No, just your assumptions on difficulty. I don't expect any of these 60 GH range ASIC products to make even $1000 in 2013 alone. By design, difficulty will always bring the cost-to-mine very close to power costs. When it catches up, power costs will be all that matters. Until then, delivery dates before it adjusts are the key importance.
Agreed. The people who have the view that the power doesn't matter only have it because they have high profitability expectations. The people saying that power is all that matters are expecting very long payoff horizons due to difficulty increases. I think the latter view is safer and also more correct, especially if you're not counting on being very early in your deployment. Which I think results is another interesting bit to take away from the discussion: BFL is apparently not expecting their customers purchases to pay for themselves for quite a long time. This is fine by me, as it's also what I expect— an I mine for fun and to support Bitcoin... but if you were thinking otherwise you might want to carefully review your expectations. Although I generally try to take a conservative approach, in this particular case I would tend more towards the former idea: that power is not likely to play a significant role immediately (or more accurately, it is only PART of the equation). Rather, it is the ROI - Return On Investment - that matters. ROI takes into account both net income (from mining) and expenses (capital cost and running cost). Consider this: I a purchased a mining device 6 months ago for $600 that generates $3/day and costs $0.25/day to operate. Assuming these numbers don't change over time, I can expect an ROI of about 220 days. Not bad. Now consider a hypothetical device identical to the one above, except that it uses TWICE the power. Thus costing me $0.50/day to operate. Now instead of an ROI of 220 days I can expect an ROI of 240 days. A 10% increase. Is this significant? I would argue that it is not; there is little difference between 220 days and 240 days. Both are equally reasonable. Let's take this further: if my running costs (electricity) ever became a SIGNIFICANT percentage of the mining income, I would stop mining altogether. I imagine many other miners would as well. For instance, back to my original device making $3/day and costing $0.25/day to operate : if difficulty went up so high to reduce my mining income to, say, $1/day, I would be unlikely to purchase another one. The ROI at that point would become too long (800 days) and it would not be worth buying such a device. Especially in an unstable and risky ecosystem as bitcoin. So I don't believe that difficulty will ever rise to a level where running costs become a very significant percentage of the gross income; the ROI would simply become too long and many people would stop mining (or, perhaps more pragmatically, people would become increasingly unlikely to invest in new mining hardware, thus capping difficulty). In this case, the first ASIC manufacturer to diversify into different cyrptocurrencies would have the edge in second and third gen mining devices. |
|
|
|
No, just your assumptions on difficulty. I don't expect any of these 60 GH range ASIC products to make even $1000 in 2013 alone. By design, difficulty will always bring the cost-to-mine very close to power costs. When it catches up, power costs will be all that matters. Until then, delivery dates before it adjusts are the key importance.
Agreed. The people who have the view that the power doesn't matter only have it because they have high profitability expectations. The people saying that power is all that matters are expecting very long payoff horizons due to difficulty increases. I think the latter view is safer and also more correct, especially if you're not counting on being very early in your deployment. Which I think results is another interesting bit to take away from the discussion: BFL is apparently not expecting their customers purchases to pay for themselves for quite a long time. This is fine by me, as it's also what I expect— and I mine for fun and to support Bitcoin, and not because I'm expecting to make a a lot of funds doing it... but if you were thinking otherwise you might want to carefully review your expectations. Of course, at the moment— there is still the potential of getting ASIC based miners before they are widely deployed and the difficulty catches up, so thats a competitive factor thats hard to reason about. Though you can probably count on it not happening for you if you're last on a long preorder backlog… +1 I agree with this. Seems they (BFL) don't expect the payoff to be very sudden. I just wonder what the PR jackknife will be like if the difficulty suddenly skyrockets on a mere 60GH rig. The electricity argument them becomes very valid in the immediate future rather than in the further future. The ultimate question is whether people will make their deadlines or not. I am thinking...not. |
|
|
|
Their website is directly connected to the hashing speed firmware calculator:)
So they just have to update the statement for GH/s. voila a new examhash miner is born....
Uhhh what might happen when bitcoin becomes self aware?:/
  Someone must stop BFL before it is too late. Someone contact D.A.R.P.A....quickly! http://www.darpa.mil/http://en.wikipedia.org/wiki/DARPA |
|
|
|
If the answer were no, why would we be doing it? The answer must necessarily be yes.
From my perspective, putting the emotional issues (including history) aside, it's completely feasible that a company could produce ASIC-based mining systems without the investments Enigma mentioned. You or I could do it in our garages.
So the answer turns out to be... Enigma is full of it. Response Enigma? No, not even close.... Enigma mentioned plenty of real points that someone working at one of those fabrication plants would recognize as "an issue". He points out that more demanding hardware fabrication projects would need better hardware than those cited by BFL. Assuming that BFL hired a good consultant that understood their current and future needs, there should be no problems. If the consultant didn't do a good job, then BFL just spent about 50k on inadequate equipment. (Thats Enigmas [professional] opinion) Though Enigma does not know what techniques are employed to build the project that BFL is working with, so he does not ultimately know (for sure) if BFLs in-house equipment is adequate for the job. He only states that it would be a poor choice for most jobs hes worked on. Again, it all depends on whether BFL purchased the right equipment for the project or if it will be inadequate for the task. They (BFL) could be using very rudimentary fab techniques and therefore there is no issue. Or they could have a "Oh Sh*t!" or "Doh!" moment when they try to do the fab in-house. Hopefully the consultant or advice they were given was good advice...then there would be no problems. ------------------------ As BFL_Josh/Inaba said, if there is a problem it won't show up in the first batch (that is being done by a separate fab house). If they do it for the second batch and there are problems, then they will default to plan B and use a separate fab house if the equipment doesn't do what they hope it will do. |
|
|
|
But carrying the analogy, would Enigma say that what you're doing in your garage is simply not enough and you need to get a million dollars worth of tools and equipment from AMG? That's the analogy.
I don't know if he's correct is my point. Assuming that he is correct if the average BFL customer will settle for nothing less than 100% perfect, is he still correct if they are willing to restart their machines every week or so?
You can carry most analogies too far, and I think we can see the breaking point from here. Walk into your average small dealership service department and see the equipment they have. Then walk into a small but sustainable garage. You'll see that people can get work done in both, but there's a lot less bodging in the dealership. In the small garage, one employee has to wear multiple hats (hey, I do startups, that's how we roll) which decreases depth of knowledge and ability to foresee potential problems but minimises cost. This can result in a bunch of "oh shit" moments where valuable lessons are learned that would have been avoided by spending hundreds of thousands or millions of dollars on a specialised piece of equipment and at least one trained tech. Thats what Enigma was aiming to put down in his points/posts. |
|
|
|
But carrying the analogy, would Enigma say that what you're doing in your garage is simply not enough and you need to get a million dollars worth of tools and equipment from AMG? That's the analogy.
I don't know if he's correct is my point. Assuming that he is correct if the average BFL customer will settle for nothing less than 100% perfect, is he still correct if they are willing to restart their machines every week or so?
I think Enigmas point is that there are many challenges and quality issues to work through. He feels in his opinion that the commercial components they bought are inadequate to truly provide a high level of quality control and internal fabrication that BFL sought to aquire All the other technical jargon supports his opinion. (much appreciated technical jargon!) |
|
|
|
1) When BFL's line of ASIC's hit the wild they must not consume more than 1.1 watt of electricity per 1Gh/s of Bitcoin mining speed, and according to everything you are telling us 1.1 watts per 1Gh/s is the most electricity they will use.
2) You will no longer post in my thread, if you post in my thread again you automatically lose the bet.
Now things get interesting. I would recommend to set an end date when the BFL ASICs have to be delivered (this year in larger quantity?). If they hit the wild in two years this 1.1W/GH/s might be no problem anymore. Josh, it's your turn - words or money? BTW two weeks until BFL mass shipments (end of Oct / beginning of Nov), so the first devices are already on your table. That should be save money. Or it's not that save which means no device on the table and no mass shippments within two weeks... +1 Interesting point your making |
|
|
|
How do you get that out of what I said? CPU->GPU->FPGA->ASIC->  ? Well, I don't intend to mine with first gen hardware for very long. I thought most users would be replacing the first gen hardware by selling it off to other interested parties. Buying up newer hardware would then be relatively cheap when going from first to second gen. Of course this would likely create a Gray Market that wouldn't involve any of the ASIC manufacturers. I see there is still plenty of ideas to be innovated through. Perhaps multi-core ASIC chips with more than one viable cryptocurrency. ------------------------------ From what you've said though, it almost sounds like you think most of your customer base would stay with first gen devices for the duration of it's useful life. Sure, denying others easy access to discounted hardware through a Gray Market would probably be profitable. (Both for miners and ASIC vendors) But....there is no guarantee you would keep your prices high enough to prevent new miners from entering into the fray. (Like the "Little Single SC") I guess we may have to ask you to give the community a heads up when you decide to do the next round of "price chopping"? ------------------------------- By the way, Can I get you "on the record" stating that the "Little Single SC" wasn't derived from re-flashing of 40GH/s spec'ed hardware? I have this notion that you might have underclocked hardware from the original 40GH/s spec to make the Little Single SC series. I would like to know if this untrue or not. Do all "Little Single SC" come with 4 or 8 chips? (Or is it mixed lots?) |
|
|
|
Ok Inaba.
