 Show Posts
|
|
Pages: « 1 [2] 3 »
|
My GTX 1060 6G got to 387h/s with i7-2700k @4.6Ghz in Linux Mint 18.3 But it doesn't get near 360h/s when i use Ubuntu 18.04 !  By the way, in the screenshot you can see that the miner use only 1 thread, it does not profit from all cores  i think it's not optimized well for multi thread. PS: Mint use driver 384.113 with Cuda 8 Ubuntu use driver 390.59 with Cuda 9 have you tried the 2700k on stock clocks? or more cards? id be intrested to know how an older i7 scales with 4 cards |
|
|
|
|
@gameboy366
All 4x 1060s on risers.
When I run the miner with -E 1,2,3 so it only runs on one card I get 250hs on card 0... however I get 140hs when I run on Gpu 4 . Which is bizzar .
When running 1 x 1060 card on i7 in x16 slot I got 350hs.
|
|
|
|
I think this is possible only if all GPU's are installed into x16 slots, but I have not heard about the existence of motherboards with 6x PCIE x16 slots. The best result I saw using I7-8700K and 1x1060 was 500KH @PCIEx16, 420KH @PCIEx4 and 230KH @PCIEx1. I7-8700K. 1060@PCIEx16 + 1060@PCIEx4: 0-509.1 1-412.5 Total: 921.6 H/s 2018/05/30 11:36:13 0 - Yay, solution found! G3258. 1070@PCIEx16 + 1060@PCIEx1: 0-492.3 1-223.7 Total: 716.0 H/s 2018/05/30 13:37:15 1 - Yay, solution found! I7-2600: 1060@PCIEx16 + 1060@PCIEx1 +1060@PCIEx4 0-350.3 1-235.5 2-306.7 Total: 892.5 H/s 2018/05/30 13:38:43 0 - Yay, solution found! Can you make test with I7-8700K and 1070@PCIEx16? Once the miner is optimised in the next few months it will be able to run at full speed on the GPU with low cpu load. It's just the way the miner is currently written thats making it cpu / pci-e bus bound. If you have the hardware and you don't mind swapping the cpu on then it makes sense to do a swap and mine at the best hash rate. It's a combination of Cpu speed(DDR3 ram cpus take a big hit) 80% x pci-e width 15% x 5%GPU. So with 2x GTX 1060 on a i7 you get more hash than 4x1060's on a i5 2500s . ~700h/s More than 3 cards and the performance gains are reduces too.. I'm getting about 8 btm a day on i5 2500s with 1060 x 4. More profitable than most on crypto coinz. So I have the option of getting a i72600 (max supported by Mining rig) or using two cards on 16x and 8x mode on a i7 3770. To get 700h/s. What's the best watt/hash? As I doubt my i7 is very efficient Cheers |
|
|
|
Hello guys, after other days of tests here are my results:
Today i upgraded the miner to the v2 , this version seems to use all resources in just one instance (so no need to run more than one),
My initally setup was of 3x gtx 1060 and 1x gtx 1070, doing some tests i have seen that the 1070 was not working as expected so i removed it and i put a 1050 NON ti.
The hashrate was around 700 h/s to 1000 h/s with 3x 1060 and 1x 1050 (non-ti) and g4440 as cpu
I changed my g4440 to i5 6400 and hashrate is now stable on 1000 h/s:
The 1050 (non ti) is doing 230 h/s (i was like OMG)
First 1060 3 gb is doing around 320 h/s
Second 1060 3 gb is doing around 230 h/s
Third 1060 3 gb is doing around 200 h/s
I tried to mount the 1070 but global hashrate got compromised so i removed it again (maybe 1070 is not good atm with this miner), however in single test (only 1070 mounted), it was doing 355 h/s
I run other single tests (only 1 gpu connected), here are the results:
1050 (non ti) : 255 h/s
1060 3 gb : 340 h/s
1070 355 h/s
would be fun to compile a list of GPU hashrates w/ the system their on included. If others want to submit date like gregory did above ^ 4 x 1060 GPUs 700h/s (To note: my hash rate is restricted by the CPU i5 2500s, I tried an i7 3770 yielded 330h/s for 1 GPU with 13% CPU usage) The miner does not use the GPU's equally i have one 1060 6gb gpu at 60watts doing 230 h/s and another one at 55watts doing 151h/s  |
|
|
|
So you mean 5gpus running two instances is the best choice in your case?
I read the whole thread and see that 1050ti is the best p/p vga for this bytom mining, right?
Correct 5gpus and 2 instances was best case scenario for me. 1050ti and 1060 3gb are the best as they are both cheap and as the miner is not optimised the big gpus don't have any hashrate advantage. I think with a professional miner, this problem will be solved. Just the news I needed to start my day. Thanks. B3 will go down in history as the ASIC killed by the smallest of gpus. Let's mine the hell out of this coin before Bitmain releases the real deal. The B3 is not an asic.. hence why its not leagues more powerful than the gpus. It's a deep learning chip. And they can upgrade the firmware. The latest b3 firmware upgrades it from 750 h/s to 1000h/s. I think a really well written software could blast the b3 away.. if you were cynical you could say this nvidia miner is purposely a non optimised no we released to get people intrested but not effect the bitmain sales. And when a good nvidia miner is released they will release the b3+ |
|
|
|
Latest mining stats summary Antminer b3 https://ibb.co/nayeFoGTX 1060 rig x4. At 250w with i5 quad https://ibb.co/hwwwT8Considering the pc rig takes less juice I would say they are pretty close in total hash rate per watt. |
|
|
|
I just swapped out a pentium G620 2 core on a 4 x 1060 rig i was getting a jumpy av 335h/s both cores around 95%
ive put in a i5 quad core 2400s and i get a much more stable 380h/s all 4 cores at 71%
i tried two threads but hash rate went down to 135h/s on each thread and cpu was 100% on all cores.
The -I intensity doesn't seem to be the best thing in the world, defiantly far too cpu bound for a gpu miner. but its v1 prob so its expected.
I suggest you to try three instance of ./miner,look my reply above Sorry I meant not 2 threads but I meant two instances. I logged into Linux as miner and ran the miner. (All cores at 70%.) Then fired up another ssh logged in as root and ran the miner. And the sum of the two hash rates was less than just 1. I don't know how yours can be so much better with similar systems? I'm running on smos. And killed all the processes. Is the i5 2500s just not as powerful? It has 4 real cores I don't know the exact reason, i will leave you my setup: System: HiveOs Ram : 4 gb(was not enough so i created another 4 gb of swap) To start the miner i use this script: https://pastebin.com/gR2e1bp5 (i created a file called start.sh) Every start.sh has my wallet and the worker number, example: bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.001 bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.002 bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.003 I think i could do more hash if only i had a quad core. Oc: +150 clock Hmm weird why at one instance I'm at 4 cores at 70%. Do you remember your cpu usage with just one instance. Would me running 1 instance as user miner. And an other as root cause a problem? Surely it just multitasks the same. No matter how the instances are run I was getting 120% cpu with one process (200% limit because cpu is dual core), with three process i'm getting 70% - 70% - 60& (again, 200% limit), so i was planning to get a quad core (i3 or i5) to give each process the max requested (120% in my case). EDIT: i tried with 6 core cpu and was using the same 100% of cores, so i think it's a combined cpu-gpu Sorry to labour the point but I was at 70%-70%-70%-70% on one instance with 4 x1060 rock steady 380h/s Adding an extra Instance yielded 100% across 4 cores and 135h/s on each core. Thus a drop in gross hash My fault, this time i was talking about cpu. 70% was referred to the cpu. Btw i ordered an i5 and should be delivered next week.. i will update my results Regarding you, i think your main reason why the first instance lose power is due cpu. Which cpu are you using? and or curiosity, how much ram? Lol I meant on my i5 I got 4 x 70% cpu usage on its 4 cpu cores with 1 instance of the miner. When i went to 2 instances total cpu usage went to 4 x100%. I.e full cpu maxed out on two instances with no increase in hash. 4gb syatem ram. Simple miner on usb. I have an i7 I could pop in and try |
|
|
|
I just swapped out a pentium G620 2 core on a 4 x 1060 rig i was getting a jumpy av 335h/s both cores around 95%
ive put in a i5 quad core 2400s and i get a much more stable 380h/s all 4 cores at 71%
i tried two threads but hash rate went down to 135h/s on each thread and cpu was 100% on all cores.
The -I intensity doesn't seem to be the best thing in the world, defiantly far too cpu bound for a gpu miner. but its v1 prob so its expected.
