Just wondering, what temperature are you guys seeing on your BJs? I'm @ 80C with stock settings (550 MHz): HFA 0: 80C/.79V 80C/.79V 73C/.79V 74C/.79V 452.8G/423.4Gh/s | A:3112200 R:8550 HW:5363 WU:5473.7/m
I assume that since there are 4 cores we get temp for each one? So my hottest core is 80C and my coolest 73C. In other news, my BJ still goes belly up after ~30 hrs of stable mining. I'm still on cgminer3.9.0h2 so I might try compiling the 3.11 version to see if that fixes my stability issues. |
|
|
|
Overclocking is the ONLY way you can even hit their promised performance!
If you clock to 610, cgminer will show it runnign at about 450 GH/s. But we know now this is not the REAL hashrate. Eligius will show it running at 425 GH/s. Why? Because cgminer is currently incorrectly counting bad shares (unless you upgrade to 3.11).
Upgrade cgminer to 3.11 and you'll see the hashrate drop to 425, the REAL HASHRATE.
Now, put the settings back to stock (clock 550), cgminer will show 420 GH/s but the real hashrate will be <395 GH/s. Not even the 400 GH/s we were promised.
Once again HashFast has over-promised and under-delivered.
btw, the BJs are not very overclockable. You probably won't be able to go higher than 615 clock without crashes every 30 seconds. I was hoping the BJs would be overclockable to at least 500 GH/s but now it looks like not even 450 GH/s is possible and overclocking is necessary just to hit 400 GH/s.
It is not really overclocking. HF themselves announced it will do 500 GH/s while still running underclocked:  HALF A TERAHASH/s (500 GH/s) on a single chip, while still running underclocked. In any case, 395.80 Gh/s is the average hashrate I'm seeing at the pool with default settings. |
|
|
|
BTW. The BJ is silent when I turn it on. There's a light on the front panel that come on, but nothing out of the fans... is this the way they behave? No fan until hashing?
Yes, the fan only turns on when the hashing starts. When I initially turned on my BJ without having properly configured the RPi the same thing happened (only the light on the BJ is on, no fans) |
|
|
|
I highly recommend using the latest cgminer 3.11x with these devices, quite a bit of work to the hf driver was done. Also you do not need to supply the --hfa-hash-clock if you are just going to run at 550 since thats default, a good number to try is 610, of course will use more power but should get you more hashing and not many if all hw errors..
Thanks. I turned on automatic HF updates again and after a reboot I see I now have a new version: cgminer3.9.0 h2 It is still not 3.11.x but I'll leave the 'h2' version running for a bit to see if that improves stability. I did notice it caused my hashrate to drop slightly from 422 to 417. I assume HF will push 3.11 automatically once it is ready for primetime. Will post future updates in the HF user thread as not to spam this one. |
|
|
|
My BJ hashes @ 422 GH/s consuming 407W at the wall using stock settings:  #!/bin/bash
sleep 10
/usr/bin/screen -dmS miner /opt/minepeon/bin/cgminer -c /opt/minepeon/etc/miner.conf --hfa-hash-clock 550 Even at stock it is not very stable as it has gone belly up 2 times over the last 3 days. It requires a manual reboot to get it running again. Minepeon graphs also go crazy when this happens:  |
|
|
|
My main concern is that the BITMAIN issue has to be clarified. Their products look exactly like ASICMINER's, and the specs are perfectly Gen2.
Good point. There is certainly an uncanny resemblance in the products (blades, sticks, cubes) and the timing & specs fit what we would have expected the Gen2 BE chips. Same sales approach (group-buys on bitcointalk, no pre-orders, etc). Even the announcement post with the blue headings and writing style is similar to FC. Maybe I'm hallucinating. The failed AM Gen2 chips must have been really abysmal for ASICMINER to decide to go for half a year without any product at all. |
|
|
|
I was considering taking a position in IXC, however the coin generation properties worries me: Also it is not exactly like bitcoin as it makes 96 coins per block and does not lower that over time but instead stops boom directly down to zero at some specific block number. there are currently ~13M iXcoins (as of June 2013) in existence and it is expected that all 21 million iXcoins will have been generated by 2015. What will happen to the network hashrate in 2015? There will be no reason to merge-mine it anymore and it won't be widely adopted enough for fees to provide an alternative... |
|
|
|
|
Oh, how the mighty have fallen! How did we get here and will we ever see those glory days again?