When your fixed cost is extremely low and your initial capital outlay is high, that fixed cost becomes largely irrelevant.
It doesn't matter anyway, because you're deluded if you think any ASIC which will ship in the next couple months will be competitive with future products in three years.
Try doing an Internal Rate of Return and plug in your expected device power usage and then do one with power doubled. You'll see that when all factors are considered, electricity is largely an irrelevant cost.
See, that is the thing you don't seem to understand: If the device is profitable, it's still competitive. There's nothing to "upgrade" to after ASIC, so as long as it's profitable, it will be mining. Ergo, the power is the single most important aspect of an ASIC mining device. Are you going on record to say that you have no plans for a second generation device to replace the first generation? |
|
|
|
|
@ Cablepair
That is a crazy amount of money...
Therefore you should be extremely specific as to the nature of the electrical use.
When you cite 1.1 watts of electrical use, you should specify WHERE the electrical current will be measured and HOW. I know it sounds stupid of me to bring this up but these are points that the loser and winner would look over the fine print.
------------------------------
For example,
Inaba might choose one of his BFL Single SC that has premium chips and actually does 1.1watts per GH/s. So you might want to specify if this is from an average sampling or if Inaba (BFL_Josh) can choose which which sample he wants. (Not all rigs are equal at the chip level.)
-----------
Another point that might later be brought up is WHERE the measurement occurs. At the Chip Socket level?
The Power supply electrical socket? You won't know if the BFL Single SC is using something like a nanoATX power supply with an efficiency of 65% or if it is a quality PSU of 95%.
You should be very specific as to where the measurement occurs and what kind of measurement device is used. (Lab quality or if it is something like a Kill-A-Watt meter.
-----------
You should also specify WHO will do the measurement and how many samples are necessary to verify the electrical draw. (Perhaps you can send each other samples or ask a third party for verification? (Youtube Video recording of the measurement process without editing?)
It might make good PR or bad PR depending on how the details come together.
Oh and last but not least, whether "Tethered wattage" is considered part of the final result or not.
|
|
|
|
Sounds like you have possibly found a viable candidate to remove the extra tethered wattage.
While reducing 'tethered wattage' is always desirable in a general sense, don't lose sight of the specific numbers; the cost of running 'inefficient' hosts may be less than you think. Take an example: an 800Mhps FPGA miner will generate $3/day. My i7 3770K full-size PC uses 90W at idle (I have measured it with a Kill-A-Watt meter). Of course I would not suggest using this as a host, but for illustrative purposes a 90W PC running 24 hours/day will cost about $0.25/day in electricity (2.2 kWh, $0.11/kWh). That's 8% of the mining income. If you happen to run multiple miners off that same PC, the cost of running it becomes an even smaller portion of the mining income. And if you are talking about running a bASIC, the cost of running a 90W PC becomes largely insignificant. So if you were to use something like a Raspberry Pi, that uses only a few watts, to host your mining hardware, you'd be able to save at most $0.20/day or $0.25/day compared to a full-blown PC. Monthly, that's a savings of $6-$7. That may be significant to some people, and insignificant for others. I host my hardware using a small 25W Zotac Zbox, which I also use for my day-to-day computing tasks; this works well for me. I agree with everything you said, as I mentioned the same point you made a few pages ago when Inaba (BFL_Josh) was bringing up the power use issue. You should contact Inaba (BFL_Josh) and ask him to be nicer to Tom over the power issue. The power issue is really only relevant for a mining device when it nears its End Of Life cycle. |
|
|
|
Respectfully, the first issue is that BFL and bASIC are both selling a tethered device which means it must be connected to a desktop PC, Laptop PC, or other mobile device.
...
The average desktop PC at idle consumes about 150watts.
...
So you are never actually running a BFL device with a mere 60 watts. You are consuming 60 watts plus the overhead for the tethered PC. At best your total power consumption is greater than 60watts. Either it is at 30watts extra or closer to 210watts.
This little gem should have more than enough CPU power to run typical mining software at less than 0.5 W. It includes both wired and wireless network in addition to an USB port. I just got one off eBay at less than 25 USD. The board itself is so small, it is probably enough space for it inside the ASIC miner box. Wifi makes the miner a nice portable space heater. (If the miner is in a metal box, you need to add an external antenna to get reliable wifi.) Edit: Fixed link Sounds like you have possibly found a viable candidate to remove the extra tethered wattage. Any idea how to configure one for bASIC or BFL mining hardware? |
|
|
|
[..]
