ASICMINER: Entering the Future of ASIC Mining by Inventing It

[SebastianJu]

Check out this latest change to 0.8.2:
https://github.com/bitcoin/bitcoin/pull/2577

This pull defines 'uneconomic dust' as 54.3 uBTC (5430 satoshis, about $0.007 at current prices), and treats any transaction with outputs less than 5430 satoshis as non-standard (won't be relayed, won't be mined). 5430 satoshis is derived from the cost (in fees) to spend a TxOut/TxIn. See https://people.xiph.org/~greg/txouts2.png for proportion of recent outputs this will (eventually) affect.

Will this make the initial Satoshis sent on dividends no longer relay/get mined?

I believe this only means that the border where one has to pay a minimum fee was set down from 0.01 to 0.00005430. Im not sure about this, but it sounds this way.

Variance will only remain a non-issue if they keep adding hashrate so that they remain at a constant proportion (or increased proportion) of the network. As soon as the their proportion decreases, variance will increase.

But will it matter? You have the chance to earn more or less in a week, but mostly its an average. Because it is more or less you wont lose anything in the long run. Of course you can earn less in the first week and more in the next week when the difficulty raised but it can be the opposite way too, so i dont see that this is a big problem.

What i wonder is if this ip is attackable then. It would be good to have some more ips to chose from or a pool to switch too in case of an attack.

The transparency isnt a problem for me too because we have to trust friedcat and co anyway. We would not be able to see if they throw some hashpower to another pool or solo mine. The numbers shown at a pool are in no way safer than solomining. And if its about the general hashingpower... i wonder if it would be hard to make a website that is showing the complete asicminer-datacenter-hashrate. I mean in the datacentre theres probably a software that has a overview about all the blades to see if one is failing. So this software probably has the hashrates too. It wouldnt be a border to implement this into a website then.

Only saying...

[lan787]

But will it matter? You have the chance to earn more or less in a week, but mostly its an average. Because it is more or less you wont lose anything in the long run. Of course you can earn less in the first week and more in the next week when the difficulty raised but it can be the opposite way too, so i dont see that this is a big problem.

It matters for someone. Consider your employer pays you either $200 or $300 this week depending on pure luck. Somebody would not take that risk(for ex. he has a loan to pay). Risk alleviating instruments cost real money.

[Rodyland]

But will it matter? You have the chance to earn more or less in a week, but mostly its an average. Because it is more or less you wont lose anything in the long run. Of course you can earn less in the first week and more in the next week when the difficulty raised but it can be the opposite way too, so i dont see that this is a big problem.

It matters for someone. Consider your employer pays you either $200 or $300 this week depending on pure luck. Somebody would not take that risk(for ex. he has a loan to pay). Risk alleviating instruments cost real money.

No (responsible) company in the world guarantees dividends.  They may try to always pay the same (or even an increasing) amount, but a guarantee - to me that would be a sign of management with their head in the clouds.

As investors and owners, we bear some risk.  If we can average a few percent higher net profit, at the cost of a little bit higher week-to-week volatility, then as long as that volatility doesn't put at risk the company's ability to pay its bills as they come due, I say do it.

[Mccanical]

Solo mining seems like a no brainer to me. Might be more/less than the average each week but at least we're not relying on any pool to handle our bitcoins for a percentage.

[kano]

You should consider p2pool instead of solo. Your traffic stays local like solo, but your variance is decreased like it would be on a pool.

Come on Diablo, with 8 Th/s, variance is a non-issue. They mine 500+ blocks per month on average. They have more chances of being hit by thunder 10 times than to mine zero blocks during a specific month and being unable to cover their monthly expenditures.

p2pool leads to a higher orphan rate than solo mining.

Solo mining is obviously the best choice with 8 Th/s.

I totally agree with this. Solo is definitely the best option for ASICMINER and it's shareholders. Please don't go into experiments like p2p pool

They won't do p2pool.
They only need to point one 10GH/s board at it and they'll see the current problems with p2pool.
Then imagine 8TH/s ...
Read the p2pool thread for more details.

[OneMINER]

So ASICMINER don't have secondary pool in case 1st pool went offline . Isn't that like.... easy to setup or something?

We are switching to solo mode since we could not find enough pools having stable connection from China to distribute the hashpower.

The solo solution is being tested/done along with the deployment. The only problem is transparency. We plan to do it with writing information to the coinbase transaction to let everyone check.

You should consider p2pool instead of solo. Your traffic stays local like solo, but your variance is decreased like it would be on a pool.

+1, and increase the income by using this https://bitcointalk.org/index.php?topic=62842.0

Not suggesting we actually invest in alt-coins (Although it wouldn't be a bad idea), Just suggesting collecting them and selling them daily on vircurex or something.

So we merge mine with these asics for alt currencies and then dump them on exchanges? Do you realize how fast the value will fall for these alt currencies doing this? It will quickly become worthless.  I cant imagine this is the direction ASICminer will go.

The difficulty for the alt chains would quickly adjust to bring the rate of coin generation to normal. IMO the markets will soak up low priced coins quickly. It would be near 100% profit on merged mining coins. Extra effort on trying to not 51% the alt coins may take up some time though.

You should consider p2pool instead of solo. Your traffic stays local like solo, but your variance is decreased like it would be on a pool.

Come on Diablo, with 8 Th/s, variance is a non-issue. They mine 500+ blocks per month on average. They have more chances of being hit by thunder 10 times than to mine zero blocks during a specific month and being unable to cover their monthly expenditures.

p2pool leads to a higher orphan rate than solo mining.

Solo mining is obviously the best choice with 8 Th/s.