I suggest you to try three instance of ./miner,look my reply above Sorry I meant not 2 threads but I meant two instances. I logged into Linux as miner and ran the miner. (All cores at 70%.) Then fired up another ssh logged in as root and ran the miner. And the sum of the two hash rates was less than just 1. I don't know how yours can be so much better with similar systems? I'm running on smos. And killed all the processes. Is the i5 2500s just not as powerful? It has 4 real cores I don't know the exact reason, i will leave you my setup: System: HiveOs Ram : 4 gb(was not enough so i created another 4 gb of swap) To start the miner i use this script: https://pastebin.com/gR2e1bp5 (i created a file called start.sh) Every start.sh has my wallet and the worker number, example: bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.001 bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.002 bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.003 I think i could do more hash if only i had a quad core. Oc: +150 clock Hmm weird why at one instance I'm at 4 cores at 70%. Do you remember your cpu usage with just one instance. Would me running 1 instance as user miner. And an other as root cause a problem? Surely it just multitasks the same. No matter how the instances are run I was getting 120% cpu with one process (200% limit because cpu is dual core), with three process i'm getting 70% - 70% - 60& (again, 200% limit), so i was planning to get a quad core (i3 or i5) to give each process the max requested (120% in my case). EDIT: i tried with 6 core cpu and was using the same 100% of cores, so i think it's a combined cpu-gpu Sorry to labour the point but I was at 70%-70%-70%-70% on one instance with 4 x1060 rock steady 380h/s Adding an extra Instance yielded 100% across 4 cores and 135h/s on each core. Thus a drop in gross hash |
|
|
|
I just swapped out a pentium G620 2 core on a 4 x 1060 rig i was getting a jumpy av 335h/s both cores around 95%
ive put in a i5 quad core 2400s and i get a much more stable 380h/s all 4 cores at 71%
i tried two threads but hash rate went down to 135h/s on each thread and cpu was 100% on all cores.
The -I intensity doesn't seem to be the best thing in the world, defiantly far too cpu bound for a gpu miner. but its v1 prob so its expected.
I suggest you to try three instance of ./miner,look my reply above Sorry I meant not 2 threads but I meant two instances. I logged into Linux as miner and ran the miner. (All cores at 70%.) Then fired up another ssh logged in as root and ran the miner. And the sum of the two hash rates was less than just 1. I don't know how yours can be so much better with similar systems? I'm running on smos. And killed all the processes. Is the i5 2500s just not as powerful? It has 4 real cores I don't know the exact reason, i will leave you my setup: System: HiveOs Ram : 4 gb(was not enough so i created another 4 gb of swap) To start the miner i use this script: https://pastebin.com/gR2e1bp5 (i created a file called start.sh) Every start.sh has my wallet and the worker number, example: bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.001 bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.002 bm1qnsk9rl45s4kps6mm97393gdjr5hslf4hvu9ru2.003 I think i could do more hash if only i had a quad core. Oc: +150 clock Hmm weird why at one instance I'm at 4 cores at 70%. Do you remember your cpu usage with just one instance. Would me running 1 instance as user miner. And an other as root cause a problem? Surely it just multitasks the same. No matter how the instances are run |
|
|
|
I just swapped out a pentium G620 2 core on a 4 x 1060 rig i was getting a jumpy av 335h/s both cores around 95%
ive put in a i5 quad core 2400s and i get a much more stable 380h/s all 4 cores at 71%
i tried two threads but hash rate went down to 135h/s on each thread and cpu was 100% on all cores.
The -I intensity doesn't seem to be the best thing in the world, defiantly far too cpu bound for a gpu miner. but its v1 prob so its expected.
I suggest you to try three instance of ./miner,look my reply above Sorry I meant not 2 threads but I meant two instances. I logged into Linux as miner and ran the miner. (All cores at 70%.) Then fired up another ssh logged in as root and ran the miner. And the sum of the two hash rates was less than just 1. I don't know how yours can be so much better with similar systems? I'm running on smos. And killed all the processes. Is the i5 2500s just not as powerful? It has 4 real cores |
|
|
|
|
I just swapped out a pentium G620 2 core on a 4 x 1060 rig i was getting a jumpy av 335h/s both cores around 95%
ive put in a i5 quad core 2400s and i get a much more stable 380h/s all 4 cores at 71%
i tried two threads but hash rate went down to 135h/s on each thread and cpu was 100% on all cores.
The -I intensity doesn't seem to be the best thing in the world, defiantly far too cpu bound for a gpu miner. but its v1 prob so its expected.
|
|
|
|
Makes me laugh how people put "Crack" and also "fork from an asic" lol. The instructions on how to create the miner are detailed in the white paper, its like a builder building a house from an architectures plans.. Also you can't fork from an asic because you'd have to physically change the asic chip to do the new algorithm  , whereas to change algo on a GPU is a relatively quick software mod. anyway rant over, happy mining  |
|
|
|
I have seen some people running 2 sessions of this on the same machine essentially doubling the hash rate.
does anyone know how to do that?
you need to fuck with the script a bit, ive pm'd you some instructions good luck! (as buddy said above tho, its not very optimized, not really profitable right now  ) Can you PM them me too please? |
|
|
|
Any updates on this?
Update You don't have to register on uupool to get paid out. They pay out a minimum of 1 btm The miner is poorly optimised im guessing 4x 1060 could match the speed of the antminer if it was optimised as well as say x17 on ccminer. I don't have time to code the miner but there are people probably already mining on better versions. And we just get the previous versions for now. |
|
|
|
|
could you create a dns mask on ubuntu to redirect btm.uupool.cn:9220 to stratum-btm.antpool.com:6666
not sure how you would change the port?
The easiest way to come about this is properly installing dnsmasq (which is a DNS cacheing server) then in the folder /etc/dnsmasq.d add a file for each domain you want to redirect.
For instance this is the contents of /etc/dnsmasq.d/hotmail.com on my system:
address=/hotmail.com/127.0.0.1
address=/www.hotmail.com/127.0.0.1
If I then ask the DNS server for the IP of hotmail.com i get 127.0.0.1 like this:
llp@cerberus:~$ nslookup hotmail.com
Server: 192.168.0.10
Address: 192.168.0.10#53
Name: hotmail.com
Address: 127.0.0.1
llp@cerberus:~$
There are a bit more options you can set, they can all be found described in /etc/dnsmasq.conf
|
|
|
|
something changed in hive it was working in previous versions, now it wont prob that ninja "0.52" update im too lazy to downgrade to see if "0.5" or earlier still works but it might be worth a try i switched to vm  cant make it utilize more than 40% of my 1080ti's tho you can use antpool, just use your desktop wallet address, seems to work for me (i couldnt recieve the SMS to register @uupool) How did you change to the antpool stratum? I changed the run.sh to include -url stratum+tcp://stratum-btm.antpool.com:6666 but it didnt work |
|
|
|
|
Are you getting paid from UU pool?
[/quote]
it says paid but how to check, the wallet showing nothing on block explorer,
the link on offcial website, is not even clickable, can you provide me link of block explorer so ican check status?
i can check thro personal wallet too but its taking long to sync.. and its eating my hashrate..
[/quote]
you can download about 90% of the block chain off the bytom website and paste it into your wallet appdata directory to speed the sync up
use tunnel bear free edition and VPN to hongkong you get a faster experience
-edit
now getting this error
root@simpleminer:~# tail -fn 1 /var/tmp/miner.log
0-62.9 1-73.1 2-75.8 3-78.0 Total: 289.8 H/s 2018/05/19 13:44:55 0 - Yay, solution found!
0-72.4 1-71.8 2-70.9 3-68.7 Total: 283.8 H/s 2018/05/19 13:45:05 0 - Error submitting solution - Timeout
0-77.7 1-82.1 2-83.9 3-79.3 Total: 323.0 H/s 2018/05/19 13:47:54 0 - Yay, solution found!
0-79.7 1-81.4 2-82.3 3-77.4 Total: 320.8 H/s 2018/05/19 13:48:04 0 - Error submitting solution - Timeout
0-82.9 1-76.4 2-85.0 3-82.8 Total: 327.0 H/s
[signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0x5cace2]
goroutine 100 [running]:
github.com/leifjacky/btmgominer/clients/stratum.(*Client).Close(0x0)
/data/go/src/github.com/leifjacky/btmgominer/clients/stratum/stratum.go:94 +0x22
github.com/leifjacky/btmgominer/algorithms/bytom.(*StratumClient).Start(0xc420454500)
/data/go/src/github.com/leifjacky/btmgominer/algorithms/bytom/siastratum.go:98 +0x7aa
created by github.com/leifjacky/btmgominer/algorithms/bytom.(*StratumClient).Start.func1
/data/go/src/github.com/leifjacky/btmgominer/algorithms/bytom/siastratum.go:75 +0x104
|
|
|
|
Any way to find the setup for gpu mining this? I would love to test it out.
You can load the miner on a linux box and run it. I converted one of my SMOS miners over to it. The link to http://uupool.cn/course/btm is not working - can you load it on some other site for download to help us out? Thank you in advance The site seems to be working now - anything special to do for installing the miner ... youtube video you've seen in the past (obviously not this miner but a how to .. otherwise I can check .. just hoping you had a better resource than random looking around. I downloaded and extracted the miner on (smos/Ubuntu) wget http://7fvi19.com1.z0.glb.clouddn.com/btm-miner-1.0.tar.bz2Changed the address.txt file to my wallet. I killed the miners that were running. (SMOS Only) Ran the miner. Watch the performance 'tail -fn 1 /var/tmp/miner.log' I had some help with killing the miners. I had to create a bash script to do that. If you are running SMOS then let me know and I will show you how. Im in the process of dl smos now Fairly new to Linux - How do you kill miners with a bash script? (I assume is some sort of startup.sh in some folder?) - I currently have HiveOS guess no one has tested on it yet? - Once you have the kill scrip, how do you run btm-miner from the web interface? or are you doing it all on the actual mining box? So I used SMOS just because it was convenient. I had to create a bash script to kill the miners and keep them dead. Then it's just a pure manual process. The web interface is pretty much dead except I can tell if the rigs are on. Create the file touch killminers Edit the file sudoedit killminers Paste this text and save. #!/bin/bash killall xterm -9 killall screen -9 screen -wipe ps -efw | grep CRON | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep fanspeed | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep watchdog | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep update_status | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep emergency | grep -v grep | awk '{print $2}' | xargs kill Add execute to the file chmod +x killminers Execute the script ./killminers If you have monitor on the rig then you will see some errors on the screen. I was logged in as root for all of this so I logged as 'miner' and ran 'screen -wipe' and that cleared it up. Are you getting paid from UU pool? |
|
|
|
Any way to find the setup for gpu mining this? I would love to test it out.
You can load the miner on a linux box and run it. I converted one of my SMOS miners over to it. The link to http://uupool.cn/course/btm is not working - can you load it on some other site for download to help us out? Thank you in advance The site seems to be working now - anything special to do for installing the miner ... youtube video you've seen in the past (obviously not this miner but a how to .. otherwise I can check .. just hoping you had a better resource than random looking around. I downloaded and extracted the miner on (smos/Ubuntu) wget http://7fvi19.com1.z0.glb.clouddn.com/btm-miner-1.0.tar.bz2Changed the address.txt file to my wallet. I killed the miners that were running. (SMOS Only) Ran the miner. Watch the performance 'tail -fn 1 /var/tmp/miner.log' I had some help with killing the miners. I had to create a bash script to do that. If you are running SMOS then let me know and I will show you how. Im in the process of dl smos now Fairly new to Linux - How do you kill miners with a bash script? (I assume is some sort of startup.sh in some folder?) - I currently have HiveOS guess no one has tested on it yet? - Once you have the kill scrip, how do you run btm-miner from the web interface? or are you doing it all on the actual mining box? So I used SMOS just because it was convenient. I had to create a bash script to kill the miners and keep them dead. Then it's just a pure manual process. The web interface is pretty much dead except I can tell if the rigs are on. Create the file touch killminers Edit the file sudoedit killminers Paste this text and save. #!/bin/bash killall xterm -9 killall screen -9 screen -wipe ps -efw | grep CRON | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep fanspeed | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep watchdog | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep update_status | grep -v grep | awk '{print $2}' | xargs kill ps -efw | grep emergency | grep -v grep | awk '{print $2}' | xargs kill Add execute to the file chmod +x killminers Execute the script ./killminers If you have monitor on the rig then you will see some errors on the screen. I was logged in as root for all of this so I logged as 'miner' and ran 'screen -wipe' and that cleared it up. can you do all this this by ssh? using miner@192.168.0.12 password: miner |
|
|
|
Got it working on my 6x1060, hashrate isn't great at 500+ h/s. Temps are really low though, looks like the miner can be optimized more.
What is your intensity set to? Had it at the default of 8, now I'm trying to increase it see if it helps. Are you using simplemingos? I only get this output: Driver = 390, CARD COUNT=6, WK=1x21 Any clues? tail -fn 1 /var/tmp/miner.log read the thread.. then it start showing you the data.. my work flow to test it was to ctrl + C to stop simple miner, install the miner. nano address.txt, change the wallet address, then run the ./run.sh tail -fn 1 /var/tmp/miner.log then wait about 15 mins before upool updated. |
|
|
|
|
