ASICMINER: Entering the Future of ASIC Mining by Inventing It

[necro_nemesis]

How do you arrive at AM being Allied Control?

[1713857223]

1713857223

[1713857223]

1713857223

[1713857223]

1713857223

[maxl]

I think they worked together to make the immersion cooled datacenter in Hongkong.

[spartan82]

That's not enough to say that asicminer is manufacturing and supplying these containers. The hardware that will be housed by these containers on the other hand....

[necro_nemesis]

It's a concept.

Why don't we just build a sea lab near a fissure in the earth's crust to house a data center. We can use the ocean's cool water to dissipate heat from our Novec and power the facility via geothermal energy rising from the fissure. When boards fail Jar Jar can swim down and swap them out.  

[bitcoin.newsfeed]

So AM is selling containers!

No. Its not. But ASICminer paid for R&D of these "proof of concept" 2-phase immersion cooling datacenters, but Friedcat suddenly decided to abandon this manufacturing and mining path and only sell chips. 

[Caesium]

[..] Friedcat suddenly decided to abandon this manufacturing and mining path and only sell chips.  

Source? Or is this you presenting your own assumptions as cast in iron facts?

[arnuschky]

What's with fuse F1? Bunch of diodes on this side of the board not showing, ICs U10, U9 and U6, switch SW1, resistor R22, and that fuse looks to be coming up short.

What is your point? And where did this image come from?

[ning]

Here's the source of the image above: http://blog.rockminer.com/#!/2014/04/11/Rockminer_update_of_11_April

[CMMPro]

Check it against the reference design posted earlier.

I haven't done so, but unpopulated components on boards is not something to be worried about at this point.
In my design iterations I have done this as well...where certain board features were added for testing purposes but not for large scale production. You aren't going to make several different board designs at this low level of production.

The missing features could also be for in circuit programming or electrical testing...or they may have simplified the design.

It could also be that they found out that since some components were faulty (as they have stated publicly) so they stopped the line and pulled these out not fully populated.

I really don't think this is a labcoin situation, those boards were so obviously fake garbage it wasn't funny.

These at look completely legit to me other than being missing some components.


Edit: After looking a the pictures more closely it looks to me like most of the power source circuit is missing.
The large connector missing at the board edge is the power input and you see the ground plane, large diodes, mosfet/transistor etc all entering the board and in that area. what they are doing is likely using a bench power supply to test the design and forgoing the use of the production physical connectors etc. The more precise bench supply allows them to skip all the circuits related to smoothing and level shifting the power source. 


 

[necro_nemesis]

I came to many of the same conclusions about omitted components but then there's F1 which leads me to believe what was on the Gerber for the board is not what got ordered and attempted to be stuffed.

Also all boards are seemingly identical so if they pulled components they already did all. Conclusions about the boards simply provide further evidence that the ASIC is not the immediate concern.

@rockxie : You're on here trolling for IPO money and potential miner buyers; in return in all fairness what's currently up with the ASIC?

[Draino]

http://blog.rockminer.com/#!/2014/04/13/Testing_Results_Of_BE200.md


Testing Results Of BE200

We've got the results of one good testing board,it seems not  good,but sill can be accepted.We will receive more chips at next weekend if things are going well .

Results:

Board:one chip testing board
Frequency:360Mhz
Volt:0.72V
Hashrate per chip:11.52Ghash
Power consumption:6.375W per chip
Power consumption per Ghash:6.375/11.52=0.5539W/Ghash
After power supply changeover:0.5539/81% = 0.684W/Ghash（at blade）
Power consumption on wall:0.684/0.8 = 0.855W/G
Adding other components loss about 1KW/Thash
Tips:this result is not very accurate just for reference.

[chriswilmer]

http://blog.rockminer.com/#!/2014/04/13/Testing_Results_Of_BE200.md


Testing Results Of BE200

We've got the results of one good testing board,it seems not  good,but sill can be accepted.We will receive more chips at next weekend if things are going well .

Results:

Board:one chip testing board
Frequency:360Mhz
Volt:0.72V
Hashrate per chip:11.52Ghash
Power consumption:6.375W per chip
Power consumption per Ghash:6.375/11.52=0.5539W/Ghash
After power supply changeover:0.5539/81% = 0.684W/Ghash（at blade）
Power consumption on wall:0.684/0.8 = 0.855W/G
Adding other components loss about 1KW/Thash
Tips:this result is not very accurate just for reference.

Seems pretty good to me! Good enough to mass produce 100,000+ chips go go go go go g og ggogogogogo!

[Chris_Sabian]

11.52Ghash at 6.375W...

Here I am hashing away at 37 Ghash at 320W

[SebastianJu]

So AM is selling containers!

No. Its not. But ASICminer paid for R&D of these "proof of concept" 2-phase immersion cooling datacenters, but Friedcat suddenly decided to abandon this manufacturing and mining path and only sell chips. 

The decision of friedcat to not make enduser sales anymore but switch to b2b only doesnt mean that friedcat wont host his own miner farm. If you want to claim that you would need to say where you got your info from.

In fact its a good decision. Selling to resellers means more profit with less work and you havent to deal with refunds, support an all that kind of really time intensive work.

[lophie]

11.52Ghash at 6.375W...

Here I am hashing away at 37 Ghash at 320W

what is your hardware?

[Chris_Sabian]

11.52Ghash at 6.375W...

Here I am hashing away at 37 Ghash at 320W

what is your hardware?

AM Cube. 

Bought a few in November.  Sold a couple and mined with the others.

Paid bitcoin for them and was able to get a positive ROI.  Also was able to partially heat part of the house in the winter so the cost of heating was less. (still paid for the electricity but recouped the cost by mining bitcoins)

[chriswilmer]

http://blog.rockminer.com/#!/2014/04/13/Testing_Results_Of_BE200.md


Testing Results Of BE200

We've got the results of one good testing board,it seems not  good,but sill can be accepted.We will receive more chips at next weekend if things are going well .

Results:

Board:one chip testing board
Frequency:360Mhz
Volt:0.72V
Hashrate per chip:11.52Ghash
Power consumption:6.375W per chip
Power consumption per Ghash:6.375/11.52=0.5539W/Ghash
After power supply changeover:0.5539/81% = 0.684W/Ghash（at blade）
Power consumption on wall:0.684/0.8 = 0.855W/G
Adding other components loss about 1KW/Thash
Tips:this result is not very accurate just for reference.

Seems pretty good to me! Good enough to mass produce 100,000+ chips go go go go go g og ggogogogogo!

Seriously... shouldn't we be more excited about this? I mean, even 1 KW/Thash is fine. As far as I know nobody else can do better and AM can probably out cost anyone. AM should start super mass producing these right? Right?

[Franktank]

For those wondering comparisons with the competition:

Note: "Efficiency" is comparing the the ROCKMiner test blade with all other hardware. Data was obtained from The Genesis Block. ROCKMiner blade is only after very few, unoptimized tests. It is subject to change.

EDIT: Here are the updated numbers, please let me know of any more issues. Sorry for the confusion all.

EDIT2: I've taken down the chart. I'll repost it again when they have the final specifications. I apologize for any issues.

[eastpk]

I think that Spoondoolies is missing from that list. Granted they are pretty new - I'm fairly certain their numbers come close to AM.

[Franktank]

I think that Spoondoolies is missing from that list. Granted they are pretty new - I'm fairly certain their numbers come close to AM.

I'll take a look at it, I basically copied the data from The Genesis Block and from Rockxie's initial tests post. I'll reiterate that all numbers from the chart are purely hypothetical, with the exception of the ROCKMiner blade (based off of initial tests).