One interesting thing to note about ROI is that for some reason bitcoin miners have an unreasonably high resell value on ebay. So an S7 in hand today (or soon) doesn't have to mine the full value back if you can resell it and:
resell amount + mining returns > cost + expenses
Of course with the halving coming up there may actually be a glut of miners hitting the resell market next year some time so timing a sale might be part of the equation if you go that route. But it's hard to predict what the halving will actually do. I would have thought the LTC halving would increase the value of LCT but it didn't really. If BTC follows the same pattern then mining will get much less profitable.
Topic: ANTMINER S7 is available at bitmaintech.com with 4.86TH/s, 0.25J/GH - page 205. (Read 527809 times)
November 17, 2015, 01:32:46 PM
November 17, 2015, 01:17:38 PM
I wager the data sheet published for these 1385 chips was not based on nominal chips but exceptional chips and perhaps some exaggeration.
IIRC I don't think we've seen a unit come in at spec, but I can't say I've ready EVERY post on the S7's.
My units are exceptional 4.86 rock solid B1's and I'm excited to see what they can do once warranty is up, and closer to ROI.
An associate has several he's been clocking at 625mhz since day one and they still have .00xx HW rates, when he clocked them higher, HW rates suffered and the hash did not increase.
All the empirical evidence so far points to the released performance data being slightly in-accurate.
Has anyone (novak / sidehack) gotten any raw 1385's to play with?
P.S. I just noticed that... IF the 1385's were producing 30 gh/s instead of 32.5 gh/s...
162 Chips @ 30gh/s = 4860
IIRC I don't think we've seen a unit come in at spec, but I can't say I've ready EVERY post on the S7's.
My units are exceptional 4.86 rock solid B1's and I'm excited to see what they can do once warranty is up, and closer to ROI.
An associate has several he's been clocking at 625mhz since day one and they still have .00xx HW rates, when he clocked them higher, HW rates suffered and the hash did not increase.
All the empirical evidence so far points to the released performance data being slightly in-accurate.
Has anyone (novak / sidehack) gotten any raw 1385's to play with?
P.S. I just noticed that... IF the 1385's were producing 30 gh/s instead of 32.5 gh/s...
162 Chips @ 30gh/s = 4860
November 17, 2015, 01:09:39 PM
I'm wondering if batch 7 will handle OC over 650 MHZ ?
Unlikely, as even the "gems" of the S7 world start to lose hash and see increased HW errors above 625.
The best way to think about this would be that the batch sevens are "factory overclocked"
I think it's the other way round and Batches 1 to 6 are actually "factory underclocked"
Even batch 7 is underclocked relative to the Bm1385 Data Sheet. If we look at the numbers.
With a 12V supply and an 18 Node String that gives a 0.667V Core Voltage / Chip.
Data Sheet Hash / Chip at 0.66V is 32.5GH/s * 162 Chips = 5265GH/s / 8.1 = 650MHz clock
Whereas we have with Batch 7 5063GH/s /8.1 = 625MHz clock
So we are getting close but still underclocked relative to the Data Sheet.
Rich
November 17, 2015, 01:08:02 PM
almost impossible to ROI (before halving) @ current btc price with such high % diff jump.
PLEASE add in : miner + psu + shipping costs + tax + elec cost (if any) + ISP + Setup costs (if any) + DIFFICULTY ! many forget to add in DIFFICULTY ! next jump is over 10%
even with avg +5% diff jump @ current prices - (pool luck + variance . . . one will almost NEVER get 100% all the time) + FREE power = VERY long time to ROI & NOT to mention halving.
with the math above i am seeing above 200 days range MINIMUM.
btc price @ 334 USD , free internet, 0 breakdowns, USD 150 for shipping & psu, 0 tax, FREE power, pool fee of 2.5% (PPS), 0 setup + maint. costs.
I'm NOT trying to discourage but please do your math.
best part is that you WILL NOT get your miner until shipped & the delivery man knocks on your door so add in a -15% lost due to diff jump till the time you plug it in & start mining.
