[GUIDE] GridSeed GC3355 5 Chip Setup/power/windows/linux/rpi by UnicornHasher

[BotwinBG]

My units are red - what are the practical differences - should I feel speshul compared to the golden peasants or just jump off a bridge?

Sometimes the units do this :

Code:

0: >>> LTC : 55aa1f2816000000
0: >>> LTC : 55aa1f2817000000
0: >>> LTC : 55aa1f000000000000000000000000000000000000000000000000000000c0ff3f00000009a1993dbcaaeb744a2f84d36bd2540e97e4df56486ff3ead1a12a70d0db998700000002d50931d314c0e2070a387b02a95ca2605c1e61b569d425fc9a34ac01588e7f0aecdc35f508449ec1eaf003bfa0a02374a6f0470e0b9719c94201c16e6941a0155313d8f31c08174500000000ffffffff12345678
[2014-03-03 02:21:04] Stratum detected new block
[2014-03-03 02:21:04] 0: Dispatching new work to GC3355 LTC core

for like an eternity , with pools not reporting any hashrate. Any magic tricks to make them do the YAY or is everything as it should be?

There is some difference which I have yet to find.  The lack of yay usually means you are not submitting shares.  You could try another pool but it sounds like you are submitting shares as you said sometimes.

Yeah, they seem to get stuck somehow in that mode, sometimes a powercycle helps, sometimes not. Getting slightly irritating, cause of the constant babysitting.

[jane2]

 accepted: 1166/1166 (100.00%), 0.00 khash/s (yay!!!)
i want to chack hashrate cpuminer 2.3.2...

is not there any way other than to check pool?

[philipma1957]

This is a real thing?

I thought the whole thing was a parody of ASIC vaporware.

Lol.  Just because I sell live unicorns doesn't mean all ASIC's are vaporware

Get back to me when you have Petite Lap Giraffes.

that was a very good commercial .

[miaviator]

accepted: 1166/1166 (100.00%), 0.00 khash/s (yay!!!)
i want to chack hashrate cpuminer 2.3.2...

is not there any way other than to check pool?

There is currently no way to check hashrate aside from using the controller or checking at the pool.

It'll likely be 0-3 months before full support is added to make that possible.

[jane2]

accepted: 1166/1166 (100.00%), 0.00 khash/s (yay!!!)
i want to chack hashrate cpuminer 2.3.2...

is not there any way other than to check pool?

There is currently no way to check hashrate aside from using the controller or checking at the pool.

It'll likely be 0-3 months before full support is added to make that possible.

okay...
thx

[bryanlee38]

wow. good work on the setup guide. The time and patience it would have taken you! You SIR are a good man!

[miaviator]

Added wireless info:

Part 4: Setting up the controller

If you do not have wired internet available for the controller box you can easily setup a wireless bridge to run these or any other miners more reliably over wireless:
http://www.trendnet.com/products/proddetail.asp?prod=180_TEW-432BRP
http://www.trendnet.com/products/proddetail.asp?prod=230_TEW-651BR

I use one on my laser printer and have used dozens of the 432s for various projects.  I have a box of them here that I'm not using at the moment.

[EricKennedy]

Thanks miaviator for this complete and accurate tutorial. I'd managed to get the ASICs hashing on Windows, but only got them running on Linux thanks to you.

You state that "There is currently no way to check hashrate aside from using the controller or checking at the pool".

Actually, there is one : cpuminer has a -S option, which allows writing its output to syslog. You then need to write a syslog filter to isolate output from the different cpuminers and output to a separate logfile, and some shell/perl/ruby script to parse the logfile and produce relevant statistics. You won't get directly the hash rate, but using the accepted lines, you can make an accurate guess.

Fully agree with you that this souldn't be an issue if Jack was providing proper support to the product he sells.

Eric

[miaviator]

Thanks miaviator for this complete and accurate tutorial. I'd managed to get the ASICs hashing on Windows, but only got them running on Linux thanks to you.

You state that "There is currently no way to check hashrate aside from using the controller or checking at the pool".

Actually, there is one : cpuminer has a -S option, which allows writing its output to syslog. You then need to write a syslog filter to isolate output from the different cpuminers and output to a separate logfile, and some shell/perl/ruby script to parse the logfile and produce relevant statistics. You won't get directly the hash rate, but using the accepted lines, you can make an accurate guess.

Fully agree with you that this souldn't be an issue if Jack was providing proper support to the product he sells.

Eric

Thanks Eric,

I'm going to leave out that last part unless we can make an automated method of showing the estimated hash rate

Can you link to this guide in your Tech Support OP?

[EricKennedy]

Can you link to this guide in your Tech Support OP?

OP updated

[BrandonMcPherson]

Thanks for the setup guide!

I have a single unit, no controller. How in the heck do you A.) know what mode it's in, and B.) change modes?

I have scoured the Interwebs and not found an answer to this question. I assume that means I'm missing something incredibly obvious.

[csmflynt3]

My units are red - what are the practical differences - should I feel speshul compared to the golden peasants or just jump off a bridge?

