Omg9500
Newbie
Offline
Activity: 112
Merit: 0
|
|
March 11, 2018, 02:18:08 PM |
|
effective hashrate has been greatly reduced with latest update. anyone else?
Yes, my RX580s are now 18Mh/s and my RX480s are now 14MH/s. This is the same on 3 of my rigs. I looking into how to revert the update or I'll blow these away and rebuild on a linux distro or Windows 7. Updated: I went to settings and recovery and rolled back to the previous update. I'm back to 33Mh/s and 30Mh/s. Thanks Microsoft for your amazing fail. Probably it's not the windows update but windows update updated you GPU driver. You can prevent this by running gpedit.msc administrative tools and windows update .... https://www.itechtics.com/disable-automatic-driver-updates-windows/
|
|
|
|
|
|
|
|
|
"With e-currency based on cryptographic proof, without the need to
trust a third party middleman, money can be secure and transactions
effortless." -- Satoshi
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
peterboy1
Newbie
Offline
Activity: 168
Merit: 0
|
|
March 11, 2018, 02:24:16 PM |
|
effective hashrate has been greatly reduced with latest update. anyone else?
Yes, my RX580s are now 18Mh/s and my RX480s are now 14MH/s. This is the same on 3 of my rigs. I looking into how to revert the update or I'll blow these away and rebuild on a linux distro or Windows 7. Updated: I went to settings and recovery and rolled back to the previous update. I'm back to 33Mh/s and 30Mh/s. Thanks Microsoft for your amazing fail. you must be talking about reported/client's hashrate, not the effective/pools calculation hashrate which is the most important. i was talking about average EFFECTIVE hashrate which has decreased much in this version.
|
|
|
|
janding
Jr. Member
Offline
Activity: 170
Merit: 6
|
|
March 11, 2018, 02:33:09 PM |
|
effective hashrate has been greatly reduced with latest update. anyone else?
Yes, my RX580s are now 18Mh/s and my RX480s are now 14MH/s. This is the same on 3 of my rigs. I looking into how to revert the update or I'll blow these away and rebuild on a linux distro or Windows 7. Updated: I went to settings and recovery and rolled back to the previous update. I'm back to 33Mh/s and 30Mh/s. Thanks Microsoft for your amazing fail. you must be talking about reported/client's hashrate, not the effective/pools calculation hashrate which is the most important. i was talking about average EFFECTIVE hashrate which has decreased much in this version. After a windows update I uninstalled the AMD drivers and installed them again. That fixed all the issues with hash rates that dropped.
|
|
|
|
whoismoses
Member
Offline
Activity: 247
Merit: 59
|
|
March 11, 2018, 02:54:27 PM |
|
11.3 helps RX550 cards a lot. Among all my cards I have 18 Asus RX 550 4GB. Each card is up 1.5-2.5 MH/s...
It is like adding a whole another RX 580!
|
|
|
|
Claymore (OP)
Donator
Legendary
Offline
Activity: 1610
Merit: 1325
Miners developer
|
|
March 11, 2018, 04:22:24 PM |
|
Hello, @Claymore! Since it's getting warmer outside it would be fine to consider -ttli affect while rebooting by -minspeed. E.g. "-minspeed 10250" will reboot rig slower than 250Mh, but only if any of GPUs are throttling.
PS My previous problem is still there. New miner instance cannot take 3333 port after restarting in reboot.bat without actual rig reboot. Different rigs, different Windows installations, days of waiting - socket is busy anyway after "-minspeed" restarting. Also I noticed if I close miner with "X" button above, it shows "Quit, please wait..." for a while, after this sockets close immidiately. Maybe if you could execute procedures hidden in this "please wait" sign after minspeed detection, it will solve the problem?
Ok I will check it. PPS -dcri autodetection is great! Got +12 Mh from my rigs. Thanks! ) BTW, is best dcri different for different DAGs? Last year I was measuring best dcri for each GPU by myself, but yesterday realized all of them are wrong...
