Pages:
Author

Topic: SRBMiner Cryptonight AMD GPU Miner V1.9.3 - native algo switching - page 66. (Read 237261 times)

hero member
Activity: 2548
Merit: 626
Looks like GPU 0, which would be a RX470, 4gb card. All the other algos I tested worked fine.
I have 4 other 470s in this rig that are the same.

So that's why it is interesting, if the other cards are same.
Can you try to add --nocache in start.bat ?
newbie
Activity: 13
Merit: 0
Hey there can you please tell me how many hashes you are getting with Vega 56/64?

Did you find any solution for vega 64 or 64 FE? Is this CBR giving higher hashrate than cast XMR?
jr. member
Activity: 195
Merit: 4
Working on V177 for the moment.

I've tested several algos and was working on Bittube, running from CMD Line like so..

Code:
--ccryptonighttype bittubev2 --cgpuid 0,1,2,3,4,5 --cgpuintensity 0,0,0,0,0,0 --cgputhreads 2,2,2,2,2,2 --cpool [Server]:[Port] --cwallet [User] --cpassword [Password]

I'm getting the following error:

Error CL_COMPILE_PROGRAM_FAILURE when calling clCompileProgram [ctx->Program] for DeviceID 0 (Thread 0)

I dont understand, you get this on 1.7.8 ?
Which GPU ?

Sorry, no meaning.. I'm currently working/benching on V 1.7.7 -- I didn't see that 178 was out until after I started all my benching.

NP, but im still interested why you get that.
Which GPU can't compile ?

Looks like GPU 0, which would be a RX470, 4gb card. All the other algos I tested worked fine.

I have 4 other 470s in this rig that are the same.
hero member
Activity: 2548
Merit: 626
V1.7.8
- Added new algo 'CN-Gpu' (faster than original implementation in Xmr Stak)
- Much better auto setup on some cards for 'CN-Turtle'
- Added check for last sent share time
- Added aliases for algos, to be more compatible with other miner names
- Prettier stats (when pressing 's')
- Minor bug fixes


+ Ryo will fork soon to CN-Gpu algo
+ Added check for last sent share time, if no share was sent to pool for 30 minutes, miner will restart.
+ To make it easier for everyone, lots of algo name aliases were added, you can now use algo names from Stak, Xmrig, etc... Also you can use coin names, and if an alias exists, it will select the right algo for you.

Examples for "cryptonight_type" :

- tube
- monero
- cryptonight_turtle
- masari

etc..

Also added aliases for Masari, Stellite and Turtle for MoneroOcean, so miner will now switch to those coins.
is there any coin already mining on this algo cn-gpu?
any number for the benchmark cn-gpu?

only testnet : https://tnpool.ryo-currency.com
hero member
Activity: 2548
Merit: 626
Working on V177 for the moment.

I've tested several algos and was working on Bittube, running from CMD Line like so..

Code:
--ccryptonighttype bittubev2 --cgpuid 0,1,2,3,4,5 --cgpuintensity 0,0,0,0,0,0 --cgputhreads 2,2,2,2,2,2 --cpool [Server]:[Port] --cwallet [User] --cpassword [Password]

I'm getting the following error:

Error CL_COMPILE_PROGRAM_FAILURE when calling clCompileProgram [ctx->Program] for DeviceID 0 (Thread 0)

I dont understand, you get this on 1.7.8 ?
Which GPU ?

Sorry, no meaning.. I'm currently working/benching on V 1.7.7 -- I didn't see that 178 was out until after I started all my benching.

NP, but im still interested why you get that.
Which GPU can't compile ?
jr. member
Activity: 195
Merit: 4
Working on V177 for the moment.

I've tested several algos and was working on Bittube, running from CMD Line like so..

Code:
--ccryptonighttype bittubev2 --cgpuid 0,1,2,3,4,5 --cgpuintensity 0,0,0,0,0,0 --cgputhreads 2,2,2,2,2,2 --cpool [Server]:[Port] --cwallet [User] --cpassword [Password]

I'm getting the following error:

Error CL_COMPILE_PROGRAM_FAILURE when calling clCompileProgram [ctx->Program] for DeviceID 0 (Thread 0)

I dont understand, you get this on 1.7.8 ?
Which GPU ?

Sorry, no meaning.. I'm currently working/benching on V 1.7.7 -- I didn't see that 178 was out until after I started all my benching.
hero member
Activity: 2548
Merit: 626
Working on V177 for the moment.

I've tested several algos and was working on Bittube, running from CMD Line like so..

