Topic: Would i be better off waiting for the 16nm miners? (Read 3671 times)

It's hard to say if bad or good.  We really need to know release date for general public.  And if they lower quantity from 333 units.

We also need an idea of difficulty when miner is released.  Just still a lot of unknowns.

it appear to be the same for me

we have 3tera vs 4.8 tera, let's say 50% more

and 800w vs 1200w, again 50% more

the only real difference is that their price are 2.61 vs 7.7 and 7.7 isn't 50% more, so ironically this 14nm asic is not better because more efficent but because it cost less...

so a 3th miner will drains 780w and costs 2.61btc, not bad, cheaper than the s7, now it costs 6.15btc, or 1.27btc/th...

so a 3th miner will drains 780w and costs 2.61btc, not bad, cheaper than the s7, now it costs 6.15btc, or 1.27btc/th...

~

Compare to original Chinese verion of the news, it's 0.87B/T

http://www.8btc.com/bw-14nm-chip

I don't see it being a stick math below:

1PH = 1000TH
1000TH / 333 Units = around 3 TH  per unit.   The best stick miner out there does around 16 GH (the compac).  There is just not a way they fit 3T (like 300 GH less then 3 S5's) on a thumb drive.

On price you could read it a few way's but i read 0.87 BTC per unit.  So:

333 X .87 BTC = 289.71 BTC or around 76773.15 USD or 1 PH.  Which sounds pretty reasonable.   But again anyone feel free to check my work.

It seems will be a war of new machines, the hashrate will be double in a while, the diff will rise and the earnings will drop deep into hell... Bitmain starts the war with new S7, and Spodoolies counter-atack with the SP50 release -but only bulk sales- and now BW launch this announcement, but more cheaper than others!!
Meanwhile, BITFURY is now the biggest pool with +-95 PH (quiet but running).

I think the bitcoin hardware is just began a war, but this time between big farms and manufacturers, and with this, the difficulty will be double in some months (and there will be the halving in july).  

Bad news to home miners, i think...

Also it is the 333 miners they talk about 1PH for that price. I mean really? They would have to be imho usb sticks even if you got 333 of them at that price.

yeah point to you .i inferred that it would be a usb stick with the values they can get from 14nm..but it would not have to be such.

I stand corrected. But I think it likely as a 'consumer product' at that speed and electric use.

Also it is the 333 miners they talk about 1PH for that price. I mean really? They would have to be imho usb sticks even if you got 333 of them at that price.
Still IMHO the price is PER miner. So a bulk order would be at 333 miners (sticks or whaever) approx at 231.00 usd $76,923.00 usd or so. The sSondoolies
SP50 data hall commercial product they likely will get out in a month or so if not sooner (or at least to their data hall) is 110TH for prob about 40 to 50k imho.

If these are usb sticks and my guess on the spondoolies sp50 110 TH unit is correct. Then we are looking at Winter 2016....so....anyway my view....

Just don't see it. Still see this as a USB stick gimmick to move a few and show off the flagship product 14nm chip to the masses in very small quanties.

The will use this as you state (chips) big massive (at cost) (cheap elec in china) data hall.What they sell here (imho usb sticks) for that price per miner
is simply to wave the corp flag.

Then again I really, really want to be wrong on this. But I doubt it

I might be missing something obvious but where do they announce a USB stick?  I see a big product they just want you to order a lot of them.

333 units = 1 PH sounds like a big miner not usb stick.  But I could me misreading something.

It is all gonna going to (or mostly) their data hall imho. The usb stick announce above is just to get the word out on their product. It is a gimmick. The big money will
be made internally with these chips (at cost) with cheap electric in china.

I can't imagine anyone pre-ordering gear at this point with just renderings.  And it's a crazy quantity to buy looks like they are shooting for data centers as first customers with the amount.

I am hoping we see sales a lot different after it launches and they fill their data centers up.

Here you go 14nm usb chip for the masses..you pay thru the nose for it....but a cute toy (and the joys it looks like of a pre-order raffle)

Info on BW company itself


https://bitcoinmagazine.com/articles/the-unknown-giant-a-first-look-inside-bw-one-of-china-s-oldest-and-largest-miners-1444675310

on the supposed 14nm product

https://bitcoinmagazine.com/articles/bw-to-launch-nm-chip-and-miner-for-general-population-1445278227



from the above

quote


Along with the announcement of its bitcoin chip, the company has announced that it will be releasing a bitcoin miner for consumers that will launch in Winter 2016.
“The miners are now available for pre-order at a price of 0.87 BTC with a minimum order of 333 miners which is 1 petahash,” Virgilio Lizardo Jr., Head of International at Bitbank told Bitcoin Magazine in an exclusive interview. “This price will be available until November 11, 2015. This winter, the miners will be available with no minimum order required; the price at this time has not been decided yet.”

According to BW, each 14nm chip in the miner will be able to attain anywhere from 34.6-63GH/s. The power consumption is 18W based on a voltage of 0.59V to 0.76V.

end quote

So the above says (assuming you could get one w/o bulk at the bulk price above) it would be as I type this $230.99 for a 34.6GH to 63.0GH USB Stick!

Sheesh. I can get any number of ASIC miners NOW ..electric be damned for a better return then that! Never ROI indeed!

But anyway another toy....but right now you could get 10 8gb usb stick Sidehack Gekko Miners (overclocked say to 10gh each) for 250 bucks would be 100gh NOW TODAY!
..so I sure as hell am not sure why the hell you would do the above on these BW sticks arriving in Winter 2016. The ROI is the same. A smile

but toys be toys I guess.

Sidehack sales thread here

https://bitcointalksearch.org/topic/run-2-closedsidehack-stickgekkoscience-compac-official-sales-thread-1126705 

I can't imagine anyone pre-ordering gear at this point with just renderings.  And it's a crazy quantity to buy looks like they are shooting for data centers as first customers with the amount.

I am hoping we see sales a lot different after it launches and they fill their data centers up.

 BW *IS* LK - in part.
 LK owns part of BW IIRC "Bitcoinbank" owns the rest.

 I'm curious where the claim that LK/BW are designing their own chip comes from - that would be a complete break with their previous business model, and they have NOT said or announced anything about such a thing that I've been able to find (though they also have never specifically said they are planning to use Innosilicon A3 chips).

LK announced LONG ago at this point.  Everyone assumed it was with Innosilicon as they have always been partners in past.  Also they said same NM as each other.

BW miner is kinda new and assumed with Innosilicon.  So no public info really on what's happening with LKETC original gear if still planned, or BW stole Innosilicon.

That is what I get out of it anyways.

Avalon is a project name. My company is Canaan-creative. Bitcoin mining is still my primary job.

LK / BW is totally different form Innosilicon. As I know, LK/BW is designing the chips under their own control.  

Good luck to their first chips.  

ng

you say soo many miners like i get fuzzed betewewn the S5 S6 S7 and the A ant miners are best the A or S and major difrences?

Thanks all for elaborating. Innosilicon's job on A1 project was to do the proper ASIC and and ship them to whoever wanna design the miners . By the way, we are free to ship in the open market from day 1 and we never shipped one region with better quality ASIC than another region. Any such claim is just fking lie and malicious. Why would we do that? Never! European region to us, are just as important as any other regions in the world. The two chip Reference board we did for A1 we did was only served to evaluate and demo our ASIC and not for production use. As it turned out, companies like Bitmine had very little prior engineering experience and they are not competent to handle the miner production in the end. All in all, engineering is all about knowledge, process and experience. Hard lessons are learnt and it is never easy to get-rich-quickly if you don't really know it.

BTW, Martin from IMET is a cheater and liar. He begged us to ship him a few thousand ASICs to make boards back in March 2014 and promised to pay later. After ASIC arrived, he disappeared. His A1 Miner was too extreme and too hot to be reliable.  He should not be trusted.

After A1 phase, we did 28nm A2 ASIC PLUS the official A2 Terminator miner design together so that there won't be another Bitmine like miner fiasco which hurt ASIC sales, which was better.

The latest development is that we are moving on to the A3 ASIC and A4 ASIC design already. Both in 14nm and already taped out with over 60% power saving over the 28nm generation. Anyone interested in those are free to contact us. I guess we are the 1st to move to 14nm at this point and we are confident A3 and A4 are both gonna work great.

Thanks all for the spirited debate and sharing. God bless!

Lk - https://bitcointalksearch.org/topic/lk-group-limited-will-start-bitcoin-miner-14nm-asic-project-1033676

BW  - https://www.bw.com/pool/miner

They teamed up with BW.   There are some big moves that happened.  Are you still with avalon?

Lk - https://bitcointalksearch.org/topic/lk-group-limited-will-start-bitcoin-miner-14nm-asic-project-1033676

BW  - https://www.bw.com/pool/miner

They teamed up with BW.   There are some big moves that happened.  Are you still with avalon?

sources?  

I have already decided to put my antminer U3 on for lottery mining, have named it blocky incase he finds me a block!... 63GH/s... I am interested in mining in a pool as well with a bigger rig as well... Would I be better off waiting for a later miner? I seem to find the antminer s7 is very high priced given the difficulty level keeps rising. I want to do it mainly for fun but also I do want to minimize losses if possible. Either getting a later rig or use that moment to cash in on the price drop of the s7?


What do you guys recommend? hang fire and wait? or go for it?

Jacob

Because they are the only company who have publicly announced consumer sales of 14nm mining chips with their A3's, unlike Bitmain and SP-Tech who are staying with 28nm for the time being.

Because they are the only company who have publicly announced consumer sales of 14nm mining chips with their A3's, unlike Bitmain and SP-Tech who are staying with 28nm for the time being.

Also, why you metioned Innosilicon? I haven't hear about them for really a long time. 

ng,

Thank you for coming in and commenting on many of the threads, it is nice to have industry insight from professionals such as yourself.

I personally believe that the "arms race" with Bitfury and KNC is more PR and marketing than real facts. All we really know for certain is that they are increasing their hashrate, not that they are meeting the specs that they claim. 

Would you agree that with the low price of electricity many of these large farms are likely able to obtain, the $/GH can be just as or more relevant as the W/GH in terms of profit? For that reason, I wouldn't think 14/16nm is going to be a cost-effective before the next ~12 months given it's infancy compared to 28nm at this time.  If it was, one would think Innosilicon would have taped out by now.

I think Bitcoin mining will finally move to 14/16nm at early or mid 2016, in the retail market.

The major barriers for us to move to 14/16nm node are the cost is ~X2 and the manufacture time is ~X2.5 compare to 28nm. This means you spend even more on same THs on the chips, and get them months later. Also, cash before order.

The only benefit is ~30-50% power saving. Do you really think this is worthy?

also, ALL bitcoin chip designs are cell based. This impossible to do a  good "full custom" design in 1or 2years. The word "Full custom" turns into a advertisement words now.

 

ng

My opinion is we wont see any 16nm until 2016 so if you want to wait that long to get into mining or you take the S7 which as of right now is the best option available to the public

droid, do you have a source for the comment that 16nm is not much more efficient than the 28nm BM1385.

All the sources I have seen say there is a considerable performance difference, especially with the 16 nanometer 3D FinFET chips. The target figures at the chip level being quoted by KNC and Bitfury are 0.07J/GHs and 0.06J/GHs respectively.  The BM1385 is 0.216J/GHs.  

Presumably only KNC knows the actual specs of the 16nm miners they seem to be rolling out.  The last 4 weeks have seen a 10-15 petahash  increase in the KNC pool.

In summary the 16nm chips could be between 2 and 3 times more efficient and likely around 0.1J/GHs at the system level.  But for the immediate future they will not be available to the public.

from what ive heard the 16nm chips are not that much more efficient than the 28nm fully custom BM1385

I'd wait not so much on 16nm, but on "what other companies come out with to compete with the S7" right now.

Bitfury and KnC aren't going to be selling to the public, and BitFury in particular isn't going to be selling 16NM at all for months at best (they just announced tape-out earlier this month, it takes MORE months sometimes a year+ to get working production hardware from that point).

which will be nice i can't wait!!! .  I think nothing but positive  thought for that. .

Any miner to be released in 2016 will be crossing over the halving around July in it's effective life. So I think price of miners are going down as we creep closer to the halving day as revenue from the miners will drop. I would not speculate price vs hash rate now because it's a completely new market in 6 months.

Agreed. I don't think there will be a 16nm miner in a while (some years, maybe). The develop cost is too much, and when it will be launched the diff will be launched too, so you will get no money. And next year, the reward will be only 12.5BTC per block, so i don't think we will see an early 16nm chip. Not in a long time. 

Agreed its basically an ASIC arms race each leap in efficency follows with a leap in difficulty to compensate so over all mining sooner rather then later in my onion is always the better option and forgetting the daily volatility of BTC the halving will come again and the rewards will once again be reduced and as the rest of the world jumps on board the value of BTC is poised to go up

16nm miners are still in fact quite far away from a mass release. When it gets released, if the efficiency is high, the miner would be sold out in a fast rate. Even without it, the difficulty would increase as the time goes and it wouldn't gain too much profit.

I have already decided to put my antminer U3 on for lottery mining, have named it blocky incase he finds me a block!... 63GH/s... I am interested in mining in a pool as well with a bigger rig as well... Would I be better off waiting for a later miner? I seem to find the antminer s7 is very high priced given the difficulty level keeps rising. I want to do it mainly for fun but also I do want to minimize losses if possible. Either getting a later rig or use that moment to cash in on the price drop of the s7?


What do you guys recommend? hang fire and wait? or go for it?

Jacob