Odd. Looking at the manual it looks like the PCI-E power connectors from the board are 4 pin into 6 pin from the supply, leaving two pins overhanging the edge. Okay, that would work.
Here's a problem. I don't see the wires to the connectors color coded, yellow for 12v and black for ground.
I saw this on ebay and bought one: NEW PCI-E 8 PIN DAISY CONNECTOR---6"+18"---SEE ACTION IN VIDEO CARD---USA MADE
I had assumed the "clip" side would be the same polarity as the "clip" side of the 4 pin. A couple of things: a) the PCI-8 has on one side 4 black wires and on the other 3 yellow wires and a black; b) when compared to a PCIE-4 the "clip" side of the PCIE-4 has 2 yellow and the non-clip side has 2 black. The clip side of the PCIE-8 are ALL black?!
This would make it very easy to apply the wrong polarity to the board or even short out one's supply.
Are the PCI-E-8 correct for what they are?
I can see that if consistently wrong at all connections the jumper could be put in place and work just fine given that the hole flow doesn't recognize color but if one were to CUT the wires and connect to the positive and negative of a power supply, one could be wrong.
Also, I can determine which pins are 12v and which ground by buzzing continuity from the 4 pin connector.
Checking an old motherboard having a 4 pin PCI-E connector, the clip side is not ground while the non-clip side is common with the metal shield covering the USB stack. |
|
|
|
|
Odd. Looking at the manual it looks like the PCI-E power connectors from the board are 4 pin into 6 pin from the supply, leaving two pins overhanging the edge. Okay, that would work.
Here's a problem. I don't see the wires to the connectors color coded, yellow for 12v and black for ground.
I saw this on ebay and bought one: NEW PCI-E 8 PIN DAISY CONNECTOR---6"+18"---SEE ACTION IN VIDEO CARD---USA MADE
I had assumed the "clip" side would be the same polarity as the "clip" side of the 4 pin. A couple of things: a) the PCI-8 has on one side 4 black wires and on the other 3 yellow wires and a black; b) when compared to a PCIE-4 the "clip" side of the PCIE-4 has 2 yellow and the non-clip side has 2 black. The clip side of the PCIE-8 are ALL black?!
This would make it very easy to apply the wrong polarity to the board or even short out one's supply.
Are the PCI-E-8 correct for what they are?
I can see that if consistently wrong at all connections the jumper could be put in place and work just fine given that the hole flow doesn't recognize color but if one were to CUT the wires and connect to the positive and negative of a power supply, one could be wrong.
Also, I can determine which pins are 12v and which ground by buzzing continuity from the 4 pin connector.
|
|
|
|
Miner protection offered by KnC would be a huge plus.
So today I'm thinking about the possibility of my Mercury getting fried from lightening. I recall a Presario that was destroyed by lightening. The RJ-45 was molten copper at the motherboard. The power in jack was blackened. The power brick was dead. Here in the US a home will usually have two phases of 120VAC relative to neutral and neutral is tied to ground in the service and then ground to a ground-rod driven into the ground properly at the service and the service has only those two phases coming in from the electric utility. Perhaps the power company asked the cable company to pay a fee for connecting to their ground, was refused and got cut. So, the jolt of lightening came in via the cable, thru the cable modem to the RJ-45 ethernet connection, out the power-in jack thru the power brick trying to get to neutral in the 120VAC outlet which is tied to ground in the service. If the computer was using Wifi it wouldn't have fried. They should call it WhyFry. So, I was thinking about my Mercury. Since it has an operating system I'd need do something like take a laptop and configure its RJ-45 as the downlink of a router and the laptop's wifi as the uplink. My very first router was an older system with multiple ethernet cards running either Slink or Potato. That I recall was not for me an easy thing to do. I wonder if it's any easier with Wheezy. I'd probably take a communication penalty in metrics. Think I'll try cable box -> 802.11 linksys wireless router -> (wireless) -> bridge -> my regular network routing. I'll simply be unplugging my regular network routing uplink RJ-45 from the 802.11 and putting the bridge in place of the uplink cable, the uplink cable then going the bridge. My entire regular network will then not be wired to the cable box. This instead of wifi on each of the miners. No threat of molten RJ45 copper and lesser fried power in on boards. Well, bridge isn't what I want. Having two E900 Linksys routers, looking at the config I see "bridge" offered as an uplink option. Trying that I lost the network for a while. Seems bridge as they see it is taking the uplink and disabling it, then allowing the downlink ports and the wireless to be connected as if connecting a high speed switch. There isn't a capability of *wireless* uplink to the upstream router. If, as their example shows, there is still a hardwire ethernet link then I don't have wireless isolation of the downlink router. Which is the correct forum for this? I'm only doing this because the KnC will be the highest cost device on my network. |
|
|
|
Miner protection offered by KnC would be a huge plus.
