Author

Topic: HashFast announces specs for new ASIC: 400GH/s - page 374. (Read 880816 times)

RHA
sr. member
Activity: 392
Merit: 250
They have got two dies to achieve 250 GH/s using 300 W.
It seems their PR simply multiplied the speed by two to make great announcement but forgot to think about power requirements.
Now the technicians are still trying to find a way to supply 600 W (for which two PCI-e could be enough, yet near the limit, depending on the PSU used)
and dissipate it (which can be quite hard as the cooling system is, as I remember, projected for less power),
and the chip has to withstand all this power going through it.
newbie
Activity: 28
Merit: 0
I just realized this, 6pin PCI-e connectors are only rated for 75W, and you could probably get MAX 200W out of them.
In this blog post: https://hashfast.com/second-tests/ HF says the power consumption for just 2 dies is 300W. Maybe this is why they haven't said anything about running all 4 dies. They are probably running into limitations on how much power they can push through those 2 connectors.
sr. member
Activity: 462
Merit: 250
BJ upgrade price gone up to 1900.00 usd today?

Looks like it. Probably because they realized their pricing scheme made no sense. The BJ with upgrade was cheaper per GH than the Sierra.

They're a bunch of baffoooooonss
legendary
Activity: 1904
Merit: 1007
BJ upgrade price gone up to 1900.00 usd today?


So instead of tweaking their chips they adjust prices? Go go marketing team! First chips received 10 days ago and still not a single proof of them hashing.
newbie
Activity: 28
Merit: 0
BJ upgrade price gone up to 1900.00 usd today?

Looks like it. Probably because they realized their pricing scheme made no sense. The BJ with upgrade was cheaper per GH than the Sierra.
sr. member
Activity: 420
Merit: 250
BJ upgrade price gone up to 1900.00 usd today?
donator
Activity: 1218
Merit: 1080
Gerald Davis
Maybe because calling something a single chip when it is four different chips on the same board is at best dishonest. Four knc chips are producing 700gh/s, they just keep them on separate boards so that cooling is easy and single chip failures can be handled without losing the entire board.

It is 4 dies operating as a single chip.   KNC uses the same method BTW so it takes 16 dies for a KNC (Nov) Jupiter to produce ~650 GH/s.
legendary
Activity: 1904
Merit: 1007
Still no picture of 500GH/s speed?

It's all over the internet.  Don't you read Coindesk?

400 BC called.  Gorgias wants his solipsism back.   Roll Eyes

I don't read Coindesk. Did a search there and found nothing. Want to point me to where exactly is the proof?
legendary
Activity: 1176
Merit: 1001
ToS says that we have to send them a letter by the 15th of Jan if we don't receive the miner, yes, so?
legendary
Activity: 1890
Merit: 1003
Irony, it now seems you all are seemingly disappointed IF you recieve your miner BEFORE New Years....

Don't the Terms of Agreement/Service state you have to notify HashFast with some letter that you have not recieved your order "within a reasonable period of time"?
legendary
Activity: 1176
Merit: 1001
You are right about customs, it could be a problem, but the packet size doesen't really make any difference for DHL, and they are able to sustain the load they have during Christmas. So i hope all the best for you, but don't be too disappointed when you receive the miner before the end of the year.

(thinking about it, DHL should handle customs directly, (your package won't be sitting in customs, there will be a custom employe on the plane and they will do their checks there)).
legendary
Activity: 1120
Merit: 1002
TIL I learned Coindesk is a reputable source for ASIC hardware news.
I've learned it the hard way as well.

example: i leave in france , and if they dont shipped tomorow ( monday) : IT'S IMPOSSIBLE TO RECEIVE THE MACHINES IN TIME :  then REFUND !!!

cedivad , we dont talk about only one  smalls packets ... , and  having to anticipate customs issues , and we are in a critical period too ( christmas),  Wink
legendary
Activity: 1176
Merit: 1001
TIL I learned Coindesk is a reputable source for ASIC hardware news.
I've learned it the hard way as well.

example: i leave in france , and if they dont shipped tomorow ( monday) : IT'S IMPOSSIBLE TO RECEIVE THE MACHINES IN TIME :  then REFUND !!!
legendary
Activity: 1120
Merit: 1002

That should be deliver "deadline *".  The "deadline" (until they decide to retroactively change it after the fact) is based on delivery date not shipping date.  

* Of course the deadline for pre Aug 15th orders has long since passed, the company however has fraudulently decided to steal customers funds by not providing refund in violation of their own pre Aug 15th TOS.

absolutly !! this what i have in the footer of my oficial invoice ( after payement) Grin

then they have to deliver before 31 dec, not shipping before 31 ! that's make a difference for those who want their money back !

example: i leave in france , and if they dont shipped tomorow ( monday) : IT'S IMPOSSIBLE TO RECEIVE THE MACHINES IN TIME :  then REFUND !!!
sr. member
Activity: 308
Merit: 251
The realist
Well it's mid December. Where are the miners or any word when they'll start shipping?

And I wonder how Hashfast_CL is going with that Eligius stats request. I'm really interested to see a BJ's performance  Cool

Quote
Even so, the latest delivery estimates for the modified, improved substrate design are three weeks from today. This means that we would be shipping our first rigs mid-December.
legendary
Activity: 1652
Merit: 1067
Christian Antkow
Still no picture of 500GH/s speed?
It's all over the internet.  Don't you read Coindesk?

TIL I learned Coindesk is a reputable source for ASIC hardware news.
legendary
Activity: 2492
Merit: 1491
LEALANA Bitcoin Grim Reaper

That should be deliver "deadline *".  The "deadline" (until they decide to retroactively change it after the fact) is based on delivery date not shipping date.  

* Of course the deadline for pre Aug 15th orders has long since passed, the company however has fraudulently decided to steal customers funds by not providing refund in violation of their own pre Aug 15th TOS.

I haven't been following this that closely but if D&T claims it I'd say it's worth considering.

If it is true...shame on you HashFast.
legendary
Activity: 1890
Merit: 1003
Some are saying that if you ordered later with the direct wallet (not the bitpay ordering that was only available on day 1), you might be able to get your original BTC back. 

Personally I like mining, though I will almost definitely not recover the 60 BTC spent on day 1 (even with timely MPP). 
And yes... you would be an idiot to ask for a refund if all you are going to get back is a locked in 90% loss.

Bingo, you got it.
legendary
Activity: 1890
Merit: 1003


Then you are in a better position then most of us.. because since the TOS clearly states that it must be delivered by dec31, then you will be able to  request a refund on Jan 1st.  AND  IF you paid btc directly to HF's wallet, then you will be able to get your btc back.

question is, if they come to your door with the equipment on Jan 2nd, will you accept delivery?


That is an important (very important) point to keep in mind. If you sent a BTC payment to their wallet and not through a payment processor that converts the money automatically to fiat...then you are deserving of every last BTC as originally paid.

You paid 60 BTC to their bitcoin wallet (sans any payment processor) you should get 60 BTC back. No reason for them to claim otherwise.
donator
Activity: 1218
Merit: 1080
Gerald Davis

That should be deliver "deadline *".  The "deadline" (until they decide to retroactively change it after the fact) is based on delivery date not shipping date.  

* Of course the deadline for pre Aug 15th orders has long since passed, the company however has fraudulently decided to steal customers funds by not providing refund in violation of their own pre Aug 15th TOS.
Jump to: