Topic: AntMiner S9 rampant speculation thread, have to be quick though - page 6. (Read 12633 times)

This is probably nonsense, but someone on OpenBazaar is offering an S9:

http://bazaarbay.org/17144de3790b134b05a0fca102ff8b8887d08e27/listing/c6bbb66fd2041bde823ad70df9bdb757600aafee

They claim:

BITMAIN ANTMINER S9 Batch 1 with 9.63th/s
Shipping from 26th May 2016
Price per unit   4.8BTC

It's probably a scam. But I'm posting it here on the off chance that it turns out to be accurate.

The only current worry about Scrypt-based coins is the expected large difficulty ramp-up when the A4 starts selling this summer.
 Main question comes down to "how much will it cost" and the resulting "how many will they manage to sell over the first few months".
 My A2s haven't hit ROI yet - but I'm fairly comfortable that they WILL manage it, expecially after I move to a land of VERY VERY cheap electric over the next couple months (barring disasters).

 As far as the S7 ROI - in my case, with slightly LOWER than average electric, no S7 batch would have ever managed to achieve ROI for me. The FIRST batch came the closest, but only due to the big Bitcoin price runup last winter - and even WITH that factor, I'd still be pretty far in the hole.

 Folks with VERY VERY cheap electric (under 5c/KWH) might have managed to ROI their S7s without having to sell them off, but I really don't feel comfortable being FORCED to count on selling off miners while they're still at least marginally profitable to achieve ROI. I came VERY close to taking a batch on my S5s + SP20 'cause of that, if I'd waited even 2 more WEEKS to sell them I'd have lost money on them (and as it is, I BARELY broke even on them and GETTING some of them sold was a bit of a nightmare).

I made the same mistake. My only out was to sell off my S7s and the earned BTC to realize a fiat profit. But that was only possible because BTC nearly doubled in value.

It would have been a much better investment to just to buy BTC, speculating on price, instead of buying S7s, speculating on difficulty.

Like you, I won't make this mistake again. I'm already assuming that the S9 will be insanely priced. In which case I'll hang on to my crappy Scrypt ASICs just to keep my hand in. They're paid for and with the relatively flat Scrypt difficulty will continue to profit because they don't care about the BTC halving.

Your analysis using the basis that everyone has free electricity lol.  Of course if you had free power every miner would ROI, factor into that the 10 cents per kw/h which is average for the United States and there will be no ROI lol

The S7 should of ROI by now. with USD

Look at my analysis thread

https://bitcointalksearch.org/topic/lifetime-hardware-roi-profits-for-all-bitcoin-mining-hardware-1477198

I spent my hard-earned BTC on Batch 1 and it still hurts. Anyway, lesson learned, luckily I didn't go all in with the S7, and I'm not repeating this mistake with the S9.

I purchased my coins from coinbase and then purchased my batch 1.

I live in USA  so legally I need to use cash basis.  My batch 1 made a cash profit.
I would have been better off holding the coins as they 2x the price I paid for them.

But I also purchase a lot of bitmain contracts with 230 btc  that  when they matured they paid in btc that rose to 450-500.

I have no complaints about my batch 1 in terms of profit.  It was an under hasher that  was compensated for on May of this year after complaining since Oct 15 or so.

roi on B1 S7 in btc is close to being impossible (unless you sold for btc sometime ago), much better in $.

This was in their s-2.

They must have a connection with the company.

I would buy that s-7 lite with a no psu option and add a titanium 800 watt from silverstone.

Then down clock it just a bit. Then put it in my friends office for free power.

And from NewEgg pricing on that Enermax (I've NOT had good luck with their power supplies, 50% fail rate while in warrenttee though fairly small sample size) they could have gone for an EVGA or Seasonic with near-Platinum efficiency and a MUCH better track record on longevity for about the same price.

there is no lost profit until you reach roi, and roi on new equipment is usually ery high

now think about those that bought the s7 early, at the unbelievable price of 8 btc at that time, they are still trying to reach roi

the same will happen with the s9 the s11 etc...nothing will change

It is a quite a risk to buy a miner now - but thinking about lost profit...

Buying a miner now is really a big risk with the halving approaching so soon, I would wait for after the halving to see how the difficulty is affected and the price of bitcoin, I'm sure Antminer will wait until after the halving because they know a lot of people are probably thinking this as well.  Although just last year BTC was at like $230 so who knows maybe the halving has already been factored into the Bitcoin price.

I just hope that it does not look like that abomination with enermax PSU attached on top.

Or maybe they'll have a couple of options, like a water-cooled system similar to the C1 (I loved that C1. So sad it's sitting on a shelf doing nothing. *sniff*). Or an immersion kit.

My guess for the S9 would be somewhere along 10 TH/s and it's price somewhere in $1200 to $1400 region. It will be the most efficient one so a power consumption of 1000W could be their target for this.
The looks of it could be like an S4 body type and the noise would be the most among all the antminer versions.

I will buy this S9 if there is a PSU integrated into it same as they did with S4.

In the name of decentralization I would love to see a 5 TH/s @ 800W for $1k so that people can run it at home with a 1KW PSU. But, I think it's more realistic with around 9 TH/s @ $1500-$2k and @ 13-1500w

 Pretty close to my early guess. I've not seen specs on their next chip yet to make any informed guesses though - I was pretty bloody close on the S7 *EXCEPT* I didn't predict they would move to 3 strings/board and 3 boards/miner. There isn't enough space on the board to move to more strings/board - and they might have to move back to 2 if the 14/16nm chips run at low enough voltage. I also doubt they can cram 4 boards AND enough heat sinking into a unit the size of the S7 or even close, and I don't see them changing the form factor a lot due to what kind of BAD effects it would have on their Antmine infrastructure.


 VERY optimistic. I'd guess $2000 ballpark, the new node is a LOT more expensive to develop for and the S7 started out quite a bit higher than that. Also, there won't be any "massively new massively more efficient" gear showing up for a few years, so they can afford to change their "180 day ROI assuming no diff increase price basis" structure to something a bit higher - unless Bitfury beats them to the punch and they have to be competative.


 I suspect those dates are a bit optimistic, doesn't give them much time to liquidate used S7s or ramp up enough S9s to replace THEIR S7s internally before they start selling them to the public. They haven't even announced the CHIP yet, on the BM1385/S7 the chip announcement as I recall was a month or two before they started blowing out used S5s.


 Bitmain announced they were working on a 14 or 16 nm full custom chip design at the same time (and in the same announcement IIRC) as the original S7 announcement.


I doubt they'll have enough built by then.
I would be a LITTLE supprised if they are even deploying the S9 internally in significant quantity by the halfing, given they've not even announced the CHIP yet,

"ldzly Official 2016-05-16 20:45:39

Of course we will publish our S9(Yeppppp , S9 is the name , because in Bitmain"s history, odd number means good luck =antminer beijing hot items in a selling season) which is made of our next generation chips. And we are lack of accurate information when it will release ,before that we need a long term test process to make sure the stable perfomance. So I have no idea where the information of mining sweden form ? also, I dont know why some users choose to believe him instead of inquiry form Bitmain.

have a nice day !"