AMD 6600xt in hiveOS

[Oinq]

With all the respect @blackfiveo1, there is nothing to agree or disagree on my post.

All the YouTube videos I have been seeing or other people’s posts are pointing between 75~90, but my test shows something different.
Initially, I thought it was my wall meter that was wrong/broken. I added a second in line.
Is it possible I have 2 different wall meters, measuring wrong?

Tomorrow I’ll take my wall meter to my friend’s and a test against his can be done. He also has a digital meter from the utility company, where he can see actual energy consumption. I also want to test this.

Something that might be tricking this is the power factor. Some wall meters don’t measure it well…

[georgebg]

hi, there is my config:

Working rock solid for day.
Miner is teamredminer.
At the wall when I remove cards :33 W

[georgebg]

At the wall when working:

70W / card is the real result !

[blackfiveo1]

gminer

[blackfiveo1]

70w would be about what I am seeing

if you do 70*10 = 700 that allows 52 for mobo/cpu, seems reasonable.

[niexasadas]

1st of all, you are using ridiculously high core clock, you dont limit any woltage at all, using dpm states, which is irrelevant. 2nd. maybe your PSUs are working at 90%+ capacity? 3rd. yes hiveos (better yet AMD) do not report correct wattage, but it is 20-30W diff. at most. try settings as 90% of the posts above suggests and see what those cards consume then.

[atilatheunos]

Tank you man

[Rabbi530]

So I am relatively new to mining. The rig in question has the following specs:

MSI MPG z490 gaming edge wifi
10th gen I7 Intel cpu LGA1200
Corsair Vengeance DDR4 Ram 16gb
Corsair RM1000x 80 plus gold PSU

I bought to 6600 XTs one Asus and one Red Devil. I installed the Asus first on the board. I had no issues. I used TeamRedMiner and everything worked great. I then connected the red devil to a riser card. The rig appeared to boot okay and TRM booted. They both began mining but within minutes I got a temp warning and I realized the Red Devils fan was not spinning. I shut the rig off checked the connections and restarted. On the second restart both cards fans started spinning on boot and then stopped as usual during the boot but never started again once mining. Now I can’t get either cards fans to spin. They both will mine but the temp quickly heats up and I don’t want to let them get past 90 degrees Celsius. I tried switch flight sheets, mining software, and turning on auto fan in both TRM and Hive.

Any help would be greatly appreciated!

[Oinq]

Hello @niexasadas, thank you for your answer.
I’ll try to address your observations:

1. High clock. After fiddling for hours tho OC settings, I tried the “popular presets” in Hive.
   At the time this was the most popular, and it’s the one that took me above 30 MH.
2. Power limit. The only cards I power limit are 3080s and 3090s because of the memory temperature; I wasn’t aware I had to do it also on AMD. Also, I don’t see other profiles which power limits these cards.
3. DPM states. They came in with the loaded profiles, I can’t get rid of them.
4. I have 2x 1200W server PSUs here. I imagine they are working under 50% or close to it.

From your speech I imagine you are forward to the settings people use 950s core clock.
Just tested it.

Do people here have a tutorial how to overclock these cards? What to change first, second, third?
Nvidia cards are easy because there are only 2 parameters, but these Amd need much more fiddling, and I’m lost in what to do…

Thank you all in advance and in special to @niexasadas for helping.
Nuno

[niexasadas]

Power Limit is always bad idea. For NVIDIA, setting core clock automatically limits the power, but for AMD you want to enter VDD at least.

There are two popular settings:
1100/600/1150 or 940/660/1150

Both gives me good hash rates. I am using GMINER.
Both cards using around 75W from the wall.

[Pilasz]

@Oinq

rd to stabilize. Nvidia are so MUCH easier…
Either way, I think I’m goin

thanks for screenshot. i have 480 reading from wall, total system 6 6600 xt. before it was 440-430m but different MH seutp. aroud 30, now aroud 31. testin new pool/miner.

[Oinq]

And if u force 100% in hive oc profile?

[Oinq]

I’m reaching there!
Thanks for the heads-up.
Something I was doing wrong was not wait long enough.
On my 3070s you change something in the OC and in 10 sec it becomes effective.
At these, it takes a few minutes to stabilize.
940/660/1150 gives me lower power and higher hash rate.

Any tips no the last card to bring it up a bit more?

Best regards and thank you
Nuno

[niexasadas]

I would say start with 1140
also try 1100/600/1140 setup with it too, best is to restart the rig/miner to see effect quicker. If hash rate is going down, it seems memory clock is too high

[nipunanilaksha]

thanks for the update

[bcminer]

worked for me after a whole day of troubleshooting, thanks pobreta

[Oinq]

Same here, Thanks @niexasadas

[jmg05006]

More info please. I have 6 sapphire nitros and can’t get them to hash. What version of hiveos and miner are you using?

[Oinq]

[Lalading]

All working fine with actual hiveos and gminer … only had many problems with hardware like bullshit msi board … now using a 12gpu board, no more problems