Topic: Ethash DAG epoch RX470/480/570/580 FIX !! - page 19. (Read 151478 times)
I updated the drivers some days back and all of the GPU speed increased back to normal. Thank you Robinhood.
But I have a lot of Incorrect ETH shares message especially on one of the GPU (RX 580 Gaming).
About 15%. The others is about 1%. But I had none some time back.
I don't know if this has to do with the new drivers or something I did.
Already tried to put things back to normal but no luck.
Did anyone had this problem with the drivers so that I can discard this possibility?
Hi Guys,
Watch how to upgrade the driver for newbies and results on youtube : AMD RX 470/480/570/580 Hash Rate New Driver Guide
https://youtu.be/b8J_ZtLTUyc
Rebel
Kudos
Watch how to upgrade the driver for newbies and results on youtube : AMD RX 470/480/570/580 Hash Rate New Driver Guide
https://youtu.be/b8J_ZtLTUyc
Rebel
Kudos
I think I am not a newbie account, and for the second time I will declare that the drivers are good and working (for me on 570 and 580).
About the hashrate - see the picture of one of my rigs (6x RX 580 8 Red devil): average is constantly improving ad at the moment is even higher then the reported!!! (never seen this before)
http://imgur.com/a/LEqwj
Robinh00d - thank you very much, appreciate it. There is still hope for the community...
About the hashrate - see the picture of one of my rigs (6x RX 580 8 Red devil): average is constantly improving ad at the moment is even higher then the reported!!! (never seen this before)
http://imgur.com/a/LEqwj
Robinh00d - thank you very much, appreciate it. There is still hope for the community...
I applied the driver to one of my rigs, and i see improved hashrate but wattool refuses to work, I managed to get undervolting under control with MSI afterburner beta 16, but i'd rather have wattool on my AMD rigs, any way to get wattool to work with newer drivers ? I only have rx 470s.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
In my test I used two identical rigs.
One with an old driver, one with the new one. Both with new wallet addresses (I wanted to remove suspicion about pool luck and variable difficulty)
Start simultaneously in the same pool.
After 6 hours I saw the positive difference in favor of the new driver ... the small difference, but visible difference.
My only fear is of "bugs in sleep".
appreciate in sharing your observation.. we can't remove doubts. i am doing the same as you did.. looking forward to see the results tomorrow..
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
In my test I used two identical rigs.
One with an old driver, one with the new one. Both with new wallet addresses (I wanted to remove suspicion about pool luck and variable difficulty)
Start simultaneously in the same pool.
After 6 hours I saw the positive difference in favor of the new driver ... the small difference, but visible difference.
My only fear is of "bugs in sleep".
I've tested this on all my rigs and at first glance all seemed great.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
I was thinking the same thing, but did not compare the before and after number of shares
I've tested this on all my rigs and at first glance all seemed great.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
Im doing daily 24hrs log to my 3 rigs.. i installed this to one of my rigs.(new wallet address) i will try to share my reading and compare my hashrate 24hrs before i install, and after i installed this driver. i am still doubtful of course but impressed whoever created this.. and i hope "HOPE" that the creator has a clean intention to help the mining community... i am not saying to use this driver. (use it at your own risk)..
I've tested this on all my rigs and at first glance all seemed great.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
Same situation here... Reported hash rate is like before but efective has rate i think same. Also number of shares same. Testing it for 8 hours so far so maybe need more time. Btw how you undervolt carda as i was unable to do it with claymore neither with afterburner newest beta 16?
I've tested this on all my rigs and at first glance all seemed great.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
I did have to bypass windows not allowing me to assign a unsigned software.
My hash rates appeared to be the same as before the nasty Dag creation started.
My power usage is the same after under volting.
My REPORTED hash rate to the pool of course matching my Rigs hash rate.
BUT here's the kicker my effective hash rate on the pool is about the same if not a little less than it was before applying the driver. So my question is this.
Is the driver merely creating false readings on the cards with no actual effect on total shares submitted to the pool much like when you think you have found the golden card that reports a hash rate of 45Mh/s when in reality its only hashing at the same speed as the others. Had the new Driver running for 7 hours now ....so of course I need test some more just incase its the pool on a go slow....although it normally only takes a couple of hours to get fairly accurate readings.
