HashFast announces specs for new ASIC: 400GH/s - page 579.

Ytterbium

full member

Activity: 238

Merit: 100

Quote from: eve on July 28, 2013, 04:26:23 AM

No body want a pre order , Avalon and Butterfly have turn out to be nothing but S..M with all their non deliveries and excuses.

I just said I want a pre-order. I pre-ordered Avalon batch 2 and have my unit, and I just pre-ordered from KnC. These guys seem legit as well.

If you can figure out which guys are the scammers and which are legit then per-ordering is a much cheaper way to get hardware. If a company uses it's own money (or an IPO, like ASICminer) to develop chips then they can keep them all for themselves rather then sell them, or sell them at prices that can make it hard to get an ROI.

eve

full member

Activity: 210

Merit: 100

No body want a pre order , Avalon and Butterfly have turn out to be nothing but S..M with all their non deliveries and excuses.

qwk

donator

Activity: 3542

Merit: 3413

Shitcoin Minimalist

Quote from: Ytterbium on July 28, 2013, 01:17:36 AM

Quote from: eve on July 27, 2013, 09:39:17 PM

Remember we don't want any pre order or IPO.

Uh, I totally want a pre-order.

Me too.

Ytterbium

full member

Activity: 238

Merit: 100

Quote from: eve on July 27, 2013, 09:39:17 PM

Remember we don't want any pre order or IPO.

Uh, I totally want a pre-order.

Luckybit

hero member

Activity: 714

Merit: 510

Quote from: Simon Barber on July 27, 2013, 09:59:08 PM

Quote from: Luckybit on July 26, 2013, 11:20:31 PM

In my opinion Bitcoin being deflationary does not automatically lead to it being a moral hazard. I think it's a possibility but it's not the result of the cap or it being deflationary. I don't consider it moral hazard because the early adopters are also the ones taking the biggest risks if government decides to crack down. In order to actually defend something like Bitcoin from the established players within the industry who would seek to politically and socially destroy it (by claiming all Bitcoiners are child pornographers, druggies, terrorists, etc), there must be deflation which rewards people for holding Bitcoin. This makes Bitcoin sticky, and only by making it sticky can it defeat increasing social stigma.

If enough people get rich because they saved Bitcoins, those people will have stories to tell which wont involve them being a child pornographer, terrorist, drug dealer (or user). Those stories are critical for the success of any currency. Just as the US dollar relied on the story of the American dream where anyone could get rich if they just work hard, only with Bitcoin it's currently true while with the dollar at this point it's a blatant lie.

I do agree with you on the idea of organic inflation but I don't think we will need that for many years. Netcoin intends to do what you're talking about by building a democratic mechanism into the design. These mechanisms can allow for stakeholders to eventually vote on the inflation rate.

Please have take a look at Netcoin http://www.netcoin.io/wiki/Netcoin:Community_Portal
and if you have time give a review of some of the proposed features, in particular the features I proposed:

Having a stakeholder based voting mechanism to determine money supply is a very interesting idea. The exact process by which this works is critical, to avoid gaming of the system. The biggest issue with alt-coins is that the network effect applies very strongly to moneys in general - i.e. the more users a currency has, the more useful it is. In the history of technologies with strong network effects there are very few (if any?) cases where a later technology that only improves in a small way compared to an earlier one takes over. I believe that only something massively disruptive could displace bitcoin.

I'm not too worried about deflation - deflation in technology products is well known (PCs have been decreasing in price for years), and it doesn't stop people from buying them. The limited liquidity has made bitcoin particularly susceptible to momentum driven bubbles, and I look forward to higher volume and more derivative products to help stabilize the value. This will help adoption.

Enough for this thread - getting dangerously off topic!

Myspace had the network effect but it was replaced by Facebook. Myspace had superior technology. I think the threat to Bitcoin is stigma and reputation, it's associated with Wikileaks, Silk Road, and the dark net. It's reputation may allow for other coins to catch up merely because those other coins have no reputation and can be offered as "underlying technology" or pure protocols. Bitcoin can adapt by going stealth, becoming an underlying protocol instead of being marketed as "Bitcoin". Amazon coin for instance could be powered by Bitcoin without anyone knowing it. The ideal situation in my opinion is the PaySwarm approach which is to build cryptocurrencies into the web standard itself and then not set Bitcoin up to be the single dominant cryptocurrency but to let the user choose which underlying cryptocurrency to use.

If you don't want to use Google as the default search engine there is Bing, Yahoo or whatever. I realize we are getting off topic, I apologize for that.

Bicknellski

hero member

Activity: 924

Merit: 1000

If we are looking at October delivery on these... that means any other chip players in the market could be pooched.

gateway

hero member

Activity: 552

Merit: 500

Quote from: Simon Barber on July 27, 2013, 10:18:37 PM

Quote from: rograz on July 27, 2013, 06:04:56 AM

what I don't get is why anyone so experienced would build such a massive chip for something like btc mining. With big chips you get worse yields, usually lower clock potential and more thermal issues. I can only really think of getting slightly better performance/area from going big but then you might not gain anything due to smaller chips normally clocking higher/better yield.

A bitcoin chip does not suffer from nearly the same level of yield issues that regular ICs do. With regular ICs if there is the slightest imperfection anywhere on the chip, then the whole chip becomes useless. E.G. A CPU that executes one particular instruction wrong is pretty much useless to anyone. A bitcoin chip is composed of many replicas of a single mining core or engine. A tiny defect is very likely to only affect a single core or engine, so reduce the chips performance by only a tiny fraction.

In a normal chip it's necessary to build a clock tree to maintain clock synchronization over much of the chip. This is because there are lots of dependencies and connections between logic in one part of the chip and logic somewhere else. In a bitcoin chip each engine or core is mostly independent from the others, so the clock only needs to be synchronous within one engine - much smaller than the whole chip. Effectively it's lots of small chips in a big chip.

What's important in designing and optimizing a bitcoin chip to produce this sort of result is a very different set of skills from what's important in the normal process of SOC design (system on a chip design), and takes a highly experienced team.

+1

Simon Barber

newbie

Activity: 56

Merit: 0

Quote from: rograz on July 27, 2013, 06:04:56 AM

what I don't get is why anyone so experienced would build such a massive chip for something like btc mining. With big chips you get worse yields, usually lower clock potential and more thermal issues. I can only really think of getting slightly better performance/area from going big but then you might not gain anything due to smaller chips normally clocking higher/better yield.

A bitcoin chip does not suffer from nearly the same level of yield issues that regular ICs do. With regular ICs if there is the slightest imperfection anywhere on the chip, then the whole chip becomes useless. E.G. A CPU that executes one particular instruction wrong is pretty much useless to anyone. A bitcoin chip is composed of many replicas of a single mining core or engine. A tiny defect is very likely to only affect a single core or engine, so reduce the chips performance by only a tiny fraction.

In a normal chip it's necessary to build a clock tree to maintain clock synchronization over much of the chip. This is because there are lots of dependencies and connections between logic in one part of the chip and logic somewhere else. In a bitcoin chip each engine or core is mostly independent from the others, so the clock only needs to be synchronous within one engine - much smaller than the whole chip. Effectively it's lots of small chips in a big chip.

What's important in designing and optimizing a bitcoin chip to produce this sort of result is a very different set of skills from what's important in the normal process of SOC design (system on a chip design), and takes a highly experienced team.

Simon Barber

newbie

Activity: 56

Merit: 0

Quote from: Luckybit on July 26, 2013, 11:20:31 PM

In my opinion Bitcoin being deflationary does not automatically lead to it being a moral hazard. I think it's a possibility but it's not the result of the cap or it being deflationary. I don't consider it moral hazard because the early adopters are also the ones taking the biggest risks if government decides to crack down. In order to actually defend something like Bitcoin from the established players within the industry who would seek to politically and socially destroy it (by claiming all Bitcoiners are child pornographers, druggies, terrorists, etc), there must be deflation which rewards people for holding Bitcoin. This makes Bitcoin sticky, and only by making it sticky can it defeat increasing social stigma.

If enough people get rich because they saved Bitcoins, those people will have stories to tell which wont involve them being a child pornographer, terrorist, drug dealer (or user). Those stories are critical for the success of any currency. Just as the US dollar relied on the story of the American dream where anyone could get rich if they just work hard, only with Bitcoin it's currently true while with the dollar at this point it's a blatant lie.

I do agree with you on the idea of organic inflation but I don't think we will need that for many years. Netcoin intends to do what you're talking about by building a democratic mechanism into the design. These mechanisms can allow for stakeholders to eventually vote on the inflation rate.

Please have take a look at Netcoin http://www.netcoin.io/wiki/Netcoin:Community_Portal
and if you have time give a review of some of the proposed features, in particular the features I proposed:

Having a stakeholder based voting mechanism to determine money supply is a very interesting idea. The exact process by which this works is critical, to avoid gaming of the system. The biggest issue with alt-coins is that the network effect applies very strongly to moneys in general - i.e. the more users a currency has, the more useful it is. In the history of technologies with strong network effects there are very few (if any?) cases where a later technology that only improves in a small way compared to an earlier one takes over. I believe that only something massively disruptive could displace bitcoin.

I'm not too worried about deflation - deflation in technology products is well known (PCs have been decreasing in price for years), and it doesn't stop people from buying them. The limited liquidity has made bitcoin particularly susceptible to momentum driven bubbles, and I look forward to higher volume and more derivative products to help stabilize the value. This will help adoption.

Enough for this thread - getting dangerously off topic!

eve

full member

Activity: 210

Merit: 100

Remember we don't want any pre order or IPO.

HashFast

full member

Activity: 154

Merit: 100

Quote from: thejestre on July 27, 2013, 07:05:36 PM

Quote from: namecoin on July 27, 2013, 04:15:11 PM

Quote from: conceptxp on July 24, 2013, 08:02:43 PM

can I come visit tomorrow? since I am 5 min from your location,

There should be quite a few folks here who are really interested in this in Bay Area. It would be interesting if someone could organize a group visit.

San Francisco Bay Area? I work just down the street from Stanford. I'd love to be part of a visit, especially when FastHash has a prototype ready and wants some users here to independently post pics or video of their unit kicking ass Wink

. Might help to quell suspicions and gather more interest closer to product launch.

So, when I buy a unit can I just pick it up from the FastHash office?

_theJestre
edit: I can't spell

If anyone is interested visiting our office and seeing what we're doing, hit me up privately. I will be coordinating several office visits during the next 2 weeks...

John
HashFast.com

googlemi

member

Activity: 152

Merit: 10

come on

cypherdoc

legendary

Activity: 1764

Merit: 1002

Quote from: thejestre on July 27, 2013, 07:05:36 PM

Quote from: namecoin on July 27, 2013, 04:15:11 PM

Quote from: conceptxp on July 24, 2013, 08:02:43 PM

can I come visit tomorrow? since I am 5 min from your location,

There should be quite a few folks here who are really interested in this in Bay Area. It would be interesting if someone could organize a group visit.

San Francisco Bay Area? I work just down the street from Stanford. I'd love to be part of a visit, especially when FastHash has a prototype ready and wants some users here to independently post pics or video of their unit kicking ass Wink

. Might help to quell suspicions and gather more interest closer to product launch.

So, when I buy a unit can I just pick it up from the FastHash office?

_theJestre
edit: I can't spell

count me in for a visit as well.

thejestre

member

Activity: 76

Merit: 10

Quote from: namecoin on July 27, 2013, 04:15:11 PM

Quote from: conceptxp on July 24, 2013, 08:02:43 PM

can I come visit tomorrow? since I am 5 min from your location,

There should be quite a few folks here who are really interested in this in Bay Area. It would be interesting if someone could organize a group visit.

San Francisco Bay Area? I work just down the street from Stanford. I'd love to be part of a visit, especially when FastHash has a prototype ready and wants some users here to independently post pics or video of their unit kicking ass Wink

. Might help to quell suspicions and gather more interest closer to product launch.

