Topic: CoinTerra announces its first ASIC - Hash-Rate greater than 500 GH/s - page 4. (Read 231002 times)

Regardless of there almost non-existent customer service and joke of an ordering system....they did answer the phones though, they did ship systems. they also have avoided bankruptcy so far.

Their systems run albeit under spec and most systems got delivered as much as I can tell.

I own both HF and CT and the ratings between the 2 companies are off.  CT should go up some in comparison.

both have 1/10 for on time delivery.....the rating is not binary(did or did not)....it looks like a scale. the rating should reflect it.

both have 10/10 for delivered miners....CT seems to have delivered theirs. HF is in involuntary bankruptcy because of the thousands they did not deliver.

both have 1/20 for uses preorders....again a scale.




As much as CT annoys me... either have a guide that cuts out some of the subjectivity and tries to true up the ratings or don't have one at all....
 

they've got 4 chips in a box that they've sold thousands of.  for them to do one chip on a card wouldve been easy peasy.   but i suspect they tested the market by pre-announcing it, to decide whether to make it, and the market told them they didn't want it... so they didn't go ahead with it.

What ever happened to Cointerras mining card? Guess they couldn't get that working either.

how many asic companies are there that never took pre-orders?  I only know of a couple and AsicMiner is the main one, right?  And they were in early, when it was low hanging fruit and low barrier to entry.   almost everyone else that came later took pre-orders, or sold their chips to system integrators/assemblers that DID take pre-orders.

whereas... 21e6 - the private mine with silicon valley investors' backing - didnt take pre-orders.. they took investors.  And surprise surprise, theyre a super secret private mine.

True. Cointerra does seem to be one of the more fortunate preorders.

I just assumed they were short of cash due to the denial of refunds/compensation although this could have changed since I've stopped following this thread a long time ago.


IMO if you don't have the money or can't get investors then you shouldn't be making a multimillion dollar asic.

The problem with preorders is that they require investor level risks for customer level rewards.

btw.. dont lump HF and CT together as bankrupt.. thats just silly.   they're completely different.  HF *is* facing bankruptcy.. actually, I've not kept up to date, it may already be bankrupt or in liquidation as we speak.  HF had huge numbers of orders, and barely shipped any of them (lucky me, i was one of the few customers that received their order).   By contrast, Cointerra also had large numbers of orders but DID ship them.  Sure, they earned a bad rap by shipping a system that was 20% off its expected performance.. but they still had the fastest and cheapest bitcoin miner available for the time that it was launched (early 2014) and did in fact sell - and *ship* thousands of them.  I suspect a major chunk of today's bitcoin network is currently powered by Cointerra devices (guessing circa 10%).  After BitFury, KnCMiner and Bitmain i think they're the next in popularity (measured in PH).

i hear you!   its unfortunately a feature of the pre-order business model that companies have to state their specs in advance in order for anyone to place an order.  And they have to appear rock solid and uber confident of their specs or people won't place an order.  If they actually said that they weren't 100% confident what the final performance would be, til the chips arrive... which is the reality... then they wouldn't get any orders, so its commercial suicide for them to be anything other than bullish about their performance specs.

its easy to say that these companies should be selling from stock... but the huge multi-millions of dollars that it takes to make a chip preclude people from doing that.  the only reason asicminer and the other early guys could afford to was simply that they were early... AND, on an old process node thats not expensive (130nm, 110nm, 65nm, 40nm.. these are cheap NREs compared to 28nm or smaller geometries).  the other guys are on newer, smaller process nodes that are much more expensive NRE (millions of dollars) to start.. and none can afford to 'fabricate' their chip without pre-orders... OR...the only other way is to raise money from... Investors.  If they have investors you can be damn sure they aren't going to sell their hardware to customers and the investors will prefer they start a private mine instead.. which is much more cost effective and profitable than selling to customers.  Thus the only way customers will get to buy miners are from the companies that haven't been funded by investors (aka, Pre-orders)

-- Jez

First sentence in the first post of that thread:


Regardless you are completely missing the point.

The reason AM/bitfury are not penalized for inaccurate estimations is because they didn't sell preorders based off those estimations.

Spondoolies/hashfast/blackarrow/bfl/cointerra repeatedly told us that their specs/dates were rock solid and that there is practically no chance their estimations are off and they took peoples money based off that.

Had any of those companies only sold hardware from stock then nobody would have been ripped off and I'd bet that CT/HF wouldn't be facing bankruptcy if that was the case.

ok, so maybe promised is the wrong word.   All power and performance specs that are stated in advance are an estimation based on simulation using asic design cad tools and data provided by the fab.

whether it was promised, or estimated isn't of issue.  Just as much as any other asic company can predict in advance its performance and power consumption, so can asicminer, and to be honest, so did everyone else.  we shouldn't be holding asicminer to any standard thats different from any other asic company.  they suffered pretty much the same fate that others have done before them and will no doubt do after them.

read the spec that asicminer posted on their first post... it doesn't say estimated anywhere.  it doesn't leave much room for doubt.   it says what they expected the spec to be, and the reality, when the chips arrived back, was far from the expectation - perhaps as much as DOUBLE the power consumption of what they expected.  Not, that any other asic company got it any more right.  BitFury expected their chip to hash at 5 GH and even labelled it on the top and the chip can do approx 2.5 GH.  Cointerra expected their system to be 2 TH (using four 500 GH chips) and it ended up 1.6 TH (four 400 GH chips).  Spondoolies has also pre-warned us a few days ago that we should expect their spec will not fully live up to their expectations, etc.  in short, its not uncommon for asic companies to say what their expected performance and power requirements are, in advance - with their best will and best intentions - and when the reality sets in (after the chips arrive) then they have to announce what it actually is, rather than what it was predicted to be, and there's often a disconnect.

anyway, the first post in their thread shows what asicminer expected their perf/power of their chip to be...

https://bitcointalksearch.org/topic/annasicminer-publicly-looking-for-potential-customerspartners-for-new-chips-438359

Please show me a single person who complained about receiving used hardware.


How many asics were less than 10w/gh before june?

Do the specs really matter when every single person who bought from AM turned a profit in USD and many even in BTC?


Please stop saying it was promised because that's just not true. It was an estimation and as we all know estimations are wrong.

and, how's business?

i didn't make it up, but you're totally right.. people don't seem to complain when they've bought inefficient old chips, as new.  I've seen people in these forums praising AM to the stars for shipping 'excellent' products that consumed for some reason significantly more power than any other system available by a large margin, and calling them rock solid and fantastic systems.   even those that consumed multiple watts per gigahash.   but you're absolutely right, people seem happy with buying AM's old stock 'as new'.  that all changed with the current AM chip which is being sold as 'fresh' and seems to be a new business model for AM however even that one was considerably out of spec when it was released.  (promised to be 0.35 watts/GH in the spec, and ended up consuming a lot more).

No. AM missed estimated specs. AM doesn't sell preorders and doesn't guarantee estimated specs. Investors simply asked FC to gaze in to his magic crystal ball and he happened to be wrong just like every single manufacturer to date. Nobody was ripped off or scammed. Only a bunch of disappointed investors.


So you just made up a hypothetical situation and expect AM to get penalized for it?

Please show me a single person complaining about late shipping or used hardware.

because they still are actively selling systems from their web store, from stock.

Agreed, Cointerra shipped a large number of systems in the month they were sposed to be shipped and were relatively on time.  To give them a score of 1/10 for 'on-time' delivery - the same score that Hashfast, ButterFlyLabs, BitMine, and BlackArrow also got  - when they delivered much much later - if at all - makes it hard to agree with the scoring system (the methodology is fine, but the actual scoring sseems to be totally at the whim of Dogie and a vocal minority).  Cointerra certainly doesnt deserve 10/10 for timeliness, but since majority of the (thousands of) systems they shipped were pretty much delivered on-time, i think it would be hard to justify anything other than a 7 or 8 out of 10 on that score.

its a somewhat selectively accurate chart that has downgraded some companies more than others for making identical mistakes.   for instance, AsicMiner's latest chip ended up much worse than double its power consumption from the announced spec to the delivery, yet that hasnt been recognised.  I also object that AsicMiner in the past has sold off old stock and perhaps even used systems that were under performers and over power consumers, and yet theyve been classed as shipping on time, when in reality they were simply upgrading their own private mine, and selling off the old boards in new boxes.. and yet that was interpreted as 'can do no wrong' by the asicminer fanboys.

why even rate them anymore?

your rating for them does not make sense.   they delivered to me, yes late but less late than avalon and in line with when KNC delivered to me their "in stock" jupiters that took three weeks to arrive because they were not assembled.
they also have done RMAs and have responded to problems etc.   They delivered everyone's order also.   
I do not think they deserve to be right up beside spondoolies but they certainly do not deserve to be where you have ranked them.

This company's rating has changed in the relaunched manufacturer trustworthiness thread.

Thanks for this, I'm trying to modify my cgminer start file using PuTTY and this gets me about halfway there.
Could you please tell me how I can get to the file?
Thanks in advance, J

I hear crickets.