Are the Antminers still worth purchasing?

[TheSnappyTrader]

So, after reading all over the web there is such a mix of opinions of whether or not mining can still be worth it.

Here's what I'm wondering. The Antminers keep coming out and don't even ship for months. To me it seems like by the time you get one the difficulty will make it only profitable for a short time or obsolete very soon. Is it even worth purchasing one or these or what? Let's assume electricity is cheap or even free.

[promojo]

So, after reading all over the web there is such a mix of opinions of whether or not mining can still be worth it.

Here's what I'm wondering. The Antminers keep coming out and don't even ship for months. To me it seems like by the time you get one the difficulty will make it only profitable for a short time or obsolete very soon. Is it even worth purchasing one or these or what? Let's assume electricity is cheap or even free.

It can be... the only part is... waiting THREE months for them to ship.   Not to mention all the new gear that will be coming out in 2018.   If you don't get at least 6-8 months you won't ROI.

[drubas]

I would recomment to wait until a new one comes. The efficiency will again be much better than previous ones. Therefore I would say... save money, wait and purchase one or two.

All the best

B.D.

[NotFuzzyWarm]

I have one antminer s9 and am happy with it.  interested in acquiring more.  I think it is possible to be profitable since I have cheap electricity to run a few from my office, as we have electricity included in rent, within reason.

Where is best place to purchase, bitmain.com...these are the people who actualy make them?

Yes that who makes them and yes they should be THE ONLY PLACE you buy them from.
Bitmain does NOT use any Authorized resellers and buying from anyone other than Bitmain means:
a. There is NO factory warranty! PERIOD.
b. You will be paying far far more then you will when buying direct from Bitmain.
c. Strong chance the resale site is a scam.

As for 'wait for new ones"... Such as what?
Wildly optimistic speculation as to the next mining ASIC chips and when they will be released is totally unfounded on what the semiconductor Foundries are able to produce. Frankly I do not see a move to lower size nodes for at least a year if not more, even then it will mainly depend on there being a real need for it. So far there is NO reason for Bitmain et al to push it. Throw in the birthing pains that Bitmain and Canaan had with their 16nm chips and it is very safe bet that both companies are going to be far more cautious.

Even at the current 16/14nm node there is not enough capacity to satisfy the market. The current bleeding-edge 10nm node currently in production by Samsung is ONLY for their internal use and will be for a long  long time before they open their doors to other companies. Yes TSMC and GloFo have their versions of that gate size but so far it is purely at testing levels and is reserved strictly for their main customers and that ain't Bitmain or Canaan.

[promojo]

Even at the current 16/14nm node there is not enough capacity to satisfy the market. The current bleeding-edge 10nm node currently in production by Samsung is ONLY for their internal use and will be for a long  long time before they open their doors to other companies. Yes TSMC and GloFo have their versions of that gate size but so far it is purely at testing levels and is reserved strictly for their main customers and that ain't Bitmain or Canaan.

Did you read the news on Canaan?  7nm chips coming in 2018 - TSMC is one of their direct partners as well doing some AI stuff....   Still vague stuff but we should expect something come early-mid 2018.

From their site - "The next big jump in capability won’t happen until production moves to 7nm in 2018, so it makes sense to move the prices down and open up more opportunity for individual miners.”

[NotFuzzyWarm]

Even at the current 16/14nm node there is not enough capacity to satisfy the market. The current bleeding-edge 10nm node currently in production by Samsung is ONLY for their internal use and will be for a long  long time before they open their doors to other companies. Yes TSMC and GloFo have their versions of that gate size but so far it is purely at testing levels and is reserved strictly for their main customers and that ain't Bitmain or Canaan.

Did you read the news on Canaan?  7nm chips coming in 2018 - TSMC is one of their direct partners as well doing some AI stuff....   Still vague stuff but we should expect something come early-mid 2018.

Yes I read that and it is meaningless. Bitmain has said the same thing. Those PR's are strictly that -- PR announcements of intended road map and have nothing to do with the Real World much less what the Foundries are actually able to produce. Back in late 2014/early 2015 both had announced work on 16/14nm with release dates in mid to late 2015 and we see how that worked out...

