[2018-02-12]Bitmain To Release Ethereum ASIC Miner F3 With 72Gb DDR3 Dram In Q2

[hl5460]

According to Technews, Bitmain is about to release the Ethereum ASIC miner F3 in Q2 or Q3 2018.

“Bitmain is about to release F3, the ethereum ASIC miner. It’s reported that every miner is mounted with 3 mainboards. On each mainboard there are 6 ASIC processors, each of which has 32 1GB DDR3 memory. Therefore, one unit of F3 miner contains 72 Gigabyte DRAM memory.”

I have doubts on the math of the report because the numbers doesn’t add up.But 72G DRAM seems reasonable and already outperforms most of the GPU-based mining rigs in the market now.
The bottleneck of Bitmain’s mass production is supply of Dram memory, which explains why Bitmain is frequent visiting storage supplier in Taiwan. The demand of Bitmain is around 3,000 pieces of 12 inch wafers per month. Ethereum ASIC chips will enter mass production stage in February 2018 by TSMC’s 28nm process. The F3 miner is expected to hit the market in Q2 or Q3 2018.
The ethhash power algorithm of Ethereum is designed to be ASIC proof. But just like Charlie Lee put it:

“I think ASICs are inevitable when the coin is valuable. But that’s not a bad thing.”

20180212132759

Bitmain has developed ASIC miner for Litecoin and Siacoin.


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

[ismail_btc]

R.I.P my rigs......

[clems]

If news is confirm it is the end of GPU mining for AMD.

All people mining ETH will go to Cryptonight and diff gonna explose...

[Avada77-Corporation]

rip eth   (edit : miner)

[hatshepsut93]

I though that developing Ethereum ASIC would be unprofitable because of their plans to switch to PoS, but it seems like Bitmain has calculated otherwise. Maybe it will be usable to mine other coins with the same algorithm. But since it won't be able to mine all GPU algorithms, GPU mining will still be around for a long time, so if this ASIC will take over ETH mining, current ETH miners will just switch to other coins.

[Ligareaux]

Planning a ETH Asic with POS plans?

[Pesmunar]

I though that developing Ethereum ASIC would be unprofitable because of their plans to switch to PoS, but it seems like Bitmain has calculated otherwise. Maybe it will be usable to mine other coins with the same algorithm. But since it won't be able to mine all GPU algorithms, GPU mining will still be around for a long time, so if this ASIC will take over ETH mining, current ETH miners will just switch to other coins.

The ETH PoW is designed to be ASIC resistant. So if there is an ASIC, it will change the algorithm.

[tbates76]

here is a copy of the article

China Bit mainland will launch the new ether ether mining machine, TSMC and memory plant will benefit
Author Atkinson | Posted on February 12, 2018 15:59 | Category Fintech , GPU , Processors

leed_case1-3-624x416
Science and Technology Daily News] [Technews encryption because although recent currency Bitcoin price breakouts, become the subject of investor sell-off, but it made money for the encryption of new business opportunities, related investment so far still in the ascendant. A few days ago, foreign media reported that Terry Gou, president of Hon Hai Group, will invest in the world's first commercial bank of cryptocurrencies. Recently, a new type of cryptocurrency mining machine will be launched soon. This type of mining machine will appear larger memory, lower energy consumption in the market, which will also bring more business opportunities for foundry leading TSMC and related memory manufacturers.




Bitmain in China is among the market leaders in the mining machine market. Not only in the global mining machine market share as high as 80% to 90%, the monthly net profit also reached 30 million US dollars, but also by virtue of its huge demand for mining machine market, to design bitcoin, Litecoin ASIC and other mining ASIC chip, making the bit mainland China has leapt to China's top five IC design firms, but also contribute to the TSMC nearly 10% of total revenue.

Antminer-S9-e1515121136127

Bitmain S9 designed by Bit Continent is currently the main type of machine that is specially designed for the mining of Pittcoin. However, since the hash calculation adopted by bitcoin is mainly based on the SHA-256 algorithm, it not only consumes computing power but also consumes energy. In addition, as the number of bitcoins in the future becomes more and more scarce, The harder it is, the bitter continent begins to develop its own ether for the future.

It is understood that the ethernet ETHASH algorithm used, originally designed for the memory bandwidth (band width), which is the demand for graphics chips. So it is suitable for PC mining, mining machine is not good at this. However, because Ether is likely to adopt a PoS proof of stack model by mid-2019, which will make the current mining machine more difficult to enforce, the continent will start designing a new type of mining machine, To meet after the market demand.

According to "Science and Technology News," the exclusive grasp of the news, bit mainland China is about to launch mining for the Eupean F3 models, the design is to increase the DRAM bus bandwidth, but also to increase the size of the memory. In the future, each miner has three mainboards, each with six dedicated ASIC processors for mining, and each dedicated ASIC processor for mining features 32 1Gb DDR3 memory. In total, there will be 72 GigaByte of DRAM memory on a F3 miner, which is significantly different in memory size compared to the current S9 miner with bitcoin only 512 MB of DDR3 memory.

Market estimates, because the F3 of small energy consumption, great benefits, the future will be the formal introduction of the future, will impact the original graphics chip PC mining market. Also optimistic about such a market, the recent frequent bit mainland China to Taiwan to meet the memory plant, to discuss whether it can provide the relevant memory capacity. It is understood that the demand for this part of the bit continent is about the amount of 3 thousand 12-inch wafers per month. In this part, the current domestic memory factories are still trying to squeeze out the production capacity supply. As for the TSMC chipmaker that will be handed over to TSMC's foundry 28nm process, it will be mass-produced and shipped in February, while the bitcoin-based mining machine in the mainland of the continent is expected to ask Q2 and Q3 of 2018 .

[J. Cooper]

Planning a ETH Asic with POS plans?

I think the release of ethereum asic miners going into Q2 illustrates that a PoS Ethereum network is still extremely distant. Otherwise I don't see why Bitmain would invest money and time into these miners.

[jimmykl]

I though that developing Ethereum ASIC would be unprofitable because of their plans to switch to PoS, but it seems like Bitmain has calculated otherwise. Maybe it will be usable to mine other coins with the same algorithm. But since it won't be able to mine all GPU algorithms, GPU mining will still be around for a long time, so if this ASIC will take over ETH mining, current ETH miners will just switch to other coins.

The ETH PoW is designed to be ASIC resistant. So if there is an ASIC, it will change the algorithm.

Yes ASIC *resistant* not ASIC *proof*. What do you mean change the algorithm?

[codewench]

In the future, each miner has three mainboards, each with six dedicated ASIC processors for mining, and each dedicated ASIC processor for mining features 32 1Gb DDR3 memory. In total, there will be 72 GigaByte of DRAM memory on a F3 miner,

The math does work out:

(3 boards) * (6 ASICs) * (32 RAMs) * (1 gigabit) = 576 gigabits = 72 gigabytes

[Mattthev]

DDR3 is pretty slow for ETH mining, look at the GDDR5X it's slower than GDDR5 and DDR3 is way slower for mining, but it would be probably cost effective with those DDR3, seems like DDR4 are too expensive that's plus.

[Bojcha]

DDR3 is pretty slow for ETH mining, look at the GDDR5X it's slower than GDDR5 and DDR3 is way slower for mining, but it would be probably cost effective with those DDR3, seems like DDR4 are too expensive that's plus.

true about gddr3 however 5x can work good
1070 with custom bios and 4GB (512MB per chip) Micron GDDR5X giving over 40MH/s with same 256bit bus

[sovacar]

DDR3 is pretty slow for ETH mining, look at the GDDR5X it's slower than GDDR5 and DDR3 is way slower for mining, but it would be probably cost effective with those DDR3, seems like DDR4 are too expensive that's plus.

true about gddr3 however 5x can work good
1070 with custom bios and 4GB (512MB per chip) Micron GDDR5X giving over 40MH/s with same 256bit bus

Sooooo, how does one edit nvidia bios?

[khufuking]

Even if this is true , Why you guys think it will kill GPU mining on ethereum ? . There is no information on how much hashrate those things will give and how much they will cost . I bet that the ratio between those ASIC and the current rigs we have will not be much . And even if that is not  the case , There is always another coin that will take ETH place as number 1 option for GPU mining .

Also making ASIC for sia and before it dash was already big fail . Look at the current profit . ( I know a lot might disagree with me on this ) But that what I think .

[POM]

DDR3 is pretty slow for ETH mining, look at the GDDR5X it's slower than GDDR5 and DDR3 is way slower for mining, but it would be probably cost effective with those DDR3, seems like DDR4 are too expensive that's plus.

true about gddr3 however 5x can work good
1070 with custom bios and 4GB (512MB per chip) Micron GDDR5X giving over 40MH/s with same 256bit bus

Are you referring to the P104s?

[does]

According to Technews, Bitmain is about to release the Ethereum ASIC miner F3 in Q2 or Q3 2018.

“Bitmain is about to release F3, the ethereum ASIC miner. It’s reported that every miner is mounted with 3 mainboards. On each mainboard there are 6 ASIC processors, each of which has 32 1GB DDR3 memory. Therefore, one unit of F3 miner contains 72 Gigabyte DRAM memory.”

I have doubts on the math of the report because the numbers doesn’t add up.But 72G DRAM seems reasonable and already outperforms most of the GPU-based mining rigs in the market now.
The bottleneck of Bitmain’s mass production is supply of Dram memory, which explains why Bitmain is frequent visiting storage supplier in Taiwan. The demand of Bitmain is around 3,000 pieces of 12 inch wafers per month. Ethereum ASIC chips will enter mass production stage in February 2018 by TSMC’s 28nm process. The F3 miner is expected to hit the market in Q2 or Q3 2018.
The ethhash power algorithm of Ethereum is designed to be ASIC proof. But just like Charlie Lee put it:

“I think ASICs are inevitable when the coin is valuable. But that’s not a bad thing.”

20180212132759

Bitmain has developed ASIC miner for Litecoin and Siacoin.


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

why so much ram ?

[icedlink]

DMM Japan most likely have it already ...
and they are test running it in their farms.

F3 specs..
650MH/s  @  750Watts

 

[bbc.reporter]

I though that developing Ethereum ASIC would be unprofitable because of their plans to switch to PoS, but it seems like Bitmain has calculated otherwise. Maybe it will be usable to mine other coins with the same algorithm. But since it won't be able to mine all GPU algorithms, GPU mining will still be around for a long time, so if this ASIC will take over ETH mining, current ETH miners will just switch to other coins.

I reckon Ethereum's switch to POS is still years away, or to some critics, it might never happen. But if it does make the switch, there's always Ethereum Classic waiting, which has a capped supply and it is deflationary in nature.

Also with Asic miners around, ETH's price might crash because of its unlimited supply and the developers' unwillingness to start the Ice age, or did it already start? Sorry, I don't follow Ethereum anymore hehehe.

[mohic]

does that mean we can use these asic for gaming somehow? O.o