ASICMINER: Entering the Future of ASIC Mining by Inventing It

[dexX7]

Does this count?:

so Gen2 is skipped?

I'd also be very interested to hear the answer to this.

It's frustrating seeing responses which indicate that all the questions in this thread are being read, yet ignored.

Yes, it was skipped. FC mentioned this in the past. Apparently the performance wasn't as good as they hoped and they just decided to move on... probably a good move.

According to friedcat Gen2 is not worthwhile for 2 reasons:
1) Gen2 would come out relatively late due to some factors outside of AM's control
2) Gen3 shows promising results and the roll-out seems to be better/more reliably organized

Thus funds are better to be spent on Gen3 hardware. The production schedule will depend on a risk/reward ratio determined by friedcat. Deviations from the schedule need to be anticipated when rolling out a new product technology, so hedge accordingly when speculating on AM.

Thanks for the quotes.

I'm not sure.. the problem I have is this: There are some sources (like the reputable, Chinese interview girl) and I really don't doubt that what they say is the truth, but it seems there is much lost in translation for example and it's not really an official source at all. Just think about the confussion that was created from this "1000 PH" statement. I just wish the first source would be friedcat and not a third party.

But thanks again.

[empoweoqwj]

Bring on the full-time PR person from AM. Sorely needed.

I'm sore tired of discussions of Gen1, Gen2, Gen3, and now today Gen1.5 AND Gen2.5 have been introduced lol

Won't even start about the nm discussions.

If you don't have facts, please stop making up version numbers of GenX.X - its turned into a bad farce.

[binaryFate]

Bring on the full-time PR person from AM. Sorely needed.

I'm sore tired of discussions of Gen1, Gen2, Gen3, and now today Gen1.5 AND Gen2.5 have been introduced lol

Won't even start about the nm discussions.

If you don't have facts, please stop making up version numbers of GenX.X - its turned into a bad farce.

+1
(reminds me we established a convention some time ago, something like g1b2 for the gen1-new blades, g1usb2, etc. I don't remember it...  That didn't account for the 1.5 and half-made 2.8 that apparently just appeared though...)

[Zubilica]

So much money for cooling . . .  for such inefficient chips.

I would rather see the 200TH mine that Bitfury is running.  I would bet it is all air cooled, and uses 1/5th the power of ASICMINER.

[Alphi]

So much money for cooling . . .  for such inefficient chips.

what you fail to realize is that just like stock coolers and water coolers for home PCs.. these coolers can be used to COOL any kind of chips provided the form factor is the same..
This means that the money spent on COOLING will pay off in the long run as each new generation of chips gets moved into/onto the submersion coolers..

looking at how much effort they went to in order to set up their immersion cooling.. you can bet that ASICminer plans to be around for a very long time...
even if eventually they fail at producing competitive chips and end up buying 3rd party chips.. they will still have a competitive advantage because they are way ahead of the competition in terms of cooling on a huge scale..

lets just hope they dont have any catastrophic failures in the cooling system leading to a Chernobyl style meltdown lols

[binaryFate]

lets just hope they dont have any catastrophic failures in the cooling system leading to a Chernobyl style meltdown lols

It seems much much more secure than any air-cooled data-center regarding fire or overheat actually! No risk of fire for all parts that are immerged,  and even if the pump/cooler breaks you have plenty of time to react.

[Kosmatos]

Judging from this, the dividend this week should be noticeably better than last week's.

https://blockchain.info/address/1HtUGfbDcMzTeHWx2Dbgnhc6kYnj1Hp24i

[Alphi]

It seems much much more secure than any air-cooled data-center regarding fire or overheat actually! No risk of fire for all parts that are immerged,  and even if the pump/cooler breaks you have plenty of time to react.

I was thinking more along the lines of a leak cause by an earthquake or some other disaster causing some of the vats to drain.. or someone sabotaging the system.. its pretty cut throat out there and there are more than a few dodgy 3rd party contractors who cut corners and/or can be bribed..

EDIT.. not saying it would happen simply playing devils advocate.
liquid cooling systems require far more maintenance and are far more prone to catastrophic failure than air cooled systems...

[dexX7]

https://blockchain.info/address/1HtUGfbDcMzTeHWx2Dbgnhc6kYnj1Hp24i

479.4131794 BTC / 400,000 shares = 0.00119853 BTC

Not much, but more indeed.

[explorer]

https://blockchain.info/address/1HtUGfbDcMzTeHWx2Dbgnhc6kYnj1Hp24i

479.4131794 BTC / 400,000 shares = 0.00119853 BTC

Not much, but more indeed.

Is that the solo address, or cumulative with franchisee revenues?

[romerun]

Knc coming with 20nm

[jbest8283]

MacDaddy coming with .67nm

[Zubilica]

Knc coming with 20nm

this is true they have send a notification

[jimmothy]

Knc coming with 20nm

this is true they have send a notification

Source?

[Zubilica]

Knc coming with 20nm

this is true they have send a notification

Source?

https://www.kncminer.com/news/news-69

[jimmothy]

Knc coming with 20nm

this is true they have send a notification

Source?

https://www.kncminer.com/news/news-69

Interesting thanks for sharing. Still not convinced this will result in significantly more efficient chips. As we have seen how AM 144nm tech competes with KNC 28nm, I wonder how 40nm vs 28nm will compare.

[strmir]

Knc coming with 20nm

this is true they have send a notification

Source?

https://www.kncminer.com/news/news-69

Interesting thanks for sharing. Still not convinced this will result in significantly more efficient chips. As we have seen how AM 144nm tech competes with KNC 28nm, I wonder how 40nm vs 28nm will compare.

There's a notice on the linked page that KnC's 20nm design should reduce power draw per GH by some 30%. That would be somewhere around 0.7W/GH, given that 28nm Jupiter is listed with ~1W/GH (at wall).

Now, IIRC 40nm ASICminer's chip is expected to perform at 0.2W/GH.

[jimmothy]

Knc coming with 20nm

this is true they have send a notification

Source?

https://www.kncminer.com/news/news-69

Interesting thanks for sharing. Still not convinced this will result in significantly more efficient chips. As we have seen how AM 144nm tech competes with KNC 28nm, I wonder how 40nm vs 28nm will compare.

There's a notice on the linked page that KnC's 20nm design should reduce power draw per GH by some 30%. That would be somewhere around 0.7W/GH, given that 28nm Jupiter is listed with ~1W/GH (at wall).

Now, IIRC 40nm ASICminer's chip is expected to perform at 0.2W/GH.

Is 0.7 w/gh for knc chip or entire asic?

[strmir]

Knc coming with 20nm

this is true they have send a notification

Source?

https://www.kncminer.com/news/news-69

Interesting thanks for sharing. Still not convinced this will result in significantly more efficient chips. As we have seen how AM 144nm tech competes with KNC 28nm, I wonder how 40nm vs 28nm will compare.

There's a notice on the linked page that KnC's 20nm design should reduce power draw per GH by some 30%. That would be somewhere around 0.7W/GH, given that 28nm Jupiter is listed with ~1W/GH (at wall).

Now, IIRC 40nm ASICminer's chip is expected to perform at 0.2W/GH.

Is 0.7 w/gh for knc chip or entire asic?

They list Jupiter efficiency for the entire miner (at wall), I don't know how it's in reality though . So, I'd say that 0,7W/GH should be for entire miner also. But can not speak for themselves.

[data]

Not wanting to ridicule KnC or anything, but once again ASICMINER is the only company not taking preorders. With the current increases in network and unclear deployments by others, you are taking a huge gamble on ordering anything that would not be able to make back it's money even in a 1EH-network.