BITMAIN launches 4th generation Bitcoin mining ASIC: BM1385

[dogie]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

[1714002464]

1714002464

[1714002464]

1714002464

[1714002464]

1714002464

[brontosaurus]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

Thats why you employ people that know what they are doing and use MPW runs. Not rocket science, is it?

[brontosaurus]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

Oh, and 'nearly eight figures'?

[mavericklm]

waiting for the s7+ monster!!! 15th and 3kw

[SockPuppetAccount]

Did you buy your used A2 for the same or more than the guy who sold it to you bought it new? The point is not that selling hardware secondhand is in general a bad thing - by no means am I saying that.
The point is selling hardware secondhand for more than it is likely actually worth in order to recoup your investment is a bad thing - which is to say, "making about 1/2 or so back on your money and you sell them used for just about the price you paid for them new" (if not more).

Unless something atypical like flat diff is happening (which can't be counted on enough to "rinse and repeat") and the viability of the machine is actually retained during that time, selling it for near new price after several months means ripping someone off. You get your "ROI" by effectively stealing it from someone else. That's exactly what's happening with Bitmain selling used S5 for the same or more than they sold them new several months ago. If the diff had gone up more than about 10% in the last six months (like the 50-100% that would have been expected had the trends of the previous two years kept steady), nobody in their right mind would be buying S5 for the same money they paid at Christmas.

I feel the argument should be well-qualified enough to require no further defense, and as it's pretty off-topic at this point, I'll say nothing further on the subject. How 'bout them BM1385 chips, eh?

Huh?  I use the same strategy with my ASIC purchases and obviously there is a lot of guesswork and a little bit of luck involved, but I have profited off the S1, S3, S5, and SP20 by using this same strategy.  When I decide to sell my old gear, I do so based on what I paid for it, how quickly difficulty is rising, what I have mined so far, the season of the year (save on heating bill in winter), and the approximate going rate for the miner on the second hand market.  I list the miner on ebay at 0.01 cents with no reserve, offer free shipping, set it for a week, and thats it.  Knowingly misrepresenting a miner as new, something it isn't, higher performance or better condition than it is, etc are all bad things.  But if the free market is overvaluing mining gear at a particular moment in time, I'm stealing from someone if I sell it?  That doesn't make sense.  Some people have cheap or free electricity, some people dont care about ROI and it's just a toy for them, some people are simply making a poorly timed purchase because they don't know how to properly assess the value of the mining gear.  It doesn't matter.  As long as they are freely choosing to purchase the item and I am not misleading them, there is nothing wrong with it.  It's value is exactly what the free market will pay for it.   Sometimes the free market overvalues a product or service, sometimes the free market undervalues a product or service.  If you are able to accurately gauge the true value of an item better than the rest of the market, whether it is the ASIC market, stock market, or hell, the grocery market, you will gain anytime you are able to sell something while it is overvalued and buy something when it is undervalued.  This is just how a free market capitalistic economy and natural price discovery functions.

[HyperMega]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

Oh, and 'nearly eight figures'?

If Bitmain spend really nearly $10,000,000 NRE for this 28nm ASIC they either have very well paid designers or were completely robbed by TSMC for the masks. 

[windpath]

Time to start saving coin

I like this reply best.

I am glad that:

a) I did not crack and buy the s-5++
b) I grabbed some coin when prices dropped yesterday.

I like both A & B...

Between Bitmain, and whatever Spondoolies and 21 has up there sleeves, it looks like we are on track for some record difficulty increases again, and just before the halving....

[sidehack]

Did you buy your used A2 for the same or more than the guy who sold it to you bought it new? The point is not that selling hardware secondhand is in general a bad thing - by no means am I saying that.
The point is selling hardware secondhand for more than it is likely actually worth in order to recoup your investment is a bad thing - which is to say, "making about 1/2 or so back on your money and you sell them used for just about the price you paid for them new" (if not more).

Unless something atypical like flat diff is happening (which can't be counted on enough to "rinse and repeat") and the viability of the machine is actually retained during that time, selling it for near new price after several months means ripping someone off. You get your "ROI" by effectively stealing it from someone else. That's exactly what's happening with Bitmain selling used S5 for the same or more than they sold them new several months ago. If the diff had gone up more than about 10% in the last six months (like the 50-100% that would have been expected had the trends of the previous two years kept steady), nobody in their right mind would be buying S5 for the same money they paid at Christmas.