I did not test it, but for recent drivers it must not depend on DAG much.
|
|
|
|
Ursul0
|
|
March 11, 2018, 04:44:07 PM |
|
@Claymore can you maybe query system for the driver version and show it in Remote Manager ?
|
|
|
|
bithead007
Newbie
Offline
Activity: 22
Merit: 0
|
|
March 11, 2018, 04:53:42 PM |
|
Ok I will check it.
<snip>
I did not test it, but for recent drivers it must not depend on DAG much.
Thanks for the update. I'm noticing crashes. RX570 and 1060 card are all that are installed, dual mining DCR, mem clock only 2000 on the 570. This was stable on 10.5. 09:36:56:840 77c ETH: GPU0 15.884 Mh/s, GPU1 19.378 Mh/s <snip> 09:38:50:088 77c ETH: GPU0 0.000 Mh/s, GPU1 19.371 Mh/s 09:39:04:762 f0 WATCHDOG: GPU 0 hangs in OpenCL call, exit Windows 10, 18.1.1 AMD driver CJ
|
|
|
|
Teress
|
|
March 11, 2018, 04:55:57 PM |
|
@Claymore I switched to linux version of Claymore some time ago. But am I blind or why is linux version storing in log files less info then win version? For example one of my rigs is rebooting. I want to analyze problem reading logs. On windows version I saw in that case for example "Gpu0 failed..". On linux I can see only "NVML: cannot get fan speed, error 999" but NO info about which gpu crashed the driver. Can anybody point me right direction? Thanks a lot...
BTW: v11.2 & v11.3 is crashing after a while all my Nvidia rigs, previous version was absolutelly stable.
|
|
|
|
Frippster
Newbie
Offline
Activity: 27
Merit: 0
|
|
March 11, 2018, 05:10:00 PM |
|
11.3 crashed all three rigs within 5 minutes. 11.2 has been perfectly stable between my 72 hour reboots.
|
|
|
|
andrenym00
Newbie
Offline
Activity: 25
Merit: 0
|
|
March 11, 2018, 05:21:00 PM |
|
Which driver is better, blockchain driver or current amd driver set to compute?
|
|
|
|
Vann
|
|
March 11, 2018, 05:25:23 PM |
|
The current AMD drivers in compute mode are optimized for Windows 10 v1709 FCU. Those are the fastest drivers to use with RX 570/580 cards. The only reason to use the blockchain drivers on Windows v1709 with RX 570/580's is some old miners like SGminer don't work with the latest drivers.
|
|
|
|
big al
Member
Offline
Activity: 140
Merit: 10
|
|
March 11, 2018, 05:58:54 PM |
|
11.3 crashed all three rigs within 5 minutes. 11.2 has been perfectly stable between my 72 hour reboots.
I'm finding the same. 11.3 is faster but less stable. If I slow down the cards a bit it becomes stable. But that negates the speed gain.
|
|
|
|
Ursul0
|
|
March 11, 2018, 06:05:02 PM |
|
GPU hangs are most often from to a high overclock, especially on the memory or too much undervolt for that card. Sometimes draining all the power from the capacitors also helps reset the GPU's and risers after a crash. Shutdown down the rig, cutoff the power to the PSU with the switch on the back and press the power button to drain all the power. Then after 1-2 minutes turn the rig back on.
yep. this can be healthy and helpful sometimes. you don't need to wait 1 minute though:) As for the 32MH - I'd guess it's rx*80 Polaris, so you should expect it to be unstable at this rate in any case just as undervolt could cause it the stock voltages could be also too much for the card to handle, so just start from lower mem overclock
|
|
|
|
cryptmaniac123
Newbie
Offline
Activity: 3
Merit: 0
|
|
March 11, 2018, 06:09:39 PM |
|
Most pools have a min. payout hurdle.... I am on Nanopool and there for example the min. payout is 0.2 ETH. Could that be the reason? sounds like that could be the reason! i did not think of that, ill look into it. Thanks !!
|
|
|
|
ildotoress
Newbie
Offline
Activity: 77
Merit: 0
|
|
March 11, 2018, 06:30:51 PM |
|
The current AMD drivers in compute mode are optimized for Windows 10 v1709 FCU. Those are the fastest drivers to use with RX 570/580 cards. The only reason to use the blockchain drivers on Windows v1709 with RX 570/580's is some old miners like SGminer don't work with the latest drivers.
In my experience, in what mining is concerned, the Beta Blockchain drivers are identical to all new drivers. No speed difference whatsoever. If someone has proof otherwise please come forward.
|
|
|
|
WhackOBill
Jr. Member
Offline
Activity: 73
Merit: 3
|
|
March 11, 2018, 06:39:56 PM |
|
I have a six 470 rig, modded BIOS with Claymore 11.2 and the following config:
-dcri 1 -cclock 1240 -mclock 2200 -cvddc 950 -tt 72 -fanmin 50 -fanmax 90 Does anyone know why I'm getting about 27MH/s for the 5 GPUs and 32MH/s for 1 of them? The GPUs are identical and bought on the same batch.
Because you're only overclocking one of them? Is that a rhetorical question? Perhaps try -cclock 1240,1240,1240,1240,1240,1240 to overclock all 6 at the same level.
|
|
|
|
Omg9500
Newbie
Offline
Activity: 112
Merit: 0
|
|
March 11, 2018, 06:50:34 PM |
|
The current AMD drivers in compute mode are optimized for Windows 10 v1709 FCU. Those are the fastest drivers to use with RX 570/580 cards. The only reason to use the blockchain drivers on Windows v1709 with RX 570/580's is some old miners like SGminer don't work with the latest drivers.
In my experience, in what mining is concerned, the Beta Blockchain drivers are identical to all new drivers. No speed difference whatsoever. If someone has proof otherwise please come forward. Beta blockchain drivers have a limit of upto 8 GPUs. New drivers don't... Speedwise maybe no difference so what is the harm of using newer drivers?
|
|
|
|
sandbagger
Newbie
Offline
Activity: 18
Merit: 0
|
|
March 11, 2018, 07:03:21 PM |
|
Win10/Defender problem here that has been driving me nuts and cant find any info
Has happened on all the versions of Claymore that I have run so its not a claymore problem
Every day at 12:30Pm claymore crashes and restarts, when this happens one of my cards(random card) does not come back with the overclocked settings from Afterburner. Using latest version of Win10
I have Defender turned off along with auto updates under group policy and used regedit too. Under Security Center it also says its off, but every day under device Performance and Health it says last scan with the current date.
I even added AVAST a few days ago and that wont stop it from updating under device perf either I also have periodic scanning turned off under virus and threat protection
anyone have any clues as to how to fix this? Thanks
|
|
|
|
rockmoney
|
|
March 11, 2018, 07:25:16 PM Last edit: March 11, 2018, 08:39:28 PM by rockmoney |
|
11.3 crashed all three rigs within 5 minutes. 11.2 has been perfectly stable between my 72 hour reboots.
@Claymore I switched to linux version of Claymore some time ago. But am I blind or why is linux version storing in log files less info then win version? For example one of my rigs is rebooting. I want to analyze problem reading logs. On windows version I saw in that case for example "Gpu0 failed..". On linux I can see only "NVML: cannot get fan speed, error 999" but NO info about which gpu crashed the driver. Can anybody point me right direction? Thanks a lot...
BTW: v11.2 & v11.3 is crashing after a while all my Nvidia rigs, previous version was absolutelly stable.
I have 2 rigs, each with 6 NVIDIA GTX 1070 8GB cards, and my MSI laptop has 1. A few months ago, I decided to add Windows 10 OS, as I was having problems with Linux on the x6 card rigs (I had large capacity drives, so partitioned them in order to keep Linux in case I ever want to test anything and compare results). Sounds like we likely experienced the same and/or similar problem, as all 3 of my machines were running fine with previous versions. After an hour of testing, adjusting, and tweaking some things, everything has been stable on v11.3 for almost 24 hours now. For me, I believe the majority of the crashing was caused by the auto fan control feature in MSI Afterburner (which I prefer for overclocking, and read a few posts elsewhere in this thread from others about that program causing conflicts). Anyway, the fix was to disable fan control in MSI Afterburner, and let Claymore miner control the fans. I have x3 powerful fans controlled by the motherboard on each rig anyway, as I like to keep my cards cool. Obviously, use caution, keep a close eye on temps, and make sure the fans on your cards are working correctly if you attempt this fix on your NVIDIA cards! I was also running the latest Windows 10 x64 391 official NVIDIA drivers, but rolled them back to the 388.71, although I don't believe this was any cause of the problem. I will reinstall the latest drivers later today and report back after running stably for at least 24 hours.
On a different subject, I did download a virus from the latest zip file found in Claymore's link to the Google Drive download. After extracting the download, there were x3 folders within the main one named "cuda6.5", "cuda7.5", and "cuda9.1". Inside each of those folders were x1 .dll file and a separate "EthDcrMiner64.exe". I always scan compressed files downloaded from the internet after extraction, and thankfully it paid off! The "EthDcrMiner64.exe" file in the "cuda6.5" folder was flagged as a backdoor trojan virus - NOT the usual false positive "bitcoin/coin miner" virus. Not sure how or why that file got in there, but it was definitely a very real threat that still has me on edge and VERY uneasy! I just attempted to download the same file found in the Google link in the OP, and now it gets flagged as a virus and won't allow me to download. I have used Claymore's miners for quite a while and never had any problems with his programs, so I do NOT believe Claymore would purposely and knowingly include a virus which allows remote access to your pc (ultimately for the purposes of stealing your coins, digital assets, credit card numbers, personal details, etc...). The infected file was downloaded via the Google link in the OP approximately 3 hours after its release. If the people who create and use these viruses spent half the time it would take to create such a thing, they could make plenty of money on their own (legally, and without having to steal). The thieving scums who do this continue giving Bitcoin/digital assets a bad rep and seriously impact this emerging sector in a very negative way which hurts all of us (not to mention the lack of words that describe all lowlife bottom feeding thieves). Anyway, USE EXTREME CAUTION & ALWAYS SCAN DOWNLOADED FILES BEFORE USING!Here is a screenshot of Claymore's Miner showing as a false positive "Trojan.BitcoinMiner". Although correctly flagged (as a miner could be negligently installed on a persons computer to mine for the attacker), it is a false positive in this case, as we want the file to mine for ourselves. The screenshot shows what your antivirus program should falsely flag. In my case, the antivirus program found two separate threats - the one seen in screenshot, and another which was a very real and dangerous threat that would have unwillingly allowed the attacker remote access to my computer (sorry, but I do not have a screenshot of the real virus. as I immediately removed it). Screenshot of false positive on imgur.com via or https://imgur.com/a/XH75d
|
|
|
|
Vann
|
|
March 11, 2018, 07:32:52 PM |
|
The current AMD drivers in compute mode are optimized for Windows 10 v1709 FCU. Those are the fastest drivers to use with RX 570/580 cards. The only reason to use the blockchain drivers on Windows v1709 with RX 570/580's is some old miners like SGminer don't work with the latest drivers.
In my experience, in what mining is concerned, the Beta Blockchain drivers are identical to all new drivers. No speed difference whatsoever. If someone has proof otherwise please come forward. I have a six card RX 570/580 rig with dual boot Windows 10 v1709 partitions. The latest Adrenaline drivers on one and the Aug 23rd blockchain drivers on the other, so I can run older miners that are not compatible with the current drivers if I need to. Using the same settings for both, in addition to the blockchain drivers being noticeably slower by 3-4 MH/s, they have frequent drops in hash rate and issues with Windows as well. The AMD blockchain drivers came out before the Windows 10 v1709 FCU and are NOT supported. The v1709 FCU was a major update to how Windows interacts with GPU's and also expanding the previous GPU limit from 8 to 13, which is why AMD had to release new drivers that were compatible with the FCU.
|
|
|
|
|