has been growing fast in the Bitcoin ASIC miner industry and they have recently taken another step forward with their Rockminer T1 miner. This miner starts things off with a new design that steps away from the tube design that many of the Bitcoin ASIC mining manufacturers are starting to use.
The new design allows for stacking and ease of use in that you can easily access cables and connections while maintaining a hot and cold isle in large installations, as well as general simple ease of use in smaller installations. The T1 also has a new controller, instead of a Raspberry Pi, it comes with a BE200 Jet Stratum Miner V 5.47 controller. This new controller can run 3 T1 miners. The specs of the T1 are good as well with Rockminer upping the speed again over the R3-Box.

[Trends]

I didn't see any Rockminer logo's or name on any pictures of the blades in your write up so I going to guess that what I mentioned before is correct!
Rockminer is repackaging someone else's build. I'm guessing FC's team did the R&D (Asicminer Tube Blade design was posted by FC on GIT 1-2 months ago), someone other than Rockminer build the pc boards to tha Aicminer design and Rockminer just assembles the pieces.
So I don't think all the credit should go to Rockminer. They certainly weren't much help with their RK boxes or their R3's, just a bunch of empty promises.

Trends

[Swimmer63]

My review of the T1 went live this morning.
I really enjoyed working with the miner it works quite well and is a great evolution of the Rockminer series.
http://www.cryptocoinsnews.com/news/review-rockminer-t1-800-ghs-bitcoin-asic-miner/2014/08/25

Rockminer has been growing fast in the Bitcoin ASIC miner industry and they have recently taken another step forward with their Rockminer T1 miner. This miner starts things off with a new design that steps away from the tube design that many of the Bitcoin ASIC mining manufacturers are starting to use.
The new design allows for stacking and ease of use in that you can easily access cables and connections while maintaining a hot and cold isle in large installations, as well as general simple ease of use in smaller installations. The T1 also has a new controller, instead of a Raspberry Pi, it comes with a BE200 Jet Stratum Miner V 5.47 controller. This new controller can run 3 T1 miners. The specs of the T1 are good as well with Rockminer upping the speed again over the R3-Box.

Good writeup, but no pictures of the miner as a whole.  Just insides and one profile.  What the heck does the thing look like?

[klintay]

Just finishing setting up my Rockminer T1 800-900 Gh/s ASIC Bitcoin Miner!

Here is a pic of my  hashing away:

And the speed:

898 GH/s!!!

Thank you Rockminer and Alex for helping me set it up

Each board has one or two PCIe power connectors ?

Two PCIe connectors. You can use 1 connector if you are using a single 1000 watt psu or you can use 2 if you only have two 500 watt psu like me.

[dogie]

Good writeup, but no pictures of the miner as a whole.  Just insides and one profile.  What the heck does the thing look like?

Thats what your community dog is for, pictures will be up tonight or tomorrow.

Two PCIe connectors. You can use 1 connector if you are using a single 1000 watt psu or you can use 2 if you only have two 500 watt psu like me.

Do NOT plug two different PSUs into one board, never ever ever ever do that. Thats how you end up with two dead PSUs and a dead hashing board.

[klintay]

Do NOT plug two different PSUs into one board, never ever ever ever do that. Thats how you end up with two dead PSUs and a dead hashing board.

Good advice! I am glad i didn't do that. I plug one psu into two boards and the other into the rest.

[proxyme]

I just put a Rockminer T1 in my server room yesterday. It seems to be hashing fine but this thing makes more noise than all my ProLiant servers combined. Sounds almost like a jet engine. BE200 Jet Stratum Miner. The humor does not escape me. When I'm in there I have to shut it down.

What are other peoples experience with the sound level? Any ideas on how to lower it? On the Statistics page there are two buttons: Clock Up and Clock Down, but they dont seem to have an effect on the fan speed.

It would be cool with an option to cut the speed down to half for example while I'm in there.

[ManeBjorn]

Sorry I was taking such close pics of it.  Mine does not have the full cage on it so mine as a whole is as you see it.
I will take some more pics in the morning and add them for you here and to the review.
Thanks

My review of the T1 went live this morning.
I really enjoyed working with the miner it works quite well and is a great evolution of the Rockminer series.
http://www.cryptocoinsnews.com/news/review-rockminer-t1-800-ghs-bitcoin-asic-miner/2014/08/25

Rockminer has been growing fast in the Bitcoin ASIC miner industry and they have recently taken another step forward with their Rockminer T1 miner. This miner starts things off with a new design that steps away from the tube design that many of the Bitcoin ASIC mining manufacturers are starting to use.
The new design allows for stacking and ease of use in that you can easily access cables and connections while maintaining a hot and cold isle in large installations, as well as general simple ease of use in smaller installations. The T1 also has a new controller, instead of a Raspberry Pi, it comes with a BE200 Jet Stratum Miner V 5.47 controller. This new controller can run 3 T1 miners. The specs of the T1 are good as well with Rockminer upping the speed again over the R3-Box.

