Cairnsmore1 - Quad XC6SLX150 Board - page 43.

ebereon

sr. member

Activity: 397

Merit: 500

Update on working miner soft with the new bitstreams:
Makomk's bitstreams:

latest MPBM git testing branch with icarus worker
latest cgminer
latest bfgminer

Glassworker's bitstream

latest MPBM git testing branch with glasswalkers worker
latest cgminer should work as Kano wrote it supports 57600 baud with the new options

If other miner soft works with one of the bitstreams, please info to me and i will update this.

eb

testconpastas2

full member

Activity: 199

Merit: 100

Quote from: Entropy-uc on August 01, 2012, 04:11:50 PM

Quote from: testconpastas2 on August 01, 2012, 04:06:33 PM

well after 24h testing one board with makomks 160 it worked well for a while but after 8-10h (in the morning) ICA 1 was hashing at half its power... after reseting board with SW1 1-off it started to run again at full speed. it seems as one of the 4 fpga had stopped.

Are ICA 0 allways the same 2 fpgas? 0-1? 0-3? or they are randomly assigned ?

can you detect wich fpga is frozen apart from watching which fpga green led is not blinking? ( my boards are too high and i need to use a ladder).

I think my issues can be temperature ones. Here in Seville 40ºC (outside) is normal in summer. I would like to watch temp in my cgminer so I could analize behavior ( maybe next controller?)

I bougth a new conceptronic 4 port powered usb hub and it works perfectly in my aspire one. My old soyntec doesn't even work with my laptop

In windows ICA 0 will always be the com port that you point CGminer to with -S first. I think linux might swap port assignments around if you restart.

40C is probably hot enough that you are overheating the FPGAs. Enterpoint's thermal solution is very good, but it's constrained by cheap packaging of the Spartan-6. With 8-9W I think you would be above the rated t-j for this device.

Thank you.

there isn't 40º at home, maybe 25-27 .
After resetting it seems it's working ok, maybe was bad luck I'll keep an eye on it. twin_test sometimes give me problems so maybe is not a bitstream issue. if it last more than two days then I'll think about using makomks'.

Entropy-uc

hero member

Activity: 756

Merit: 501

Quote from: testconpastas2 on August 01, 2012, 04:06:33 PM

well after 24h testing one board with makomks 160 it worked well for a while but after 8-10h (in the morning) ICA 1 was hashing at half its power... after reseting board with SW1 1-off it started to run again at full speed. it seems as one of the 4 fpga had stopped.

Are ICA 0 allways the same 2 fpgas? 0-1? 0-3? or they are randomly assigned ?

can you detect wich fpga is frozen apart from watching which fpga green led is not blinking? ( my boards are too high and i need to use a ladder).

I think my issues can be temperature ones. Here in Seville 40ºC (outside) is normal in summer. I would like to watch temp in my cgminer so I could analize behavior ( maybe next controller?)

I bougth a new conceptronic 4 port powered usb hub and it works perfectly in my aspire one. My old soyntec doesn't even work with my laptop

In windows ICA 0 will always be the com port that you point CGminer to with -S first. I think linux might swap port assignments around if you restart.

40C is probably hot enough that you are overheating the FPGAs. Enterpoint's thermal solution is very good, but it's constrained by cheap packaging of the Spartan-6. With 8-9W I think you would be above the rated t-j for this device.

ebereon

sr. member

Activity: 397

Merit: 500

Update on latest bitstreams after 17 hours:
Total 40 FPGA's = 10 boards SN#400+.

Makomk's oc bitstreams:
14x 200Mh, 22x 190Mh, 1x 180Mh, 3x 150Mh. That means 90% of all fpga's running @190Mh+ with <2.8% invalids!

Working bitstreams in percent:
200Mh oc = 35%
190Mh oc = 90%
180Mh oc = 90%
170Mh oc = 90%
160Mh = 100%
150Mh = 100%
140Mh = 100%

Next update with non oc'd bitstreams Wink

Update Glasswalker's bitstreams:
not changed will update when new bitstream is out

eb

testconpastas2

full member

Activity: 199

Merit: 100

well after 24h testing one board with makomks 160 it worked well for a while but after 8-10h (in the morning) ICA 1 was hashing at half its power... after reseting board with SW1 1-off it started to run again at full speed. it seems as one of the 4 fpga had stopped.

