[ANN] TeamRedMiner - Old Thread

[bbvedf]

I appreciate the advice. I will take a look.
Thanks!

I'm sorry again for the offtopic.
Just confirm that with the recommendation of pbfarmer I have gone from 1.60gps to 1.95gps in Sapphire Pulse RX580.
Thanks!

[freedom777]

So you  switch core voltage in bios to +/- 1000mV(reprogramming stock bios)? Or just in any software program in OS(like -200mV, i think what it is not possible for stable work)?

I have modified bios yeah: unlocked 60CU, applied 400Mhz mem timings to 500MHz, undervolted for gaming, but for mining in most cases you can even lower on voltage with the same clock. And since I had to go lower on clock(P7-state) as well I just adjusted P-states and voltages in Wattman. I dunno if OverdriveNTool supports Fiji, when I was using one of my furies in my PC it hadn't existed yet). But AB is definitely no-go for tuning Fury since it doesn't read its P-states - basically same thing as curve OC for nvidia.

Do you have a ready mod BIOS, or only manual tuning? What program is relevant for this today, I have not modified the BIOS for a long time (the last time it was Hawaii and I don’t remember exactly which programs were used). Maybe you have some link with manual of this question?

[Divinity666]

Do you have a ready mod BIOS, or only manual tuning? What program is relevant for this today, I have not modified the BIOS for a long time (the last time it was Hawaii and I don’t remember exactly which programs were used). Maybe you have some link with manual of this question?

No, I don't have bios ready to mine lyra right on - I still gotta use wattman to set core clocks and voltages. For bios editing check here: https://www.overclock.net/forum/67-amd/1592384-fiji-bios-editing-fury-fury-x-nano-radeon-pro-duo.html Though you dont really need it unless your cards can be ulocked, as far as I know only Asus Strix and Sapphire Tri-X(not Nitro) can.

[fancyIX]

@kerney666, what about MTP algo, do you plan to add it?

The main part of MTP is dead simple, I'm truly amazed there aren't good open source miners out there already.

For the algos we add to this miner, we only select those where we truly feel we can add enough persistent value to warrant a higher-than-normal dev fee. MTP should/will have good miners shortly, or it's really an embarrasing fail for open source in crypto. Moreover, with the hack they've added with a binary stratum protocol to keep effective bandwidth down, this is a ton of work for us in this miner. Otoh, it would be very simple to add it to e.g. a sgminer clone. Last, I think AMD cards will have a hard time beating Nvidia from an efficiency perspective for this algo, which also means adoption might be much smaller than you'd hope for as a miner dev.

Given all this, the probability that we would ever get the time spent on MTP back is tiny, so we've opted not to do it. Tbh, if I had the time I would rather spend it on an open source implementation.

 

Kerney666 its been more than 2 months, i think a high fee can be warranted still no stable amd miners available bro.

Yeah, I'm still amazed. I haven't kept track of what djm34 and fancyIX ended up with though, is it still crap? Last I checked they were getting closer to the nrs I would expect, 580 maybe 1.4-1.6 MH/s, Vegas should be able to do 3.1-3.5 MH/s. You need the compute for MTP, so 64s will outperform 56s.

Only djm34 is working on the kernel for now. So there is no "they".
Anyway, I have a stable release:
https://github.com/fancyIX/sgminer-phi2-branch/releases

[fancyIX]

Lyra2REv3 TEST

VEGA 56 84mh/s 1408/900 950/910 200w

VEGA 64 air coole 96mhs/s 1408/900 1000/900 200w

rx570 45mh/s 1290/900 2000/880  110w

Wow, really good rate.

[ku4eto]

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

[iRonNuke]

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

Yeah this algo isn't about memory it seems. I also dumped my Vegas at 500 MHz and RX series at 300 MHz. No big difference in hashrate.

My rig have the following cards;
3 x RX 560 4 GB 18 MH/s
4 x RX 570 4 GB 39 MH/s
2 x RX 580 8 GB 43-44 MH/s
2 x Vega 56 85-86 MH/s

My total rig wattage is exactly 1000W at 470 MH/s. (Lower power consumption than at CryptoNight V8!). Because I have quite expensive electricity I have quite aggressive undervoltating, 820 mV at core for RX 570/580 and memory all down to 600 mV. The Vega's have like 880 mV at core and 800 mV at memory (can't set lower because restriction in drivers apperantly. At this rates the rig is rock solid stable, havn't crashed one single time since like the fork of Vertcoin.

Till now, with my expensive electricity 0.1268$ kWh it have been profitable, like 1.5$-0.3$ (yesterday), but not anymore. So now the fun is over

[dragonmike]

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

How do you actually prevent it to go to anything higher than P0? Do you just add ";0" at the end of P1, P2 and P3 lines in Overdriventool? Will it even start if you do that? Mem clock is set at something ludicrously low (167 MHz) at P0...

[dragonmike]

Bloody hell it does work... Memory is shown at 500MHz (on the Vega 56's).
I'm testing on x16RT currently (which isn't very memory bound either) and I've dropped power draw massively. Hallelujah.

EDIT: on the Vegas, disabling P1 to P3 works, on the Polaris cards it won't, so I set the values of P1 and P2 equal to P0.

[UnclWish]

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

Yeah this algo isn't about memory it seems. I also dumped my Vegas at 500 MHz and RX series at 300 MHz. No big difference in hashrate.

My rig have the following cards;
3 x RX 560 4 GB 18 MH/s
4 x RX 570 4 GB 39 MH/s
2 x RX 580 8 GB 43-44 MH/s
2 x Vega 56 85-86 MH/s

My total rig wattage is exactly 1000W at 470 MH/s. (Lower power consumption than at CryptoNight V8!). Because I have quite expensive electricity I have quite aggressive undervoltating, 820 mV at core for RX 570/580 and memory all down to 600 mV. The Vega's have like 880 mV at core and 800 mV at memory (can't set lower because restriction in drivers apperantly. At this rates the rig is rock solid stable, havn't crashed one single time since like the fork of Vertcoin.

Till now, with my expensive electricity 0.1268$ kWh it have been profitable, like 1.5$-0.3$ (yesterday), but not anymore. So now the fun is over

How you set memory voltage to 600mV on RX580? It's minimal 750... I tried to set less than 750mV but it really stay at 750mV.

[fluxy12]

How you set memory voltage to 600mV on RX580? It's minimal 750... I tried to set less than 750mV but it really stay at 750mV.

It doesn't change anything, same comsuption at 600 than 820.

[ku4eto]

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

How do you actually prevent it to go to anything higher than P0? Do you just add ";0" at the end of P1, P2 and P3 lines in Overdriventool? Will it even start if you do that? Mem clock is set at something ludicrously low (167 MHz) at P0...

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman. Seems like going below 1750Mhz on memory forces the driver to use P0 if the load is sufficiently low for some reason.

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

Yeah this algo isn't about memory it seems. I also dumped my Vegas at 500 MHz and RX series at 300 MHz. No big difference in hashrate.

My rig have the following cards;
3 x RX 560 4 GB 18 MH/s
4 x RX 570 4 GB 39 MH/s
2 x RX 580 8 GB 43-44 MH/s
2 x Vega 56 85-86 MH/s

My total rig wattage is exactly 1000W at 470 MH/s. (Lower power consumption than at CryptoNight V8!). Because I have quite expensive electricity I have quite aggressive undervoltating, 820 mV at core for RX 570/580 and memory all down to 600 mV. The Vega's have like 880 mV at core and 800 mV at memory (can't set lower because restriction in drivers apperantly. At this rates the rig is rock solid stable, havn't crashed one single time since like the fork of Vertcoin.

Till now, with my expensive electricity 0.1268$ kWh it have been profitable, like 1.5$-0.3$ (yesterday), but not anymore. So now the fun is over

How you set memory voltage to 600mV on RX580? It's minimal 750... I tried to set less than 750mV but it really stay at 750mV.

You can always mod the BIOS and set the IMC voltage lower. Although, i wouldnt recommend going too low.

[dragonmike]

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman.

It's really the first time EVER I've read the above.
If anything is annoying as hell, and buggy to death and beyond, it's Wattman. Overdriventool is the easiest, most straightforward and logical piece of software to set clocks for AMD. The first thing I do when I install new drivers is to make 300% sure "Radeon Settings" is unchecked.

[ku4eto]

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman.

It's really the first time EVER I've read the above.
If anything is annoying as hell, and buggy to death and beyond, it's Wattman. Overdriventool is the easiest, most straightforward and logical piece of software to set clocks for AMD. The first thing I do when I install new drivers is to make 300% sure "Radeon Settings" is unchecked.

Sorry, but i do not rely a lot on 3rd party tools, that i would need to wait for a new version, if they suddenly stop working because of the new drivers. Wattman is easy to use and has all the bells and whistles i need. If you do not like it, do not use it. There is nothing buggy about it.

[fluxy12]

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman.

It's really the first time EVER I've read the above.
If anything is annoying as hell, and buggy to death and beyond, it's Wattman. Overdriventool is the easiest, most straightforward and logical piece of software to set clocks for AMD. The first thing I do when I install new drivers is to make 300% sure "Radeon Settings" is unchecked.

Sorry, but i do not rely a lot on 3rd party tools, that i would need to wait for a new version, if they suddenly stop working because of the new drivers. Wattman is easy to use and has all the bells and whistles i need. If you do not like it, do not use it. There is nothing buggy about it.

It is impossible you tried OverdriveNTool you would not have said such foolishness

[pbfarmer]

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman.

It's really the first time EVER I've read the above.
If anything is annoying as hell, and buggy to death and beyond, it's Wattman. Overdriventool is the easiest, most straightforward and logical piece of software to set clocks for AMD. The first thing I do when I install new drivers is to make 300% sure "Radeon Settings" is unchecked.

Sorry, but i do not rely a lot on 3rd party tools, that i would need to wait for a new version, if they suddenly stop working because of the new drivers. Wattman is easy to use and has all the bells and whistles i need. If you do not like it, do not use it. There is nothing buggy about it.

It is absolutely buggy - stopped using it a couple days after first trying (for mining) because of all the bugs.  The only time ODNT has required an update was during the transition from 18.x to 19.x drivers, due to the introduction of mem profiling and some other stuff.

[ku4eto]

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman.

It's really the first time EVER I've read the above.
If anything is annoying as hell, and buggy to death and beyond, it's Wattman. Overdriventool is the easiest, most straightforward and logical piece of software to set clocks for AMD. The first thing I do when I install new drivers is to make 300% sure "Radeon Settings" is unchecked.

Sorry, but i do not rely a lot on 3rd party tools, that i would need to wait for a new version, if they suddenly stop working because of the new drivers. Wattman is easy to use and has all the bells and whistles i need. If you do not like it, do not use it. There is nothing buggy about it.

It is impossible you tried OverdriveNTool you would not have said such foolishness

Pretty sure i did. I did not like it. Do i have to like it? Its called personal opinion.

I do not use OverDrivenTool, its annoying as hell. Simply downclocked via Wattman.

It's really the first time EVER I've read the above.
If anything is annoying as hell, and buggy to death and beyond, it's Wattman. Overdriventool is the easiest, most straightforward and logical piece of software to set clocks for AMD. The first thing I do when I install new drivers is to make 300% sure "Radeon Settings" is unchecked.

Sorry, but i do not rely a lot on 3rd party tools, that i would need to wait for a new version, if they suddenly stop working because of the new drivers. Wattman is easy to use and has all the bells and whistles i need. If you do not like it, do not use it. There is nothing buggy about it.

It is absolutely buggy - stopped using it a couple days after first trying (for mining) because of all the bugs.  The only time ODNT has required an update was during the transition from 18.x to 19.x drivers, due to the introduction of mem profiling and some other stuff.

The only bug that i have encountered - the Zero Fan one, was fixed with the last update. So, still, no bugs for me.

[pbfarmer]

Boy, dumping the memory down to P0 saves A LOT of power.

RX 580, at 1240/1750 does 49.6Mh/s at ~100W per GPU (as per GPU-Z).
Rig has 3x 580, 1x570 and was doing ~530W at wall.

Entered memory P0 state, and lost ~20W per 580 according to GPU-Z. Wall power consumption dropped from 530 to 440W.
Performance dropped with ~0.6MH/s per card.

Best performance/power consmption seems to be around 1200-1250Mhz.

Getting 190.8Mh/s for 440W at wall with 4 cards.

RX 570 - ~45.9MH/s
3x RX 580 - ~49.08MH/s.
Core clocks are all 1240Mhz. Core voltage is ~0.875V-0.893V (2x 0.875, 1x0.887, 1x0.893V).

Raising core voltage to sustain 1280Mhz core raises performance to 50.45MH/s per RX 580, but raises power consumption by ~30W (from 440 to 470). So, 10% power increase for 1% performance increase.

Yeah this algo isn't about memory it seems. I also dumped my Vegas at 500 MHz and RX series at 300 MHz. No big difference in hashrate.

My rig have the following cards;
3 x RX 560 4 GB 18 MH/s
4 x RX 570 4 GB 39 MH/s
2 x RX 580 8 GB 43-44 MH/s
2 x Vega 56 85-86 MH/s

My total rig wattage is exactly 1000W at 470 MH/s. (Lower power consumption than at CryptoNight V8!). Because I have quite expensive electricity I have quite aggressive undervoltating, 820 mV at core for RX 570/580 and memory all down to 600 mV. The Vega's have like 880 mV at core and 800 mV at memory (can't set lower because restriction in drivers apperantly. At this rates the rig is rock solid stable, havn't crashed one single time since like the fork of Vertcoin.

Till now, with my expensive electricity 0.1268$ kWh it have been profitable, like 1.5$-0.3$ (yesterday), but not anymore. So now the fun is over

How you set memory voltage to 600mV on RX580? It's minimal 750... I tried to set less than 750mV but it really stay at 750mV.

Mem voltage in the profiling tools is not really mem voltage - it's just a floor for core voltage (it's really for the SOC / mem controller.)  So unless you're planning on trying to run your core at 600, there's no use setting mem that low.  Simply make sure it's below core voltage, and it's basically unused.

If you really want to change your mem voltage (which is about 1.6v) you need to get into the bios.

[UnclWish]

How you set memory voltage to 600mV on RX580? It's minimal 750... I tried to set less than 750mV but it really stay at 750mV.

Mem voltage in the profiling tools is not really mem voltage - it's just a floor for core voltage (it's really for the SOC / mem controller.)  So unless you're planning on trying to run your core at 600, there's no use setting mem that low.  Simply make sure it's below core voltage, and it's basically unused.

If you really want to change your mem voltage (which is about 1.6v) you need to get into the bios.

Of course I was mentioned memory controller voltage not memory itself... But I can't set any voltage lower 750 mV on it. If I set lower it stays 750.

[Drakyla]

VEGA FE 93 MH/s
VEGA 56 86-89 MH/s

anyone know how to control the fan manually? my vega FE runs at 40%, I'd like it to run at 50%. tia

You can try this command (keep in mind its global for all cards , this sh file will not control only one of them) .

wget http://drakyla.org/atifans.sh && chmod 755 atifans.sh

After this do ./atifans.sh -s 50                    (where 50 is the % of the fans you want to be used .. can  be 60/70/100 or 0 .. up to you) .

Works fine on all 470/480/570/580 .. should work on vegas too .

Regards