kano
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
December 19, 2013, 02:48:34 AM |
|
It depends on the pool. On some pools, for example p2pool, a stale share may still be a valid block and so if you don't submit stales, you risk missing out on BTC. Historically, there have also been pools that paid for stale shares and so you always wanted to submit them. I don't know that any current pools still do that though.
Ok, thanks. That clears it up a bit. I'm not sure how a stale share can be a valid block though. If I'm not mistaken, a stale share is the result of an "old job" from the pool. In other word, a job that was submitted after someone else had already solved the block. P2Pool has it's own share chain - on p2pool you solve high difficulty shares, not blocks. Thus a share can be stale due to another share being accepted ... even if the network BTC block hasn't changed. What this means is that any block found when mining on P2Pool will actually still be a valid BTC block until the next BTC block is found on the network - even if it is a stale share. P2Pool will send the stale block to the bitcoind in the hope it is actually a valid network block - which it will be ~95% of the time. This change to p2pool (quite a while ago) meant that you were no longer earning less mining on p2pool - however it does require EVERYONE mining on p2pool to submit stale blocks - which is done by submitting stale shares in cgminer.
|
|
|
|
dE_logics
|
|
December 19, 2013, 09:01:57 AM |
|
@ckolivas Hi! Great job with CGminer and BFS! I was wondering if you're still giving support for cgminer 3.7 which supports GPU and scrypt mining. My miner is running and showing the speed in kh/s (scrypt) but till date it has not been able to solve any block. Running CPU miner (which runs at a fraction of the pace) has solved many blocks in the mean time with the same coin. I'm solo mining many alts. I'm getting a lot of hardware errors also -- 2013-12-17 23:31:30] GPU 0 found something? [2013-12-17 23:31:30] OCL NONCE 129605 found in slot 0 [2013-12-17 23:31:30] [thread 0: 1048576 hashes, 119.0 khash/sec] [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 183595 found in slot 1 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 205151 found in slot 2 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 348676 found in slot 3 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 361138 found in slot 4 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 455330 found in slot 5 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 516499 found in slot 6 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 526673 found in slot 7 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 557550 found in slot 8 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 600413 found in slot 9 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 610612 found in slot 10 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 656621 found in slot 11 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 783660 found in slot 12 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 939194 found in slot 13 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 993888 found in slot 14 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] Discarded cloned or rolled work [2013-12-17 23:31:30] GPU 0 found something? [2013-12-17 23:31:30] OCL NONCE 129605 found in slot 0 [2013-12-17 23:31:30] [thread 0: 1048576 hashes, 119.0 khash/sec] [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 183595 found in slot 1 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 205151 found in slot 2 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 348676 found in slot 3 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 361138 found in slot 4 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 455330 found in slot 5 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 516499 found in slot 6 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 526673 found in slot 7 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 557550 found in slot 8 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 600413 found in slot 9 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 610612 found in slot 10 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 656621 found in slot 11 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 783660 found in slot 12 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 939194 found in slot 13 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] OCL NONCE 993888 found in slot 14 [2013-12-17 23:31:30] GPU0: invalid nonce - HW error [2013-12-17 23:31:30] Discarded cloned or rolled work (5s):71.89K (avg):87.38Kh/s | A:0 R:0 HW:15 WU:0.0/m [2013-12-17 23:31:39] GPU 0 found something? [2013-12-17 23:31:39] [thread 0: 1048576 hashes, 119.1 khash/sec] [2013-12-17 23:31:39] GPU 0 found something? [2013-12-17 23:31:39] [thread 0: 1048576 hashes, 119.1 khash/sec] (5s):118.3K (avg):119.1Kh/s | A:0 R:0 HW:15 WU:0.0/m [2013-12-17 23:31:39] OCL NONCE 1055154 found in slot 0 [2013-12-17 23:31:39] GPU0: invalid nonce - HW error Thanks!!
|
|
|
|
mdude77
Legendary
Offline
Activity: 1540
Merit: 1001
|
|
December 19, 2013, 10:47:06 AM |
|
@ckolivas
Hi!