Code:
--ccryptonighttype bittubev2 --cgpuid 0,1,2,3,4,5 --cgpuintensity 0,0,0,0,0,0 --cgputhreads 2,2,2,2,2,2 --cpool [Server]:[Port] --cwallet [User] --cpassword [Password]

I'm getting the following error:

Error CL_COMPILE_PROGRAM_FAILURE when calling clCompileProgram [ctx->Program] for DeviceID 0 (Thread 0)

I dont understand, you get this on 1.7.8 ?
Which GPU ?
legendary
Activity: 1764
Merit: 1002
V1.7.8
- Added new algo 'CN-Gpu' (faster than original implementation in Xmr Stak)
- Much better auto setup on some cards for 'CN-Turtle'
- Added check for last sent share time
- Added aliases for algos, to be more compatible with other miner names
- Prettier stats (when pressing 's')
- Minor bug fixes


+ Ryo will fork soon to CN-Gpu algo
+ Added check for last sent share time, if no share was sent to pool for 30 minutes, miner will restart.
+ To make it easier for everyone, lots of algo name aliases were added, you can now use algo names from Stak, Xmrig, etc... Also you can use coin names, and if an alias exists, it will select the right algo for you.

Examples for "cryptonight_type" :

- tube
- monero
- cryptonight_turtle
- masari

etc..

Also added aliases for Masari, Stellite and Turtle for MoneroOcean, so miner will now switch to those coins.
is there any coin already mining on this algo cn-gpu?
any number for the benchmark cn-gpu?
jr. member
Activity: 195
Merit: 4
Working on V177 for the moment.

I've tested several algos and was working on Bittube, running from CMD Line like so..

Code:
--ccryptonighttype bittubev2 --cgpuid 0,1,2,3,4,5 --cgpuintensity 0,0,0,0,0,0 --cgputhreads 2,2,2,2,2,2 --cpool [Server]:[Port] --cwallet [User] --cpassword [Password]

I'm getting the following error:

Error CL_COMPILE_PROGRAM_FAILURE when calling clCompileProgram [ctx->Program] for DeviceID 0 (Thread 0)
hero member
Activity: 2548
Merit: 626
V1.7.8
- Added new algo 'CN-Gpu' (faster than original implementation in Xmr Stak)
- Much better auto setup on some cards for 'CN-Turtle'
- Added check for last sent share time
- Added aliases for algos, to be more compatible with other miner names
- Prettier stats (when pressing 's')
- Minor bug fixes


+ Ryo will fork soon to CN-Gpu algo
+ Added check for last sent share time, if no share was sent to pool for 30 minutes, miner will restart.
+ To make it easier for everyone, lots of algo name aliases were added, you can now use algo names from Stak, Xmrig, etc... Also you can use coin names, and if an alias exists, it will select the right algo for you.

Examples for "cryptonight_type" :

- tube
- monero
- cryptonight_turtle
- masari

etc..

Also added aliases for Masari, Stellite and Turtle for MoneroOcean, so miner will now switch to those coins.
jr. member
Activity: 230
Merit: 1
what is the best coin to mine for vegas? have some vega rigs

Loki, Tube, X-Cash or Uplexa
legendary
Activity: 1764
Merit: 1002
what is the best coin to mine for vegas? have some vega rigs
hero member
Activity: 2548
Merit: 626
daj otvori PP jer ti nemrem poruku poslat

Salji, otvorio sam



Hi. I have problem with litev7 (only litev7). Other algo works super
SRBMiner v1.7.7 (1.7.6)

[2019-02-05 15:55:03] Warning! Could not connect to a devfee pool for a long time. Please check you firewall/antivirus is not blocking connections. Until then i will take a break.
[2019-02-05 15:55:03] GPU devices disabled!

I change DNS servers to  1.1.1.1 or 8.8.8.8
does not help anything

thanks


There are 3 pools for litev7 algo, i checked they are all online and miner can connect to them without problem. IDK what could cause the problem at your end.



check power on different algo
same card frequency and power
rx570\rx470 4gb

on tube2
10500 h\s 1185 watt

on fast2
21700 h\s 1405 watt

why?Huh fast2 realy heavy algo as v8?

yes, needs more power.
Turtle needs even more.
The new CN-GPU will need even more, it's crazy like +75% more than heavy algos and +33% more than v8.....



Anyone here got the same problem?
the GPUs are mining and at some moment they are dead  Huh

I checked the hardware, is everything in order....


You could lower the memory clocks for those cards (bus 10,3) by 10-20mhz and see what happens.



