Topic: Swedish ASIC miner company kncminer.com - page 728. (Read 3050074 times)

Just be very careful with this step:
4. If increasing the SPI voltage by itself does not fix your issue, increase the voltage on each affected die, this should then bring the die back to life (it is best to not change the voltage on any working die)

Keep a close eye on the current and make sure it doesn't go above 60A. To be safe, try to keep it in the range 40 - 50A.

I found that adjusting the voltage of one of my dies to around -0.025v gave me 58A! If I'd kept going to 0 or up into the positive numbers, I'd probably have blown it all to shit.

Every miner is different, I have seen guys with settings of +0.8v etc, but keep a close eye on the values on your particular machine.

Why there is no such warning in that "official" post I have no idea. knc really should warn people that the settings are wide enough to do real damage to their hardware and what to look out for to avoid that.


Edit: Yeah, this has been annoying me:

Present, stable:


Adjusted to -0.025v (my mistake, 57A, not 58A as I said)


What exactly would happened if I set it to the full +0.1458?

Q1 was never a possibility. Ever.

$2 in parts, a little soldering skills....

Thanks!   I tried messing with that a few times and it killed more dies.   I'll give it another whirl!

Orama might correct me if I am wrong, but I believe KnC had an internal sale for 3500$ per Jupiter. I'm not sure how much they sold, but I would imagine it could be a lot!

Some of the Oct Jupiters were shipped with only 4 ports as well. Not really fair for those that planned to upgrade them.

Yeap. Our 3TH miners will only be able to do 0.00000005 BTC a day by time we get them, but in the meantime Knc Miner is likely hashing with them right now!

Then KnC should release more Jupiter's now because the landscape is unfolding into something that nobody can later compete in except for cloud networks.

I am really starting to wonder if they might have produced a batch of Jupiters and decided to just keep them to mine privately. Despite their 28nm competitors lack of delivery, the network has still managed to grow rapidly and there is an obvious demand for their already designed and successful Jupiter miner. They even told us to get some BTC ready and implied they would sell from stock instead of taking pre-orders which was followed by complete silence while somebody very steadily added a lot of hashrate to the network. If they were able to manufacture another batch of Jupiters using the proceeds from the first 2 batches they would be foolish to sell them if they believe they would turn a profit.

I think a problem we will face with any ASIC manufacture is that if they are successful with their initial pre-order > development > product launch, they will no longer need to sell them to the public and just go private at that point. It would be the smart thing to do IMO. After getting us to take all the risk on the expensive RnD they can then build their own farms for a fraction of the price/risk...

I hope I'm wrong!

why would you guess BBB when he said Controller Boards? lol

damnit knc, why did you have to remove ports 5 & 6 from the Nov Jups. 

 Ugh. This statement does not inspire confidence for shipping Neptunes by end of Q1

amazing you're the first to see the typo.  and now that you spelled it out as HigaGash it sounds like XXX porn name   
thanks anyway

Agreed on all points.  We're getting buried quick by the difficulty, and if KnC doesn't step it up so the Neptunes ship very early in Q2 we're going to be in trouble.

+1

I appreciate the update from Bitcoinorama but my feet are starting to get a little cold after this info. With the difficulty still climbing fast despite the failure of Hashfast and Cointerra to ship thus far, it's obvious there's a lot more incoming hashrate than I was aware of and they will ship eventually as both have working prototypes at this point. Could KNC be producing Jupiters for in house mining? It's hard to believe bitfury/gen 2 avalon could be responsible for it all...

I expected a least a little more concrete Nuptune info from KNC by this point as we're still left with a huge 5 month delivery window that ranges from a nice ROI to a massive loss for us customers that are in effect bankrolling the development of their 20nm miner. His update gives me the impression that they are not weighing the ROI of their customers in the schedule as much as they did for their 28nm development. I really hope that the "Q1/Q2" delivery is sincerely due to their uncertainty instead of intentional misleading like almost all other ASIC venders have been guilty of.

I am also really curious about what the actual manufacturing cost of a Jupiter is now that the RnD has already been paid for. It seems like it would have been a much better idea to refine their existing 28nm product which would allow them to have a 100% solid March 1st ship date as I feel a 700+Gh/s Jupiter should still be competitive by then and very easy to pull off.

I think I will wait until the end of the month and if there is still no official information about their progress I will consider asking for a refund or selling my pre-order.

You might want to fix your tag line.. I haven't seen an 3000HG/s (HigaGash?) ASIC miner before!

if the beaglebone was the bottleneck I'm sure he would of said just that...
There's more (or less) to it they're not revealing.  This seems a bit smelly also.

@ 'Orama/KnC: Could sales of a Jupiter with controllerboard minus BBB be an option ?

@bitcoinorama And so... when will they put up for sale the update modules? next week? price? Payment will only be possible in bitcoins? And finally... only for the old costumers

Thanks for the update on the neptunes, Sr. Orama. I appreciate the candor and the details. Happy Friday!