HashFast Sierra's Owners Thread

[-ck]

New version of cgminer, 4.1.0, out with dramatic improvements and new features for these devices:

https://bitcointalk.org/index.php?topic=28402.msg5581480#msg5581480

[1714025811]

1714025811

[1714025811]

1714025811

[1714025811]

1714025811

[bkminer]

I'm a Batch 3 Sierra customer and the email I received only said "we anticipate shipping by March 31st."  I talked to Chris on the phone and he suggested (no confirmation, he doesn't know for sure) that Ciarra might be producing in batches, and then re-tooling for other customers and then back to HashFast orders.  That would explain the erratic deliveries.

On the IceDrill thread it seems they've received 169THs(?) out of 253TH/s...their order was smaller than originally desired because they didn't raise enough BTC for the full 500TH/s... check the thread to be sure.

[kevin911]

Their anticipating doing a lot by March 31st, batch 2 sierra's still haven't been built/shipped , they sent me an e-mail saying that they targeted to ship by end of March - beginning of April. + batch 3 sierras & babyjets, batch 4 by end of April beginning of may then the new evo units. somewhere inbetween all that is the MPP units. they got a seriously big rabbit to pull out of this hat trick

[bkminer]

I'm still looking for where they said everything through Batch 4 will be shipped by March 31st... maybe Facebook

[winit]

Having an issue where windows closes Cgminer due to USB error

HFHash USB write err:<-7> LIBUSB_Error_Timeout
attempted reset got err:<0> LIBUSB_SUCCESS
hfa_send_frame: USB Send error, ret 0 amount 0  vs. tx_length 64, retrying

CGminer then crashes and has to close.

Also have one Sierra that doesnt power at all, no LED lights and not recognised when USb connected to windows.

Any help appreciated.

[winit]

Can someone treat me like an idiot and tell me exactly where I enter --hfa-hash-clock 600?

Im using CGminer 4.1

I've managed to get 4 Sierra's hashing from one PC, adding more Sierra's just causes it to crash. Is this normal?

[jddebug]

Can someone treat me like an idiot and tell me exactly where I enter --hfa-hash-clock 600?

Im using CGminer 4.1

I've managed to get 4 Sierra's hashing from one PC, adding more Sierra's just causes it to crash. Is this normal?

What OS? How are you currently starting cgminer?

If from command line linux: ./cgminer --hfa-hash-clock 600

If using config file: "hfa-hash-clock" : "600",

[winit]

Hi JD, thanks for the reply.

I downloaded CGminer 4.1 to windows 7 - inside the zip is CGminer.exe which I run from a batch file along the lines of

@echo on
cgminer.exe -o pooldetails:3333 -u myusername -p mypassword

I've tried adding in the '--hfa-hash-clock 600' on the above string and it causes CGminer to load and immediately quit.

Obviously I'm a moron for not knowing this stuff so really do appreciate the help.

[jddebug]

Hi JD, thanks for the reply.

I downloaded CGminer 4.1 to windows 7 - inside the zip is CGminer.exe which I run from a batch file along the lines of

@echo on
cgminer.exe -o pooldetails:3333 -u myusername -p mypassword

I've tried adding in the '--hfa-hash-clock 600' on the above string and it causes CGminer to load and immediately quit.

Obviously I'm a moron for not knowing this stuff so really do appreciate the help.

cgminer.exe -o pooldetails:3333 -u myusername -p mypassword --hfa-hash-clock 600

[winit]

Thanks, just crashes CGminer when I do that.

[minternj]

Dont run it from a batch file so you can see why its closing.

[winit]

So after a little over 24hrs I've learned a little and been angry a lot.

Of the 9 Sierra's 2 are useless. 1 fails to power up although did power for a short time today, on connecting it to CGminer it immediately gave an error along the lines of 'Main Board 12V error' and crashed CGminer.

The other problem Sierra was running fine but now does nothing but crash CGminer with Libusb errors.

I've had a few hours of easy running with the other 7 machines but now im having to restart every 20 mins or so.

Whilst the .4 firmware may solve some of this, there appears to be no support for windows users with the instructions only talking about Unix or babyjet users flashing their cards.

[SolarSilver]

So after a little over 24hrs I've learned a little and been angry a lot.

Of the 9 Sierra's 2 are useless. 1 fails to power up although did power for a short time today, on connecting it to CGminer it immediately gave an error along the lines of 'Main Board 12V error' and crashed CGminer.

Those quality numbers seem to reflect what other customers have been suffering

Out of the 4 BBJets I got, 1 was a dud. And Icedrill had about 25% of their 250 TH arrive not working, so they needed to RMA 60 units.

[winit]

Losing the will to live here. Having gone through the ballache of installing Ubuntu in order to update the firmware, in trying to update all I get back is this

('confirm is ', False)
('FIRMWARE_DIR is ', '.')
UC_HFU_FILE at './uc3.cropped.hfu'.
READSERIAL found at 'x86_64/readserial'.
HFUPDATE found at 'x86_64/hfupdate'.
ENTERLOADER found at 'x86_64/enterloader'.

HashFast Firmware Updater

Reading serial number from module.
x86_64/readserial: symbol lookup error: x86_64/readserial: undefined symbol: libusb_strerror
Entering Boot Loader...
x86_64/enterloader: symbol lookup error: x86_64/enterloader: undefined symbol: libusb_strerror
Traceback (most recent call last):
  File "./field_firmware_update.py", line 184, in <module>
    firmware_updater()
  File "./field_firmware_update.py", line 143, in firmware_updater
    enterloader()
  File "./field_firmware_update.py", line 62, in enterloader
    subprocess.check_call([ENTERLOADER])
  File "/usr/lib/python2.7/subprocess.py", line 511, in check_call
    raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '['x86_64/enterloader']' returned non-zero exit status 127
Command '['x86_64/enterloader']' returned non-zero exit status 127
Firmware Update had an error.  Please retry or report to HashFast Support.

Same response for a machine that works and the machine that has the LIBUSB error.

So now what?

[-ck]

Losing the will to live here. Having gone through the ballache of installing Ubuntu in order to update the firmware, in trying to update all I get back is this

Code:

Reading serial number from module.
x86_64/readserial: symbol lookup error: x86_64/readserial: undefined symbol: libusb_strerror

Same response for a machine that works and the machine that has the LIBUSB error.

So now what?

Looks like you're on an older ubuntu. For whatever reason, hashfast's loader is not that portable and looks like it needs a recent ubuntu, probably 13.10

[winit]

Hey Thanks CK... I'll see about updating my Ubuntu tomorrow and see if I have any joy there.

Do you think the firmware will resolve the machine with the LIBSUB error?

[-ck]

Hey Thanks CK... I'll see about updating my Ubuntu tomorrow and see if I have any joy there.

Do you think the firmware will resolve the machine with the LIBSUB error?

Doubt it. From what I've seen of the devices out there, some have unfixable problems and need to be RMAd.

[aasl]

shouldn't hf test the devices before shipping? the fault rate seems so high! and I wonder how long the RMA takes?

[HashFast_CL]

shouldn't hf test the devices before shipping? the fault rate seems so high! and I wonder how long the RMA takes?

We do test the devices before shipping.  But vendor implementation of the USB spec varies.  If your laptop doesn't work, try your desktop.

A high quality PCIE/PCI USB adapter often helps.

If the boards are having trouble running in a chain, try running one or two independently and isolating the problematic board on its own USB and cgminer instance.

Always use the latest firmware (0.4) and cgminer (4.1.0).

[winit]

Well so far 6 emails across 48 hrs to HF and not one reply so RMA is looking like taking a while.

Phoning tech support just gets a 'We're not able to take your call right now'