I got my BabyJet on the 29th. I wasn't expecting the RPi or that the warranty only supports Linux. Somewhere during setup, I read that WiFi could not be used for the initial setup. I also read that not too many USB WiFi adapters work with RPi.
I went to Staples, found an ASUS N-53, plugged it into the RPi, and the BabyJet was mining, running cgminer 3.9, which was on the SD card, clocking 421 GH. Overnight, it updated cgminer or MinePeon, and was clocking about 380 GH, then, HF Engineer Phil pushed-out cgminer 3.11, and the BabyJet couldn't even get as far as cgminer. Phil rolled the changes back to cgminer 3.9, the the BabyJet was running, and clocking 421 GH, again.
I read in the setup instructions that a 4 GB SD card is required to get the latest version
When I setup my BabyJet, the first thing I did was copy all of the files to one of my PCs, and noted that the SD card was only 2 GB. After reading the HashFast warranty, there was no way I was going to put a non-HF SD card in the BabyJet, so I Emailed tech support, and asked for them to send me an SD card that met their own setup instructions, since I didn't want to void my warranty. It got shipped Thursday, and FedEx left a delivery notice on Friday. That night, the BabyJet tried to update, and it went off-line. When I put the SD card in my PC, it said that it needed to be formatted--it appears to have been wiped clean, and I have no idea which file system to use to reformat it, or if it needs to be bootable. So, my BabyJet has been idle since midnight Friday night. I don't know if the update was pushed-out, assuming I got my SD card or what.
FedEx says the SD card will be here by 5PM. I have been checking the lobby every 20 minutes since 10:00 AM, this morning. I've lost three good days of mining because of this, and the difficulty went up again, on Friday.
At present, the RPi no longer recognizes the USB WiFi adapter. Last week, it knew it was an ASUS device. Now, the RPi thinks it's a Ralink device. If I have to order an Ethernet WiFi adapter for this thing, it will be at least another day and more money that this situation has cost me.
Unbelievable.
--Patrick
UPDATE: The new SD card has arrived. When I installed it, MinePeon got as far as the Ethernet address and re-booted, before I could even get to setup.hashfast.com.
I'm done messing with these Raspberry Pies and a boot process that wipes-out my WiFi drivers every time it re-boots. I've lost four (4) days of mining time due to messing with the RPi. The 10-day warranty is up. Now, I'll do what I should have done from the very beginning...
Don't worry about temperature. While the silicon has a wide variation, in general they seem to be work best at between 70-85C. My experience has been that below 70C there are more errors, so you'll usually see more until the system warms up. Be sure that if you can set your difficulty, you set it for at least 256. That's the ideal setting for a single board. Dual boards can go to 512, and Sierras should be at about 1024. On some pools, such as Eligius, every few minutes or so the pool looks at your hashrate and automatically bumps you to the optimal.
We are working tracking down the cause of some of these issues. We are dealing with 4 different areas:
1. The Raspberry Pi USB subsystem (both hardware and driver code).
2. The cgminer software
3. The system control firmware (living in the on-board Microcontroller)
4. The Silicon itself.
We've identified some issues in all of these areas. For instance, the CRC errors seem to be an anomaly in the RPI's USB stack. Of course it's normal for silicon of this high-performance design to exhibit some errors, but it should be such a small percentage that it isn't statistically relevant. However, it looks like all 4 parts are conspiring together to compound the problem. We have identified some issues with the on-board firmware, so we will be pushing an update soon. I'll let you know when it goes out. If you are mining with a system other than our RPI distribution, then there will be some delays as we still need to finalize and test the update mechanism. Until then you can always get a RPI and load our image (see a few posts above) and let it run long enough to update. The update mechanism checks for updates every hour, and within 5 minutes of boot up, so you shouldn't need much longer than that to get an update once it's made live.
If this new version of cgminer (3.11.0hf1) seems worse, let me know. I've found it more reliable here in the lab, but I'd love to hear your experience. Like I said, for now, disregard it's reported hashrate and use what your pool is reporting. And yes, it's normal to see around 5% variation throughout a day. We should have new version to push soon, and if that doesn't improve things, I'll be happy to roll you back to the old one. All of these versions contain experimental code to try to reduce these problems, that's what the "HF" designation means.
Right now, my main goal is to ensure you keep hashing away. So anything we can put into place, even if it's only a temporary band-aid, we won't hesitate to do it. We don't want you to lose any revenue while we get to the bottom of this!
I've been an Engineer for a long time, and I've never worked on a project this difficult, so we appreciate you patience. There aren't too many chips out there the size of your fingernail that suck down over 100 amps a pop, and then we crammed 4 of them together!
BTW: We just hired 2 more people so we can improve support. Hopefully you will soon see how awesome it's getting around here!
-Phil
i
