I was having the same problem with 1 of my boards. I tightened all the heat sinks and checked temperatures and it still didnt help. Then I lowered the clock without lowering the voltage and it stopped the resets. If I remember correctly, my boards are currently running at 860 clock with 780mV and are quite stable, but only doing about 340GH/s each. I just tell myself that it is more power efficient this way.

When I set up my first miner, I was unable to get 400 gh with 1024hz x .820v, (It, in-fact was doing about 1.5gh, dropping and then becoming a zombie) but that is probably because I am using a 520w psu with only 16A + 18A power. I was, however able to get a good 360gh by setting the frequency to 920 and the voltage to .780. The miner has been up and stable now for 3 days. I think it is worth a try.

My miners are on the way. I'm confused, though... am I supposed to get 1150 compatible fans or 2011 compatible fans?

bobsag3, I sent you a PM, haven't heard back... I made payment on another X-1, still haven't heard anything about it... something weird happened on your side with coinbase.

Well, based on some stats about Avalon group purchase costs:
10,000 avalaon chips are capable of 2.82 TH/s.
10,000 chips costs 782.1 BTC (or thereabouts) this is roughly $78,210
In the past, it has taken 7.15 TH to increase the network difficulty by 1 million
7.15 TH/2.82 TH = 2.53 avalon batches of 10,000 chips.
Convert into dollars at 1 btc = $100.00: 2.53 * $78,210 = $197871.3
so that is $197,871.3 worth of only chips (no boards included) to increase network difficulty by 1 million.

TheJuice (and apparently, powpow) predicts a 576 million difficulty by December 29th... this is about 540 million higher difficulty than it is now:
540 * $197.871.3 = $106,850,502, which is probably on the high side of how much it would cost (assuming there are people who can get hashing power for a little cheaper)

Obviously, this is just a ball-park figure.

So you guys really think there will be an additional $92,000,000 (give or take 10,000,000) invested in mining by the end of the year?

This post is incredibly long, the end goal is to say "Yes, there is hope that bitcoin mining profitability is not dead yet." Bear with me here.

To all of our friends who are saying profitability is dead....

Not yet. Maybe "insanely quick profitability" is dead, but not profitability. some quick figures:
it costs about $300.00 to purchase a fully functional K16. Steamboat has said that the chips should be able to punch out 325 MH/s each.... a single miner then comes out to 5200 MH/s.

Now for some conjecture:

Difficulty adjusts every 2016 blocks.
as of lately, blocks have been about 8 minutes long. (a result of a 20% increase in hashing power each set of 2016 blocks; the last 6 difficulty changes have averaged out to about 20.79%; the last 17 have been 15.89%)
It takes then, about 10.9 days for 1 difficulty adjustment to go through.
if the % increase remains constant at 20%, there will be 33.5 difficulty adjustments in a year
that means that a machine which could get 1 bitcoin in the first difficulty set would be able to get only 1/(1.2^33.5) bitcoins at the end of the year, or about 0.00223 bitcoins. In other words, difficulty increases to 449x what it was at the beginning of the year.
our January 8, 2013 difficulty was 3,249,550.
449 * 3,249,550 = 1,459,047,950 Projected difficulty on Jan 8, 2014 if current growth had been sustained from the beginning to the end of the year.

Ok, we are off-target for hitting 1.46 billion this year. if growth increases at 20% per difficulty adjustment for the rest of the year, we would hit about 500,000,000 difficulty. (This would require $99,149,202 worth of network chip hashing power to be in-place and running by the end of the year. [more info on how I came up with that figure below])

all the calculations below are done with avalon chips and avalon chip equivalents. This is done because an assumption is made that chips by other companies will be approximately the same in a chip's Mh to cost ratio.

theoretically, if a 20% change had been sustained through this whole year (it has not been, see 3 paragraphs above), the below 1.46 billion difficulty would be hit. For this example, let's say it has been and will be sustained through the end of the year.

I would be very surprised if the difficulty continued to increase up to 1.46 billion by Jan 2014, this is why:
Based on historic data, 7.15 TH is enough to increase difficulty by 1 million... (equivalent avalon asic chip cost of $198,298 not including boards)
Thus far, Avalon has received about 331.9 TH worth of chip orders, worth $7,577,980. The current difficulty of 37,508,353 represents approximately that same amount of cost ($7,510,356).
so, a 1.46 billion difficulty would require 10,439 TH to be achieved... (equivalent avalon asic chip cost of $288,722,476 not including boards)
I doubt another $280,700,000 will be invested into buying chips by Jan 2014 to create the 1.46 billion difficulty that would result.

In short, this current rate of growth is not sustainable without either an increase in bitcoin's per unit worth, or an increase in MH/unit cost for chips. I think it is pretty likely that the current growth rate is not here to stay, look for it slowing down within the next 6 or 7 difficulty adjustments. (7 more adjustments at 20% would result in a difficulty of 134,399,210 for an approximate network cost of $19,000,000 in chips alone.) People already seem to be slowing down in how fast they are snatching up more available hashing power, and how fast they are snatching up hashing power that will be available in the future.

Here are some ROI calculations:
assume your electricity costs $0.15 per kWh and assuming a single miner uses 50 watts per hour to run.
assume a very steep .1 profitability decline per year e.g. difficulty increased to 10x the original difficulty after 1 year.

At difficulty 37.4 million:
over 12 months of running you would earn 692.67$. It will take about 1.64 months to pay for your miner, the rest is income.

At difficulty 100.4 million:
over 12 months of running you would earn 18.91$. It will take about 7.5 months to pay for your miner, the rest is income.

That may only be a 6% ROI, but it is still profitable.

Some figures for when to pull the plug and chuck for scrap. ROI calculations are much more difficult, and were covered above. Also note, this table below does not take into account the cost for cooling. This table compares electricity costs and difficulty.
The question: "Should I keep mining with my K16?"

Difficulty	$.25 per kWh	$.20 per kWh	$.15 per kWh	$.10 per kWh
2.25 Billion	no	no	no	no
2.2 Billion	no	no	no	yes
2.1 Billion	no	no	no	yes
2.0 Billion	no	no	no	yes
1.9 Billion	no	no	no	yes
1.8 Billion	no	no	no	yes
1.7 Billion	no	no	no	yes
1.6 Billion	no	no	no	yes
1.5 Billion	no	no	yes	yes
1.4 Billion	no	no	yes	yes
1.3 Billion	no	no	yes	yes
1.2 Billion	no	no	yes	yes
1.1 Billion	no	yes	yes	yes
1.0 Billion	no	yes	yes	yes
0.9 Billion	yes	yes	yes	yes
0.8 Billion	yes	yes	yes	yes
0.7 Billion	yes	yes	yes	yes
0.6 Billion	yes	yes	yes	yes
0.5 Billion	yes	yes	yes	yes
0.4 Billion	yes	yes	yes	yes
0.3 Billion	yes	yes	yes	yes
0.2 Billion	yes	yes	yes	yes
0.1 Billion	yes	yes	yes	yes

sokol815

long live bitcoin!

consuming a modest 18.7 kw/h!

How about preventing a 51% attack? I'm sure that is a good cause, no? Recent developments with FTC are indicating it was 51% attacked. Some people are speculating that LTC will be next. purchasing even 9.2gh/s will help to secure the bitcoin network and your bitcoins.

I'm curious as to if these boards will fit in a standard optical slot on my computer... (right by the CD-Rom drive) that would be very convenient. Anyone know?

do the k16's require a raspberry pi, or will a desktop PC be able to run them?

yeah, 204 isn't much in the big game. I would also like to point out that this probably doesn't include the actual avalon miners that they have been shipping out. Just the DIY stuff people have been making.

Yeah, steamboat said he would purchase the order when half of it had been filled. So the order has already been purchased, just around 4000 chips have not been claimed from the order, yet.

The order is about half completed. (see thread title)

uh oh... I don't know whether this statement is trustworthy or not! 

50 TH on TRC would be insane.... I don't think TRC will end up being worth anywhere near that much.

Hey everyone, I was looking at the Avalon bulk chip order bitcoin receiving address:
https://blockchain.info/address/1FGAftzSTztFSB8LMwsrdCKTyqGY6zr3sU

this address alone has 55,462 BTC.

Assuming about 782.1 BTC/batch of 10000. (which is being a little conservative, as I am sure really large orders-like some of the 5000 BTC ones-are probably getting some kind of discount)

This comes out to be approximately 71 batches of 10,000 avalon chips sold. resulting in a massive creation within the next few months of a total of 204.48 TH on the bitcoin network. Some of these may already be operational, but I am willing to bet that there is at least 2 or 3 times more hashing power (probably another 500 TH) coming online within the next year.

It is not unreasonable that the bitcoin network will increase in difficulty nearing the 100 million mark within the next year. this is probably really good news for the security and value of bitcoin.

for the most part, an fpga is used as a stepping stone towards an asic... you would be very hard pressed to find any kind of fpga the size of a thumb-drive that is able to get a small percentage of a normal avalon chip (288 MH/s) (or even one that you can put the chip into the fpga).... for the most part, you just don't see fpga boards that small.

check this out:

yet on their website they have posts from december 2012? hehehe. nicely done scammers!