HashFast announces specs for new ASIC: 400GH/s

[dexX7]

Riddle me this: Can a bankrupted company give refunds? If no, than they have to deliver before Dec 31st => no refunds. If they manage to deliver batch #1 they will most probably deliver batch #2 also => no refunds.

This could buy more time to accumulate fresh, new funds.. but let's not get all too crazy. What's your reason to believe HF is unable to ship before Dec 31 anyway?

[1715141203]

1715141203

[1715141203]

1715141203

[1715141203]

1715141203

[RHA]

Cedivad, look at my history of posts - I'm in no way associated with HashFast (as a shill or a customer), yet I still believe Perozoso's  point is valid.

Another thing:

2) just after the delay, people realised that the delay will cost them 75%+ of their investment.

The delay is about three weeks (20-30 Oct to middle Nov). Let's say it is a whole month - you are going to lose mining in November. It would be around 12 BTC or less.
Is 12 BTC equal to 75%+ of the investment?
I estimate 400 GH/s machine to earn up to 35 BTC between Dec 2013 and the end of 2014, and then still earn more (not hosted, with 0.15 $/kWh electricity).
Even if my estimates are wrong, add MPP to that. You won't lose (at least in USD).

My conclusions: Don't use TGB calculator and don't exaggerate. This delay is nothing. More dangerous would be delaying two or three months, but then MPP or refunds would kick in.

[ImI]

Cedivad, look at my history of posts - I'm in no way associated with HashFast (as a shill or a customer), yet I still believe Perozoso's  point is valid.

Another thing:

2) just after the delay, people realised that the delay will cost them 75%+ of their investment.

The delay is about three weeks (20-30 Oct to middle Nov). Let's say it is a whole month - you are going to lose mining in November. It would be around 12 BTC or less.
Is 12 BTC equal to 75%+ of the investment?
I estimate 400 GH/s machine to earn up to 35 BTC between Dec 2013 and the end of 2014, and then still earn more (not hosted, with 0.15 $/kWh electricity).
Even if my estimates are wrong, add MPP to that. You won't lose (at least in USD).

My conclusions: Don't use TGB calculator and don't exaggerate. This delay is nothing. More dangerous would be delaying two or three months, but then MPP or refunds would kick in.

Also estimate rather 500 GH instead of 400. They already hinted that peformance will be higher.

[cedivad]

This delay is 3 difficult adjustments in the best case (the mid november deadline will be heavily missed imo, we have yet to see pics of everything). This is why, with this delay, i call these machines worthless. (1*1.5^3 more hashrate at the beginning). Your extrapolations are different, that's all.
I'm sorry that you had to begin your posts with a suggestion to look at your post history. Please accept my apologies.

[RHA]

Also estimate rather 500 GH instead of 400. They already hinted that performance will be higher.

I prefer to be pessimistic in estimates.

[Minor Miner]

TACOTIME;

Can you get me an answer to these questions on the sierra case?
1.  What are the case dimensions (l x w x h)?  I am NOT looking for "it's a standard 4U case", need dimensions.
2.   On the mounting flanges, what is the diameter of the holes and what is the center to center distance across the face?   I am assuming from the pictures that are now out that they are going with holes.

[Puppet]

This delay is 3 difficult adjustments in the best case (the mid november deadline will be heavily missed imo, we have yet to see pics of everything). This is why, with this delay, i call these machines worthless. (1*1.5^3 more hashrate at the beginning). Your extrapolations are different, that's all.
I'm sorry that you had to begin your posts with a suggestion to look at your post history. Please accept my apologies.

2 weeks is not 3 difficulty adjustments, its barely more than one. Difficulty adjustments are also not 50% (yet) and particularly the upcoming one is likely relatively modest; KnC is done shipping, the rest isnt ready yet, so you will only get some BFLs and Bitfury's being added to the network. Honestly, if a 2 week delay wasnt factored in to your purchase decision, you should never have  bought; Ive had shipments at customs for longer than that. And that would be a lot worse than when everyone is delayed 2 weeks.

ANyway, best I can tell, the FTC is on your side and you should have the right to get a refund.  I will support anyone trying to get it, but at the same time I find it hard to crucify hashfast over what is so far a very very modest delay, certainly when compared to BFL customers that have been waiting 6 or 12 months.

[cedivad]

>2 weeks is not 3 difficulty adjustments
From the 20th of december to the 15th (or 20th) of november, maybe even more than that.

>KnC is done shipping
As long as it's profitable to produce them, they will be produced. I don't see the hashrate increases to slow down anytime soon; we won't see exahashes next year, either.

>Honestly, if a 2 week delay wasnt factored in to your purchase decision, you should never have  bought
2 weeks is a good delay. The problem is that they will delay more than that (where are the pics of the PCBs?)

>I find it hard to crucify hashfast over what is so far a very very modest delay, certainly
They are laying to us and buying time. I don't see the future bright in any way.

(Email style quoting for short)

[Visesrion]

News !

https://hashfast.com/golden-nonce-interface-protocol-released/

[Puppet]

From the 20th of december to the 15th (or 20th) of november, maybe even more than that.