I feel the argument should be well-qualified enough to require no further defense, and as it's pretty off-topic at this point, I'll say nothing further on the subject. How 'bout them BM1385 chips, eh?

Huh?  I use the same strategy with my ASIC purchases and obviously there is a lot of guesswork and a little bit of luck involved, but I have profited off the S1, S3, S5, and SP20 by using this same strategy.  When I decide to sell my old gear, I do so based on what I paid for it, how quickly difficulty is rising, what I have mined so far, the season of the year (save on heating bill in winter), and the approximate going rate for the miner on the second hand market.  I list the miner on ebay at 0.01 cents with no reserve, offer free shipping, set it for a week, and thats it.  Knowingly misrepresenting a miner as new, something it isn't, higher performance or better condition than it is, etc are all bad things.  But if the free market is overvaluing mining gear at a particular moment in time, I'm stealing from someone if I sell it?  That doesn't make sense.  Some people have cheap or free electricity, some people dont care about ROI and it's just a toy for them, some people are simply making a poorly timed purchase because they don't know how to properly assess the value of the mining gear.  It doesn't matter.  As long as they are freely choosing to purchase the item and I am not misleading them, there is nothing wrong with it.  It's value is exactly what the free market will pay for it.   Sometimes the free market overvalues a product or service, sometimes the free market undervalues a product or service.  If you are able to accurately gauge the true value of an item better than the rest of the market, whether it is the ASIC market, stock market, or hell, the grocery market, you will gain anytime you are able to sell something while it is overvalued and buy something when it is undervalued.  This is just how a free market capitalistic economy and natural price discovery functions.

Alright, maybe I should simplify a bit. Selling used miners at fair market value is good. In the past five months or so, with relatively flat diff, the FMV for something like the S5 has been relatively unchanged, so running one for five months and then selling it for new price actually works.
However, for the several years before then, and for the future probably starting within the next two months, the diff is not flat. So the current conditions are somewhat of an aberration.
The guy is saying the proper course of action has been, and will be, to buy a miner, run it, and sell it for approximately new price.
So let's say someone buys a $500 miner, expected breakeven about 9 months out with typical power cost. He then runs it for three months. The diff goes up 25% in that time (4% biweekly increase, historically fairly reasonable). He then tries to sell it for $500. Would not you assume he's a fool? Would not you assume anyone buying it would also be a fool? Would not the expected market value for that $500 machine have dropped, probably to below $400? With the exception of the past few months, this has been the case since the first day of ASIC mining (or FPGA mining, probably even GPU days). Depreciation keeps pace with diff increase (and is slowed by increasing coin price, but historically increasing coin price increases the diff even faster). Take a look at what S1 prices did in early 2013. Diff went way up really fast, coin prices went down, and the price for an S1 dropped about 90% in something like four months. You think my December S1 would have sold in March for the 4BTC paid for it, or more like 8BTC if priced in dollars? Only to a fool.

So I'd say that, in general, with some exceptions, someone who takes a sizeable chunk out of the viable life of a miner and then sells it at new price is not selling at expected fair market value, and someone that buys it probably didn't do his research to know he was being ripped off.

The point is not, and never has been, that selling secondhand hardware at fair market value is bad. The point is, and always has been, that successfully selling used hardware at well above market value is taking advantage of a fool, which is unethical. Implying that the proper strategy for making positive returns on mining is (and always has been) taking advantage of fools, well I like to think this community is better than that.

[dogie]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

Oh, and 'nearly eight figures'?

If Bitmain spend really nearly $10,000,000 NRE for this 28nm ASIC they either have very well paid designers or were completely robbed by TSMC for the masks. 

I'm talking about smaller process nodes and in huge batches. Ie its a safer bet to get not quite as good 28nm performance then risk a lot on trying something new will full custom.

[alienesb]

Did you buy your used A2 for the same or more than the guy who sold it to you bought it new? The point is not that selling hardware secondhand is in general a bad thing - by no means am I saying that.
The point is selling hardware secondhand for more than it is likely actually worth in order to recoup your investment is a bad thing - which is to say, "making about 1/2 or so back on your money and you sell them used for just about the price you paid for them new" (if not more).