Anyone mining Turtle with RX580? what your hash rates?

Auto setup for Ellesmere is not very good in 177, i fixed that in the upcoming version.
You can set it by hand to :

intensity : 104-112
worksize: 8
threads: 2
fragments: 4
newbie
Activity: 3
Merit: 0
Does it work on Ubuntu ?
jr. member
Activity: 66
Merit: 4
Anyone mining Turtle with RX580? what your hash rates?
hero member
Activity: 1680
Merit: 583
xUSD - The PRIVATE stable coin - Haven Protocol
Anyone here got the same problem?
the GPUs are mining and at some moment they are dead  Huh

I checked the hardware, is everything in order....

Code:
[2019-02-06 09:04:24] GPU0 :    1013 H/s [T: 69c, RPM: 976, CC: 1190 MHz, MC: 2200 MHz][BUS:10]
[2019-02-06 09:04:24] GPU1 :    1022 H/s [T: 71c, RPM: 1039, CC: 1190 MHz, MC: 2200 MHz][BUS:5]
[2019-02-06 09:04:24] GPU2 :    1022 H/s [T: 74c, RPM: 982, CC: 1190 MHz, MC: 2200 MHz][BUS:8]
[2019-02-06 09:04:24] GPU3 :    1012 H/s [T: 74c, RPM: 1071, CC: 1190 MHz, MC: 2200 MHz][BUS:7]
[2019-02-06 09:04:24] GPU4 :    1016 H/s [T: 74c, RPM: 962, CC: 1190 MHz, MC: 2200 MHz][BUS:2]
[2019-02-06 09:04:24] GPU5 :    1024 H/s [T: 75c, RPM: 1078, CC: 1190 MHz, MC: 2200 MHz][BUS:6]
[2019-02-06 09:04:24] GPU6 :    1006 H/s [T: 68c, RPM: 1074, CC: 1190 MHz, MC: 2200 MHz][BUS:1]
[2019-02-06 09:04:24] GPU7 :    1019 H/s [T: 65c, RPM: 995, CC: 1190 MHz, MC: 2200 MHz][BUS:3]
[2019-02-06 09:04:24] Total:    8134 H/s
[2019-02-06 09:04:31] Pool accepted result 0x0000181F [52ms]
[2019-02-06 09:05:00] Warning: GPU0 [BUS: 10] hashing speed is 0 H/S
[2019-02-06 09:05:06] Pool accepted result 0x000037DF [53ms]
[2019-02-06 09:05:26] Warning: GPU0 [BUS: 10] hashing speed is 0 H/S
[2019-02-06 09:05:26] Warning: GPU7 [BUS: 3] hashing speed is 0 H/S
[2019-02-06 09:05:28] Pool accepted result 0x00000D4C [104ms]
[2019-02-06 09:05:49] Pool accepted result 0x00006F87 [54ms]
[2019-02-06 09:05:52] GPU0 [BUS: 10] DISABLED [DEAD]
[2019-02-06 09:05:52] Warning: GPU7 [BUS: 3] hashing speed is 0 H/S
[2019-02-06 09:06:06] Pool accepted result 0x000042CF [53ms]
[2019-02-06 09:06:18] GPU7 [BUS: 3] DISABLED [DEAD]
newbie
Activity: 417
Merit: 0


check power on different algo
same card frequency and power
rx570\rx470 4gb

on tube2
10500 h\s 1185 watt

on fast2
21700 h\s 1405 watt

why?Huh fast2 realy heavy algo as v8?



fast v7 use more power than v7 algo
now fast v8 use more power than v8.
newbie
Activity: 2
Merit: 0
Hi. I have problem with litev7 (only litev7). Other algo works super
SRBMiner v1.7.7 (1.7.6)

[2019-02-05 15:55:03] Warning! Could not connect to a devfee pool for a long time. Please check you firewall/antivirus is not blocking connections. Until then i will take a break.
[2019-02-05 15:55:03] GPU devices disabled!

I change DNS servers to  1.1.1.1 or 8.8.8.8
does not help anything

thanks
jr. member
Activity: 288
Merit: 1


check power on different algo
same card frequency and power
rx570\rx470 4gb

on tube2
10500 h\s 1185 watt

on fast2
21700 h\s 1405 watt

why?Huh fast2 realy heavy algo as v8?
hero member
Activity: 2548
Merit: 626
Will you be releasing an update with CN-GPU? They will be forking at block 228750 in about 10 days.

Yes in a few days if everything goes well.
Pages:
Jump to: