Pages:
Author

Topic: CoinTerra announces its first ASIC - Hash-Rate greater than 500 GH/s - page 83. (Read 231002 times)

donator
Activity: 1218
Merit: 1079
Gerald Davis
How's the power consumption?

At the moment, I don't have anything to measure it, I will come back to you.
But it can't be much, the whole setup doesn't even get lukewarm Smiley




Somewhat off topic but 517 GH/s?  I thought it was more like 400 GH/s stock.  Did you modify the boards for overclocking?
legendary
Activity: 2156
Merit: 1072
Crypto is the separation of Power and State.
large investors of cointerra, I welcome to contact me privately

What large investors?  They don't have any.

CryptX is the closest thing to one, but their IPO has barely raised a couple million.

The market was simply tapped out and IPO fatigued by the time Cointerra arrived late to the game.

There's only so much room for firms competing to be The Next ASICminerTM, and Cointerra didn't find a seat when the music stopped.
hero member
Activity: 702
Merit: 500
large investors of cointerra, I welcome to contact me privately

thats a bit cryptic...  i think you'd have more chance of them contacting you if they knew why

and this forum may not be the place to find them... 
hero member
Activity: 588
Merit: 504
large investors of cointerra, I welcome to contact me privately
legendary
Activity: 1904
Merit: 1007
Ok my mistake then!
legendary
Activity: 1904
Merit: 1007
I'm curious... I've heard people say that the KnC chips are just fpga copies.. but I've not heard it from any legitimate nor direct sources.  Do you have a source for that info?   Its often said by hashfast as one of the FUDs that they spread, but i really haven't seen it directly mentioned anywhere.

And anyway, if they've got them running at less than 1 Watt per GH, at the wall.. thats pretty acceptable to me.  Its not as low power as hf or ct, but then, they're nit shipping yet... so the difference is moot.

I don't have a source for that.  KnC may be standard cell, but getting the same efficiency at 28nm as Bitfury got at 55nm could be responsible for the popular 'FPGA copy' assumption.

Anyway, this is a Cointerra thread so let's get back to speculating about the reasons for their failed tape-out and (possibly related) decision to hire new people for their next chip.   Cool

Are you really that bad intended? Bitfury is ~2W/GH while KNC is ~1W/GH-~1.3W/GH(reports vary)

Bitfury, 400 GH system (16x H-boards) uses a lot less than 800W at the wall it is closer to 400W.  Some boards which overclock and overvolt the chips have lower efficiency although I haven't seen one that is as bad at 2 W/GH.  Not sure where you are getting your info.

I remember that i read some posts in the US Bitfury thread. Too lazy to search for them. I can admit that i was wrong, no problem there. Waiting for someone else do the dirty job of searching posts.

Edit: Or maybe it was some cryptx thread. I just knew those numbers. Any proof is welcomed.
member
Activity: 83
Merit: 10
Ok, awesome "I did a science!" graphic. That made me laugh.

As a Cointerra customer, I'm pissed about their 30 day warranty. I was really galled by their announcement that they'd extend it to 60 days if I bought hosting from them. A 1 month warranty is a joke. It basically says, "We're not sure if this is going to be working in 2 months, but if its not, that's your problem."

What if they release a firmware update on day 31 that bricks my rig? What if their water block cooling springs a leak? Maybe you don't think that these particular scenarios are likely, but the experienced among us know that things can break in ways we can't predict. This policy passes increased risk to individual miners. In this ASIC crapshoot, we're already putting enough money at risk.

As a contrast, my KnC rig is backed by a 12 MONTH warranty. If it frys after 1 year, I've already made all the coins its going to make. That's how you treat your customers, in my opinion. If Cointerra doesn't change this policy, guess who's getting my 2nd generation ASIC order?

If Cointerra is confident that they're building a good product, then they should stand behind it.
donator
Activity: 1218
Merit: 1079
Gerald Davis
I'm curious... I've heard people say that the KnC chips are just fpga copies.. but I've not heard it from any legitimate nor direct sources.  Do you have a source for that info?   Its often said by hashfast as one of the FUDs that they spread, but i really haven't seen it directly mentioned anywhere.

And anyway, if they've got them running at less than 1 Watt per GH, at the wall.. thats pretty acceptable to me.  Its not as low power as hf or ct, but then, they're nit shipping yet... so the difference is moot.

I don't have a source for that.  KnC may be standard cell, but getting the same efficiency at 28nm as Bitfury got at 55nm could be responsible for the popular 'FPGA copy' assumption.

Anyway, this is a Cointerra thread so let's get back to speculating about the reasons for their failed tape-out and (possibly related) decision to hire new people for their next chip.   Cool

Are you really that bad intended? Bitfury is ~2W/GH while KNC is ~1W/GH-~1.3W/GH(reports vary)

Bitfury, 400 GH system (16x H-boards) uses a lot less than 800W at the wall it is closer to 400W.  Some boards which overclock and overvolt the chips have lower efficiency although I haven't seen one that is as bad at 2 W/GH.  Not sure where you are getting your info.
legendary
Activity: 1904
Merit: 1007
I'm curious... I've heard people say that the KnC chips are just fpga copies.. but I've not heard it from any legitimate nor direct sources.  Do you have a source for that info?   Its often said by hashfast as one of the FUDs that they spread, but i really haven't seen it directly mentioned anywhere.

And anyway, if they've got them running at less than 1 Watt per GH, at the wall.. thats pretty acceptable to me.  Its not as low power as hf or ct, but then, they're nit shipping yet... so the difference is moot.

I don't have a source for that.  KnC may be standard cell, but getting the same efficiency at 28nm as Bitfury got at 55nm could be responsible for the popular 'FPGA copy' assumption.

Anyway, this is a Cointerra thread so let's get back to speculating about the reasons for their failed tape-out and (possibly related) decision to hire new people for their next chip.   Cool

Are you really that bad intended? Bitfury is ~2W/GH while KNC is ~1W/GH-~1.3W/GH(reports vary)

Edit: Is iCEBREAKER the cypherdoc of Cointerra?
legendary
Activity: 2156
Merit: 1072
Crypto is the separation of Power and State.
Posting a picture of HashFast's CEO holding a box of their finished wafers in poor lagging Cointerra's thread is just plain mean.

From my point of view, this could as well be HashFarts CEO getting out the garbage.

Then your point of view is woefully under-informed.

Had you bothered to pay attention, you would know HashFast taped-out to TSMC in August.

Fact: HashFast now has chips in-hand, but Cointerra won't for at least 2 months.

FUD: your POV.
legendary
Activity: 974
Merit: 1000
Posting a picture of HashFast's CEO holding a box of their finished wafers in poor lagging Cointerra's thread is just plain mean.

From my point of view, this could as well be HashFarts CEO getting out the garbage.
sr. member
Activity: 462
Merit: 250
What you mean?   All systems go on the Cointerra forum, just ask Rich!

http://forum.cointerra.com/threads/cant-wait-to-get-my-rig.48/

http://forum.cointerra.com/threads/howdy.6/#post-105


"Oh yeah. The support team was excellent to me, as well. And I feel ya'ACN... The creds of the staff and team here are above and beyond! After comparing machine numbers, once I saw the backgrounds of the Cointerra Team, I KNEW that this was where my money was going. "
legendary
Activity: 2156
Merit: 1072
Crypto is the separation of Power and State.


Posting a picture of HashFast's CEO holding a box of their finished wafers in poor lagging Cointerra's thread is just plain mean.

Have you no pity for Cointerra's ASIC design team, even as they are being replaced with more competent engineers?   Undecided
legendary
Activity: 974
Merit: 1000
If you got your imaginary HF miner in Jan you save $100 each month but down $30,500 for being a derp (over 25 years....)

I'm not getting my currently-under-construction HF miner, which is based on chips that have been delivered to San Jose, "in Jan."

If HF doesn't deliver by Dec 31, I get a refund.

HashFast gets 300% more hashing power per 28nm wafer than KnC.  That gives them great flexibility WRT MPP and delay compensation.

My dear iCEBREAKER,

I am proud to announce that inside this box, I hold your long awaited mining-unit.



Sadly, it has been run over by a truck and now is broken beyond repair (see left side). But don't fear, our MPP (Miner Protection Plan, aka MyPePe) will save you from any further chagrin, and we will REFUND you the original price you paid for this unit 4 times, which equates to $10 at the current $/BTC exchange rate.)