So today I'm thinking about the possibility of my Mercury getting fried from lightening. I recall a Presario that was destroyed by lightening. The RJ-45 was molten copper at the motherboard. The power in jack was blackened. The power brick was dead. Here in the US a home will usually have two phases of 120VAC relative to neutral and neutral is tied to ground in the service and then ground to a ground-rod driven into the ground properly at the service and the service has only those two phases coming in from the electric utility. Perhaps the power company asked the cable company to pay a fee for connecting to their ground, was refused and got cut. So, the jolt of lightening came in via the cable, thru the cable modem to the RJ-45 ethernet connection, out the power-in jack thru the power brick trying to get to neutral in the 120VAC outlet which is tied to ground in the service. If the computer was using Wifi it wouldn't have fried. They should call it WhyFry. So, I was thinking about my Mercury. Since it has an operating system I'd need do something like take a laptop and configure its RJ-45 as the downlink of a router and the laptop's wifi as the uplink. My very first router was an older system with multiple ethernet cards running either Slink or Potato. That I recall was not for me an easy thing to do. I wonder if it's any easier with Wheezy. I'd probably take a communication penalty in metrics. Think I'll try cable box -> 802.11 linksys wireless router -> (wireless) -> bridge -> my regular network routing. I'll simply be unplugging my regular network routing uplink RJ-45 from the 802.11 and putting the bridge in place of the uplink cable, the uplink cable then going the bridge. My entire regular network will then not be wired to the cable box. This instead of wifi on each of the miners. No threat of molten RJ45 copper and lesser fried power in on boards. |
|
|
|
How do we configure our miners? Can you plug in a monitor, mouse and keyboard setup to them directly, or are settings accessed via LAN?
Exactly what I was wondering. KNC said it would be "stand alone". And what pool would we join to get running? Oh boy, this'll be fun.  pretty sure its all done by lan, ssh login to the beaglebone then edit settings form there. should be quite simple really. Use any pool you like! seems like everyone knows about ssh but me, how do you access a linux machine, using ssh from a windows (XP) machine? Putty. Options are ssh, telnet, rlogin, raw, not in that order. |
|
|
|
Seriously you are a member of the foundation and have no clue about normal hashrate variance?? Only thing thats "suspicous" is the drop but i guess that were the ddos attacks on guild etc.  Those guys were taking in a couple of grand dollars a day with a botnet? If they stay dead that's a nice potential hash drop. Gotta figure they're no dummies. Some probably bought KnC miners and post here. Maybe we'll hear, Oh so and so is gone from the forum, hauled off to the slammer for tax evasion. |
|
|
|
Damn, my computer is louder than that. same here, my workstation makes definitely more noise I'm also interested in having more details about this (like anybody else fwiw). Anyway the avg hash rate for the last 3hs is 683.99. if confirmed it would be impressive. If you remember what Orama observed when he visited about the guys there and how they were really into what they were doing, they were enthusiasts and keen to see what they could make these rigs do...so it looks like his impression was spot on and now they have something to play with they are in their element. Maybe they're pushing one to destruction to see what the limit is? Speaking of destructive testing, I'd like to see them put one of each model in a well insulated large box outside having an air conditioner. Have the air conditioner cycle on full cold, 2 hours on 2 off, letting the running miners heat warm the box while the AC is off. The expansion and contraction will show what to expect regarding cycles and extremes before failures, what it takes/costs to repair, how best to adjust the soldering of the ASICs to minimize failure, things like that. Of course an environmental chamber would work too. I remember days of testing a radio frequency modulator for space.... |
|
|
|
just a reminder,
50BTC and BTCguild are both under DDOS attack.
I was checking why my miner doesn't work. I though it was the hardware. Now it seems that it is the pools that are not able to respond.
Sit tight and wait.
Earlier, happened with Slush's stratum yet I felt it was the miners because only most stopped, those on stratum mining proxys on more than one machine for best diff differentiation, the raspberry pi running minepeon successfully switched to stratum.bitcoin.cz while previously it was sharing a mining proxy with a ztex. So, I tried stratum+tcp from various miners directly to stratum.bitcoin.cz and that wouldn't work for some time, then did. So, if Slush was under attack he recovered in under a half hour I think. |
|
|
|
So, almost the first (within 1 hr 40 min)
bitbet lost
no hash-rates posted
No hashing machine posted.
Alot of concerned looking faces in the pics.
Am I missing something?
Set up a webcam once. Later a glitch with one of my routers. The webcam caught me troubleshooting. For some reason I looked furious. If someone else it'd scare me. I KNOW I wasn't that angry, just perplexed. |
|
|
|
|
Maybe at one of the computer shows somebody from a competing miner manufacturing company said a successful September delivery would be inauspicious, looking at Black September and 9/11.
|
|
|
|
I dont mind any minor delays of a few days. Thats not the problem. I dont like the non-existent communication-policy.
We had a on-off communication the last months. Sometimes very detailed then again absent for several weeks. And now its just non-existent again.
Either dont communicate details at all and stay silent or communicate and tell your customers whats going on.
But this inbetween of broken promises (Video, China-Show etc) and silence is just not good imo.
Yeah I agree.....but then again....I'll probably stick with it like the titanic ..(I'm a bfl order of april.....with nothing yet there either so probably looking at November for those folks) the below clip applies http://www.youtube.com/watch?v=nZIlAExvneosigh....but then again....if this all goes to crap with both BFL and KNC I've learned a valuable life lesson...ie I'm going AMISH screw technology...hot amish girls (who have no media to know they are hot)....clean air...good food...HARD BRUTAL LABOR WITH MY FRIEND THE MULE 12hr days 6 days a week...er what was that last part ...er nevermind.... it is what it is ....should just write this all off ..... look at it simplifying my lifestyle...(ie less $$$) see I'm more 'Zenlike" already (probably better I have less ego now anyway.....after all this.......(start the denial here........) Searing Raise chickens. |
|
|
|
Yep the silence is weird. Orama has been on a few times and has also said noting. Im pretty sure if there was bad news though that he would let us know. I can only assume that they are holding back until they have a machine hashing which should be today at some stage imo
With gambling involved, bitbet, who can say what's right. |
|
|
|
I remember a statement from KnC that those machines can even run in an apartment in Dubai without air-condition, there are margins upon margins.
That might address overheating rather than temperature extremes. |
|
|
|
|
Given the large size of the ASIC, heating a cool room with these miners will expose the ASICs to a range of temperatures and the large size expanding and contracting could be a problem. Might see lots of failures from those conditions while temperature controlled rooms see the miners go unscathed.
|
|
|
|
|
I'd really like an assurance from KnC that there were no difficulties soldering the ASICs to the boards. That large size ASIC is novel.
|
|
|
|
|
11:03 PM Sunday night there. Or will they be shipping from China?
|
|
|
|
Think they underestimated the time it takes to put the ASICs on boards?
Maybe there are unexpected quirks to such a large size ASIC. Getting all the ball soldered on the grid might be more difficult given the large size. |
|
|
|
|
Think they underestimated the time it takes to put the ASICs on boards?
|
|
|
|
|
Recording daily noon network hashrate, 9/27 topped 1,324,213Gh/s then dropped back yesterday and today is 1,272,428. Could that 9/27 peak have been KnC?
|
|
|
|
Nice idea. Avoid the cable to powerline potential at the miner. If a cable to powerline high potential, molten copper potential, occurs it will happen where the ethernet is injected onto the powerline and show no cable to powerline potential at the miner. |
|
|
|
|
Show Posts