Unless something atypical like flat diff is happening (which can't be counted on enough to "rinse and repeat") and the viability of the machine is actually retained during that time, selling it for near new price after several months means ripping someone off. You get your "ROI" by effectively stealing it from someone else. That's exactly what's happening with Bitmain selling used S5 for the same or more than they sold them new several months ago. If the diff had gone up more than about 10% in the last six months (like the 50-100% that would have been expected had the trends of the previous two years kept steady), nobody in their right mind would be buying S5 for the same money they paid at Christmas.

I feel the argument should be well-qualified enough to require no further defense, and as it's pretty off-topic at this point, I'll say nothing further on the subject. How 'bout them BM1385 chips, eh?

Huh?  I use the same strategy with my ASIC purchases and obviously there is a lot of guesswork and a little bit of luck involved, but I have profited off the S1, S3, S5, and SP20 by using this same strategy.  When I decide to sell my old gear, I do so based on what I paid for it, how quickly difficulty is rising, what I have mined so far, the season of the year (save on heating bill in winter), and the approximate going rate for the miner on the second hand market.  I list the miner on ebay at 0.01 cents with no reserve, offer free shipping, set it for a week, and thats it.  Knowingly misrepresenting a miner as new, something it isn't, higher performance or better condition than it is, etc are all bad things.  But if the free market is overvaluing mining gear at a particular moment in time, I'm stealing from someone if I sell it?  That doesn't make sense.  Some people have cheap or free electricity, some people dont care about ROI and it's just a toy for them, some people are simply making a poorly timed purchase because they don't know how to properly assess the value of the mining gear.  It doesn't matter.  As long as they are freely choosing to purchase the item and I am not misleading them, there is nothing wrong with it.  It's value is exactly what the free market will pay for it.   Sometimes the free market overvalues a product or service, sometimes the free market undervalues a product or service.  If you are able to accurately gauge the true value of an item better than the rest of the market, whether it is the ASIC market, stock market, or hell, the grocery market, you will gain anytime you are able to sell something while it is overvalued and buy something when it is undervalued.  This is just how a free market capitalistic economy and natural price discovery functions.

Alright, maybe I should simplify a bit. Selling used miners at fair market value is good. In the past five months or so, with relatively flat diff, the FMV for something like the S5 has been relatively unchanged, so running one for five months and then selling it for new price actually works.
However, for the several years before then, and for the future probably starting within the next two months, the diff is not flat. So the current conditions are somewhat of an aberration.
The guy is saying the proper course of action has been, and will be, to buy a miner, run it, and sell it for approximately new price.
So let's say someone buys a $500 miner, expected breakeven about 9 months out with typical power cost. He then runs it for three months. The diff goes up 25% in that time (4% biweekly increase, historically fairly reasonable). He then tries to sell it for $500. Would not you assume he's a fool? Would not you assume anyone buying it would also be a fool? Would not the expected market value for that $500 machine have dropped, probably to below $400? With the exception of the past few months, this has been the case since the first day of ASIC mining (or FPGA mining, probably even GPU days). Depreciation keeps pace with diff increase (and is slowed by increasing coin price, but historically increasing coin price increases the diff even faster). Take a look at what S1 prices did in early 2013. Diff went way up really fast, coin prices went down, and the price for an S1 dropped about 90% in something like four months. You think my December S1 would have sold in March for the 4BTC paid for it, or more like 8BTC if priced in dollars? Only to a fool.

So I'd say that, in general, with some exceptions, someone who takes a sizeable chunk out of the viable life of a miner and then sells it at new price is not selling at expected fair market value, and someone that buys it probably didn't do his research to know he was being ripped off.

The point is not, and never has been, that selling secondhand hardware at fair market value is bad. The point is, and always has been, that successfully selling used hardware at well above market value is taking advantage of a fool, which is unethical. Implying that the proper strategy for making positive returns on mining is (and always has been) taking advantage of fools, well I like to think this community is better than that.

So if a farmer buys a tractor and uses it a couple of seasons then decides to sell it, and because that tractor is no longer made but is still in demand, he manages to get just about what he paid for it back is that wrong too? It's supply and demand.

At one point Bitmain sold miners in the $340 range then at another they were selling for over $450ish and at times they just were not available. It's about percieved value to the buyer and that is not ripping someone off.

You have a rather jaded view of the secondary market. I bet there will be people selling your stick at higher than sticker price on eBay.

[flikflak]

Wouldnt it be nice to have some kind of standard in the BITMAIN world?
I'm thinking of just changing the blades and the firmware to upgrade an S5 to an S7. It would reduce the prices a bit (lets say 10-15%, its shipping costs, VAT ..) and it would also kinda bond the customer to the company as well.

[sidehack]

If another reputable manufacturer makes a tractor with 2 more cylinders, an air-conditioned cab and burns less fuel for the same amount of work, and sells it new at the same price the farmer wants for his, which would you recommend someone buy? If the market value of the used tractor is now 2/3 what he paid new, and he asks new price for it, would you buy it?

If, however, the several-year-old tractor is pretty much exactly equivalent to current offerings, your comparison falls into the market conditions of the last few months and is therefore irrelevant to the argument as a whole since the last few months' conditions are not representative of the market so far, nor the expected market in the future. You might look instead at the time depreciation of miner prices between early 2013 and early 2015, which instead of being a bad analogy of the argument, is the actual argument.

When Bitmain raised the S5 price from $320 to $420 in a day, did you see anyone at all happy that they'd readjusted their prices to what customers thought was more fair, or was literally everyone complaining about being ripped off?

Anyone who wants to buy one of my sticks from me at what I consider a fair price is welcome to do so. Anyone who instead buys from a reseller will probably still get a decent deal; if not, a little bit of looking will lead them to a better price and anyone asking for grossly more than fair market [for profit, not expenses] will not have my endorsement. Even if the reseller market determines they can be sold for more than the $25 I'm asking, my price will not change. But considering that my sticks, even at the best price I the manufacturer can offer, will not produce a positive return (and that is not their intention), the point is erroneous.

I hate greed, and I hate opportunism. I know it's the way of the world (and especially the bitcoin economy) but that doesn't make it right.

What I'm really annoyed by right now is that we're arguing two different points. People keep insisting on counterexamples where the fair market value of the object is retained, but my argument is that, for most of the history of ASIC sales and for the perceived future, the fair market value depreciates rapidly and therefore selling depreciated hardware (probably to someone unaware of the market value) without a reduced price is wrong (by taking advantage of the unaware purchaser). Just because depreciation stalled in the last few months (compared to the two full years before it, and the expected future) doesn't mean a strategy that works right now has been by any means a fair strategy thus far in the game, or will continue to be so. If you haven't comprehended that by now, you're never going to.

[WBF1]

...taking advantage of the unaware purchaser...

Lots of different sectors of the economy rely on scarcity of information or unwillingness of consumers to shop around. Off the top of my head, I can think of auto mechanics (not all of them obviously, but $80+ for an oil change with an extra $80 on top to replace the cabin air filter after they show you the external "dirty" side of it is a common scheme among the "drive thru" style oil change and maintenance shops), most things related to home maintenance and improvement (windows, doors, HVAC equip, roofing, etc), medical care, etc. Basically anything you cannot easily and readily do a cost comparison of is an industry that (at least from a sales aspect) relies on scarcity of information.

Put differently, they rely on the laziness of the consumer.

Sidehack I really like everything you have done and am excited to see what you come up with in the future, but you gotta realize that the price of goods is equal to what people are willing to pay for them. Full stop.

Am I advocating people try to rip off unknowing consumers? No. but I'm also not advocating overt hand-holding for the sake of "fairness". There's an aspect of Darwinism at play here I guess is what I'm arguing. If someone wants to "get rich quick" on this "bitcoin thingy" and is too lazy to do their research, then maybe they didn't "deserve" (from a survival of the fittest standpoint, not from a morality or "fairness" standpoint) to be sitting on a bunch of cash to begin with.

All that being said, being a white knight and pointing out bad deals and attempting to educate consumers is fair game as well.

[MidwestMiner]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

Oh, and 'nearly eight figures'?

If Bitmain spend really nearly $10,000,000 NRE for this 28nm ASIC they either have very well paid designers or were completely robbed by TSMC for the masks. 

$10 million NRE is 16 nm costs ffs. Of course they didn't drop that on 28 nm Outside of development costs (time/staff) NRE for 28nm is sub $5 million with margin.

[HyperMega]

Before anyone gets too excited over this development I would wait until you know what the die size is and the operating frequency at quoted power consumption.

I am surprised that they say that full custom poses a higher 'risk' - that's only true for very complex chips like cpu's, not for the very simple (and I mean very simple) functions found in SHA256.

Good luck to them though, anything that puts a spanner in the works of KNC or 21 has to be welcomed.

Well when you're considering laying down nearly 8 figure sums without seeing the chip work first, you better be damn sure its going to work.

Oh, and 'nearly eight figures'?

If Bitmain spend really nearly $10,000,000 NRE for this 28nm ASIC they either have very well paid designers or were completely robbed by TSMC for the masks. 

$10 million NRE is 16 nm costs ffs. Of course they didn't drop that on 28 nm Outside of development costs (time/staff) NRE for 28nm is sub $5 million with margin.

Of course, I was just kidding, because the "8 figure sum" was mentioned together with the risks of a 28nm custom design.

[sidehack]

WBF1 - if it explains my mindset a bit, my entire family are relatively self-sufficient (which is to say, employed or at least skilled in auto mechanics, construction, farming and other trades) Christians and tend to be both educated and educators. I'm very much in favor of treating people ethically whether they ask for that treatment or not. I'm in favor of helping people improve their stations rather than taking advantage of their lack of awareness. I'm in favor of people learning to do things for themselves rather than relying on others (who will, per your apt examples, likely seek opportunistic profit dishonestly). I will never not advocate for those things, and I will never endorse an action which results in profit through deliberate deception or fleecing the undereducated. I'm just as embarassed by the people making bad decisions because they wanted to "get rich quick" and didn't take time to properly research as I am by the people wanting to "get rich quick" and finding ways to capitalize on someone's lack of knowledge. The one is a fool, the other is a thief. So that's where I'm coming from when I say selling well above fair market value is unethical.



I wonder what machine prices for the future 14- and 16-nm stuff will be, given the high initial costs for chips and the increased cost due to low chip yields? As efficiency asymtotically approaches a practical minimum value, gear should be viable for longer periods especially if undervolting is commonplace. I expect to see low-clocked high-density 28nm machines still be very competitive with at least the first generation of top-clocked 14nm gear, based heavily on low initial cost rather than low operating costs. Increasing initial investment for chips by approximately an order of magnitude will definitely have a large effect on the resulting machine price. The questions's going to be the halving, and whether super-low-power gear is available before that occurs.

I really hope Bitmain lets out some sample chips for us to play with sometime soon.

[Tupsu]

Wouldnt it be nice to have some kind of standard in the BITMAIN world?
I'm thinking of just changing the blades and the firmware to upgrade an S5 to an S7. It would reduce the prices a bit (lets say 10-15%, its shipping costs, VAT ..) and it would also kinda bond the customer to the company as well.

You want  standard ?
Let me give one free.

Tested. S5 + works with  S5 18 pin controller board.  If this is not the standard?
You do not even need to change the software.

Now you can do from yor  S5+  three miners .

[dogie]

Wouldnt it be nice to have some kind of standard in the BITMAIN world?
I'm thinking of just changing the blades and the firmware to upgrade an S5 to an S7. It would reduce the prices a bit (lets say 10-15%, its shipping costs, VAT ..) and it would also kinda bond the customer to the company as well.

This was discussed a while ago in design, and the conclusion was that its not actually that desirable for the same reason that S2 upgrade kits never appeared. By the time the next generation appears, the previous generation is still mining away perfectly happily and so you can't immediately replace those hashing boards.

Ie if the S5 was modular to the S7, you'd still buy a full S7 because any S5s you had would still be fine to run for the foreseeable future. No one would scrap their S5, they'd simply run both or sell the S5.

[Meech]

Wouldnt it be nice to have some kind of standard in the BITMAIN world?
I'm thinking of just changing the blades and the firmware to upgrade an S5 to an S7. It would reduce the prices a bit (lets say 10-15%, its shipping costs, VAT ..) and it would also kinda bond the customer to the company as well.

This was discussed a while ago in design, and the conclusion was that its not actually that desirable for the same reason that S2 upgrade kits never appeared. By the time the next generation appears, the previous generation is still mining away perfectly happily and so you can't immediately replace those hashing boards.

Ie if the S5 was modular to the S7, you'd still buy a full S7 because any S5s you had would still be fine to run for the foreseeable future. No one would scrap their S5, they'd simply run both or sell the S5.

So did you know for sure about the S2 uprade kits not being produced and held back a response?  They only failed do to Bitmain not following through on not one but both announcements.  It's a shame to let a good case and design that's accomodating and easily upgradeable die. 

[flikflak]

@dogie

Of course, an S5 isnt obsolete when the S7 has been released - you have your points here.

@Tupsu

Thanks for the pics- that looks cute. So it is quite possible.