BITMAIN launches 4th generation Bitcoin mining ASIC: BM1385 - page 3.

Biodom

legendary

Activity: 3892

Merit: 4331

Quote from: suchmoon on August 28, 2015, 12:25:55 PM

Quote from: notlist3d on August 28, 2015, 12:14:55 PM

But it is hard to tell if they come out and no one else has something to compete. They can do what every they really want.

Exactly. Why would they sell it cheap if there is enough demand at a higher price. And if $700 sounds like a good price for many posters here then likely there is going to be enough demand at $1000.

they were in a similar position with S3 and S5 (a little less so since they probably wanted to undermine Sp20 as a competitive machine), yet the prices stayed somewhat within reason. Weak yuan/renminbi is also a factor in our favor. When s5 was introduced, everybody thought that it would be more expensive in $$.
I will probably not buy anything above ~3 BTC for ~2.4Th/550-600W.

suchmoon

legendary

Activity: 3654

Merit: 8909

https://bpip.org

Quote from: notlist3d on August 28, 2015, 12:14:55 PM

But it is hard to tell if they come out and no one else has something to compete. They can do what every they really want.

Exactly. Why would they sell it cheap if there is enough demand at a higher price. And if $700 sounds like a good price for many posters here then likely there is going to be enough demand at $1000.

notlist3d

legendary

Activity: 1456

Merit: 1000

Quote from: wlefever on August 28, 2015, 10:14:19 AM

Quote from: Biodom on August 28, 2015, 10:10:20 AM

Quote from: QuintLeo on August 28, 2015, 04:02:38 AM

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

The 230watt / th is THE hard figure they've mentioned, though the fine print on that specified chips running at .66 volts.
Presuming they stick with string design, that DOES specify 18 chips per string.
It would be VERY VERY iffy for them to try to cram 54 hash chips on one hash board.
Therefore, the general consensus has been 2 hash boards, 2 strings per chip, appx. 2.3 TH at appx 550 watts ballpark for an S7, at around 4 BTC PLUS shipping (or around $930ish at current BTC pricing). Call it $1000 or so shipped.

There is absolutely no consensus that it will cost $930, none at all.
$700-750 is much more reasonable for a 2.3Th machine, and will sell twice as many units.
They can get extra $100 by delaying shipment by 2-3 weeks while "testing"

I was getting ready to type something up to the tune of $930 sounding on the high side. It really wouldn't follow their pricing trend. I would expect the range you suggest Biodom, and as I mentioned above anything under $700 would be a bonus.

It will likely follow all previous generations. I suspect a 110/120 2 blade machine on launch with new chip. I think 1k is on high side. I suspect it will be a decent amount lower then that.

But it is hard to tell if they come out and no one else has something to compete. They can do what every they really want.

wlefever

legendary

Activity: 1174

Merit: 1001

Quote from: Biodom on August 28, 2015, 10:10:20 AM

Quote from: QuintLeo on August 28, 2015, 04:02:38 AM

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

The 230watt / th is THE hard figure they've mentioned, though the fine print on that specified chips running at .66 volts.
Presuming they stick with string design, that DOES specify 18 chips per string.
It would be VERY VERY iffy for them to try to cram 54 hash chips on one hash board.
Therefore, the general consensus has been 2 hash boards, 2 strings per chip, appx. 2.3 TH at appx 550 watts ballpark for an S7, at around 4 BTC PLUS shipping (or around $930ish at current BTC pricing). Call it $1000 or so shipped.

There is absolutely no consensus that it will cost $930, none at all.
$700-750 is much more reasonable for a 2.3Th machine, and will sell twice as many units.
They can get extra $100 by delaying shipment by 2-3 weeks while "testing"

I was getting ready to type something up to the tune of $930 sounding on the high side. It really wouldn't follow their pricing trend. I would expect the range you suggest Biodom, and as I mentioned above anything under $700 would be a bonus.

Biodom

legendary

Activity: 3892

Merit: 4331

Quote from: QuintLeo on August 28, 2015, 04:02:38 AM

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

The 230watt / th is THE hard figure they've mentioned, though the fine print on that specified chips running at .66 volts.
Presuming they stick with string design, that DOES specify 18 chips per string.
It would be VERY VERY iffy for them to try to cram 54 hash chips on one hash board.
Therefore, the general consensus has been 2 hash boards, 2 strings per chip, appx. 2.3 TH at appx 550 watts ballpark for an S7, at around 4 BTC PLUS shipping (or around $930ish at current BTC pricing). Call it $1000 or so shipped.

There is absolutely no consensus that it will cost $930, none at all.
$700-750 is much more reasonable for a 2.3Th machine, and will sell twice as many units.
They can get extra $100 by delaying shipment by 2-3 weeks while "testing"

Starin

sr. member

Activity: 266

Merit: 250

Infected Mushroom

Quote from: Chris_Sabian on August 28, 2015, 09:18:43 AM

Quote from: QuintLeo on August 28, 2015, 04:02:38 AM

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

The 230watt / th is THE hard figure they've mentioned, though the fine print on that specified chips running at .66 volts.
Presuming they stick with string design, that DOES specify 18 chips per string.
It would be VERY VERY iffy for them to try to cram 54 hash chips on one hash board.
Therefore, the general consensus has been 2 hash boards, 2 strings per chip, appx. 2.3 TH at appx 550 watts ballpark for an S7, at around 4 BTC PLUS shipping (or around $930ish at current BTC pricing). Call it $1000 or so shipped.

That appears to be a fair assessment. With those number, with no difficulty increase, ROI is ~300 days. That is going to be tight with the halving in less than a year and possible difficulty increases. Bitcoin price again is a huge unknown.

Doing the math, what was your energy cost? Considering the halving and diffuculty increase 300 days are too much.

Chris_Sabian

legendary

Activity: 896

Merit: 1001

Quote from: QuintLeo on August 28, 2015, 04:02:38 AM

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

The 230watt / th is THE hard figure they've mentioned, though the fine print on that specified chips running at .66 volts.
Presuming they stick with string design, that DOES specify 18 chips per string.
It would be VERY VERY iffy for them to try to cram 54 hash chips on one hash board.
Therefore, the general consensus has been 2 hash boards, 2 strings per chip, appx. 2.3 TH at appx 550 watts ballpark for an S7, at around 4 BTC PLUS shipping (or around $930ish at current BTC pricing). Call it $1000 or so shipped.

That appears to be a fair assessment. With those number, with no difficulty increase, ROI is ~300 days. That is going to be tight with the halving in less than a year and possible difficulty increases. Bitcoin price again is a huge unknown.

QuintLeo

legendary

Activity: 1498

Merit: 1030

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

The 230watt / th is THE hard figure they've mentioned, though the fine print on that specified chips running at .66 volts.
Presuming they stick with string design, that DOES specify 18 chips per string.
It would be VERY VERY iffy for them to try to cram 54 hash chips on one hash board.
Therefore, the general consensus has been 2 hash boards, 2 strings per chip, appx. 2.3 TH at appx 550 watts ballpark for an S7, at around 4 BTC PLUS shipping (or around $930ish at current BTC pricing). Call it $1000 or so shipped.

notlist3d

legendary

Activity: 1456

Merit: 1000

Quote from: dogie on August 27, 2015, 07:32:01 PM

Quote from: VirosaGITS on August 27, 2015, 06:07:16 PM

Quote from: flikflak on August 27, 2015, 05:55:15 PM

^^ you guys really think so? Even if they got 200-1000 sample chips, that would make 32.5 TH. So what? Antpool is still @70 PH/s. And afaik other big pools haven't increased their hashrate significantly (well maybe a few smaller ones did).

I wouldn't put aside the probability of Bitmain owning private mines. Keeping them off "the grid" is probably the best way to keep good control over Bitcoin from the miner side. Those mines would give them a lot of leeway on their actions.

That's not really how they roll. They likely manage a few farms for private clients (because why not, its cheaper) but they're reasonably transparent about their hashpower. They're also not interested in bullying the network, if the overwhelming majority [of the other power] is voting one way, they'll follow unless its suicide.