I'll be more curious to see if the miner chip makes have learned their lesson on what happens when you pack max number of cores that will fit into low-node size chips INTENDED FOR LOW POWER MOBILE APPLICATIONS drawing at most 1W and then try to push >10W through them, eg they are unstable and have specs all over the map.

As for Canaan's foray into AI chips - useless for mining and this has been discussed elsewhere in the Forum. Those are slated for autonomous vehicles/drones.

[QuintLeo]

There won't be significant production capacity on 7nm (or even 10nm) that is available to a SMALL company like Caanan untill at best late 2018 - and that's assuming the node even goes into production AT ALL in 2018.

 Keep in mind that 14/16nm "production" was first announced to exist almost 2 YEARS before any major capacity showed up, due to yield issues with the new node and the amount of work to get a line up and running at ALL for those nodes, plus the HUGE equipment investment needed.
7nm/10nm node is looking to be even WORST about all of that by a lot, at least in part due to the issues with getting EUV up to the power level needed on a reliable PRODUCTION basis.

 I wouldn't bet on ANY miner manufacturer having access to more than "tape-out" level "production" before 2019, or MAYBE the very end of 2018.


 Even INTEL hasn't announced firm dates on it's introduction of 10nm process devices, and they're YEARS ahead on that node vs anyone that has announced 7nm "being worked on" (and were the first to actually start shipping 14nm production chips IIRC).

[Aurel577]

thanks notfuzzywarm.  I got my first one from ebay and did pay over market price, but it seems like it is very difficult to get an order on bitmain website.  I will wait and see what happens.

I agree as I bought both my S9 off EBay and did pay more than Bitmain but have made that back by being able to mine instead of waiting the couple months for shipping. Just have to look for the bargains on EBay they do pop up every now and then.

[fanatic26]

I agree as I bought both my S9 off EBay and did pay more than Bitmain but have made that back by being able to mine instead of waiting the couple months for shipping. Just have to look for the bargains on EBay they do pop up every now and then.

Considering the unreliability of mining hardware this is not a good idea. You have zero warranty this way and are probably paying a significant markup on the actual unit cost.

[xIIImaL]

I have one antminer s9 and am happy with it.  interested in acquiring more.  I think it is possible to be profitable since I have cheap electricity to run a few from my office, as we have electricity included in rent, within reason.

Where is best place to purchase, bitmain.com...these are the people who actualy make them?

Thanks mate. I knew this site to purchase the antminer series. Please let me how much you are making with this miner? I am asking because I am planning to invest and buy one or two antminer s9. That is why I am gathering more information about. Please PM me the information or if you don't mind please quote below. I will check then.

[promojo]

Thanks mate. I knew this site to purchase the antminer series. Please let me how much you are making with this miner? I am asking because I am planning to invest and buy one or two antminer s9. That is why I am gathering more information about. Please PM me the information or if you don't mind please quote below. I will check then.

There are so many factors; roughly.  Anywhere from 200-500USD/month.

[jasper1234]

Even at the current 16/14nm node there is not enough capacity to satisfy the market. The current bleeding-edge 10nm node currently in production by Samsung is ONLY for their internal use and will be for a long  long time before they open their doors to other companies. Yes TSMC and GloFo have their versions of that gate size but so far it is purely at testing levels and is reserved strictly for their main customers and that ain't Bitmain or Canaan.

Did you read the news on Canaan?  7nm chips coming in 2018 - TSMC is one of their direct partners as well doing some AI stuff....   Still vague stuff but we should expect something come early-mid 2018.

Yes I read that and it is meaningless. Bitmain has said the same thing. Those PR's are strictly that -- PR announcements of intended road map and have nothing to do with the Real World much less what the Foundries are actually able to produce. Back in late 2014/early 2015 both had announced work on 16/14nm with release dates in mid to late 2015 and we see how that worked out...

I'll be more curious to see if the miner chip makes have learned their lesson on what happens when you pack max number of cores that will fit into low-node size chips INTENDED FOR LOW POWER MOBILE APPLICATIONS drawing at most 1W and then try to push >10W through them, eg they are unstable and have specs all over the map.

As for Canaan's foray into AI chips - useless for mining and this has been discussed elsewhere in the Forum. Those are slated for autonomous vehicles/drones.

That was kind of rude and abrupt..sheez!

[NotFuzzyWarm]

Even at the current 16/14nm node there is not enough capacity to satisfy the market. The current bleeding-edge 10nm node currently in production by Samsung is ONLY for their internal use and will be for a long  long time before they open their doors to other companies. Yes TSMC and GloFo have their versions of that gate size but so far it is purely at testing levels and is reserved strictly for their main customers and that ain't Bitmain or Canaan.

Did you read the news on Canaan?  7nm chips coming in 2018 - TSMC is one of their direct partners as well doing some AI stuff....   Still vague stuff but we should expect something come early-mid 2018.

Yes I read that and it is meaningless. Bitmain has said the same thing. Those PR's are strictly that -- PR announcements of intended road map and have nothing to do with the Real World much less what the Foundries are actually able to produce. Back in late 2014/early 2015 both had announced work on 16/14nm with release dates in mid to late 2015 and we see how that worked out...

I'll be more curious to see if the miner chip makes have learned their lesson on what happens when you pack max number of cores that will fit into low-node size chips INTENDED FOR LOW POWER MOBILE APPLICATIONS drawing at most 1W and then try to push >10W through them, eg they are unstable and have specs all over the map.

As for Canaan's foray into AI chips - useless for mining and this has been discussed elsewhere in the Forum. Those are slated for autonomous vehicles/drones.

That was kind of rude and abrupt..sheez!

Please ref here for why... The OP asked, if worth it to purchase a miner because of... and I debunked a false 'reason' given to not buy one.

Being a technical Forum many topics here are NOT open to discussion -- something either is or is not currently possible. When people with zero technical background in what they are talking about start trying to justify a belief by continually saying 'but what if they could... and then if they could only do...' is bad enough, topping that off by putting a timeline to something not yet possible is not intelligent discourse: it is either just wishful thinking or willful ignorance. ref NEAN

[margana]

I've been told 8-10 weeks to get a miner from bitmain. Is that not to be believed?

What are the recommended tools to accurately model futures diffficulty and mining profitability over time (to the extend that's possible)?

[QuintLeo]

I've been told 8-10 weeks to get a miner from bitmain. Is that not to be believed?

What are the recommended tools to accurately model futures diffficulty and mining profitability over time (to the extend that's possible)?

 Bitmain has been selling ALL of their models on a "2-3 months preorder" basis for several months now - demand exceeds supply that much.

 WHEN you can get into their site at all to order anything right after they release a new batch (optimally their current "revamping the site" will DEAL with the traffic load a lot better).


 There are no "tools" to accurately model future profitability - prices vary too much for any "model" to work due to factors that often CAN'T be predicted ahead of time (like the Chinese Government).
 "Technical Analysis" on cryptocoins is a JOKE, pricing is far too "news driven" for TA to WORK on a market as immature as cryptocoins.
 As best, you can estimate it - and hopefully allow ENOUGH to make up for unexpected changes along with expected ones.

 The model for future difficulty on SHA256 coins comes down to seeing WHEN a new Bitmain batch is going to be shipped, and how big it is, then adding a "fudge factor" for the apparently much lower sales of gear by Caanan, BW.com, and Bitfury-based miners.
Since the introduction of the S9, difficulty increases on SHA256-based coins have been limited to how many chips the various miner makers can get out of the foundries since they are now competing with the BIG BOYS like Apple, Samsung, Nvidia, AMD, and IBM (but not Intel due to Intel having THEIR OWN foundries) on how much foundry production they can get out of TSMC/Samsung/GF on lines that are probably sold out months ahead of time and running flat-out processing the CURRENT state-of-the-art semiconductor technology.