Topic: KnCMiner Openday Wednesday 5th & Monday 10th June - page 3. (Read 91301 times)

"Virgin Islands "  maybe it's the KnCMiner guys with a bag full of bitcoins -- they taken the Gold and nerver to be seen again


just joking

Are you aware you are posting this at 5 in the morning Swedish time?

He's stuck at the Virgin Islands customs, trying to explain what "Mars" and "btc" mean =D

Where is Bitcoinorama? We need your down to earth feedbacks .. we are waiting and waiting ..

I'm also looking forward to some updates possibly with pictures.

I sifted through the last 10 pages and saw little concerning Openday reports.

Thanks in advance for sharing.

What's the latest on the openday?

LOL, "Package type: Standard BGA 144" ...what is "STANDARD" BGA? Is that 1.27mm pitch? 1.25mm pitch? Or 1mm pitch? Wait -- the dimensions are 10x10mm and the balls are 12x12 array, so this is NOT standard by any means! Is it 0.8mm BGA or 0.75mm BGA then? At least it's not ucBGA with 0.4 or 0.5mm pitch! There are so mayny pearls of useless "wisdom" on that page, it is very, very entertaining material. What do they mean "Global Foundries IBM core"? The page is full of nonsense... and the 65nm technology they use has some issues with power/performance ratio the TSMC process is superior to that one, sorry folks. Also, the leakage with rising voltage is horrendous!

And most of all: where are the capacitors? Which retard designed a custom ASIC substrate for low-speed operation only? WHERE are the capacitors??? Look on any graphics card chip and you will understand me. Or under the heat spreader of your CPU. Or on the ceramic or organic substrate of your old Athlon/Duron/etc ...the capacitors there are there to allow it to reach SPEED. True, they may have used far more advanded technologies... but that would be in the brag-description, wouldn't it?

Wait, you would pay in chips more than for the whole Jupiter unit? Of course, the AMD 65nm technology IS costly, that's one of the reasons AMD dropped the PhenomII CPUs so fast, even when they were pretty great. Basically the BFL 65nm is the same process as AMD  Phenom CPUs, right? And KnC 28nm is likely to be the same process as the HD7970 chips. That is a difference of 4 generations, plus we all know the praises sung to the TSMC 28nm process by all their customers (Altera, Xilinx and AMD loves it a lot too)

I'm not saying you are slow in the head, but their cost is around 1-4 million USD in research, development and production, you are saying this like you can get any amount of money you want for yourself, and I believe you that, but most of us can not do that and have to do things the normal way. Their only luck is that they have already the tools to design 28nm products, because they have done so for their previous customer.
And this is not "a retail".

Basically, you have not done your research at all. And it is "cost", not "cost price". You are not able to prove that they would make 5-20x more by mining than by selling, we are two who provided you with reasons it is not so.

https://products.butterflylabs.com/65nm-asic-bitcoin-mining-chip.html

BFL 4gh/s chips 65nm

xing fingers that they deliver on time.  have nice investment in this.  this could be the next Avalon!  Prices could easily go up 10X on each unit.

Something like that, but... it is NOT the difficulty OR the BTC price, it is a combination of both. Plus the market volatility.
The BTC price will start crashing repetitively very soon, and by the time the ASICs will start hashing there will be enough panic to force the miners to sell immediately what they mine to get more return than they would get the next week.
It will also serve to kill the last paradigm of GPU hashing permanently. GPU hashing will become the same joke as CPU hashing is now in few months of time.

Basically, what awaits us, is simultaneous combination of outrageous difficulty and very low bitcoin price.

This situation may last for several months and will kill off many people attempting to mine and the bad sentiment will keep this situation the way it will be for a long time.

Expect the Avanon2 chips to be DOA. (I am exaggerating of course, their price is fantastic. (~$10 per chip)

To quote wikipedia: "when there is blood on the streets"

https://bitcointalksearch.org/topic/asic-hardware-buyers-association-231806

mining hardware customers unite !

1) they already made a device utilising the said 28nm technology for their customer, so...
1b) I believe they are going to run the simple 32-bit implementation on the 28nm, as there is almost nothing that can go wrong
1c) Not true, I am hashing on a 28-nm technology chip made by TSMC, and it works very nicely...
1d) People buying HD9970 on Christmas may be hashing with 20-nm technology, but that one is more expensive at it needs triple-mask exposure which is getting really annoying.
2) far less hashrate per total network hashrate than at the first, second or even third set of Avalon units. So... boo! Invalid argument...
3) tiny power consumption? Not really, not really, completely in the line with 28-nm process, I expect the shipped units to take ~400W, but that depends on the tweaking, clocking, Vcc.
4) the prices are not cheap, plus the devices they sell are not cheap either, they are reasonably expensive for a small business enterprise to invest in.
5) not true, 3-month lead time is for the chip manufacturer to make the chips and package them into the BGA, their complete lead time is around 6-7 months when they started making this product. Reasonable -- I have been down this path too.
5b) the ASIC design has been closed, laid out and simulated and tested a long time ago!
5c) the order for the ASICs will be placed tomorrow; I'm not sure if the files are already uploaded in the mask-maker factory or is some masks have been already manufactured, but from what I heard, they have already arranged for a tight delivery (assured delivery) schedule, which the Avalon team did not. Avalon team went for the cheap option "it will be done when it will be done", that is how they were able to deliver at all in the first place.

I hope this explains a lot about the technology side.

+ + +
Totally my words, man!
Greed is the great driver of boom and bust cycles, currently we are heading straight head-on into a big, big bust of the BTC/USD.
They wouldn't make enough money for the next batch by simply mining with all the units they will manufacture, because you need the money well in advance before you start any manufacture.
Also, most manufacturers work way faster if you walk into their shop with CASH in hand!

excellent heating. kinda fancy though.

Looking forward:

Only 3600 coins per day to mine, if the exchange price of bitcoin stays at $100, and there are 36000 ASIC devices mining the coins, their daily return will be only $10, 10 month $3000, less electricity cost, not possible to ROI after one year

This means, if ASIC manufacturers delivered more than 36000 devices of decent hash rate (9000 devices per manufacturer, given 4 manufacturer now), then some of their late customer will not be able to ROI if bitcoin price does not rise. If bitcoin price rise another 10 times, then it is possible to have 360000 customer. So ASIC manufacturer's sale is tightly related to the exchange price of bitcoin. And they have motivation to promote the bitcoin's public value

They canceled the Mars to go straight to ASIC production. No stops in between. That's kind of a BIG risk.

But hey, who cares. OPM right?

Man, you are really hurting about that, are you? Relax, you'll live.

So how realistic is the September/October delivery frame? To my understanding ORSoC is a reputable company and I assume it would be considered highly unprofessional by them to deliver months late, yet I don't quite understand how they will pull it off unless they are already pretty far into the development process, and if they are, why have we not seen any proof/prototypes yet?

I live in Europe and orders in a week I get maximum 10 days.
I hope you can minig much with your asics v2 in February 2014