Thank you for your business and please come back every time,
Yours sincerely,
Eduardo Castrato

P.S.: About this refund thing, you obviously got something wrong, these where custom made stuff, additionally, money ran out, sorry.
legendary
Activity: 2156
Merit: 1072
Crypto is the separation of Power and State.
HashFast gets 300% more hashing power per 28nm wafer than KnC.  That gives them great flexibility WRT MPP and delay compensation.

For the moment HashFast have 0% more hashing power than KNC.
No miner = no hash power
End of line.

Sorry, I'll use the future tense next time for the sake of you easily confused types.

This moment won't last forever.  Some of like to look down the road, instead of staring at rear-view mirror.

End of line?  Hardly.   Roll Eyes

The ASIC wars have barely begun!   Cool
hero member
Activity: 504
Merit: 500
HashFast gets 300% more hashing power per 28nm wafer than KnC.  That gives them great flexibility WRT MPP and delay compensation.

For the moment HashFast have 0% more hashing power than KNC.
No miner = no hash power
End of line.
legendary
Activity: 2156
Merit: 1072
Crypto is the separation of Power and State.
If you got your imaginary HF miner in Jan you save $100 each month but down $30,500 for being a derp (over 25 years....)

I'm not getting my currently-under-construction HF miner, which is based on chips that have been delivered to San Jose, "in Jan."

If HF doesn't deliver by Dec 31, I get a refund.

HashFast gets 300% more hashing power per 28nm wafer than KnC.  That gives them great flexibility WRT MPP and delay compensation.
sr. member
Activity: 462
Merit: 250

I know math is hard for you and you don't understand what an "FPGA copy" is, much less why HashFast's chip is more advanced.

Just forget about all the confusing things going on under the hood and focus solely on the output, like a good consumer.  Leave the icky details to us hardware nerds and you'll do fine!


You brought math up in your defense??  Let's see how well that works out for you.


KNC 1Th example using about 1W/Gh (rough numbers to prove the point)

Oct Output:  50 BTC (use a low $300 per BTC) for $15,000, less the $150 power costs.  (1% of revenue)
Nov Output:  30 BTC for $9,000 less the $150 power costs  (1.7%)
Dec Output:  15 BTC for $4,500 less the $150 power costs  (3.3%)


So if you got your wonderful 1/3 less power costs in October you save $100 per Th. which is less than 1% for that month

But of course you didn't..If you got your wonderful HF miner in Nov you save $100 per Th per month but down $15,000 for being late to the party (over 12 years to make up difference @ $100 month savings)

If you got your wonderful HF miner in Dec you save $100 each month but down $26,000 for being late (over 21 years to recoup)

If you got your imaginary HF miner in Jan you save $100 each month but down $30,500 for being a derp (over 25 years....)







legendary
Activity: 2156
Merit: 1072
Crypto is the separation of Power and State.
I'm curious... I've heard people say that the KnC chips are just fpga copies.. but I've not heard it from any legitimate nor direct sources.  Do you have a source for that info?   Its often said by hashfast as one of the FUDs that they spread, but i really haven't seen it directly mentioned anywhere.

And anyway, if they've got them running at less than 1 Watt per GH, at the wall.. thats pretty acceptable to me.  Its not as low power as hf or ct, but then, they're nit shipping yet... so the difference is moot.

I don't have a source for that.  KnC may be standard cell, but getting the same efficiency at 28nm as Bitfury got at 55nm could be responsible for the popular 'FPGA copy' assumption.

Anyway, this is a Cointerra thread so let's get back to speculating about the reasons for their failed tape-out and (possibly related) decision to hire new people for their next chip.   Cool
sr. member
Activity: 378
Merit: 250
what is making me curious, who were all those people who purchased first CoinTerras 16000 usd per 2trhs. If those 16K invested in btc, back then it is now usd 66000. That way I can order now 11 CoinTerras 2trhs with delivery at February for same amount of bitcoins. Seems like 22 trhs is better then 2trhs. Maybe waiting make even more sense now too? I will buy Cointera's miners at some point, almost certain about it, but question is when, not sure about that.
Pages:
Jump to: