ASICMINER: Entering the Future of ASIC Mining by Inventing It

[rudi]

No, there won't be any stratum problem using cgminer with Raspberry PI.

So, we now have software that doesn't have the stratum problem. Meanwhile, the 2.4PH/s btcguild userid 566620 has disappeared in the past 48 hours, causing btcguild pool hash rate to dip significantly. And the 1H7FpBV8huVmPvt5eRzszAkgnERH2FaWvq address which luffy27 and others speculate to belong to AM has likewise stopped receiving payouts.

Disclaimer: correlation != causation.

This might have to do with my ealier prediction that AM's hashrate will soon be large enough to start self-mining again. They might be quitting BTCGuild in order to prepare for self-mining.

Disclaimer: Very speculative.

[hdbuck]

No, there won't be any stratum problem using cgminer with Raspberry PI.

So, we now have software that doesn't have the stratum problem. Meanwhile, the 2.4PH/s btcguild userid 566620 has disappeared in the past 48 hours, causing btcguild pool hash rate to dip significantly. And the 1H7FpBV8huVmPvt5eRzszAkgnERH2FaWvq address which luffy27 and others speculate to belong to AM has likewise stopped receiving payouts.

Disclaimer: correlation != causation.

This might have to do with my ealier prediction that AM's hashrate will soon be large enough to start self-mining again. They might be quitting BTCGuild in order to prepare for self-mining.

Disclaimer: Very speculative.

hell yea, plug em self-bitches in!

Disclamer: very enthusiatic

[rudi]

Brief
ASICMiner Prisma is based on a proprietary X48 design for BE200. It greatly reduces power
consumption of the device per GH/s compared to the X24 design.

...

Power consumption: ~0.67 W/G board-wise. On-wall power consumption relies on PSU
efficiency. (typically 1050-1100 W per full device, 0.75-0.78 W/G)

Nice. I wonder how they managed to squeeze improve power efficiency by ~25% compared to the X24 design.

But even though this further destroys the mining competition, it probably requires nerves of steel to continue investing in miners at this time.

I'm also very impressed by this. Not only that they managed to squeeze so much power efficiency, but also the size of the device.

Regarding efficiency: Do I understand correctly that 0.75 W/G puts the Prisma at the top in terms of efficiency among devices that are currently shipping?
Is there any other company currently shipping devices with efficiency equal or greater to 0.75 W/G?

[funkymunky]

This I have been wondering too.

The fact RS hasn't been commenting too much could mean this is the case 

Brief
ASICMiner Prisma is based on a proprietary X48 design for BE200. It greatly reduces power
consumption of the device per GH/s compared to the X24 design.

...

Power consumption: ~0.67 W/G board-wise. On-wall power consumption relies on PSU
efficiency. (typically 1050-1100 W per full device, 0.75-0.78 W/G)

Nice. I wonder how they managed to squeeze improve power efficiency by ~25% compared to the X24 design.

But even though this further destroys the mining competition, it probably requires nerves of steel to continue investing in miners at this time.

I'm also very impressed by this. Not only that they managed to squeeze so much power efficiency, but also the size of the device.

Regarding efficiency: Do I understand correctly that 0.75 W/G puts the Prisma at the top in terms of efficiency among devices that are currently shipping?
Is there any other company currently shipping devices with efficiency equal or greater to 0.75 W/G?

[xhomerx10]

Brief
ASICMiner Prisma is based on a proprietary X48 design for BE200. It greatly reduces power
consumption of the device per GH/s compared to the X24 design.

...

Power consumption: ~0.67 W/G board-wise. On-wall power consumption relies on PSU
efficiency. (typically 1050-1100 W per full device, 0.75-0.78 W/G)

Nice. I wonder how they managed to squeeze improve power efficiency by ~25% compared to the X24 design.

But even though this further destroys the mining competition, it probably requires nerves of steel to continue investing in miners at this time.

I'm also very impressed by this. Not only that they managed to squeeze so much power efficiency, but also the size of the device.

Regarding efficiency: Do I understand correctly that 0.75 W/G puts the Prisma at the top in terms of efficiency among devices that are currently shipping?
Is there any other company currently shipping devices with efficiency equal or greater to 0.75 W/G?

 There is nothing so surprising about this design.  They have simply down-clocked the chips and doubled the number per board.  They could have done this in June and had the most efficient machine on the market but the chips were worth more at the time (at least that was the thinking) so it didn't make sense.  