Great job with CGminer and BFS!
I was wondering if you're still giving support for cgminer 3.7 which supports GPU and scrypt mining.
My miner is running and showing the speed in kh/s (scrypt) but till date it has not been able to solve any block. Running CPU miner (which runs at a fraction of the pace) has solved many blocks in the mean time with the same coin.
I'm solo mining many alts.
I'm getting a lot of hardware errors also -- Thanks!!
Check your intensity. It's probably too high. Could be faulty GPU as well. M
|
I mine at Kano's Pool because it pays the best and is completely transparent! Come join me!
|
|
|
Fletch
Full Member
Offline
Activity: 168
Merit: 100
I'll have a steak sandwich and a... steak sandwich
|
|
December 19, 2013, 10:57:24 AM |
|
P2Pool has it's own share chain - on p2pool you solve high difficulty shares, not blocks. Thus a share can be stale due to another share being accepted ... even if the network BTC block hasn't changed.
What this means is that any block found when mining on P2Pool will actually still be a valid BTC block until the next BTC block is found on the network - even if it is a stale share. P2Pool will send the stale block to the bitcoind in the hope it is actually a valid network block - which it will be ~95% of the time.
This change to p2pool (quite a while ago) meant that you were no longer earning less mining on p2pool - however it does require EVERYONE mining on p2pool to submit stale blocks - which is done by submitting stale shares in cgminer.
Thanks again. I think I got it now. Based on your description, I will continue to submit stales when I'm mining on p2pool, but enable the --no-submit-stale option for other pools.
|
|
|
|
dE_logics
|
|
December 19, 2013, 02:38:26 PM |
|
@ckolivas
Hi!
Great job with CGminer and BFS!
I was wondering if you're still giving support for cgminer 3.7 which supports GPU and scrypt mining.
My miner is running and showing the speed in kh/s (scrypt) but till date it has not been able to solve any block. Running CPU miner (which runs at a fraction of the pace) has solved many blocks in the mean time with the same coin.
I'm solo mining many alts.
I'm getting a lot of hardware errors also -- Thanks!!
Check your intensity. It's probably too high. Could be faulty GPU as well. M Reducing the intensity will still generate hardware errors . There appears to be some hardware error generated per KH, i.e. halving the KH/s will reduce the no. of errors per second to half. I still don't get any share with lower intensity.
|
|
|
|
Krak
|
|
December 19, 2013, 05:45:21 PM |
|
P2Pool has it's own share chain - on p2pool you solve high difficulty shares, not blocks. Thus a share can be stale due to another share being accepted ... even if the network BTC block hasn't changed.
What this means is that any block found when mining on P2Pool will actually still be a valid BTC block until the next BTC block is found on the network - even if it is a stale share. P2Pool will send the stale block to the bitcoind in the hope it is actually a valid network block - which it will be ~95% of the time.
This change to p2pool (quite a while ago) meant that you were no longer earning less mining on p2pool - however it does require EVERYONE mining on p2pool to submit stale blocks - which is done by submitting stale shares in cgminer.
Thanks again. I think I got it now. Based on your description, I will continue to submit stales when I'm mining on p2pool, but enable the --no-submit-stale option for other pools. There's really no harm in letting it submit stales regardless of the pool you're on. Either you'll get paid for the share or it'll be rejected.
