Same here; will be rocking 5xAX1200i if KfC ever dispatch bonus miners.

KfC asked me to send the burnt out cubes back to them at my expense so they can inspect them and decide if they should RMA. That's not an acceptable level of service right? They are essentially saying I have to solder new power sockets myself and hope the cubes still work or suffer further down time and shipping costs?

Yes, it's an over current issue alright! Over twice the current the socket is designed for was an over current issue from day one.

As for overclocking; there is no overclocking with Neptune. Everything from the power connection to the VRM's to the chips are already running at or above their maximum spec. The web interface allows users to change from 475Mhz to 500Mhz but this would not make enough difference speed wise to warrant the risk of changing stock settings.


In that situation you can usually see what happened; if there were two pins making good contact and one making poor contact it would be usual for the failing pin to be a lot more blackened / melted than the other two which burn out quickly afterwards. If there were two pins making contact and one with none then we would likely see two very blackened and melted and one intact? However it's neither of these cases; it had been running for a good while before it melted and all three pins melted consistently to each other.

u27

...and now for lots of nice high res photos:





Since KnC didnt bother to respond to me, let alone RMA, last time this happened; here's some warranty voiding nudes:



(nice to see they wrote "2014 KnCMiner" right alongside the socket so there is no confusion)

u27

I requested refund before they started shipping which was refused. The fire hazard miner was then delivered weeks late. RMA was refused on dead dies on the grounds that the remaining dies were running slightly over 3TH. To top it all off this is now 2/5 cubes burnt out with 16AWG cables.

KnC have two arguments in this regard;
1. Low quality cables
2. Improperly seated cables i.e. not all six pins making contact

Real world;
1. Clearly not a low quality cable with 16AWG; it's a design issue where each 50w pin is expected to handle >100w
2. As shown all three of the positive pins burnt, not just two, as it would if one of them wasn't seated properly

I have been lucky so far in that my house hasn't burnt down yet.

If *everybody* could post this link on the KnC fourms *every day* that would be good, thanks. https://bitcointalk.org/index.php?topic=170332.msg8498317#msg8498317

...and now for lots of nice high res photos:





Since KnC didnt bother to respond to me, let alone RMA, last time this happened; here's some warranty voiding nudes:



(nice to see they wrote "2014 KnCMiner" right alongside the socket so there is no confusion)

u27

[/quote]

August... they can't seriously suggest that this meets their obligations?

It appears to have leveled out; around 20%.

At this rate it could be record diff increase.

Well it could be that yea. I'm sure it has nothing to do with KfC jumping in from of Eliguis hash rate wise:


Last week KfC mined 38 out of the 1081 blocks mined (yes, that is over half a million dollars for the week) and that worked out at almost exactly 3.5% of the network. No wonder people are still waiting for batch one ($13k) neptunes.

u27

Jesus, could they get any more vague? When did they change "expected August" to "after the summer" which could literally mean any future date?

Also, has anybody else noticed that the difficulty is right around vertical since the last diff change? Have KfC actually started shipping batch 2 ($10k) or is it just ants and SP30 units coming online?

The ant "testing pool" is up to 3% of the network and that is a lot of 441GH/s miners!

u27

Yea, with the best will in the world and no major issues you're looking at a month from wafer before you can reasonably expect to see miners delivered (excluding the one off "look, we met our deadline" miner which may arrive a week or so earlier).

To give them credit this kind of timescale was unheard-of before KnC; which makes it all the more amazing that they shipped with one fire hazard PCI-e power socket.

u27

A few people asked him for contact details and none were forthcoming.

Yes, they will likely rip us off, but that doesn't make it ok.

Even by their twisted interpretation of the wording KfC still delivered all of batch one ($13k) late without offering free hashing as promised; some of them have been very late. If KfC now tag batch two right on the end of batch one it effectively punishes batch one customers (to the tune of $3,000) for having a little faith in KnC and ordering Neptune as soon as they were open to the public.

It's just not right. I would be very happy for batch two and batch three to ship this week but KnC need to acknowledge that they were late and as such they need to lump all the batches together in one big batch for continuous shipping. Now that all the batches have become one big late batch they need to sort out compensation for the people who paid more. How can they justify the last guy in batch one not paying the same as the first guy in batch two?

I could understand if the last guy in batch one got $3,000 refund and everybody else got it pro-rata; so where "n=umber of days between delivery and start of batch two" it would be $3,000 - (0.1BTC * n) refunded to each batch one customer. That wouldn't provide any incentive for being in an earlier batch but at least the faithful wouldn't be worse off.

u27

There had better not be! They have to leave a month after batch one don't they?

Either that or send me $3,000 right about now.

like this?

http://www.etb-tech.com/poweredge-750w-redundant-power-supply-2950.html?gclid=Cj0KEQjw3vyeBRDt673h5ZTYrdQBEiQAzhkf8ibziN_iuniQ32q7SSyrYIrzPczmNKUu6DyEDNq6eyoaAs0q8P8HAQ#.U9-Yo2Ms-qk

I'm getting 94% with the Corsairs. 92% is really good for a server PSU so you must be running them right in the sweet spot; I have never seen them run better than 90% myself.

I had a bunch of nice 1U server supplies that would have covered Neptune before I ordered the Corsair ones but I wouldn't be overly happy running without any OCP in a consumer environment; those server supplies will just keep pumping 12v out until any possible short reaches 2kW and that could easily mean a fire. They are designed to run in server rooms with fire suppression; not domestic dwellings.

Just my two cents.

u27

I just figured that running 4kW+ of Neptunes (when the other five cubes arrive), expending energy to waste the heat generated by them and then paying to run a 3kW immersion heater at my house for our hot water is kinda stupid.

No reason I can't run the lowest coil in an evap. cooling system through a big tank of water 24/7 and then only waste the energy if the 60 degree 7100 vapor gets to the next coil up; at least then I get free hot water for running miners.

I would have it up and running by now if I wasn't so busy with work. Likewise I would love to get involved with what you guys are doing; but I just don't have the time.

u27

Does your mac have 16 processor threads and 84699mb total graphics memory? I thought not. My video rendering box does.


Looks like two of the same box laying on it's side to me.

u27

No need, covered it ages ago: https://bitcointalk.org/index.php?topic=170332.msg7508465#msg7508465

You clearly don't need five cables with your AX1200i because you will need more than one (ideally three) anyway.

With two cubes per AX1200i I'm getting very good efficiency and 100% stability to date; not to mention OCP which may very well prevent my house burning down given this KfC "design".

This guy recommends AX1200i with splitter cables; you should be using splitter cables anyway since I have seen those EVGA boxes without OCP melt PCI-e plugs at the PSU end.

u27

It is my understanding that the efficiently recovered energy in the form of hot water is then wasted with your systems? For instance I recall seeing a couple of external cooling units at asicminer?

Have you looked into using the energy?

Approx. 4000kcmil