Topic: [ANN] US/North American Bitfury sales NEW STOCK ***NOW SHIPPING*** - page 24. (Read 576772 times)

XZ site is here:http://tukaani.org/xz/
I do believe they have binaries for win32/64.

ah, can that extraction be done on a Windows machine or does it have to be linux?

You don't open the image on the raspi.
First the image is compressed with .xz so first extract it to get the actual .img file
Then use whatever disk util you prefer to clone the image to the SD card.

the *.img.xz won't open up on my raspi.

V2 img https://www.dropbox.com/s/zjk90fjra5bwhnd/rPi_V2.img (3.64GB)

It's the same stock image.   

FYI: I got it to work on P2pool and put in instructions on the FAQ page:  https://megabigpower.com/shop/index.php?route=information/contact

compared to v2 h-boards, yes. But owning a v2 mboard (requires v1.2 hboards) i considered the option of a 100GH v3 system instead, but:
1) its $25 more per board compared to v1.2 sales
2) it does not include the rPi or SD card (an extra $50 and sourcing)

Im quite happy with the v1.2 H-boards i eceives, averaging 32.8GH each across 6 of them, all pencil modded to *approximately* 1.7 kOhm

Hey guys,

if anyone needs an mboard, Iam currently selling a spare one:
https://bitcointalksearch.org/topic/m.4410833

as Dave is not offering them individually.

  nothing news on chainminer and we need someone to improve bgminer or cgminer.

if you are going to order more than 4 or 8 Hboards and better to  get 100Gh/s with mboard or 2x100Gh/s with 2 Mboards.

any new/good V2 M-board (v1.2 H-cards) rPi image?

  here link: https://megabigpower.com/downloads/rPi_V3.img

Here is the V3 image from the manual on the site.  https://megabigpower.com/downloads/rPi_V3.img
Says you need an 8GB card for it.

This is what I'm running.  It's set up to use bfgminer, not chainminer.

https://dl.dropboxusercontent.com/u/57535575/bitfury-sd-card-customized-2.img.xz

Once you have it installed, you might want to update bfgminer:

Sorry to post technical on the sales thread but does anyone have or know where I can get an image for the V3 M board.  Nothing on the OP and not getting answers from the other threads.
Thanks!!

Cool.. I am wondering if they truly intend to charge me $650 per card if I order 10, lmao.
j.k..
Maybe that should be $300/per 10 by now.

So it was a bad SD card that your rig wasn't firing up?

Thanks for info.

Wondering how late can I order for them to ship today.

based on previous thread advice, i pencil modded all the R02F resistors from 1.8K to 1.7K (+/- 0.025) and with the SD image remade, the unit is running well. heres some observations (Ill update with more as the system reaches stable operation):
note: all chips except for a few exceptions are Autotune-54 (a few are Autotune-53 due to high initial error levels)

5min: Original card that ran between 35-39GH before is at 37.5GH right now -GOOD
         New cards are all running between 30-34.5GH (average of about 32GH)
        In the past, the pencil mod often 'intensifies' with age and heat, so in an hour or so I'll post more results, and am expecting the new cards to all increase by 1-2GH as .    .       tuning takes place
1Hour: the new cards are settling in around Autotune 55-57 with a few exceptions that are around 53 and >30% HW errors
         New cards are  averaging mildly higher, about 32.5GH average. Between 7 cards i have 234GH
1Day:  The new cards are all running between 33-36GH, with an average of 34.5GH - 244GH across all 7
        

I don't have a solution to SD card failure (it's not just corruption), so I just buy lots of SD cards and keep at least one new card ready to go for each rig. At any hint of SD card problems, I pop in a new card and test the old card by putting it in a card reader on another pc and hashing its contents twice and comparing the results, by doing something like 'dd if=/dev/sdb | md5sum'. Usually multiple passes hash differently and I toss the old card and image another backup card for the next time it happens.

Heh, because exponential systems cannot continue unfettered.  People keep posting TGB mining calculator saying, "Look, Look!  Look what the difficulty is going to be, you'll never break even!!!" without giving a moments thought to what the predicted values are and what would have to realisitically occur to support them.  The point is, that the foundation of everyone's "IT WILL NOT ROI" claims are flawed because they're implying a continual exponential curve/model.  What SHOULD be used is a Sigmund Curve, or S-Curve.



As to why I call it failing, because that's what uncontrolled exponential growth systems do.  They fail, as in fail to be exponential over time.  Either the phsyical system fails or the mathematical model no longer represents the physical world.  Take your pick.  It has nothing to do with the arbitrary upper limit of difficulty, but the tangible/physical requirements to support an exponentially growing difficulty.  

The problem is that we *dont* know what's going to happen, but anyone familiar with mathematical models will realize that an unchecked exponential curve is not it.  We are not going to add 14PH of equipment in the next 30 days, 28PH in the days following that, 56PH in the 30 days following that ad nauseum.  I recognize this, and I would reckon that you do as well.  It's the inability of the system to operate in this matter that opens up the possibility of reaching a positive ROI if the consumer makes educated predictions and assumes a sizeable level of risk.

That said, I'm not going to continue my diatribe since I'm driving this thread way off topic.

I don't know why you are insisting so hard on your "exponential growth will fail" thing. Yes, of course it will fail, if will fail much before infty because there is an upper limit for difficulty, otherwise it would not be computable. Furthermore, why do you call it "failing"? We will see saturation, but no "failing".

That's how saturation (or, in your terms, "failing") looks like:

http://home.arcor.de/wetec/rechner/cladekurve.png

I see no saturation on the Bitcoin difficulty graph. Not even a clue.

http://bitcoin.sipa.be/speed-small-lin.png