I appreciate your answer, but your tone seems worn down and angry. I'm not the enemy like you consider iCEBREAKER, but I won't blindly defend a company while others rise around them. 1.) Lost regarding revenue, not "where are the chips". The hashrate increased a lot more than I think people expected from "Non-Avalon" related miners coming online. Also I believe this was supposed to generate more revenue for ACTM and help with costs (Is ACTM doing alright without this?) Also where (Or near where) on this list is ACTMs Chips https://docs.google.com/spreadsheet/ccc?key=0AiLYkKIHJaIsdHpIaGdUOWRYVUdncTNpNlVKbVhCbEE#gid=0 This should give us a more definitive answer as to when chips should come. 2.) I'd like an official statement on this. 3.) This is common knowledge, please don't assume that because we ask question we must be stupid. You know exactly what is meant when the word "Tape-out" is used with regarding ACTM. 4.) It's been about 3-4 Weeks now, I don't think anyone is saying it is easy as flipping a switch but enough time has passed now that it should be done or close. We don't need Fluff in this thread, we need hard facts even if the answers are not what shareholders want to hear. I'd rather have a "I don't know" than an answer that provides no substance or a speculated answer. There are people interested in buying more shares but it's been so quiet lately and with the Avalon Delays having working Avalon "Boards" ready isn't very exciting (Group buys have had these ready for weeks now). I have a substantial amount of shares and would love to see the price go up but the price won't go up until the speculation is gone and facts are put on the table. We need an updated timeline based on current situations. My tone is not worn down or angry, if anything it's "tired" of repeating the same information. I don't cultivate enemies and certainly Icebreaker is not my enemy. I focus on discussing points, not persons.1) "Lost regarding revenue" was exactly what I addressed. ActM is not receiving revenue from them, same as the rest of Avalon buyers, so the network total didn't also rise from everyone receiving those chips. It's all relative. 3) This is not common knowledge it seems. Anyone that really knows eAsic's process knows that "tape-out" is a non-issue. It does not represent any kind of bottleneck and it's achievement is mostly irrelevant on the overall pipeline. The chip design is not part of their critical path for manufacturing. 2) and 4) are still under NDA, so while you may not like it (understandable) there's really not much more to be said until the light goes green.  |
|
|
|
How much did we lose on the Avalons?
Where is eASIC's press release?
Will ACTM ever tape-out?
Is there a problem with the BTC conversion?
These are valid concerns with the influx of new ASIC manufacturers in the market. Every day that passes ACTM loses their edge on quicker delivery. There's been enough time now that we should have more information. 1) Avalons are not "lost", ActM just didn't receive them yet, same as everybody else, so the total network hashrate didn't increase as much as it could have if those chips had all been delivered to every Avalon customer. 2) eAsic's PR will be released as soon as possible ofc 3) ActM doesn't "tape-out" in the conventional IC serial development sense, wafer manufacturing can be started in paralell to chip design, this is one of the key eAsic advantages 4) Why should there be one? Signing a $1M deal is not flipping a switch. |
|
|
|
Not sure why this is still a question (goes beyoond just HF). Remember even excluding water cooling pumps, fans, and controller boards there are a lot of efficiency loss between between the wall and the chip. "This is still a question" because Vbs is here to spread FUD about HashFast. He's on ACTM's Board of Cheerleaders, invested more than he can afford to lose, and got married to a competitor's stock that not only became obsolete soon after launch but is also enduring disasters WRT BTC conversion and bulk Avalon orders. At this point, asking dumb questions which have already been answered repeatedly is all he can do to prop up his failing stillborn venture. I'm sorry, who are you again? This kind of ad hominem argumentation is indeed lamentable, so if you have nothing substantial to discuss, please keep to yourself. I'll reiterate my question about the power figures: Simon, the store page says "* Real silicon power consumption may vary from simulation results by +/- 20%" (I assume this means the chip can be between 200-300W), you would really have no problem if it ever comes to be 300/324 = 0.926W/mm^2? It's still 12% more than an o/c SB-E even at an extreme 350/425 = 0.824W/mm^2. Using some quick math for forced convection water cooling, with: Water: 50 - 10.000 (W/m^2K) Chip area (m^2): 0.000324 t_surface = 80ºC t_air = 30ºC q = 10.000*0.000324*(80-30) = 162W, which is a water heat transfer figure under chip requirements, so the heatspreader has to really compensate for the so small chip area! |
|
|
|
350W is a rounded up number for the whole system, including the power lost in the 2 stages of power supply, and fans etc. The chip itself draws 250W @ nominal.
