Bitmain & Canaan To Release 5nm chip ASIC Miners in Q1 2020!

[DaCryptoRaccoon]

I still run my S3'+s.  Some of the toughest miners they made in my view I still have around 10 of them with not one chip down.

It would be very nice to see bitmain give us a miner with less draw but I cannot see this being the case from what I have been reading it's going to be a trade off between better hashing power or better efficiency.

Do we know any more info on this or a rough date they may be hitting the store?

[mikeywith]

So you think they're gonna start building miners with under 2000W power draw again? Have they ever gone backward in power draw, since the S3?

Not exactly under 2000w , but most likely nothing above 3kw , current list is a follows:

Code:

S17 >   52th at 2600w
S17e>  64th at 2880w
S17+>  75th at 3500w

20% power reduction without increasing frequency  ( the way I imagine it)

Code:

S19 >   52th at  2080w
S19e>  64th at   2300w
S19+>  75th at  2800w

Bitmain might have a different way of looking at things, but the way I see it is that there is not much time left for industrial mining gears, based on my thoughts and other's mentioned here >https://bitcointalk.org/index.php?topic=5208911.0 , mining farms' days are going to be over soon.

As a result of the above theory, It makes sense that Bitmain should start focusing on home-friendly mining gears, those gears need to draw as less power as possible and be as quite as possible, if they don't do that post halving, they will have to do it close to the 2024 halving.

Again I am just speculating on what I would do if I was Jihan wu, obviously, i am more likely to be wrong than right when i speculate

[sidehack]

Focus on home-friendly miners, you say? That sounds familiar.

[Searing]

Crap! Did the math on my above on having too much equipment depreciation and mining died on me this year and likely next year. Damn you IRS!
 
Anyway, from what I can tell I need to make $13.5k or so a year in ADJUSTED gross income for the next say 3 years counting this year in Home Business Mining or whatever to keep this 'bug a boo' at bay. Ok, this year I have it covered ...barely I think and will show a profit. I showed a loss in 2016. Will beat the 13.5K equip deduction hard floor in 2019. Whew! So I really, really kinda have to show a profit in 2020 via (1) eBay sales of crap in attic maybe $5k worth?  (2) mining (3) crypto trading (un-frigging-likely) (4) build a data hall someplace (but it would also have to make income and then I'd have more equipment/building depreciation) Ack! I suppose a person could (5) consult as a BTC expert *but hell, I don' listen to myself most times with crypto*

So thinking I may have to get equipment in 2020, and just NOT take the 25% equipment deduction, maybe not even take off electric. Just so I can show an adjusted Net Income from the business (with equipment deductions on the books till probably thru 2021) or I will be slapped into 'hobby mode' so fast by IRS it will make my 'head spin'. The IRS, alas, will not take an attic full of obsolete ASIC equipment from 2013 onward in my attic, by that point in time, as 'proof', it was not a hobby. If you read the IRS rules they also can say you are a 'hobby' if your business skills are 'unrealistic' and/or bad, or the business itself is 'dubious' at best. Ack! As far as I can tell it is NOT illegal for me to screw myself and not take deductions I'm entitled to, on electric from mining and/or equipment.

However, there is a fine line between, say 13.5K not used and exposing myself to IRS as a hobby, when and if I fall off the profit cart in 2020. I will have to do the math, however. It is not the first time, I've gotten equipment rather than more to IRS (how I got in this mess with the 70% or more BTC/Crypto crash in the frigging first place!) (Bitcoin: You merciless bitch!)

But again, probably gonna have to do something like this, very carefully, after halving, at the end of 2020, and used equipment maybe. It is all very dubious if this is gonna work or not. But at least I'm clear for 2019. But have no frigging idea yet besides eBay store sales on how the f*ck I can pull the same out in 2020. Anyway, advice welcome via PM or whatever. (I'm probably off-topic ..kinda sorta....but sh*t 5nm chips in Q1 has got to put a lot of us on Bitcointalk here in the same frigging boat as myself!) whatever, bitcoin always drama!

Brad

[stompix]

As a result of the above theory, It makes sense that Bitmain should start focusing on home-friendly mining gears, those gears need to draw as less power as possible and be as quite as possible, if they don't do that post halving, they will have to do it close to the 2024 halving.

Again I am just speculating on what I would do if I was Jihan wu, obviously, i am more likely to be wrong than right when i speculate

Wishful thinking...
I will side with philipma1957 and say this is a desperate move to unload as much gear as they can if the halving price screws mining gear sales for a year or two.
And if you want to unload a ton of miners as fast as you can before the shit hits the fan you can't rely on home miners, you will have to force the hand of industrial-size business that will either upgrade to stay in the game or choose to close, which would be also a win since they will get more % with their own farms.

As much as I would really love to buy a home miner that would be tolerable if put in a good case in one of my storage closets (no wood, we're all concrete here in central Europe) I doubt we will get anything this year.
An S19 hydro at 900W so I could enjoy a toast without toasting my breaker not have the feeling I'm eating it near a damn J9 turbofan? ...

[mikeywith]

I understand where you are coming from, but you are missing one key point, which is:

Home-friendly Miners > work just fine for industrial mining farms.
Industrial Miners > do not work for home.

Since efficiency is nearly 100% the same, what would stop Bitmain from reducing the risk of price collapsing prior/after the halving where only home miners might be able to survive the hit?

Let's assume you have an efficient chip and can build a miner that does 20w/th , you can make

3000 watts and get 150th miner

Or

2000 watts and get 100th miner

The latter could be sold to both home and industrial miners, it will of course be a little bit more expensive as in price per terahash since the cost of the frame, fans and wires will be nearly the same for both gears, but the first option limits your potential clients.

I don't know what I don't know, but if Bitmain thinks that there is a 30% chance the following halving will bring industrial mining down, they would really start focusing on home-friendly mining gears.

[philipma1957]

or better yet drop to 1400 watts. or even 1000 watts.

L3+  are pretty much bullet proof running at 700 watts and doing 500mh.

[mikeywith]

Yes they can go that law, but the lower they go the more expensive it gets, there is an intial cost for every miner they make, be it 1000w or 2000w, so they would of course try to maximize the amount of chips they can put in a single miner so that the intial cost spreads across them making cost per terahash as low as possible, while trying to maintain a low power consumption.

probably go back to 2 fans only design? Or even something similar to the R4? Except the latter does not fit well in data centers.

This could be way too early, 6.25btc a block is still somehow safe, but the following halvings will cause more damage, bitcoin price has to go to unbelievably high prices for indusrsial mining to still be a thing, and if that does not happen - bitmain has no choise but to focos on effienct home-friendly mining gears, we could see once that look like a real heater, some might have a pink color on them for the ladies  , anything that does not make a lot of noise, keeps the house a little warm and pays it's own bill.

[philipma1957]

Yeah  putting in more chips  is better for them.

Well we  will get to see what happens as time moves on.

Would have been better for   longer ½ ing times say every 7 years ,  but who knows maybe someone figures out a way to scale this.

[wndsnb]

Not likely that the big manufacturers will spend any resources to make a home-friendly miner. Just too small a percentage of their sales to make it worth it for them. Industrial miners will prefer the higher density even if the price/TH and efficiency is the same.

Maybe a 3rd party could make a chassis+power supply and buy controllers and hashboards from one of the big manufacturers. A miner with a single hashboard would be in the 1000W range. If you import hashboards and controllers into the USA, do you still get hit with the extra 25% tax? It would not be a huge business, but maybe a small operation could make a profit.

[philipma1957]

Do it all through Malaysia  tax drops to 2.6%.

[Searing]

So you think they're gonna start building miners with under 2000W power draw again? Have they ever gone backward in power draw, since the S3?

I don't think so either...that they will build under 2000 watt miners again either, however, you COULD make a 'small' case that Bitmain is in trouble. They just laid off another 50% of their workforce on top of a previous 50% layoff of the workforce in 2019. So if you figure they have to have a 'certain' amount of Toaster's (miners) to get out the door with the economy of scale vs profits to keep the plants open/the designers of chips happy/the data halls/etc... and you'd probably have to have Bitcoin POP significantly to grease the wheels on the above. IT COULD happen? Nah? But playing contrary...the above would have to line up in the above for anything like that to happen.

Bitmain: Making ASIC Toasters in Amounts that allow us to stay at the leading edge of 'evil'! ) (tm Bitmain)

Brad

[philipma1957]

If they were smart they would send out a 1000 watt  sLn17 miner  using:

1000 watt to do 25th   40 watts a gh
1400 watts to do 33th  42 watts a gh

do them from Malaysia market at home lower noise miners.

