Topic: Swedish ASIC miner company kncminer.com - page 1584. (Read 3049505 times)

(...)

Where is the photo of the chips, I can't find it?

(...)

Not trolling just disappointed in their business practice when they say they or someone says they are supposedly offering refunds up until shipping, and I actually asked for the refund before they decide to post the shit on their site that they are in full production and I confront them here in the forum with a pic of their e-mail reply. and I have other emails asking about vat info and sales conformation on my phone. Sad part is im needing it because of a family matter.

We did see a photo of the chips. And if they do arrive on Sunday wherever they have to be to put some into a prototype maybe they will photograph them, maybe even hashing in the prototype.

Opening them up on the airplane in transit might not be all that cool an idea, afterall...

Heck maybe they'll arrive friday night, or saturday, allowing many hours to get that prototype ready by sunday ... while production units spew out of conveyor belts...

Gosh they might even arrive tonight! If so why would it take all the way to sunday to get a prototype ready?

-MarkM-

So why haven't we seen a photo of the chips yet? If they haven't arrived yet, then it's unlikely cloudhashing will be doing much on Tuesday.

The prototype seems to be ready on 29th (Sunday)
https://bitcointalksearch.org/topic/m.3240278

they might want to show one to Bitcoinorama as well to chill down everyone and on Monday 30th the first assembled miners are being shipped.

When cloudhashing then receives units on next day, Oct. 1st,  puts them immediately online (pre-configured)  during the morning, the count down on their website would fit into the picture.

maybe byteminr http://eepurl.com/FRl6f has no day1/2 orders, therefore he is a little bit later in business ?

Correct, they could still 'deliver' this month.

I think there have been valid concerns voiced in this thread lately.
Yes there are the obvious BS trolls, par for course.

It comes down to what can be viewed as KnC moving the goalposts.

Let us review the facts we have:

KnC
on: June 26, 2013, 02:14:10 PM
"I would like to say that EVERYTHING that needs to be done at this point in time for a September delivery has indeed been done.
We are on track for delivery and we know of nothing currently to the contrary."

this becomes

KnC
on: July 30, 2013, 01:41:06 PM
"We are still on schedule to begin shipping in September."

'delivery' becomes 'shipping'

Sam/KnC
on: May 30
"We see no reason why it cant work like that, the only addition I will add is that the refund will be 100% with the order cancelled, We wont be offering a discount/refund and still shipping you the boxes. Refunds are offered up to the day of shipment of your order.

Above all else we will be fair.
Sam "

Today this becomes

KnC
"Therefore, any changes of orders cannot be accepted."

It is speculated that this includes changes of the refund variety.

Today at 01:39:25 PM
Kaliecious starts a thread
https://bitcointalksearch.org/topic/m.3239687

"KNC miner is refusing refunds who needs to be contacted to inform they are not adhearing to the laws of it hasn't shipped they need to give a refund?"

Kaliecious is asked to provide evidence and posts this pic.



I have not personally recieved confirmation from KnC that this is a fact.
(I have emailed asking for confirmation)
If the thread and pic are a troll it is a clever one and perfectly timed.
There was a well timed malicious troll (since edited from thread I think)

While this is being discussed over 150 posts vanish from the thread.
Coinkydink? Could be!

I understand and share some customers concern that similar behavior observed in other vendors has not ended well for customers or schedules.

I think it unlikely that all who ordered day 1 & 2 based on originaly stated 'delivery' in September will recieve units in September. It is possible, it is only the 26th.

The conditions of 'the bet' are not the conditions pre-orders were solicited under.

People who ordered after the day 1 & 2 ALSO did so based on the original terms.
That includes refunds up to day before shipping.
KnC stated the original terms.
KnC took the risk and stood to reap the rewards.

If I have misstated any facts let me know and I will edit and explain my edits as I always do.
If I am flat out wrong I will eat my crow when it is served.

For the record I have not decided that I _DO_ want a refund.
I do expect that option to remain until shipping as per the origional terms of the agreement.

YMMV

you mean they will pay an extra to get "the factory" to work over the weekend? it is not China, it is Sweden.

And Saturday and Sunday, or do you live in a parallel dimension that doesn't have weekends? They worked weekends building the data centre, you can be sure they also will for production.

The big news is that all the production orders that are due for delivery before October 15th have now been confirmed with the factory for final assembly and our production schedule is now locked. As such, we are not able to make any more changes.

Well probably not.  Physics starts to get into the way.  20 and 14nm are prohibitively expensive and probably will be for 2 years.   28nm is probably as good as it gets.  It remains to be see how efficient 28nm tech can be made but it is interesting that Cointerra and HF have roughly the same die density (GH/mm2) and roughly the same efficiency (J/GH).   For the record BFL claims similar but I don't put too much stock into it.  Still the odds that two teams with reasonable industry talent would achieve roughly the same performance and miss something obvious for higher efficiency is unlikely.  So no 14/20nm and pretty reasonable that any other future offerings will be ~0.8 J/GH (say +/- 50%).  That means even if the chips are free 1 PH/s is going to be expensive.  Converting, regulating and routing 800 KW isn't cheap and neither is getting rid of the 800 KW of heat it produces, even with refrigeration. 800 KW = >220 tons of AC.

Agreed.  Also Bitfury might also be able to get to ~0.4 J/GH with a die shrink to 28nm as they have very impressive specs for 55nm.  Still anything outside the realm of a magnitude or more in efficiency improvement means that 1 PH/s is going to consume an insane amount of power.

They have Friday and Monday. unlikely.

Agreed.  Also Bitfury might also be able to get to ~0.4 J/GH with a die shrink to 28nm as they have very impressive specs for 55nm.  Still anything outside the realm of a magnitude or more in efficiency improvement means that 1 PH/s is going to consume an insane amount of power.  Cut the 800KW in half, even 400KW is still 1,700 amps @ 240V.  Then remember energy in = energy out.  Most people don't think about heat in KW but 400KW is 1.4 million BTU/hr.   To put that number an average size furnace is 60K to 80K BTU/hr.

We are talking about something like this for cooling.

Oh, cloudhashing (big knc customer) has in its FAQ stated


but on the other hand the timer in the head banner shows

now it's 26th, 21:40 BST meaning => count down ends ~ on Oct. 1st 10:20 a.m. GMT+1

Wow I did not know of the huge resale value of bitcointalk.org accounts ...

I guess when they are new & free they are not worth anything ...but after they have got a few posts under their belt thats when the $$ & offers start flooding in to sell your identity

It all makes sense

its also fair to say current designs and the selected process nodes are more likely geared for performance/mm² (per $)  than power efficiency. Not that I expect an order of magnitude difference, but more than doubling efficiency at the expense of performance per mm² or $ might not be out of reach with downclocking/volting and a LP process Maybe even a redesign if that would be worth it.

IS Bitcoinorama, Bargraphics??