Depends on how hard you cool and overclock your Sandy Bridge E. The company that is assembling our systems specializes in overclocking. They run Sandy Bridge Es overclocked to 350W (CPU power alone, not whole system), using the same cooling system we are using. We are also using a heatspreader.
Metal migration is a well understood phenomenon. We have followed all the fab's rulesets for electromigration so that the current levels we're going to see will not be a problem (even current distribution, and thicker metal layers). Currently in the simulator for EM our chip passes the test for a 5 year lifetime, but fails the 11 year test - and that is running somewhat overclocked, at about 540GH/s.
Thanks for the info. What's the size of the package/heatspreader and what are you using between it and the chip? Solder? TIM? W/mK? Not sure why this is still a question (goes beyoond just HF). Remember excluding water cooling there chips doesn't run on 12V. A good 12VDC to ~1VDC PSU (on the board) is at best 90% efficient. So if the chip uses 250W @ ~1VDC then it will require at least ~277W @ 12VDC. Now the ATX PSU which converts the 120V AC isn't 100% efficient either. If it 90% efficient then to supply 277W @ 12VDC requires 308W @ 120VAC. Add in 20W for pump and radiator fans plus a margin fro safety and ~350W is more than reasonable. Once again this isn't unique to just HF, it applies to all ASICs by all companies. Board Wattage = Chip Wattage / DC PSU efficiency System Wattage = ASIC Board Wattage + Controller Wattage + Fan Wattage + Auxiliary Wattage Wall Wattage = System Wattage / ATX PSU efficiency The wattage at wall is NEVER going to be the wattage at the chip level. Yes, you are quite correct, two power stages at η around 0.9 are enough to have 19% in power losses, plus the rest of the cooling extras. Still, the specs say " Under 350 watt power draw*" and "* Real silicon power consumption may vary from simulation results by +/- 20%", so we're talking apples and oranges here.  |
|
|
|
Interesting so, KnC is a 250/900 = 0.28 W/mm^2 chip and HashFast is a 350/324 = 1.08 W/mm^2 chip, requiring a cooling solution that can transfer 5.6x the heat per mm^2. I really hope your cooling solution holds up 24/7!
Out of comparison, an ATI 7970 is a 250/365 = 0.68 W/mm^2 chip and an nVidia GTX Titan is 250/561 = 0.45 W/mm^2 chip.
Hashfast's chip at nominal (400GH/s) is expected to consume about 250W of power, not 350W. i.e. 0.77W/mm^2. Overclocked Sandy Bridge E in shipping commercial products with the same cooling system runs 350W with a 425mm^2 die, 0.82W/mm^2. OK, so two questions: 1) https://hashfast.com/shop/babyjet/ states 350W power draw (+/- 20%), but the chip only consumes 250W? 100W for cooling/misc? 2) An overclocked Sandy Bridge-E doesn't "run" at 350W, for example check this (pic below). The 349W here is System Peak Power, a very different metric ( system vs CPU and peak vs sustained). Subtracting the system idle power ( 85W@4.7GHz), even at peak usage it would run at 264W (and it would die from electromigration if run like this 24/7), so 264/425 = 0.621W/mm^2, still 20% below 0.77W/mm^2.  Edit: Let's not forget that a Sandy Bridge-E uses a high quality heatspreader (IHS) with fluxless solder, so the actual contact area with the cooler is much bigger than the die size, reducing the W/mm^2 requirements by a large amount. |
|
|
|
I'd like to write today about a small piece of why we are confident our product is better than KnCs. So today's topic: Our silicon design is superior. Both are 28nm designs, but HashFast's is far more powerful and energy-efficient. Let's look at KnC's 28nm ASIC, and some basic details as we can pull from their documentation. https://www.kncminer.com/news/news-25First let's calculate the hash rate per square millimeter of silicon. This is a measure of the efficiency of the design. Honestly, we don't need much to estimate this. The lid size for their chip is enough to make some good estimates. KnC's diagram shows their chip has a 41.2mm lid, and implies that the silicon under that lid may be between 30mm x 30mm, and 36mm x 36mm. (The additional space is needed for decoupling capacitors and such.) Let's use those two numbers as bounds for the size of the silicon under the lid. If the die(s) take up just 30x30mm of the space under the lid, then: 30x30mm = 900mm^2 100 GHash / 900 mm^2 = 0.11 GHash/mm^2 Or if the die takes up a bit more of the space under the lid, 36x36mm = 1296mm^2 100 GHash / 1296mm^2 = 0.077 GHash/mm^2 HashFast's Golden Nonce chip: I don't have to estimate the size because I work at HashFast. One 18x18mm die is able to do 400 GHash (nominal - more overclocked**) Hashing per square mm: 18x18mm = 324mm^2 400 GHash / 324mm^2 = 