gecox22
|
|
September 25, 2015, 02:10:01 AM |
|
Just follow the instructions, it's not a firmware, it's an update to any firmware (replaces a few files) [/quote] Pretty easy to do. Thanks for the help there. But now I have a question please see screen shot below of the first miner I updated. It is reporting I am pushing 500+ ghs. how is that possible? I didnt over clock it or change any settings at all. I just followed the instruction and did a reboot. My pool stats are reporting about 60 ghs under what my miner is reporting. about 445 currently. I am also running only running this miner with 2 power connectors instead of 4 since I haven't figured out if I need to over clock them or not. Just trying to figured out if I did something wrong or if I am really putting out that much hashing power on a s3 or is something glitched. After running for a bit it has came down alittle bit around 500 but still seems a big differnce between the two units back to back running off the same power supply. They both are now running the same lastest firmware from bitmain. and only only has been updated to the kano cgminer-binaries.
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
September 25, 2015, 03:05:28 AM |
|
It's simply an issue of variance. Also your time frame is WAY too small to make any judgement of the performance. The antminer replies with work at the difficulty it's told to mine at. It doesn't report work done other than the shares it finds. I guess my difficulty setting is more accurate than bitmain's - thus cgminer shows higher variance. https://github.com/ckolivas/cgminer/blob/master/driver-bitmain.c#L592It means that the miner is doing less nonce checking work for exactly the same amount of valid shares. The Paid figure is the shares accepted by the pool over the time you have been submitting shares, so is exactly what a (non-solo) pool will pay you. It can take a few hours to average out to the actual value. Yeah I've had a 440GHs show over 500GHs for over an hour. Again, it's just variance.
|
|
|
|
gecox22
|
|
September 25, 2015, 03:45:40 AM |
|
It's simply an issue of variance. Also your time frame is WAY too small to make any judgement of the performance. The antminer replies with work at the difficulty it's told to mine at. It doesn't report work done other than the shares it finds. I guess my difficulty setting is more accurate than bitmain's - thus cgminer shows higher variance. https://github.com/ckolivas/cgminer/blob/master/driver-bitmain.c#L592It means that the miner is doing less nonce checking work for exactly the same amount of valid shares. The Paid figure is the shares accepted by the pool over the time you have been submitting shares, so is exactly what a (non-solo) pool will pay you. It can take a few hours to average out to the actual value. Yeah I've had a 440GHs show over 500GHs for over an hour. Again, it's just variance. Sounds good and thank you for the explanation. Just never seen that much speed before kinda freaked me out. I trouble shot my second miner for a bit to find out why it was having so many hardware errors. I updated the firmware to match the first one and the loaded the kano binary, and found at a freq of 225 I am getting a x on one of the chips and causes the miner to really slow down. So I took it offline did some minor detailing to it. Looking for oil residue or anything abnormal put it back together stuck it back in run and still have the x . I changed the freq from 225 to 218.75 and it seems to have fixed the issue just lost about 40 ghs. So now I am wondering if my solder connection on 1200 watt power supply are good enough or not. I figured a 1200watt should power both no problem. Maybe I just have a bad solder join or something. I will trouble shoot alittle tomorrow after work and maybe put the funky miner on it own power supply and see how it does. Oh and after backing off the freq the hardware errors back off as well. Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
|
|
|
|
PPOC
|
|
September 25, 2015, 03:51:53 AM |
|
It's simply an issue of variance. Also your time frame is WAY too small to make any judgement of the performance. The antminer replies with work at the difficulty it's told to mine at. It doesn't report work done other than the shares it finds. I guess my difficulty setting is more accurate than bitmain's - thus cgminer shows higher variance. https://github.com/ckolivas/cgminer/blob/master/driver-bitmain.c#L592It means that the miner is doing less nonce checking work for exactly the same amount of valid shares. The Paid figure is the shares accepted by the pool over the time you have been submitting shares, so is exactly what a (non-solo) pool will pay you. It can take a few hours to average out to the actual value. Yeah I've had a 440GHs show over 500GHs for over an hour. Again, it's just variance. Sounds good and thank you for the explanation. Just never seen that much speed before kinda freaked me out. I trouble shot my second miner for a bit to find out why it was having so many hardware errors. I updated the firmware to match the first one and the loaded the kano binary, and found at a freq of 225 I am getting a x on one of the chips and causes the miner to really slow down. So I took it offline did some minor detailing to it. Looking for oil residue or anything abnormal put it back together stuck it back in run and still have the x . I changed the freq from 225 to 218.75 and it seems to have fixed the issue just lost about 40 ghs. So now I am wondering if my solder connection on 1200 watt power supply are good enough or not. I figured a 1200watt should power both no problem. Maybe I just have a bad solder join or something. I will trouble shoot alittle tomorrow after work and maybe put the funky miner on it own power supply and see how it does. Oh and after backing off the freq the hardware errors back off as well. Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday. I am voting that this block will be 168% diff block. We should start a "cast your guess" and closest gets a tip from those that submitted a guess. Keep things fun around here.
|
BTC: 1Bo6YsPeHCrVRygHLJg9BwHeaLSQpppcJi "Lost coins only make everyone else’s coins worth slightly more. Think of it as a donation to everyone."
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
September 25, 2015, 03:55:11 AM |
|
... Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
Fortunately, no matter how good or bad we did yesterday, it will have no effect on the next block. The 'amusing' side effect of understanding that is to also understand that at any point in time, the next block is expected to be 100% Diff from 'now'. Even though we've spent over 90% Diff so far, looking for our next block, it doesn't actually affect, in any way whatsoever, when we will find it.
|
|
|
|
gecox22
|
|
September 25, 2015, 04:16:26 AM |
|
... Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
Fortunately, no matter how good or bad we did yesterday, it will have no effect on the next block. The 'amusing' side effect of understanding that is to also understand that at any point in time, the next block is expected to be 100% Diff from 'now'. Even though we've spent over 90% Diff so far, looking for our next block, it doesn't actually affect, in any way whatsoever, when we will find it. You are correct. I was just speaking in general Since we have such a good day. This block will take for ever. I know how fast we find a block does not account for how long the next one will take. We could find 5 block all in green back to back and the next block would still have just as much chance a the last 5 blocks to be green red or black. It is all a lotto crap shoot. Thus why people mining on solo can still find a block with a s1 or a s3. The likely hood is slim but there is still a chance.
|
|
|
|
|
gecox22
|
|
September 25, 2015, 10:12:30 AM |
|
This is running all four power connections correct.
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
September 25, 2015, 10:35:37 AM |
|
Well, no software changes are necessary with my version, I've put a large range of clock speeds in the web page. If any of those pages have hardware voltage change info, for S3s that ignore the voltage, that may be helpful.
|
|
|
|
Jake36
|
|
September 25, 2015, 12:12:17 PM |
|
This is running all four power connections correct. Yes, all 4 power connectors plugged in.
|
|
|
|
carlosmnk
|
|
September 25, 2015, 04:45:45 PM |
|
Well, no software changes are necessary with my version, I've put a large range of clock speeds in the web page. If any of those pages have hardware voltage change info, for S3s that ignore the voltage, that may be helpful. Is there any versión for the Antminer S5??
|
Played with: USB RedFury - BlackArrows Prospero X1/X3 - Butterfly Monarch- Spondoolies SP20E - Avalon 6 - Antminer U3/S3/S3+/C1/S5/S7 Batches 3-7-8 - Sfards SF100 - Innosilicon A2 Terminator - Alcheminer 96/256 - KNC Titan - Etherum Rigs
|
|
|
VirosaGITS
Legendary
Offline
Activity: 1302
Merit: 1068
|
|
September 25, 2015, 05:14:27 PM |
|
Well, no software changes are necessary with my version, I've put a large range of clock speeds in the web page. If any of those pages have hardware voltage change info, for S3s that ignore the voltage, that may be helpful. Is there any versión for the Antminer S5?? Not that i know of, initially S5s were not undervoltable, but some users found out the latest PCBs, v1.91 that is, can run undervolted. A few oscilliator were added from the v1.90 version to the v1.91 version. This however does not mean you can change settings or pencil mod it, the design does not seem to allow this that we can tell of, the way to run it undervolted it is using an adjustable PSU, ideally within the 9.5-11.5 volt range.
|
|
|
|
Mikestang
Legendary
Offline
Activity: 1274
Merit: 1000
|
|
September 25, 2015, 09:30:13 PM |
|
Well, no software changes are necessary with my version, I've put a large range of clock speeds in the web page. If any of those pages have hardware voltage change info, for S3s that ignore the voltage, that may be helpful. Right, I neglected to mention all those links were from before Kano wrote updated cgminer software for the S3 that makes overclocking much easier, but they still make good reading. And yes, if you're going to OC then you should have all 4 pci-e connectors plugged into a psu. I use an 800w gold rated atx psu, it's overkill for 1 S3 but gives me room to power other stuff as well.
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
September 26, 2015, 04:07:01 AM |
|
...
Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
I am voting that this block will be 168% diff block. We should start a "cast your guess" and closest gets a tip from those that submitted a guess. Keep things fun around here. Close, 160%, but more a dark blood red than a bright red as was suggested
|
|
|
|
|
PPOC
|
|
September 26, 2015, 04:09:50 AM |
|
...
Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
I am voting that this block will be 168% diff block. We should start a "cast your guess" and closest gets a tip from those that submitted a guess. Keep things fun around here. Close 160% but more a dark blood red than a bright red as was suggested Just saw that, hey not bad only 8% off. To bad others did not cast a vote. Think I would have won that round. My next prediction is 88% diff. Well either 88 or 37%, but I'm leaning closer to 88 but would like 37.
|
BTC: 1Bo6YsPeHCrVRygHLJg9BwHeaLSQpppcJi "Lost coins only make everyone else’s coins worth slightly more. Think of it as a donation to everyone."
|
|
|
markBG
Newbie
Offline
Activity: 29
Merit: 0
|
|
September 26, 2015, 07:12:33 AM |
|
How can know my reward from block which is not matured when I mining with BTC address
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
September 26, 2015, 10:46:33 AM |
|
How can know my reward from block which is not matured when I mining with BTC address
You can't The information available when mining with an address is not a lot more than on the solo pool. The only extra you can see is only what anyone can see on the web site without logging in.
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
September 26, 2015, 02:32:16 PM |
|
...
Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
I am voting that this block will be 168% diff block. We should start a "cast your guess" and closest gets a tip from those that submitted a guess. Keep things fun around here. Close 160% but more a dark blood red than a bright red as was suggested Just saw that, hey not bad only 8% off. To bad others did not cast a vote. Think I would have won that round. My next prediction is 88% diff. Well either 88 or 37%, but I'm leaning closer to 88 but would like 37. 32% - close
|
|
|
|
gecox22
|
|
September 26, 2015, 06:43:13 PM |
|
...
Now lets start finding some more blocks. I bet this one is going to be a bright red. Since we had such good luck yesterday.
I am voting that this block will be 168% diff block. We should start a "cast your guess" and closest gets a tip from those that submitted a guess. Keep things fun around here. Close, 160%, but more a dark blood red than a bright red as was suggested I am happy to see it wasn't a bright red.. Humm I say this block will be black around 80%...
|
|
|
|
|