PLEASE add in : miner + psu + shipping costs + tax + elec cost (if any) + ISP + Setup costs (if any) + DIFFICULTY ! many forget to add in DIFFICULTY ! next jump is over 10%
even with avg +5% diff jump @ current prices - (pool luck + variance . . . one will almost NEVER get 100% all the time) + FREE power = VERY long time to ROI & NOT to mention halving.
with the math above i am seeing above 200 days range MINIMUM.
btc price @ 334 USD , free internet, 0 breakdowns, USD 150 for shipping & psu, 0 tax, FREE power, pool fee of 2.5% (PPS), 0 setup + maint. costs.
I'm NOT trying to discourage but please do your math.
best part is that you WILL NOT get your miner until shipped & the delivery man knocks on your door so add in a -15% lost due to diff jump till the time you plug it in & start mining.
Mining is just as much of a speculative market as the exchanges are when speculating on price. I imagine some, if not most here, are "speculating" the price of bitcoin [Over time] will increase at a higher percentage rate than the percentage rate in which bitcoin price increases. That's what I'm speculating. I have my reasons for investing more in rigs than buying BTC alone. We will see over time if my speculation is right. Although, I must say, I have plans on getting to cheaper power costs soon as well.
i don't think that you can predict anymore. i run the numbers based on 10% rise during next 1-3 cycles, then 4%. As a result, there are $$ deficits by July IF bitcoin price is the same. If bitcoin goes to $700 by July, this will all be in +. I decided not to buy either bitcoin or miners anymore. I had my fill for 2015 already.
November 17, 2015, 12:50:23 PM
I'm wondering if batch 7 will handle OC over 650 MHZ ?
Unlikely, as even the "gems" of the S7 world start to lose hash and see increased HW errors above 625.
The best way to think about this would be that the batch sevens are "factory overclocked"
They have a decent amount of time to perfect as shipping is a way's off. I'm really surprised at speed change, seems they can not be happy with a single design. This will make market on 2nd hand S7's so much harder to shop for and sell with so many variations and no + or other thing to differentiate it.
I wish they would have kept old price though. With difficulty turning nasty... it could be a hard ROI to do.
November 17, 2015, 12:46:36 PM
I'm wondering if batch 7 will handle OC over 650 MHZ ?
Unlikely, as even the "gems" of the S7 world start to lose hash and see increased HW errors above 625.
The best way to think about this would be that the batch sevens are "factory overclocked"
November 17, 2015, 12:01:45 PM
Batch 7 is available, but I can't seem to make a purchase. They kicked it up to 5T.
November 17, 2015, 11:33:41 AM
So, I have a spare 1600W PSU and a few BTC to buy a B7. If I buy a B7, I would be hosting it (already have 6 S7s, some are on their way). Would you buy one if you had the PSU and BTC handy or save the BTC?
November 17, 2015, 11:29:44 AM
almost impossible to ROI (before halving) @ current btc price with such high % diff jump.
PLEASE add in : miner + psu + shipping costs + tax + elec cost (if any) + ISP + Setup costs (if any) + DIFFICULTY ! many forget to add in DIFFICULTY ! next jump is over 10%
even with avg +5% diff jump @ current prices - (pool luck + variance . . . one will almost NEVER get 100% all the time) + FREE power = VERY long time to ROI & NOT to mention halving.
with the math above i am seeing above 200 days range MINIMUM.
btc price @ 334 USD , free internet, 0 breakdowns, USD 150 for shipping & psu, 0 tax, FREE power, pool fee of 2.5% (PPS), 0 setup + maint. costs.
I'm NOT trying to discourage but please do your math.
best part is that you WILL NOT get your miner until shipped & the delivery man knocks on your door so add in a -15% lost due to diff jump till the time you plug it in & start mining.
PLEASE add in : miner + psu + shipping costs + tax + elec cost (if any) + ISP + Setup costs (if any) + DIFFICULTY ! many forget to add in DIFFICULTY ! next jump is over 10%
even with avg +5% diff jump @ current prices - (pool luck + variance . . . one will almost NEVER get 100% all the time) + FREE power = VERY long time to ROI & NOT to mention halving.
with the math above i am seeing above 200 days range MINIMUM.
btc price @ 334 USD , free internet, 0 breakdowns, USD 150 for shipping & psu, 0 tax, FREE power, pool fee of 2.5% (PPS), 0 setup + maint. costs.
