[ANN] TeamRedMiner v0.10.10 - Ironfish/Kaspa/ZIL/Kawpow/Etchash and More

[livada]

i read this but no understud

 --eth_hashwatch(=M,M) Defines a min/max watchdog hashrate in MH/s expected when gpus are hashing.  If a GPU falls outside
                              of this range for too long it will be considered faulty and the watchdog will be triggered.  The
                              default min/max values are 2 MH/s and 150 MH/s.  To activate with default values, use
                              --eth_hashwatch.  To set your own values, use e.g. --eth_hashwatch=1.25,60. The same range is used
                              for all gpus in the rig.  A negative value for one side disables it, e.g. use --eth_hashwatch=-1,1000
                              to handle gpu resets that result in huge fake hashrates but no minimum check.

what is  what in this? --eth_hashwatch=1.25,60

1 is gPU1 ?
25 is minimum 25mh ?
60 is maximum 60mh?
or 1.25 is MH min?

i have 6 vega and 1 vega give 45-50mh/s - all rig have 280mh

this option use only1 gpu HR or all rig HR ? what command i must use?
or
this option use for 1,vega- and i use only default command --eth_hashwatch??

i wont restart miner if my rig give 200mh or less. pls what comand i use for this?
this command --eth_hashwatch=200,300 is good? 200hr min 300 max? why i must use MAX?
or i must use for 1 gpu limit?
this command --eth_hashwatch=40,51

--eth_hashwatch=1.25,60 means

every gpu follows this range in your mining config.
if any GPU falls under 1.25 MHs it would trigger watchdog.  If it goes beyond 60 MHs it would trigger the watchdog.  Ethash is in MHs for almost every card hence the MHs is unit of measurement.
You can use anything below 40 for min and anything above 60 or a more flexible range like 35 as min,70 as max.

thx ...

[weifa]

Great job with TeamRedMiner!  I found your miner on a list of Etchash supported miners, great job supporting the new Ethash variant.

I was wondering if you had any interest in adding support for Ubqhash?

Here's some of the technicals to help explain the algo changes more in depth:  https://blog.ubiqsmart.com/introducing-ubqhash-8fa515befd7

Ubiq has quite a large community of miners, but the current number of mining programs is somewhat limited.  I'm reaching out on behalf of the Ubiq community to see if you're interested in potentially adding Ubqhash support, thanks!

[miner29]

someone can mine with the rx 5500 4gb ethash?

i know on linux you can for a few more weeks.

[kerney666]

Great job with TeamRedMiner!  I found your miner on a list of Etchash supported miners, great job supporting the new Ethash variant.

I was wondering if you had any interest in adding support for Ubqhash?

Here's some of the technicals to help explain the algo changes more in depth:  https://blog.ubiqsmart.com/introducing-ubqhash-8fa515befd7

Ubiq has quite a large community of miners, but the current number of mining programs is somewhat limited.  I'm reaching out on behalf of the Ubiq community to see if you're interested in potentially adding Ubqhash support, thanks!

Hi weifa! Well, if it's only keccak -> blake in the dag cache build that differs it's quite simple. Our TODO list just keeps on growing though so I won't make any promises, but hopefully we'll find some time in the not-too-distant future.

[kerney666]

someone can mine with the rx 5500 4gb ethash?

The driver support for the 5500 4GBs just seems horrible. Haven't ran enough tests, but it fails miserably at allocating a large portion of the available vram. Most 8GB gpus have a limit at 7.75GB, i.e. 256MB is reserved. It feels like this might have leaked over to the 5500 4GBs as well, but I'm just guessing.That said, afaik I don't think anyone is able to mine the current ETH epoch (379) with their 5500 4GBs.

[bassistas]

Hi guys I am trying to mine eth with my r9 390 8GB gpus
but I get this error. Does anyone know how to fix this?
                                                                                                                                                                                                                                       
[2020-12-02 21:48:54] Auto-detected AMD OpenCL platform 0                                                                                                                                                                                                                                 
[2020-12-02 21:48:55] Initialize error, unknown device name: 'Hawaii'
[2020-12-02 21:48:55] Failed to initialize device idx 0 (-8)
[2020-12-02 21:48:55] Successful clean shutdown.                                                                                                                                                                                                                                         
[2020-12-02 21:48:55] Watchdog script executor thread exiting.                                                                                                                                                                                                                           
                                                                                                                                                                                                                                                                                         
teamredminer exited, waiting to cooldown a bit

[weifa]

Great job with TeamRedMiner!  I found your miner on a list of Etchash supported miners, great job supporting the new Ethash variant.

I was wondering if you had any interest in adding support for Ubqhash?

Here's some of the technicals to help explain the algo changes more in depth:  https://blog.ubiqsmart.com/introducing-ubqhash-8fa515befd7

Ubiq has quite a large community of miners, but the current number of mining programs is somewhat limited.  I'm reaching out on behalf of the Ubiq community to see if you're interested in potentially adding Ubqhash support, thanks!

Hi weifa! Well, if it's only keccak -> blake in the dag cache build that differs it's quite simple. Our TODO list just keeps on growing though so I won't make any promises, but hopefully we'll find some time in the not-too-distant future.

Hey kerney thanks for your reply!  Yes I believe Ubqhash does involve switching over the keccak -> blake as you said.  One thing to mention is I believe the hard fork from Ethash to Ubqhash occurred at block 660,000 (epoch 22).  I hear you on the TODO list, no rush on our front.  If you get a chance in the future to add support I'm happy to help try it out on some of my miners, cheers!

[kerney666]

Hi guys I am trying to mine eth with my r9 390 8GB gpus
but I get this error. Does anyone know how to fix this?
                                                                                                                                                                                                                                       
[2020-12-02 21:48:54] Auto-detected AMD OpenCL platform 0                                                                                                                                                                                                                                 
[2020-12-02 21:48:55] Initialize error, unknown device name: 'Hawaii'
[2020-12-02 21:48:55] Failed to initialize device idx 0 (-8)
[2020-12-02 21:48:55] Successful clean shutdown.                                                                                                                                                                                                                                         
[2020-12-02 21:48:55] Watchdog script executor thread exiting.                                                                                                                                                                                                                           
                                                                                                                                                                                                                                                                                         
teamredminer exited, waiting to cooldown a bit

Hi bassistas! Sorry, we only support Fiji/Ellesmere generation and later gpus, Hawaii is not included.

[XxXBigDickXxX]

Guys i have idea-- if 6800 have cache like processor, is it possible to make them mine randomX ?!!

[kerney666]

Guys i have idea-- if 6800 have cache like processor, is it possible to make them mine randomX ?!!

It most probably means RandomX runs better on a 6800 than e.g. a 5700XT. That said, I don't think it will be the rational choice as a miner, there aren't enough 2MB pads available in 128MB to even run a single hash on each compute unit on a 6800XT.

[GreyGhost_rc]

Hi, I have this weird problem, on a 6 card RIG, 4 of the cards gives me a hashrate of 20mh/s (down from 27mh/s, they are 4gb cards so I think it is that), but the 5th card gives me 30 mh/s, but poolside it is 500kh/s. The 6th card is a vega which gives me 42mh/s. Any ideas why this should be? It does not matter how I swap out the cards, which card is used for display or anything. Also tried the whole eth_epoch 274,274,274,274,273 etc.

[NCarter84]

Hi, I have this weird problem, on a 6 card RIG, 4 of the cards gives me a hashrate of 20mh/s (down from 27mh/s, they are 4gb cards so I think it is that), but the 5th card gives me 30 mh/s, but poolside it is 500kh/s. The 6th card is a vega which gives me 42mh/s. Any ideas why this should be? It does not matter how I swap out the cards, which card is used for display or anything. Also tried the whole eth_epoch 274,274,274,274,273 etc.

Hi!

What OS and Drivers are you using?

[sxemini]

Hi, I have this weird problem, on a 6 card RIG, 4 of the cards gives me a hashrate of 20mh/s (down from 27mh/s, they are 4gb cards so I think it is that), but the 5th card gives me 30 mh/s, but poolside it is 500kh/s. The 6th card is a vega which gives me 42mh/s. Any ideas why this should be? It does not matter how I swap out the cards, which card is used for display or anything. Also tried the whole eth_epoch 274,274,274,274,273 etc.

Hi!

What OS and Drivers are you using?

I had the same bug, only a visual bug. Card has not enough Vram for ETH, this card only hash with 0 - 2 mh. I had this problem on 2 of 10 rigs, all with same OS win10 and driver 19.7.4.
I don´t why 1 card has not enough vram, display gpu is the integrated gpu and all cards the same for me. I checked gpu z and i see that one gpu use more vram as the other cards.  

But i think it is not a miner problem, same is happen with lolminer for example.

[Yirvenpaeae]

hi,

i'm new on TRM. i have 4 vega red evil 56 card hynix ,with my settings from claymore i get a hashrate on 43mh/s
930mhz mem 1050 core on 900mv can't get the mem higher.
 i try modded the mem timings with the amdmemorytweakxl tool.

i always get the 43mh/s
i'm not sure i've im doing it right. try the changes from the ethgenreraltuning list in tweak tool. apply.

does TRM auto tuning reset this settings? or do i have to add anything in the command line?

thy

[kerney666]

hi,

i'm new on TRM. i have 4 vega red evil 56 card hynix ,with my settings from claymore i get a hashrate on 43mh/s
930mhz mem 1050 core on 900mv can't get the mem higher.
 i try modded the mem timings with the amdmemorytweakxl tool.

i always get the 43mh/s
i'm not sure i've im doing it right. try the changes from the ethgenreraltuning list in tweak tool. apply.

does TRM auto tuning reset this settings? or do i have to add anything in the command line?

thy

The 43 MH/s is the classic limit at stock timings, so it really sounds like the changed timings aren't being applied properly for you. The miner doesn't touch anything related to clocks/voltages/timings, only the fan curve if you have fan control enabled.

It's hard to say why the timings aren't being applied. If you start the miner and let it run for a few mins, then open amdmemtweakxl and e.g. boost tREF to 17000 or make a slight adjustment down for tRCDWR and apply the changes, does the hashrate move in the miner? If not, what timing are displayed of you close+reopen amdmemtweakxl, are your changes there?

[GreyGhost_rc]

Hi, I have this weird problem, on a 6 card RIG, 4 of the cards gives me a hashrate of 20mh/s (down from 27mh/s, they are 4gb cards so I think it is that), but the 5th card gives me 30 mh/s, but poolside it is 500kh/s. The 6th card is a vega which gives me 42mh/s. Any ideas why this should be? It does not matter how I swap out the cards, which card is used for display or anything. Also tried the whole eth_epoch 274,274,274,274,273 etc.

Hi!

What OS and Drivers are you using?

I had the same bug, only a visual bug. Card has not enough Vram for ETH, this card only hash with 0 - 2 mh. I had this problem on 2 of 10 rigs, all with same OS win10 and driver 19.7.4.
I don´t why 1 card has not enough vram, display gpu is the integrated gpu and all cards the same for me. I checked gpu z and i see that one gpu use more vram as the other cards.  

But i think it is not a miner problem, same is happen with lolminer for example.

Hi, am using Win10 with newest Adrenaline drivers. Mine is fine on other miners, but I like TRM. The giving <1mh/s is not the display gpu, seeing as I also use integrated gpu. And I do not think it is the VRAM issue, because if I swap that card out with another the other card, ALWAYS the LAST 4gb card gives me <1mh/s

[sxemini]

Hi, I have this weird problem, on a 6 card RIG, 4 of the cards gives me a hashrate of 20mh/s (down from 27mh/s, they are 4gb cards so I think it is that), but the 5th card gives me 30 mh/s, but poolside it is 500kh/s. The 6th card is a vega which gives me 42mh/s. Any ideas why this should be? It does not matter how I swap out the cards, which card is used for display or anything. Also tried the whole eth_epoch 274,274,274,274,273 etc.

Hi!

What OS and Drivers are you using?

I had the same bug, only a visual bug. Card has not enough Vram for ETH, this card only hash with 0 - 2 mh. I had this problem on 2 of 10 rigs, all with same OS win10 and driver 19.7.4.
I don´t why 1 card has not enough vram, display gpu is the integrated gpu and all cards the same for me. I checked gpu z and i see that one gpu use more vram as the other cards.  

But i think it is not a miner problem, same is happen with lolminer for example.

Hi, am using Win10 with newest Adrenaline drivers. Mine is fine on other miners, but I like TRM. The giving <1mh/s is not the display gpu, seeing as I also use integrated gpu. And I do not think it is the VRAM issue, because if I swap that card out with another the other card, ALWAYS the LAST 4gb card gives me <1mh/s

It is vram   i have checked it, and in my rigs it was the 6. card. I don´t know why people don´t believe such things, i have tested, controlled and i am not a noob, but yeah do what you do, you will never find a solution.

I wrote, i don´t know why, but one gpu (always the 6. gpu) has not enough vram. check it by yourself or leave. why are so many people so ignorant

[Yirvenpaeae]

hi,

i'm new on TRM. i have 4 vega red evil 56 card hynix ,with my settings from claymore i get a hashrate on 43mh/s
930mhz mem 1050 core on 900mv can't get the mem higher.
 i try modded the mem timings with the amdmemorytweakxl tool.

i always get the 43mh/s
i'm not sure i've im doing it right. try the changes from the ethgenreraltuning list in tweak tool. apply.

does TRM auto tuning reset this settings? or do i have to add anything in the command line?

thy

The 43 MH/s is the classic limit at stock timings, so it really sounds like the changed timings aren't being applied properly for you. The miner doesn't touch anything related to clocks/voltages/timings, only the fan curve if you have fan control enabled.

It's hard to say why the timings aren't being applied. If you start the miner and let it run for a few mins, then open amdmemtweakxl and e.g. boost tREF to 17000 or make a slight adjustment down for tRCDWR and apply the changes, does the hashrate move in the miner? If not, what timing are displayed of you close+reopen amdmemtweakxl, are your changes there?

hi, thx

now i got 46 mh/s with 1700 on ref and 20rcdrd and 4 rcdwr. there was no diference with rcdwr starts on 10.

does the settings from tweak tool reset after pc restart? watchdog is enabled with restart the system.
can i set a target temp for fan with command line ? or just settingfan speed?

[kerney666]

hi,

i'm new on TRM. i have 4 vega red evil 56 card hynix ,with my settings from claymore i get a hashrate on 43mh/s
930mhz mem 1050 core on 900mv can't get the mem higher.
 i try modded the mem timings with the amdmemorytweakxl tool.

i always get the 43mh/s
i'm not sure i've im doing it right. try the changes from the ethgenreraltuning list in tweak tool. apply.

does TRM auto tuning reset this settings? or do i have to add anything in the command line?

thy

The 43 MH/s is the classic limit at stock timings, so it really sounds like the changed timings aren't being applied properly for you. The miner doesn't touch anything related to clocks/voltages/timings, only the fan curve if you have fan control enabled.

It's hard to say why the timings aren't being applied. If you start the miner and let it run for a few mins, then open amdmemtweakxl and e.g. boost tREF to 17000 or make a slight adjustment down for tRCDWR and apply the changes, does the hashrate move in the miner? If not, what timing are displayed of you close+reopen amdmemtweakxl, are your changes there?

hi, thx

now i got 46 mh/s with 1700 on ref and 20rcdrd and 4 rcdwr. there was no diference with rcdwr starts on 10.

does the settings from tweak tool reset after pc restart? watchdog is enabled with restart the system.
can i set a target temp for fan with command line ? or just settingfan speed?

My bad, I wrote rcdwr but really meant rcdrd. That's the one you want to lower slowly. Also, did you do 1700 or 17000 for ref?

For fan control, you can target a core temp, junction temp, or mem temp, set the initial fan speed or a static fan speed. Use --fan_control --fan_default_vega=::70:60 to target 70C mem temp with fans initially at 60%.

[Yirvenpaeae]

hi,

i'm new on TRM. i have 4 vega red evil 56 card hynix ,with my settings from claymore i get a hashrate on 43mh/s
930mhz mem 1050 core on 900mv can't get the mem higher.
 i try modded the mem timings with the amdmemorytweakxl tool.

i always get the 43mh/s
i'm not sure i've im doing it right. try the changes from the ethgenreraltuning list in tweak tool. apply.

does TRM auto tuning reset this settings? or do i have to add anything in the command line?

thy

The 43 MH/s is the classic limit at stock timings, so it really sounds like the changed timings aren't being applied properly for you. The miner doesn't touch anything related to clocks/voltages/timings, only the fan curve if you have fan control enabled.

It's hard to say why the timings aren't being applied. If you start the miner and let it run for a few mins, then open amdmemtweakxl and e.g. boost tREF to 17000 or make a slight adjustment down for tRCDWR and apply the changes, does the hashrate move in the miner? If not, what timing are displayed of you close+reopen amdmemtweakxl, are your changes there?

hi, thx

now i got 46 mh/s with 1700 on ref and 20rcdrd and 4 rcdwr. there was no diference with rcdwr starts on 10.

does the settings from tweak tool reset after pc restart? watchdog is enabled with restart the system.
can i set a target temp for fan with command line ? or just settingfan speed?

My bad, I wrote rcdwr but really meant rcdrd. That's the one you want to lower slowly. Also, did you do 1700 or 17000 for ref?

For fan control, you can target a core temp, junction temp, or mem temp, set the initial fan speed or a static fan speed. Use --fan_control --fan_default_vega=::70:60 to target 70C mem temp with fans initially at 60%.

I start with 17000 on ref an go down to 18 on rcdrd but it was  not stable. ref is now on 15600 and rcdrd 20 is stable for a few hours. hashrate is 46mh/s. i set the target temp on 55
But on restart the settings are lost.