ETH GPUs miners beware!

[keystoneminer]

Deep Learning Accelerating Card SC1
Pre-Batch of Cyber Monday Nov. 8 ~ Nov. 18, each user can buy up to 10 units of SC1. You will get the same amount coupons equalling to the quantity of SC1s purchased successfully at Cyber Monday (each coupon valued 50% of the price of SC1). These coupons can be used to purchase our next generation DL card SC3. SC3 will be equipped with BM1682, and scheduled to ship in 2018 Q2!

$589.00

Weight: 0.613 kg

Quantity:-  
1
 +
Sold out
While Sophon SC1 features abundant functions supporting multiple application scenarios, due to the large number of design-in customers, we may not have enough resources to provide further technical support for the standard features beyond those we announce. Please purchase with caution. Keep tune, looking forward to the upcoming new features!

*** anyone have any info on the above or have one on order ?

[1714632218]

1714632218

[1714632218]

1714632218

[1714632218]

1714632218

[Squintz]

16x mining rig for sale 1070ti we're about to wiped off the earth

[rdluffy]

But, they will produce an Asic to a coin that will move to POS soon?

[kukelka]

so it means that it's not worth to start buying gpus and building rigs?

[bitcoinman93]

There are rumours that Bitmain has set its sights on ETH and will be releasing an ASIC in Q2 or Q3 of this year!

https://www.cryptoinfomag.com/2018/02/12/eth-asic-arriving-soon/

Just rumors for now, but thanks for sharing anyway as it's something that might happen. It would really shake whole eth mining community. Any chance of some real info from bitmain?

[StGermain]

I think. The main developers of Ethereum are very well prepared for such a thing. And it would be very crazy to launch a dedicated machine at this point in which the plasma protocol on the proof of stake are about to be implemented

[xenomorphe1]

I think. The main developers of Ethereum are very well prepared for such a thing. And it would be very crazy to launch a dedicated machine at this point in which the plasma protocol on the proof of stake are about to be implemented

They are going to mine ETC or other Ethash algos coins if ETH is only POS.

[d57heinz]

I think. The main developers of Ethereum are very well prepared for such a thing. And it would be very crazy to launch a dedicated machine at this point in which the plasma protocol on the proof of stake are about to be implemented

They are going to mine ETC or other Ethash algos coins if ETH is only POS.

Yes looking like a good time to buy up some etc🤓🤣.  But seriously even looking at goin hybrid this makes sense. Rewards will drop during the conversion forcing use of asic to keep chain alive or price will have  to rise according to percentage of pow/pos.  Going to be an interesting transition if it ever happens. 

BR

[xenomorphe1]

ETH/ETC are going to be more centralized... The concept of decentralization is going to die?

[vx0713]

So when F3 is launched it will be the most sought after Asic and like A3 will give 500$ per day for early adopters and will gradually come down to 30$ per day.

Please stop spreading bullshit, and do your math. For the F3 to give you $500 it will have to hash at a 6000 Mh/s at current difficulty (and I'm not even tanking into consideration the cost of electricity).
Even if the news is true there is no effin way it can pull this off...

 http://whattomine.com/coins/151-eth-ethash?utf8=%E2%9C%93&hr=6000&p=405.0&fee=0.0&cost=0.0&hcost=0.0&commit=Calculate

[Travis726]

OH No, this is gonna be bad, what will happen to GPU miners it will go down to algo and the current of price? i've heard also 20 series nvia will released not compatible for the mining holy crap.

Lol on the 20 series. Nvidia answers to shareholders.  They talk like crypto wasn’t the bulk of sales but that’s bs.  They will work fine for crypto or watch their share price and sales tumble.  I honestly don’t believe this one bit.

BR

nvidia would be just fine without miners.. I say would because Im not sure I believe they will somehow make their cards not work at mining without killing their performance at the same time.

So when F3 is launched it will be the most sought after Asic and like A3 will give 500$ per day for early adopters and will gradually come down to 30$ per day.

