OS
Pool
ASIC Firmware
ASIC Hub
More
blog
Blog
News and Articles
forum
Forum
Top discussions, announcements, help
Show all articles
Referral Program
referral
Increase your income with Hive. Invite your friends and earn real cryptocurrency!
Get your referral link

So much invalid shares?

AMD Cards
#1

Hey so my rig uses all AMD cards on TeamRedMiner, one card in particular has been so much trouble, I have tried tons of OC settings, flashing with one click (which made the gpu crash constantly) and I just can’t seem to stop it from producing invalid shares?

Then the beginning of the week two other cards started producing invalid shares (GPU3 & 4) but not as bad. These cards do have the one click bios and are OC’d

Now the strange thing is, on Ethermine it is showing no invalid shares? Just 2 stale ones?

Is this a TeamRedMiner issue/gpu issue? I have no idea.

Thanks in advance.

Screenshot_4
Screenshot_41520×508 104 KB
Screenshot_5
Screenshot_5772×145 10.1 KB

#2

Hy!
TeamRedMiner issue. Please try Phoenixminer. AMD GPUs perform much more stably. :slightly_smiling_face:

1 Like
#3

Hey! Just switched to Phoenixminer and now got these errors coming up:

20211105_142350
20211105_1423502016×1134 856 KB

#4

And Invalid shares starting already

Screenshot_8
Screenshot_81525×520 115 KB

#5

Have you tried deleting OC? Is it invalid even without OC?

#6

Phoenixminer —> Delete OC —> web.: Hive OS —> Shutdown In 30s.

#7

Is this problem recent or it has been that way? I had the same problem with my rig and the issue for me was the uneven power distribution.

#8

The OC settings that I have now are the most stable, it seems when I mess with them the GPU just keeps crashing

#9

And this is the second option

#10

Always been that way, apart from the other 2 gpus randomly starting to produce invalid shares, as far as power distribution goes, I have two 1000w PSU powering 8 GPUs and their risers, all using GPU cables

#11

I have two rigs on a same plug. That made both rigs having too many invalid shares. One rig, I hooked up 1 PSU wire to 2 cards. Some AMD cards draw more power than the others so using 1 psu wire for 2 was not ideal. That was my experience.
Now, my fix was: 1) put 2 rigs in 2 different location 2) I rewired. Still using 1 wire for 2 cards but connecting to the lesser power hunger Nvidia 2060 with just 75 W.

In short, mine was the even power. And two days ago, I had to replace a riser.

#12

Screenshot_9
Screenshot_91419×79 20 KB
Power consumption

#13

Hey so I have a 12 GPU slot MB so running all from 1 rig, I can see what was happening with you but it doesn’t make sense in my situation, GPU 0 & 1 are share 2 cables (1 for each riser and 1 for each card) which is the exact same setup as GPU 6 & 7 yet 0 & 1 aren’t failing?

Screenshot_10
Screenshot_101535×512 118 KB

#14

20211105_144629
20211105_1446294032×2268 1.54 MB
20211105_144650
20211105_1446504032×2268 1.5 MB
20211105_144718
20211105_1447184032×2268 1.49 MB
20211105_144728
20211105_1447282016×1134 609 KB
20211105_145411
20211105_1454114032×2268 1.39 MB
20211105_145450
20211105_1454502944×2208 984 KB
20211105_145504
20211105_1455042016×1134 572 KB

The rig

1 Like
#15

Your rig is nice. Also, in another room someone told me that running 10 12 cards also tend to be unstable. And there is always 1 card that continues to be a problem. It might be the same card but the internal of each is different and the way it draws power is different. Since everything is running well, try to see if replacing riser would help. Just process of elimination from now on. hahaha

#16

all miners used to be stable!
with latest releases, they are getting more fragile, and the worst thing is, those bad shares!
i know most of the time is the OC settings, but i think that the developers need to update us when a update is up and if we need to change the OC settings!
i got 3 rigs and they used to run just fine, now, its hard to get the sweet spot and not to have those bad shares or miner restarts and so on…

#17

PhoenixMiner has faced lots of reboots due to its latest update. I agree with you. I changed PhoenixMiner to TeamRed to evade enormous Phoenix reboots. Any version of that reboots repetitively.
About invalid shares, anybody should decrease OC to face no invalid share until the future updates. Reset Miner is not the solution.
I know that sometimes all cards, cables, risers, OC, etc. are the same but one card faces invalid shares, it is inevitable.

#18

TRM is best for AMD gpu´s…where did u get that phoenix miner is better ? :roll_eyes:

#19

true, if u find vers. that is stable for you…dont upgrade unless there is some critical change that is mandatory for your mining

#20

yes, i guess that is the best way to go right now…
we gotta wait till its confirmed that the new version is enough stabe, so its worth upgrading to…