1.23 GHash/mm^2 Let's compare those numbers, for the high and low values for KnC's chip: 1.23 / 0.11 = 11.2 1.23 / 0.077 = 16 So HashFast's chip is between 11 and 16 times more efficient, in hashing per square mm, than KnC's chip. This has an impact on how fast we can deliver units to customers. One wafer of HashFast's chips has the same capacity as 11 to 16 wafers of KNCs. The initial engineering run from TSMC is limited to 12 wafers, out of which KNC will be able to satisfy 11 to 16 times fewer customers than HashFast will be able to. You'll get your units faster once production starts from us. In addition, the HashFast chip operates much more efficiently. You get four times the hash rate for the same amount of power (250W). That's 1.6 GHash/W for HashFast, and 0.4 GHash/W for KnC. Calculations such as this are a small part of why we are confident that we are delivering a quality product to our customers. We figure it's time to start sharing such information. Amy Woodward VP Engineering HashFast ** P.S. Simon made me put in the line about overclocking. But no one would ever do that to our beautiful chips, right?  Interesting so, KnC is a 250/900 = 0.28 W/mm^2 chip and HashFast is a 350/324 = 1.08 W/mm^2 chip, requiring a cooling solution that can transfer 5.6x the heat per mm^2. I really hope your cooling solution holds up 24/7! Out of comparison, an ATI 7970 is a 250/365 = 0.68 W/mm^2 chip and an nVidia GTX Titan is 250/561 = 0.45 W/mm^2 chip. |
|
|
|
Even with the eASIC deal, won't ActM chips be slower and more $/GH than those of CoinTerra and other companies? The Avalon situ has really screwed ActM.
In a condensed answer, no. There's much more to designing chips than just stating wishful specs and eAsic has more design experience than any of them (they've been working on 28nm at least since 2010). |
|
|
|
Interesting isn't it? It seems ActM is still in scam-territory.  Seems after the eAsic deal is complete ActM will rise from "scam" to having the "best ASIC designer team" overall. |
|
|
|
The 2 TH/s miner will be targeted towards retail customers and housed in a 2U air-cooled container. Early estimates of power efficiency indicate 0.6 Watts/GHs, 1,200 Watts total, though Cointerra wants to remain conservative with power estimates until manufactured chips can be measured. The miner will be self contained with a power supply and controlled through an ethernet port. 1200W of heat inside a 2U case (more like 1333W considering η=0.9)?  That'll definitely fry some eggs!  Example of a 2U case with PSU:  |
|
|
|
Yeah I was referring to BFL being completely and utterly retarded beyond the point of recovery... unless they suck unawares in with google ad's and the like.
Most unawares don't have $5000 (with taxes) to blow on a boondoggle like this. The low price point of BFL's other devices is in part why they had so many orders and also why the company is in trouble as the units ended up costing them money due to the redesign. (What I'm saying is that once you factor in the cost of delays + new parts + Wages + fixed costs that they sold the original units effectively below cost.) This is why a couple months ago they offered a $100 upgrade option for a 2nd chip in Jalapeno's. And are doing another cash grab now. They aren't only doing a cash grab, they are eventually pushing all their pre-order customers to upgrade else they are left in the dust. Imagine you had pre-ordered a 500GH/s Mini-Rig for $22.5k, now to be outdated by a 600GH/s Monarch for $4.7k, ~82.7% cheaper per GH/s. They are basically shifting their orders into the future (upgrade or be left with a worse deal), reducing their costs now. |
|
|
|
Thanks for that. It all makes sense. You really do know your stuff! EDIT: I do have one question though. When BFL stated, "The PCIe format used is 1X for maximum compatibility." I read that not as them using a pci-e x1 connector as seen in the image below, but more that they wanted to ensure that it wouldn't take a top of the range motherboard supporting pci-e x 16 standard to be compatable with their product, however would use the same connectors commonly used in current grapics cards. Can a heavy weighted dual card like BFL's be safely mounted to a motherboard given the low surface area?  They only need a x1 connector, just like that one. Anyway, there's already a thread to discuss this! https://bitcointalk.org/index.php?topic=276692(check https://bitcointalk.org/index.php?topic=276692.msg2954571#msg2954571 and https://bitcointalk.org/index.php?topic=276692.msg2955061#msg2955061, I still believe >250W on a blower cooler is insane for anything running 24/7) |
|
|
|
I'm no expert but have built a few pc gaming systems in my time. Please correct me if I'm wrong..