Please stop spreading bullshit, and do your math. For the F3 to give you $500 it will have to hash at a 6000 Mh/s at current difficulty (and I'm not even tanking into consideration the cost of electricity).
Even if the news is true there is no effin way it can pull this off...

 http://whattomine.com/coins/151-eth-ethash?utf8=%E2%9C%93&hr=6000&p=405.0&fee=0.0&cost=0.0&hcost=0.0&commit=Calculate

he is basing his numbers on sia (and the recent bitmain miners performance) and nothing else...

[pickleburglar]

Did some quick maths

ASSUMING that they use quad channel interface for each chip with the fastest ddr3 in existence and its 96GB not 72GB, that would make the total memory bandwidth of the rig 1TB/s which is 4 times faster than a RX580 or around 120MH/s. That would be the worst case scenario.

[jstefanop]

Did some quick maths

ASSUMING that they use quad channel interface for each chip with the fastest ddr3 in existence and its 96GB not 72GB, that would make the total memory bandwidth of the rig 1TB/s which is 4 times faster than a RX580 or around 120MH/s. That would be the worst case scenario.

Thats incorrect, if they are using quad channel spread to 8 16bit 512MB DDR3 chips for each core thats ~14mh per core (assuming around a 1GHZ ram speed, some DDR3 can go up to 1600mhz. 14x 6 cores x 3 boards ~ 250 MH.

[passteque]

Asic can just be more efficicent, no need to have supérior hashrate to call it an Asic. Look skein asic, they are 20% costlier than GPU. But use 5 times less electricity

[drawingthemoon]

If that is true the GPU rigs and GPU will be back to normal price finally. But I can not imagine that this is true. is anybody here who understand if it is technically possible? Everytime interesting how the ASIC resistant coins get an ASIC miner from bitmain

[xleps81]

Thats the worst thing to happen to ETH minning...

Just bad news this 2018....

What will you mine?

[Johny101010]

ASIC usually raise the price of a coin. So I'm ok with it.
But, what I don't understand is why would you buy an Eth ASIC when they are about to go POS?
Once ETH does it other coins based on the same algorithm will too. And is not like you can mine Cryptonight , Equihash or any other coins with that machine. You just have to wait for new coins and hope they are the next big thing.

[brewlig]

ASIC usually raise the price of a coin. So I'm ok with it.
But, what I don't understand is why would you buy an Eth ASIC when they are about to go POS?
Once ETH does it other coins based on the same algorithm will too. And is not like you can mine Cryptonight , Equihash or any other coins with that machine. You just have to wait for new coins and hope they are the next big thing.

When is it planned to move to PoS?  

[QuintLeo]

Asic can just be more efficicent, no need to have supérior hashrate to call it an Asic. Look skein asic, they are 20% costlier than GPU. But use 5 times less electricity

 The only Skein ASIC I am aware (Baikal) of has 2.5-3 TIMES the performance of the 1080 ti - and Skein is one of it's "minor" algorithms that it gets lower hashrate on.

 ASIC do NOT "usually raise the price of a coin" - they are normally developed when a coin has a high enough total market cap AND network hashrate to make it possible for an ASIC miner company to sell enough to PAY for the development of the things.
 You have the relationship backwards - it's high PRICE that drives creation of an ASIC, NOT the other way around - and in the case of the Gridseed GC3355 based ASIC gear the ASIC showed up after Litecoin price had COLLAPSED and Bitcoin price was dropping.

[Chromexnet]

Did some quick maths

ASSUMING that they use quad channel interface for each chip with the fastest ddr3 in existence and its 96GB not 72GB, that would make the total memory bandwidth of the rig 1TB/s which is 4 times faster than a RX580 or around 120MH/s. That would be the worst case scenario.

Thats incorrect, if they are using quad channel spread to 8 16bit 512MB DDR3 chips for each core thats ~14mh per core (assuming around a 1GHZ ram speed, some DDR3 can go up to 1600mhz. 14x 6 cores x 3 boards ~ 250 MH.

How did you calculate 14 MH/s per core? 14 MH/s would require roughly 140 GB/s bandwidth which seems unrealistic even for a quad channel DDR3. A quad channel 1866MHz DDR3 can reach 59.7 GB/s (source https://www.microway.com/knowledge-center-articles/performance-characteristics-of-common-transports-buses/). Therefore one core hashrate should be close to 6 MH/s. 6 MH/s * 6 * 3 = 108 MH/s.