You mean from end of october to middle of november, no? Thats 2 weeks in my book. Even if its three, thats still no 3 difficulty adjustments and certainly not 50% ones.

As long as it's profitable to produce them, they will be produced.

No one every said KnC got out of the business, but they finished batch 1, batch 2 will only start shipping mid November, by their own account. Hence the next 2 weeks I dont expect spectacular increases in the network hashrate. As for what will happen after that, everyone would have told you it would go ballistic in December and January, but you cant blame HF for that. If anything their delay will also delay the ballistic trajectory.

2 weeks is a good delay. The problem is that they will delay more than that (where are the pics of the PCBs?)

I can have a PCB prototypes made in 2 days and volume production in a week. Anything can cause delays, but not seeing a PCB is not a worry, nor is seeing one any reason to breath easily (think BFL monarch PCB pics).

They are laying to us and buying time. I don't see the future bright in any way.

I see no evidence of lying at this point. Certainly not more than one might have "seen" months ago.

[cedivad]

News!
https://forums.butterflylabs.com/announcements/597-bitforce-sc-communication-protocol-draft.html

Attached is a copy of our draft of the communication protocol used for the SC line of products.
I look forward to developer feedback, please post any questions you may have.

Josh, 12-28-2012. (months before working chips)

[cedivad]

>You mean from end of october to middle of november, no?
Well, we will see that when they finally deliver.

>I can have a PCB prototypes made in 2 days and volume production in a week.
This makes me suppose that they don't have a draft of the PCB ready yet.

>I see no evidence of lying at this point.
You are right. It's not lying, but the next time i have an evident example, i will quote that for you. Anyway no lying, sorry for that.

[iCEBREAKER]

>Honestly, if a 2 week delay wasnt factored in to your purchase decision, you should never have  bought
2 weeks is a good delay. The problem is that they will delay more than that (where are the pics of the PCBs?)

>I find it hard to crucify hashfast over what is so far a very very modest delay, certainly
They are laying to us and buying time. I don't see the future bright in any way.

Did you have Dec 31 in your ToS?

I did.

I think we're done here!   

[dexX7]

That was fast. The Genesis Block also published an article about the interface protocol.

HashFash Announces Open Sourcing of Bitcoin Mining ASIC Interface Protocol

Today HashFast announced they are open sourcing the interface protocol used for controlling their bitcoin mining ASICs. Open sourcing this design gives significant insight into how their bitcoin miners operate and provides a platform for driver developers. More users collaborating in driver development should allow for a more diverse set of options, and a better […]

http://thegenesisblock.com/hashfash-announces-open-sourcing-bitcoin-mining-asic-interface-protocol/

[cedivad]

Oh god, here, too.
I offered you something that you will trow around for months, didn't i?

[cedivad]

Anyway i would have never said it, but that pdf (https://hashfast.com/wp-content/uploads/2013/10/gn_protocol.pdf) is really good.
Finally some proofs that they have been working on something. But this good feeling is still overcomed by the incredibly similarity with the BFL paper released months before anything... BFL's paper was poorer, at least.

[iCEBREAKER]

Anyway i would have never said it, but that pdf (https://hashfast.com/wp-content/uploads/2013/10/gn_protocol.pdf) is really good.
Finally some proofs that they have been working on something. But this good feeling is still overcomed by the incredibly similarity with the BFL paper released months before anything... BFL's paper was poorer, at least.

Glad to see you're feeling better, coming out of the crazy zone, and joining us in the real world.



Here's to better luck with internet lawyering in the future!  Cheers!

BTW, the Global Work Queue described in the protocol document is awesome!

This is just a preview of the high quality of innovative technology HashFast has been busy developing, while ActiveMining played with logos and mucks about with facilitating the trade of unregistered securities on illicit exchanges.

[gateway]

damn someone beat me to the punch..

anyhow..

The HF Golden Nonce interface protocol is out and ready for your reading enjoyment! https://hashfast.com/golden-nonce-interface-protocol-released/

[Ytterbium]

>Honestly, if a 2 week delay wasnt factored in to your purchase decision, you should never have  bought
2 weeks is a good delay. The problem is that they will delay more than that (where are the pics of the PCBs?)

>I find it hard to crucify hashfast over what is so far a very very modest delay, certainly
They are laying to us and buying time. I don't see the future bright in any way.

Did you have Dec 31 in your ToS?

I did.

I think we're done here!  

Except for the whole "US law requiring a refund" thing.  Are you retarded or something?  Illegal terms of sale aren't enforceable.

It doesn't matter if HF delivers in the middle of November like they say, you'll never ROI. It's completely worth it to try for a refund before they deliver. Unless you like burning your money.

[ninjarobot]

The protocol doc is pretty good indeed.

So one HF ASIC actually has 4 dies and will come out of the box @ 422.4 GH/sec (@550MHz), and max overclockable to 537.6 GH/sec (@700 Mhz).

While many will be eager to overclock from the get-go to make up for lost hashing-time, please take note that this will likely void your warranty:

Modules which are over-clocked will contain logs of such operation, which may void warranty.

Since the KnC 1st batch units seem to be failing all over the place, I for one will probably run my HF unit at the nominal speed for the first week (or two) and monitor the threads for HF units failing before even considering overclocking.