Yep.  No one will mine to lose money because that makes no sense.  The more people who drop out, the lower the difficulty goes.  Mining has seldom been unprofitable since the beginning of bitcoin.

That being said, we will probably see BTC mining profit approaching that of other popular investments (3-7% per annum) after ASICs come out.  I always expected mining would likely become centralized to large mass ASIC possessing corporations as profit reduced, as a corporation with $50 million invested would be happy to make $1.5 million in profit per year or $51.5 million in revenue.  By contrast, the average joe probably wouldn't care to generate $15 of profit from a $500 investment in an ASIC miner.

Right, I guess I can just use the block hashes themselves then.  Reading up on it, it doesn't seem like multiple hash algorithms should be a problem, although I don't think it's been implemented before (PPC uses only SHA256 as a hash function; the proof of stake blocks are just very low difficulty apparently).

Yeah, I won't deny that ASIC will be more energy efficient.  I'm working on developing a fork right now that overcomes this problem without  severely impacting GPU performance.  In the meantime, I'm mining LTC because I can see it continuing to be a major force in the land of alt. cryptocurrency.

Hi,

I'm working on a bitcoin fork right now and I'm trying to figure out something related to data structures.

I want to develop a blockchain (hash tree) whose algorithm for hashing subsequent blocks is determined as a function of the nonce of the 0th, 0th + c, 0th + 2c, ... , etc block (leaf) where c is a constant.  I assume that the nonce values are pseudorandom and non-predictable, but I don't know if that's actually true.

Which leads to the question: Can I compose a hash tree like this?  What kind of problems do you think I'll run into, and are they impossible to overcome?  I want to see if anyone knows the answer to this before I begin trying.

I don't know if this kind of hash tree has a name or not either, or if it's been implemented before.  I guess it might be called a polymorphic hash tree.

It requires roughly 128kb per process; for 1600 processes (eg 5870) you then need about 200mb of memory, although the algoritm as implemented uses a bit more typically. If you want to figure out what effect that the gddr5 memory is having on mining scrypt, turn your memory down to about 25mhz; at that speed you'll be pulling about the same quantity of bandwidth as the ddr memory you're refering to.

Memory bandwidth in the internal bus is critical, hence why GPUs mine so much faster than CPUs (GPU memory bandwidth is 5-10x that of the L2 cache).

Now if you can put 8000 scrypt cores on an ASIC clocked at ~150 hz with 8gb ddr3 memory at 25gb/s, congrats, you've made an ASIC exactly as fast as a GPU that maybe uses a little less power.

I welcome an FPGA miner to the scene, but don't be surprised if you build one and the results are underwhelming.

Looks good, am getting 0.48% stales with the proxy and reaper

Be sure to get 12v from the PSU rather than the board for the PCIe slots, any more than 3x cards on a single board without that tends to toast the 24 pin connector

9k shares are fine -- what matters is the quantity of coins you get since different diff shares have different payouts.

http://mining.bitcoin.cz/stratum-mining

It's a new low bandwidth network implementation for mining.  It also reduces stales as it takes way less time for work to be sent to mining nodes.  One of the problems with the litecoin network over the old JSON protocol is the time it takes to send new work to mining clients when a new block is found.  The protocol was designed to facilitate high throughput mining operations on the bitcoin network but works well for the litecoin network too.

Kay, updated!

edit: It went down to -30C recently and I checked the power usage of my outdoor rig with 3x 7950s; it went from 750w to 650w, an almost 15% decrease in power consumption!  Anyone looking to save on power costs should consider outdoor mining.

It was on one of those quad a9s yeah.. but I heard they were pulling 2w and not 6w like that person indicates and pulling ~8KH/s. But maybe they didn't check power consumption accurately.

If you don't pay for power, 6950s that are unlocked are the best. If you pay for power, 7950s.

Can someone please explain algorithmically how the reward algorithm works? I'm not understanding the 0% interest rate or statements like

Value of what?  So 5% of any coins I hold are destroyed each year?  Isn't that a disincentive to do anything but spend coins?

Yes, you need really fast RAM hence video cards work so well.  They're actually quite a lot faster sequentially than on-die L2 cache on most modern CPUs.

You can get really high power efficiency with small ARM chips with NEON and L2 caches like Cortex A9s (~2-3x that of video cards) but the chips are expensive.  You could either make an ASIC with the cache on-die (lots of space used...) or an ASIC that accesses GDDR5 through a high performance bus, but you will probably never see the 20-100x increase in efficiency you would with SHA256 only mining.  You need about 45k circuits for a scrypt core (Salsa20/SHA256) versus 25k for a SHA256 core as well, so you use up more die space there too.

Most people who keep calling LTC a "scam" haven't looked at the encryption algorithm thoroughly.

Updating the mining guide with more information on how to run stratum correctly with ltcmine.ru

if you are getting high stales make sure you are using the mining proxy!

My bad, I misread jeratuls post.

the url for the pool you are trying to is malformed or you have entered the wrong port, username, or password.

if you have http in front of the pool name it will not work either

personally for a 7750 because of the low gcn core count I would use cgminer and the following config:

Make sure you set the memory and gpu speeds correctly too, you should get ~150 kh/s

CoinLab   I've got no desire to put my wealth into any centralized organization after Bitcoinica etc

With the number of blocks it's mining per day, it represents about 2 TH/s.  It's possibly a pool.  It's not http://pool.itzod.ru/ because that's too small, so it may be a Russian private pool.

Also like a pool, the block generation rate seems highly variable.

It's impossible to tell if BFL is a scam or not at this point, since we haven't seen a product from them at this point or any reasonable indication that they even ordered wafers from a fab (unlike Avalon, who has showed their TSMC inventory sheets).