They're laying down BGA parts, and probably QFN parts.. Do they have X-Ray inspection? Do they have 3D solder paste inspection?
[..]
In the spirit of bringing the discussion to a higher level, the above is "translated" into links for further reading. BGA parts: http://en.wikipedia.org/wiki/Ball_grid_arrayA ball grid array (BGA) is a type of surface-mount packaging used for integrated circuits. BGA packages are used to permanently mount devices such as microprocessors. A BGA can provide more interconnection pins than can be put on a dual in-line or flat package. The whole bottom surface of the device can be used, instead of just the perimeter. The leads are also on average shorter than with a perimeter-only type, leading to better performance at high speeds.
Soldering of BGA devices requires precise control and is usually done by automated processes. A BGA device is never mounted in a socket in use. What is X-Ray Inspection? http://en.wikipedia.org/wiki/Automated_X-ray_inspectionAutomated X-ray inspection (AXI) is a technology based on the same principles as automated optical inspection (AOI). It uses X-rays as its source, instead of visible light, to automatically inspect features, which are typically hidden from view.
The increasing usage of ICs (integrated circuit) with packages such as BGAs (ball grid array) where the connections are underneath the chip and not visible, means that ordinary optical inspection is impossible. Because the connections are underneath the chip package there is a greater need to ensure that the manufacturing process is able to accommodate these chips correctly. Additionally the chips that use BGA packages tend to be the larger ones with many connections. Therefore it is essential that all the connections are made correctly.[1]
AXI is often paired with the testing provided by boundary scan test, in-circuit test, and functional test. QFN parts: http://en.wikipedia.org/wiki/Quad-flat_no-leads_packageFlat no-leads packages such as QFN (quad-flat no-leads) and DFN (dual-flat no-leads) physically and electrically connect integrated circuits to printed circuit boards. Flat no-leads, also known as MicroLeadFrame and SON (small-outline no leads), is a surface-mount technology, one of several package technologies that connect ICs to the surfaces of PCBs without through-holes. Flat no-lead is a near chip scale package plastic encapsulated package made with a planar copper lead frame substrate. Perimeter lands on the package bottom provide electrical connections to the PCB.[1] Flat no-lead packages include an exposed thermal pad to improve heat transfer out of the IC (into the PCB). Heat transfer can be further facilitated by metal vias in the thermal pad.[2] The QFN package is similar to the quad-flat package, and a ball grid array. Automated optical inspection (AOI) http://www.goepel.com/en/optical-inspection/aoi-systems/solder-paste-inspection.htmlhttp://www.youtube.com/watch?v=wGCQEpCD8ts (Step 5) http://www.youtube.com/watch?v=t8SidZ3wLvQhttp://www.youtube.com/watch?v=3ThMLs-Cuekhttp://www.youtube.com/watch?v=-QQihWCpN9I |
|
|
|
Removed my GPU, still nothing, I think i fried my MB
Whatever you do, don't panic. Disconnect all power leads except for those connected to the motherboard (24pin plus any extra 4 pin plugs). As strange as it sounds, sometimes a fan might fail and it shorts causing strange behavior throughout the system. (Including no POST) If you pull out all cards and leave the system as if it were a barebone, then you should know if it is the motherboard. Also retrace the steps you might have done in the last 24 hours. Usually it might be something pligged in badly. |
|
|
|
@ BFL_Josh (Inaba)
The biggest concern of mine is not electricity use if it remains within a certain range. What concerned me the most is what was the safe operating temperature for the mining hardware.
Winter is coming to the northern Hemisphere. So the heat put out by these devices will not be noticable if BFL or bASIC ships out in the next 2 to 3 months. But after Febuary, if one of these devices has to be cooled with Air Conditioning, then that would make a big dent in the profit margin. It would have been a deal breaker to add 1 to 2 kilowatts of power consumption during the following spring, summer and fall. As well as affect the reliability of the device.
Thank you for your recent answers on your mining device. That helped me understand that you have taken this into account and have designed safety features (well BFL_Engineer did at least) so that the device would not suffer from a heat related issue.
Those of us coming into summer will have this well tested before your summer rolls around  (assuming delivery before your summer)  +1 Good point, keep us updated on this! |
|
|
|
By the way, if anyone is wondering where I got my estimates from, I have a device called a Kill-A-Watt meter that tells me what the power consumption is at the electrical socket. http://en.wikipedia.org/wiki/Kill_A_WattI have one of the P4400. |
|
|
|
|
@ BFL_Josh (Inaba)
The biggest concern of mine is not electricity use if it remains within a certain range. What concerned me the most is what was the safe operating temperature for the mining hardware.
