elzium
Member
Offline
Activity: 96
Merit: 10
|
|
December 08, 2016, 03:59:51 PM |
|
All u with psu problems, test ur equipment with eth+sia dualmining, if that doesnt hold, ur system is just underdimensioned.
Best you start rereading the posts the when v9 was released so you actually understand what people are complaining about. I've read them spike issues, thing is, software makes components work within its limits and so far czm used less than max power, now as i understand there are short periods of time that max is used and weak systems fail.
|
|
|
|
smaxz
Sr. Member
Offline
Activity: 430
Merit: 253
VeganAcademy
|
|
December 08, 2016, 04:00:13 PM Last edit: December 08, 2016, 04:51:00 PM by smaxz |
|
Yea. The spikes are quite real. One of my pc's has a 730w psu powering a rx 480 and a 280x. This combination has been running stable for months without problems (even eth/sia dual mining). A few minutes ago, the pc rebooted and said "a power surge was detected - press F1 to continue". There is nothing else running on the system except claymore v9. there are crashes often leading upto the donation mining portion of mining which makes me wonder if some dual mining mechanism isnt triggered during the donation period which is more taxing on the system. how old is your power supply? did it you manage tog et up and running again without issue? My PSU is new (Thermaltake SmartSE 730W). System is running again - stable so far. Mainboard, CPU and RAM are new too. My farm is littered with older gpu's so i share an evga supernova between multiple rigs on the 6/8pin side of the gpu's and leave the risers and gpu0 powered fully by psu plugged into the motherboard (usually around 500 watt) I have yet to experience any spikes that would take my system down but my motherboards are mainly higher end (but outdated) asus which may increase my immunity to this phenomenon. Some of those with the solely next gen gpu's may benefit from this strategy also if spikes continue to result in a power flutter.
|
- NGdTwHRSdnThdi1drQuHGT3khAHRtZ1HMq -
|
|
|
CoolnQuiet
Sr. Member
Offline
Activity: 388
Merit: 288
Living the dream
|
|
December 08, 2016, 04:18:13 PM |
|
I also have two RX480s and I am experiencing sudden power spike issue on v9. It happens randomly, however it causes system to restart. There was no such problem on v8. My PSU is Corsair 1200W (gold).
|
|
|
|
JuanHungLo
|
|
December 08, 2016, 05:08:15 PM |
|
Interesting problem that has shown up in 3 of my 280x rigs with 4 GPUs each: Out of 8 rigs, 3 of them have had the first GPU (as displayed by Claymore's Ethereum Dual Miner Manager 2.5), actually GPU 2, go up to about 325 Sol/s with errors in red. This did not happen until about 22-24 hours after release of the V. 9.0 beta. Restarting the miner software results in 0.00, obviously. Using Device manager disable/enable resets it and it works again. Any idea what is happening?
I set these rigs to run cool with undervolt to 1125, 1000/1500, i 4.
Anybody else experience this?
|
Bull markets are born on pessimism, grow on skepticism, mature on optimism, and die on euphoria. - John Templeton
|
|
|
kmarx
Newbie
Offline
Activity: 6
Merit: 0
|
|
December 08, 2016, 05:15:14 PM |
|
Just to share what I've been observing with the "power spikes" and HWINFO64. Spikes seem to occur every 5-6 polling intervals (time = 100ms)
V9 (Intensity 8 ):
Fury Nitro:
1110/550, -108mv, ~352 S/s;
GPU Core Current: AVG 153A, but spikes as high as 247A
GPU Core Power: AVG 173W, but spikes as high as 280W
1140/550, -12mv, ~360 S/s;
GPU Core Current: AVG 175A, but spikes as high as 310A
GPU Core Power: AVG 210W, but spikes as high as 370W
RX470 Nitro 8GB:
1300/2100, -100mv, ~223 S/s;
GPU Core Current: AVG 78A, but spikes as high as 126A
GPU Core Power: AVG 78.3W, but spikes as high as 126W
V8 (Intensity 4 ):
Fury Nitro:
1110/550, -108mv, ~300 S/s;
GPU Core Current: AVG 145A, but spikes as high as 192A
GPU Core Power: AVG 165W, but spikes as high as 218W
1140/550, -12mv, ~310 S/s;
GPU Core Current: AVG 180A, but spikes as high as 224A
GPU Core Power: AVG 207W, but spikes as high as 274W
RX470 Nitro 8GB:
1300/2100, -100mv, ~190 S/s;
GPU Core Current: AVG 75A, but spikes as high as 99A
GPU Core Power: AVG 74.5W, but spikes as high as 99.6W
|
|
|
|
vatusasid
Newbie
Offline
Activity: 54
