I don’t believe a riser is going to solve the problem, I have AMD 6800s in riserless motherboards with all the same behaviors described in this thread.
I’d love a riser to solve the problems, but I have had more luck running the worst 6800s:
Tks! 
When I found that we can flash the BIOS from RX6800XT to normal RX6800 I noticed that the huge difference is in the “F state” in the XT version is “1941MHz” while in the normal 6800 it is a maximum of 1551 but we never reach it unless we flash it with BIOS from the XT version and then bring it back. But then I installed a new card and noticed that the hashrate (and F state) dropped again. That’s why I installed UPP (a program similar to MorePoweTool but for Linux) and started playing with the frequencies, the rest you can read in my post from a few days ago
When I increased the frequency to only F state, everything fell into place. The only problem was that after each restart I had to re-apply the settings and when changing the setting through UPP the overclock settings are lost and you have to confirm them again via the dashboard (but you also have to change something because it gives you that nothing has changed). So I was looking for a method to apply the settings for “F state” at startup as this had to happen before the overclocking settings so this makes this script. Now through Nbminer I have a stable 62.5MH / s at quite low frequencies of SOC and Dcef
Can we also set the “SoC Frequency” with UPP?
Yes you can but I found difference only when change F state command for SOC frequency will be:
upp -p /sys/class/drm/card1/device/pp_table set smc_pptable/FreqTableSocclk/0=1200 --write
1200 is new frequency.
Hi. which miner are you using? Do you get these great MH on NBminer also?
I’ve only used TeamRedMiner with my AMD cards. 
I also have an XFX RX6800XT, what would be the best configuration for it?
By the way, the XFX RX6800 script worked perfectly on my rig, thanks for sharing your knowledge.
WOOOOOOOOOOW
Awesome tutorial!!
I had diferent hasrates on all my six 6800. Now its solid 62,5 for all.
Using teamredminer.
THANK YOU VERY MUCH, INCREDIBLE
Be carefull if you have enabled intel GPU.
Intel gpu is card0
Reorder your script
Reboot, dont shutdown, only reboot
I am using this commands in teamredminer:
–eth_dag_slowdown=9 --eth_stagger --eth_ramp_up
for 6800 non xt I have this settings:
1350 775/775/1300 1075
and watch with amd-info command if your F state is at 1551 Mhz.
Prove your script with “bash /opt/script_upp/script.upp.sh” and watch with amd info if all your 6800 non xt are at f state 1551 Mhz.
If you have Intel GPU on thats card0, start config with card1.
And when you prove your script see that there is no errors.
thanks for the fast reply,
my main concern is the high temp I am getting I don’t know if I am stacking too much cards in a small area or is it some bad config.
btw I saw that you too have some cards running on 46MH is it normal?
but yours running at a low temp and low wattage which is awesome
These two cards are 6700 XT runing at 46,5 Mh/s and I have one 5700 XT running at 52 Mh/s.
All my 6 x 6800 non xt are runing at 62,5 Mh/s.
10 hours now rock solid.
Copy my voltage settings, I think memory is “protecting” at 1250 mv and memory controller you just need 775.
Copy exact settings and I think that you have problems with power supply, can not go from 0% load to 100% load to quickly so put my teamredminer settings:
–eth_dag_slowdown=9 --eth_stagger --eth_ramp_up
Could you please help me? I did everything like your guide. But when I give the command ‘bash script.upp.sh’ I receive the error :
Traceback (most recent call last):
File “/usr/local/bin/upp”, line 11, in
sys.exit(main())
File “/usr/local/lib/python3.6/dist-packages/upp/upp.py”, line 373, in main
cli(obj={})()
File “/usr/local/lib/python3.6/dist-packages/click/core.py”, line 1137, in call
return self.main(*args, **kwargs)
File “/usr/local/lib/python3.6/dist-packages/click/core.py”, line 1062, in main
rv = self.invoke(ctx)
File “/usr/local/lib/python3.6/dist-packages/click/core.py”, line 1668, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File “/usr/local/lib/python3.6/dist-packages/click/core.py”, line 1404, in invoke
return ctx.invoke(self.callback, **ctx.params)
File “/usr/local/lib/python3.6/dist-packages/click/core.py”, line 763, in invoke
return __callback(*args, **kwargs)
File “/usr/local/lib/python3.6/dist-packages/click/decorators.py”, line 26, in new_func
return f(get_current_context(), *args, **kwargs)
File “/usr/local/lib/python3.6/dist-packages/upp/upp.py”, line 334, in set
res = decode.get_value(pp_file, var_path)
File “/usr/local/lib/python3.6/dist-packages/upp/decode.py”, line 696, in get_value
pp_bytes = _read_binary_file(pp_bin_file)
File “/usr/local/lib/python3.6/dist-packages/upp/decode.py”, line 42, in _read_binary_file
f = open(filename, ‘rb’)
FileNotFoundError: [Errno 2] No such file or directory: ‘/sys/class/drm/card0/device/pp_table’
I am not a linux habitual user, so I can’t figure out what’s going wrong.
Perfect!
can you please explain what is the procedure for installing ppu on hive os drive?
Can you show me output of amd-info
@bunker8 read my posts from few days ago first you need to install “python3” next with command pip3 install upp