I don't know if I would say they are transparent at all. They are pretty darn secretive on their mines. I remember when they tried to sell one they said you could not even visit it even if you bought the whole thing.

So open and clear I don't think so. Unless your talking about something like hashnest. But the private farms are pretty secretive.

dogie

legendary

Activity: 1666

Merit: 1185

dogiecoin.com

Quote from: VirosaGITS on August 27, 2015, 06:07:16 PM

Quote from: flikflak on August 27, 2015, 05:55:15 PM

^^ you guys really think so? Even if they got 200-1000 sample chips, that would make 32.5 TH. So what? Antpool is still @70 PH/s. And afaik other big pools haven't increased their hashrate significantly (well maybe a few smaller ones did).

I wouldn't put aside the probability of Bitmain owning private mines. Keeping them off "the grid" is probably the best way to keep good control over Bitcoin from the miner side. Those mines would give them a lot of leeway on their actions.

That's not really how they roll. They likely manage a few farms for private clients (because why not, its cheaper) but they're reasonably transparent about their hashpower. They're also not interested in bullying the network, if the overwhelming majority [of the other power] is voting one way, they'll follow unless its suicide.

notlist3d

legendary

Activity: 1456

Merit: 1000

Quote from: OgNasty on August 27, 2015, 06:20:44 PM

Quote from: flikflak on August 27, 2015, 05:55:15 PM

^^ you guys really think so? Even if they got 200-1000 sample chips, that would make 32.5 TH. So what? Antpool is still @70 PH/s. And afaik other big pools haven't increased their hashrate significantly (well maybe a few smaller ones did).

I didn't say I thought they had a huge amount of hashrate with the new chips already.

If I were to make a guess, it would be that they got working sample chips, tested them, and then converted a massive amount of BTC for a huge order of BM1385 chips. Hence why we saw an enormous rate drop around the time of the press release. If that were the case, I would find it hard to believe that they could have miners shipping out prior to October. They even could have generated a nice discount for themselves by shorting the market just prior to cashing out for the new chips.

Granted this is all just speculation from my imagination, but it appears to be the most reasonable scenario in my mind and makes much more sense than XT (which was DOA in my opinion) being the cause for the recent "flash crash."

I am kinda tired of XT to be honest. I really wish I could know if those main people thought they were going to get adoption. It seems like a way to get certain people in the news and make them seem important.

I believe in optional acceptance of size, not forced. But that is for another day I guess. But as far as bitmain it is scary they have enough internal information that could easily be used in gaming the market. They have got there through business moves. But I'm not sure I want them at that position long term.

OgNasty

donator

Activity: 4760

Merit: 4323

Leading Crypto Sports Betting & Casino Platform

Quote from: flikflak on August 27, 2015, 05:55:15 PM

^^ you guys really think so? Even if they got 200-1000 sample chips, that would make 32.5 TH. So what? Antpool is still @70 PH/s. And afaik other big pools haven't increased their hashrate significantly (well maybe a few smaller ones did).

I didn't say I thought they had a huge amount of hashrate with the new chips already.

If I were to make a guess, it would be that they got working sample chips, tested them, and then converted a massive amount of BTC for a huge order of BM1385 chips. Hence why we saw an enormous rate drop around the time of the press release. If that were the case, I would find it hard to believe that they could have miners shipping out prior to October. They even could have generated a nice discount for themselves by shorting the market just prior to cashing out for the new chips.

Granted this is all just speculation from my imagination, but it appears to be the most reasonable scenario in my mind and makes much more sense than XT (which was DOA in my opinion) being the cause for the recent "flash crash."

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: flikflak on August 27, 2015, 05:55:15 PM

^^ you guys really think so? Even if they got 200-1000 sample chips, that would make 32.5 TH. So what? Antpool is still @70 PH/s. And afaik other big pools haven't increased their hashrate significantly (well maybe a few smaller ones did).