Are ICA 0 allways the same 2 fpgas? 0-1? 0-3? or they are randomly assigned ?

can you detect wich fpga is frozen apart from watching which fpga green led is not blinking? ( my boards are too high and i need to use a ladder).

I think my issues can be temperature ones. Here in Seville 40ºC (outside) is normal in summer. I would like to watch temp in my cgminer so I could analize behavior ( maybe next controller?)

I bougth a new conceptronic 4 port powered usb hub and it works perfectly in my aspire one. My old soyntec doesn't even work with my laptop

hm

member

Activity: 107

Merit: 10

board 62-0017 using cgminer
no problems with flashing, no lost usb-connections.

shipping_build.bit	don't know how well it worked
twin_test.bit	sometimes stable at around 380mh/s for up to two days, but most of the time between 40 and 200mh/s after some hours or minutes at around 380mh/s, never 0mh/s
makomk_icarus_cm1_140_test.bit	0mh/s after 2 to 4 "Accepted" messages
glasswalkers fpgaminer_top.bit	0mh/s
glasswalker2.bit (current)	running >1d21h, pool reports between 100 and 300mh/s, cgminer output see below

disabling usb power by taping the 5v connector insibe the host-side usb plug did not resolve the stability issues.

=> unstable board

latest cgminer output (glasswalker2.bit):

Code:

 cgminer version 2.6.1 - Started: [2012-07-31 00:28:42]
--------------------------------------------------------------------------------
 (5s):1509.8 (avg):948.9 Mh/s | Q:5119  A:10827  R:70  HW:0  E:212%  U:4.0/m
 TQ: 6  ST: 6  SS: 2  DW: 635  NB: 260  LW: 89625  GF: 108  RF: 13
 Connected to http://us2.eclipsemc.com:8337 with LP as user hm_worker
 Block: 000003525bdd224714873325e187e633...  Started: [21:44:19]
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 ICA 0:                | 187.7/186.7Mh/s | A:3542 R:21 HW:0 U:1.30/m
 ICA 1:                | 189.3/188.7Mh/s | A:3621 R:21 HW:0 U:1.33/m
 ICA 2:                | 194.7/193.8Mh/s | A:3664 R:28 HW:0 U:1.34/m
 ICA 3:                | 380.0/379.7Mh/s | A:   0 R: 0 HW:0 U:0.00/m
--------------------------------------------------------------------------------

spiccioli

legendary

Activity: 1378

Merit: 1003

nec sine labore

Quote from: kano on August 01, 2012, 09:59:45 AM

I've written an Icarus change for cgminer that will support 3 new options:
baud rate (115200 or 57600) work divisor (1, 2, 4 or 8 ) and number of FPGA
This should even work with the old setup with only 1 of 2 FPGA working

https://github.com/ckolivas/cgminer/pull/283
Anyone interested come visit #cgminer as usual ... tomorrow ...

kano,

I'd like to ask one thing: the HW: value gives the number of invalid hashes that have been returned by a GPU, can this control be enable for FPGAs as well?

MPBM has a column in its web-page interface which tells how many invalid shares have been returned, can cgminer test returned shares to see if they're valid?

spiccioli

Isokivi

hero member

Activity: 910

Merit: 1000

Items flashing here available at btctrinkets.com

This is a make-shift guide for windows users for faster, permanent flashing. I have not yet tested it, but apparrentley Slipbye has had succes with it. It also gets us out of the virtual machine (for good ?)

21:13] http://sourceforge.net/projects/libusb-win32/files/libusb-win32-releases/1.2.6.0/libusb-win32-devel-filter-1.2.6.0.exe/download
[21:14] install that, make sure the board to be flashed is plugged, start the libusb filter wizard
[21:14] select the USB composite device which shows an ID of 0403 8350
[21:14] install the filter driver for that
[21:14] download this: https://xc3sprog.svn.sourceforge.net/svnroot/xc3sprog/trunk/xc3sprog.exe
[21:14] create a new file called cablelist.txt in the same directory
[21:15] put this line inside that file:
[21:15] cm1 ftdi 20000000 0x0403:0x8350:
[21:15] open a command prompt in the directory where the files are and run these commands:
[21:15] set CABLEDB=cablelist.txt
[21:15] xc3sprog -c cm1
[21:15] it should detect the fpgas
a this point you want to copy the .bit files you'll be using to the same folder as xc3sprog is in.
[21:16] if that worked, you can go ahead with flashing like usual
[21:16] xc3sprog -c cm1 -p 0 -Ixc6lx150.bit file_to_be_flashed.bit