Good writeup, but no pictures of the miner as a whole.  Just insides and one profile.  What the heck does the thing look like?

[ManeBjorn]

I look forward to your guides as always Dogie.
 

Good writeup, but no pictures of the miner as a whole.  Just insides and one profile.  What the heck does the thing look like?

Thats what your community dog is for, pictures will be up tonight or tomorrow.

Two PCIe connectors. You can use 1 connector if you are using a single 1000 watt psu or you can use 2 if you only have two 500 watt psu like me.

Do NOT plug two different PSUs into one board, never ever ever ever do that. Thats how you end up with two dead PSUs and a dead hashing board.

[klintay]

I just put a Rockminer T1 in my server room yesterday. It seems to be hashing fine but this thing makes more noise than all my ProLiant servers combined. Sounds almost like a jet engine. BE200 Jet Stratum Miner. The humor does not escape me. When I'm in there I have to shut it down.

What are other peoples experience with the sound level? Any ideas on how to lower it? On the Statistics page there are two buttons: Clock Up and Clock Down, but they dont seem to have an effect on the fan speed.

It would be cool with an option to cut the speed down to half for example while I'm in there.

clock up and clock down change the frequency (hashing speed) of the chip. What hashing speed are you getting?

Hey Rockminer,

Thanks to Eyeboot, I got a new heatsink for that R3. However I think you'll notice something wrong in the picture below. Same unit. Not even 30 days in. Using a pretty nice 80plus gold PSU to run it at stock speed. I don't expect much given how the overheating issue was handled but thought others might like to see this. Again, I got 2 and the 1 running right next to it has been fine. I've tried to RMA this twice now, and people just quit talking to us.

https://i.imgur.com/oOlQNGU.jpg

Can't see in the picture - is either of the burned pins (in the red PCI-E plug or in the white socket) pushed in? Typically this happens when there is poor contact and that can be caused by slightly misaligned pins. They don't go into each other, you force the plug, and one of the pins gets pushed in or bent resulting in bad connection, heat, sparks, and melted plastic.

I'll take a closer look but tbh, this unit has been a dud since the beginning, see my earlier post. I've connected enough cables to enough miners to know not to force and make bad connections, and really RockMiner needs to replace this unit for me.

I'm sorry, I'm not sure I understand. You mentioned one of the blades had a different heatsink and was overheating. You got a replacement heatsink - did that fix the problem? If it did and now the issue is the burned cable it would be counterproductive for you to lose mining time waiting for a replacement. Check the connector on the blade and the blade itself if there aren't any other visible problems with it. Clean the scene of the accident. Connect another PCI-E cable. Check if it doesn't get hot. That's what I would do anyway. Unless there are other issues with the miner of course.

The heatsink did not fix the issue, in fact. One blade still runs much hotter than all others causing the unit to reboot, and now the power connection issue. At this point, any mining time I would lose, is balance by the amount of my own time I have spend babysitting, adding additional fans, replacing the SD card in the Rpi, re-doing thermal compound twice, and now running on 3/4 capacity because of this fried connection.

Given that the R3 sitting right next to it is 100% fine, this one seems defective.

If you think this is acceptable, you have a much lower standard for ASICs than I do.

Minersource should be really sorting this out for you as they were the reseller. How long have you had the miner for before it went wrong? I believe Minersource.net have a warranty on their products normally.


Also i have notice that www.rockminer.com does not have a list of authorised resellers anymore...maybe it is me who can't find the link   or can nobody else find it too

[proxyme]

Hi klintay,

I'm getting roughly the same hashing as you. I haven't tried higher than 300MHz. Oddly enough, my miner reports 100% hardware error on board 9 & 10 and 0% for the other 2, but it seems fine and Slush's pool reports around 892GH/s as i type this.

Yes, the Clock up/down are in increments of 10. But what I was hoping for was to lower the speed to say half and thereby slowing down the fan speed and noise. This would be for an hour or two while I sit next to it. Afterwards I could go back to full speed, leave the room and lock the door.

I know this might be bit of a luxury problem, but as it is now, I have to completely cut the power temporarily.

Tomorrow I'll open the box and take a look at those 4 fans. I'm thinking they may be getting full power no matter what hashing speed I'm at. Perhaps I can find some sort of regulator to put in between. I don't know. Hardware is not my strongest side.

On a final note, I think the web setup is a bit on the simple side compared to say an Antminer.
It does have this mysterious system port that I have to look into and also run a port scanner against it to find any other ports and if I can shell into it.


http://imgur.com/OF9obV8

[proxyme]

Screen shot

https://i.imgur.com/OF9obV8.png

https://i.imgur.com/OF9obV8.png

[klintay]

Hi klintay,

I'm getting roughly the same hashing as you. I haven't tried higher than 300MHz. Oddly enough, my miner reports 100% hardware error on board 9 & 10 and 0% for the other 2, but it seems fine and Slush's pool reports around 892GH/s as i type this.

