Topic: ASICMINER: Entering the Future of ASIC Mining by Inventing It - page 401. (Read 3917543 times)

It proves that when market forces are in place to run an ASIC inefficiently it will be run inefficiently. Reduce the impact of the cost of the ASIC and your considerations as to where to optimally run it changes. The cost of building power supplies to run it,  the space required to locate it and the measures required to cool it all change. For all we know this could be a very efficient ASIC at lower power which isn't the end of the world given what's planned to go into mining and franchising but we don't know.

What are you talking about?

What exactly does bitmain chips meeting advertised specs prove?

Can you see past your simple arithmetic for a moment and understand it's the principle being demonstrated in the example?

Now let's go further using the principle. If you're making your own ASICs you're paying somewhere in the neighborhood of the square root of dirt. Got that math? If the relationship between power and performance is not linear then there's potentially a sweet spot somewhere below where you flex your muscles and advertise your ASICs at when you're attempting to sell each one for top dollar. If you're building say, 23p of mining facilities that you're not flogging on eBay when you think you're done with it; you're going to build it to work for you as long as possible. Who's to say 11.52 Gh/s is the sweet spot when we don't have the other numbers? It may be a good point for RM to operate his first equipment based on his profit margins at the time his equipment will hit market but it's a snapshot in a much longer time line.

OK

As usual, it "depends". If it were an efficient system where only the high cost electricity zones would deploy the new hardware generations, you could probably maintain a decent amount of asic generations operating between the high and low cost electricity zones (a new generation could push the current one to a lower cost zone). But this would assume no new generation is deployed directly in the low cost zones.

I would think everyone investing significant amounts will tend to operate in the lower/lowest costs zones, so it would skew the system dramatically (and the "steady state" could only work with few generations of asics).

But is a "semi steady state" really a "steady state"?

Here you're opening a whole new can of worms. I think that would depends on your OPEX. CAPEX is high with the tanks but if maintenance (OPEX) is low enough, it would be a decent target for countries where labor and space are expensive (but what of electricity costs, what of heat dissipation power usage?). If OTOH you have cheap labor and space, the tanks aren't looking that good anymore.

Also, unless there is a way to stack them, they won't really help you improve the density of your deployment (they have other issues with connectors, plumbing, etc) vs a regular datacenter (more on that after the end of June, I guess). I'm not really sold on the tanks, I think the idea is neat, but the numbers don't convince me so far (probably need more data).

What about it? bitmain always advertised up to 0.68w/gh.

Here's a real world example where three S1's are consuming a total of 500w hashing at 140gh/s each:

https://bitcointalksearch.org/topic/guide-undervolt-antminer-s1-119wgh-at-the-wall-526060

Yes this is exactly what I am assuning due to the fact that no asic manufacturer has advertised below 0.5w/gh (bitfury excluded) and nobody has been able to significantly undervolt below advertised limit.



Heres a hint:

knc can be overclocked 1.7 times  (700gh/s vs advertised 40gh/s)
Hashfast can be overclocked 2 times (800gh/s vs 400gh advertised)
And bitmain can be overclocked 2 times (200gh vs 100gh advertised)

All of which have no problem hashing at advertised speeds.

Again, got any evidence that other manufacturers are not on the far left side of that chart?

Either you cant read a chart or you are assuming current asics run at a point to the extreme left of that chart, which would also imply those other asics can be overclocked by factor of 2-3x. Here is hint: almost none can. KnC, HF, CT, bitmain, Bitmine .. you'd be lucky to hit advertised speeds, let alone 2-3x more.

Either way, welcome to my ignore list.

I think it illustrates the exact opposite. Notice the diminishing returns at around 0.3w/gh? Other manufacturers may have hit that limit at 0.6w/gh (bitfury excluded).

ITs not meant as promotion for BFL, Id be the last person to suggest you order anything from there. But the chart is interesting and relevant to earlier discussions about how you can scale power efficiency of every ASIC. Even if you take those numbers with a table spoon of salt, it nicely illustrates my point.  There is about a factor 3x difference in power efficiency depending what voltage/frequency point you pick. That spread is not going to be vastly different for any other bitcoin asic.

They've posted some web interface on their twitter feed so I assume they'll do standalone miners (as well)

Been away a while....so, when/why have we stopped receiving weekly dividends?

Cheers,

 Two pre-orders, many lies and delays later, I do not trust anything they have to say.
You need only look at this disinfographic to understand their ways.  It's more lies.  Everyone knows that KNCminer shipped ~ 1W/Gh at the wall for example.

BFL, your efficiencies - real or imagined - are already lost to your competition.  Mining for longer is moot since the difficulty will be so high you will be making NOTHING but heat and an infinitesimally small fraction of a Bitcoin per day.

By the way, the very day before the disinfographic was posted, Josh said that the new BFL chip was more than 2x more efficient than anything our there.

https://bitcointalksearch.org/topic/m.6449710

 A day later and it's already 3 to 5 times more efficient?  Imagine how efficient it will be when it finally delivers in two weeks^tm

Caveat Emptor!  Information asymmetry detected.

 

RM is only one AM customer and one board designer of many. Seems his boards are using USB.

There are other board designs in progress from other board designers. With Ethernet and with a different power consumption. And a different number of chips per board.

FWIW, this is what BFL published recently:

http://www.cybtc.com/article-900-1.html

?

Good enough to plan to produce a significant number of them according to a recent report.

as we've talked (in length!) when the results came out, its not stellar, but not terrible..more testing/optimisation required.

Is this good?

RM's results as of the 14th of April.