ASICMiner BE300S Samples Arrived, <0.2W/G Achieved at Board Level

[sidehack]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

[1714601374]

1714601374

[1714601374]

1714601374

[1714601374]

1714601374

[raskul]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

[Chris_Sabian]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

I disagree.

These chips are ~2-3x more efficient than the next competitor's chips. With mining margins very thin, if you can get 2-3x the hashrate with the same electricity cost, why wouldn't you?  Of course, the price needs to be competitive as well.

[raskul]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

I disagree.

These chips are ~2-3x more efficient than the next competitor's chips. With mining margins very thin, if you can get 2-3x the hashrate with the same electricity cost, why wouldn't you?  Of course, the price needs to be competitive as well.

i think we are getting ahead of ourselves a little. on-paper specs show these chips to be ~2-3x more efficient than the next competitor chips, but while competitors do not have the requirement to release their own specs, then by the time these are baked and ready, that ~2-3x will certainly not be the case.
then we have every h/w manufacturer on the planet vying to build 500TH/s rigs which in turn puts difficulty way up to a level where no mining is profitable.
we are approaching times where manufacturers will need to remain cautious. Super-efficient chips are all very well, but they still have to sell them to a market who are looking for profit.

[Chris_Sabian]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

I disagree.

These chips are ~2-3x more efficient than the next competitor's chips. With mining margins very thin, if you can get 2-3x the hashrate with the same electricity cost, why wouldn't you?  Of course, the price needs to be competitive as well.

i think we are getting ahead of ourselves a little. on-paper specs show these chips to be ~2-3x more efficient than the next competitor chips, but while competitors do not have the requirement to release their own specs, then by the time these are baked and ready, that ~2-3x will certainly not be the case.
then we have every h/w manufacturer on the planet vying to build 500TH/s rigs which in turn puts difficulty way up to a level where no mining is profitable.
we are approaching times where manufacturers will need to remain cautious.  Super-efficient chips are all very well, but they still have to sell them to a market who are looking for profit.

I agree.

The spring will be an interesting time for sure.  I hope FC and AM build a decent size self mine with their cheap power in China.

Right now we are looking at the 2nd difficulty that will +/- 1% or so.  So that would suggest that mining has reached an equilibrium with price for the current generation of chips.  I fear that if price doesn't increase, not many people would be willing to purchase new chips.

[raskul]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

I disagree.

These chips are ~2-3x more efficient than the next competitor's chips. With mining margins very thin, if you can get 2-3x the hashrate with the same electricity cost, why wouldn't you?  Of course, the price needs to be competitive as well.

i think we are getting ahead of ourselves a little. on-paper specs show these chips to be ~2-3x more efficient than the next competitor chips, but while competitors do not have the requirement to release their own specs, then by the time these are baked and ready, that ~2-3x will certainly not be the case.
then we have every h/w manufacturer on the planet vying to build 500TH/s rigs which in turn puts difficulty way up to a level where no mining is profitable.
we are approaching times where manufacturers will need to remain cautious.  Super-efficient chips are all very well, but they still have to sell them to a market who are looking for profit.

I agree.

The spring will be an interesting time for sure.  I hope FC and AM build a decent size self mine with their cheap power in China.

Right now we are looking at the 2nd difficulty that will +/- 1% or so.  So that would suggest that mining has reached an equilibrium with price for the current generation of chips.  I fear that if price doesn't increase, not many people would be willing to purchase new chips.

i concur.

But therein could lie another issue, with spring comes the warmer climate in a lot of the world. it's going to be a very interesting time when it comes to deployment.
I, for one think that it's time all manufacturers should be working in tandem, working together to assure their own profits... the pulling ahead by any great margin by any one could cause problems for the entire network. To be fair, difficulty can be commanded as I believe we are currently seeing with what you comment on regarding difficulty remaining pretty stagnant is something which should be worked on by all. I want every miner to make their profits, and to ensure this, I feel that some form of collaboration between manufacturers is going to be required. what is your take on this?

[Chris_Sabian]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

I disagree.

These chips are ~2-3x more efficient than the next competitor's chips. With mining margins very thin, if you can get 2-3x the hashrate with the same electricity cost, why wouldn't you?  Of course, the price needs to be competitive as well.

i think we are getting ahead of ourselves a little. on-paper specs show these chips to be ~2-3x more efficient than the next competitor chips, but while competitors do not have the requirement to release their own specs, then by the time these are baked and ready, that ~2-3x will certainly not be the case.
then we have every h/w manufacturer on the planet vying to build 500TH/s rigs which in turn puts difficulty way up to a level where no mining is profitable.
we are approaching times where manufacturers will need to remain cautious.  Super-efficient chips are all very well, but they still have to sell them to a market who are looking for profit.

