What do you mean by connect? Connect to the pool? That doesn’t sound like a GPU issue, sounds like a pool/config issue.
Yeah, somehow they didn’t connect anymore and I didn’t change anything. I just changed miner and working again.
But now, facing another issue. Fans of one of the Vegas are not working and again I didn’t change anything. So weird… and not even downgrading to a previous HiveOS version fixed the damned issue with the fans, which worked perfectly yesterday…
I will try again with TeamRedMiner or other miners, something weird must be going on here, jeez.
Vegas can be touchy, first try turning the system off and removing power for a min or 2 and rebooting. That does it sometimes.
If that doesn’t do it hopefully you have a bios backup, I have had to reflash the bios on the vega several times in cases like this and it would work fine after. If you have a bios switch can also try flipping that.
1 Like
Ok thanks I will try that. I think I have bios somewhere (I just flashed these Vegas months ago and didn’t touch them since). And I will try with the switch as well, just in case.
EDIT:
Well, still the same result.
I reflashed bios with original one and fans not working. Flashed Bios with the one I had before, which is Vega 56 version for Vega 64 Strix and still fans not working (this is the one that have been working for months).
I tried as well using bios switch with same results.
So… I don’t know what else I should do. I could try opening the card and cleaning it, but I don’t think that´s the problem. I always use an air blower with the cards from time to time, so they are more or less dust-clean.
I will turn it off 1 day and check again tomorrow. But I mean, it´s a Vega 64 Asus Strix which was new and running for months without any issue. Others are still working fine, it´s just this one.
Obviously as fans don’t work, as soon as it starts mining, temperature goes to the moon
UPDATE: I just turned the rig off for 15 mins and now, suddenly it´s working again. We´ll see how long it lasts.
Greetings all. In trying to mirror ss-silver ergo MH rates, I seem to fail. Wondering if anyone has insight or suggestions please. Below is what I gleamed from my OC settings and amdmemtweaks.
As you can see my hash rate can’t go much higher without dead GPU errors. This is stable except for GPU 0 that goes triggers the restart script with GPU dead every 12hrs or so. But watchdog.sh reboots my rig and GPU 0 starts mining ok again.
GPU 0 Hynix mem per ss-silver: amdmemtweak --CL 20 --RC 36 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 22 --REF 65535 --RFC 239 --RTP 5 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 13 --RCDWR 12
GPU 1-5 Samsung mem per ss-silver: amdmemtweak --CL 20 --RC 36 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RTP 5 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
I want to move on th adding mo GPU but until I figure out why such low hash rates I don’t want to complicate the rig.
Currently no GPU has been flashed. They are used & stock as I rcvd them. I have 4 Vega 56’s in a dead pile that I am hoping to resuscitate them by flashing the GPU BIOS.
How can I get my hash rate to reach 180’s MH as I see in this forum?
Any suggestions are welcome. For guidance I have used teamredminer_autolykostuning guide and ss-silvers mining Ergo w/ Vega 56 post.
My setup is:
HIVEOS V. 0.6-213@220221 linux
ASUS B250 MB, 250GB SSD, 8GB Mem
(3) HP 1200w PSU’s using 2 now. 4GPU’s on one PSU and 2GPU’s on the second PSU
teamredminer v.0.9.3
My at-the-wall wattage is 1075watts. For the low MH rates I am not running well.
Thanks for allowing me to join your community and I hope I can one day help others as I hope others can help me.
Cheers, CommShop
Never mind: According to Nov '21 post by bonezy, no one is getting more than 140MH since ERGO update. I guess we just wait. I am still wondering if I should flash my GPU’s or since they are performing nominally as bonezy says, I should just continue adding GPU for now? Any feedback is welcome. Thx
The days of 180+mh are over I am sad to say, there was a change to the network last year that dropped the hashrates down to ~120-140 IIRC. I have not tried since the change.
After very very quick test this is the highest point i can get out from my vega, i tested only 1 which is samsung as shown in image
Yep, quite disappointing to be honest
Yes, seriously torn about selling my vegas but who knows if there will be another ergo like coin that they dominate in and I regret it. They are some of the most temperamental cards I have that is for sure.
Hi, do you know if there´s any issue right now with Vegas?. They just don’t mine anymore and I didn’t touch anything.
Weird. I am using nanopool + teamredminer.
Well, somehow now they are working. Weird things that sometimes happen and you can’t find the reason.
I know there’s no update to get these cards back up to full hash like they were before but I have been able to get 154mhs out of all of my cards all of them are in the 145w range too. There may be some more room to tweak them but that’s the best I have been able to do.
Its not that they have been nurfed its the way the older cards reference the memory. This is how I understood it: Ergo uses memory tables, 6 months ago you could fit 4x memory tables on the Vegas HBM2 8GB. Then the tables simply didn’t fit after that fateful day. So now the Vegas can only now fit 3x memory tables on the HBM2. That fateful day we all lost around 25% hash.
I wish they could pool the cards HBM2 to one larger memory or offload to the systems RAM.
Definitely not the end of the world for them. They still preform pretty nice for what they are. This is also one of the best descriptions of what happened that I have heard.
1 Like
Does anyone know if the vega 64 FE cards are affected like the other cards are? The have 16 gigabyte memory not the standard 8 that the other cards have.
If they are affected has anyone tried dual mining ergo and ergo on it to see if there is a increase on hash rate that way? I know you can dual mine ergo to two different pools on one card because I tried it but mine are 8g cards and I did not see a increase in hastate.
Hi again!! xD well since Eth days are almost over the last month I worked on Erg a bit regarding eff mostly.
From the side of vegas I’m almost done.
Got 2x 3080s tweaked as well, was nightmare to set them up.
Any suggestions to improve eff without losing hashrates are welcome xD
1050w from the wall (hot day). LF winter xD
-Ryzen 3600 PBO off (all energy savings @bios are on), except Ram erg <3 ram speeds 3733@cl17 @1.36V. =>Next step test undervolt cpu or+ disable cores.
-Superflower 1600 platinum (had dual psu +2000w but removed it since erg doesn’t need it) got sweet 30-50w returns.
RTX 3080s p2 state off, power management adaptive (ain’t make difference unless idle…).
1 Like
those are some impressive power usage numbers you got on the vegas there, i see you were able to drop the vddc quite low to 800 on most, im running mine on 920 currently but havent played around on it much, can you use R mode with ergo or not really?
Thx mate 
, no I can’t use R mode at all I’m a windows user.
Tried your settings in HiveOS, they work fantastic! you sir are a genius!
2 Likes
Can you share the oc how it turned out on hiveos?
Using TRM with r-mode?
I found that R mode does not work so dont turn that on otherwise it wont start mining.
What works best are these settings and if you get GPU DEAD error then just bump up the core voltage until it is stable
Hive shows 100w but in reality its more like 140w per GPU and these settings yield about 131mhs
Core Clock: 1280
Core Voltage: 830
Memory clock: 830
SOC frequency: 899
amdmemtweak --CL 20 --RC 37 --RP 12 --WR 12 --CWL 8 --FAW 12 --RAS 26 --REF 22600 --RFC 234 --RTP 5 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
2 Likes
This topic was automatically closed 416 days after the last reply. New replies are no longer allowed.