Even though BFL labs state, "PCI Express - Monarch cards consume two PCI slots when installed in a standard ATX montherboard. The PCIe format used is 1X for maximum compatibility.", does this not mean that they will use the pci-e x16 slot (not to be confused with a pci-e x1 slot) and run it as pci-e x1 in the bios? Would this not make both the images and their reference to pci-e x 1 marry up?
It means they will use a PCI-E x1 connector (as they should), because the x1 bandwidth is more than enough for bitcoin mining. You can slot a PCI-E x1 card in any x1, x4, x8 or x16 slot. You can also use a PCI-E x16 card (GPU) to mine on PCI-E x1, this is a known trick way back into 2011 at least, for example, https://bitcointalk.org/index.php?topic=6982.0Also the power connectors are not visible in the four images because none of the images show that side of the card? If we argue that the first image does, then it's fair to say this BFL card doesn't have any power connectors.
The first image of the card shows enough to see there are no power connectors there (it's a CAD oversight no doubts there). You need two 8-pin PCI-E power connectors to be able to reach 350W+. I'm not sure how accurate your stats are for power draw as according this source you can see a 7970 uses 362 watts, with the 7990 drawing 491 watts. http://www.techspot.com/review/663-amd-radeon-hd-7990/page8.htmlIn terms of the air cooling solution, what I see in those images is no different to high end dual slot gpu's not using after market cooling. All that said you would be mad to go for one of these cards from BFL. I hope they get their arses kicked by NY Department of Financial Sevices. EDIT: I thought I was mistaken by the link mentioning 'system power consumption' meaning entire sytem but it's not, it is indeed the cards only. Those pics show total system power, a 7970 GHz edition there would be consuming at max 362-74 = 288W (which it isn't because there's also CPU+RAM load on the Crysis 3 stress test). You can check that the official TDP of the 7970 is 250W. Keep in mind that their goal is to go for an external exhaust card to be used on 4U cases. External exhaust cards blow all the hot air to the outside of the case so that you mostly only need to worry about intake fans for cool air. A 7990, for example (as all dual GPU cards) use cooling solutions that leave a lot of hot air inside, and this hot air re-circulation is what makes it very difficult to have 3 cards inside one case, because now you need intake fans and outake fans capable of removing all that excess heat. Also, let's not forget that a 7990 at 375W TDP is actually two chips (~187.5W each), so I'd say BFL will probably use a solution with several chips inside. In the end, if they want to keep an EE solution (and they should) they will probably have to lower specs on the cards to ~250W total each (underclocking/undervolting/less chips per card) for the EE cooling to be effective. |
|
|
|
Now I understand that VBS probably holds a lot of shares in Actm but you are starting to look silly. When AMC was new and was looking like scam more than anything you were supporting it and hyping it all around. Never ever have I heard you say anything wrong. But on every other company that came on scene from then, you just try to look for what is wrong with it (not that it is bad thing overall) and point it out.
Not like I wouldnt actually hate BFL and their shit btw.
I dont think you meant 'silly', did you? You meant 'biased', surely? Thing is, being biased is unavoidable if you have a given interest. The thing to note is Vbs's total grasp of presenting stated stats, as a means to compare and grade. And because of his knowledge of chip design and form, he is even able to present examples of whats gone before/precedence. To be silly, would be to ignore Vbs....whatever his interests may or may not be. I think the first thing anyone needs to do when looking at specs of anything is compare them to what others have done/achieved. I've been looking at promises from a lot of the current chip designs and the first thing I notice is they look way out of what's been accomplished, some even on the same or older technology nodes. I may be biased, but I still think ActM's flip-chip (FC672, 27×27), at 16GH/s and ~1GH/s/W has some of the most reasonable specs around here. |
|
|
|
Any insight into this volume on BTCT right now? Lots of 30 share buys followed by 30 share sells. Why do this?
Someone is executing PUT options, look closer at the prices! |
|
|
|
So are you saying that either what BFL has released spec wise is overly optimistic or they're not nearly as far along with development as they're leading people to believe?
I'd say why release specs on a PCI-E card that already defies what the actual cooling hardware can do in that form factor? When aiming for an external exhaust card, why not release a 250W, 420GH/s card that is much more believable when compared with the rest of the hardware out there? Now I understand that VBS probably holds a lot of shares in Actm but you are starting to look silly. When AMC was new and was looking like scam more than anything you were supporting it and hyping it all around. Never ever have I heard you say anything wrong. But on every other company that came on scene from then, you just try to look for what is wrong with it (not that it is bad thing overall) and point it out.
Not like I wouldnt actually hate BFL and their shit btw.
Please give factual points where I'm looking silly, other that a circular definition. You don't agree with my views? Why? Which ones? Which points was I making that later didn't turn out accurate? Debate my points, not myself. |
|
|
|
I thought this was the ActiveMining thread.. not the BLF...lol
Indeed, let's bring OOT discussion to the speculation thread. |
|
|
|
Some thoughts about the BFL Monarch, https://products.butterflylabs.com/600-gh-bitcoin-mining-card.html- The four CAD pics don't correspond to the final product (PCI-E x16 on pics, PCI-E x1 on product; no PCI-E power connectors on pics, each card should require two PCI-E 8-pin plugs)
- External exhaust blower-type cooling solution is unsuitable for 350W TDP. This kind of TDP requires either water-cooling or getting about half of the heat back into the case (e.g., check dual-gpu cards). On a 4U with 3 of those cards, that's ~0.5*350*3 = 525W of heat recirculating inside!
- The margin on the hashrate specs is very large: "600 GH/s nominal performance ( + - 20% )", meaning a 600*0.8 = 480GH/s card still falls into specs.
- Power usage is a gamble for this form factor: "350w (0.6w/GH conservative estimate)". Depending on the final cooler solution they decide on, they are already operating on higher cooling performance assumptions than anyone else. Blower-type cards (GTX Titan, GTX 780, AMD 7970) are all around a TDP of 250W. If more dissipating power is needed, the cooling solution may not allow for it, requiring cards to be run underclocked/undervolted.
|
|
|
|
So, a 600GH/s PCI-E card? Although they have in the specs +/- 20%, so it can turn out to be a 480GH/s card while falling in-spec. Great wording there! Even better, the picture is misleading as it's a 350W PCI-E card with a blower-type fan? That cooling solution doesn't work for a 350W TDP card. Even a GTX Titan/780 has a great vapor chamber cooler and it's a 250W TDP GPU. |
|
|
|
On this topic, what is the Activemining plan for mining with their machines?
Sure the 28nm will be more efficient, but how much hashing could Ken even operate at his facility?
This is a very interesting topic. That is why I wonder. It is highly likely that the network is 1,000 Th/s (1,000,000 GH/s) in 45 days from now. So, say you want to be a player like friedcat and have 20%. With avalon chips drawing 9W per Gh/s, you need 900 kw just to power the avalons to get 10% of the market. How you going to do that? lot's of buildings with that power coming in? How you going to distribute the power? Oh, and guess how much heat 900 kW of power puts out when it goes into something that basically just heats up? But to get 20% you need 1800kW PLUS cooling. Even at 1W, it is an enormous undertaking. It will be cool to see how people over come these hurdles. Someone is going to have 500 TH/s rolled out by November? They best call the electric company now because ordering new transformers takes a while. Indeed, most of the current mining proposals seem to skimp on that part. The Avalons work at 6.6W/GH/s, so for the 20,000 ordered chips, that's 5640*6.6 = 37,224W or around 41kW on 90% efficient power, a small figure. The current reserved facility is 100ft (30m) underground, naturally achieving a constant 58ºF (14ºC), which saves immensely on cooling costs. The rest of the mining operation will depend on the number of ordered new Fast-Hash-One chips, although I'm expecting the sales figures will always vastly surpass the mining income. Let's not forget the >$12 million Avalon made in bulk chip sales. |
|
|
|
I think we are about to see the total hash rate growth rate actually start to cool off in the coming weeks..thoughts? http://blockchain.info/charts/miners-operating-profit-marginCould be totally naive..but you would have to be fairly confident you could compete at a top level to start getting into the mining game now correct? Or will we just have a bunch of idiots come out and run non-profitable operations? Would be good for ACTM hardware sales.. No cooling down in the foreseeable future, as that chart is based on GPU efficiency. Things will only start to cool down as all 28nm manufacturers flood the market and push out everything that's less efficient. https://blockchain.info/statsElectricity consumption is estimated based on power consumption of 650 Watts per gigahash and electricity price of 15 cent per kilowatt hour. In reality some miners will be more or less efficient.** For profit margin hardware costs are estimated to be $1000 per gigahash every 2 years, and bandwidth $1 per gigahash per year. Anyone that gets into the mining game now without some small measure of "future-proof hardware" (whatever that stands for) will be kicked out faster than someone who pays attention to that. All hardware is profitable until it doesn't pay the electricity bill anymore, that's the point where the switch is flipped.  |
|
|
|
|
Show Posts