So, when I buy a unit can I just pick it up from the FastHash office?

_theJestre
edit: I can't spell

namecoin

jr. member

Activity: 38

Merit: 2

Quote from: conceptxp on July 24, 2013, 08:02:43 PM

can I come visit tomorrow? since I am 5 min from your location,

There should be quite a few folks here who are really interested in this in Bay Area. It would be interesting if someone could organize a group visit.

Ytterbium

full member

Activity: 238

Merit: 100

Quote from: Loredo on July 27, 2013, 01:05:22 PM

This is the second instance of this question being raised; the other instance being on the KnC thread. So that makes two times that well known, highly experienced - maybe, some might even say, leading edge - chip design firms have gone with "massive" chip topologies.

As I said before, I like it - I think it would be cool to see Bitcoin mining start to drive advances in IC tech, as people try to extract more performance and efficiency. With bitcoin, the improvements just from better ICs immediately give you an edge, while with other hardware you have all the other limiting factors: drives, memory, bandwith, etc plus the fact that software might not be able to take advantage of your hardware easily. With bitcoin, simply having a faster chip gives you an advantage.

It's definitely cool to see bitcoin ICs being designed in an ultra-high end way, comparable to top end CPUs and GPUs (or network switch chips I'd bet), instead tiny 110 and 130nm chips, equivalent to the kind of chips people use for low end ICs in toys and support chips like USB controllers, etc.

Quote

All I'm sayin' here - because I don't know the answers - is maybe we should recall what many experts said about IBM a few decades back. I'm paraphrasing here, but:

What I don't get is why anyone so experienced would build such as small and underpowered computer and suggest it has business uses. With small computers you cannot mount and read tapes or card decks, and have to rely on memory discs, which are small, slow, and very expensive. I can only really think of not having to cool the iron core with water as an advantage, but then your processor is so slow and can't process even a moderate sized COBOL program.

See what I'm sayin'?

Still though, don't forget the Carl Sagen quote, though: "But the fact that some geniuses were laughed at does not imply that all who are laughed at are geniuses. They laughed at Columbus, they laughed at Fulton, they laughed at the Wright Brothers. But they also laughed at Bozo the Clown."

All great successes have their doubters, but there are far more failures then successes.

Still there's every reason to think these guys will be able to deliver.

jmaccoin

newbie

Activity: 43

Merit: 0

Quote from: HashFast on July 27, 2013, 11:47:05 AM

Quote from: jmaccoin on July 27, 2013, 05:41:27 AM

So I googled fasthash. I am pretty sure I can handle that. Seems there is some water/ice cooling. But a drill for the power supply?

There was no talk of gigahashes at all, it was measured in grams?!?

Also, there was no talk of nanometers, it was all in microns. There was something about a 220, right down to 20 microns in one process?

And where do you input your wallet address? Then there was talk of a press?

This is going to knock bitcoin mining on its ass.

Hi JMA,

Not sure what site you landed on, but it's not our's. Our name is HashFast, not the reverse.
We will be rekeasing more info soon...

Please send me the link privately?

John

Sorry it was a joke, I meant no offence.
http://www.youtube.com/watch?v=GXp0BmLFsjw

Loredo

full member

Activity: 238

Merit: 100

Quote from: rograz on July 27, 2013, 06:04:56 AM

what I don't get is why anyone so experienced would build such a massive chip for something like btc mining. With big chips you get worse yields, usually lower clock potential and more thermal issues. I can only really think of getting slightly better performance/area from going big but then you might not gain anything due to smaller chips normally clocking higher/better yield.

This is the second instance of this question being raised; the other instance being on the KnC thread. So that makes two times that well known, highly experienced - maybe, some might even say, leading edge - chip design firms have gone with "massive" chip topologies.

All I'm sayin' here - because I don't know the answers - is maybe we should recall what many experts said about IBM a few decades back. I'm paraphrasing here, but:

What I don't get is why anyone so experienced would build such as small and underpowered computer and suggest it has business uses. With small computers you cannot mount and read tapes or card decks, and have to rely on memory discs, which are small, slow, and very expensive. I can only really think of not having to cool the iron core with water as an advantage, but then your processor is so slow and can't process even a moderate sized COBOL program.

See what I'm sayin'?

HashFast

full member

Activity: 154

Merit: 100

Quote from: Ytterbium on July 27, 2013, 12:14:36 PM

Quote from: ?? on ??

I'd be hard pressed to think that any ASIC design company would allow a chip to be manufactured that couldn't be cooled correctly...... That is their business. Leave this kind of stuff up to them, they know what they are doing.

Between the type of packaging and custom heatsinks even if it were 400W per chip I'm sure they figured it out.

Well, just look at Butterfly labs, lol Grin

But seriously - one thing that might be interesting to see done is have the chips simply run at higher temperatures. The greater the difference in temp between the heatsink and the air, the higher the rate of energy transfer to the air.

It seems like there are two limiting factors, first the operating temp of the IC itself, since the resistance changes as it gets hot, chips will stop working properly - but couldn't you design the chip to operate correctly at higher resistance levels? "Millitary grade" ICs are rated to 125°C. for example.

The other problem is simply the solder holding the chip on the board, lead free solder melts at around 200°C, apparently. But couldn't you design a package using a different material, like gold? If you look at modern socketed Intel chips they have gold pins on the board, and simple flat areas the pins connect too. The sockets are probably attached to the boards using BGA, of course but couldn't you put gold contacts on the board and attach the chip some other way?

One thing I think is cool about bitcoin is the fact that because the chip design is simple, it probably takes a lot less effort to actually design the IC itself, which can mean faster cycle times to innovate new technology more quickly then with other kinds of technology that gets made on the bleeding edge of IC tech, like CPUs and GPUs. (I think I've mentioned this a couple times before)

Quote from: HashFast on July 27, 2013, 11:47:05 AM

Not sure what site you landed on, but it's not our's. Our name is HashFast, not the reverse.
We will be rekeasing more info soon...

Please send me the link privately?

John

He's making a joke about the fact that if you google hashfast, you find marijuana related material Wink

LMAO!

Ytterbium

full member

Activity: 238

Merit: 100

Quote from: ?? on ??

I'd be hard pressed to think that any ASIC design company would allow a chip to be manufactured that couldn't be cooled correctly...... That is their business. Leave this kind of stuff up to them, they know what they are doing.

Between the type of packaging and custom heatsinks even if it were 400W per chip I'm sure they figured it out.

Well, just look at Butterfly labs, lol Grin

But seriously - one thing that might be interesting to see done is have the chips simply run at higher temperatures. The greater the difference in temp between the heatsink and the air, the higher the rate of energy transfer to the air.

It seems like there are two limiting factors, first the operating temp of the IC itself, since the resistance changes as it gets hot, chips will stop working properly - but couldn't you design the chip to operate correctly at higher resistance levels? "Millitary grade" ICs are rated to 125°C. for example.

The other problem is simply the solder holding the chip on the board, lead free solder melts at around 200°C, apparently. But couldn't you design a package using a different material, like gold? If you look at modern socketed Intel chips they have gold pins on the board, and simple flat areas the pins connect too. The sockets are probably attached to the boards using BGA, of course but couldn't you put gold contacts on the board and attach the chip some other way?

One thing I think is cool about bitcoin is the fact that because the chip design is simple, it probably takes a lot less effort to actually design the IC itself, which can mean faster cycle times to innovate new technology more quickly then with other kinds of technology that gets made on the bleeding edge of IC tech, like CPUs and GPUs. (I think I've mentioned this a couple times before)

Quote from: HashFast on July 27, 2013, 11:47:05 AM

Not sure what site you landed on, but it's not our's. Our name is HashFast, not the reverse.
We will be rekeasing more info soon...

Please send me the link privately?

John

He's making a joke about the fact that if you google hashfast, you find marijuana related material Wink

Topic: HashFast announces specs for new ASIC: 400GH/s - page 579. (Read 880461 times)