Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
|
November 25, 2013, 06:24:25 PM Last edit: November 25, 2013, 06:48:23 PM by Phoenix1969 |
|
single rails provide much cleaner power and are able to supply the extra amperage You keep saying this. Can you provide some support for this statement please? okay, I had a brand-new 700watt atx, that had 5 12v rails, which totally fried just trying to run a Saturn for 5 lousy minutes. Why?...each of the two pci feeds had a 150 watt rating, and the boards draw very close to that amount. It fried. I'm trying to save you from my mistakes.... Why would anyone want a multirail psu anyway? one rail overloads....you're done. with a single 12v rail, no single pci feed can overload the psu...nor hamper the hashrate from limiting the current flow due to overdraw....period! Hope that helps.. *Some may forget, but I'm an energy researcher, and recently consultant to a large energy corporation in development of fuel gasification technologies, more specifically in the development of more efficient systems using resonant energies through the application of harnessing the bounce of signal between a HV coil, and a PWM, as well as the enhancement of combustion through the use of adding Direct Injection, plasma ignition, and Hydroxy gasses made through the harnessing of the coil HV signal bounce, and shunted to a 143 plate cell where it is produced, and then metered into the Gaseous injection system, mixed with the vapors from whatever your fuel is in the gassifier... enough on that... but I can't stress enough how the actual current reaching any electronic(especially digital)device is critical to it's performance.
|
|
|
|
btc_uzr
Sr. Member
Offline
Activity: 476
Merit: 250
let's have some fun
|
|
November 25, 2013, 06:27:12 PM |
|
No details. Coming Soon TMWow! 20nm AND early 2014! That MUST be the "incoming news" announcement from 'orama. Bingo! Great news, 20nm! Bad news 'Orders opening soon'! Bitcoinorama, in hope I get not ignored I'd like to ask the following: 1) Why is the pre-order game continuing ? Didn't they collect enough money to fund the next gen ?! 2) Do you have to pay the full price when opening the order books or is it like pay 10-20% up front and then the rest when it's close to shipping...? 3) How much will they approximately cost ? Would be very nice to get a clue before you actually have to pay.. 4) ANY advantage for the early investors who took a high risk and making KnCMiner's success possible in first place ? 5) Will there be another batch of upgrade boards for October gen ? Thank you very much in advance for your answers! **For those who want to pay in BTC, be aware here's a guy offering you a 60% return of your coins without additional costs for electricity, tranquilizers, waiting, or else...might be the better choice
|
..and Thou shalt spread the coin in the name of cryptography for eternity
|
|
|
Henchman24
|
|
November 25, 2013, 06:37:58 PM |
|
single rails provide much cleaner power and are able to supply the extra amperage You keep saying this. Can you provide some support for this statement please? okay, I had a 700watt atx, that had 5 rails, which totally fried just trying to run a Saturn for 5 lousy minutes. Why?...each pci feed had a 150 watt rating, and the boards draw very close to that amount. It fried. I'm trying to save you from my mistakes.... Your argument is a red herring. The OP was very specific about the PSU he was using, which supplies more than adequate amperage per rail. The above is a story about you choosing an inadequate multi-rail PSU. To use that one example to support your above statement is just silly. Surely you must see that? I can see why Edgar gets frustrated with some of your posts. You have a hard time admitting when you're wrong. The plural of anecdote is not data.
|
|
|
|
xyzzy099
Legendary
Offline
Activity: 1065
Merit: 1077
|
|
November 25, 2013, 06:42:42 PM |
|
yep, ouch. single rails provide much cleaner power and are able to supply the extra amperage during the spikes in hashrate... and two 1500watt supplies on a 20 amp circuit is not so good either...(20 amps x 110v = 2200 watt max per 20 amp circuit, (or in EU...220v x 10amp = 2200watt)and you are saying there is more than two jupiters on that circuit?....I really think it's your power situation now for sure...
I think these 'spikes in hashrate' you see is just variance. The chips run at a constant speed. Variance does not cause your miner to draw more or less current. Okay spikes are variance....? yes... and what causes that variance?...who knows...have you used a kil-o-watt to see that what you are saying is correct? Yours stay rock-solid on a single wattage reading? Mine don't. The reading on mine at the wall goes up & down in direct correlation with the variance... and when I had them on a "Shared circuit" it was terribly obvious the other things going on were effecting it when the kil-o-watt dropped when other things were used in the house, like microwave oven, coffee pot, toaster, aircon, etc.... That's just my observations here, which is why I say that. You can even watch bertmod to verify what I'm saying.... the vrm output goes up & down with hashrate as well My VRM's are outputting a FULL 12v, @ 50amps now btw... and if I said how...Edgar would complain.... but it has to do with temps and clean, unwavering, unlimited power.... Variance is caused by luck. Spikes (both up and down) in your apparent hashrate can (and do) happen based on sheer luck. Your miner could, say, generate 10 diff 256 shares in 10 hashes, and it will look to the pool like your hashrate is much higher than it really is, briefly. The converse is also true. One thing that could possibly explain your observation though is the enabling and disabling of cores - that would affect both your hashrate and the current draw, I imagine, but it seems like the difference would be pretty small unless a number of cores get enabled/disabled at once. Each core is only 1/192 of one modules hashpower and probably an even smaller fraction of it's current requirements.
|
Libertarians: Diligently plotting to take over the world and leave you alone.
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
|
November 25, 2013, 06:54:16 PM |
|
yep, ouch. single rails provide much cleaner power and are able to supply the extra amperage during the spikes in hashrate... and two 1500watt supplies on a 20 amp circuit is not so good either...(20 amps x 110v = 2200 watt max per 20 amp circuit, (or in EU...220v x 10amp = 2200watt)and you are saying there is more than two jupiters on that circuit?....I really think it's your power situation now for sure...
I think these 'spikes in hashrate' you see is just variance. The chips run at a constant speed. Variance does not cause your miner to draw more or less current. Okay spikes are variance....? yes... and what causes that variance?...who knows...have you used a kil-o-watt to see that what you are saying is correct? Yours stay rock-solid on a single wattage reading? Mine don't. The reading on mine at the wall goes up & down in direct correlation with the variance... and when I had them on a "Shared circuit" it was terribly obvious the other things going on were effecting it when the kil-o-watt dropped when other things were used in the house, like microwave oven, coffee pot, toaster, aircon, etc.... That's just my observations here, which is why I say that. You can even watch bertmod to verify what I'm saying.... the vrm output goes up & down with hashrate as well My VRM's are outputting a FULL 12v, @ 50amps now btw... and if I said how...Edgar would complain.... but it has to do with temps and clean, unwavering, unlimited power.... Variance is caused by luck. Spikes (both up and down) in your apparent hashrate can (and do) happen based on sheer luck. Your miner could, say, generate 10 diff 256 shares in 10 hashes, and it will look to the pool like your hashrate is much higher than it really is, briefly. The converse is also true. One thing that could possibly explain your observation though is the enabling and disabling of cores - that would affect both your hashrate and the current draw, I imagine, but it seems like the difference would be pretty small unless a number of cores get enabled/disabled at once. Each core is only 1/192 of one modules hashpower and probably an even smaller fraction of it's current requirements. okay...I'm just dreaming, which is why I can get 143GH+ from every single board, even those with the "Die 0" problem.
|
|
|
|
xyzzy099
Legendary
Offline
Activity: 1065
Merit: 1077
|
|
November 25, 2013, 06:56:41 PM Last edit: November 25, 2013, 07:32:39 PM by xyzzy099 |
|
yep, ouch. single rails provide much cleaner power and are able to supply the extra amperage during the spikes in hashrate... and two 1500watt supplies on a 20 amp circuit is not so good either...(20 amps x 110v = 2200 watt max per 20 amp circuit, (or in EU...220v x 10amp = 2200watt)and you are saying there is more than two jupiters on that circuit?....I really think it's your power situation now for sure...
I think these 'spikes in hashrate' you see is just variance. The chips run at a constant speed. Variance does not cause your miner to draw more or less current. Okay spikes are variance....? yes... and what causes that variance?...who knows...have you used a kil-o-watt to see that what you are saying is correct? Yours stay rock-solid on a single wattage reading? Mine don't. The reading on mine at the wall goes up & down in direct correlation with the variance... and when I had them on a "Shared circuit" it was terribly obvious the other things going on were effecting it when the kil-o-watt dropped when other things were used in the house, like microwave oven, coffee pot, toaster, aircon, etc.... That's just my observations here, which is why I say that. You can even watch bertmod to verify what I'm saying.... the vrm output goes up & down with hashrate as well My VRM's are outputting a FULL 12v, @ 50amps now btw... and if I said how...Edgar would complain.... but it has to do with temps and clean, unwavering, unlimited power.... Variance is caused by luck. Spikes (both up and down) in your apparent hashrate can (and do) happen based on sheer luck. Your miner could, say, generate 10 diff 256 shares in 10 hashes, and it will look to the pool like your hashrate is much higher than it really is, briefly. The converse is also true. One thing that could possibly explain your observation though is the enabling and disabling of cores - that would affect both your hashrate and the current draw, I imagine, but it seems like the difference would be pretty small unless a number of cores get enabled/disabled at once. Each core is only 1/192 of one modules hashpower and probably an even smaller fraction of it's current requirements. okay...I'm just dreaming, which is why I can get 143GH+ from every single board, even those with the "Die 0" problem. I don't understand that response. How is this related to hash-rate spikes being caused by variance?
|
Libertarians: Diligently plotting to take over the world and leave you alone.
|
|
|
arlekyn13
|
|
November 25, 2013, 06:56:48 PM Last edit: November 25, 2013, 07:09:34 PM by arlekyn13 |
|
What I tried to underline is that: a) I have no idea what caused my Saturns problems, my Jupiter running on the same line and same PSU type was not affected b) The Saturns came with all boards affected by die #0 issue, (same as my Jupiter), they are running with die #0 off on 0.98 c) Both Saturns ran pretty well on 0.98.1 beta, one of them easily powering all dies, the other required more heat and was really hard to keep all dies powered on d) Once 0.99 was released, i switched all my machines to it and for a day or two, they were running the same as on 0.98.1 beta e) All of a sudden, both Saturns became affected with disabled cores, but in a different way: one of them simply has all cores disabled at startup (and keeps them like that) on 0.98.1 beta or official and 0.99. The other one starts normally, but after a while almost all cores on the same ASIC (coincidence?), the lower one on the web interface, gets disabled, only 0-5 cores remaining enabled on each die. Again, disabled cores are present on just one ASIC, the lower one in the web interface page. This started at a day or 2 after 0.99 and the problem remained even after hard reset and affecting 0.98.1 as well, beta or official. f) Both Saturns run normally (with their die #0 issue) on 0.98
My guess was that 0.99 FW somehow caused this issue, but it just persists now on 0.98.1 as well.
|
1CmrswU7JYpi9WNC8EHWCV3aam1FJsW2Zu - to show appreciation for my work
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
|
November 25, 2013, 07:05:03 PM Last edit: November 25, 2013, 07:47:55 PM by Phoenix1969 |
|
What I tried to underline is that: a) I have no idea what caused my Saturns problems, my Jupiter running on the same line and same PSU type was not affected b) The Saturns came with all boards affected by die #0 issue, (same as my Jupiter), they are running with die #0 off on 0.98 c) Both Saturns ran pretty well on 0.98.1 beta, one of them easily powering all dies, the other required more heat and was really hard to keep all dies powered on d) Once 0.99 was released, i switched all my machines to it and for a day or two, they were running the same as on 0.98.1 beta e) All of a sudden, both Saturns became affected with disabled cores, but in a different way: one of them simply has all cores disabled at startup (and keeps them like that) on 0.98.1 beta or official and 0.99. The other one starts normally, but after a while almost all cores on the same ASIC (coincidence?), the lower one on the web interface, gets disabled, only 0-5 cores remaining enabled on each die. Again, disabled cores are present on just one ASIC, the upper one in the web interface page. This started at a day or 2 after 0.99 and the problem remained even after hard reset and affecting 0.98.1 as well, beta or official. f) Both Saturns run normally (with their die #0 issue) on 0.98
My guess was that 0.99 FW somehow caused this issue, but it just persists now on 0.98.1 as well.