Yes, the Clock up/down are in increments of 10. But what I was hoping for was to lower the speed to say half and thereby slowing down the fan speed and noise. This would be for an hour or two while I sit next to it. Afterwards I could go back to full speed, leave the room and lock the door.

I know this might be bit of a luxury problem, but as it is now, I have to completely cut the power temporarily.

Tomorrow I'll open the box and take a look at those 4 fans. I'm thinking they may be getting full power no matter what hashing speed I'm at. Perhaps I can find some sort of regulator to put in between. I don't know. Hardware is not my strongest side.

On a final note, I think the web setup is a bit on the simple side compared to say an Antminer.
It does have this mysterious system port that I have to look into and also run a port scanner against it to find any other ports and if I can shell into it.

Good to hear that the hw error is an anomaly...what speed does Slush pool report and how many rejects?

I ditto your idea about the fans, i notice they don't seem to have much suction power. Also with regards to the fan speed control that is a good idea and would be useful. Lets both email Alex at Rockminer and make the suggestion. Maybe they can release a firmware update for the BE controller that can do this.  

[R-T-B]

It does have this mysterious system port that I have to look into and also run a port scanner against it to find any other ports and if I can shell into it.

I'd be careful trying a traditional TCP/IP ssh shell to it...  knowing what "SYS" means in router nonemclature, it could be a cisco-style rollover port.  This is basically full serial port, voltages and all routed over ethernet (lookup a "cisco rollover cable").  If that's what it is, it will kill the ethernet of whatever you hook it up to!

EDIT:  Oh, you don't mean that second port I've seen in the pictures?  Do you mean a standard TCP/IP port?  If so ignore.

[dogie]

You can get 100% HW errors if a board has dropped out. The hash rate is very averaged so can take much longer to drop to 0 than for HW errors to rise. Just restart the controller once you're sure the PSU isn't misbehaving.

[proxyme]

It does have this mysterious system port that I have to look into and also run a port scanner against it to find any other ports and if I can shell into it.

I'd be careful trying a traditional TCP/IP ssh shell to it...  knowing what "SYS" means in router nonemclature, it could be a cisco-style rollover port.  This is basically full serial port, voltages and all routed over ethernet (lookup a "cisco rollover cable").  If that's what it is, it will kill the ethernet of whatever you hook it up to!

EDIT:  Oh, you don't mean that second port I've seen in the pictures?  Do you mean a standard TCP/IP port?  If so ignore.

Yeah, I meant the system port 7478 from the settings page here:

https://i.imgur.com/XxxnyHr.png

I ran a port scan against the controller and nmap doesn't tell much except this:

Code:

7478/tcp open  unknown

1 service unrecognized despite returning data. If you know the service/version, please submit the following fingerprint at http://www.insecure.org/cgi-bin/servicefp-submit.cgi :

SF-Port7478-TCP:V=6.40%I=7%D=8/27%Time=53FD9167%P=i686-pc-windows-windows%

SF:r(GenericLines,1,"E")%r(GetRequest,1,"E")%r(DNSStatusRequest,1,"E")%r(H

SF:elp,1,"E")%r(X11Probe,1,"E");

MAC Address: 00:1E:C0:CD:1D:35 (Microchip Technology)

Warning: OSScan results may be unreliable because we could not find at least 1 open and 1 closed port

Device type: specialized|WAP|phone

Running: iPXE 1.X, Linksys Linux 2.4.X, Sony Ericsson embedded

OS CPE: cpe:/o:ipxe:ipxe:1.0.0%2b cpe:/o:linksys:linux_kernel:2.4 cpe:/h:sonyericsson:u8i_vivaz

OS details: iPXE 1.0.0+, Tomato 1.28 (Linux 2.4.20), Sony Ericsson U8i Vivaz mobile phone

So far I haven't had any luck with that port. I'll have a go at it some more when I get the time.
I'm assuming the other physical port on the controller is a serial port. I think I recognize it from some of my network equipment. I have some serial cables from a Procurve switch that will fit. Later this week I will try it and fire up Hyperterm.

Anyone know what the last 3 parameters on the settings page are for?

RollNtime
WakeUp period
Normal/Smart[0/1]

I'm having a hard time finding any documentation for this box. If anyone has any I'll appreciate it. It doesn't look like many are running it.

Regards and thanks

[proxyme]

Regarding the HW errors:

From what I can tell, it appears to be a display glitch. If a board drops out it would be indicated on the test status page, I assume. Unless there's a display glitch on that page also.

Slush's pool reports roughly the same hashing rate as the controller status page. I can't seem to find the rejects anywhere on the pool site.

[dogie]

Dogie's comprehensive setup guide in super HD!