Cairnsmore1 - Quad XC6SLX150 Board

[dirtycat]

rampone/kano, thanks for the pointers above, im now mining away in icarus mode;

Code:

cgminer.exe -o http://xxxxxxxxxxxxxxxx:xxxx -u xxxxxxxxx -p xxxxxxxxx --disable-gpu -S "\\.\COM22" -S "\\.\COM23" --icarus-timing=short

Code:

 cgminer version 2.4.2 - Started: [2012-06-13 20:21:48]
--------------------------------------------------------------------------------
 (5s):968.3 (avg):689.6 Mh/s | Q:16  A:9  R:0  HW:0  E:56%  U:1.7/m
 TQ: 2  ST: 3  SS: 0  DW: 2  NB: 2  LW: 62  GF: 0  RF: 0
 Connected to http://xxxxxxxxxxxxxxxx:xxxx with LP as user xxxxxxxxx
 Block: 00000a82d52e60e136003fbe9fa2fc95...  Started: [20:26:46]
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 ICA 0:                | 354.7/349.5Mh/s | A:3 R:0 HW:0 U:0.58/m
 ICA 1:                | 304.8/340.0Mh/s | A:6 R:0 HW:0 U:1.15/m
--------------------------------------------------------------------------------

awesome!

[1713565729]

1713565729

[1713565729]

1713565729

[1713565729]

1713565729

[1713565729]

1713565729

[daemonic]

not quite as awesome as before, but it could be down the to icarus-timing stuff. After a 2 hour run so far, its only submitting about 4.5% the number of shares as my 6970 running at 440MH/s, didnt expect miracles this early on though

Code:

 cgminer version 2.4.2 - Started: [2012-06-13 20:33:04]
--------------------------------------------------------------------------------
 (5s):181.8 (avg):181.6 Mh/s | Q:232  A:68  R:0  HW:0  E:29%  U:0.6/m
 TQ: 2  ST: 3  SS: 0  DW: 33  NB: 10  LW: 487  GF: 1  RF: 0
 Connected to http://xxxxxxxxxxxxxxxx:xxxx with LP as user xxxxxxxxx
 Block: 0000055f327705a5b8ed12192763a641...  Started: [22:24:54]
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 ICA 0:                |  94.0/101.6Mh/s | A:63 R:0 HW:0 U: 0.51/m
 ICA 1:                |  90.7/ 80.0Mh/s | A: 5 R:0 HW:0 U: 0.04/m
--------------------------------------------------------------------------------

[norulezapply]

Currently hashing away using 2 FPGAs after installing the drivers Excellent work Enterpoint!

Time to get it working correctly with my RaspberryPi..

[ebereon]

ebereon,

I have Vista SP2 on this laptop atm, and have done the same as yourself;

1) i installed the latest catalyst drivers from http://support.amd.com/us/gpudownload/Pages/index.aspx (Desktop Graphics -> Radeon HD Series -> Radeon HD 6xxx Series -> Win Vista 32 Bit), this was purely to satisfy the missing OpenCL.dll error on cgminer start.

2) Connected usb of cairnsmore1 and installed the drivers for Port A/B, SPI, JTAG, COM20-23.

3) Opened a command prompt as Admin, (so as to make sure there wouldnt be any com port access issues)

4) Navigated and ran cgminer as below;

Code:

cgminer.exe -o http://xxxxxxxxxxxxxxxx:xxxx -u xxxxxxxxx -p xxxxxxxxx --disable-gpu -S "\\.\COM22" -S "\\.\COM23" --icarus-timing=short

As you have experienced, the 2 yellow led's closest to the usb/power turn off, but for me cgminer then starts chugging along. the blue led seems to flash when a share is found.

thank you for this explanation but it don't work.
I tested it on win7 x64, win7 32, winXP.

The cgminer's output is:

Code:

[2012-06-13 22:02:56] Started cgminer 2.4.2
[2012-06-13 22:02:56] Loaded configuration file cgminer.conf
[2012-06-13 22:02:56] Icarus Detect: Test failed at \\.\COM20: get 00000000, should: 000187a2
[2012-06-13 22:02:56] Icarus Detect: Test failed at \\.\COM21: get 00000000, should: 000187a2
[2012-06-13 22:02:56] Icarus Detect: Test failed at \\.\COM22: get 00000000, should: 000187a2
[2012-06-13 22:02:56] Icarus Detect: Test failed at \\.\COM23: get 00000000, should: 000187a2

And then cgminer exits with "no device" message.

My board shows a orange led on every fpga and a blink red led on the spartan 3.

can someone please post the switch configuration? i don't know if thats the problem?

[Keninishna]

Got my board today, and plugged it in and installed drivers. Now mining away in cgminer at about 200 mh/s can't wait to get the tricone bitstream on this badboy.

[dirtycat]

Got my board today, and plugged it in and installed drivers. Now mining away in cgminer at about 200 mh/s can't wait to get the tricone bitstream on this badboy.

tricone is nice but I wouldnt want to donate 5% of my hashing power I want it all.. cant wait for enterpoint to get them running at full power.

[Gomeler]

Just got home, 5 minutes fiddling with installing the drivers and it's up and running with cgminer 2.4.2 in Windows 7 x64. I'm quite happy with board "G2-0020" and it's blinking LEDs. Nice work guys!

[Doff]

Has anyone been able to get it going on Linux yet? I think I'm having an issue with the Driver on debian and haven't quite figured it out yet.

Maybe I should be more specific.

dmesg | tail gives me

545.907681] usb 3-6: new high-speed USB device number 6 using ehci_hcd
[  546.043623] usb 3-6: New USB device found, idVendor=0403, idProduct=8350
[  546.043627] usb 3-6: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[  546.043631] usb 3-6: Product: Cairnsmore1
[  546.043633] usb 3-6: Manufacturer: FTDI
[  546.043635] usb 3-6: SerialNumber: FTVIDJBE
[  702.225149] usb 3-6: USB disconnect, device number 6

I also added the file to /lib/udev and restarted udev, however when I go to look for ttyUSB there are none listed. So somewhere I have something wrong obviously.

I also tried loading the libftdi1 libs from the Debian repository with no luck thinking I was missing a driver file or something.

If anyone can help point me in the right direction that would be cool.

[kano]

Has anyone been able to get it going on Linux yet? I think I'm having an issue with the Driver on debian and haven't quite figured it out yet.

Maybe I should be more specific.

dmesg | tail gives me

545.907681] usb 3-6: new high-speed USB device number 6 using ehci_hcd
[  546.043623] usb 3-6: New USB device found, idVendor=0403, idProduct=8350
[  546.043627] usb 3-6: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[  546.043631] usb 3-6: Product: Cairnsmore1
[  546.043633] usb 3-6: Manufacturer: FTDI
[  546.043635] usb 3-6: SerialNumber: FTVIDJBE
[  702.225149] usb 3-6: USB disconnect, device number 6

I also added the file to /lib/udev and restarted udev, however when I go to look for ttyUSB there are none listed. So somewhere I have something wrong obviously.

I also tried loading the libftdi1 libs from the Debian repository with no luck thinking I was missing a driver file or something.

If anyone can help point me in the right direction that would be cool.

Hmm - maybe try:
 sudo modprobe ftdi_sio vendor=0x0403 product=0x8350
and see if that creates the /dev/ttyUSBx
modprobe is required on kernel 2.6.38-8 (11.04)

[Doff]

Has anyone been able to get it going on Linux yet? I think I'm having an issue with the Driver on debian and haven't quite figured it out yet.

Maybe I should be more specific.

dmesg | tail gives me

545.907681] usb 3-6: new high-speed USB device number 6 using ehci_hcd
[  546.043623] usb 3-6: New USB device found, idVendor=0403, idProduct=8350
[  546.043627] usb 3-6: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[  546.043631] usb 3-6: Product: Cairnsmore1
[  546.043633] usb 3-6: Manufacturer: FTDI
[  546.043635] usb 3-6: SerialNumber: FTVIDJBE
[  702.225149] usb 3-6: USB disconnect, device number 6

I also added the file to /lib/udev and restarted udev, however when I go to look for ttyUSB there are none listed. So somewhere I have something wrong obviously.

I also tried loading the libftdi1 libs from the Debian repository with no luck thinking I was missing a driver file or something.

If anyone can help point me in the right direction that would be cool.

Hmm - maybe try:
 sudo modprobe ftdi_sio vendor=0x0403 product=0x8350
and see if that creates the /dev/ttyUSBx
modprobe is required on kernel 2.6.38-8 (11.04)

That worked, now I get a new error,  getting closer!

This is the new error.

Icarus Detect: Test failed at /dev/ttyUSB2: get 00000000,
 should: 000187a2

 For all of them

[yohan]

I'm not best person to advise on Linux but we have been runing Ubuntu for our testing if you want to try that as an alternative.

Hopefully in the CGminer support materials we have coming today or tomorrow you have 2 options either to run our slow bitstream on 4 FPGAs or a standard Icarus at full speed on 2 FPGAs. The latter will earn you more return right at the moment.

Yohan

[kano]

Has anyone been able to get it going on Linux yet? I think I'm having an issue with the Driver on debian and haven't quite figured it out yet.

Maybe I should be more specific.

dmesg | tail gives me

545.907681] usb 3-6: new high-speed USB device number 6 using ehci_hcd
[  546.043623] usb 3-6: New USB device found, idVendor=0403, idProduct=8350
[  546.043627] usb 3-6: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[  546.043631] usb 3-6: Product: Cairnsmore1
[  546.043633] usb 3-6: Manufacturer: FTDI
[  546.043635] usb 3-6: SerialNumber: FTVIDJBE
[  702.225149] usb 3-6: USB disconnect, device number 6

I also added the file to /lib/udev and restarted udev, however when I go to look for ttyUSB there are none listed. So somewhere I have something wrong obviously.