I'm still holding AM, mainly because I am too lazy to sell. But if Gen 3 fails, I fear all is lost.
FC mentioned that he does not like to work for others, what is preventing him from neglecting AM and letting it die a slow death while rebooting privately? His identity is still unknown so he can do whatever he pleases. Keeping my shares on faith alone is tough.
|
|
|
|
BTW, Ninjarobot, did you get that call about the MPP today as promised?
Nope. |
|
|
|
we’ve delivered on our promise Some balls... Did they hire Inaba? Yes, this is a bit arrogant. As if it was their only promise. The promise was 400 GHs shipping October 20-30. They conveniently left out the delivery promises and they make all the difference to HF customers. The burn-in 'gift' is also a bit of a joke consideringing ever reaching ROI is completely out of the window for any Batch 1 customer. After running my BJ on Eligius for an entire night it is reporting 399.23 Gh/s average hashrate. (Minepeon is reporting 425).  Have not checked power consumption yet. Overall it is still a clear case of overpromise & underdeliver with some arrogance sprinkled on top. |
|
|
|
One interesting thing I noticed in the minepeon graphs is that you can see the burn-in history. In my case the unit started hashing on Jan 17: https://i.imgur.com/sOzYWL0.png?1Actually this only proves the image they duplicated on the SD card was used on Jan the 17th, it does not need to be your machine (unless all machines come with a unique hashing account preset) They do come with their own hashing account. When you first log in to minepeon you will see the following disclaimer: For factory testing an initial Bitcoin wallet was generated on this device in order to perform a live test of your product. As a gift to you, we are providing you this wallet so you can claim any mining proceeds from the mining pool (Eligius) that may have accrued, but we do not imply that any have accured, or that the mining pool, which is beyond our control, will release any proceeds We DO NOT adivse using the pre-configured test wallet for mining production, as it's possible the keys could have been exposed during manufacturing, assembly, distribution or shipping. Please claim any proceeds as soon as possible in order to ensure they are not lost. The mining pool may not transfer funds into this wallet until a certain minimum payout amount is reached, or until after a certain time has elapsed. Please see the Eligius mining pool home page for more information.
We have provided this system pre-configured to check for any updates that may be needed to from time to time, and as such, this system will attempt to connect to setup.hashfast.com during startup. Your IP address, the factory generated public key, module serial numbers, and version numbers may be sent to facilitate this. If this is undesired, you can remove this functionality by clicking the "Default cgminer" button and then the "Save" button under the settings tab in your control panel. Note that once this is done, your system and your HashFast modules will not receive any further updates. You can re-enable the update system at any time by restoring the original settings; click the "Default HashFast cgminer" button, then the "Save" button, and then restart your system. During update requests, at no time is any sensitive information sent, such as private keys, any mine pool settings, passwords, etc.
NOTE: There is no password initially set on the web-based control panel, so we recommend setting one as soon as possible. The base Arch Linux system also has a defualt password, please see the MinePeon documentation for more information. We recommend having a properly qualified network engineer assess your network and systems for any security issues before using any systems for production mining. We also highgly recommend you make a backup of all settings (can be done fron the settings tab) so you can restore your setup in the event of SD card corruption or failure.
At startup the system will attempt to begin mining on the Eligius pool with the factory generated test wallet. Please change this as soon as possbile if you wish to keep any mining proceeeds generated by your system. This can be done on under the "pools" tab of the control panel. Note that by default there is a watchdog timer configured on this system that will reboot it in the event that mining hasn't started for 5 minutes, or within 1 minute after mining has failed. There is information about how this works available on the internet. But, yes - the account is associated with the SD card not the BJ. |
|
|
|
Mine has been running very stable, hopefully yours does the same.
Did HashFail ever call back about the MPP?
They have not called yet. One interesting thing I noticed in the minepeon graphs is that you can see the burn-in history. In my case the unit started hashing on Jan 17:  |
|
|
|
|
My BJ arrived today. Just picked it up from FedEx (never got the tracking number and missed delivery earlier today).
Unit is pretty nice and is mostly a plug&play affair. Took me less then 30m to install, set-up raspi with minepeon and configure pools.
By default it will mine on eligius at the burn in address associated with your account. The account will already contain any coins the unit might have mined during the HF burn in testing. In my case around 70 mBTC. It is running @ 422 GH/s
Will let it run overnight to see how stable it is.
|
|
|
|
At least thanks to ninjarobot we will know more about the MPP in the next 24 hours. (or maybe that call will never arrive).
Indeed, grill them my friend! (And record the convo if possible (for future litigation) if legal. I hope they follow through, but would not be surprised if they did not given their track record of over-promise & under-deliver. The only thing I'd like to know about MPP is when and in what quantity they will deliver after the 90 day MPP window ends. The ROI calculation only makes sense if you actually deliver immediately afterwards. I assume they have worked out most of the hardware and supply kinks by now and should be able to start shipping in volume including MPP. Difficulty is nearing 10x from what it would have been on Oct 20 when the BJ should have started shipping, so 4x with MPP will help a little if shipped early Feb, but in no way will it fix the damage already done by the delays. If they call, I'll ask if they are okay with the conversation being recorded. |
|
|
|
|
FYI:
Erin called me yesterday asking if I wanted to void the USD check and get reinstated in the shipping queue. I answered YES if you can ship before end of next week otherwise NO.
Today, I got a call from Rodrigo to let me know my order (1 BJ) has shipped. He will be providing the tracking number over email.
Like jspielberg I paid 60 BTC for a unit that will likely only produce 6 BTC (10%), so the timing of MPP is crucial to getting closer to 50% (best case).
I asked Rodrigo about MPP and he told me someone at HF will call me within 24h with more details.
|
|
|
|
- Shipments begin: October 20th-30th, in order of purchase
So much for shipping in order. Ah well...so many promises were broken, might as well break another one. |
|
|
|
Congrats with your BJ! To get a sense of where they (HF) are in the shipping queue, would you mind sharing your order#? http://hashfast.com/order-chain/Thanks & please keep us in the loop regarding stability, power use and overclockability  |
|
|
|
jjiimm_64 is just trying to get a full refund before the others. That's dirty, a fair share would be the best thing to wish for, as you're all in the same boat.
That would also probably more efficient.
He's just a prick who tries to steal money from you before it's too late, but I doubt he's been bribed by HF.
To be honest, if jjiimm_64 is just trying to get his original BTC payment back I would not call that 'stealing money'. Of course it would be nice if he could share relevant information with other customers so we don't have to fight each other. |
|
|
|
I am hoping they will do something for us before all the cases really start..
I also wish HF would be willing to meet their their Batch 1 customers in the middle and work something out. For example, if HF could guarantee to deliver the ordered units + 4 MPP boards before the end of January, I would accept it. Personally, I don't think this is an outrageous request as the original unit was supposed to ship in October of last year and the 90 day MPP window ends in 8 days. In other words, I'm willing to take a 50%+ loss and accept what I ordered even though the original unit is 3 months late. However, the ball has been in HashFast's court for a long time now and they are not playing. We know nothing apart from the occasional hostile ToS change, release form or unsolicited refund. The clock is ticking, and I can not wait forever. Is it too much to ask a company to work *with* their customers instead of against them? |
|
|
|
TL;DR: No matter what a company does, there will always be the whiners and criers about how it's done... that being said, it's best to do what's good for the continued existence of the company.
That is exactly the problem. HF is doing what is good for HF, not for their customers. They retroactively rewrite the rules of their own game to suit their needs. Offsetting risk and losses from the company to the customer. That might be acceptable if we were investors or 'backers'. But we are not. If you're going to have to deal with the disgruntled people no matter what, then keep the company a float so you can handle the problems instead of going out of business trying to satisfy the unsatisfiable.
You use heavily biased words like whiners, criers, unsatisfiable... But all people are asking is for HF to live up to their own promises. Nobody is asking for more than they put in. Customers will be happy to break even or take modest losses at this point. I would not call that unreasonable or unsatisfiable. I believe that's what might be happening with Hashfast right now - they are doing the best they can in a really crappy situation.
This situation is entirely of their own making. Selectively forcing USD refunds on their earliest BTC customers while not communicating is not what I call doing 'the best they can'. |
|
|
|
|
Show Posts
