HashFast BabyJet users thread

[HF-Engineer]

Several things:

1. I see this a lot: "it keeps rebooting".   If the RPI reboots, it will show a "rainbow" image on the monitor (if you have one connected), and if you check the status page on your web interface, you'll see the "System Uptime" value reflect this reboot.  You should not be seeing the RPI reboot often, if at all.  If you are, it indicates a problem possibly with your SD card, it's filesystem, or possibly the power supply for the RPI.  For instance, plugging in some USB periperals are known to overload the RPI.  We recommend to only keep the BJ plugged in when mining.  Some people have installed WiFi adapters, and some of these can draw more power peak than the RPI can output, resulting in a reboot.  Now, cgminer restarting is somewhat normal.  We've set it up so that in the event of any problems it restarts cgminer.  Until all the bugs get worked out, this is so you at least keep mining.  You'll see this reflected in the "miner uptime" on the status screen.  Mine usually lasts a day or 2.  Sometimes when cgminer restarts it can take up to a minute to get going again.  This seems to be due to a USB locking problem, which we are aware of and are trying to fix.  If you unplug your USB cable from the RPI and plug it back in, this delay is usually reduced, but either way, it will start mining again if left to it's own devices.  If you believe it really is your RPI rebooting, then maybe it's best to backup your settings and reload the SD card.  It's also a good idea to go get a spare SD card so you have on on hand, and this way you can test a "virgin" image if problems develop.  Please be clear when you post here in the future as to the difference between RPI system reboots and cgminer restarts.  This will help me to help you if I better understand exactly what's happening on your system. 

2. Errors and "work watchdog resets" are not an indication your system is broken.  Most of these are due to bugs which are being worked on and will be fixed in upcoming software/firmware releases.  Please be patient!   As long as your average hashrate as reported by the pool is around 400Gh/s, you are ok.  The point of the watchdog is to ensure, regardless of glitches, you keep mining.  If your average pool reported hashrate is below 400Gh/s, then there may actually be a problem.  Post it!  Be clear about the symptoms, and post a log in verbose mode if possible.  All the engineers here want to see them!

3. If your die temps vary from each other more than about 4 degrees C, there is likely something up with your cooler.  Note that you must be running the custom hf version of cgminer to see all 4 die temps at once.  The standard version only shows the max per board in the status.  The temps seem to prefer being in the 70-80C range.  Don't think cooler is better, because it usually isn't!  (Though this varies somewhat from die to die)  Get a tube of hi-quality (grey) thermal compound, remove your cooler head, re-apply the compound (read online how to do this, but a little goes a long way!) and re-install the cooler head.  There should be washers under each screw, and the screws should be tightened until the stop.  Do not force them beyond snug!

4. Unfortunately we cannot support you if you are running Windows or some other miner program.  We only officially support you mining with the RPI, but if you are running our version of cgminer (right now it's 3.9.0h2) on a Linux box, I'll still try to help you.  If you are running anything else, don't bother asking, as there are plenty of known issues that we are aware of that can cause problems.  If you want to make the most BTC and not have your miner dead when you check it, you should be running what we recommend.  If you are looking at this post and it's more than a week old, this information may be out of date.  Read newer posts!

5. Tech Support.  I plan to remain active here as much as I can.  I'll help if I can.  But we are also improving tech support.  We have hired new people, developed new procedures, and engineering is monitoring TS and providing guidance.  If you didn't get good help, try again!  Please be patient and CIVIL.  Everyone wants to help, make easier for them not harder.

6. Warranty; I am not authorized to give you any information on what voids it or how long it lasts, etc.  But if you have a broken system, most likely we can get you going again.  Don't panic, be patient!

-Phil
 

[Micky25]

^ you're doing a great job here, thanks again. What a difference to the previous chapters of this story...

[Shaxz]

Hi Phil,

Thanks for all of your help.  Do you happen to have the Samtec part no for that 2.0mm pitch 16 pin IDC ribbon cable for daisy chaining boards together?

Thanks in advance

[legendroute]

Hi Everybody who has a BJ running

I have got my BJ(400GB) up and running for about 1 day, I am using Eligius as the pool. I can mine about 0.08BTC a day(Which I think it will take very long to get my investment back). Is this a good profit? What is your profit a day? Is there any other more profitable pool you  recommend? Thank you very much!

[-ck]

I've finally posted an official version that is designed to make the most of the new firmware coming up, cgminer 3.12.1

Announce and details here:
https://bitcointalk.org/index.php?topic=28402.msg4989440#msg4989440

This should also fix the windows reliability issues for those poor souls mining with their BlowJobs on windows.

[HF-Engineer]

Hi Phil,

Thanks for all of your help.  Do you happen to have the Samtec part no for that 2.0mm pitch 16 pin IDC ribbon cable for daisy chaining boards together?

Thanks in advance

Here's the info:

Product family:
http://www.samtec.com/technical-specifications/default.aspx?SeriesMaster=TCSD
Part number decoder: http://www.samtec.com/documents/webfiles/pdf/TCSD.PDF

Spec: TCSD-08-D-03.00-01-N-SR
BabyJet chaining cable: Three-inch long ribbon cables
2.00 mm pitch 2x8, double ended, notch polarized, with strain relief.

-Phil

[HF-Engineer]

Hi Everybody who has a BJ running

I have got my BJ(400GB) up and running for about 1 day, I am using Eligius as the pool. I can mine about 0.08BTC a day(Which I think it will take very long to get my investment back). Is this a good profit? What is your profit a day? Is there any other more profitable pool you  recommend? Thank you very much!

I personally use Eligius.  Mine is showing a 3-hour average hash rate of 408.35 Gh/s which is worth 0.07834091 BTC per day at the present difficulty, which amounts to about $56 and change per day at the present (declining) exchange rate on coinbase.

-Phil

[tacotime]

Still no per die voltage control?

[perezoso]

This should also fix the windows reliability issues for those poor souls mining with their BlowJobs on windows.

Hey now.  I never thought I'd be the Windows Defender(tm), but in the interest of fairness, the hashrate graph from Eligius for my Blowjobs is hard solid as a rock.  Really.  Cranking along at about 430gh each with no problems to speak of for what, almost 10 days now?  I can't complain a single bit about that aspect of the BlowJob experience.

[-ck]

This should also fix the windows reliability issues for those poor souls mining with their BlowJobs on windows.

Hey now.  I never thought I'd be the Windows Defender(tm), but in the interest of fairness, the hashrate graph from Eligius for my Blowjobs is hard solid as a rock.  Really.  Cranking along at about 430gh each with no problems to speak of for what, almost 10 days now?  I can't complain a single bit about that aspect of the BlowJob experience.

I'm pleased to hear that. There are two camps with the windows people: everything's hunky dory and everything's funky shit - and that group is more common with 3.12.0. I'm hoping to get more of the former with the latest release.

[Micky25]

only good exp. with windows and 3.12 since days and 3.12.1 since today as well (way better than with the RPi). Many thanks to you too.

[sillywhim]

Scrap the NSA-spy device aka HashFast custom minepeon raspi.

Windows => zadig => cgminer => Always Up => BJ'ing All Nite Long

[dbbit]

only good exp. with windows and 3.12 since days and 3.12.1 since today as well (way better than with the RPi). Many thanks to you too.

I can't say whether it's actually working better. It's different. Instead of getting the familiar:


 [2014-02-04 09:45:09] HFA 0 NOTICE: ######################### WARNING: Work Watchdog Reboot Imminent!
 [2014-02-04 09:45:11] HFA 0 HFGetHeader usb read err:(-99) LIBUSB_ERROR_OTHER
 [2014-02-04 09:45:11] HFA 0 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND
 [2014-02-04 09:45:11] HFA 0: device disappeared, disabling
 [2014-02-04 09:45:11] HFA 0 failure, disabling!
 [2014-02-04 09:45:11] HFA 0: hfa_send_frame: USB Send error, ret -4 amount 0 vs. tx_length 8
 [2014-02-04 09:45:17] Hotplug: Hashfast added HFA 1



Which we've all come to know and love, I now get pages and pages of the following spew on reset:

attempted reset got err:(0) LIBUSB_SUCCESS
HFGetHeader usb read err:(-1) LIBUSB_ERROR_IO
attempted reset got err:(0) LIBUSB_SUCCESS
HFGetHeader usb read err:(-1) LIBUSB_ERROR_IO
attempted reset got err:(0) LIBUSB_SUCCESS
HFGetHeader usb read err:(-1) LIBUSB_ERROR_IO
attempted reset got err:(0) LIBUSB_SUCCESS
...



I don't think it resets any less frequently, and because of the long spew, cgwatcher takes a little longer to reset it in fact.

There's probably no way around the firmware watchdog, it's just the cgminer now behaves differently

[Morblias]

Here is a new one... lol. Note: I only have 1 baby jet, not 2.



Seems like somehow it got disconnected or not seen by cgminer, then hot-plugged it to add it back, but never removed the first one that shows OFF now.

[storm2k5]

I got my babyjet yesterday, and I'm having problems with it. It came very battered, screws falling out, front door broken.
Initially I was pleased, cause cgminer said it reaches ~490ghps.