Hmm, that is indeed strange, and I had a similar incident... which is why I stay on the beta.... I had to reboot/reflash one of the machines a few times... I 'm clueless on the actual problem, but as I stated yesterday, that very same machine/hashing board was fixed in a "desperation move" that I can't recommend... It hasn't acted up in over 12 hours now so far.... (The tower grab incident) but... one of the other main techniques I employ, is keeping the temps to a specific "Sweetspot" which also seems to help stability, and help avoid the"Dwindledown"(also linked to mem issue)The "Spot" is also in a tight 5 degree range. Moreover, a board with troubles is likely to stay fully operational for longer periods, with a lower error rate. I probably shouldn't mention this, but I've moved my target temps even higher on every board, which is how I achieved a full 12v & 50amp on almost every vrm, on every single board, which I agree...doesn't really make sense, but it is what it is... discovered it fiddling with the Die0, or bad/unop vrm, whichever you deem it. The board with the "Dead vrm" became the best one when I accidentally reached 80C a certain temp on that board. I thought that it may be specific to that board, but decided to try to duplicate it on the other boards, on that machine, to see if the behaved the same way....they did....so I did them all. And before Edgar someone starts screaming I'm giving "Bad advice", stop right there... I'm NOT giving advice here. I'm simply telling you what I'm doing, and what the results are. I wonder if There's a way I could get ahold of some of the "Bad" boards... RMA'd ones.... to try this on.... but I doubt it...they probably repair them...lol
|
|
|
|
jxz
Newbie
Offline
Activity: 26
Merit: 0
|
|
November 25, 2013, 07:16:10 PM |
|
A question for those with November orders. When I log in to the knc site, on my common information tab, it shows my customer group as "Upgrade Modules". Is this how it reads for everything or is something on my account screwed up?
Edit to add: My order still is listed as paid status, I'm in the 106XX range.
|
|
|
|
KGBSlim
Full Member
Offline
Activity: 126
Merit: 100
CoinSlingin
|
|
November 25, 2013, 07:19:22 PM |
|
A question for those with November orders. When I log in to the knc site, on my common information tab, it shows my customer group as "Upgrade Modules". Is this how it reads for everything or is something on my account screwed up?
Edit to add: My order still is listed as paid status, I'm in the 106XX range.
Mine says the same and its order 102xx
|
Tips - 16AbA3b2GRR37t6RasVJmoxXTGuGUE1PAP
|
|
|
lemonte
|
|
November 25, 2013, 07:20:57 PM |
|
A question for those with November orders. When I log in to the knc site, on my common information tab, it shows my customer group as "Upgrade Modules". Is this how it reads for everything or is something on my account screwed up?
Edit to add: My order still is listed as paid status, I'm in the 106XX range.
Mine says the same and its order 102xx Mine says the same.
|
|
|
|
Biffa
Legendary
Offline
Activity: 3220
Merit: 1220
|
|
November 25, 2013, 07:21:22 PM |
|
COOL, I just noticed they only mention a single model...."Neptune" This will solve many of the issues selling smaller machines... and make Customer service a breeze!! I really hope that's what it means!...
Heh, I hope Neptune "IS" the smaller model
|
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
|
November 25, 2013, 07:23:52 PM |
|
COOL, I just noticed they only mention a single model...."Neptune" This will solve many of the issues selling smaller machines... and make Customer service a breeze!! I really hope that's what it means!...
Heh, I hope Neptune "IS" the smaller model Not a bad idea really......! one easy size & model to deal with.....
|
|
|
|
fragout
Legendary
Offline
Activity: 1279
Merit: 1018
|
|
November 25, 2013, 07:26:22 PM |
|
O'rama, be nice: can you give us a hint of hashrate? What kind of ASIC will it be?
There will be more info tomo. Figures have been mentioned, but I'm not going to steal anyone else's thunder. They're working hard for this, 20nm is just being released for commercial use, and we'll be one of the first companies in the world to half-step down to this process node. The figures have to make sense, Alchip flew here Friday week before last, and we are proceeding with the design. define commercial use ?
|
|
|
|
Biffa
Legendary
Offline
Activity: 3220
Merit: 1220
|
|
November 25, 2013, 07:31:23 PM |
|
A question for those with November orders. When I log in to the knc site, on my common information tab, it shows my customer group as "Upgrade Modules". Is this how it reads for everything or is something on my account screwed up?
Edit to add: My order still is listed as paid status, I'm in the 106XX range.
Mine says the same and its order 102xx Mine says the same. 102xx and "Upgrade modules" I think they changed us to that group just before they had the flashmob sale of upgrade modules. btw, I don't think I said it at the time, but just for those of you that missed out on the upgrade modules, I think they should have been limited to only Saturn and Mercury buyers, even though I didn't have either model.
|
|
|
|
Micky25
Legendary
Offline
Activity: 974
Merit: 1000
|
|
November 25, 2013, 07:34:24 PM |
|
btw, I don't think I said it at the time, but just for those of you that missed out on the upgrade modules, I think they should have been limited to only Saturn and Mercury buyers, even though I didn't have either model.
+1 on all points
|
|
|
|
Gator-hex
|
|
November 25, 2013, 07:35:43 PM |
|
No details. Coming Soon TMWow! 20nm AND early 2014! That MUST be the "incoming news" announcement from 'orama. Bingo! Great news, 20nm! Bad news 'Orders opening soon'! Bitcoinorama, in hope I get not ignored I'd like to ask the following: 1) Why is the pre-order game continuing ? Didn't they collect enough money to fund the next gen ?! 2) Do you have to pay the full price when opening the order books or is it like pay 10-20% up front and then the rest when it's close to shipping...? 3) How much will they approximately cost ? Would be very nice to get a clue before you actually have to pay.. 4) ANY advantage for the early investors who took a high risk and making KnCMiner's success possible in first place ? 5) Will there be another batch of upgrade boards for October gen ? Thank you very much in advance for your answers! **For those who want to pay in BTC, be aware here's a guy offering you a 60% return of your coins without additional costs for electricity, tranquilizers, waiting, or else...might be the better choice DONT GET TOO EXCITED!I don't think it will be a true 20nm ASIC, looking at how their 28nm compares to 55nm BitFury in speed/efficiency, I think they are using FPGA to ASIC HardCopying http://www.altera.co.uk/devices/asic/hardcopy-asics/hardcopy-v/hcv-index.jspEven if they bring out a 20nm it's doubtful it will be anymore efficient than a true 28nm ASIC so they're no further along than the competition.
|
|
|
|
Mota
Legendary
Offline
Activity: 804
Merit: 1002
|
|
November 25, 2013, 07:40:59 PM |
|
O'rama, be nice: can you give us a hint of hashrate? What kind of ASIC will it be?
There will be more info tomo. Figures have been mentioned, but I'm not going to steal anyone else's thunder. They're working hard for this, 20nm is just being released for commercial use, and we'll be one of the first companies in the world to half-step down to this process node. The figures have to make sense, Alchip flew here Friday week before last, and we are proceeding with the design. define commercial use ? Commercial use means they will stand by their illegal ToS and will try not to be held responsible for warranty issues and losses incurred by delayed production. Same old same old. And again with the preorder bullshit because if they let you make a normal order they have to deliver in time. All risks to the customers as well as the funding. That's the way to build lasting business relationships!
|
|
|
|
vesperwillow
|
|
November 25, 2013, 07:43:51 PM |
|
I don't think it will be a true 20nm ASIC, looking at how their 28nm compares to 55nm BitFury in speed/efficiency, I think they are using FPGA to ASIC HardCopying http://www.altera.co.uk/devices/asic/hardcopy-asics/hardcopy-v/hcv-index.jspEven if they bring out a 20nm it's doubtful it will be anymore efficient than a true 28nm ASIC so they're no further along than the competition. This is part of the reason I was a bit shocked with the claim, and the reason I pointed to commercial GPU's and other high-end products which favor 28 lately. My 2 concerns with this would be the issues faced with 20nm tapeout, and firmware. The mercury's I've seen seem to have no issues whatsoever, but some Saturns and most Jupiters do, so it makes me wonder if they'll be able to stabilize the next generation before shipping.
|
|
|
|
Micky25
Legendary
Offline
Activity: 974
Merit: 1000
|
|
November 25, 2013, 07:45:35 PM |
|
|
|
|
|
|