I wouldn't put aside the probability of Bitmain owning private mines. Keeping them off "the grid" is probably the best way to keep good control over Bitcoin from the miner side. Those mines would give them a lot of leeway on their actions.

flikflak

sr. member

Activity: 408

Merit: 259

^^ you guys really think so? Even if they got 200-1000 sample chips, that would make 32.5 TH. So what? Antpool is still @70 PH/s. And afaik other big pools haven't increased their hashrate significantly (well maybe a few smaller ones did).

OgNasty

donator

Activity: 4760

Merit: 4323

Leading Crypto Sports Betting & Casino Platform

Quote from: RoadStress on August 27, 2015, 04:46:04 PM

Quote from: bigblind on August 27, 2015, 04:23:22 AM

but there is no information about releases yet?

They are either building/upgrading their own mine or they have the results from a small batch of test chips which means that it will take 2-3 more months for the real miners.

Test chips is what I assumed. While I'm sure their test batch of chips is going strong already in their own mine, I don't think we'll see an S7 miner ship before October 1st.

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: RoadStress on August 27, 2015, 04:46:04 PM

Quote from: bigblind on August 27, 2015, 04:23:22 AM

but there is no information about releases yet?

They are either building/upgrading their own mine or they have the results from a small batch of test chips which means that it will take 2-3 more months for the real miners.

Assuming they released their chip data as soon as they got it, which i highly doubt. They have no incentives to release something new at the moment. The chip data just seemed timed to hit the lesser ASIC maker with "We'll kill you again if you try anything."

Really.

RoadStress

legendary

Activity: 1904

Merit: 1007

Quote from: bigblind on August 27, 2015, 04:23:22 AM

but there is no information about releases yet?

They are either building/upgrading their own mine or they have the results from a small batch of test chips which means that it will take 2-3 more months for the real miners.

wlefever

legendary

Activity: 1174

Merit: 1001

Quote from: toptek on August 27, 2015, 02:48:13 PM

i well aware of that and that's why but not the same power usage is my point it does say s3 + aren't they really the same s3 and s3 + . at least i replaced a s3 + board with a reg s3 board i bought off ebay that works just like a S3 + but that's not the point im trying to make I'm trying to be very positive and not prove any one wrong or right .

Sorry, I had a hard time understand what you were trying to get across. I'm with you on the being positive!

mavericklm

hero member

Activity: 742

Merit: 500

Quote from: Swimmer63 on August 27, 2015, 02:37:25 PM

Quote from: wlefever on August 27, 2015, 02:14:41 PM

..........................

Can we assume this miner will generate half the amount of heat as an S5 if were to have the same number of EDIT ~~chips~~ Th/s?

a bit over half, but it wont be at the same hash speed 1000% Wink

toptek

legendary

Activity: 1274

Merit: 1000

Quote from: wlefever on August 27, 2015, 02:14:41 PM

Quote from: toptek on August 27, 2015, 11:49:21 AM

Quote from: wlefever on August 27, 2015, 10:52:55 AM

Do they ever mention what the hashrate will be? Or, just the power consumption per Th/s as 230 watts?

I may be stretching, but a 2.4+ Th/s machine for <$700 SHIPPED would be amazing.

very first page says how much per TH

.. of his thread.

That release date reminds me of a MS release or game release . lol

At the wall power consumption AntMiner S7 230W/TH

were all guessing on the a mount of speed it ships with . why not 340 to 400, S3 and S5 were amazing to and priced right .

If anyone noticed they left out the S2 and S4 and S5 + on the front page which make is even more positive for everyone . but I'm hopping

I know it shows how much per Th/s, but not how many Th/s Wink

. My hope is that it is over a 2 Th/s machine with those efficiencies.

They leave out the S2 bc it uses the same chip as the S1 (BM1380), S4 uses the same as the S3+ (BM1382), and S5+ uses the same as the S5 (BM1384).

i well aware of that and that's why but not the same power usage is my point it does say s3 + aren't they really the same s3 and s3 + . at least i replaced a s3 + board with a reg s3 board i bought off ebay that works just like a S3 + but that's not the point im trying to make I'm trying to be very positive and not prove any one wrong or right . i did miss word my comment my bad and sorry .

Topic: BITMAIN launches 4th generation Bitcoin mining ASIC: BM1385 - page 3. (Read 39340 times)