Topic: A Public Plea to Bitcoin Developers and Supporters alike - page 3. (Read 10863 times)

Ice_chill - Thank you.   I understand your point of view.  

Thanks for adding that. That was kind of the conclusion I had reached after more thought, but failed to write because I had already rambled far too long.

asic is SO bad for bitcoin.  NOT good at all!!!!!!

There is nothing wrong with only one company developing ASICs right now for 2 reasons.
1. We don't know when a second company will come up with their own ASIC.
2. ASIC is the natural evolution from FPGA, and it would take a lot of luck and great coincidence to have 2 different companies develop or release ASICs at the same time, it's logical that one company will be the first and the only one for a while, before other companies step in.

Regarding having only a few MegaMiners, these are smart and successful business men (if they have the money to invest) and they will be interested in making a profit.
This is where it gets interesting, the more trust there is in Bitcoin, the more people use it and the more the price of each Bitcoin goes up, thus if the people with 1TH rigs start messing about, then they are destroying their own wealth.

Also you'll be surprised how many people around the world are more than happy to make $500 per year from mining, I think the $150 unit will sell in enough quantities to make Bitcoin well decentralized.

Meta:  Not sure where the "g" came from.   I'll assume you meant "within".   

Back to your point, use your imagination a little more.   

This announcement pushes the bar so high that there is effective no company that is going to step-up to make the investment in developing their own custom chip (ASIC).  So the centralization issue is two-fold.  One, being that only a single company will be effectively supplying all the hashing power for Bitcoin until we have a true competitor.   Second, being that the profits for mining that were and are more decentralized at this moment, will now go to the miners that will make the capital investment to keep up with this new hashing power cold war.   Not at first but give it a little time as more large units ship out, the difficultly will increase to keep up and eventually you will only have a few mega miners that can have operations that are profitable enough to keep them mining (ie: incentive).   

Now you have two more points of attack along with the exchanges to really give Bitcoin a lot of hassle.   I am not sure what the answer is at this point, I would like to have Gavin or other core developers weigh on this.  To try and ignore that this could be a potential issue is disingenuous.

The code does not allow "from withing" attacks.

Why do you assume the attack isn't already in the works?   What I am saying is that one of the effective ways to attack Bitcoin is from within.

Just wait for the quantum computing

Well on the bright side the ASIC is the last mega leap, after that advancements will not be x20 like it is now.

If bitcoin needed my help to succeed, it would not succeed.
Bitcoin is fundamentally a good idea and superior to other options. No one is going to ruin it by producing products to use it. Of course it may eventually be the end of casual mining, but mining is just a distribution method.
IMO, we are about to go mainstream and unless you scale up $20k or so your mining dreams will just be a fantasy.

No.  BFL has not yet entered the ASIC market; they have issued a press release.  Others have spewed hot air about doing an ASIC as well, and no doubt a few others are working silently on one.


Another erroneous assumption.  There is only a single hurdle to leap -- money -- in building an ASIC.  Once you've handled that problem, the rest -- using the latest industry process and die size, etc. -- will come naturally to any new ASIC competitor.

Its a free market. I wonder how much it would cost to buy bfl out by the community.

Sure, if it's competitive. Otherwise, why bother?

Changing the proof of work algorithm periodically would do nothing more than promote centralization. Why? Simple: I don't know of any developer that will continually rewrite a GPU kernel or FPGA bitstream from scratch "just because". They need a damn good reason because they put a lot of time and effort into it. Therefore, the only ones with optimized hardware would be those like ArtForz who own the hardware already and are capable of writing fast kernels to run on it.

We even saw previously him (and others?) selling access to faster kernels, just because he had the expertise to write them in the first place. Don't kid yourself that an algorithm change periodically just for the fun of it would in any way promote decentralization - even though a kernel may be written to support the new proof of work and distributed to the miners, it will never be the fastest available in the first iteration.

Who ever believes otherwise, hasnt thought it through. Think of it this way, if you have an asic, you could sell a relatively small amount of them for a few months and pocket a huge markup on them, and then immediately turn on 5x or 10x more (virtually free) asics and mine with that yourself, ensuring your customers essentially threw their money away. Sure, you could call that evil, but those customers did throw their money your way, so do you have any doubt it would earn you more than mining yourself from the getgo without selling anything?

6 or 12 months after the first asic comes to market, things might change, as price per GH may have come down so much it may not be as profitable to sell asics, but who ever comes to market first would be nuts or mathematically impaired to not sell.

I know.. Just $150 for such a massive amount of additional security in the block chain.  Isn't it awesome?

Oh.. you think that's not a good thing?



No, I'm not getting it.  Why is it bad that miners can now buy reasonably priced dedicated mining hardware, that gives an incredible boost to bitcoin security for the rest of us?  That the block rewards actively incentivises them to do so is bad?  Nope.  Just don't see it.



Oooooh... you've bought a load of GPUs and they're looking a bit worthless?  Awwwww; shame.  It's not really the downfall of bitcoin if you get stung a bit is it?  Still, look on the bright side: when the next computational leap in bitcoin hashing power comes, all those who carelessly buy these BFL products will be in the same position as you.

Your concerns about centralisation are unjustified.

• 3.5 GH/s for $149 is $42 per GH/s
• 40 GH/s for $1,299 is $32 per GH/s
• 1 TH/s for $29,899 is $29 per GH/s

Perfectly reasonable economies of scale; and all in the same order of magnitute.  There is nothing to stop 1000 small investors either buying one mega rig between them, or 1000 mini rigs of their own.  Either of which will represent "decentralisation".

Fruit stocks skyrocket on breaking news that they are the wave of future currency technology.

True, but there is a dynamic to Bitcoin where this developing situation could be tragic. First, consider these points:

1) Every government institution has passed laws that directly and/or indirectly enforce their monopoly over money (or the monopoly of their banking buddies). Bitcoin is the first real contender to compete given its technology is pear to pear and can run on equipment available to all.

2) The ASIC market is a high barrier entry especially if you consider the Bitcoin economy size. BFL entry was premature and it was done on purpose to remove the incentive from potential competitors.

So, what are the possible outcomes?

1) BFL plays their cards right and serves the community just good enough to guarantee there will be no competitors (remember silicon technology is high barrier).
    They will make bitcoins cheaper for themselves compared to everyone else. They will always have the first chunk of the pie with each advancement.
    But the real problem is the source for bitcoin mining technology will have become 100% centralized. I hope I don't need to explain the problem there.
    Too much power or influence over any monetary system will always lead to abuse and corruption down the road. Also, it's a single entity for government to go after.

2) The Bitcoin community forfeits some of its power and control over Bitcoin. BFL at any time could take us for a ride. That is happening now; but just you wait a year down the road when they are the only vendor on the block and then they decide to decrease their hardware cost by 25%. Again, too much power and influence over the market. Insider trading that can't be punished! At least if they do something too stupid, another competitor will be much more likely.

I must confess, my last rant was full of emotions and I was grumpy. I was dumping on the community so to speak.
This led me to ponder the alternatives more. Here are my thoughts:

Ideally, Bitcoin mining should always be done on equipment that has a different primary use. For example, mining on GPUs.
If the algorithm was changed to be hostile to specific hardware, it would still only be a matter of time before someone finds a new approach
to make specific hardware successful again. So, what does that leave us?
Like the OP proposed: The algorithm would need to be altered on a periodic basis or from time to time. This would discourage anybody from making specific hardware.
This comes with its own set of challenges. A trustworthy source by the community would have to lead the way in order for each change to be successful. Changing the core components to Bitcoin is extremely difficult. I just don't see this happening.

So, the reality: the majority are going to go with the flow. BFL coin here we come. I remember my friends advice that said life is too short to run against the current. I'm going to take his advice. I just hope the community and all of you won't be oblivious to the developing situation. Better yet, I really hope BFL becomes a better community player and makes some sacrifices like making their ASIC design open source. Now that would be awesome!

Just ordered some more BFL equipment BTW to support BFL coin. :-|
Plan to follow the open source ASIC initiative much more. Hopefully it will find some venture capitol.

Thanks for reading my thought and sorry about my previous post.

Edit: minor grammar edits

How much cheaper do you want your mug warmers to be? If the only option was huge rigs then I'd agree but there will always be the inexpensive options not to mention those still getting free electricity, better video cards, and improved FPGA designs and hardware. I don't get your "big spenders" idea with ASICS I can get in for half the price of GPUs? My guess is the ease of use and relative cheapness of these new ASICS will increase adoption amongst the general population.

Even ASICS will have risk, the plug-able engine might be upgraded or precision increased (which, by the way, from Zeno's paradox will make mining never really end), it is an open source project after all.

Thing is, if someone made a big investment in GPUs recently, it will likely never pay for itself. I thought that looked quite likely for a while, since newer cards don't improve on the 5-series cost/hash-rate wise. When BTC was at 5$ it was pretty obvious that you'd rather buy coins than cards. Now it can be tight, but ASICs or other tech can always turn up soon.

As with everything, don't bet more than you can afford to lose.