Merit: 0
|
|
December 08, 2016, 05:39:41 PM |
|
Just to share what I've been observing with the "power spikes" and HWINFO64. Spikes seem to occur every 5-6 polling intervals (time = 100ms)
V9 (Intensity 8 ):
Fury Nitro:
1110/550, -108mv, ~352 S/s;
GPU Core Current: AVG 153A, but spikes as high as 247A
GPU Core Power: AVG 173W, but spikes as high as 280W
1140/550, -12mv, ~360 S/s;
GPU Core Current: AVG 175A, but spikes as high as 310A
GPU Core Power: AVG 210W, but spikes as high as 370W
RX470 Nitro 8GB:
1300/2100, -100mv, ~223 S/s;
GPU Core Current: AVG 78A, but spikes as high as 126A
GPU Core Power: AVG 78.3W, but spikes as high as 126W
V8 (Intensity 4 ):
Fury Nitro:
1110/550, -108mv, ~300 S/s;
GPU Core Current: AVG 145A, but spikes as high as 192A
GPU Core Power: AVG 165W, but spikes as high as 218W
1140/550, -12mv, ~310 S/s;
GPU Core Current: AVG 180A, but spikes as high as 224A
GPU Core Power: AVG 207W, but spikes as high as 274W
RX470 Nitro 8GB:
1300/2100, -100mv, ~190 S/s;
GPU Core Current: AVG 75A, but spikes as high as 99A
GPU Core Power: AVG 74.5W, but spikes as high as 99.6W
I have the R9 Nano. The GPU will have flucuate between 0% to 100% power all the time, and then it crashes.
|
|
|
|
citronick
Legendary
Offline
Activity: 1834
Merit: 1080
---- winter*juvia -----
|
|
December 08, 2016, 05:48:44 PM |
|
Just to share what I've been observing with the "power spikes" and HWINFO64. Spikes seem to occur every 5-6 polling intervals (time = 100ms)
V9 (Intensity 8 ):
Fury Nitro:
<snipped>
I have the R9 Nano. The GPU will have flucuate between 0% to 100% power all the time, and then it crashes. My Nano rigs got the greatest boost from v9 but some of them went as high at 350H/s then died.... died as in gone from device manager -- so hard reboot and reduced the intensity to 4. Some Nanos wre still coping at i=6 I think the "maximal" that Claymore uses the i=7 and i=8 to give that 1% additional boost -- maybe causing the spikes. I can only imagine he may be exploring deeper into the power architecture to get that extra boost. i=6 is safe zone for me and I will set this as the gold standard for v9, else i=4 then i=0 -- last resort v8. v8 is , IMHO, the most stable version of all. Well to be fair, v9 is labeled BETA....
|
If I provided you good and useful info or just a smile to your day, consider sending me merit points to further validate this Bitcointalk account ~ useful for future account recovery...
|
|
|
Xardas2014
|
|
December 08, 2016, 05:52:20 PM |
|
I decided to give dwarfpool a try for ZEC since other pools were showing a lot less hash than my miners were producing. It has been about 24 hours of testing and the results are rather dramatic. Rigs tested averaged ~1080 sol/s shown in Claymore.........on coinmine.pl I was paid for ~830 sol/s.......at dwarfpool I got paid for ~1100 sol/s. Dwarfpool FTW!
|
|
|
|
Hotmetal
|
|
December 08, 2016, 05:53:59 PM |
|
Anyone having these Power Spikes/Shutdowns with Tahiti GPUs?
I monitored mine at the wall, and the power was constant the entire time. It never went more than 5% off from the average wattage.
Don't have any issues however I got like 2000 Watts of PSU power, powering only 5x 280x since I needed the extra PCIe plugs.
Maybe its only with the RX series, however for those its not worth the switch since ETH is just as profitable at the moment.
You're not going to catch these spikes at the wall. Use Corsair link if you have that PSU's and you can see it easily. Some people have reported HWINFO catches it as well but i haven't tested that.
|
|
|
|
ffthomas
|
|
December 08, 2016, 05:56:54 PM |
|
Same problem here.. But the interesting thing is that I don't experience this random reboot on all of my rigs. I use default intensity (-i 6) on all rigs and tried to set it lower on the crashed rigs.. it didn't help. All rigs are built with the same hardware: 6*RX470 with 1600W Enermax PSU.
|
|
|
|
IOTUSA
|
|
December 08, 2016, 05:57:00 PM |
|
I decided to give dwarfpool a try for ZEC since other pools were showing a lot less hash than my miners were producing. It has been about 24 hours of testing and the results are rather dramatic. Rigs tested averaged ~1080 sol/s shown in Claymore.........on coinmine.pl I was paid for ~830 sol/s.......at dwarfpool I got paid for ~1100 sol/s. Dwarfpool FTW!
Does Dwarfpool offer a U.S. node? I was considering moving my rigs there myself.
|
|
|
|
Cakalasia
Member
Offline
Activity: 162
Merit: 10
|
|
December 08, 2016, 06:00:06 PM |
|
Same problem here.. But the interesting thing is that I don't experience this random reboot on all of my rigs. I use default intensity (-i 6) on all rigs and tried to set it lower on the crashed rigs.. it didn't help. All rigs are built with the same hardware: 6*RX470 with 1600W Enermax PSU.
Same here. I even set the intensity to 3 or 4, there are still reandom boots. I have to reduce the frequency, that solve the problems.
|
|
|
|
Qazo
|
|
December 08, 2016, 06:15:17 PM |
|
You're not going to catch these spikes at the wall. Use Corsair link if you have that PSU's and you can see it easily. Some people have reported HWINFO catches it as well but i haven't tested that.
I have RM1000 and downloaded Corsair Link where I can see those spikes?
|
|
|
|
Xardas2014
|
|
December 08, 2016, 06:20:38 PM |
|
I decided to give dwarfpool a try for ZEC since other pools were showing a lot less hash than my miners were producing. It has been about 24 hours of testing and the results are rather dramatic. Rigs tested averaged ~1080 sol/s shown in Claymore.........on coinmine.pl I was paid for ~830 sol/s.......at dwarfpool I got paid for ~1100 sol/s. Dwarfpool FTW!
Does Dwarfpool offer a U.S. node? I was considering moving my rigs there myself. Yes, there .bat setup is different from the ETH pool there, so I'll post a sample .bat........ ZecMiner64.exe -wd 1 -r 0 -i 8 -zpool zec-us1.dwarfpool.com:3336 -zwal t1cPySgKnxaW85tPLJX5nrRXoMiyeMRDX5h.PC -zpsw <your email here to enable email monitoring alerts, or use x> -allpools 1
|
|
|
|
beIbox
Newbie
Offline
Activity: 21
Merit: 0
|
|
December 08, 2016, 06:36:57 PM |
|
Starting from v9 my system hangs and reboot automatically even if I reduce intensity. VGA: VTX3D R9 390 at default clocks 1000/1500 No problem with v8 even at max intensity. Waiting for v. 9.1 to fix this...
|
|
|
|
Prelude
Legendary
Offline
Activity: 1596
Merit: 1000
|
|
December 08, 2016, 06:41:24 PM |
|
Starting from v9 my system hangs and reboot automatically even if I reduce intensity. VGA: VTX3D R9 390. No problem with v8 even at max intensity. Waiting for v. 9.1 to fix this...
9.1 won't fix it. V9 is harder on GPUs, they need more voltage to be stable vs V8.
|
|
|
|
ffthomas
|
|
December 08, 2016, 06:42:44 PM |
|
Same problem here.. But the interesting thing is that I don't experience this random reboot on all of my rigs. I use default intensity (-i 6) on all rigs and tried to set it lower on the crashed rigs.. it didn't help. All rigs are built with the same hardware: 6*RX470 with 1600W Enermax PSU.
Same here. I even set the intensity to 3 or 4, there are still reandom boots. I have to reduce the frequency, that solve the problems. Thanks, I will try to undervolt and/or underclock the GPUs. But I prefer stock settings and after everything is stable with stock bios then I can modify and optimize clock, timing, voltage, etc. But now I have issues when using full stock settings. I hope someone will figure it out why this random reboot happens. I think it is related somehow to the wires, breakers, or something with the electricity and the PSU reacts this way to protect the rig. (My other rigs are in the garage (different wires and breakers) and they are stable with -i 6 and all same hardware and settings as my rigs in the house. I don't overload any breakers not even close.) I'm not an electrician genius so that's not my table.
|
|
|
|
Gagana
Member
Offline
Activity: 80
Merit: 10
|
|
December 08, 2016, 06:50:58 PM |
|
From 8.0 to 9.0
6*RX 470 power consumption raised from 800W to 1000W from the wall.
|
|
|
|
xeridea
|
|
December 08, 2016, 06:57:32 PM |
|
From 8.0 to 9.0
6*RX 470 power consumption raised from 800W to 1000W from the wall.
Odd, mine went from ~860 to ~860.
|
Profitability over time charts for many GPUs - http://xeridea.us/chartsBTC: bc1qr2xwjwfmjn43zhrlp6pn7vwdjrjnv5z0anhjhn LTC: LXDm6sR4dkyqtEWfUbPumMnVEiUFQvxSbZ Eth: 0x44cCe2cf90C8FEE4C9e4338Ae7049913D4F6fC24
|
|
|
|
|