Sometimes the units do this :

Code:

0: >>> LTC : 55aa1f2816000000
0: >>> LTC : 55aa1f2817000000
0: >>> LTC : 55aa1f000000000000000000000000000000000000000000000000000000c0ff3f00000009a1993dbcaaeb744a2f84d36bd2540e97e4df56486ff3ead1a12a70d0db998700000002d50931d314c0e2070a387b02a95ca2605c1e61b569d425fc9a34ac01588e7f0aecdc35f508449ec1eaf003bfa0a02374a6f0470e0b9719c94201c16e6941a0155313d8f31c08174500000000ffffffff12345678
[2014-03-03 02:21:04] Stratum detected new block
[2014-03-03 02:21:04] 0: Dispatching new work to GC3355 LTC core

for like an eternity , with pools not reporting any hashrate. Any magic tricks to make them do the YAY or is everything as it should be?

There is some difference which I have yet to find.  The lack of yay usually means you are not submitting shares.  You could try another pool but it sounds like you are submitting shares as you said sometimes.

I had this issue as well seems unplugging/replugging usb connection resolves it. I changed the port speed to 115200 and run it at 800 MHZ which seems to make me have to do this with much less frequency. Anything higher and its every few hrs I must reset it seems.

[miaviator]

Thanks for the setup guide!

I have a single unit, no controller. How in the heck do you A.) know what mode it's in, and B.) change modes?

I have scoured the Interwebs and not found an answer to this question. I assume that means I'm missing something incredibly obvious.

Check the windows/linux sections and follow the steps.  To hash in scrypt only mode you just start using the scrypt instruction set.  To hash in dual mode you turn on BTC hashing then scrypt hashing with the --dual flag.  If anything is missing or unclear let me know and I'll correct it.

[BrandonMcPherson]

Thanks, mviator—I read right past the part where you explained precisely what I was looking for. Thanks for the setting me straight!

[wolfey2014]

My units are red - what are the practical differences - should I feel speshul compared to the golden peasants or just jump off a bridge?

Sometimes the units do this :

Code:

0: >>> LTC : 55aa1f2816000000
0: >>> LTC : 55aa1f2817000000
0: >>> LTC : 55aa1f000000000000000000000000000000000000000000000000000000c0ff3f00000009a1993dbcaaeb744a2f84d36bd2540e97e4df56486ff3ead1a12a70d0db998700000002d50931d314c0e2070a387b02a95ca2605c1e61b569d425fc9a34ac01588e7f0aecdc35f508449ec1eaf003bfa0a02374a6f0470e0b9719c94201c16e6941a0155313d8f31c08174500000000ffffffff12345678
[2014-03-03 02:21:04] Stratum detected new block
[2014-03-03 02:21:04] 0: Dispatching new work to GC3355 LTC core

for like an eternity , with pools not reporting any hashrate. Any magic tricks to make them do the YAY or is everything as it should be?

There is some difference which I have yet to find.  The lack of yay usually means you are not submitting shares.  You could try another pool but it sounds like you are submitting shares as you said sometimes.

I had this issue as well seems unplugging/replugging usb connection resolves it. I changed the port speed to 115200 and run it at 800 MHZ which seems to make me have to do this with much less frequency. Anything higher and its every few hrs I must reset it seems.

I think I'm having a UART problem in that my cheapo $20 Inland 10 port hub's UARTs ain't keeping up because they aren't really HIGH SPEED chips , just FULL SPEED just 1.1 compatible shyt... which I don't frikkin need...that is if its really the hub....might have only one downstream UART and 1 upstream instead of a high speed full duplex uart at each port, which I am thinking costs more than $20 ;]. I don't mind spending the money for a good hub but they gotta stop BSing people about them being real 2.0 High Speed hubs.....UGH!

Anyway, anyone know what a good REAL 10+ port 2.0 high speed hub costs? Brand name? Model number would be great, too! Wink

Was hashing great downstream 'receiving and dispatching work to GC3355 but would get hung up with the comeback downstream work...didn't do work, nothing. Just keep receiving and going DDDDUHHHHHHHHH!
So I lowered the port speed on all comm ports to 38400...and lowered the clock speed from 850 to 800 for now until I am sure I've solved the port issues.

So far Winblows 7 is holding up otherwise.
Forget about solving comm port issues on XP machines. Never ending battle, I fear. Fret not, Win7 has no comm port issues, assignment wise. Sees device, assigns comm port automatically, no hassles.

While I was hashing away at 850 for around 6 hours before the UARTs bottle necked and said "WHAT?" I was doing up to 496KH/s! All 5 total hashed at 1950+KH/s at times per pool stats. Ran at around 85* F. Whole smash running at no more than 25 to 30 Watts total for all 5 pods. Nice! And that's with only 2 fans cooling all 5 pods.
They're made to run much warmer - full time - so, 30+ Watts ain't nuttin! Saves around 10.65 Watts of energy which translates into more coin for my little machines to pay themselves off in a few months!

