Topic: Swedish ASIC miner company kncminer.com - page 433. (Read 3050075 times)

it's dead and bloody fucked

RMA it and they'll send it back with 3 dead dies.. same board!!

Dead die?

This is an October Jupiter that's been running from 24/7. Noticed I have a dead die. I already increased the SPI voltage all the way up to 3.3V and it wouldn't kick it on. Then I slowly increased the individual die voltage to the MAX, which is +.1465. Didn't awake up the die either. Is there anything else that can be done to kick this die on?  I would really thought that the rig would last a lot longer then 8 months.

While probably true for Franken & Neptune...
If you didn't ROI on their first 28nm batch... You probably didn't buy one.

Try turning the overclocking off?

I believe the only people making ROI on the equipment KfC sells is KfC.

Difficulty has risen again. Super Jupiter does now 0.15238969 BTC per day. (3.17 TH/s)

In July +/- 0.065 BTC per day.

In August +/- 0.04366 per day.

Difficulty has risen again. Super Jupiter does now 0.15238969 BTC per day. (3.17 TH/s)

In July +/- 0.065 BTC per day.

In August +/- 0.04366 per day.

Dead die?

Dead die?

This is an October Jupiter that's been running from 24/7. Noticed I have a dead die. I already increased the SPI voltage all the way up to 3.3V and it wouldn't kick it on. Then I slowly increased the individual die voltage to the MAX, which is +.1465. Didn't awake up the die either. Is there anything else that can be done to kick this die on?  I would really thought that the rig would last a lot longer then 8 months.

Dead die?

This is an October Jupiter that's been running from 24/7. Noticed I have a dead die. I already increased the SPI voltage all the way up to 3.3V and it wouldn't kick it on. Then I slowly increased the individual die voltage to the MAX, which is +.1465. Didn't awake up the die either. Is there anything else that can be done to kick this die on?  I would really thought that the rig would last a lot longer then 8 months.

Found what I had read: http://www.coindesk.com/kncminer-offers-plan-b-case-neptune-miner-delivery-delay/


This is where it said a choice...

I don't mean to rain on anyone's parade but I thought Sam said it's either or, Plan B or wait, did he change it?

“Plan B is simple,” Sam Cole, one of KnCMiner’s co-founders, told CoinDesk: “If we face a delay or any type, we will give each person a choice.”

This is a plan that the company is putting in place for paying customers of its latest model, the 3TH/s Neptune miner.

The options are as follows:

1. Continue to wait.

2. Take a USD refund.

3. Convert to 3TH of hosting performance for a period (still to be worked out, but is likely to be around 6 months).

I'm beginning to wonder if some of the FrankenJup owners haven't given up too quickly trying to get their Jups running. Some of the orders were indeed total shit and unusable. But I've managed to get another 6 boards running from the second batch for a total of 24 boards and 3.7THs*. I expect to easily get another 6 boards running for another ~900GHs. Maybe more if I bought another controller, which I won't. I possess not special technical prowess either. So if I can do it, anyone can... if it's at all possible.

Of course I was lucky to be in the first 'assembled' batch too and received an extra 18 boards to work with. So there's that, lol.

I can't say I'm completely dissatisfied although the design of the FJ's are crap. I'm hashing now with more THs than I originally paid for. Although I did have to buy 2 extra Controller boards (at an exorbitant price), 2 additional PSUs and 12 i30s. Waiting for the IDC cables was the biggest headache. KNC should supply those with the Controller boards since they are rather uncommon and difficult to find.

If knew then what I know now, I would have certainly cancelled or never ordered from KNC in the first place. But I consider myself to be one of the lucky ones. I could have made many other choices that would have been much worse, such as preordering from HashFast, BitMine.ch, VMC, etc. or continued to wait for a Neptune. Maybe I'll be jealous if/when they are finally delivered. Maybe not.

On the dark-side... I'm still waiting for 2+ Monarchs that probably will never be delivered before BFL goes tits-up. Lol.



* underclocked... Otherwise I'd get very close to 1THs per 6 modules. But the FrankenJup @ stock clock runs too hot for my tastes. Also they seem to be much more efficient at lower clock speeds and since I'm maxing out my circuits, I need to watch this carefully. Plus it's so DAMN hot in my house despite the recent addition of supplemental A/C for my "server" room (spare bedroom). I have to keep the house A/C at 85 during the day so it doesn't run constantly.

I don't think you're dreaming. I think you're just high. Can I have some of what you're smoking, please?

Even if KNC could get it to 10THs, there'd be virtually no way for you to supply enough power to it even if they got power consumption down to 0.5w/GHs (which they won't) that would be 5,000 watts. Going from 28nm to 20nm isn't enough of a die shrink to increase that much hashing power AND reduce power consumption. Impossible.

Btw, what does "8tg/s" mean? That is a unit of measurement I've never seen before?

Why I'm thinking Nipy will come with 10th/s all day, this let me couldn't get sleep, with 20nm chip, at least 30% potential enhancement, and about 20% power reduce, at higher integrated circuit cluster, am I day dreaming~  

Tity = 400mh/s, so Nipy at least 8tg/s at same 28nm scale, and Nipy is in 20nm, so at least 9.6th/s, am I dreaming~