|
BTC: 1KrakenLFEFg33A4f6xpwgv3UUoxrLPuGn
|
|
|
momchilandonov
Jr. Member
Offline
Activity: 121
Merit: 1
|
|
December 19, 2013, 10:16:54 PM |
|
Here is what I see with -n command. Problem is my R9 290 isn't monitored no matter what --gpu-platform or --gpu-map commands I make. I have the same issue if there is a HD5870 but even worse because it doesn't start if a monitor isn't plugged in. R9 290's fan is working on low rpm so I am sure the card can be damaged very easily that way in like 5-10 minutes maybe( I wonder if there is some throttling solution to this integrated in the GPU?)? I want it to be monitored by cgminer. Please tell me the right command in this scenario. Also will there be a difference if I don't attach a monitor on my integrated Intel HD Graphics GPU too? Only ethernet and power cables and autorun file . It works on --gpu-platform 1 but it doesn't show temperatures. If I change 1 to 0 cgminer shuts down or if I choose -d 1 or more... maybe I need to use a combination of gpu-map and --gpu-platform?
|
|
|
|
TurdHurdur
|
|
December 20, 2013, 12:24:44 AM |
|
When specifying a worksize of 512 on 3.7.2, the .bin file created in the program directory has w256 in the filename. Does that mean there's a hardlimit of 256 worksize? Is there a way to go beyond that?
|
|
|
|
-ck (OP)
Legendary
Offline
Activity: 4256
Merit: 1645
Ruu \o/
|
|
December 20, 2013, 12:54:14 AM |
|
I've finally bitten the bullet and killed off GPU mining from the code and it will not be in the release [..] I am making a conscious decision and taking a stance to only support bitcoin by doing this and will consider all discussions regarding alternative cryptocurrencies [and GPU mining] as offtopic from here on.
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
TurdHurdur
|
|
December 20, 2013, 12:59:04 AM |
|
I've finally bitten the bullet and killed off GPU mining from the code and it will not be in the release [..] I am making a conscious decision and taking a stance to only support bitcoin by doing this and will consider all discussions regarding alternative cryptocurrencies [and GPU mining] as offtopic from here on. Alright, can someone else provide some insight, a time machine, or a free ASIC rig for me?
|
|
|
|
-ck (OP)
Legendary
Offline
Activity: 4256
Merit: 1645
Ruu \o/
|
|
December 20, 2013, 03:35:10 AM |
|
I've finally bitten the bullet and killed off GPU mining from the code and it will not be in the release [..] I am making a conscious decision and taking a stance to only support bitcoin by doing this and will consider all discussions regarding alternative cryptocurrencies [and GPU mining] as offtopic from here on. Alright, can someone else provide some insight, a time machine, or a free ASIC rig for me? Try litecoin forums.
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
chadtn
|
|
December 20, 2013, 04:54:04 AM |
|
Here is what I see with -n command. Problem is my R9 290 isn't monitored no matter what --gpu-platform or --gpu-map commands I make. I have the same issue if there is a HD5870 but even worse because it doesn't start if a monitor isn't plugged in. R9 290's fan is working on low rpm so I am sure the card can be damaged very easily that way in like 5-10 minutes maybe( I wonder if there is some throttling solution to this integrated in the GPU?)? I want it to be monitored by cgminer. Please tell me the right command in this scenario. Also will there be a difference if I don't attach a monitor on my integrated Intel HD Graphics GPU too? Only ethernet and power cables and autorun file . It works on --gpu-platform 1 but it doesn't show temperatures. If I change 1 to 0 cgminer shuts down or if I choose -d 1 or more... maybe I need to use a combination of gpu-map and --gpu-platform? My Radeon 290's have to have at least one monitor plugged in or ADL monitoring will not work. Using the onboard video doesn't count. I have to have at least one monitor on one of the 290 cards. If the monitor turns off for any reason while it is hashing the system will eventually lock up. Works the same way in Linux and Windows. AMD/ATI needs to fix that ASAP. I just downloaded the latest WHQL drivers for Windows that were released today, but haven't tested them yet to see if it was fixed. Nothing is mentioned about it in the tiny little change log they included. What's strange is something like 40MB's has been removed from the driver package. Chad
|
|
|
|
momchilandonov
Jr. Member
Offline
Activity: 121
Merit: 1
|
|
December 20, 2013, 08:56:49 PM |
|
Here is what I see with -n command. Problem is my R9 290 isn't monitored no matter what --gpu-platform or --gpu-map commands I make. I have the same issue if there is a HD5870 but even worse because it doesn't start if a monitor isn't plugged in. R9 290's fan is working on low rpm so I am sure the card can be damaged very easily that way in like 5-10 minutes maybe( I wonder if there is some throttling solution to this integrated in the GPU?)? I want it to be monitored by cgminer. Please tell me the right command in this scenario. Also will there be a difference if I don't attach a monitor on my integrated Intel HD Graphics GPU too? Only ethernet and power cables and autorun file . It works on --gpu-platform 1 but it doesn't show temperatures. If I change 1 to 0 cgminer shuts down or if I choose -d 1 or more... maybe I need to use a combination of gpu-map and --gpu-platform? My Radeon 290's have to have at least one monitor plugged in or ADL monitoring will not work. Using the onboard video doesn't count. I have to have at least one monitor on one of the 290 cards. If the monitor turns off for any reason while it is hashing the system will eventually lock up. Works the same way in Linux and Windows. AMD/ATI needs to fix that ASAP. I just downloaded the latest WHQL drivers for Windows that were released today, but haven't tested them yet to see if it was fixed. Nothing is mentioned about it in the tiny little change log they included. What's strange is something like 40MB's has been removed from the driver package. Chad My machine is working like a charm without a monitor! Latest drivers from AMD. I didn't installed SDK additionally (comes with the driver). I don't plug in monitor! Just used my TV to install Windows 7 but I think the VGA starts without an attached monitor. Could be a motherboard issue. I am using FM2A88X Extreme 6+ which supports 3 PCI-E!!! Also a friend is using two r9 290 without a monitor attached and they both work again on Windows 7 latest drivers (could be using the beta 13.11 or something like that). Mine aren't beta! My R9 290 wasn't starting up even with monitor the first couple of days because I was trying it with a weak Fortron 350 watts but I think it has to do with the 6-8 pins VGA power cables.
|
|
|
|
x0rcist
|
|
December 21, 2013, 11:24:19 PM |
|
Hi there. I am here to report what looks to be a bug in cgminer. This bug manifests itself if you run cgminer pointed at the stratum port of the bfgminer proxy. This has been reported by numerous users, usually trying to point their KnC miner at the bfgminer proxy: https://bitcointalk.org/index.php?topic=370365.msg3970326#msg3970326To reproduce: bfgminer --stratum-port 4444 -o some.miningpool.com -u worker -p password cgminer --o 127.0.0.1:4444 -u doesnt -p matter The output will repeatedly read: Rejected a18df423 Diff 2/1 GPU 1 (H-not-zero) I worked with Luke to debug both bfgminer and cgminer and he was able to make a few changes in the cgminer source to get accepted shares. Here is a summary of what he had to say: cgminer is broken if nonce2_size is anything other than 4 (32-bit) A proxy must always reduce the nonce2_size from the upstream server BFGMiner always uses 2 (16-bit) cgminer is memcpy-ing nonce2 with a fixed 32-bit size, even if nonce2_size is smaller If you were to just memcpy the nonce2 with an adjusted size, it will fix little endian but break on big endian (to confirm the bug) I'd really like to keep this civil and hopefully get a legitimate look at this bug as I am getting numerous users reporting it as I mentioned before. Thanks Con! I'm having to exact same issue wich makes it unable to properly manage my asics in a fast way. Would be nice if this bug gets priority because after searching it looks like i'm not the only one Thanks!
|
|
|
|
-ck (OP)
Legendary
Offline
Activity: 4256
Merit: 1645
Ruu \o/
|
|
December 22, 2013, 12:01:06 PM |
|
I'm having to exact same issue wich makes it unable to properly manage my asics in a fast way. Would be nice if this bug gets priority because after searching it looks like i'm not the only one Thanks! Should be fixed in git now.
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
dant
Newbie
Offline
Activity: 7
Merit: 0
|
|
December 22, 2013, 09:24:22 PM |
|
Using cgminer to connect to bitminter. [root cgminer-2.5.0]# ./cgminer -o http://mint.bitminter.com:80 -u doesnt -p matter [2013-12-22 21:17:21] Started cgminer 2.5.0 [2013-12-22 21:17:21] Probing for an alive pool [2013-12-22 21:18:21] Pool 0 slow/down or URL or credentials invalid [2013-12-22 21:18:21] Unable to get work from pool 0 http://mint.bitminter.com:80 [2013-12-22 21:18:21] No servers were found that could be used to get work from. [2013-12-22 21:18:21] Please check the details from the list below of the servers you have input [2013-12-22 21:18:21] Most likely you have input the wrong URL, forgotten to add a port, or have not set up workers [2013-12-22 21:18:21] Pool: 0 URL: http://mint.bitminter.com:80 Any ideas? I've tried the other ports listed for bitminter too, nothing.
|
|
|
|
-ck (OP)
Legendary
Offline
Activity: 4256
Merit: 1645
Ruu \o/
|
|
December 22, 2013, 09:25:12 PM |
|
Using cgminer to connect to bitminter. [root cgminer-2.5.0]# ./cgminer -o http://mint.bitminter.com:80 -u doesnt -p matter [2013-12-22 21:17:21] Started cgminer 2.5.0 [2013-12-22 21:17:21] Probing for an alive pool [2013-12-22 21:18:21] Pool 0 slow/down or URL or credentials invalid [2013-12-22 21:18:21] Unable to get work from pool 0 http://mint.bitminter.com:80 [2013-12-22 21:18:21] No servers were found that could be used to get work from. [2013-12-22 21:18:21] Please check the details from the list below of the servers you have input [2013-12-22 21:18:21] Most likely you have input the wrong URL, forgotten to add a port, or have not set up workers [2013-12-22 21:18:21] Pool: 0 URL: http://mint.bitminter.com:80 Any ideas? I've tried the other ports listed for bitminter too, nothing. Mate, 2.5.0 is ANCIENT. It won't connect to shit worth shit.
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
dant
Newbie
Offline
Activity: 7
Merit: 0
|
|
December 22, 2013, 09:32:18 PM |
|
Using cgminer to connect to bitminter. [root cgminer-2.5.0]# ./cgminer -o http://mint.bitminter.com:80 -u doesnt -p matter [2013-12-22 21:17:21] Started cgminer 2.5.0 [2013-12-22 21:17:21] Probing for an alive pool [2013-12-22 21:18:21] Pool 0 slow/down or URL or credentials invalid [2013-12-22 21:18:21] Unable to get work from pool 0 http://mint.bitminter.com:80 [2013-12-22 21:18:21] No servers were found that could be used to get work from. [2013-12-22 21:18:21] Please check the details from the list below of the servers you have input [2013-12-22 21:18:21] Most likely you have input the wrong URL, forgotten to add a port, or have not set up workers [2013-12-22 21:18:21] Pool: 0 URL: http://mint.bitminter.com:80 Any ideas? I've tried the other ports listed for bitminter too, nothing. Mate, 2.5.0 is ANCIENT. It won't connect to shit worth shit. I need to use CPU for mining, I have machines that I'm paying for regardless and they don't have GPU's. How can I use cgminer to connect to bitminter and use CPU's?
|
|
|
|
-ck (OP)
Legendary
Offline
Activity: 4256
Merit: 1645
Ruu \o/
|
|
December 22, 2013, 09:38:39 PM |
|
I need to use CPU for mining, I have machines that I'm paying for regardless and they don't have GPU's. How can I use cgminer to connect to bitminter and use CPU's?
Cgminer hasn't mined on CPUs in ages for a reason. You'll get no support for that here. It doesn't even mine on GPUs any more either. Feel free to do some research before asking why as well.
|
Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel 2% Fee Solo mining at solo.ckpool.org -ck
|
|
|
|
|