But my pool only shows ~250ghps for the last 20 hours.

I believe I might have a damaged core or two, cause the temp section in cg miner reads like this:

Code:

HFA 1: 77C/0.80V 73C/0.79V 66C/0.79V 51C/.82V

So the last 1-2 cores seem pretty cold compared to the first two which indicates a problem to me.

I already checked screws on cooler -> tightened them -> same results.
I reapplied thermal paste -> same results.

In cgminer log I'm getting no errors at all.

Anybody have an idea what could be wrong with it and what steps I could try?

[rMuD]

This should also fix the windows reliability issues for those poor souls mining with their BlowJobs on windows.

Hey now.  I never thought I'd be the Windows Defender(tm), but in the interest of fairness, the hashrate graph from Eligius for my Blowjobs is hard solid as a rock.  Really.  Cranking along at about 430gh each with no problems to speak of for what, almost 10 days now?  I can't complain a single bit about that aspect of the BlowJob experience.

I'm pleased to hear that. There are two camps with the windows people: everything's hunky dory and everything's funky shit - and that group is more common with 3.12.0. I'm hoping to get more of the former with the latest release.

Now if we can get Phil to open up his cgminer mods so it can be merged back in to the root build, and the non-rpi miners can get the benefits. 

[CurcO]

I got my babyjet yesterday, and I'm having problems with it. It came very battered, screws falling out, front door broken.
Initially I was pleased, cause cgminer said it reaches ~490ghps.

But my pool only shows ~250ghps for the last 20 hours.

I believe I might have a damaged core or two, cause the temp section in cg miner reads like this:

Code:

HFA 1: 77C/0.80V 73C/0.79V 66C/0.79V 51C/.82V

So the last 1-2 cores seem pretty cold compared to the first two which indicates a problem to me.

I already checked screws on cooler -> tightened them -> same results.
I reapplied thermal paste -> same results.

In cgminer log I'm getting no errors at all.

Anybody have an idea what could be wrong with it and what steps I could try?

You either need to change clock speeds, either OC or underclock use in cgminer "--hfa-hash-clock 600", I see from ex that the die's like to run over the 70-80 C , from your post you have one core in in the 51 C/.82V which tells me that core that is not working also you can see that the V is  high as to the nominal output in V is .79..

--hfa-hash-clock <arg> Set hashfast clock speed (default: 550)

[storm2k5]

You either need to change clock speeds, either OC or underclock use in cgminer "--hfa-hash-clock 600", I see from ex that the die's like to run over the 70-80 C , from your post you have one core in in the 51 C/.82V which tells me that core that is not working also you can see that the V is  high as to the nominal output in V is .79..

--hfa-hash-clock <arg> Set hashfast clock speed (default: 550)

CurcO thanks for your answer.

I found that
-using HF image on RPI with cgminer 3.09 -> cgminer says 490ghps
-using cgminer 3.12 on windows -> cgminer says 250ghps
-pool says 250ghps

I tried over/underclocking it in the range from 500-600mhz, underclocking makes it slower, overclocking makes it faster (max average I saw was 300ghps - far away from the 400+ other are getting)

Could it be an issue with PSU / connectors?

[CurcO]

You either need to change clock speeds, either OC or underclock use in cgminer "--hfa-hash-clock 600", I see from ex that the die's like to run over the 70-80 C , from your post you have one core in in the 51 C/.82V which tells me that core that is not working also you can see that the V is  high as to the nominal output in V is .79..

--hfa-hash-clock <arg> Set hashfast clock speed (default: 550)

CurcO thanks for your answer.

I found that
-using HF image on RPI with cgminer 3.09 -> cgminer says 490ghps
-using cgminer 3.12 on windows -> cgminer says 250ghps
-pool says 250ghps

I tried over/underclocking it in the range from 500-600mhz, underclocking makes it slower, overclocking makes it faster (max average I saw was 300ghps - far away from the 400+ other are getting)

Could it be an issue with PSU / connectors?

What Pool ? , I found that on P2pools the like coinex.pw it preforms at that range, gives a lot of errors from Pool submit errors to watchdog errors, so I have to stick to a BTC pool like , slush, Eligius , or similar. Don't know why the miner preforms slower , hmm  Phil might know why ( aka the HF Engineer that has been an excellent source for help).

[storm2k5]

CurcO, thanks again

I'm on bitminter.com.
I think the fact that 3.12 on windows shows the same value the pool shows means that the value with HF cgminer 3.09 on the rpi is probably wrong?

I also would like HF Engineer to comment on this