Never saw this anywhere, but today I noticed (and this could be a waiting tempfix of sorts for the dag problem for amd), that with slowdown due to next epoch it is possible to significantly raise dcri value and get much more power for the second coin - in my case currently dcr. Yes, from 14 to 32 dcri and voila, doubled hasrate for dcri, while eth hashrate shows no significant dip (well, except to epoch dip). And it seems to have its logic behind it: dag-epoch slowdown apparently frees some computing capacity for the second coin.
Try this and thank me later.
Try this and thank me later.
Anyone tried this with a miner other than claymore?
AMD might provide the fix the same week or day of RX Vega release.. "Maybe"
If they keep up with their 2 weeks release, then yes. I think that too, they might release it in few days with VEGA.
no wallets
Well, I guess as long as you mine on a pool my tests showed this driver to be safe, now if you mine solo then I would not install this driver either.
Has anyone installed this on their pandaminer yet? If so how did it go? My hashrate has remained the same with DAG change on 10+ machines. The only machines that went down are my pandaminers. I have not tried installing this on them yet.
What are your other 10+ machines running? Nvidia or AMD GPUs?
Might sound silly being AMD thread but just making sure
Has anyone installed this on their pandaminer yet? If so how did it go? My hashrate has remained the same with DAG change on 10+ machines. The only machines that went down are my pandaminers. I have not tried installing this on them yet.
Am I the only one noticing the trend of noob accounts saying this driver works great.
I have missed some activity on a couple of pages but it seems odd to me.
I have missed some activity on a couple of pages but it seems odd to me.
Look #109
Edit: But I tested.
I put a backup of empty windows (no browsers, no wallets) only Claymore.
I used a new mining address.
After the test I deleted the entire HDD ... and I changed the MAC to my network card (and a new IP, of course)
I admit: the test was positive.
But ... I prefer to wait for the official version.
Am I the only one noticing the trend of noob accounts saying this driver works great.
I have missed some activity on a couple of pages but it seems odd to me.
I have missed some activity on a couple of pages but it seems odd to me.
I'm a noob here too and absolutly with you! It's strange! I will wait for the official release or someone, like Claymore, say a word to this "beta".
Am I the only one noticing the trend of noob accounts saying this driver works great.
I have missed some activity on a couple of pages but it seems odd to me.
I have missed some activity on a couple of pages but it seems odd to me.
There's definitely something wrong with your setup. You should be getting about 24 at stock. EDIT: or maybe 22 since they are 4gb
I really do not think so, I have 6 Asus Rog Rx 470 4G hynix, 1 Gygabyte Rx 470 4Gb gaming 1 hynix, and 2 XFX Rx 470 4Gb samsung dual BIOS.
Except for the XFX (Which seemed to be already overclocked), all of them hits 20 at stock, I tried claymore 9.5, 9.7 and windows and linux.
Overclocked all cards gets 24,5 in windows (linux stays 20)
This is what a 470 can do.
http://imgur.com/a/8ajS4
You can say there's nothing wrong when you achieve this.
Can you share your mod? My HIS rx470 (Hynix) can only reach 21 MH/s (stock), 22.x (modded) & 23.x (DAG fix). Thank you.
Mine are Sapphire with Samsung. They key is custom memory straps. There's a thread here about it: https://bitcointalksearch.org/topic/custom-ram-timings-for-gpus-with-gddr5-download-links-updated-1758267
Once you give that a read you should be able to do it yourself.
Also I found that -asm 2 helps a lot with the 470's
Your are talking about a 4Gb card?
On my rig with 5 x saphire rx 480 I firstly instaled newest afterburner beta 16, driver, patched all goes perfectly. After restart i started claymore my hash rate completely returned to 145 mhz.. Each card giving 29.263.. My setup for afterburner are core voltage -60, power limit 0, core clock 1150, memory clock 2000.. in claymore -cvddc 1025 Rig is working stabile for few hours so far...
But what i noticed in afterburner is that there is big fluctating in gpu power usage it goes from 85-110w and also gpu core voltage is fluctating from 0.956 to 1.075... This big oscilating are not good or it is normal as before on same rig i was getting 87-90w power usage and core coltage betwen 0.981 with little oscilating to 0.987.. Didnt have device to measure real usage from wall...
How to solve this?
But what i noticed in afterburner is that there is big fluctating in gpu power usage it goes from 85-110w and also gpu core voltage is fluctating from 0.956 to 1.075... This big oscilating are not good or it is normal as before on same rig i was getting 87-90w power usage and core coltage betwen 0.981 with little oscilating to 0.987.. Didnt have device to measure real usage from wall...
How to solve this?
Jump to: