Bitcoin Forum
November 10, 2024, 09:10:59 AM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 [2] 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 ... 106 »
  Print  
Author Topic: T-Rex 0.26.8 ETHW, ETC (+dual mining) GPU miner  (Read 174959 times)
nitrobg
Member
**
Offline Offline

Activity: 413
Merit: 17


View Profile
June 10, 2018, 08:26:28 AM
 #21

Many thanks for reporting this. Unfortunately I couldn't replicate the crash you've experienced, tried it on 1050, 1070 and 1070ti and it all worked fine with -i 17. Was it 1060 or 1080ti that crashed?
The API, along with the windows version, should be available in a couple of weeks.
Also, I was wondering if you could benchmark phi1612 on your 1080ti for our stats?

Maybe it is a too high overclock, I really don't know as my rig is mixed with many different cards. tpruvot's ccminer worked fine, though. I've lowered to overclock on all cards that ran past 2050MHz, we'll see if crashes are still there.
The default ccminer api would be lovely, so the hashrate monitoring would work on Hive.

Lyra2z doesn't show improvement over high intensities, but it works up to -i 26 on 1080 Ti.
Phi however gets a 4% boost from a higher intensity. Testing on 1080 Ti @ 200W, -1000 memory, +120 core.
-i 20 (default): 37.8-38.1MH/s
-i 27: 39.5MH/s
trexminer (OP)
Member
**
Offline Offline

Activity: 283
Merit: 63


View Profile
June 10, 2018, 09:22:27 AM
 #22

Added Tribus algo. See download links in the first post.
trexminer (OP)
Member
**
Offline Offline

Activity: 283
Merit: 63


View Profile
June 10, 2018, 09:33:36 AM
 #23

Maybe it is a too high overclock, I really don't know as my rig is mixed with many different cards. tpruvot's ccminer worked fine, though. I've lowered to overclock on all cards that ran past 2050MHz, we'll see if crashes are still there.
The default ccminer api would be lovely, so the hashrate monitoring would work on Hive.

Lyra2z doesn't show improvement over high intensities, but it works up to -i 26 on 1080 Ti.
Phi however gets a 4% boost from a higher intensity. Testing on 1080 Ti @ 200W, -1000 memory, +120 core.
-i 20 (default): 37.8-38.1MH/s
-i 27: 39.5MH/s

As I don't own a 1080 Ti, this is very valuable info, once again thanks for that. I'll see what I can do about the API.
I know it might be a lot to ask but could you also benchmark Tribus algo with your 1080 Ti that I just added?
nitrobg
Member
**
Offline Offline

Activity: 413
Merit: 17


View Profile
June 10, 2018, 10:02:32 AM
 #24

As I don't own a 1080 Ti, this is very valuable info, once again thanks for that. I'll see what I can do about the API.
I know it might be a lot to ask but could you also benchmark Tribus algo with your 1080 Ti that I just added?


Lovely, thanks. Can you somehow link the libraries into the binary?
Tribus - Default -i 23: 116MH/s. -i 27: 120MH/s.
mtx_demon
Full Member
***
Offline Offline

Activity: 273
Merit: 100


View Profile
June 10, 2018, 04:18:52 PM
 #25



[2018-06-10 09:17:55] Starting on stratum+tcp://us-west01.miningrigrentals.com:50243
[2018-06-10 09:17:55] GPU monitoring enabled.
[2018-06-10 09:17:55] 1 miner thread started, using 'lyra2z' algorithm.
[2018-06-10 09:17:56] New difficulty: 1.2 (0.00469)
[2018-06-10 09:17:56] lyra2z block 88917, diff 84.704
[2018-06-10 09:17:57] GPU#0: Intensity set to 17, 131072 cuda threads

[2018-06-10 09:18:02] [FAIL] [A/R]: 0/1, diff: 0.773, 1958.08 kH/s
[2018-06-10 09:18:02] Reject reason: Worker is not Authorized, e8.
[2018-06-10 09:18:02] Stratum connection interrupted
[2018-06-10 09:18:04] [FAIL] [A/R]: 0/2, diff: 0.006, 2533.79 kH/s
[2018-06-10 09:18:04] Reject reason: Worker is not Authorized, e8.
[2018-06-10 09:18:04] Stratum connection interrupted
[2018-06-10 09:18:05] lyra2z block 88917, diff 84.704
[2018-06-10 09:18:07] [FAIL] [A/R]: 0/3, diff: 0.026, 2819.10 kH/s
[2018-06-10 09:18:07] Reject reason: Worker is not Authorized, e8.
[2018-06-10 09:18:07] Stratum connection interrupted
[2018-06-10 09:18:08] lyra2z block 88917, diff 84.704
[2018-06-10 09:18:11] [FAIL] [A/R]: 0/4, diff: 0.005, 2992.80 kH/s
[2018-06-10 09:18:11] Reject reason: Worker is not Authorized, e8.
[2018-06-10 09:18:11] Stratum connection interrupted
mtx_demon
Full Member
***
Offline Offline

Activity: 273
Merit: 100


View Profile
June 10, 2018, 04:55:11 PM
 #26

I was able to connect manually

but pretty annoying the fact that I cannot connect to MRR
mtx_demon
Full Member
***
Offline Offline

Activity: 273
Merit: 100


View Profile
June 10, 2018, 04:59:59 PM
Last edit: June 10, 2018, 06:02:56 PM by mtx_demon
 #27

Speeds are amazing!!

EVGA GTX 1070, 2117.66 kH/s
Gigabyte GTX 1070, 2046.16 kH/s
Zotac GTX 1070, 1957.97 kH/s
ASUS GTX 1080 Ti, 3738.15 kH/s


Berni5813
Newbie
*
Offline Offline

Activity: 3
Merit: 0


View Profile
June 10, 2018, 05:08:25 PM
 #28

Hi! Thx for the cuda 9.2 version. I tested on 1070ti and the hashrate is awesome Cheesy
The bad is that I have all my rigs working with cuda 8, can we get a version compiled for it? my other 74 1070ti will be very happy!
mtx_demon
Full Member
***
Offline Offline

Activity: 273
Merit: 100


View Profile
June 10, 2018, 05:16:17 PM
Last edit: June 10, 2018, 06:23:30 PM by mtx_demon
 #29

I am testing on a 12 rig

The miner start when crashes without an error MSG

Built with CUDA Toolkit 9.1

[2018-06-10 10:18:46] Starting on stratum+tcp://pool.blockmaster.eu:4553
[2018-06-10 10:18:46] GPU monitoring enabled.
[2018-06-10 10:18:46] 12 miner threads started, using 'lyra2z' algorithm.
[2018-06-10 10:18:47] New difficulty: 1 (0.00391)
[2018-06-10 10:18:47] lyra2z block 88950, diff 100.524
[2018-06-10 10:18:48] GPU#5: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#4: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#2: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#0: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#8: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#6: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#10: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#1: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#3: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#7: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#9: Intensity set to 17, 131072 cuda threads
[2018-06-10 10:18:48] GPU#11: Intensity set to 17, 131072 cuda threads


UPDATE

I got this error
23e000000-23ea00000 ---p 00000000 00:00 0
23ea00000-23ea04000 rwxs 00000000 00:06 464                              /dev/nvidiactl
23ea04000-23ec00000 ---p 00000000 00:00 0
23ec00000-23f000000 rwxs 00000000 00:06 464                              /dev/nvidiactl
23f000000-23fa00000 ---p 00000000 00:00 0
23fa00000-23fa04000 rwxs 00000000 00:06 464                              /dev/nvidiactl
23fa04000-240600000 ---p 00000000 00:00 0
240600000-240604000 rwxs 00000000 00:06 464                              /dev/nvidiactl
240604000-241200000 ---p 00000000 00:00 0
241200000-241204000 rwxs 00000000 00:06 464                              /dev/nvidiactl
241204000-241400000 ---p 00000000 00:00 0
241400000-241800000 rwxs 00000000 00:06 464                              /dev/nvidiactl
241800000-241c00000 rwxs 00000000 00:06 464                              /dev/nvidiactl
241c00000-242600000 ---p 00000000 00:00 0
242600000-242604000 rwxs 00000000 00:06 464                              /dev/nvidiactl


Can you fix that??
Also try to fix the fact that I wanna use a username not a wallet address
and I don't know why is not accepting
F848
Newbie
*
Offline Offline

Activity: 10
Merit: 0


View Profile
June 10, 2018, 06:02:24 PM
 #30

From the program output and strace it looks somewhat like you plugged some optimized kernels into ccminer, added a dev fee, and finally compiled and obfuscated the whole thing.
But you are technically not allowed to do that. ccminer is licensed under GPL. All modifications have to also be open source.
Sorry if I’m mistaken and you actually built the whole pool protocol handling and all yourself. The algo speedup is certainly nice and impressive. It’s just that everyone who’s dabbled in closed source miner development knows that the kind of most annoying part is building the whole shell around the actual GPU code. Building on top of a GPL-licensed open source miner is a cheap and not actually legal way around that.


Getting this error on a couple machines one Ubuntu 16.04 the other 18.04

./t-rex: /usr/lib/x86_64-linux-gnu/libcurl.so.4: version `CURL_OPENSSL_3' not found (required by ./t-rex)
If anyone cares, on Arch you get around this problem by doing
export LD_PRELOAD=libcurl.so.3
with installed libcurl-compat. How did I know? Ran into the same issue with ccminer already.
minimumnz
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
June 10, 2018, 11:04:06 PM
 #31

Is there any chance to get it compiled for cuda 9.2 or cuda 8?

Thank you! nice work  Smiley

Here you go, cuda 9.2 version - https://mega.nz/#!qJsQlAKB!rOpHtsqWjwvKzIljvL1uXO4bv3G-ZeWEB1XAMLWvpgk
I haven't tested it properly though, may not be stable or as fast as 9.1. Let me know how it goes.

Cuda 8 please Smiley
mtx_demon
Full Member
***
Offline Offline

Activity: 273
Merit: 100


View Profile
June 10, 2018, 11:47:41 PM
 #32

Is there any chance to get it compiled for cuda 9.2 or cuda 8?

Thank you! nice work  Smiley

Here you go, cuda 9.2 version - https://mega.nz/#!qJsQlAKB!rOpHtsqWjwvKzIljvL1uXO4bv3G-ZeWEB1XAMLWvpgk
I haven't tested it properly though, may not be stable or as fast as 9.1. Let me know how it goes.

Cuda 8 please Smiley

Just update to cuda 9.1 it's way better!!

I am using NV/OC with Cuda 9.1 and nvidia drivers 387
trexminer (OP)
Member
**
Offline Offline

Activity: 283
Merit: 63


View Profile
June 11, 2018, 03:28:57 AM
 #33

Replying to all of you:

Quote
But you are technically not allowed to do that. ccminer is licensed under GPL. All modifications have to also be open source.
You are right, ccminer was chosen as a base for this miner which is a GPL violation. All I can say is the miner is being rewritten from scratch at the moment, so in two weeks which is our current target the whole miner is going to be a brand new thing. CCMiner code was chosen solely to speed up the delivery of the first release.

Quote
Cuda 8 please  Smiley
I tried compiling it against cuda 8 but unfortunately the hashrate was far from what is achievable on cuda 9.1/9.2. I'm sure it can be fixed but requires some thorough investigation. I would really recommend moving to cuda 9 as it's very stable, and I think you'll eventually have to do the switch anyway as 8 becomes legacy.

Quote
but pretty annoying the fact that I cannot connect to MRR
It seems MRR is using a slightly different network protocol, we'll try to make supported in that new version of the miner I mentioned above when it comes out.

Quote
Can you somehow link the libraries into the binary?
I tried a few options to make the miner work on a system that doesn't have CUDA toolkit installed, but only this one worked for me:
I provide the binary and libcudart.so.9.1 (or libcudart.so.9.2). Then you copy this libcudart file to your preferred location, say "/home/john/lib64/". Edit /etc/ld.so.conf, append your chosen location to the end of the file. Then run "sudo ldconfig". Would that work for you?

I've created a discord channel - https://discord.gg/gj7jcYf. Let's discuss issues there.
djentmining
Newbie
*
Offline Offline

Activity: 38
Merit: 0


View Profile
June 11, 2018, 04:22:32 AM
 #34

Bruh what's the usage on Phi2 algo ?
trexminer (OP)
Member
**
Offline Offline

Activity: 283
Merit: 63


View Profile
June 11, 2018, 05:20:10 AM
 #35

Bruh what's the usage on Phi2 algo ?

This miner doesn't support Phi2 at the moment. We'll add it in future versions.
tazmako
Full Member
***
Offline Offline

Activity: 350
Merit: 100


XDNA - Most innovative cryptocurrency in 2018


View Profile
June 11, 2018, 03:54:01 PM
 #36

wait for windows version

XDNA ❱❭ Dynamic. New. Adaptable.
Revolution in PoW GPU mining | Innovation in Masternode investing
° BitGun ° T.N.T. ° XDNA Foundation °
Stable block reward even with a 10000% increase of hashrate![/url]
                 |Website| Whitepaper| Discord| Telegram | Github|                
preda
Sr. Member
****
Offline Offline

Activity: 756
Merit: 250


View Profile
June 11, 2018, 03:57:27 PM
 #37

Bruh what's the usage on Phi2 algo ?

This miner doesn't support Phi2 at the moment. We'll add it in future versions.


the chain i s issued i think will need time for phi2 at thi moment, rather id like to have the windows version dev, would be great
trexminer (OP)
Member
**
Offline Offline

Activity: 283
Merit: 63


View Profile
June 12, 2018, 12:49:25 AM
 #38

In the meantime, I just published a new version that speeds up Lyra2z, Tribus and Phi1612. Updated the benchmarks and download links in the first post.
djentmining
Newbie
*
Offline Offline

Activity: 38
Merit: 0


View Profile
June 12, 2018, 01:12:17 AM
 #39

Thanks for the update dev Smiley
mtx_demon
Full Member
***
Offline Offline

Activity: 273
Merit: 100


View Profile
June 12, 2018, 12:33:24 PM
 #40

So far using the new miner over 24 hours

Increase of 20-30% of speeds

Amazing miner thanks so much dev for the hard work!!
Pages: « 1 [2] 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 ... 106 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!