i have rx 470 4g spphire+ and afterburner 4.4 beta 16 don't increase m clock higher than 1300 !!!
i enabled extend overclocking limit with no change
Topic: Ethash DAG epoch RX470/480/570/580 FIX !! - page 27. (Read 151564 times)
Any way to under voltage core on afterburner more than -96? If not then is a problem.
How can this be fixed with a new driver, when they said this hash slowdown occured because of 2gb memory modules being used, and like because not all data can fit in one module, it has to be spread on multiple, so the data read time is increased, hence hash drops.
How can this be fixed with a software ?
This doesnt apply for HBM i doubt is about memory modules. How can this be fixed with a software ?
This is my first message! 
I try the driver and it's ok, my ring is two RX 580, from 56.8Mh/s I go to 61.5Mh/s (mining only eth - bios modded for gpu/memory frequency/voltage and memory time... so I don't use any app for set this)
The card is set to go at 1150 gpu/2200 memory/900mV
I notice more share on ethermine (demonstration of the increased Mh/s) and for the stale there is no changes. My average effective hashrate was 51.8/51.9, now is 52.4 (driver was installed at 17:30 Rome time so I hop go up in the night)
On the system before install the driver I have installed 17.7.2 (driver version 22.19.662.4 from device manager). This one install version 22.19.659.1
So thanks!
I try the driver and it's ok, my ring is two RX 580, from 56.8Mh/s I go to 61.5Mh/s (mining only eth - bios modded for gpu/memory frequency/voltage and memory time... so I don't use any app for set this)
The card is set to go at 1150 gpu/2200 memory/900mV
I notice more share on ethermine (demonstration of the increased Mh/s) and for the stale there is no changes. My average effective hashrate was 51.8/51.9, now is 52.4 (driver was installed at 17:30 Rome time so I hop go up in the night)
On the system before install the driver I have installed 17.7.2 (driver version 22.19.662.4 from device manager). This one install version 22.19.659.1
So thanks!
Why do you bitch about difficulty? U know that nvidia have advantage and with this everyone will have equal earnings?
Are you talking to me? I'm not bitching. I was suggesting the difficulty will plummet after all the AMD GPU's go offline because this "driver" might fry them.
works fine. thanks a lot.
straight forward. but these new crap apis are shit.
need afterburner to lower -100mv on all and underclock core & overclock mem in claymore config.
back to 28mh dual on etc.
using amd internal fans control, yet cannot set my target temps - any ideas?
updated watttool anyone?
2*480 msi refs + 2*470 refs (Hynix+Samsung) + 470 Nitro+
straight forward. but these new crap apis are shit.
need afterburner to lower -100mv on all and underclock core & overclock mem in claymore config.
back to 28mh dual on etc.
using amd internal fans control, yet cannot set my target temps - any ideas?
updated watttool anyone?
2*480 msi refs + 2*470 refs (Hynix+Samsung) + 470 Nitro+
Why do you bitch about difficulty? U know that nvidia have advantage and with this everyone will have equal earnings?
I am pretty sure this guy is some legendary member, he had to make a fake new account probably because he didn't want to get introuble with AMD for pre-releasing these drivers.
I highly doubt a AMD driver can become some executable virus.
I highly doubt a AMD driver can become some executable virus.
Maybe its not a virus but a GPU destroyer? Watch that difficulty plummet
Not compatible with msi afterburner, high consumption, not stable clock speed 
Download New After burner 4.40 beta work fine
I can confirm on 470/480/570 that it is working! Tried benchmark 199 and the result is 29mgh not the bloody 16mgh! Good work!
I am pretty sure this guy is some legendary member, he had to make a fake new account probably because he didn't want to get introuble with AMD for pre-releasing these drivers.
I highly doubt a AMD driver can become some executable virus.
I highly doubt a AMD driver can become some executable virus.
I've noticed this driver gives also less stales. anyone noticed this also?
nothing change i returned back
I heard that the AMD Fire Pro 17.Q3 driver already has the DAG size fix.
You can install it for RX 4xx/5xx without a problem AFAIK.
You can install it for RX 4xx/5xx without a problem AFAIK.
I tried the latest fire pro drivers and they would not install on a 480/470 box
confimed to work
and with the latest beta version of AB you can adjust clocks and voltage like before
will send .1 eth
thank you
AMD is back!!! suck it NVIDIA lol
and with the latest beta version of AB you can adjust clocks and voltage like before
will send .1 eth
thank you
AMD is back!!! suck it NVIDIA lol
I heard that the AMD Fire Pro 17.Q3 driver already has the DAG size fix.
You can install it for RX 4xx/5xx without a problem AFAIK.
You can install it for RX 4xx/5xx without a problem AFAIK.
I notice miming ETC with the new AMD Crimson ReLive Edition 17.7.2 driver, it only went down one 1 MH @ 141DAG sense etc is that high now.
can't say if that's the drivers or what, an you need after burner beta 15 or higher 16 is out live now for the drivers to work right with over clocking with afterburner and driver Crimson ReLive Edition 17.7.2.
can't say if that's the drivers or what, an you need after burner beta 15 or higher 16 is out live now for the drivers to work right with over clocking with afterburner and driver Crimson ReLive Edition 17.7.2.
How can this be fixed with a new driver, when they said this hash slowdown occured because of 2gb memory modules being used, and like because not all data can fit in one module, it has to be spread on multiple, so the data read time is increased, hence hash drops.
How can this be fixed with a software ?
How can this be fixed with a software ?
You guys are really silly.
Date Registered: August 05, 2017, 11:47:49 PM
4 posts...
robin hood is here to ripp of your ass with backdored driver.
Why AMD did not post this before, instead of this you trust to some guy who uploads drivers to dropbox.
Why not direct link from AMD?
lol
Date Registered: August 05, 2017, 11:47:49 PM
4 posts...
robin hood is here to ripp of your ass with backdored driver.
Why AMD did not post this before, instead of this you trust to some guy who uploads drivers to dropbox.
Why not direct link from AMD?
lol
I understand your opinion, I am OK with that, it is normal that people thinking like that, but let me tell you that I don't need to do this, but I did it, I want to help all the mining community.
Try it, if it is working for you maybe you can tell me just Thank You, or even sent me some donation.
Once again people, you don't need to use it, it is up to you, I use it on more then 600 gpu's, all of them are 470/480, working perfect for a week, you don't need to trust me, but as I can see there is more and more confirmation about this driver, it is working of course.
There is no link to official amd site.
If you have 600 gpu's than you would be here long time member and have had more than 5 posts
I agree, you are not script kiddie, you probably developed this good into driver, so you are good coder
Nice work, but i still believe this is malware.
Can anyone here using this solution confirm one thing that I've noticed for awhile now? I usually gauge my rig's performance by the # of shares submitted / hour. It's usually a rough estimation after a couple of hours of mining and I notice often the highest hash rate doesn't necessarily correlate with share submit rate. The pools I use usually are prop or PPLNS and really sharefinding is how we all find blocks. The reading from claymore/ethminer is a just a spit out of the current calculation our gpu thinks it can do and reports to the pool that later on will make a judgement on. I've done all the fine tuning on my cards to the point where I trim back the mem overclock just enough to not incur mem errors on HWIN64 and max out my share rate. So my hash rate maybe a bit under avg yet my % for the block round is higher than expected. Essentially i try to punch above my weight class.
So back to my orig question: Have you guys noticed a commensurate increase in share sub/accept rate along with the uptick in 'reported' hash rate? Because if it's all the same I'd rather not push my cards further if it doesn't actually give me any tangible benefit other than bragging rights.
Thanks
So back to my orig question: Have you guys noticed a commensurate increase in share sub/accept rate along with the uptick in 'reported' hash rate? Because if it's all the same I'd rather not push my cards further if it doesn't actually give me any tangible benefit other than bragging rights.
Thanks
Jump to: