Topic: [GUIDE] GridSeed Blade (80 Chip) Miner Support/Tuning - page 2. (Read 38912 times)

HI Everyone,

I did some extensive work on my Gridseed blades and found now the perfect tuning / modification combination.

If you replace the resistor to an 40KOhm resistor I highly recommend a better capacitor, PSU and also better mosfet.

Here are some photos of my latest Blade I modded. Also highly recommended you better replace the Power infeed like I did otherwise your Power connector will melt.  

https://www.dropbox.com/s/trfftgyykag742g/Photo%20May%2011%2C%204%2036%2047%20PM.jpg?dl=0

We do also offer this service at Lumanet.info just let us know and we work with you. Even on repair.

https://www.dropbox.com/s/to43143yrk08d7f/Photo%20May%2011%2C%204%2037%2011%20PM.jpg

https://www.dropbox.com/s/vufe69q5dwo8bt9/Photo%20May%2011%2C%204%2048%2020%20PM.jpg

https://www.dropbox.com/s/ojfstren7ibwfi8/Photo%20May%2011%2C%205%2043%2058%20PM.jpg

oh eee!

sry for bumping this, but i need the command line for this version of cgminer, especially for the com port, how to set it via batch file(not config)

update on my problem - I ended up destroying the ferrite bead on the other half of the blade and now I am simply going to rule it out as a bad connection on my psu or perhaps on the molex adapter I made.  I purchased 10 new ferrite beads from Digikey for maybe 6 dollars.  I've replaced them using the super lazy method of wire conductive glue (something I had around the house)  One of the two halves of the dead blade now works!  However the other half refuses to power on (I first plugged them into the usb to see if they can communicate with the pc - this broken blade refuses to work - I've tried to reinstall the bead, press the reset button on the asw100 and also even went so far as to try simply bridging that place where the bead is to be installed (from my understanding it simply acts as a low pass filter and the asic should function at least still) with the glue and nothing seems to get any connection.  I've examined the board carefully and see no damage so I'm at a complete loss for explanation)

Hey all, I've seen peeps in here running massive amounts of blades, and I'm having an issue running 3 gridseeds (6 blades). 

Part of my issue, is one of them is modded and overclocked, so it requires individual settings.  But I managed to get that done, and managed to get all 6 blades running mining all one particular coin.  They ran fine like this for nearly a month.

But now I've decided that it's too much magahash to throw all at one coin, since the overall mining of the coin is lowered..  (no sense throwing all 18 mh/s at a coin that's only got 6 mh's besides me mining it)  So I've created separate batch and config files for each gridseed.  This was sort of a pita to do as it required me tagging all the cables so I could plug them in at the proper time, but once it was running it was solid for a day or so.

Well, as I've noticed from time to time, a blade will eventually quit producing work, even though it appears to be doing so.  Before this would have just taken killing and restarting cgminer then unplugging and replugging all the blades back in using hotplug.

With the 3 separate instances, hotplug has to be turned off and all the blades called out individually, so when one drops, you only kill one other blade and replug both blades called out by that instance.  That was great, it kept the other 4 blades mining when only one needed a restart.  And it worked for a couple days.....

.....then this morning all hell broke loose.  I went to restart a blade, and could not get CGminer (the windows ALT version) to recognise both blades at once, it would crash.  I could start one or the other, but both plugged in at once would crash.

Natural first solution is "have you turned it off and on again", so I shut everything down and rebooted the rig.   After that, I could not get any instance of CGminer to recognize 2 blades at startup without crashing, and if I started CGminer with an individual blade, I could start 2 instances, but the 3rd would crash.

So, I try the Jmordica version of CGminer, swap USB cables, ports, and hubs, reinstall the USB drivers in winblows, and still get the same thing..

After 3 hours of screwing around, the only way I could get all the blades running at once was to use hotplug, when I had it running for days before without an issue.

Any clues as to the root cause and how to fix my issue?

Seriously though - im in the EXACT same boat.  I came to the last page hoping to see some sort of hardware support back and forth here...I'll place an update AFTER I have to sift through the posts.

UPDATE:  Just as I suspected - no support that I could see except for software based support and tuning suggestions. 

@eraser - I fixed it at least by making a new set of wiring for my units (bought a 10 pack dc barrel plug adapters purchased from amazon and used the pcie 6x to molex connectors I had from some gpu boxes.  Now it works again.  I would suggest you check your wiring.

in other news, I completely broke one of my blades (both sides) by burning out the ab126 (IIRC?) ferrite bead next to the female barrel jack where the plug fits into.

would it work to simply throw a lot of conductive glue over the area?  (what is this ferrite bead for?)

I'm sorry for my outburst...my fault.
I was  hoping for anyone with same problems that found a way to resolve it.
ty

this is a support/tuning thread for the blades not .... a vendor support thread (zoomhash or whatever) ....

Dear all,
i ordered 2 blades from zoomhash/sevengnomes on 27 May, received 19 June.
it works 4 days than the first blade stopped to work and the day after the second too.
The fan doesn't work , no green light on board ( only red light ).
I can see the hashpower on hashra dashboard but it doesn't mine any shares.
I'm waiting a definitive response from sevengnomes support from 4 days.
Spent 1700$ , 4 mining days and no warranty on that products....
very BAD customer service....


Marco

Ok I am at hair tearing out time.

Trying to run a gridseed blade with cgminer.  I cant seem to get cgminer to find the blades itself.  I have to start cgminer then plug the blades in 1 by 1.  Which is a pain and should be unnecessary.  I am also wanting to use cgwatcher so that in the event a power cut or whatever everything will start mining again by itself.

I'm using cgminer 3.7.2 blade version
Used zadig to change drivers
this is the line currently using      -o stratum+tcp://stratum.doge.hashfaster.com:3339 -u nzgrant.5 -p2 --scrypt --gridseed-options=freq=850,chips=40


Any thoughts?

Hi,

Can you please let me know the command line(exact one) to pass in cgminer batch file to run it with your fork in? I tried but my g-blade is either not running or when it runs, it's on default config. i.e. freq to be 888 with other similar parameters.


Thanks

P.S. Please help, as I've gone crazy trying to run my g-blade with bfgminer but got a lot(!) of errors and no acceptable hash-rate(2-5.1mh/s at max.).

jmordica,

Thanks for this. I love it verses cpuminer. I only have one problem. I can get my full advertised hash rate for a good 4-6 hours and then my HW rate increases and my hash rate drops by almost half.

Is there something I'm doing wrong???

Here's my current config file:

{
        "pools": [
                {
                        "url": "stratum+tcp://uswest.wafflepool.com:3333",
                        "user": "1QJoj61hmoBYQg8j4YEhxLJDxRjGVyYMtw_RBP1",
                        "pass": "d=1024"
                },
                {
                        "url": "stratum+tcp://us-west2.multipool.us:7777",
                        "user": "RBP1",
                        "pass": "123"
                },
                {
                        "url": "stratum+tcp://scryptguild.com:3333",
                        "user": "RBP1",
                        "pass": "123"
                },
                {
                        "url": "stratum+tcp://us.clevermining.com:3333",
                        "user": "1QJoj61hmoBYQg8j4YEhxLJDxRjGVyYMtw",
                        "pass": "d=1024"
                }
        ],
        "gridseed-options": "baud=115200,freq=838,chips=40,modules=1,usefifo=0",
        "hotplug": "30",
        "worksize": "1",
        "kernel": "scrypt",
        "thread-concurrency": "8192",
        "lookup-gap": "0",
        "gpu-threads": "2",
        "queue": "0",
        "scan-time": "1",
        "expiry": "1",
        "auto-fan": true,
        "scrypt": true,
        "api-listen": true,
        "api-allow": "W:192.168.2.0/24",
        "api-port": "4028",

I'm currently running 5 Blades:

