Bitmain's Released Antminer S9, World's First 16nm Miner Ready to Order

[capaneo]

since bitmain opened the last batch it hasnt sold out... either they have increased production by a lot, or there is some backroom information that the big companies arent buying any letting the small ones take the hit, considering there are news about smaller chips and samsung coming to the game and a lot of people with big farms selling all their S9's; brings me some fear (since i got a order open for march) that they are either coming with a new miner or some other big news is coming out, anyone else with these thoughts ?

I just signed up to post this here. I think the math doesn't add up anymore for 16nm chips. Unless you are doing liquid cooling and even that is almost breakeven.

https://medium.com/@sisa83/bitcoin-mining-heading-towards-a-cliff-354961c73c26

[The Demon Slick]

Is this for real?

http://news.8btc.com/bitmain-to-release-ethereum-asic-miner-f3-with-72gb-ddr3-dram-in-q2

I have no idea if true or not but it would not shock me.

Consider:
- Bitmain recently releasing previously unannounced blake miner with shipping within 10 days
- Competition heating up in sha256 mining space so bitmain looking for new markets
- Bitmain seems to be rolling out miners for new algos in blitzkrieg fashon
- Would fit pattern of bitmain building asic chips for gpu algos
- threats from dev teams to change hashing algos or switch to Proof of Stake does not deter them

Where it ends I do not know.

Actually, I do hope Bitmain release that ETH ASIC, as it would be a total game changer.
It is totally not normal to pump for years the GPU production business, the stocks of ATI and NVIDIA skyrocket; on the other hand gaming business is declining as there are literally no video cards to be bought to play the latest top notch games.
Today I also read this article on BBC News (seems legit) which is even more sad if it is true - lack of GPU hinders space research.
https://twitter.com/coindesk/status/964039175686311938

So - "Go Bitmain"...

Nvidia cries in public about the miners buying up all the cards, but in private they high five each other and count their money.

[kobo_eth]

since bitmain opened the last batch it hasnt sold out... either they have increased production by a lot, or there is some backroom information that the big companies arent buying any letting the small ones take the hit, considering there are news about smaller chips and samsung coming to the game and a lot of people with big farms selling all their S9's; brings me some fear (since i got a order open for march) that they are either coming with a new miner or some other big news is coming out, anyone else with these thoughts ?

I just signed up to post this here. I think the math doesn't add up anymore for 16nm chips. Unless you are doing liquid cooling and even that is almost breakeven.

https://medium.com/@sisa83/bitcoin-mining-heading-towards-a-cliff-354961c73c26

Wow, interesting article that mathematically proves bitcoin mining to be useless already (or in the very near times to come). It will be interesting to hear the other guys comments here.

[castiel0504]

I'm feeling like there should to come new "S9", like there was jump from S7 to S9...4th > 13-14TH... S9 is becoming obsolete..Unless you have free electricity...Or Bitcoin jumps...But you can't just wait for it to jump...Difficulty is increasing enormous speed...And new Dragonmint...is ONLY 2TH better then S9...We need crazy jump in tech again

[capaneo]

I'm feeling like there should to come new "S9", like there was jump from S7 to S9...4th > 13-14TH... S9 is becoming obsolete..Unless you have free electricity...Or Bitcoin jumps...But you can't just wait for it to jump...Difficulty is increasing enormous speed...And new Dragonmint...is ONLY 2TH better then S9...We need crazy jump in tech again

There won't be a big jump anytime soon. The reason we had a big jump form S7 to S9 was that S7 was using 28nm chipset in 2015. And in 2015 we already had commercial production of even up to 14nm. But right now S9 is using 16nm which is pretty much the bottleneck. The next step is 10nm and that is barely getting into commercial production today and when it does the foundries will be booked by giants like Apple and Samsung for some time. So S9 is here to stay for at-least another year or two.

you can increase the hashrate by liquid cooling, better design, etc etc. But won't be anything like the S7 to S9

[philipma1957]

since bitmain opened the last batch it hasnt sold out... either they have hat  production by a lot, or there is some backroom information that the big companies arent buying any letting the small ones take the hit, considering there are news about smaller artichips and samsung coming to the game and a lot of people with big farms selling all their S9's; brings me some fear (since i got a order open for march) that they are either coming with a new miner or some other big news is coming out, anyone else with these thoughts ?

I just signed up to post this here. I think the math doesn't add up anymore for 16nm chips. Unless you are doing liquid cooling and even that is almost breakeven.

https://medium.com/@sisa83/bitcoin-mining-heading-towards-a-cliff-354961c73c26

Wow, interesting article that mathematically proves bitcoin mining to be useless already (or in the very near times to come). It will be interesting to hear the other guys comments here.

If you believe that article mathematically proves bitcoin mining to be useless right now you do not understand what a math proof is.  