They can mark up  to 1000 usd

Just sell just to usa market.

[Searing]

Remember Philipma1957 I said similar things in 2016 and Bitmain went with it.....be careful they likely may steal your ideas! (or in my case back way back a happy coincidence?)

Bitmain: Evil is as Evil does (tm: Bitmain)

[stompix]

There is one thing you must take into consideration...
Home miners not only will be hit less but they will also start to not look at mining as a revenue source....
Investing 10k in some mining gear (I consider 10k to be the maximum limit for calling it a home mining operation) to get what...500$ a month?

Note that I make a difference between home mining ( guys running gear not in data centers but with profit in mind) and hobbyists who would buy 1 or 2 small miners and run it just for the sake of having something to toy it every now and then.

How much can bitmain hope to sell to those guys? Especially since if you want something just fun you can grab an s9e for 140$ to play with?
And more important, let's say the rivals do the opposite, now you're in one not so comfy spot!
Rather than calling industrial mining dead if the price doesn't go at least to 10k this year I would look at mining gear manufacturing as a dead industry.

[mikeywith]

Investing 10k in some mining gear (I consider 10k to be the maximum limit for calling it a home mining operation) to get what...500$ a month?

To my knowledge, there was never a time where a mining gear took 20 months to ROI unless you plan on making it so by paying a fortune for a terrible mining gear at the worst time one can buy.

What about a little more reasonable numbers?

we know what the 5nm is capable of, and we know Bitmain can do.

S19- 52th at  2080w , say a home miner can manage 6kw of power, so they can buy 3 of these, using today's numbers at 6 cents kwh that gear makes about 140$ a month , so 3 of it make 420$ a month

at 1000$ per gear > 7 months to ROI
at 1200$ per gear > 8.5 months to ROI
at 1500$ per gear >10.7 months to ROI

It's highly unlikely that they will sell it for anything more than that judging by current gear prices, also for home miners it's more than just direct profit especially in the winter where they can use those miners for hear sources.

How much can bitmain hope to sell to those guys?

If mining becomes completely unprofitable for mining farms, all mining profits will move to home miners, things might be a lot more profitable than they are now, but certainly not as much as they sell now, which might also make production cost more expensive,thus makes the numbers above way off.

And more important, let's say the rivals do the opposite, now you're in one not so comfy spot!

True, It's very unlikely that Bitmain will do something different from the rest, these drastic changes will have to be universal, they won't be an option, the market will force itself upon them.

Rather than calling industrial mining dead if the price doesn't go at least to 10k this year I would look at mining gear manufacturing as a dead industry.

Yup that's another possibility, I think even if everyone who is into bitcoin decided to buy a gear or two, they won't get close to the amount of mining gears that huge farms combined buy, once we reach to that point, many things will change, some ASIC's manufacturer might shut down.


There is only one thing that needs to happen to make this discussion/debate completely irrelevant for years to come which is price moving higher gradually.

[philipma1957]

Well  from bitmain's viewpoint  isn't it easier

Pump the price of LTC  to 150-250 dollars

Then build L9 7nm or 5nm  LTC miners.

They are a business first  they have clearly shown they don't care that btc is the number one crypto-coin in the world.


From a value move boasting the price of LTC to sell newer LTC gear looks doable.

While pumping btc price to 15k  has to be harder to do

and the ½ ing cancels  the price pump to 15k.

While the price pump in ltc has a long time to be cut by the  next ½ ing

If bitmain does not do a ltc pump  canaan may do it (they have yet to do anything but sha 256)

But I feel someone (asic Builder)  will try this  switch to  LTC-Doge Idea.

Last I recall the L3+  was  a 28nm but Maybe it was a 45nm chip.

LTC- Doge are worth 2.8B + .3 b = 3.1b

BTC - NMC is worth 142.891b + .007b = 142.898b

certainly easy to pump ltc.


My fear is  rather then push  BTC/BCH/sha256 they move towards ltc/doge

with a block size of 6.25  btc will need to be 15,000 usd to stay even.

ltc has a block size of 12.5  for next 3 years. 

 If it boosts to 300 a coin  a 5nm  LTC miner  will look good to asic buyers.

[Searing]

So you think they're gonna start building miners with under 2000W power draw again? Have they ever gone backward in power draw, since the S3?

No, I really don't. However, the 'economy of scale' for bitcoin to keep its factories and employees open with the added problem of R & D development for new ASIC's along with other supply chain issues, I do know they have to move a 'minimum' amount of product (ie toasters) or the whole thing will become unraveled. So PERHAPS they might offer a low-end model for the above reasons to just keep the frigging doors open. But this would only apply if they can't move huge units fast enough at inflated prices, which will always be what they prefer. But the 'Needs must when the Devil drives"....so they might have to get creative to keep the doors open with all the overhead they have, even with a 50% cut in the workforce. I myself don't understand how they are selling the units they are now at:

(1) IMHO the inflated prices up by 1/3 more than I find in any way reasonable for ROI.

(2) Why the hell they are not shipping EVERYTHING to the USA to get around USA Tariffs via Malaysia where some in the USA are getting the product from (if lucky).

(3) How you currently can sell (mostly sell out) of Bitmain units that ROI in more than 1 year with flat difficulty. I mean are we simply dealing with clueless very rich newbies?

(4) How you can even justify ANY of the equipment above when Bitmain and Innsilicon announced they will go from 7nm to 5nm chips in Q1 of 2020. (A 15% improvement)

(5) And IF they really can get 5nm Chips out in Q1, they obviously will go to Bitmain Data Halls first and any 'overage' of 7nm will be sold to the clueless in bulk!

So yeah, I'm befuddled, to say the least on all ASIC miner prices of ANY flavor of crypto! Even without the USA 27.6% Tariff with import fees and less shipping. The math is so ugly as to be obvious IMHO, or at least at my 10c kWh electric prices.

Brad

[judypug1956]

bitmain dropped price under 800 for t17 not  subject to trump tax if you order 1 at a time

https://shop.bitmain.com/product/detail?pid=00020200107142814172GxOYpK6s06B5

they are looking to clear gear out.

the t17 can do 45 watts with correct firmware.

https://bitcointalk.org/index.php?topic=5208500.0

Antminer T17

"...Downvolts to 1900w at 42th/s - 45w/th..."

from topic page on the thread above.

bitmain has an issue of being oversold once 1/2 ing comes.

All of the above relates to the topic as they have to be afraid that btc price is not going to rise to meet miners needs.
So they are selling bigger and better gear real soon just around the corner.

I smell a big crash coming very soon.

[stompix]

To my knowledge, there was never a time where a mining gear took 20 months to ROI unless you plan on making it so by paying a fortune for a terrible mining gear at the worst time one can buy.
~
If mining becomes completely unprofitable for mining farms, all mining profits will move to home miners, things might be a lot more profitable than they are now, but certainly not as much as they sell now, which might also make production cost more expensive,thus makes the numbers above way off.

Well, that's my point, remember we're talking about a scenario with no price increase and halving drop.

If home mining will be profitable with no (bulk equipment buys, preferential electricity rates, more efficient gear) I highly doubt you will ROI under 12 months. If you start making a profit without those advantages in six months you can surely bet industrial-scale mining won't be going down which...would keep home mining at bay...never-ending circle.

Second, what is home mining?
I stopped mining when the gear hit new levels of "efficiency" because of the noise and power, right now I could still probably run 3-4 s9 in my home but probably I would look every time at the breaker when both my electric oven and dishwasher are running during summertime along with the air conditioner as our circuits here won't be able to handle more than 7.5 kw.
Now I could use my parent's farm for mining as just one cereal grinder there is 30kw and is not used permanently but...
Would then I will still be a "home" miner?

And more important, those who buy these miners, as many as they would be, they would have to buy only a few 1,2 max 4 of those,?
Because the moment you buy 20 home miners... it makes no sense why you wouldn't buy 4 big ones that are 5 times more powerful, right?

This is the problem with targeting the so-called home miners, it's a very tight niche.

Hobby mining, that's something else, and this is why I would love a cheap silent miner that anyone could run just to support the network and that wouldn't make you raise an eyebrow when the bill comes. But that's wishful thinking and I know it.

Well  from bitmain's viewpoint  isn't it easier
Pump the price of LTC  to 150-250 dollars
Then build L9 7nm or 5nm  LTC miners.

At 200$ the daily reward for LTC mining is 1.4 mil.
With no price rise and drop in reward bitcoin would still generate around 7 mils a day.

After the BCH pumping loses probably they've stopped with the stupid plans...I hope....