--------------------------------------------------------------------------------
 (5s):28.10M (avg):28.32Mh/s | A:1157632  R:10240  HW:154  WU:20.9/m
 ST: 1  SS: 0  NB: 215  LW: 17782  GF: 0  RF: 0
 Connected to uswest.wafflepool.com diff 1.02K with stratum as user 1QJoj61hmoBYQg8j4YEhxLJDxRjGVyYMtw_RBP1
 Block: 261aff9a...  Diff:25  Started: [20:01:34]  Best share: 885K
--------------------------------------------------------------------------------
 [P]ool management ettings [D]isplay options [Q]uit
 GSD  0: 48E273793137  838 MHz | 2.838M/2.831Mh/s | A: 99328 R:1024 HW:18 WU: 1.8/m
 GSD  1: 48E36F6C3137  838 MHz | 2.630M/2.832Mh/s | A:133120 R:1024 HW: 8 WU: 2.4/m
 GSD  2: 48ED655F3137  838 MHz | 2.837M/2.837Mh/s | A:132096 R:   0 HW: 6 WU: 2.4/m
 GSD  3: 48FD814A3137  838 MHz | 2.838M/2.833Mh/s | A:122880 R:   0 HW:24 WU: 2.2/m
 GSD  4: 48FB727A3137  838 MHz | 2.836M/2.829Mh/s | A:131072 R:1024 HW:17 WU: 2.4/m
 GSD  5: 48E0565B3137  838 MHz | 2.693M/2.835Mh/s | A:113664 R:1024 HW: 2 WU: 2.1/m
 GSD  6: 48FD824E3731  838 MHz | 2.837M/2.830Mh/s | A:105984 R:1024 HW:31 WU: 1.9/m
 GSD  7: 48FC6F5E3731  838 MHz | 2.839M/2.834Mh/s | A:122880 R:2048 HW: 5 WU: 2.2/m
 GSD  8: 48F975483430  838 MHz | 2.837M/2.833Mh/s | A: 82432 R:2048 HW:36 WU: 1.5/m
 GSD  9: 48F2633A3430  838 MHz | 2.839M/2.831Mh/s | A:114176 R:1024 HW: 7 WU: 2.1/m
--------------------------------------------------------------------------------

 [2014-06-29 20:00:51] Accepted 0b29edb7 Diff 5.87K/1024 GSD 7 pool 0
 [2014-06-29 20:00:56] Stratum from pool 0 detected new block
 [2014-06-29 20:01:01] Accepted 262668e6 Diff 1.72K/1024 GSD 1 pool 0
 [2014-06-29 20:01:02] Stratum from pool 0 detected new block
 [2014-06-29 20:01:02] Rejected 3219709d Diff 1.31K/1024 GSD 9 pool 0 (job not found 7a14)
 [2014-06-29 20:01:05] Stratum from pool 0 detected new block
 [2014-06-29 20:01:08] Accepted 19d84ab2 Diff 2.54K/1024 GSD 5 pool 0
 [2014-06-29 20:01:14] Accepted 11b8a5df Diff 3.7K/1024 GSD 3 pool 0
 [2014-06-29 20:01:14] Stratum from pool 0 detected new block
 [2014-06-29 20:01:15] Accepted 3421e991 Diff 1.26K/1024 GSD 9 pool 0
 [2014-06-29 20:01:19] Stratum from pool 0 detected new block
 [2014-06-29 20:01:19] Rejected 3ec0c198 Diff 1.04K/1024 GSD 5 pool 0 (job not found 7a18)
 [2014-06-29 20:01:21] Accepted 1aa8e8fb Diff 2.46K/1024 GSD 6 pool 0
 [2014-06-29 20:01:26] Stratum from pool 0 detected new block
 [2014-06-29 20:01:27] Accepted 2fedbd7b Diff 1.37K/1024 GSD 4 pool 0
 [2014-06-29 20:01:28] Accepted 2d696d3c Diff 1.44K/1024 GSD 9 pool 0
 [2014-06-29 20:01:32] Accepted 26fafaab Diff 1.68K/1024 GSD 1 pool 0
 [2014-06-29 20:01:32] Accepted 364bf0c8 Diff 1.21K/1024 GSD 7 pool 0
 [2014-06-29 20:01:32] Stratum from pool 0 detected new block
 [2014-06-29 20:01:32] Accepted 2622c618 Diff 1.72K/1024 GSD 0 pool 0
 [2014-06-29 20:01:34] Stratum from pool 0 detected new block
 [2014-06-29 20:01:37] Accepted 0c9d2b85 Diff 5.2K/1024 GSD 1 pool 0
 [2014-06-29 20:01:39] Accepted 2b7075d0 Diff 1.51K/1024 GSD 7 pool 0

Check out this thread. https://bitcointalksearch.org/topic/m.6813270, I got mine today, will post as soon as I can figure out how to attach jpg's.

bump

what is the best vardiff for these g-blades?
i notice its two boards running ~2.5mh so i need to apply settings around this?

still trying to find a guide for that. I don't know the .conf commands needed to force which units run on which instance

No, I don't sorry.  I only run blades, but I've read elsewhere that it can be done but it is more trouble than it's worth.
This thread may be helpful.  https://forum.give-me-coins.com/discussion/634/mining-litecoin-with-different-gpus/p1
Why don't you run two instances?

Any chance you know of a guide to do this ? I have given up after all my searching throughout the days, bit worn out

Yes, but it is a pain to setup and not recommended.  It's much easier to have them in separate windows.

ive compiled your miner but i still can not get it to show the proper hashrate!