paparazzo (OP)
Newbie
Offline
Activity: 3
Merit: 0
|
|
March 16, 2018, 08:21:08 AM |
|
Hello, I am new to mining I have 5 rigs- 6x RX580 8gb each and I have been mining ETC for a week now. I had some problems with two of my cards but they will be resolved today. However I would like to ask some of the more experienced of you: how do you keep your average hash rate more stable and closer to the reported hash rate? My typical average hash rate is arround 770-810 while the reported rate is 920. It is not always like that but the max average hash rate I have had was about 850-860. Thank you. I would appreciate any help you can provide
|
|
|
|
somethingiswrong
Newbie
Offline
Activity: 31
Merit: 0
|
|
March 16, 2018, 08:31:23 AM |
|
With reported hashrate you mean the hashrate shown by the pool you are mining on? Pools only estimate your hashrate based on the shares you have submitted. So the hashrate of your rig will always differ from the one shown by the pool.
|
|
|
|
manji
Full Member
Offline
Activity: 210
Merit: 174
BookiePro.Fun - The World's Betting Exchange
|
|
March 16, 2018, 11:58:02 AM |
|
if stability is your priority don't push your cards too much, in another word use stock bios with a little bit OC. the key to stability is keep your cards still cool also give them enough a power.
|
|
|
|
leonix007
Sr. Member
Offline
Activity: 1008
Merit: 297
Grow with community
|
|
March 16, 2018, 12:42:33 PM |
|
if stability is your priority don't push your cards too much, in another word use stock bios with a little bit OC. the key to stability is keep your cards still cool also give them enough a power.
Wrong, modded RX cards can keep stability more than stock ones and give you higher Hash, My modded cards are way cooler than stocks bios. I believed the Average Hashrate has been computed by the value of your Effective Hashrate, while the Effective HR is computed thru your valid shares which is dependent on the current difficulty. This might mean that your reported Hashrate could not guarantee an equal/slight difference value
|
|
|
|
manji
Full Member
Offline
Activity: 210
Merit: 174
BookiePro.Fun - The World's Betting Exchange
|
|
March 16, 2018, 01:21:25 PM |
|
--snip--
Wrong, modded RX cards can keep stability more than stock ones and give you higher Hash, My modded cards are way cooler than stocks bios. I also modified my Polaris cards but didn't give OP a personal view. So RX series with modded, actually, pushing default condition, different from with R9 and kinds. Mod bios would change default strap that has been calculated by a vendor in a safe zone, and they give a range to increase a few elements on it.That's why I give technical view to OP because of whatever when we push something to its limit, for sure will be something a bit different.
|
|
|
|
leonix007
Sr. Member
Offline
Activity: 1008
Merit: 297
Grow with community
|
|
March 16, 2018, 01:40:23 PM |
|
--snip--
Wrong, modded RX cards can keep stability more than stock ones and give you higher Hash, My modded cards are way cooler than stocks bios. I also modified my Polaris cards but didn't give OP a personal view. So RX series with modded, actually, pushing default condition, different from with R9 and kinds. Mod bios would change default strap that has been calculated by a vendor in a safe zone, and they give a range to increase a few elements on it.That's why I give technical view to OP because of whatever when we push something to its limit, for sure will be something a bit different. Yeah agree unto the perspective of Cards pushing unto its limit, however it does not affect stability versus on stock ones , from the looks of it, from temps to wattage you may underclock/overclock it depends on your references and may get amazing result that seems not pushing your cards too much compared to stocks
|
|
|
|
manji
Full Member
Offline
Activity: 210
Merit: 174
BookiePro.Fun - The World's Betting Exchange
|
|
March 16, 2018, 02:04:35 PM |
|
Yap, When talking about stability, something that been tried with a ton of test would be better (it's called standard and vendor do a ton of test, I say for sure), that's the point. I can't judge something "wrong" or "right" just because I do that. Same as you, all of my Polaris card been modded and seems stable but wouldn't so stable as stock with a bit tune-up on it. For honest, I do that for hashing speed.
|
|
|
|
leonix007
Sr. Member
Offline
Activity: 1008
Merit: 297
Grow with community
|
|
March 16, 2018, 02:27:01 PM Last edit: March 16, 2018, 03:09:38 PM by leonix007 |
|
Yap, When talking about stability, something that been tried with a ton of test would be better (it's called standard and vendor do a ton of test, I say for sure), that's the point. I can't judge something "wrong" or "right" just because I do that. Same as you, all of my Polaris card been modded and seems stable but wouldn't so stable as stock with a bit tune-up on it. For honest, I do that for hashing speed.
Honestly I doub't on the test being held, for sure standard test is for gaming, as the cards are not purely designed on mining. so the stability test is different from gaming to compute and were specifically not really a default condition for compute mode. my first rigs 2 yrs ago has been a purely stock bios RX480 reference cards and I could tell, stability is same but hashes 15% less same is true on debunking myths about stressing GPU's when modded https://www.reddit.com/r/Amd/comments/6irktq/debunking_myths_about_mining_and_gpus/
|
|
|
|
paparazzo (OP)
Newbie
Offline
Activity: 3
Merit: 0
|
|
March 16, 2018, 07:13:04 PM |
|
Thank you for the response So from what I gather it's a matter of tweaking the rigs to reach a optimum hash rate.
|
|
|
|
manji
Full Member
Offline
Activity: 210
Merit: 174
BookiePro.Fun - The World's Betting Exchange
|
|
March 17, 2018, 04:05:03 PM |
|
.....
Honestly I doub't on the test being held, for sure standard test is for gaming, as the cards are not purely designed on mining. so the stability test is different from gaming to compute and were specifically not really a default condition for compute mode. my first rigs 2 yrs ago has been a purely stock bios RX480 reference cards and I could tell, stability is same but hashes 15% less same is true on debunking myths about stressing GPU's when modded https://www.reddit.com/r/Amd/comments/6irktq/debunking_myths_about_mining_and_gpus/Whatever the different view of us, it would be inputs for OP. I stick to my opinions, didn't mean I "said" you are wrong. We both know what we said, right? What we told is our experience. Thank you for the response So from what I gather it's a matter of tweaking the rigs to reach a optimum hash rate. Stability will be more important than higher Hashrate.
|
|
|
|
Basmic
|
|
March 17, 2018, 04:19:20 PM |
|
With so many rigs, you really should have a total hashrate of at least 900. Each GPU series RX 580 must give not less than 30 mh/s. Perhaps you have a problem with the firmware of the BIOS of the GPU. In order to achieve this speed, I upgraded the BIOS of my GPU. It seems to me that the main reason for this.
|
|
|
|
paparazzo (OP)
Newbie
Offline
Activity: 3
Merit: 0
|
|
March 18, 2018, 02:02:56 PM |
|
With so many rigs, you really should have a total hashrate of at least 900. Each GPU series RX 580 must give not less than 30 mh/s. Perhaps you have a problem with the firmware of the BIOS of the GPU. In order to achieve this speed, I upgraded the BIOS of my GPU. It seems to me that the main reason for this.
My total hash rate is 960, however my average is lower between 800-860. I have upgraded my BIOS and there reported hash rate is quite good 31-32 mh/s per card. The per cent of drop shares is also in range in my opinion beteen 1-4 % . I seems i need to do some more tinkering with the cards in order to up the average hash rate.
|
|
|
|
|