expect 3-6 week delivery time with dealextreme (or anything sent via chinapost) |
|
|
|
Is there any disadvantage running a k64 as 4 single k16?
Ok you need 4 USB-cable but is there a speed decrease?
Not different. a k64 is just 4 k16, not cut up into 4 boards but kept as 1. They are connected between each board, not by usb but by a ribbon cable. All you will need is one usb cable, and i think you can connect/chain up to 8 boards in one ribbon cable or was it even 16 boards. I would assume 8 of them, since that is 128 chips (USB hosts can accept 128 devices, and i would assume that the controller on the K16 either follows this rule or applies a similar rule to the chips it controls) No, the USB protocol is not used when communicating with the chips. Each 16-chip board is shown as a single device when connected via USB. If you use the built-in chaining capability of the boards (which uses the i2c protocol, not USB) it still shows up as a single USB device, regardless of how many boards you have chained together. okay, thanks for clarifying. On some research, it looks like i2c is limited by the number of pins used. If 3 pins are used, 8 devices are the limit. 4 pins should enable 16 devices. how many are on the K16/K64? |
|
|
|
Is there any disadvantage running a k64 as 4 single k16?
Ok you need 4 USB-cable but is there a speed decrease?
Not different. a k64 is just 4 k16, not cut up into 4 boards but kept as 1. They are connected between each board, not by usb but by a ribbon cable. All you will need is one usb cable, and i think you can connect/chain up to 8 boards in one ribbon cable or was it even 16 boards. I would assume 8 of them, since that is 128 chips (USB hosts can accept 128 devices, and i would assume that the controller on the K16 either follows this rule or applies a similar rule to the chips it controls) |
|
|
|
|
^the quotes are not fully needed i guess, they were to refer to the fact that the design is fairly finalized, with the laser-etched heatsink (if you can call it that!) and 4 available colours.
I don't doubt the K1 can be made cheaper, and thats what i like to hear! By the time these enter production, difficulty will be through the roof and <0.3BTC prices may be necessary to come near break-even.
I see these 1-chip devices as more of a fun/gift item then a profitable device (unless you resell on ebay apparently), and think that it might be more important to focus on the 16-chip designs if profit is the goal (considering 5-8 of these K1s match the price of most K16 preorders)
|
|
|
|
I am interested in buying these to get rid of my GPU's. Are you still on track for an August to September delivery? I would like to get at least 20 and possibly more.
price is going to be an issue with these, since asicminer is already delivering 'professional' looking devices at 336 MHash for the same price. |
|
|
|
I have used all 28 ports on that hub with no problem for miners. The magnets are definitely a issue though.
you must be referring to BFL units that draw power from elsewhere. For block eruptors, 4A will only power 8 or 9 <500mA devices |
|
|
|
That still doesn't answer my question. Playing with the numbers or not, you asked my why 2BTC and I've answered it was a ceiling. I asked you how much would you pay for, and you point me to a mining calculator. Would you please answer how much would you pay for, then? I see 2 reasons to buy: novelty and profitability. For me, a USB miner is made for novelty, so i would find a 75% return of costs to be sufficient to be 'satisfied' and have a cool physical peice of hardware. But the asic design to have 10 chips will/should never cost 5x the cost of 2 chips, because of hardware redundancy and basic production costs being cut away. right now, I am willing to go 0.9BTC/Ghash. In a month I would expect 0.5BTC/Ghash. In 2 months, 0.35BTC |
|
|
|
|
This thing has just blasted its way up in BTCT, I can't beleive its at 3x what i bought for only a week ago, wish i had some more bitcoin in play money then the measly 05BTC i began with!
is there a forseeable roof to this thing? I'm cautiously moving my options to match the steep climb, and can't risk being caught short if this thing gains another 20% in a day.
that said, ive had some good luck selling options at a profit, re-investing the premium, and re-buying shares at little/no loss by the price when the call is carried out
|
|
|
|
there is a decent amount of PCB space between boards. Is there a reason for that, or will version 2 boards be designed more densely in order to shrink the size?
There are no plans to shrink the size right now. Of course, the closer the boards, the more difficult it will be to get the heat out. Maybe after more real-world data, a size shrink could be done. I was thinking more about shriking the H-boards, not the m-boards. right now, there is a decent gap between each bitfury chip that could surely be reduced with some design revisions or adjustment of resistor placement. However, this would still increase the heat density... |
|
|
|
I think an upper edge slot brace is a good idea. These are hand made prototypes, so Cscape used the caps he had on hand - I know he recommended a larger cap along the way too, so perhaps you are seeing the newer assembled boards vs the boards he made a couple weeks ago.
There was a mismatch between the capacitor shape on the board, and the capacitor mentioned in the parts list. I ordered the (smaller) caps on the BOM, but noticed it didn't really fit well on the footprint on the board. I put the small ones on anyway (with hand soldering it is just possible to do that), but ordered the bigger ones for a better fit, and used those on later boards. Electrically they both work, so there was no need to replace the smaller ones on the first couple of boards. thanks for the response, I wasnt sure if i was seeing things or if there were just different versions of board in there. there is a decent amount of PCB space between boards. Is there a reason for that, or will version 2 boards be designed more densely in order to shrink the size? |
|
|
|
|
Shipping to Ontario, Canada?
|
|
|
|
Is it USA-only? I'm in canada and would love to order at least 3-5 of them! |
|
|
|
Thanks to Cscape for hand-assembling these boards! This is long, neck-wrenching work to do all this by hand! As you can see, the prototype rPi connector was not the right one, so it oriented the rPi in a weird way. We've since identified a taller right-angle connector that puts the rPi vertical like the other H-boards, as originally designed. We are also replacing the power connector with Molex-Jr to match your PSU power supply GPU plugs. These are 100TH boards, which don't use PSU power supplies. Case designers: don't forget you can have a small display from the rPi. It would be great to see someone figure a way to drive external LED's from the rPi, despite that we tie up all the GPIO pins. This is what 400Gh/s looks like.  spectacular! I can see why it was said that these need some sort of stand-offs to prevent breaking/wobbling if in the direction of a fan perhaps V2 boards could have some mounts at the top the run spacers between and fortify the boards against oneanother? also, is it just my eye or the camera angle, but it seems that the capacitor on bords 7,8,&9 look slightly different in size? On boards 9 and A, the capacitor looks big and fat, but on the boards 7,8 it seems much smaller for some reason? |
|
|
|
LOL I went a bit overboard... But I had a bunch of extra AMD stock CPU coolers, so I removed the rear metal plate, thermal pad, applied fresh thermal paste, and voila: <img> https://i.imgur.com/q1yIwW7.jpg3</img> wow, that might be a little overkill! |
|
|
|
|
"best way to power an asic" = 'how do i get free electricity' = im not even going to google it for you
your question has no more relebvance to asics then to a fridge or computer. (and there are forums for those too)
|
|
|
|
|
Any updates? The last piece of real news was July 8's prototype hashing away, and time keeps rolling closer to the 'delivery date'
I'd love to hear how things are going from the K16, logistics/shipping/assembly, avalon chips, and estimated delivery date(s)
|
|
|
|
We'll just add poor grammar, spelling and homophobia to the list of things that make you a worthless piece of shit and a pathetic human being. Like the list wasn't already long enough.
So not only are you a confirmed liar, you're also a homophobe, excellent. You are a real piece of work.
Now, why don't you go back to grade school and learn basic math before talking shit on the internet?
Best. Company. Ever. I've got a paypal order that is less than 45 days old and can be disputed successfully on a whim for various paypal guidelines. right now it looks like that will be the case... what a special type of idiot Inaba/BFL_josh is to make his company look like a pile of incompetent fools who are in past their heads |
|
|
|
Please only use USPS to ship to Canada. FedEx, UPS, etc tend to charge absurd brokerage fees usually equivalent to 15-20% of the declared value of the item.
+1 +2 If it will cost several days' mining to cover the brokerage fees, no thank you. priority USPS services should deliver within a week |
|
|
|
Hi buzzdave,
i am sorry if this was already discussed but i couldnt find an exact response so hoping you can clarify again.
1) August 400GH fullkit is ~$19k & October 400GH is $8k. But August and October 25GH are both $1.3k. Why is the 400GH at >50% discount for Oct but the 25GH has the same price as Aug?
2) Per the website:
- 25 GH = 16 chips & 400 GH = 16*16 = 256 chips -> they both average ~1.5GH per chip.
- But the chip page states 2.7 GH per chip.
Does that mean the 25 GH full kit's real hashrate is ~43GH(2.7 * 16), and the 400GH is ~691 GH(2.7 * 256)?
3) Are the inventory numbers for the Oct fullkits accurate? right now it shows a 378 unit left for the 400GH, if all 378 gets sold will they all be delivered in Oct?
Thanks, i like how this is not dependent on avalon and already has asic chips in hand.
based on demand much more demand for the 25G kits then for the full 40G miners. pricing is based on demonstrated demand and presale volume/speed |
|
|
|
|
my first impressions/thoughts:
1) ditch the cutout labelling. as it was mentioned, it may make these more identifiable to thieves (as if a big humming box isnt already) and may cause the airflow to be wonky, creating random circulation paths that counteract the fans and/or lead to more noisy fan rotation. A sticker/decal is equally functional and more appropriate
2) I dig the rubber feet. Without them this could be a source of vibrations from the spinning fans and sound terrible as your big hashing box vibrates against the table endlessly.
3) I like the simple design. The more basic the shape/materials, the more easily it can be replicated by anyone with connections to a CNC or laser/water cutter and generic hardware.
4) including a cutout or two for status LEDs or even a LCD would be cool. (I dont know if bitfury designed either of these into its hardware, but surely the RPi software can be modified to output status indicators)
5) A modified case that is simply a backplane might be nice for those who simply want to mount thier hardware to a surface but dont want a full case necessarily (as mentioned in #1, a simple backplane with mounting holes for the psu/Mboard/Hboards would be cheap and easy to source from local cnc/watercutting shops)
|
|
|
|
|
Show Posts