I agree.

The spring will be an interesting time for sure.  I hope FC and AM build a decent size self mine with their cheap power in China.

Right now we are looking at the 2nd difficulty that will +/- 1% or so.  So that would suggest that mining has reached an equilibrium with price for the current generation of chips.  I fear that if price doesn't increase, not many people would be willing to purchase new chips.

i concur.

But therein could lie another issue, with spring comes the warmer climate in a lot of the world. it's going to be a very interesting time when it comes to deployment.
I, for one think that it's time all manufacturers should be working in tandem, working together to assure their own profits... the pulling ahead by any great margin by any one could cause problems for the entire network. To be fair, difficulty can be commanded as I believe we are currently seeing with what you comment on regarding difficulty remaining pretty stagnant is something which should be worked on by all. I want every miner to make their profits, and to ensure this, I feel that some form of collaboration between manufacturers is going to be required. what is your take on this?

I have some miners in my house acting as space heaters to help with the heating bill.  Once spring comes and the miners are no longer profitable (better to buy the coins), I probably won't buy many large ones because of the dual cost of electricity and cooling.

anufacturers working together to fall under some kind of Game theory model.  This would be interesting to discuss though.   I doubt it will happen.  Human greed is always present.  If I can buy more of these chips to 'print money' for myself, why wouldn't I?  Short term gains are, for the most part, more important than long terms gains for many people. 

[MichaelBliss]

I, for one think that it's time all manufacturers should be working in tandem, working together to assure their own profits... the pulling ahead by any great margin by any one could cause problems for the entire network. To be fair, difficulty can be commanded as I believe we are currently seeing with what you comment on regarding difficulty remaining pretty stagnant is something which should be worked on by all. I want every miner to make their profits, and to ensure this, I feel that some form of collaboration between manufacturers is going to be required. what is your take on this?

A Bitcoin mining cartel? I like the way your thinking on that.   And I think we're overdue for one.   They could fix the supply as well as the price...   I think we're off topic at this point though.  

[raskul]

yes, agreed, way OT. apologies, my brain often goes off on a tangent and i find it difficult to stop thinking... i'm hoping that one day it solves a block.
 

[dogie]

I like the idea of building a simple brick-powered introductory miner with USB connection. Something a n00b can buy, inexpensive but reasonably efficient and not requiring advanced computer knowledge or miscellaneous hardware stashes to use and maintain. Just saying, Dogie's not the only one thinking it. I've always said there wasn't enough (or good-enough) hardware in the mid-range consumer sector between USB stick miners and farm-grade hardware.

the problem therein lies that mid range miners, like USB stick miners are pretty much novelty items now. Sales volume won't command the demand on build investments, and making power bricks is just more manufacturing expense.
I do believe however, that there is no requirement to build humongous large-power miners with such efficient chips. FC would do better to keep product under 4TH/s... even that is a good home mining rig at these efficiencies and which can also be deployed in a farm setting.
I guess we'll have to wait and see.

I disagree.

These chips are ~2-3x more efficient than the next competitor's chips. With mining margins very thin, if you can get 2-3x the hashrate with the same electricity cost, why wouldn't you?  Of course, the price needs to be competitive as well.

No, no they're not. Its just the test range displayed is at the very low end of the chip to demonstrate best W/Gh. The actual release W/GH will be significantly higher, as will the GH/chip. But even at .25 at chip, there are already chips that can achieve that at low clock speeds, and many more to come.

[Mabsark]

No, no they're not. Its just the test range displayed is at the very low end of the chip to demonstrate best W/Gh. The actual release W/GH will be significantly higher, as will the GH/chip. But even at .25 at chip, there are already chips that can achieve that at low clock speeds, and many more to come.

According to previous updates, the production chips would have have greater efficiency:

Update

Some details of BE300:

Process: tsmc 28nm hpc

Package: fclga (5mm x 5mm)

Normal Mode:
    0.7v vdd
    6gh/s per chip
    8gh/s-12gh/s per chip for mass production
    0.343w/g on chip
    ~0.3w/g on chip for mass production

Low Power Mode:
    0.55v vdd
    4.5gh/s per chip
    6gh/s-9gh/s per chip for mass production
    0.225w/g on chip
    ~0.2w/g on chip for mass production

The schedule of BE300 producing: First batch production will be done next Feb.

So, why are you assuming that they will be less efficient?

[Dexter770221]

According to previous updates, the production chips would have have greater efficiency:

Normal Mode:
    0.7v vdd
    6gh/s per chip
    0.343w/g on chip

And looks like real performance of samples is better than predicted:

...
    6.4GH/s | 0.2363W/G
    6.8GH/s | 0.2439W/G
    7.2GH/s | 0.2495W/G
...

[dogie]

No, no they're not. Its just the test range displayed is at the very low end of the chip to demonstrate best W/Gh. The actual release W/GH will be significantly higher, as will the GH/chip. But even at .25 at chip, there are already chips that can achieve that at low clock speeds, and many more to come.

According to previous updates, the production chips would have have greater efficiency:

Update

Some details of BE300:

Process: tsmc 28nm hpc

Package: fclga (5mm x 5mm)

Normal Mode:
    0.7v vdd
    6gh/s per chip
    8gh/s-12gh/s per chip for mass production
    0.343w/g on chip
    ~0.3w/g on chip for mass production

Low Power Mode:
    0.55v vdd
    4.5gh/s per chip
    6gh/s-9gh/s per chip for mass production
    0.225w/g on chip
    ~0.2w/g on chip for mass production

The schedule of BE300 producing: First batch production will be done next Feb.

So, why are you assuming that they will be less efficient?

You're missing the point. There are 4 variables, not 2. W/GH, GH/chip and $/GH [and so $/chip]. In order to maximise $/chip, GH/chip will be raised much higher for a retail product than you see in those tests. That in turn, increases W/GH.

I'm not saying the chips can't achieve those lower W/Gh, I'm just saying don't be disappointed when the products based on those chips aren't .25W/GH. [This is of course ignoring that chip W/GH also isn't = to product W/GH.]

[aerobatic]

According to previous updates, the production chips would have have greater efficiency:

Normal Mode:
    0.7v vdd
    6gh/s per chip
    0.343w/g on chip

And looks like real performance of samples is better than predicted:

...
    6.4GH/s | 0.2363W/G
    6.8GH/s | 0.2439W/G
    7.2GH/s | 0.2495W/G
...

while the power numbers are certainly good, there are two questions i'd have.

the main goal when doing bitcoin mining is balancing the capital cost and the operational cost.

the capital cost can to a large extent by expressed as a performance/density figure, GH/MM, and the operational cost expressed as a Power figure W/GH/S (aka J/GH).  the remaining system costs (fans, power supplies, pcbs etc.. are largely constant regardless of the silicon used, so the main factor is efficiency and cost of the silicon itself)

clearly, many designers are working on asics that run at low voltages and thus achieve a headline low power figure.

The key is to achieve LOW POWER, while at the same time, ensuring that the silicon cost is low.

So it looks like the nominal performance of the be300 chip is somewhere between 3 and 6 GH/s (or 2.8 to 7.2 gh/s) per package.
And we've been told the package size is a 5x5mm square.   Since we havnt been told the die size... lets try and guess.  clearly its nowhere near 5x5 mm... so perhaps its 3x3 (=9 mm) or 4x4 (=16 mm)?

it makes a difference how big the die area is, as the gigahashes per mm need to be significantly better than last year's asics if its to be competitive in 2015.   it needs to be above 3 GH/mm to be competitively priced, and ideally north of 5 gh/mm.  And if... completely guessing... the die is... lets say its a 2.5 x 2.5mm, (= 6.25mm) then its going to be lower than 1 GH/mm, and thus will be expensive compared to other competitors coming out in 2015.   Bear in mind that last year's chips were all above 1 gh/mm, also in the tsmc 28nm process.  this year's ones should be significantly more efficient in order to bring down the cost of mining.

[Syke]

the capital cost can to a large extent by expressed as a performance/density figure, GH/MM, and the operational cost expressed as a Power figure W/GH/S (aka J/GH).  the remaining system costs (fans, power supplies, pcbs etc.. are largely constant regardless of the silicon used, so the main factor is efficiency and cost of the silicon itself)

Possibly. You still need to know the ratio of silicon to other system parts costs. If a $100 miner is made up of $1 worth of chips, and $99 worth of other parts, doubling the cost of the chips only makes the miner cost $101. Unfortunately I don't know how much the raw chips cost to know if it significant or not relative to total system cost. The silicon could be a significant portion of the total system cost by now. It didn't used to be.

[aerobatic]

Possibly. You still need to know the ratio of silicon to other system parts costs. If a $100 miner is made up of $1 worth of chips, and $99 worth of other parts, doubling the cost of the chips only makes the miner cost $101. Unfortunately I don't know how much the raw chips cost to know if it significant or not relative to total system cost. The silicon could be a significant portion of the total system cost by now. It didn't used to be.

its the goal of the system designer to minimise the system and maximise the contribution of the silicon to the system.. and asicminer have been good at this.  their recent systems are quite efficient and their new one, with string support, should be even more so  - making most of the cost of the system will be the cost of the silicon.  that silicon needs to be extremely cost efficient on a gh/mm basis (which translates directly to gh/$), especially as so many mining asic suppliers are using the same tsmc foundry and similar process, thus their wafer costs are near identical, so the only differentiator is going to be the silicon efficiency of both power and cost.  ultimately, its gigahashes per wafer that directly affect the cost.

[antirack]

@aerobatic: how much does slicing, packaging and testing contribute to the production cost of Bitcoin ASICs?

And separately, how about the effect of yields on cost?

I would assume that AM has experience now in this field, after 2 generations of chips and battling with problems?!

[aerobatic]

@aerobatic: how much does slicing, packaging and testing contribute to the production cost of Bitcoin ASICs?

And separately, how about the effect of yields on cost?

I would assume that AM has experience now in this field, after 2 generations of chips and battling with problems?!

bitcoin mining chips, with their large number of replicated cells (hash engines) are very good for yield - probably the best possible case that a foundry could wish for, as often they can still find useful and functional chips even if not all the hash engines are operational... whereas if you're making Apple A6 chips, even a single defect is fatal and the chip will be rejected.  you need every single circuit to work and any single defect on the die spoils the entire die, wheres for bitcoin mining, they can cope with some, and perhaps even a lot of defects and still have a useful and valuable asic.  for that reason, its not uncommon for bitcoin asics to have exceptional yield compared to regular asics, thus the yield factor doesnt become as important as the cost of each wafer and the efficiency of the overall design.

that said, as you know, wafers are round, and asics dies are rectangular.. thus the larger the die, the less number of dies fit onto a wafer.. because of wastage around the edges of the wafer where the circle intersects the many rectangles.. and if any part of th rectangular die touches the edge, the die is of course unusable.  asicminer uses very small dies, which should mean they will get less wastage at the edges of the wager and should have more usable dies per wafer.  on the other hand, the edges of each die where they're sliced puts a border around each die, which itself adds some wastage (die streets) and a larger area of the wafer is given over to edges versus usable die area on bigger dies.

i think asicminer uses small dies now because they have always done... whereas there would be less packaging cost, and lower board costs, jf they used larger dies with more hash engines per die.  there's some ineffiencies in using small chips that are not fully optimum.  in bitcoin asics... there are also various efficiencies of using small dies.. in that they might be able to be air cooled without individual heatsinks.. which saves cost and manufacturing time...   however, having each tiny die have its own package could become more costly than having larger dies with fewer packages.   personally, i think that larger dies are the way to go.  i think having tiny dies, thus high numbers of packages and larger boards etc.. adds more points of failure, and more board power losses.. so I'm in favour of larger dies in general.  not necessarily huge ones any longer (as they require exotic cooling, which we've all seen isn't as reliable).

[Syke]

personally, i think that larger dies are the way to go.  i think having tiny dies, thus high numbers of packages and larger boards etc.. adds more points of failure, and more board power losses.. so I'm in favour of larger dies in general.  not necessarily huge ones any longer (as they require exotic cooling, which we've all seen isn't as reliable).

Large dies are very hard to power and cool. Just compare the companies and die sizes and you'll see a very strong corelation.

Late Companies
BFL: Large die
HF: Large die
CT: Large die
KNC: Large die

On time companies
Avalon: Small die
Bitfury: Small die
Asicminer: Small die
Bitmain: Small die
Spondoolies: Medium die

It's pretty clear that the companies that tried big dies has the most delays. I do agree the Asicminer dies could grow quite a bit and still work well.

[sidehack]

I don't know about a sea of tiny chips, but I'd much rather see a multi-chip high-hashrate design than a single- or few-chip high-hashrate design. Increased surface area decreases cooling requirements, distributed power decreases PCB requirements, and increased modularity increases reliability and repairability. From an owner/hardware-maintainer standpoint, the most problematic machines I've run were the ones with hot single-die setups.