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

[mstrozier]

Ok different set of errors when running TeamRedMiner for both Ethash and Chukwa mining together.  I have one session pulling the first 5 GPU's on Callisto.   I have the 6th GPU on Turtlecoin. 

First error.  I have the Ethash portion running first and then start up the chukwa miner.  Screen shot of error: https://ibb.co/mzgh8Mm

The chukwa miner would never progress, just sat at that screen.  However on the ethash portion one by one each card went into dead status.


So next tried firing up chukwa first.  Then Ethash.  And here is that result:  https://ibb.co/VD2qgw6

[1715340565]

1715340565

[1715340565]

1715340565

[1715340565]

1715340565

[vitik]

@mstrozier
can a weak processor? or loaded with something
I had a similar one, removed the load from the cores and everything was decided

[UnclWish]

Sorry for offtopic...
What card better Vega 56 Asus Strix OC or Sapphire Pulse?

[mstrozier]

@mstrozier
can a weak processor? or loaded with something
I had a similar one, removed the load from the cores and everything was decided

It's not a CPU issue from what I can see.  Right now I have TeamRedMiner running on all 6 GPU's mining Chukwa.  It's using less than 2% of the cpu usage.   When I split the rig and put in my prior example, 5  Rx 470's on Ethash Algo mining Callisto, and put the 6th Rx 470 on Chukwa mining Turtle coin, the above posting screenshots are the errors I get.   Several posts up on my original I show the task manager window open.  CPU load at that point was only 0.3%

[Dropshot2018]

Sorry for offtopic...
What card better Vega 56 Asus Strix OC or Sapphire Pulse?

Asus Strix

[ivomm]

Hi! Could you give me an example of tuning AMD 570 for Turtle coin? More specifically, which is the most power efficient core clock/voltage? I've read the documentation and it said for the core 1050 mhz, but how low can the voltage go - 800V or even less? Currently I am 1150/850 but the power consumption is 20% more than CN-r.

Chukwa and CN/r are completely different algos, there's just no point expecting a power consumption X because CN/r draws Y.

For your question, you just have to test your way to a stable setup, impossible to say where your 570 will croak . If you have a power target you don't want to exceed you'll have to clock down the mem clk to throttle the algo, otherwise just continue to dial down the core clk and voltage and see where you end up.

I know this. It takes much time to determine for each card whether the voltage/frequence settings will cause troubles. For example, after 2 months of running one of my 13 card rig with settings for XMR, one of my cards (actually the only 560 card) started to have issues. Even after cold restart the card becomes dead quickly. Since the memory settings for this particular card I left by default, I came to the easy conclusion that the 1150/850 setting for the core is too much. So I reduced the clock to 1050 and the card seemed to like it. So far 24 48 hours without errors. My point is that it will take months to determine whether a setting 1050/800 or 1050/750 etc. could be achieved for each 570 and 580 card. I just need a confirmation from someone that 800 or less mV is achievable for chukwa, and thus it is worthy to take the time and adjust each card. I guess I can get 100-200W less power consumption for a rig, if this is the case. Someone wrote here that he managed to get equal power consumption for chukwa and cn-r without sacrificing the hash rate, so my guess is he managed to stabilize his cards with less than 850 mV on the core.

[xMindx163]

Can be, too, add?

v1. 90 GMiner
+ added CryptoNightBBC algorithm for Nvidia GPUs (BBC coin)

[reaper7881]

Can be, too, add?

v1. 90 GMiner
+ added CryptoNightBBC algorithm for Nvidia GPUs (BBC coin)

Looks like a Chinese centralised scam coin
That requires miners to register their wallets knowyourminer lol
Also blocks mine by pools other than the official one are orphaned on purpose

[dagere76]

I have a 4 GPU setup with MSI AMD RX 470. They are all the same card. I am running windows 10 64 pro with a Pentium processor and 4gb Ram and a 120 solid state hard Drive. I have my virtual memory set at minimum to 6gb and max at 20gb. I am using NiceHash with the latest version of TeamRedMiner. I for some reason cannot run more than two gpus. If I try to start with a third gpu I get the cannot initialize idx error. Any thoughts what would cause this?
Thanks

[ggarbi]

Hi there.
Im mining bittube (CNsaber algo) on my vega 64 rigs and my first GPU hashing less than others (about 1700 hash the rest are around 2100) and when i switch algo to for example CnR or CNconceal there is no problem and they are hashing the same amount.
Is there any problem with this miner on CnSaber algo ?
Im using 19.4.1 driver and clocks are the same + reg edit and weak Teamredminer timer on AMD tweaker.
And ofc Teamredminer 0.6.1 version.
Have a nice year and happy new year to all.

[ggarbi]

I have a 4 GPU setup with MSI AMD RX 470. They are all the same card. I am running windows 10 64 pro with a Pentium processor and 4gb Ram and a 120 solid state hard Drive. I have my virtual memory set at minimum to 6gb and max at 20gb. I am using NiceHash with the latest version of TeamRedMiner. I for some reason cannot run more than two gpus. If I try to start with a third gpu I get the cannot initialize idx error. Any thoughts what would cause this?
Thanks

I think the problem is virtual driver (paging)
Right clock on my computer -->properties --> go to advance system setting --> Advance tab --> under performance part, click on setting --> Go to advanced --> change virtual memory amount to something like 20000

You can watch a youtube video to see how and how much virtual memory you need for those cards
Good Luck i hope it works

[pbfarmer]

I have a 4 GPU setup with MSI AMD RX 470. They are all the same card. I am running windows 10 64 pro with a Pentium processor and 4gb Ram and a 120 solid state hard Drive. I have my virtual memory set at minimum to 6gb and max at 20gb. I am using NiceHash with the latest version of TeamRedMiner. I for some reason cannot run more than two gpus. If I try to start with a third gpu I get the cannot initialize idx error. Any thoughts what would cause this?
Thanks

Try a minimum page file size of 4 + 4 x #GPUs ==> 20GB or maybe even a touch more...

[dagere76]

Well that did the trick feel like a new cake, set the min from 8 to 16gb and the max from 20 to 32k mb...that fixed it. I have the 8gb cards so I did 8 times 4 cards and got 32k mb.
Thanks for all the help

[ivomm]

After the last fork on 2nd Jan, Ethereum difficulty dropped significantly and currently it is the most profitable to mine by far. But I am reluctant to switch back because I don't know how long the 4GB cards will work under Windows 10. They worked when the DAG was 3.3GB but it is now almost 3.4GB and Windows 10 is reserving 15-20% of the video memory. I don't know the exact percentage.

[PIOUPIOU99]

these are these ?
 


3.39 GB

[tvukoman]

One Vega 56 referent (64 bios, samsung mem) have hardware errors:
[48C, fan 71%]       ethash: 48.36Mh/s, avg 48.54Mh/s, pool 47.27Mh/s a:9112 r:0 hw:164

already lowered clocks and mem in Overdrive Ntool:
960/850 core and 1055/850 mem

memtweak:
--CL 20 --RAS 30 --RCDRD 14 --RCDWR 12 --RC 44 --RP 14 --RRDS 3 --RRDL 6 --RTP 5 --FAW 12 --CWL 8 --WTRS 4 --WTRL 9 --WR 14 --REF 17000 --RFC 249

Another card hash 49.1 with same memtweak and 976/850 1075/850 without hw errors.

Any recomendations?

txs

[BogdanCo]

One Vega 56 referent (64 bios, samsung mem) have hardware errors:
[48C, fan 71%]       ethash: 48.36Mh/s, avg 48.54Mh/s, pool 47.27Mh/s a:9112 r:0 hw:164

already lowered clocks and mem in Overdrive Ntool:
960/850 core and 1055/850 mem

memtweak:
--CL 20 --RAS 30 --RCDRD 14 --RCDWR 12 --RC 44 --RP 14 --RRDS 3 --RRDL 6 --RTP 5 --FAW 12 --CWL 8 --WTRS 4 --WTRL 9 --WR 14 --REF 17000 --RFC 249

Another card hash 49.1 with same memtweak and 976/850 1075/850 without hw errors.

Any recomendations?

txs

Try with 875 mV for both core and memory....

[tvukoman]

One Vega 56 referent (64 bios, samsung mem) have hardware errors:
[48C, fan 71%]       ethash: 48.36Mh/s, avg 48.54Mh/s, pool 47.27Mh/s a:9112 r:0 hw:164

already lowered clocks and mem in Overdrive Ntool:
960/850 core and 1055/850 mem

memtweak:
--CL 20 --RAS 30 --RCDRD 14 --RCDWR 12 --RC 44 --RP 14 --RRDS 3 --RRDL 6 --RTP 5 --FAW 12 --CWL 8 --WTRS 4 --WTRL 9 --WR 14 --REF 17000 --RFC 249

Another card hash 49.1 with same memtweak and 976/850 1075/850 without hw errors.

Any recomendations?

txs

Try with 875 mV for both core and memory....

txs, but that will increase VDDC in TRM to 906 mV and power consumption go up for 20W from the wall.
In all other algo, i can mine without hw errors with 850 mV with that card...

[pbfarmer]

One Vega 56 referent (64 bios, samsung mem) have hardware errors:
[48C, fan 71%]       ethash: 48.36Mh/s, avg 48.54Mh/s, pool 47.27Mh/s a:9112 r:0 hw:164

already lowered clocks and mem in Overdrive Ntool:
960/850 core and 1055/850 mem

memtweak:
--CL 20 --RAS 30 --RCDRD 14 --RCDWR 12 --RC 44 --RP 14 --RRDS 3 --RRDL 6 --RTP 5 --FAW 12 --CWL 8 --WTRS 4 --WTRL 9 --WR 14 --REF 17000 --RFC 249

Another card hash 49.1 with same memtweak and 976/850 1075/850 without hw errors.

Any recomendations?

txs

Not all memory is the same - some can take more aggressive timings, some less.  I would trim back your timings changes to a minimum set (and reduce REF), and build from there:

Code:

--REF 15600 --RFC 248 --RAS 30 --RP 12 --RC 44 --RCDRD 14

This will prob get you to 99.5%+ of your list anyway.  If these still don't work - try reducing REF more (try 11700), then maybe RC, then RP - those are your biggest movers.

Also, are you sure you need 850mv for those settings?  All of my vegas (including the flashed 56) are running 930cclk/1028mclk @ 800mv for 47.4MH/s.

EDIT: Take a look at your temperatures (esp mem) on the two different cards as well.  Higher mem temps cause faster charge decay requiring more frequent refreshes (i.e. lower REF).  Mine do just fine w/ REF=15600, but mem temps are all in the high 50's to low 60's.

[tvukoman]

One Vega 56 referent (64 bios, samsung mem) have hardware errors:
[48C, fan 71%]       ethash: 48.36Mh/s, avg 48.54Mh/s, pool 47.27Mh/s a:9112 r:0 hw:164

Any recomendations?

txs

Not all memory is the same - some can take more aggressive timings, some less.  I would trim back your timings changes to a minimum set (and reduce REF), and build from there:

Code:

--REF 15600 --RFC 248 --RAS 30 --RP 12 --RC 44 --RCDRD 14

This will prob get you to 99.5%+ of your list anyway.  If these still don't work - try reducing REF more (try 11700), then maybe RC, then RP - those are your biggest movers.

Also, are you sure you need 850mv for those settings?  All of my vegas (including the flashed 56) are running 930cclk/1028mclk @ 800mv for 47.4MH/s.

EDIT: Take a look at your temperatures (esp mem) on the two different cards as well.  Higher mem temps cause faster charge decay requiring more frequent refreshes (i.e. lower REF).  Mine do just fine w/ REF=15600, but mem temps are all in the high 50's to low 60's.

txs pbfarmer! With your memtweak timings just finished test 24 hour with 0 hw errors ;-)

HBM temperature is low 70's on all cards (can't do much on that, small room with no window for cooling, just one fan, and few holes for fresh air)

Never tried so low mV (800), i don't have power play table for 800, is it possible to running 800 mV with ppt build for 850 mV?