Winter is coming to the northern Hemisphere. So the heat put out by these devices will not be noticable if BFL or bASIC ships out in the next 2 to 3 months. But after Febuary, if one of these devices has to be cooled with Air Conditioning, then that would make a big dent in the profit margin. It would have been a deal breaker to add 1 to 2 kilowatts of power consumption during the following spring, summer and fall. As well as affect the reliability of the device.
Thank you for your recent answers on your mining device. That helped me understand that you have taken this into account and have designed safety features (well BFL_Engineer did at least) so that the device would not suffer from a heat related issue.
|
|
|
|
Keep in mind that even if a 54 GH bASIC uses 150W (2.5x) and a 60 GH BFL Single uses 60W, they're competitive for the first year of operation (assuming $0.12/kwh), regardless of the difficulty factor: bASIC: $1099.99 + $157.68, $23.29/GH BFL: $1333.00 + $63.07, $23.27/GH And surely both will be obsolete within a year, so don't tell me about how many pennies you'll save after the first year. And if it turns out the bASIC will use 250W, CablePair can simply lower the price by 10% or overclock to 60 GH to regain competitiveness. The important factors to be considered when investing in ASICs is price per GH and delivery speed. The big question one should ask right now, is will a new BFL order be delivered before or after a new bASIC order, as that will affect the bottom line far far more than power cost. We have some estimates when each will ship the earliest orders, but in BFL's case especially, the delivery date of new orders is a wild guess. The sooner BFL figures out and publishes estimates of when they'll catch up with all pre-orders, the more sales they'll take from bASIC.  This is a good point, but I think your missing the point of the ASICs. If you invest in a high power consumption ASIC, it ceases to become profitable much faster than a lower power one. You say that they will both be obsolete in a year, but that just doesn't make sense. They become obsolete when they no longer have a positive ROI - so naturally, the one with the lower power consumption becomes obsolete later. In the case of a 2x bump in power consumption that means your device becomes completely useless much sooner than a comparable device. I dunno about you, but I'd rather mine on a device for 3 years than 1 year and have to fork out more money to stave off obsolescence. Not everyone can keep investing money year after year. That's why power is the most important aspect of an ASIC device and it literally defines if the device is viable or stillborn. Tom is basically the only hold out on the ASIC front as far as power figures go and it makes me question why that would be. Either he or his engineers have a rough estimate (or at least I would hope so!), and putting that out there would be the transparent thing to do, even if it's with the caveat that "it's only an estimate and may change by 20%" or something similar. I understand what you are trying to say, but there are two large issues to your position on this matter. Respectfully, the first issue is that BFL and bASIC are both selling a tethered device which means it must be connected to a desktop PC, Laptop PC, or other mobile device.------------------------------ The average desktop PC at idle consumes about 150watts.http://en.wikipedia.org/wiki/Desktop_computerThe average laptop (depending on specific design) consumes anywhere from 30watts (ultra portables) to 90watts (DTR's [DeskTop Replacements]).http://en.wikipedia.org/wiki/Desktop_replacement_computerhttp://en.wikipedia.org/wiki/Ultra-mobile_PCSo you are never actually running a BFL device with a mere 60 watts. You are consuming 60 watts plus the overhead for the tethered PC. At best your total power consumption is greater than 60watts. Either it is at 30watts extra or closer to 210watts.* Not including the Host PC monitor which ads anywhere from 20 to 80watts depending on monitor quality. The same will be true for bASIC mining hardware. This is why I chose Avalon instead of BFL or bASIC. As long as their standalone system comes somewhere near 200 to 300 watts they are pretty much in the same neighborhood as BFL and bASIC actual operating expense. If they do better than that they are still competitive. ------------------------------ The second large issue is that you ?appear to assume? the individual isn't going to make a sufficient amount of BTC or Fiat Money to cover the costs of an upgrade to a second or third generation BitCoin mining device in a years time. If the Mining device is that unprofitable over the span of a single year...then why bother selling them if you can't replace the cost of the device plus a healthy profit margin within 1 year?? (Heck, even in 3 months of mining) This ultimately means that unless the device is seriously underperfoming or consuming a significant amount of power, it will make it's ROI (Return on Investment) with ample profit. There will be those with a few hundred dollars less in their pocket due to higher electrical costs. While others with a few hundred more in their pocket due to power efficiency. The difference is relatively minimal (IMO) considering the overall net income from mining over 1 year. |
|
|
|
|
Show Posts