Trust me the article shows a possible outcome that may be true or may be false. No more or less then that.


It based all calculation on a 10k btc.

And we are now over 11 k.
So it is already wrong.

Of course in the morning price may be 12k

Or 9 k

Or back to 10k

It also assumes 13% will continue well it may or may not .

But if we stay at 10 k price with 13% diff per month we will bust.

That could happen. That is possible. But the closer we get to busting out. Growth will go flat.

I can run 13kwatts 24/7/365 in a 1.2 acre solar array that has been paid off.

Yes I would earn less btc if diff goes up up and away but if I want I can use gpus and mine other coins.

I am not the only miner with gear and miners paid off.

[Raymond_B]

I have a prediction for all the naysayers. Let posit that BTC difficulty and price come to where there is basically no profit to be had. Some folks will still mine with free or almost free electricity, like Phil. But I would imagine with all this hardware out there that it would not be much more than just a firmware flash away to mine something else. Maybe that's SHA256 based or? Who knows, innovation is always driven by these kinds of events.

That's not to say buying this equipment isn't a risk. But I imagine we'll always find a use for it.

[leowonderful]

I have a prediction for all the naysayers. Let posit that BTC difficulty and price come to where there is basically no profit to be had. Some folks will still mine with free or almost free electricity, like Phil. But I would imagine with all this hardware out there that it would not be much more than just a firmware flash away to mine something else. Maybe that's SHA256 based or? Who knows, innovation is always driven by these kinds of events.

That's not to say buying this equipment isn't a risk. But I imagine we'll always find a use for it.

What do you mean? These machines can only mine SHA256 with the ASICs they are outfitted with (as ASICs are built to do one thing), and they can only hash SHA256 as a result and nothing else. Flashing them to mine something other than a SHA coin wouldn't work. Flashing for more algos only worked on the Baikal alt miners a while back as they were FPGA based iirc.

[Raymond_B]

I have a prediction for all the naysayers. Let posit that BTC difficulty and price come to where there is basically no profit to be had. Some folks will still mine with free or almost free electricity, like Phil. But I would imagine with all this hardware out there that it would not be much more than just a firmware flash away to mine something else. Maybe that's SHA256 based or? Who knows, innovation is always driven by these kinds of events.

That's not to say buying this equipment isn't a risk. But I imagine we'll always find a use for it.

What do you mean? These machines can only mine SHA256 with the ASICs they are outfitted with (as ASICs are built to do one thing), and they can only hash SHA256 as a result and nothing else. Flashing them to mine something other than a SHA coin wouldn't work. Flashing for more algos only worked on the Baikal alt miners a while back as they were FPGA based iirc.

Maybe, also maybe someone else comes out with another SHA256 coin, or maybe something we haven't thought of. Just never say never...

[leowonderful]

I have a prediction for all the naysayers. Let posit that BTC difficulty and price come to where there is basically no profit to be had. Some folks will still mine with free or almost free electricity, like Phil. But I would imagine with all this hardware out there that it would not be much more than just a firmware flash away to mine something else. Maybe that's SHA256 based or? Who knows, innovation is always driven by these kinds of events.

That's not to say buying this equipment isn't a risk. But I imagine we'll always find a use for it.

What do you mean? These machines can only mine SHA256 with the ASICs they are outfitted with (as ASICs are built to do one thing), and they can only hash SHA256 as a result and nothing else. Flashing them to mine something other than a SHA coin wouldn't work. Flashing for more algos only worked on the Baikal alt miners a while back as they were FPGA based iirc.

Maybe, also maybe someone else comes out with another SHA256 coin, or maybe something we haven't thought of. Just never say never...

I just find something like that exceptionally unlikely. If there were to be another SHA256 coin that comes out (happens very often), it'll likely be mineable by regular SHA ASICs, and there are plenty of algos besides SHA out there that are more popular for new coins. If you're looking into buying an S9 I wouldn't count on a miracle like this happening though, always plan for the worst.

[Raymond_B]

I have a prediction for all the naysayers. Let posit that BTC difficulty and price come to where there is basically no profit to be had. Some folks will still mine with free or almost free electricity, like Phil. But I would imagine with all this hardware out there that it would not be much more than just a firmware flash away to mine something else. Maybe that's SHA256 based or? Who knows, innovation is always driven by these kinds of events.

That's not to say buying this equipment isn't a risk. But I imagine we'll always find a use for it.

What do you mean? These machines can only mine SHA256 with the ASICs they are outfitted with (as ASICs are built to do one thing), and they can only hash SHA256 as a result and nothing else. Flashing them to mine something other than a SHA coin wouldn't work. Flashing for more algos only worked on the Baikal alt miners a while back as they were FPGA based iirc.

Maybe, also maybe someone else comes out with another SHA256 coin, or maybe something we haven't thought of. Just never say never...