Let me know how you do. I'm sure others will be equally curious if not as adventurous or,, brave Wink
Peace!
Wolfey2014

Pea Ess....
Running the latest version of LTC Energy Saver cpuminer - if anyone's interested...

[Powell]

I need some help with Gridseed, I'm running cpuminer and when it starts

C:\Users\mining\Desktop\cpuminer\cpuminer-2.3.2-GC3355-win32>minerd  --freq=600
--gc3355=com17 --url=stratum+tcp://litecoinpool.org:3333 --userpass=XXXXXXXX.
XX:X
[2014-02-26 18:15:31] Starting Stratum on stratum+tcp://litecoinpool.org:3333
[2014-02-26 18:15:31] GC3355 chip mining thread started, in SINGLE mode
[2014-02-26 18:15:31] Open com17 failed, GetLastError:2

Any idea on what to do? The virtual com17 say is working properly.


thanks

Use

--gc3355=\\.\COM17 instead, that should fix it. you need the direct address for COM10 onwards

This actually helped big time the \\.\COM11 (and so on) is working on my machine currently now.  So I have 10 running in Windows 7 x64 currently.

[miaviator]

I think I'm having a UART problem in that my cheapo $20 Inland 10 port hub's UARTs ain't keeping up because they aren't really HIGH SPEED chips , just FULL SPEED just 1.1 compatible shyt... which I don't frikkin need...that is if its really the hub....might have only one downstream UART and 1 upstream instead of a high speed full duplex uart at each port, which I am thinking costs more than $20 ;]. I don't mind spending the money for a good hub but they gotta stop BSing people about them being real 2.0 High Speed hubs.....UGH!

Anyway, anyone know what a good REAL 10+ port 2.0 high speed hub costs? Brand name? Model number would be great, too! Wink

Was hashing great downstream 'receiving and dispatching work to GC3355 but would get hung up with the comeback downstream work...didn't do work, nothing. Just keep receiving and going DDDDUHHHHHHHHH!
So I lowered the port speed on all comm ports to 38400...and lowered the clock speed from 850 to 800 for now until I am sure I've solved the port issues.

So far Winblows 7 is holding up otherwise.
Forget about solving comm port issues on XP machines. Never ending battle, I fear. Fret not, Win7 has no comm port issues, assignment wise. Sees device, assigns comm port automatically, no hassles.

While I was hashing away at 850 for around 6 hours before the UARTs bottle necked and said "WHAT?" I was doing up to 496KH/s! All 5 total hashed at 1950+KH/s at times per pool stats. Ran at around 85* F. Whole smash running at no more than 25 to 30 Watts total for all 5 pods. Nice! And that's with only 2 fans cooling all 5 pods.
They're made to run much warmer - full time - so, 30+ Watts ain't nuttin! Saves around 10.65 Watts of energy which translates into more coin for my little machines to pay themselves off in a few months!

Let me know how you do. I'm sure others will be equally curious if not as adventurous or,, brave Wink
Peace!
Wolfey2014

Pea Ess....
Running the latest version of LTC Energy Saver cpuminer - if anyone's interested...

Can you try running on a powered USB 2.0 hub with at least 1Amp per port?  I think the .5Amp per port or overall power of your hub may be the issue vs the speed.

[miaviator]

I need some help with Gridseed, I'm running cpuminer and when it starts

C:\Users\mining\Desktop\cpuminer\cpuminer-2.3.2-GC3355-win32>minerd  --freq=600
--gc3355=com17 --url=stratum+tcp://litecoinpool.org:3333 --userpass=XXXXXXXX.
XX:X
[2014-02-26 18:15:31] Starting Stratum on stratum+tcp://litecoinpool.org:3333
[2014-02-26 18:15:31] GC3355 chip mining thread started, in SINGLE mode
[2014-02-26 18:15:31] Open com17 failed, GetLastError:2

Any idea on what to do? The virtual com17 say is working properly.


thanks

Use

--gc3355=\\.\COM17 instead, that should fix it. you need the direct address for COM10 onwards

This actually helped big time the \\.\COM11 (and so on) is working on my machine currently now.  So I have 10 running in Windows 7 x64 currently.

Adding to the Windows section at the top!

[BrandonMcPherson]

I went back through the instructions and ran into a problem.

First, I successfully have CGminer hashing in SHA-256 mode. When I then go to run CPUminer in scrypt mode with the dual flag, I get this error:

Code:

./minerd: error while loading shared libraries: libjansson.so.4: cannot open shared object file: No such file or directory

I went through the CPUminer installation instructions again to be sure I hadn't missed something. All of the commands completed successfully, but I do still get the error above.

I also tried launching minderd without CGminer running.

Note that I know just enough about Linux to be highly dangerous, and I appreciate any help from the folks who know more.

[MZD]

Pulling my hair out. lol.

I had to go with the controller option instead of PC because of electrical wiring in my house.

So I setup a second router as an AP and used my main router to bridge it.

I can see both of the controller ip/mac addresses. When I type them in.....nothing. Cannot connect.

Any ideas guys?