Just to clarify for anyone else reading, the £25/month fee mentioned above only applies if you don't qualify for free Citigold by its eligibility criteria.  If you do, then FASTER and SEPA payments are free per http://www.citibank.co.uk/personal/banking/citigold/cgts.htm   Only saying this because so far I'm pretty happy with my experiences with my Citigold account, except for the recent announcement that they are closing their retail banking services in 11 countries including Japan.

How long have you been mining it for?  That's alot more BTC then I've gotten and it's certainly long enough that statistical fluctuations should have evened out.

I've only got one board, but it has run fairly stable since it was repaired, albeit with 15% HW error.  It still puts out a consistent ~20 GH/s at int:52:2:950.   My downtime is more caused by ISP outages causing CGMiner to freak-out than by anything else.   Mine was repaired in the UK by Milkbottlec.  I never ran it before sending it off for repair.   It's made 0.24 BTC since I started mining in early Jan.  Far away from break-even at today's BTC prices. :-)

I'll count myself lucky that I only got the one so aren't having the mutiple boards problems.  

It seems obvious to me that maybe the problems you all are having is the quality of the repair itself?   Have you considered having them resoldered or re-repaired again?   Is anyone at Drillbit responding and could maybe speak up about the design and/or parts that should be needed to fix the initial problem?  I'll admit I haven't read every message on the Drillbit Forums but I don't remember anyone but the 'repairers' being told what the parts were / what the issue was.

I *am* using 950mV.  I said so twice already. :-)

I tried int:53:2:850 earlier, the error rate was higher.

I'm using a large floor fan blowing over the whole board.  Pretty sure it is cooling the whole board pretty well.   Without a temperature sensor gun, is there a way to tell if components are getting hot and causing the HW errors?

I'm trying 53:2:950 right now actually.  I have fans on the heatsink side (the big silver one with no components) only right now.  I didn't think 35c was too hot.

What concerns me is the high HW rate though.  Over 10% seems more than most people are reporting.

I got my repaired board back on this past Sat and have finally been mining with it for the first time.   (I was away both in early Dec and Xmas, so shipping times screwed me around.  )

I've been using CGMiner 3.9.0, built myself on Ubuntu 13.04, and experimenting with settings.  I've noticed that the board really seems to take awhile to hit peak processing speeds.   I mean like 15 minutes before the 5s avg reported hash rate gets over 20GH/s.   Is this normal?  Is anyone else seeing it?

I'm also getting HW errors over 10% even when using a shielded alternative USB cable.   For example, my latest run (int:52:2:950) has been going on for 70 minutes and the CGMiner reports:

Do I just have a poorly hashing board or is it possibly related to the repairs?  Milkbottlec reported having to resolder things a second time to get an acceptable hash rate, so perhaps things still aren't quite 100% there?

Given the boards are shipping late, and I'l be away when you're saying my board should arrive (I leave this Wednesday), I'm rather disappointed.   Can you hold my express shipping package from this week and send it so that it arrives on Dec 9th?

Awesome, thanks!

Any plans to include Drillbit products, based on bitfury's, in your site?  https://bitcointalk.org/index.php?topic=266235.0

They haven't quite shipped yet, but that shouldn't stop you from listing them since you've got others that haven't shipped yet on there. :-)

I'm guessing a little bit here as I don't have DevCPP, but this sounds like it is probably because the two different compilers (as and DevCpp) don't mangle the names in a similar format.  So inlining the assembly within a DevCpp compiled file will probably work just fine to resolve this.   If not, you might be able to use compile-time flags to tell DevCpp to generate object files compatible with standard 'as' output?

Never used a RasPi myself, but I understood they ran some form of *nix, in which case if the miners pretty much run, why not just setup two logins and run a cgminer in each?   That way all the config files for one instance are completely separate from the other instance.   You can run them via cron or some sort of autostart script so that they launch when the RasPi is restarted.

2.7GH/s!!!   Awesome!   Congrats Drillbit Team on all the hardwork.   Now when do they ship again? :-D :-D :-D

Not to scare you, but I paid for both the device and shipping via paypal and I'm green already.   Did you pay for shipping as well as the original purchase?  It wasn't included in the original prices.

I'm also in the boat of wanting to pay shipping costs via PayPal.  I don't have any BTC yet -- that's kind of the point of buying this miner! :-) 

I've yet to find anywhere I feel comfortable registering my fiat details to exchange fiat for BTC, or those sites that look reasonable to me have weeks long processes for your first exchange, or they charge some very high fees when all I need is less than 1 BTC to pay for shipping.

I'm disappointed to hear you're thinking of making us pay in BTC for shipping when we were able to pay for the hardware via PayPal!

Lest this implication continue to fester in peoples' brains, I never tried, and certainly never meant, to imply Barntech was at fault here.  Only the BitFury manufacturer!  And apparently I owe them a bit of apology because I didn't know they had separate distributors who don't share an order base.  That's my own fault for not realizing that!

I do hope Drillbit goes ahead with ordering these earlier delivery chips and allows those of us who previously ordered boards or thumbs to move up to that reel for a fair cost.   I'd rather have my existing order earlier than have to put in a completely new order!

Barntech has posted repeatedly that the price went up for latter reels too.   So, again, I don't see how anyone should know that higher price means anything in and of itself here.

Look at it this way, if you were physically waiting in a queue to buy something and someone left, you'd be pretty unhappy if the vendor found somebody walking by and gave them that spot in line -- well after you've joined the queue -- just because they payed a bit more.  Especially when the vendor didn't first go down the line asking if anyone wanted to pay a bit more to move up first.   Yet that seems to me to be exactly what's happening here.   As far as I can tell, you all just don't seem to mind because you're jockeying to be one of those people taking those spots. :-)

Unfortunately, my budget is shot and I can't afford to buy more Drillibts on top of that which I've already ordered.  On the other hand, if Barntech goes ahead with the plan to buy another reel for earlier delivery AND offers everyone the opportunity to pay the difference in cost and move up, I'll have no complaint.  Then I will have been offered my spot earlier in line and the decision will be mine on whether to move up or not.  It seems clear that this is NOT what BitFury did -- though to be honest, I didn't know American vs. European distributor issues were affecting things here until your post.

If I'm wrong in someway on this, I'm all ears but just saying higher price = earlier spot isn't obvious, nor is letting those go to NEW buyers first.

Huh?  How should price matter given that they were already asking new orders to pay more per chip?  And how are they unrelated batches?  Are they not the same chip?

This sounds very customer unfriendly to me and I hope Barntech is talking to them about it because of that.  They should have either moved our reel up, or have offered to move us up for an additional fee based on the difference to what our reel cost vs what the vacated spot in production was being charged.

But no longer shipped to those people and offered openly for sale to anyone, aren't they?  In which case I still think everyone else in the queue ought to be moved up and these reels sold at the end of the queue.     But at least they aren't 'new' hashing power, so that's making me feel a bit better about it.

Apologies as maybe I'm misunderstanding BitFury's plans, but I will be rather unhappy if they allow those ordering chips now to get theirs shipped earlier than those who ordered their reel awhile ago!  They should be moving everyone's shipping date up earlier and selling the extra reels for delivery AFTER those who've already ordered!

I welcome anyone who can actually ship product! 

That said, am I getting my math wrong or is the $500 per chip, capable of 20GH/s to 40GH/s, a might bit more expansive than the BitFurys at 2GH/s for $25?  If BitFury was double cost, it could be equivalent to these and you already have an IC for them?   Yes, you could order a larger quantity and get it down to $4.50/GH/s but that still leaves alot of room for the BitFury's to cost more and still be cost effective. 

 I don't remember the the GH/W of the BitFury's though, and that's the dominating cost if you mine for awhile.  I think the BitFury's were around 2W/chip?  Which would make them 1W/GH right?  These claim somewhere between 0.6 W/GH to 1 W/GH (unless you undervolt them).  So I guess these would win out should you keep them mining for long enough.