I just find something like that exceptionally unlikely. If there were to be another SHA256 coin that comes out (happens very often), it'll likely be mineable by regular SHA ASICs, and there are plenty of algos besides SHA out there that are more popular for new coins. If you're looking into buying an S9 I wouldn't count on a miracle like this happening though, always plan for the worst.

Yes, thank you. I am not naive, just optimistic.

[halfmil]

I have a prediction for all the naysayers. Let posit that BTC difficulty and price come to where there is basically no profit to be had. Some folks will still mine with free or almost free electricity, like Phil. But I would imagine with all this hardware out there that it would not be much more than just a firmware flash away to mine something else. Maybe that's SHA256 based or? Who knows, innovation is always driven by these kinds of events.

That's not to say buying this equipment isn't a risk. But I imagine we'll always find a use for it.

What do you mean? These machines can only mine SHA256 with the ASICs they are outfitted with (as ASICs are built to do one thing), and they can only hash SHA256 as a result and nothing else. Flashing them to mine something other than a SHA coin wouldn't work. Flashing for more algos only worked on the Baikal alt miners a while back as they were FPGA based iirc.

Maybe, also maybe someone else comes out with another SHA256 coin, or maybe something we haven't thought of. Just never say never...

I think the litecoin fork they are trying to do with litecoin cash is going to be SHA256 Algo.

[lossol]

I read the article. And it's 13% per month difficulty is way off. The article stated difficulty doubling every 5-6 months. But in actuality, it more than doubled in under 3 months! I used the mining calculators in the last week of Nov. 2017. I took a snapshot of how much bitcoin an S9 makes and it was only 0.92 bitcoin back then. Today an S9 only makes 0.43 bitcoins. Regardless of the price of bitcoin, that's more than a 50% drop in under 3 months. If this continues an S9 won't make much bitcoins, but it can still be profitable if the price of bitcoin moves up exponentially. If the price stagnates, you can forget about making a profit even with so called "free electricity". If difficulty continues to shoot up exponentially, bitcoin mining will go the way of dash mining! So all cryptos designed this way will become unprofitable to mine eventually, once difficulty gets high enough.

[philipma1957]

I read the article. And it's 13% per month difficulty is way off. The article stated difficulty doubling every 5-6 months. But in actuality, it more than doubled in under 3 months! I used the mining calculators in the last week of Nov. 2017. I took a snapshot of how much bitcoin an S9 makes and it was only 0.92 bitcoin back then. Today an S9 only makes 0.43 bitcoins. Regardless of the price of bitcoin, that's more than a 50% drop in under 3 months. If this continues an S9 won't make much bitcoins, but it can still be profitable if the price of bitcoin moves up exponentially. If the price stagnates, you can forget about making a profit even with so called "free electricity". If difficulty continues to shoot up exponentially, bitcoin mining will go the way of dash mining! So all cryptos designed this way will become unprofitable to mine eventually, once difficulty gets high enough.

Yeah 

price rise > diff rise s-9 is great
price rise = diff rise  s-9  is good
price rise < diff rise  s-9 is bad

the   ratio varies

diff rise  for 2017  336,899,000,000 to 1,931,136,000,000 = 5.73 to 1
diff rise for 2016   113,354,000,000 to    336,899,000,000    = 2.97 to 1
diff rise for 2015     43,971,000,000 to    113,354,000,000   = 2.57 to 1
diff rise for 2014      1,418,000,000 to       43,971,000,000   = 31.00 to 1   damn
diff rise for 2013            3,249,000  to        1,418,000,000   = 436.44 to 1  damn
diff rise for 2012
diff rise for 2011
diff rise for 2010
diff rise for 2009

[HoleShot]

price rise > diff rise s-9 is great
price rise = diff rise  s-9  is good
price rise < diff rise  s-9 is bad

the   ratio varies

diff rise  for 2017  336,899,000,000 to 1,931,136,000,000 = 5.73 to 1
diff rise for 2016   113,354,000,000 to    336,899,000,000    = 2.97 to 1
diff rise for 2015     43,971,000,000 to    113,354,000,000   = 2.57 to 1
diff rise for 2014      1,418,000,000 to       43,971,000,000   = 31.00 to 1   damn
diff rise for 2013            3,249,000  to        1,418,000,000   = 436.44 to 1  damn
diff rise for 2012
diff rise for 2011
diff rise for 2010
diff rise for 2009

This is perfect. Thanks Phil. I had something similar to this in a spreadsheet.

i think the main reason for the massive hash rate increases were due to enterprise class ASIC centers going in worldwide. Of course every joe blow setting up his garage contributed some.

[The Demon Slick]

I'm feeling like there should to come new "S9", like there was jump from S7 to S9...4th > 13-14TH... S9 is becoming obsolete..Unless you have free electricity...Or Bitcoin jumps...But you can't just wait for it to jump...Difficulty is increasing enormous speed...And new Dragonmint...is ONLY 2TH better then S9...We need crazy jump in tech again

There won't be a big jump anytime soon. The reason we had a big jump form S7 to S9 was that S7 was using 28nm chipset in 2015. And in 2015 we already had commercial production of even up to 14nm. But right now S9 is using 16nm which is pretty much the bottleneck. The next step is 10nm and that is barely getting into commercial production today and when it does the foundries will be booked by giants like Apple and Samsung for some time. So S9 is here to stay for at-least another year or two.

you can increase the hashrate by liquid cooling, better design, etc etc. But won't be anything like the S7 to S9

This is incorrect. The ebang ebit has already been delivered, it's using 10nm chips from Samsung, but it's still only an 18T/Hs machine for around 5k USD, so it's not really the game changer I expected it to be.

[capaneo]

I'm feeling like there should to come new "S9", like there was jump from S7 to S9...4th > 13-14TH... S9 is becoming obsolete..Unless you have free electricity...Or Bitcoin jumps...But you can't just wait for it to jump...Difficulty is increasing enormous speed...And new Dragonmint...is ONLY 2TH better then S9...We need crazy jump in tech again

There won't be a big jump anytime soon. The reason we had a big jump form S7 to S9 was that S7 was using 28nm chipset in 2015. And in 2015 we already had commercial production of even up to 14nm. But right now S9 is using 16nm which is pretty much the bottleneck. The next step is 10nm and that is barely getting into commercial production today and when it does the foundries will be booked by giants like Apple and Samsung for some time. So S9 is here to stay for at-least another year or two.

you can increase the hashrate by liquid cooling, better design, etc etc. But won't be anything like the S7 to S9

This is incorrect. The ebang ebit has already been delivered, it's using 10nm chips from Samsung, but it's still only an 18T/Hs machine for around 5k USD, so it's not really the game changer I expected it to be.

Those are mostly just marketing BS. In terms of energy, the ebang machine is using .09J/GH +/- 10%. Which is only a %10 improvement over S9s. So the improvement is within their margin of error.

But even if we give them the benefit of the doubt. When you solve the equations for the ebang machine, at an optimistic rate, it will only earn 0.45BTC during its entire lifetime. (As oppposed to S9s 0.37BTC). That is the entire lifetime of the machine. Assuming the machine is operating with free electricity and is running until end of time. In reality the lifetime earning will be much less.

[kodokbuduk]

Just thinking about mining Litecoin Cash. How much we will get with this S9 13.5th?

[halfmil]

Just thinking about mining Litecoin Cash. How much we will get with this S9 13.5th?

I have been following this fork and it looks interesting. Supposedly it was created to give old bitcoin miners like s7's something to mine that would be profitable. Since the fork happened 2 days ago it has been mooning to over $9 as of now. But wallets are not available yet and therefore it can't be mined yet. But wallets should be ready shortly. There are only a couple of exchanges that support it at the moment like yobit and the only people that were able to sell it were litecoin holders on those exchanges when it forked and had LCC dropped into their accounts. I imagine when everyone is able to access their lcc and sell it the price will come down quite a bit. I'm just going off of memory from a form I read but I want to say someone calculated that they would be able to mine 26 LCC per day with about 60th but when everyone is able to mine the diff will jump a lot and price may come down a lot. Probably like the A3 did with siacoin. But it might be a good bet to mine it right now at least for a little bit.  There's a thread on it here  

https://bitcointalk.org/index.php?topic=2876538.0

[leowonderful]

Just thinking about mining Litecoin Cash. How much we will get with this S9 13.5th?

I have been following this fork and it looks interesting. Supposedly it was created to give old bitcoin miners like s7's something to mine that would be profitable. Since the fork happened 2 days ago it has been mooning to over $9 as of now. But wallets are not available yet and therefore it can't be mined yet. But wallets should be ready shortly. There are only a couple of exchanges that support it at the moment like yobit and the only people that were able to sell it were litecoin holders on those exchanges when it forked and had LCC dropped into their accounts. I imagine when everyone is able to access their lcc and sell it the price will come down quite a bit. I'm just going off of memory from a form I read but I want to say someone calculated that they would be able to mine 26 LCC per day with about 60th but when everyone is able to mine the diff will jump a lot and price may come down a lot. Probably like the A3 did with siacoin. But it might be a good bet to mine it right now at least for a little bit.  There's a thread on it here  

https://bitcointalk.org/index.php?topic=2876538.0

If it becomes possible for old miners to mine this coin then the chain will become more profitable than BTC, which'll just result in people pouring into Litecoin Cash and profitability becoming even once again all over the board. Unless you're early to the show I wouldn't expect much.