I had order #217. It's hashing with only 54 watts from the wall!!
Very much thanks to Punin & Bitfury group!
Topic: [ANN] Bitfury ASIC sales in EU and Europe - page 85. (Read 250482 times)
Hi!
I had order #217. It's hashing with only 54 watts from the wall!!
Very much thanks to Punin & Bitfury group!
I had order #217. It's hashing with only 54 watts from the wall!!
Very much thanks to Punin & Bitfury group!
Dyaheon,
did you get a tracking number when your unit was shipped and do you mind posting a photo of the assembled unit?
Thanks!
spiccioli
SMT resistor costs 0.005 in quantity. How much does thermal glue and 10 heatsinks cost?
I am not sure if I understand what you mean? Resistor is not a magic solution to generated heat? Punin was saying that by changing the resistor they are increasing the voltage and achieve higher clock rate. In expense of higher heat generation. Therefore I was wondering why wouldnt they glue heatsinks on the boards to cool them a bit. If you get heatsinks/glue in mass quantities the cost would be nearly nothing...
I was the second order (#218), and now happily hashing away 
Went for the install before instructions were available, and powered one H-board when it was oriented the wrong way
Luckily only one of the 16 chips was burned and it's hashing fine now. H-board components should be facing the raspberry. Gotta be careful with this stuff.
Ordered 6 starter kits, got 8 to hit the promised hashrate as they're slightly underperforming. Getting ~150-157GH, I think. It varies a bit and I don't have a good reading from pool yet.
Power usage is 120W at the wall, which comes to just 0.8J/GH. Very nice. No noise either obviously as it's passively cooled, which is great if you're going to be mining somewhere where that matters. Compact design is a plus too.
A crappy (sorry!) pic of a starter kit:
Edit: 2 of the boards are slow. Not sure what's the matter with those, they seem to lose speed over time, with some chips going offline. Perhaps the software needs a bit more polishing
Went for the install before instructions were available, and powered one H-board when it was oriented the wrong way
Luckily only one of the 16 chips was burned and it's hashing fine now. H-board components should be facing the raspberry. Gotta be careful with this stuff.Ordered 6 starter kits, got 8 to hit the promised hashrate as they're slightly underperforming. Getting ~150-157GH, I think. It varies a bit and I don't have a good reading from pool yet.
Power usage is 120W at the wall, which comes to just 0.8J/GH. Very nice. No noise either obviously as it's passively cooled, which is great if you're going to be mining somewhere where that matters. Compact design is a plus too.
A crappy (sorry!) pic of a starter kit:
Edit: 2 of the boards are slow. Not sure what's the matter with those, they seem to lose speed over time, with some chips going offline. Perhaps the software needs a bit more polishing
SMT resistor costs 0.005 in quantity. How much does thermal glue and 10 heatsinks cost?
By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Why not glue heatsinks on the chips?
Anyone with changed status? Mine is #26x and is still "Processing".
First orders shipped today! More to follow..
Congrats!
I'm speechless. I absolutely adore the simplicity of BitFury solution and the results of that design. You've done with 55nm chips what certain infamous American company is hoping to achieve using 28nm process. Well done, punin, well done.
First orders shipped today! More to follow..
Exciting!
Starter kits benefit the most, as we are putting 2 H-cards in until we have achieved sustained noncerate (useful nonces) of over 25GH/s on one card.
Even more exciting!!
But wait, this doesn't mean the current speed of the H-cards is only 12,5 GH/s ?
No, it means I don't have time to desolder chips from them
You'll be surprised what you get.But that's just my guess...
No, because there are like 150 interconnections. Someone posted that info somewhere, too hard to find right now
First orders shipped today! More to follow..
+1
Anyway you could ... give us the last order number you shipped ... :-) curious where I stand ...
+1, please do it =)
To complete it: 28X didn't make it into todays shipment, sadly. Still says "processing"
First orders shipped today! More to follow..
Exciting!
Starter kits benefit the most, as we are putting 2 H-cards in until we have achieved sustained noncerate (useful nonces) of over 25GH/s on one card.
Even more exciting!!
But wait, this doesn't mean the current speed of the H-cards is only 12,5 GH/s ?
No, it means I don't have time to desolder chips from them
You'll be surprised what you get.But that's just my guess...
Watching in earnest - hopefully we're entering the age of honest ASIC traders that deliver on their promises!
Ok we did some quick test now with another R01F resistor (don't try this at home please unless you're experienced in SMT soldering). This will void your warranty!
By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the powerconsumption is.... 250W AT THE WALL.
When full sized (400gh, 16 boards) fall 10-20% short, i guess so does every single h board? Wow, two times 80-90% is like 160 - 180%, thats 40-45 gh per original uni By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the powerconsumption is.... 250W AT THE WALL.
plus 250w on 400, thats 60-70w per two hboards. This. Is. Madness. Im freaking excited!
What's was/is the original resistor value and resultant voltage?
1.5k for 0V7
Ok we did some quick test now with another R01F resistor (don't try this at home please unless you're experienced in SMT soldering). This will void your warranty!
By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the powerconsumption is.... 250W AT THE WALL.
By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the powerconsumption is.... 250W AT THE WALL.
What's was/is the original resistor value and resultant voltage?
Congrats on shipping!!
In photo posted by Nico it's clearly visible that capacitors near coil are askew mounted. Also in big zoom if think shortcuts between chip pins are visible. Maybe thats a reason why some boards perform poorly...
Maybe manufacturer is quick but quality suffers greatly.
Maybe manufacturer is quick but quality suffers greatly.
Can you link post plz?
ty
https://bitcointalksearch.org/topic/m.2978673
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the power consumption is.... 250W AT THE WALL.
Thanks for the update.
In photo posted by Nico it's clearly visible that capacitors near coil are askew mounted. Also in big zoom if think shortcuts between chip pins are visible. Maybe thats a reason why some boards perform poorly...
Maybe manufacturer is quick but quality suffers greatly.
Maybe manufacturer is quick but quality suffers greatly.
Can you link post plz?
ty
Ok we did some quick test now with another R01F resistor (don't try this at home please unless you're experienced in SMT soldering). This will void your warranty!
By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the powerconsumption is.... 250W AT THE WALL.
By changing the resistor to 2.2k gives us 0V74 Vcore. This will push the hashrate up a bit and support higher clocks. We will run a production batch with this resistor to try to bring the cards within spec. Fans are highly recommended in anything but polar regions
Meanwhile full sized miner is falling quite a bit short (10-20%) on hashrate, but the powerconsumption is.... 250W AT THE WALL.
In photo posted by Nico it's clearly visible that capacitors near coil are askew mounted. Also in big zoom if think shortcuts between chip pins are visible. Maybe thats a reason why some boards perform poorly...
Maybe manufacturer is quick but quality suffers greatly.
Maybe manufacturer is quick but quality suffers greatly.
Can you link post plz?
ty
In photo posted by Nico it's clearly visible that capacitors near coil are askew mounted. Also in big zoom shortcuts between chip pins are visible. Maybe thats a reason why some boards perform poorly...
Maybe manufacturer is quick but quality suffers greatly.
Maybe manufacturer is quick but quality suffers greatly.
Jump to: