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

[dragonmike]

Devs, turtlecoin devs will implement Argon2iD at height 1 800 000, could you implement this ago in your miner ?

Hi, just saw that today as well. I can't make any promises at this point though, it's a very small use case for a new algo .

Argon2iD... Excuse my ignorance but isn't that a CPU-only algo? Maybe I'm confusing it with something else.

the difference here is that power use is so tightly coupled w/ the 'intensity' setting, and can swing so significantly w/ an incremental change.

Bottom line, that's 18.5 kh/s for ~138w vs 19.5 kh/s for ~155w on average, without all the trouble searching for intensity, clock, and power combinations.

Btw, I've also tested this on my V64 flashed ref V56, and i'm seeing similar efficiency gains (19.4 -> 18.3kh/s, 150 -> 135w,) but w/ cn_config=L22+22

Nice finds.
Btw Teamviewer screws massively with stability when using low clocks - which would correlate with the findings from your first statement above.

I've actually changed my overdriventool bat file to first set higher clocks so the miner initialises and ramps up mining properly before it lowers the clocks 60 seconds later. Will test using even lower clocks in combination with lower cn_config and see how much more efficiently I can get the cards to run.

What voltage did you set for 850 cclock (p0)? 825mV?

[1715001484]

1715001484

[1715001484]

1715001484

[1715001484]

1715001484

[seefatlow]

@pbfarmer: I'm currently testing my Vegas at 1258 core (OT setting) @ 807mV and they seem to hold for now (except for two weaker cards that won't have anything less than 850mV even at low clocks).

EDIT: 807 didn't cut it. It mined as I changed it on the fly, but would not initialise at this voltage. 815 seems to work (both init and mining) - Couple of hours going, so far so good.

EDIT2: 815 held for a few hours but ended up crashing in the end. Reducing core clocks further doesn't seem to increase stability, maybe there really is a balance to be found.

Yeah, this is a tough nut to crack - esp for those couple salty cards you inevitably run into.  I'm kinda figuring out that you have to tune this a bit backwards from the normal process - stability seems to be all about mem (SOC) speed + cn_config.  Core speed tuning on it's own doesn't do much - cclocks only real importance seems to be in determining which cn_config you can use.  This is roughly what i'm seeing:

L28+28: >= 1150mhz
L26+26: 1050mhz - 1150mhz
L22+22: 950mhz - 1050mhz (not sure about this one...)
L18+18: 850mhz - 950?mhz

The relationship isn't cleanly linear - i've seen cn_config 'holes' where for instance, in one case L26+26 is fine, L24+24 is crap, L22+22 is good again.  And in other cases, I've run into tight couplings, where even reducing cn_config incrementally caused instability.

I'm tempted to just run all my vegas like i do for ethash - 850 (core-p0) / 1107 (mem-p3) / L18+18 - and then tune voltage as needed.  It's straightforward (no acg, everything fixed except voltage,) low power, and i still get 18.5kh/s out of it.

EDIT: cn_config numbers are for V64 - maybe scale down by 4 for V56?

I just don't understand how you guys are able to pull off <850mV voltage at the said memclk . My one and only MSI Reference Vega64 at the very same setting is only stable above 870mV.

Post your overdriventool profile in here please.
Also, have you got your SoC clock set to 1200 or is it still at the stock 1107? That's been the culprit for many people, including me, to get this thing to work properly.

My ODNT profile here

Name=RX_Vega_64
GPU_P0=852;800;0
GPU_P1=991;800;0
GPU_P2=1084;805;0
GPU_P3=1138;805;0
GPU_P4=1150;850;0
GPU_P5=1202;850;0
GPU_P6=1212;850;0
GPU_P7=1408;870
Mem_P0=167;800;0
Mem_P1=500;800;0
Mem_P2=800;805;0
Mem_P3=1040;870


Oh....I checked my SoC clock, it at 1107. So how do I set it to 1200?

[seefatlow]

As an update from more testing by yours truly...

Lower clocks have a larger overall stability impact than I thought they would. It seems that going below 1250 MHz core clock on Vegas comes with LESS stability at the same voltage. pbfarmer is probably on to something by tweaking the cn_configs downwards win parallel with lowering clocks.

So it seems that 1250@825 core is a sweetspot (mem clocks always remain unchanged at 1107) for Vega56's at L24+24. They still hash north of 19.5 KH/s there.

Probably won't bother trying to go lower with lower intensities as that's where hashrate starts to drop.

My rig of 6 Vegas draws less than 1170W at the wall using the settings above on TRTL+Loki. That's quite remarkable.

Dragonmike,

6 Vegas at 1170W ATW isn't anywhere near pbfarmer's published 165W per Vega GPU. The weirdest thing is that my 870mV settings is already at the same level of efficiency as yours and i thought I am doing much worse in terms of efficiency as y'all.

[dragonmike]

@pbfarmer: I'm currently testing my Vegas at 1258 core (OT setting) @ 807mV and they seem to hold for now (except for two weaker cards that won't have anything less than 850mV even at low clocks).

EDIT: 807 didn't cut it. It mined as I changed it on the fly, but would not initialise at this voltage. 815 seems to work (both init and mining) - Couple of hours going, so far so good.

EDIT2: 815 held for a few hours but ended up crashing in the end. Reducing core clocks further doesn't seem to increase stability, maybe there really is a balance to be found.

Yeah, this is a tough nut to crack - esp for those couple salty cards you inevitably run into.  I'm kinda figuring out that you have to tune this a bit backwards from the normal process - stability seems to be all about mem (SOC) speed + cn_config.  Core speed tuning on it's own doesn't do much - cclocks only real importance seems to be in determining which cn_config you can use.  This is roughly what i'm seeing:

L28+28: >= 1150mhz
L26+26: 1050mhz - 1150mhz
L22+22: 950mhz - 1050mhz (not sure about this one...)
L18+18: 850mhz - 950?mhz

The relationship isn't cleanly linear - i've seen cn_config 'holes' where for instance, in one case L26+26 is fine, L24+24 is crap, L22+22 is good again.  And in other cases, I've run into tight couplings, where even reducing cn_config incrementally caused instability.

I'm tempted to just run all my vegas like i do for ethash - 850 (core-p0) / 1107 (mem-p3) / L18+18 - and then tune voltage as needed.  It's straightforward (no acg, everything fixed except voltage,) low power, and i still get 18.5kh/s out of it.

EDIT: cn_config numbers are for V64 - maybe scale down by 4 for V56?

I just don't understand how you guys are able to pull off <850mV voltage at the said memclk . My one and only MSI Reference Vega64 at the very same setting is only stable above 870mV.

Post your overdriventool profile in here please.
Also, have you got your SoC clock set to 1200 or is it still at the stock 1107? That's been the culprit for many people, including me, to get this thing to work properly.

My ODNT profile here

Name=RX_Vega_64
GPU_P0=852;800;0
GPU_P1=991;800;0
GPU_P2=1084;805;0
GPU_P3=1138;805;0
GPU_P4=1150;850;0
GPU_P5=1202;850;0
GPU_P6=1212;850;0
GPU_P7=1408;870
Mem_P0=167;800;0
Mem_P1=500;800;0
Mem_P2=800;805;0
Mem_P3=1040;870


Oh....I checked my SoC clock, it at 1107. So how do I set it to 1200?

No it shouldn't be at 1200, you're all fine.
Try to set your GPU_P7 and Mem_P3 voltages at 850. Don't disable the lower states, just make sure they are set to lower clocks & voltages than the highest, for example like this:

Name=RX_Vega_64
GPU_P0=852;800
GPU_P1=991;803
GPU_P2=1084;805
GPU_P3=1138;810
GPU_P4=1150;820
GPU_P5=1202;830
GPU_P6=1212;840
GPU_P7=1408;850
Mem_P0=167;800
Mem_P1=500;805
Mem_P2=800;810
Mem_P3=1040;850

[dragonmike]

As an update from more testing by yours truly...

Lower clocks have a larger overall stability impact than I thought they would. It seems that going below 1250 MHz core clock on Vegas comes with LESS stability at the same voltage. pbfarmer is probably on to something by tweaking the cn_configs downwards win parallel with lowering clocks.

So it seems that 1250@825 core is a sweetspot (mem clocks always remain unchanged at 1107) for Vega56's at L24+24. They still hash north of 19.5 KH/s there.

Probably won't bother trying to go lower with lower intensities as that's where hashrate starts to drop.

My rig of 6 Vegas draws less than 1170W at the wall using the settings above on TRTL+Loki. That's quite remarkable.

Dragonmike,

6 Vegas at 1170W ATW isn't anywhere near pbfarmer's published 165W per Vega GPU. The weirdest thing is that my 870mV settings is already at the same level of efficiency as yours and i thought I am doing much worse in terms of efficiency as y'all.

That's because he's running them at lower clocks and intensities. I haven't gone there yet.
1170W at the wall is with the Vegas at 1258MHz@825mV, with cn_config L24+24. This achieves 19.55kh/s per card, or pretty much exactly 100h/W.

I'll try to match pbfarmer's tests when I have more time later.

[pbfarmer]

What voltage did you set for 850 cclock (p0)? 825mV?

Anywhere from 806 (could prob go 800) to 825, w/ one pos at 831.  Avg across 10 Vegas is 812-818.

[livada]

how use reserve pool?

i wont use 2 o3 more pool if first dead.

[kerney666]

how use reserve pool?

i wont use 2 o3 more pool if first dead.

One of the few remaining features we need to add before dropping the beta tag. Should have done it a long time ago .

[livada]

how use reserve pool?

i wont use 2 o3 more pool if first dead.

One of the few remaining features we need to add before dropping the beta tag. Should have done it a long time ago .

ok thx and go on work

[dragonmike]

Running 6 Vega56@64 @ 850 MHz @ 825mV (850mV for two weaker cards) on core_P7.
Getting approx 98 KH/s combined at 950W at the wall.

That's actually hilarious. This is pretty much roughly 150W per Vega. 103 h/W. L18+18.

L20+20 results in roughly the same figures. More interleave adjusts though.

L22+22 gives a whole bunch of interleave adjusts (all GPUs) and no improvements in hashrate from the values tested above... and ends up crashing after a while.

EDIT: @pbfarmer, I've not managed 18.3 kh/s as you seem to though. More like 16.5 at 850/1107. Using Core_P7 though, not sure that's what's making the difference?

[livada]

Running 6 Vega56@64 @ 850 MHz @ 825mV (850mV for two weaker cards) on core_P7.
Getting approx 98 KH/s combined at 950W at the wall.

That's actually hilarious. This is pretty much roughly 150W per Vega. 103 h/W. L18+18.

L20+20 results in roughly the same figures. More interleave adjusts though.

L22+22 gives a whole bunch of interleave adjusts (all GPUs) and no improvements in hashrate from the values tested above... and ends up crashing after a while.

EDIT: @pbfarmer, I've not managed 18.3 kh/s as you seem to though. More like 16.5 at 850/1107. Using Core_P7 though, not sure that's what's making the difference?

6*vega56 nitro+(hynix) stock bios - 107800hr L28+28 (al card have diff mem and core) (mem 897-936mhz 880mv)(gpu 1450-1470 870mv)-- 1040W at the wall (Rig have  6extra fan) TEM=56° - fan= 1900RPM

6*vega56 nitro+(hynix) stock bios - 109300hr L24+24 (al card have diff mem and core) (mem 897-936mhz 880mv)(gpu 1450-1470 870mv)-- 1090W at the wall (Rig have  6extra fan) TEM=56° - fan= 2000RPM

800mv PPT table 1107SOC

Vega 56 ref(samsung) with 64bios - L24+24 best - 19500hr 1230/870mv-gpu and 1100/870mv mem - 120/127W in GPU-Z(temp=57° fan=3000pm)
this card go more up but my power is to low for 1400gpu

Vega 56 ref(samsung) with 64bios - L28+28 - 19250hr 1230/866mv-gpu and 1100/668mv mem - 112/122W in GPU-Z(temp=57° fan=2700pm)

[dragonmike]

Vega 56 ref(samsung) with 64bios - L24+24 best - 19500hr 1230/870mv-gpu and 1100/870mv mem - 120/127W in GPU-Z(temp=57° fan=3000pm)
this card go more up but my power is to low for 1400gpu

Vega 56 ref(samsung) with 64bios - L28+28 - 19250hr 1230/866mv-gpu and 1100/668mv mem - 112/122W in GPU-Z(temp=57° fan=2700pm)

You should be able to lower your core clocks even further to 850 or even 825. Also, max your mem clocks to 1107. That's as far as you can go on equivalent SoC.

I've noticed that accessing Teamviewer will crash my rig when running low clocks and high intensities. Also, I'm now using a staggered approach: I set higher clocks and voltages, start the miner, and once all cards are up and hashing, apply lower clocks. Just a little script with overdriventool on a timer. Works wonders.

[livada]

Vega 56 ref(samsung) with 64bios - L24+24 best - 19500hr 1230/870mv-gpu and 1100/870mv mem - 120/127W in GPU-Z(temp=57° fan=3000pm)
this card go more up but my power is to low for 1400gpu

Vega 56 ref(samsung) with 64bios - L28+28 - 19250hr 1230/866mv-gpu and 1100/668mv mem - 112/122W in GPU-Z(temp=57° fan=2700pm)

You should be able to lower your core clocks even further to 850 or even 825. Also, max your mem clocks to 1107. That's as far as you can go on equivalent SoC.

I've noticed that accessing Teamviewer will crash my rig when running low clocks and high intensities. Also, I'm now using a staggered approach: I set higher clocks and voltages, start the miner, and once all cards are up and hashing, apply lower clocks. Just a little script with overdriventool on a timer. Works wonders.

i try all. i try your PPT but 865mv is limit. if i go on 860 or 850  rig stop work.

(SAMSUNG)if i go on mem 1102 or higher  gpu dead.
hynix v56 nitro+ is bad card- 1 card work only on 897mhz- best nitro+ card work only on 936mhz.
nitro+ hyny with 64bios unstable and i not use this(i try but unstable)

Teamviewer crash my rig only on trtl. on other algo work fine.

this is 6* vega 56 nitro+(hynix) rig stock bios - 109300hr 1088W at the wall (100h/w)
http://imgbox.com/nrciqXaT

Name=trtl5
GPU_P0=852;800;0
GPU_P1=991;801;0
GPU_P2=1084;802;0
GPU_P3=1100;803;0
GPU_P4=1125;804;0
GPU_P5=1150;805;0
GPU_P6=1300;815;0
GPU_P7=1470;860
Mem_P0=167;800;0
Mem_P1=500;800;0
Mem_P2=800;801;0
Mem_P3=929;870


http://imgbox.com/mD5vGUkT-This is rig with 3*vega56@64 and 1*Vega 64LE. Rig problem is suply - no power for this card. 760W at the wall.(v56=L24+24 v64=L28+28)=77870hr-work only 10min with this hr and go down to 76900hr/745W

http://imgbox.com/WpzKtdL7- This is rig with 3*vega56@64 and 1*Vega 64LE. Rig problem is suply - no power for this card. 750W at the wall.(v56=L28+28 v64=L28+28)=77200hr
L28+L28 for all card. L24 is beter for vag56 but use more power adn suply reset rig.
this rig have 77200hr 750W at the wal. 4extra fan  and vega 64LE use 30W more for pump.
750W-80(MB)-30W(pump)= 640W for 3*V56 +1*V64 (160W per card) = 103h/w

vega 56
Name=trtl1
GPU_P0=852;825
GPU_P1=991;825
GPU_P2=1084;825
GPU_P3=1100;825
GPU_P4=1115;825
GPU_P5=1125;825
GPU_P6=1130;825
GPU_P7=1210;870
Mem_P0=167;800;0
Mem_P1=500;800;0
Mem_P2=800;900;0
Mem_P3=1100;870
Fan_Min=2300
Fan_Max=3200
Fan_Target=57
Fan_Acoustic=2400
Power_Temp=75
Power_Target=0

vega 64LE
Name=trtl3
GPU_P0=852;825
GPU_P1=991;825
GPU_P2=1084;825
GPU_P3=1100;825
GPU_P4=1115;825
GPU_P5=1120;825
GPU_P6=1135;850
GPU_P7=1145;880
Mem_P0=167;800;0
Mem_P1=500;800;0
Mem_P2=800;800;0
Mem_P3=1098;880
Fan_Min=1100
Fan_Max=2400
Fan_Target=49
Fan_Acoustic=1200
Power_Temp=70
Power_Target=0

all cards can be given a lot more if they are unable to power(1000Wbronce give only 760W)

[tvukoman]

I've noticed that accessing Teamviewer will crash my rig when running low clocks and high intensities. Also, I'm now using a staggered approach: I set higher clocks and voltages, start the miner, and once all cards are up and hashing, apply lower clocks. Just a little script with overdriventool on a timer. Works wonders.

Teamviewer crash my rig only on trtl. on other algo work fine.

For local network on win10 pro you should try remote desktop app. Great, simple, Microsoft app, never have problems.

On android tablet I use the same app to connect on remote rig.

[pbfarmer]

Running 6 Vega56@64 @ 850 MHz @ 825mV (850mV for two weaker cards) on core_P7.
Getting approx 98 KH/s combined at 950W at the wall.

That's actually hilarious. This is pretty much roughly 150W per Vega. 103 h/W. L18+18.

L20+20 results in roughly the same figures. More interleave adjusts though.

L22+22 gives a whole bunch of interleave adjusts (all GPUs) and no improvements in hashrate from the values tested above... and ends up crashing after a while.

EDIT: @pbfarmer, I've not managed 18.3 kh/s as you seem to though. More like 16.5 at 850/1107. Using Core_P7 though, not sure that's what's making the difference?

using p7 will engage ACG while p0 does not (or at least shouldn't), so I could see that possibly causing differences...  If 850 is your setting, I would expect you're actually running lower cclock, which may explain lower cn_config and h/r.  If you're running 850 actual, not sure why you would see different behavior.  I am on linux, but my quick tests showed windows to be similar (at least at higher clocks).

[osirus466]

I have an 8 card rig getting 8k/s total.

5- MSI RX470 8GB mining edition
3- Nitro RX570 4GB

Im getting 980 h/s from the 470s and 1k/s from the 570s but at 125 watts per card.  Total rig power at the wall is 1000watts.  Is there any way I van lower the power draw???

[kerney666]

I have an 8 card rig getting 8k/s total.

5- MSI RX470 8GB mining edition
3- Nitro RX570 4GB

Im getting 980 h/s from the 470s and 1k/s from the 570s but at 125 watts per card.  Total rig power at the wall is 1000watts.  Is there any way I van lower the power draw???

Hi! I daresay this miner is the most power efficient miner for all the variants we support. The tools at your disposal are bios and clocks tweaks. There are other people here that have trimmed their Polaris rigs much more than I have, hopefully then can pitch in.

As a start, can you tell us a little more about your setup? What clocks do you run the cards at currently?

[vmozara]

Guys,

I just came back from the ship and wanted to try mining new algos.

So far for the Vega I was using windows 10 and adrenalin 18.6.1, ppt and classis disable-enable-clocks-miner combo, and that worked perfectly for me.

But now i tried to install new adrenalin, apply PPT, and tried to mine Turtle, but i can see that my friequencies both core and memory are jumping around and i only get 14-16KH/s

From what I understood, Turtle mining is very low consumption and heating, I would like to switch to it as summer is coming, and CN-R is quite demanding for cooling.

Can somebody make a short guide what steps are necessary with new adrenalin drivers? Thanks guys!

[QuirkSilver]

I am thinking about mining Monero (cryptonight R). I have 3 290x gpu's. My electricity is about 12 cent. As far as I can see profitability will be weak to say the least, should I wait for Monero to rise in value or would the hashrate kill of profitability anyways??

Beyerd, I ABUSED 6 290x/390 , had two near new 290x gigabyte, one had  the VRM burnt for some reason (this one was unstable anyway, think the gpu had a grit of sand from factory and heat damagge happened probably with first user)
The other with the Stilt Bios and timings mods, used to do REALLY low reading, fans run lightly, it gave around 90 watts total in HWinfo mining v7 CN back in the day.
What i would do is go 1000 mhz only and undervolt-experiment (every miner is different) sometimes it can take -200 mv, the stilt bios changes VRM frequency which isa hard tweak, however they really ran with less energy.  used to do 900 h/s back in the day.... also i know what to be done with the timings for polaris but with this 290x, you need to underclock and tweak timings specifically as hell for cryptonight, what works well for Cryptonight always is, lower tRD and tRC, higher tCL and consequent timings, loosen tRFC substantially and bump Rasmactwr mainly and rasmactrd. try with 1000 mhz memory speed with tight timings and start  overclocking memory, no errors, then maybe tighten more.... elpida in polaris does miracles at lower speed than even 1750 mhz with proper timings

[kerney666]

Guys,

I just came back from the ship and wanted to try mining new algos.

So far for the Vega I was using windows 10 and adrenalin 18.6.1, ppt and classis disable-enable-clocks-miner combo, and that worked perfectly for me.

But now i tried to install new adrenalin, apply PPT, and tried to mine Turtle, but i can see that my friequencies both core and memory are jumping around and i only get 14-16KH/s

From what I understood, Turtle mining is very low consumption and heating, I would like to switch to it as summer is coming, and CN-R is quite demanding for cooling.

Can somebody make a short guide what steps are necessary with new adrenalin drivers? Thanks guys!

Hey vmozara! So, what's the cn_config and clocks you're using? Personally, my Vegas have been doing fine on 18.6.1 and 19.2.2, I haven't bumped 19.2.2 -> 19.3.x or 19.4.y on my dev workstation yet. Turtle is a little b*tchy, for many people the perf is great out of the box, for others it's a circus. Linux has been serving people better than win10. I'm not convinced about the latest and greatest Adrenalin drivers...

Myself, I can just start the miner at any clocks using L24+24 (V56 flashed V64) or L28+28 (proper V64), and I'm at 19.5-19.6 kh/s.