Mining ERGO with Vega 56 & 64

[ss-silver]

@846875 I typically see about 1.5% to 3.5% less in the miner than reported in HiveOS over the long run (after at least a few days uptime). In the short term the miner will show all sorts of results but take the long term average if your comparing vs ETH. Right now its considerably more profitable on ERG for Vegas.

@lil ERG has a short period after each block is solved where no mining occurs, and if you watch the miner it stops mining and builds a table each time a block is solved, that is why you see the drop in hashrate approximately every 2 minutes(ERG average block time). Nothing you can do to prevent it, its just how ERG works.

[lil]

thats meaning some block i got at 183MH and some block i got at 168MH like that right?

[ss-silver]

No, when its mining you’re getting 183 but between blocks since it stops mining for a couple seconds it reports less hash. It happens to all miners at the same time, but once it builds the table and starts mining again it goes back to 183.

[846875]

Yeah, on eth I see about the same, just a few percent less reported by the pool but on ergo it was getting almost 400mh in the miner but only 300-350 tops at the pool.

According to whattomine that is about the turnover point where the profits are a wash when factoring in power.

I did notice the rebuilding of the cache or something every few mins and figured that would be part of the cause for the lower reported hash at the pool.

Thats why I was wanting to know what others were seeing at the pool on ergo with that factored in.

I am mining to nicehash if that makes any difference.

[846875]

Also curious what is a safe max temp for these GPU’s? They are in the garage and still pretty new to mining, these are my old gaming cards that I decided to start mining on instead of selling.

On eth they generally sit around 60c core and low 70’s mem in the garage with ~100F ambient temps.

On ergo they are around 8-10C higher. so 70c core and low 80’s mem.

Will this hurt them long term?

[lil]

I see it. Im using your OC setting by the way my rig is restart every 30 minutes. Can you help? Maybe its because OC setting.

[ss-silver]

Probably just need to increase VDD by 12. Which settings exactly?

[lil]

i decrease the core looks like works fine. i monitoring before its shut down, it increase to 203MH then shutdown. after that i decrease core a little and its on 183.7MH.

[Smining570]

Finally stated to fine tune erg template, trying to find compromise W/hash…i think there is one or two ref 56 that needs new pads!

[ss-silver]

the rog strix 64 is running 58c? at 144 watts?! ya needs some pads and paste

EDIT: ahh fans at 50% my b, just turn the fans up a bit those temps are not bad

[Smining570]

yeah, i had fans set up bit low and allso on rig my fans were tuned down after cold nights with eth…didnt even noticed! thanks for reminder lol… btw do you remember what thicknes are pads for that rog strix?,same 1mm/1,5mm ?

[Smining570]

@ss-silver

the rog strix 64 is running 58c? at 144 watts?! ya needs some pads and paste

what straps you are usin with that stix? cant replicate those numbers with my starps. thanks

[lil]

i just change thermal paste using grizzly but temperature still same. is it ok 80c?

[ss-silver]

@Smining570
This strap and 1404 effective core:
amdmemtweak --CL 20 --RC 36 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12

If you run the amd-info command it shows you effective core, for my Rog Strix I set core to 1409 in OC to get it running effective core at 1404. That card is good with effective core only -5 from what its set at, some of my less good vegas run -70 from what I set core at.

@lil did you do the pads too? There is one tiny spot that needs 1.5mm, if 1mm is on that spot they will run hot like that. I dont like seeing them over 80c, personally if the vegas hit 85c I turn off the rigs for an hour.

[Smining570]

@ss-silver

amd-info

I remember that die with 1,5, will do pads at some point, temps are under control for now…i had my external “delta” fans set to 25% speed so thats why temps were off.
This strix is even harder to tune with erg than with eth, i have spend 2 days with it…well not first time Ill give a try with those straps, i belive i have had them on allready but not with core 1404. From this rig i have that and one and 1 ref 56 that is giving me headache…they both are stable but not optimal “greedy” !
Ill report the results, i belive im close to what im after allready, just take´s sometimes long time to hit the sweet spot. Thanks again bud edit Strix gpu 3

[akosi]

Help my rig keeps crashing. Dont know if its my straps/oc/or faulty motherboard

[akosi]

Help my rig keeps crashing.

I have these straps:
amdmemtweak --CL 20 --RC 37 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12

Tried: 1400, 1350,1300 Core
Vdd Tried: 900, 910, 920, 930
Memory: Tried 960, 940, 920, 900, 850

The issue is always on GPU 0 Keep detecting dead.

I have 2 rigs with flashed vega 56, these are vega 64 sapphire blower edition with samsung memory.

Any idea? They won’t pass 10-20 mins then crashes.

My other rig with Asus Strix vega 64 flashed to 54 are stable with only minor Voltage fix.

[846875]

In the same boat, card is in the garage and in the middle of the day it gets a bit over 80c mem temps.

Sometimes as high as 85c.

I have had trouble finding the official rated temp for samsung HBM2. I like the extra profits from ergo but don’t want to hurt the card.

[Smining570]

hbm2 is edge at 105c if i remember correctly but its usual that over 80c deg it start give issues if u Don tlower oc s… So better cooling or less hash when hot. 80cdeg wont Hurt chip

[846875]

Great, thats what I wanted to know. Still getting ~196mh at 85c, so gonna say I am happy with that.

I am having to use 950mv to do it though, seems higher then a lot of other people? I tried going down to 910 but it insta crashed. 930mv seemed possible.