I also tried loading the libftdi1 libs from the Debian repository with no luck thinking I was missing a driver file or something.

If anyone can help point me in the right direction that would be cool.

Hmm - maybe try:
 sudo modprobe ftdi_sio vendor=0x0403 product=0x8350
and see if that creates the /dev/ttyUSBx
modprobe is required on kernel 2.6.38-8 (11.04)

That worked, now I get a new error,  getting closer!

This is the new error.

Icarus Detect: Test failed at /dev/ttyUSB2: get 00000000,
 should: 000187a2

 For all of them

There's a few causes I can think for that
1) If the device is too slow replying to the test it will return a 00000000
On a standard Icarus the test is ~0.5ms but it is allowed to take 100ms before it gets that error

2) Resetting the serial/usb:
Here's a python script that I've had to use in the past when transferring an Icarus from Windows to Linux
(after testing windows code)
 http://pastebin.com/N8nYz5cn

It may (or may not) help

If you save it as usbtest.py (and python is installed) then you just:
 ./usbtest.py /dev/ttyUSB0 icarus

Python seems to clear whatever problem I get on rare occasions.
(Ignore the 2nd numbers not matching)

3) Bitstream isn't correct
Ask one of the others who actually got some results how they got the bitstream in properly.

[Doff]

Well that seems to talk with them, makes the blue lights come on anyhow, I get this result below.

Push payload to icarus: 4863f356d8d66700101e21bc87749727e47f91598f5092bb36204da68f089f5d000000000000000 00000000080000000000000000c350b1a0bba554ffdf63dd3
Result:(should be: 063c5e01):
Push payload to icarus: ce92099c5a80bb81c52990d5c0924c625fd25a535640607d5a4bdf8174e2c8d5000000000000000 00000000080000000000000000b290c1a42313b4f21b5bcb8
Result:(should be: 8e0b31c5)

I think ill just be patient and wait for the Enterpoint cgminer update. I do appreciate the help however Kano, and its always fun to mess around with these things.

I'm still blown away at the incredible turn around Enterpoint had in getting these things out to us. I mean it shipped Monday and I got it Tuesday night not to mention its already 2 weeks early.

For Yohan, Debian and Ubuntu are pretty close as far as distributions go, so anything that works in Ubuntu usually works for Debian.

Thanks again for the help, I'm sure ill need more before I can get this running.

Doff

[rampone]

As we are all waiting for better bitstreams, the ATX supply "PowerBoard" Distibution XLS-table got released on the support site:

http://www.enterpoint.co.uk/cairnsmore/POWER_SUPPLIES.xls

Gonna be interesting, what the power distribution board does, could it be usage of the 5V?

So far so good, th board kept hashing the night over, only on 2 (think I read that somewehere before) fpgas and with 50 MHz, but 100 MH/s are better then nothing and it is hashing since 18 hours or sth...

I like...


Now just better Bitstream (maybe tell Eldentyrell as a quick temporary solution until enterpoint fiddled something amazing themselves?)

Good Work, keep it up

[MXRider]

Requesting MPBM front-end interface for Cairnsmore1:

https://bitcointalk.org/index.php?topic=62823.0

Please!

A free board for TheSeven? I doubt that is going to happen if cgminer works.

[Gomeler]

Requesting MPBM front-end interface for Cairnsmore1:

https://bitcointalk.org/index.php?topic=62823.0

Please!

A free board for TheSeven? I doubt that is going to happen if cgminer works.

Since when did cgminer become the default FPGA miner?

Assinine.

I think his point was Enterpoint has no incentive to sending TheSeven a board if the cgminer developers are working with Enterpoint(even if it isn't free). Perhaps down the road but it sounds like Enterpoint has hitched their wagon to the cgminer train for the time being.

Since MPBM is written in Python, it might be possible for an Enterpoint customer who knows Python to patch in support with minimal effort.

[spiccioli]

Here we go,

my unit arrived in Italy yesterday, I won't be able to try it until tomorrow evening, though... :/

It is really a beatiful piece of hardware and I will hook it up to a HP thin client t5570 http://www8.hp.com/it/it/products/thin-clients/product-detail.html?oid=4306189 on which I've installed ubuntu server 64bit and which draws just 8W when idle

We use them at work, so I had several second hand units available, otherwise they're pricey.

spiccioli

[yohan]

The main batch of heatsinks has arrived today so will be aiming now to raise our production levels.

[dirtycat]

Requesting MPBM front-end interface for Cairnsmore1:

https://bitcointalk.org/index.php?topic=62823.0

Please!

A free board for TheSeven? I doubt that is going to happen if cgminer works.

Since when did cgminer become the default FPGA miner?

Assinine.

cgminer is the best.

[norulezapply]

That's good news Yohan

Can anyone report the power consumption of the board mining in it's current state? (2 FPGAs approx 100Mh/s)

I'm trying to measure how efficient my cheap China power supply is (i'm guessing about 65%).

Mine is pulling 45 Watts at the wall.