I'm NOT trying to discourage but please do your math.
best part is that you WILL NOT get your miner until shipped & the delivery man knocks on your door so add in a -15% lost due to diff jump till the time you plug it in & start mining.
Mining is just as much of a speculative market as the exchanges are when speculating on price. I imagine some, if not most here, are "speculating" the price of bitcoin [Over time] will increase at a higher percentage rate than the percentage rate in which bitcoin price increases. That's what I'm speculating. I have my reasons for investing more in rigs than buying BTC alone. We will see over time if my speculation is right. Although, I must say, I have plans on getting to cheaper power costs soon as well.
November 17, 2015, 11:29:09 AM
so far bitcoin has defined ANY logic. It could stay low and diff still rise, which would finish all home mining at some point.
It is entirely possible that other companies drive this change and not bitmain.
Plus the volatility that makes buying/selling unpredictable.
For 2015 it has tracked very stable. Other then the spike to 500usd.
bitcoin did stay stable (with spikes up and down), but difficulty is 62% (soon to be 72%) higher than 12 mo ago, which favors the lowest cost producers.
November 17, 2015, 11:21:16 AM
We're looking at a huge diff increase with over half a exahash in the network. This is insane. There is going to be such a large increase in hash and we're all going to look at less profit because of the build up...And I can't afford another S7 right now! 
We were at 230 usd and 55 bill diff
We are at 324 usd and 65 bill diff
So far diff has yet to catch up to sept 4 to nov 17th in terms of price gain.
Next on sept 4 the best gear did .48 watts
If you get a s7 it can do 1/2 that.
And winter is here for the north
So if sept 4 numbers were true.
Diff will need to go to 150 bill
But
This leaves out the 1/2 ing. So. I think we move to 90-100 max
so far bitcoin has defined ANY logic. It could stay low and diff still rise, which would finish all home mining at some point.
It is entirely possible that other companies drive this change and not bitmain.
Plus the volatility that makes buying/selling unpredictable.
For 2015 it has tracked very stable. Other then the spike to 500usd.
November 17, 2015, 11:14:37 AM
We're looking at a huge diff increase with over half a exahash in the network. This is insane. There is going to be such a large increase in hash and we're all going to look at less profit because of the build up...And I can't afford another S7 right now!
We were at 230 usd and 55 bill diff
We are at 324 usd and 65 bill diff
So far diff has yet to catch up to sept 4 to nov 17th in terms of price gain.
Next on sept 4 the best gear did .48 watts
If you get a s7 it can do 1/2 that.
And winter is here for the north
So if sept 4 numbers were true.
Diff will need to go to 150 bill
But
This leaves out the 1/2 ing. So. I think we move to 90-100 max
so far bitcoin has defined ANY logic. It could stay low and diff still rise, which would finish all home mining at some point.
It is entirely possible that other companies drive this change and not bitmain.
Plus the volatility that makes buying/selling unpredictable.
November 17, 2015, 11:04:52 AM
We're looking at a huge diff increase with over half a exahash in the network. This is insane. There is going to be such a large increase in hash and we're all going to look at less profit because of the build up...And I can't afford another S7 right now!
We were at 230 usd and 55 bill diff
We are at 324 usd and 65 bill diff
So far diff has yet to catch up to sept 4 to nov 17th in terms of price gain.
Next on sept 4 the best gear did .48 watts
If you get a s7 it can do 1/2 that.
And winter is here for the north
So if sept 4 numbers were true.
Diff will need to go to 150 bill
But
This leaves out the 1/2 ing. So. I think we move to 90-100 max
November 17, 2015, 10:55:54 AM
almost impossible to ROI (before halving) @ current btc price with such high % diff jump.
PLEASE add in : miner + psu + shipping costs + tax + elec cost (if any) + ISP + Setup costs (if any) + DIFFICULTY ! many forget to add in DIFFICULTY ! next jump is over 10%
even with avg +5% diff jump @ current prices - (pool luck + variance . . . one will almost NEVER get 100% all the time) + FREE power = VERY long time to ROI & NOT to mention halving.
with the math above i am seeing above 200 days range MINIMUM.
btc price @ 334 USD , free internet, 0 breakdowns, USD 150 for shipping & psu, 0 tax, FREE power, pool fee of 2.5% (PPS), 0 setup + maint. costs.
I'm NOT trying to discourage but please do your math.
best part is that you WILL NOT get your miner until shipped & the delivery man knocks on your door so add in a -15% lost due to diff jump till the time you plug it in & start mining.
PLEASE add in : miner + psu + shipping costs + tax + elec cost (if any) + ISP + Setup costs (if any) + DIFFICULTY ! many forget to add in DIFFICULTY ! next jump is over 10%
even with avg +5% diff jump @ current prices - (pool luck + variance . . . one will almost NEVER get 100% all the time) + FREE power = VERY long time to ROI & NOT to mention halving.
with the math above i am seeing above 200 days range MINIMUM.
btc price @ 334 USD , free internet, 0 breakdowns, USD 150 for shipping & psu, 0 tax, FREE power, pool fee of 2.5% (PPS), 0 setup + maint. costs.
I'm NOT trying to discourage but please do your math.
best part is that you WILL NOT get your miner until shipped & the delivery man knocks on your door so add in a -15% lost due to diff jump till the time you plug it in & start mining.
November 17, 2015, 10:45:02 AM
i wonder if they just took their 4.86th model, OC'ed it to 625, and tought hey, it works nicelly, let`s charge more and make it a default setting.. !!! 
Most likely that's what happened
November 17, 2015, 10:43:07 AM
i wonder if they just took their 4.86th model, OC'ed it to 625, and tought hey, it works nicelly, let`s charge more and make it a default setting.. !!!
November 17, 2015, 10:35:42 AM
I'm wondering if batch 7 will handle OC over 650 MHZ ?
November 17, 2015, 10:30:49 AM
We're looking at a huge diff increase with over half a exahash in the network. This is insane. There is going to be such a large increase in hash and we're all going to look at less profit because of the build up...And I can't afford another S7 right now!
November 17, 2015, 10:30:11 AM
Difficult to believe that we have yet another variant of the S7. Not sure that I can make much more sense of this one the the previous batches.
Back to the tight spec on the PSU and they have upped the default frequency. I can only think that they have done a rev on the BM1385 or are selecting?
I also can't make the numbers add up.
4860GH/s / 600MHz = 8.1 * 650MHz = 5265GH/s not 5060GH/s
Rich
Back to the tight spec on the PSU and they have upped the default frequency. I can only think that they have done a rev on the BM1385 or are selecting?
I also can't make the numbers add up.
4860GH/s / 600MHz = 8.1 * 650MHz = 5265GH/s not 5060GH/s
Rich
Number 10 on the list of Specifications
Default Frequency = 625 Mhz
It was 650 a few hours ago. Quickly corrected error.
November 17, 2015, 10:28:20 AM
Difficult to believe that we have yet another variant of the S7. Not sure that I can make much more sense of this one the the previous batches.
Back to the tight spec on the PSU and they have upped the default frequency. I can only think that they have done a rev on the BM1385 or are selecting?
I also can't make the numbers add up.
4860GH/s / 600MHz = 8.1 * 650MHz = 5265GH/s not 5060GH/s
Rich
Back to the tight spec on the PSU and they have upped the default frequency. I can only think that they have done a rev on the BM1385 or are selecting?
I also can't make the numbers add up.
4860GH/s / 600MHz = 8.1 * 650MHz = 5265GH/s not 5060GH/s
Rich
Number 10 on the list of Specifications
Default Frequency = 625 Mhz
Duh silly me, once you have read something wrong you just see it that way again......
Took a shot and grabbed 2 of them
https://blockchain.info/tx/6a4d7c58da925f75660a9857ae428c36288c12974b0e9f867b4a4b5a33f93022
https://blockchain.info/tx/23a2bc800605402e285390a35efada040bcb6818d909c42121f1f62677109f8a
fingers crossed that they are true monsters of hash.
Jump to: