Topic: ASICMINER: Entering the Future of ASIC Mining by Inventing It - page 1055. (Read 3917058 times)

Also as a side note geological/governmental diversity would not hurt.  Find the most bitcoin friendly places in the world and distribute the hardware accordingly.  This prevents "Doomsday" scenarios (government interference/geological disaster) from completely erasing the business.  Again, another issue I have not heard much about.


Anyhow that my .02 BTC hope it spawns some meaningful discussion

Everything you've said sounds reasonable, which makes me wonder if that really is the problem, as it seems like friedcat and his team would have taken steps to ameliorate it if it was the issue (OK, you can't just snap your fingers and make a data centre appear, but you still must be able to acquire rack space in a city the size of Shenzhen at short notice).

I've also read that the hardware costs about $10K per THash. AM are supposed to have about 50THash on hand that hasn't yet been deployed.

Sorry, I edited.
I was quite surprised of reading some strange notion from you, in fact.

Here it is.

Your avatar looks like a Mandelbrot set - do you have a larger version so I can see what's on the right hand side of it? Just curious.

Without the details of the psu and network problems, I'll assume them to be solvable.  At 40% AM would receive 2400 a day on average (40% of the average of 6000 coins per day), this would be worth $288,000 USD ($120 USD market).  Lets make this easy math and say we are currently doing half that $144,000. It seems feasible to scale this out to multiple data centers or separated network racks in a single center.  Essentially find the stable network, power and hardware levels and duplicate.   For the additional $144,000/1200 bitcoins this seems to be profitable scenario.

An additional solution, that would assist with time to deploy as hash rate/network luck changes, I would keep a 10% (or whatever makes sense) additional hash rate capability deployed (to be explained momentarily).  

I have bitcoind running on my desktop monitoring the blockchain inside of hand-rolled workflow framework interfacing with json rpc api of bitcoind.  It is easily doable to monitor the number of blocks being generated and (assuming there are soft controls for the mining hardware) ensure that you are receiving on average 2.4 (40%) Blocks of 6 Blocks generated per hour.  As the 10% additional hash becomes utilized consistently due to network growth, you deploy more hardware to maintain the 10% buffer.  This avoids reactionary deployment and potential loss of income.  It allows dynamic adjustment to the network and stable income.  It also saves power and wear on hardware, by powering down unnecessary hardware when not needed and maintains the bitcoin network integrity by not surpassing 40% (or whatever) targeted income goal.  Also, it should allow you stabilize buffer hardware deployments prior to the need to utilize them to maintain income. (you can test them by re-prioritizing them in the hardware pool, let me know if you want a brain dump on that)

So that being addressed, you can satisfy the core business needs (mining with hardware for BTC) with math.  Anytime hardware arrives and is not need to maintain the hash rate buffer, it can be sold.  The price should be set in a way (which again is math against supply and demand) that ensures all surplus units are sold.  This would additionally allow a certain level of SLA to be maintained with the channels producing the hardware.  My understanding is this hardware (didn't I read it costs around $100 USD to make) can be produced and priced competitively to compete with the other vendors who seem to be coming online.  

Anyhow, that's how I would do it.

Edit: grammar and clarity

I thought its pretty clear that i meant the business aspect. A company doesnt stop targetting to earn more money because its already doing fine.

Assuming that AM haven't gone higher for the reasons you've articulated, what do you think the fix would be? Deploying kit in separate data centres? Having multiple pools for solo mining? I think I've read that they are in 2 data centres at the moment, or at least they switched from one to another.

It appears that the hardware selling, while reasonably successful, is not going to shift vast quantities of blades (unless the price drops over time), so AM are going to have to solve the problem of deploying vast quantities of blades, if it is in fact the issue that they have at the moment.

+1

Comparing it to google is so senseless I'm wondering if the guy has any idea how mining works.

edit: apparently that's not exactly what he meant, sorry (still it could have been phrased more clearly )

this +++

I've been lurking here quite a while and have read most every post beginning to end.  I find this "Virtual Entity" a fascinating experiment in unregulated, unverified, and anonymous business.  I truly mean that in a positive way.

Though, there is one thing that's troubled me during this adventure.

The "network and psu" issues have happened before and there is this "we're selling surplus inventory because we don't want to scare the bitcoin community with a 51% attack" thing.

first I monitor the hash data very closely and personally verify the calculations and numbers.  There is a "theoretical" 24 th/s deployed, but in general it hovers around 18 th/s with a 4th/s swing (2 above and 2 below) occasionally spiking another 2 in either direction. Now the network has consistently stays above 70 th/s and we haven't seen 50 th/s  in ages (bitcoin time).  AM could easily (theoretically) deploy 50 th/s without risking an approach to a 51% attack.

That being said, I can't accept the "we're selling surplus inventory because we don't want to scare the bitcoin community with a 51% attack" argument (not saying I'm right, I've just been running this around alone in my head and honestly am interested in hearing others views on the subject).  With my experience with software and hardware, I can only see the network and power management complexity double with double the hashrate.  At 200 th/s, it seems odd (in my mind) to think there is feasible infrastructure in place to deal with it.

It seems strange not see this discussed more and I'm honestly interested in a fruitful discussion about it.  

50% would still be potentially devastating. Imagine that the infrastructures of a major pool are suddenly down due to power cut, earthquake, whatever. From just below 50% it will go way above for few hours.

One could say google is doing more than fine as a business too... but they shouldnt stop growing because they already earn much money. I doubt that AM will get much problems as long as they stay below 50%... of course its hard to ensure this because the network isnt as linear.

Anywhere below 50 is fine.

you are exactly right. I think the most hashrate that AM should deploy is around 35 percent of the network. More than that would be harmful to the bitcoin enviroment and economy.

Friedcat has very good reasons NOT to deploy more hash power as long as AM hold the current high % of the global Bitcoin hashrate, and it's already close to 30%, so there is little margin here. Approaching 50%, let alone going over it would definitely (and rightfully) be seen as a threat to the security of the network (basically AM could decide what makes it to the blockchain or not, and double spend at wish), creating big FUD on BTC and probably plummeting the exchange rate. Friedcat does not want that to happen of course...

And that's not the only problem, it also wouldn't make sense economically. Always keep in mind that the money supply is constant (less the random variance, and difficulty retargeting every 2000 blocks window, that is removed when considering longer time frames). Let's suppose AM gets close to 100% of the hashrate at, say 200 Thash/sec. Then if it starts hashing at 400 Thash/sec, it would still earn close(r) to 100% and earn... just the same, but with twice the hardware costs, same for electricity costs. See the problem here?

As long as we don't do parabolic bubbles like Bitcoin does now and then we'll be fine 
Of course if it happened watching this thread would probably help us as we probably started at the low end Xd

It keeps growing,steady pace

It is nice to see how resilient the market is: we are still growing at a steady linear rate despite the many overbuys that were smoothly balanced.

Can see the movie version already  

Inserts
In a world where the blockchain is in constant turmoil
Only one group of ASICS can come out on top
For the hashing power, and the block reward
Insert some (dramatic scenes)
Miner Wars 2081
Coming soon ...

Who says that I did?
Though something like this would be great.

I don't think that's gonna happen now. It would only make sense if there was at least 400THash of non-AM in June, which doesn't look likely now. I think 200THash was predicated on BFL shipping in volume and all of the Avalon batches reaching customers. I think until we know for sure that hundreds of BFL customers have hardware in their hands AM won't over commit to deploying hardware too soon.

100TH is supposed to roll out in July for their shareholders, plus several times that for the founders in the months thereafter. But it's not yet clear if they're going to hit their estimates. KnCMiner and other hopefuls are aiming for September for first deliveries. When one of these players starts producing in volume, the network hash rate will go ballistic overnight as AM try to match them.