 Here's the math:
 240MHz/270MHz(clock ratio or prisma to tube) * 800(1 Tube hash rate) * 2 (for the doubling of the chips) =1.4 TH/s

 With some extra cooling and a little know-how, you can run at 1.6+ THs with Tube-like effifciency. (NB over-clocking the proprietary design is not recommended as it could result in a damaged product)

 While I'm not impressed by the new design, I am immpressed with AMs ability to maximize the value of their product as computational difficulty rises.

[webbrowser]

There is nothing so surprising about this design.  They have simply down-clocked the chips and doubled the number per board.  They could have done this in June and had the most efficient machine on the market but the chips were worth more at the time (at least that was the thinking) so it didn't make sense.  

 Here's the math:
 240MHz/270MHz(clock ratio or prisma to tube) * 800(1 Tube hash rate) * 2 (for the doubling of the chips) =1.4 TH/s

 With some extra cooling and a little know-how, you can run at 1.6+ THs with Tube-like effifciency. (NB over-clocking the proprietary design is not recommended as it could result in a damaged product)

 While I'm not impressed by the new design, I am immpressed with AMs ability to maximize the value of their product as computational difficulty rises.

Good point.  Someone was asking a few days ago about low voltage efficiency of the chip; now we know.

Is this the lowest sensible underclocking speed, or is there further room for this in the future, i wonder?

[xhomerx10]

It's easy to be the best when you set illogical limits.
 They are certainly the only company named Asicminer shipping at this efficiency ergo we're the best! (as far as companies named Asicminer) WHOO!
 What I mean is, who cares about shipping when all a company need do is SELL?

This I have been wondering too.

The fact RS hasn't been commenting too much could mean this is the case 

Brief
ASICMiner Prisma is based on a proprietary X48 design for BE200. It greatly reduces power
consumption of the device per GH/s compared to the X24 design.

...

Power consumption: ~0.67 W/G board-wise. On-wall power consumption relies on PSU
efficiency. (typically 1050-1100 W per full device, 0.75-0.78 W/G)

Nice. I wonder how they managed to squeeze improve power efficiency by ~25% compared to the X24 design.

But even though this further destroys the mining competition, it probably requires nerves of steel to continue investing in miners at this time.

I'm also very impressed by this. Not only that they managed to squeeze so much power efficiency, but also the size of the device.

Regarding efficiency: Do I understand correctly that 0.75 W/G puts the Prisma at the top in terms of efficiency among devices that are currently shipping?
Is there any other company currently shipping devices with efficiency equal or greater to 0.75 W/G?

[Blazed]

You guys need some new signatures!  If the prices were a tad bit lower I could see a lot of people buying the "mega" tubes.

[rudi]

There is nothing so surprising about this design.  They have simply down-clocked the chips and doubled the number per board.  They could have done this in June and had the most efficient machine on the market but the chips were worth more at the time (at least that was the thinking) so it didn't make sense.  

 Here's the math:
 240MHz/270MHz(clock ratio or prisma to tube) * 800(1 Tube hash rate) * 2 (for the doubling of the chips) =1.4 TH/s

 With some extra cooling and a little know-how, you can run at 1.6+ THs with Tube-like effifciency. (NB over-clocking the proprietary design is not recommended as it could result in a damaged product)

 While I'm not impressed by the new design, I am immpressed with AMs ability to maximize the value of their product as computational difficulty rises.

Good points, thank you.

[am-informer]

Is this a new website?  

http://www.iasicminer.com/index.php/action-channel-name-home

 

No. There's already a project for translating it: https://bitcointalk.org/index.php?topic=768327.msg8848001#msg8848001

To explain it better: http://www.iasicminer.com/ is the official ASICMINER homepage made by ASICMINER. It is not fully translated to english, so the above project of mine began translating parts of the page with the plan to put it all together in a nice dashboard / news site.

Finally the official ASICMINER twitter account is the one you point to yes: https://twitter.com/ASICMinerANN

[Jayjay04]

Where can we see the prisma ?

[am-informer]

Another two news articles are ready from the translation team:

Post 1:
---
title CN : 烤猫官方矿机，终极降价大酬宾，800G+，仅0.79BTC
title EN : Friedcat official Miner, Big sale, 800G+, only 0.79BTC
url of transl. source: http://www.iasicminer.com/index.php/action-viewnews-itemid-56
url of transl. result: https://bitcointalk.org/index.php?topic=768327.msg8940043#msg8940043

Post 2:
---
title CN : Bitcointalk矿机评测第一达人doggie综合评测烤猫绿方
title EN : Bitcointalk Miner Review – Doggie Comprehensive guide on Friedcat’s AM-TUBE
url of transl. source: http://www.iasicminer.com/index.php/action-viewnews-itemid-57
url of transl. result: https://bitcointalk.org/index.php?topic=768327.msg8940141#msg8940141


PS. I know some of that info is in the forum threads already for weeks and is not news at all. The goal of the project is to translate the iasicminer.com website in the best way possible for reference.