This is a common misunderstanding. It's not your ultimate reject or orphan rate that counts on p2pool. It's your rate relative to the rest of the pool. With a well connected node it's possible to achieve greater than 100% returns. I'd like ASICMINER to do a limited test at least. If an efficient, merged mining p2pool node was set up, the upsides could be significant. The big question is if ASICMINER's custom gear would play nice with p2pool.

[JimiQ84]

Why are we even discussing this? Pool with 5TH/s hashrate (and less) have minimal to no effect on AM's variance (when AM will have 15TH/s).

[organofcorti]

Why are we even discussing this? Pool with 5TH/s hashrate (and less) have minimal to no effect on AM's variance (when AM will have 15TH/s).

I don't even

[OneMINER]

Why are we even discussing this? Pool with 5TH/s hashrate (and less) have minimal to no effect on AM's variance (when AM will have 15TH/s).

Not talking about variance. Talking about increasing profits.

Edit: I agree, variance will be a non issue if on a pool or not. Profits.... ^^^

[lan787]

But will it matter? You have the chance to earn more or less in a week, but mostly its an average. Because it is more or less you wont lose anything in the long run. Of course you can earn less in the first week and more in the next week when the difficulty raised but it can be the opposite way too, so i dont see that this is a big problem.

It matters for someone. Consider your employer pays you either $200 or $300 this week depending on pure luck. Somebody would not take that risk(for ex. he has a loan to pay). Risk alleviating instruments cost real money.

Wages are fixed because most persons give away their complete wage each month for fixed costs and so on. But the AM-Dividend cant be foreseen. At least i had to learn this when i tried to get a loan. It didnt happen and it was good this way because the dividends dropped down lower than i thought. So it wouldnt make much difference in this case i believe.

Seems I was misunderstood. "Wages" was an example to illustrate my point that profit variance is a bad thing. It may be not "a big problem" but It may as well be depending on circumstances. Thus sometimes it is better to have lower but more reliable profit. That is purely accademical statement.

[TheSwede75]

But will it matter? You have the chance to earn more or less in a week, but mostly its an average. Because it is more or less you wont lose anything in the long run. Of course you can earn less in the first week and more in the next week when the difficulty raised but it can be the opposite way too, so i dont see that this is a big problem.

It matters for someone. Consider your employer pays you either $200 or $300 this week depending on pure luck. Somebody would not take that risk(for ex. he has a loan to pay). Risk alleviating instruments cost real money.

Wages are fixed because most persons give away their complete wage each month for fixed costs and so on. But the AM-Dividend cant be foreseen. At least i had to learn this when i tried to get a loan. It didnt happen and it was good this way because the dividends dropped down lower than i thought. So it wouldnt make much difference in this case i believe.

Seems I was misunderstood. "Wages" was an example to illustrate my point that profit variance is a bad thing. It may be not "a big problem" but It may as well be depending on circumstances. Thus sometimes it is better to have lower but more reliable profit. That is purely accademical statement.

All true statements, but if I had low monthly expenses and guaranteed employment for 20 years I would still take 'between $2500-$5k a month' over 'Always $3k' and just suck up the months when I make $2800. It's all about your capital requirements and long term profit margin.

[lan787]

Regarding ASICMINER, 10% of hashing power is a very high share for me. If we end up with few players that constitute 51% of hashing power, that would be a threat to stability of the system, thus reducing it's trustworthiness.

I understand that having 10% rather than 1% is more profitable, but it may ultimately hurt the whole bitcoin ecosystem.

[furuknap]

Regarding ASICMINER, 10% of hashing power is a very high share for me. If we end up with few players that constitute 51% of hashing power, that would be a threat to stability of the system, thus reducing it's trustworthiness.

I understand that having 10% rather than 1% is more profitable, but it may ultimately hurt the whole bitcoin ecosystem.

Over time, this will be a minor problem as technology becomes more available and shared across a larger user base. In the short term, it can even be beneficial to the credibility that there are a few major players that increase the total hashrate by such an amount, thus making the network more resiliant to a rouge entity wishing to attack. Right now, we know there exists enough hashing power to take over Bitcoin; with AM, Avalon, BFL, 100TH, and others adding network power, that becomes less likely.

.b

[lan787]

Over time this may be the large problem. ASICMINER may become rouge itself.
I see little difference between central bank and dozen mining companies. Especially if half of them will be in one country.

[furuknap]

Over time this may be the large problem. ASICMINER may become rouge itself.
I see little difference between central bank and dozen mining companies. Especially if half of them will be in one country.

Over time, AM will not be able to hold 51% :-)

.b

[lan787]

Again I see little difference between one central regulator and 10 major players. It is not a peering network.
Heck I gues even 10 players won't last for long. It somehow always goes to two entities.
Republicans vs democrats
Intel vs amd
iOS vs android
Coca cola vs pepsi
Paper or plastic

Ok this is offtopic. I'm going to stop here.

[eleuthria]

I can't resist responding...I'm so sorry friedcat for putting these here, but it's irresistable:

ASICMINER may become rouge itself.

thus making the network more resiliant to a rouge entity wishing to attack.

Rouge	Rogue

[furuknap]

I can't resist responding...I'm so sorry friedcat for putting these here, but it's irresistable:

Well, I really meant the redish type of entity!

 

.b

[eleuthria]

I can't resist responding...I'm so sorry friedcat for putting these here, but it's irresistable:

Well, I really meant the redish type of entity!

 

.b

Years of playing World of Warcraft [it kept me sane during college and a bit after] have just instilled the need to troll whenever I see 'rouge' .

[furuknap]

Years of playing World of Warcraft [it kept me sane during college and a bit after] have just instilled the need to troll whenever I see 'rouge' .

Well, I guess that those college years have killed your career working in or managing a chain of cosmetics stores, then :-)

.b