Life after ASIC

[skomo]

Sorry for newbie question, but I guess they are allowed here..

So what comes next after the current generation of ASIC? Are there any new technologies in the development to make them obsolete?

[1714874145]

1714874145

[J35st3r]

Plenty.

The current ASICs are based on very old technology as the up-front tooling costs (mask manufacture) were much cheaper than state of the art processes. Avalon uses a 110nm process, BFL uses 65nm process (hence the Avalons using far more chips than BFL for the same performance, though probably a good move as Avalon got it right the first time, while BFL's development was a series of disasters).

Given the success of ASICs (to the manufacturers anyway), there is a huge incentive to build on more state of the art processes, which will be faster and lower power comsumption. AFAIK the current commodity foundries are at around 23nm, and Moore's law still has a few generations left to go yet before it hits the endpoint (transistors don't work once you're down to a few dozen atoms in width).

So there is going to be lots more excitement to come (both success and failure) over next next few years, and that's not even allowing for black swan's like quantum computing. Stock up on the popcorn!

[skomo]

Thanks for informative reply!

So mining business has very negative oultlook, as only few that will have early access to the new technology can profit from it, or if you have free electricity? It's kind of self-destructing that way, as it will become very concentrated rather then de-centralized.

[J35st3r]

So mining business has very negative oultlook, as only few that will have early access to the new technology can profit from it, or if you have free electricity? It's kind of self-destructing that way, as it will become very concentrated rather then de-centralized.

Its very difficult to guess. I've been reading over at the Mining Speculation threads https://bitcointalk.org/index.php?board=81.0 and there are many different opinions, and that's just predicting the next few months, let alone years. But one fact is that ASICMINER are currently a full 25% of the mining hash rate, so yes, the big boys would seem to be taking over. But to balance that there is a huge backlog of ASIC orders from general punters which should start to be delivered over the next few months.

As for what happens when the difficulty rises ten fold (as is commomly predicted for the end of the year), its going to cause a shake down of the current GPU miners, and all but the earliest ASIC pre-orders are looking at making a loss in BTC terms. Then next-gen ASIC will start to pre-order and the treadmill will start to spin again.

Interesting times (and my bedtime, so no more from me today).

[zendantom]

Avalon uses a 110nm process, BFL uses65nm process

How it comes the Watt/GHash of the final product is not much better with 65nm process. Doesnt it suggests the future 23nm ASIC miners would not be much more power effecient thus today ASIC will still compete fine with future ASIC?

[Salwilliam]

There is going to be lots more excitement to come (both success and failure) over next next few years, and that's not even allowing for black swan's like quantum computing. Stock up on the popcorn!

Could quantum computing crack SHA2 and ruin Bitcoin?

[Alexander The Great]

Lol I love this title "life after ASIC" Brilliant

[Frizi]

There is going to be lots more excitement to come (both success and failure) over next next few years, and that's not even allowing for black swan's like quantum computing. Stock up on the popcorn!

Could quantum computing crack SHA2 and ruin Bitcoin?

Hashing is hard on quantumn too, but there are other possibilities. However, the Bitcoin network is flexible and can change its algorithms anytime.
Read more: https://bitcointalk.org/index.php?topic=153302.0

[Dasneko]

ASIC is very much the end of the line for the foreseeable future. You can argue that the ASIC chips we have on the market is based on old tech but the ASIC tech itself is about as advanced as you can get right now with mining.

This recent scare about quantum computing is also unfounded and based on a processor utilizing a small quantum "quirk" not an actual quantum computer. It would be a million dollar machine with the power of perhaps 10 graphics cards if used for mining bitcoins.

Now an actual quantum computer that can magically pull the right answer to a math problem out of its arse might seam formidable but if my limited knowledge about Bitcoins is correct then all that would accomplish is cut off a few nano seconds or something. Perhaps it would even lose big time to a graphics card. Its simply the wrong kind of math problem where a quantum computer would be useful.

Even if they got a majority share of the processor power and initiated a "51% attack" the actual things it would allow them to do is jack shit compared to what it would cost them. Its just not worth it.

[J35st3r]

How it comes the Watt/GHash of the final product is not much better with 65nm process. Doesnt it suggests the future 23nm ASIC miners would not be much more power effecient thus today ASIC will still compete fine with future ASIC?

Good question. BFL were bragging that their technology was so much more advanced than "the competition", then when they finally got working silicon (after several expensive design iterations!!), they discovered that it was far more power hungry than their estimations, and had to redesign their boards (and whole product range, note that they are currently only shipping their smallest Jalapinos). Just too much heat being generated inside their cute little boxes.

As for why? Well a like-for-like comparison for the same hashing architecture on the 65nm vs 110nm process should give a significant efficiency gain (I've been out of this biz for a long time now, so I won't hazard an exact figure). So it looks to me that Avalon did a better design (its Ngzhang after all, who did the Icarus and Lancelot FPGAs, he's good).

As for 23nm, that will be a while off (if ever) as the mask costs are far more expensive. Expect next gen to be 65nm (possibly with improved architecture), then maybe 45nm if BTC is still with us by then.

PS I forgot kncminer. They are using a 28nm process but its all still vaporware at the moment. They have takena very few pre-orders for September delivery (they are using a mask sharing scheme to save on up-front costs, so their initial throughput is very limited). We'll just have to see what comes of it.

[skomo]

So does it make sense today to get on one of the Avalon chips group buys? Or ASIC miner shares are better alternative to see how difficulty plays out in the next 6-9 months?

[J35st3r]

So does it make sense today to get on one of the Avalon chips group buys? Or ASIC miner shares are better alternative to see how difficulty plays out in the next 6-9 months?

Your guess is as good as mine. But you're very late to the party, so if you're buying ASIC hardware to mine you are going to be way down the queue for shipment which will make a huge difference to whether you make a profit or not. As for ASICMINER shares, again its late so you take a chance that they are grossly overpriced for the expected return (I haven't researched this, so don't take my word for it, and all the standard disclaimers about advice on shares apply).

IMHO its all a big gamble, but if you're game then do the math first. From an earlier post ...

This seems like a useful overview https://bitcointalk.org/index.php?topic=209750.0

And the calculator http://coinish.com/calc will help you do the math. Use expert mode, and your crystal ball to predict the likely difficulty rise, these threads may help https://bitcointalk.org/index.php?board=81.0

Good Luck. Its all beyond my ken so I'm staying out for now.

[Eva Braun]

Sounds like "Life after death"

[ArtificiallySmart]

For the sake of the people who are now waiting to receive their ASICS, let's hope the newer generation ASICS Won't be developed very soon.

[Welsh]

For the sake of the people who are now waiting to receive their ASICS, let's hope the newer generation ASICS Won't be developed very soon.

Yeah I know what you mean there. But, I can see newer hardware being created sooner, rather than later. Bitcoin is limited to 21 million. so we are closing the gasp quicker than before with ASIC. Therefore companies will want to get their product out fast. In order to make the most of their money. Before mining pretty much becomes too difficult.

[J35st3r]

Bitcoin is limited to 21 million. so we are closing the gasp quicker than before with ASIC.

Bitcoin Comprehension Fail (and from a hero member too!!). The network averages one block every ten minutes, regardless of hashrate.