PS2. Any translation quality suggestions are welcome!


Next up for translation is the PRISMA article at http://www.iasicminer.com/index.php/action-viewnews-itemid-58 | 基于全新方案ASICMiner Prisma矿机来袭

[stompysteve]

Where can we see the prisma ?

there was a picture up earlier the link is dead now, it looks like a 1.5x tube or at least the board does

Edit: Here ya go

[whalezy]

 There is nothing so surprising about this design.  They have simply down-clocked the chips and doubled the number per board.  They could have done this in June and had the most efficient machine on the market but the chips were worth more at the time (at least that was the thinking) so it didn't make sense.  

 Here's the math:
 240MHz/270MHz(clock ratio or prisma to tube) * 800(1 Tube hash rate) * 2 (for the doubling of the chips) =1.4 TH/s

 With some extra cooling and a little know-how, you can run at 1.6+ THs with Tube-like effifciency. (NB over-clocking the proprietary design is not recommended as it could result in a damaged product)

 While I'm not impressed by the new design, I am immpressed with AMs ability to maximize the value of their product as computational difficulty rises.

XBTec(https://bitcointalk.org/index.php?topic=626694.0) who employs AM's chip has already released miners at this lever of power consumption, I guess they also followed this idea.

[MichaelBliss]

 There is nothing so surprising about this design.  They have simply down-clocked the chips and doubled the number per board.  They could have done this in June and had the most efficient machine on the market but the chips were worth more at the time (at least that was the thinking) so it didn't make sense.  

 Here's the math:
 240MHz/270MHz(clock ratio or prisma to tube) * 800(1 Tube hash rate) * 2 (for the doubling of the chips) =1.4 TH/s

 With some extra cooling and a little know-how, you can run at 1.6+ THs with Tube-like effifciency. (NB over-clocking the proprietary design is not recommended as it could result in a damaged product)

 While I'm not impressed by the new design, I am immpressed with AMs ability to maximize the value of their product as computational difficulty rises.

XBTec(https://bitcointalk.org/index.php?topic=626694.0) who employs AM's chip has already released miners at this lever of power consumption, I guess they also followed this idea.

Really?  I thought XBTec's Pacific got something around 1W/G?  Prisma's is 0.75 - 0.78 W/G.

[Gws24]

Not impressed, Bitmains s3+ is already 0.78 W at the wall and announced the S4 which should be even better (<== just my expectation; no real specs yet) although it should have come out last week.

[Chris_Sabian]

Not impressed, Bitmains s3+ is already 0.78 W at the wall and announced the S4 which should be even better (<== just my expectation; no real specs yet) although it should have come out last week.

But how long will it take Bitmain to ship?  For what I hear, they are having problems shipping on time and are having hardware issues.

[RoadStress]

I'm also very impressed by this. Not only that they managed to squeeze so much power efficiency, but also the size of the device.

Regarding efficiency: Do I understand correctly that 0.75 W/G puts the Prisma at the top in terms of efficiency among devices that are currently shipping?
Is there any other company currently shipping devices with efficiency equal or greater to 0.75 W/G?

Yes there are other companies that are currently shipping devices with efficiency equal or better than 0.75W/G.

Not impressed, Bitmains s3+ is already 0.78 W at the wall and announced the S4 which should be even better (<== just my expectation; no real specs yet) although it should have come out last week.

But how long will it take Bitmain to ship?  For what I hear, they are having problems shipping on time and are having hardware issues.

They already shipped 9 batches!

[rudi]

Yes there are other companies that are currently shipping devices with efficiency equal or better than 0.75W/G.

Ok. Which companies? What are the products? How much do they cost?

[Gws24]

Obviously the cost of the Prism is very good compared to say the 3s+ (< 1BTC/TH vs 1.29BTC/TH) so my remark of not being impressed is only aimed at the power consumption. That being said I do think there are a couple of other things that are a disadvantage for the prism when compared to other miners:

- the use of a separate controller and if you want to mine on certain pools you need a pi
- for a small miner only available through resellers

That may not be a big problem for the big boys where most of the revenue for asicminer is coming from but for a small miner like myself it was when I was looking to buy a miner last week. I Chose the s3+ for the above reasons over the tube besides the fact that the s3+ is more power efficient as the tube. Obviously, I would have had a harder choice this week with the price difference between the Prism and s3+ at comparable power consumption.

Just my two cents.