Enjoy your 1-3hours of spare time per day

daemonic

newbie

Activity: 49

Merit: 0

Quote from: ShadesOfMarble on August 01, 2012, 12:53:08 PM

Quote from: ebereon on July 31, 2012, 10:58:12 PM

my early board SN#0015 is not working at all

Are you sure this board isn't broken? I can run 150 Mhz just fine on my S/N 26 board.
Where are the other pre-S/N-50 board users? I would love to hear some more experiences so maybe we can get a better image on how "bad" pre-50 "really" are...

My board #18 is back with Enterpoint, waiting to receive my replacement

Doff

sr. member

Activity: 327

Merit: 250

Quote from: Lethos on August 01, 2012, 02:02:25 PM

Quote from: spiccioli on August 01, 2012, 12:50:09 PM

I've built cgminer 2.6.1 for linux 32bit (works ok on my ubuntu 12.04 64bit) with OpenCL/ADL disabled and just icarus support enabled.

If you trust my build you can get it from:

http://p2pool.soon.it/cgminer/cgminer-2.6.1

Download it, issue a

Code:

chmod 755 cgminer-2.6.1

to make it executable and it's ready to go.

spiccioli

I didn't notice any major improvements over 2.5, other then the addition of litcoin mining. Which while interesting development, doesn't concern me much.
In your opinion is it worth upgrade 2.6, if you got 2.5 works fine.

2.6.0 will allow anything using the Glasswalker controller to work since it has the 56k baud rate built in, whereas 2.5 does not have that yet. Since we aren't using the Glasswalker bitstreams currently it doesn't do much over 2.5.

Doff

sr. member

Activity: 327

Merit: 250

I'm sure there is a tool out there, but I'm just using trial and error like most everyone else. I know zefir had explained to me that you can remove one of the wire's within the USB cables to make it so they wont draw power which is what the Powered USB fixes, but he would be able to explain in better detail how that's. done.

Lethos

sr. member

Activity: 476

Merit: 250

Keep it Simple. Every Bit Matters.

Quote from: spiccioli on August 01, 2012, 12:50:09 PM

I've built cgminer 2.6.1 for linux 32bit (works ok on my ubuntu 12.04 64bit) with OpenCL/ADL disabled and just icarus support enabled.

If you trust my build you can get it from:

http://p2pool.soon.it/cgminer/cgminer-2.6.1

Download it, issue a

Code:

chmod 755 cgminer-2.6.1

to make it executable and it's ready to go.

spiccioli

I didn't notice any major improvements over 2.5, other then the addition of litcoin mining. Which while interesting development, doesn't concern me much.
In your opinion is it worth upgrade 2.6, if you got 2.5 works fine.

Isokivi

hero member

Activity: 910

Merit: 1000

Items flashing here available at btctrinkets.com

Copypasta from the bounty thread:

I would once again like to urge more people who own cm1's to contribute to this bounty, you know there are a handfull of brillian individuals workin very hard at the moment to provide us a more efficient tool for earning. Most of you are allready benefitting from this. Im not saying it's only because of this bounty.. but Im fairly sure the pace things have been progressing for the last week or so has much to do with this bounty and as any well-informed virtual miner knows time is everything in this game.

https://bitcointalksearch.org/topic/bountypaid-out-a-bitstream-for-better-utilizing-the-cairnsmore1-157-2945btc-94317

Isokivi

hero member

Activity: 910

Merit: 1000

Items flashing here available at btctrinkets.com

Quote from: Doff on August 01, 2012, 01:46:54 PM

Quote from: Isokivi on August 01, 2012, 01:41:09 PM

Where Im at now:

Makomk's 190mhs bitstream on all boards, two fgpa pairs only working at half the expected rate. For some ppl identifying which pair is failing has been as easy as unplugging the usb and checking what led stays on. For me they all stay on, so in order to flash a functioning bitstream to all fgpa-pairs I'd have to do a fairly annoying amount of flashing and seeing what works. Does anyone have any toughts on how I could identify the failing cores ?

Are you using Linux or windows? Also I know this is a broken record but try different USB cables, Also try a good Powered USB hub. The powered USB hub allowed Makomk's bitstream to actually work on my board 62-177. I also use the powered USB to update the controller just in case its providing a less noisy update for some reason.

Windows 7 64-bit, unpowered hub (only one working for me!), usb3 port, cgminer 2.6.1. Sadly I have a fairly limited amount of suitable-looking usb-cables lying around and after buying 2 hubs and contributing to the bitstream bounty Im actually pushing my RoI further. Does anyone know of a method suitable for testing usb-cables ?

Doff

sr. member

Activity: 327

Merit: 250

Quote from: Isokivi on August 01, 2012, 01:41:09 PM

Where Im at now:

Makomk's 190mhs bitstream on all boards, two fgpa pairs only working at half the expected rate. For some ppl identifying which pair is failing has been as easy as unplugging the usb and checking what led stays on. For me they all stay on, so in order to flash a functioning bitstream to all fgpa-pairs I'd have to do a fairly annoying amount of flashing and seeing what works. Does anyone have any toughts on how I could identify the failing cores ?

Are you using Linux or windows? Also I know this is a broken record but try different USB cables, Also try a good Powered USB hub. The powered USB hub allowed Makomk's bitstream to actually work on my board 62-177. I also use the powered USB to update the controller just in case its providing a less noisy update for some reason.

Isokivi

hero member

Activity: 910

Merit: 1000

Items flashing here available at btctrinkets.com

Where Im at now:

Makomk's 190mhs bitstream on all boards, two fgpa pairs only working at half the expected rate. For some ppl identifying which pair is failing has been as easy as unplugging the usb and checking what led stays on. For me they all stay on, so in order to flash a functioning bitstream to all fgpa-pairs I'd have to do a fairly annoying amount of flashing and seeing what works. Does anyone have any toughts on how I could identify the failing cores ?

Doff

sr. member

Activity: 327

Merit: 250

I got board 62-0013 hashing at about 120mh at 1.9U with makomk 160.bit, nothing else works on it including the original shipping test bitstream.

I should also say it only works on one pair.

norulezapply

hero member

Activity: 481

Merit: 502

Quote from: ShadesOfMarble on August 01, 2012, 12:53:08 PM

Quote from: ebereon on July 31, 2012, 10:58:12 PM

my early board SN#0015 is not working at all

Are you sure this board isn't broken? I can run 150 Mhz just fine on my S/N 26 board.
Where are the other pre-S/N-50 board users? I would love to hear some more experiences so maybe we can get a better image on how "bad" pre-50 "really" are...

Mine are being replaced as we speak. Couldn't run twin_test on them..

ebereon

sr. member

Activity: 397

Merit: 500

Quote from: ShadesOfMarble on August 01, 2012, 12:53:08 PM

Quote from: ebereon on July 31, 2012, 10:58:12 PM

my early board SN#0015 is not working at all

Are you sure this board isn't broken? I can run 150 Mhz just fine on my S/N 26 board.
Where are the other pre-S/N-50 board users? I would love to hear some more experiences so maybe we can get a better image on how "bad" pre-50 "really" are...

It's waiting for pickup from enterpoint. I got it already replaced. So yes it seems broken, but i can flash forth and back and switching what i want and led also working, but it's not hashing. Only shipping bitstream i think was working on 2 fpga's.

ShadesOfMarble

donator

Activity: 543

Merit: 500

Quote from: ebereon on July 31, 2012, 10:58:12 PM

my early board SN#0015 is not working at all

Are you sure this board isn't broken? I can run 150 Mhz just fine on my S/N 26 board.
Where are the other pre-S/N-50 board users? I would love to hear some more experiences so maybe we can get a better image on how "bad" pre-50 "really" are...

Topic: Cairnsmore1 - Quad XC6SLX150 Board - page 43. (Read 286370 times)