northcape
Member
Offline
Activity: 98
Merit: 10
|
|
April 09, 2013, 10:22:49 AM |
|
I think I had them disabled already, but just for kicks I enabled one of the buttons on one card .. it destroyed the hashrate and now cgminer reports it as "sick" whatever that means as an update, before i tried to push that blue button, i upgraded to 13.10 drivers and it got me about 5 more kh so I was at 525kh for each card... now i'm wrestling to get all 4 working again ... don't push that button! Since the button just enables a second BIOS, I’d try powering the whole machine off completely, then disable the buttons and power it back on. That should put you right back to where you were.
|
|
|
|
Crazy rider89
|
|
April 09, 2013, 01:44:43 PM |
|
does it show the GPUs in the list once it starts running?
it should show all the possible mining targets that it is setup for, so if you had 2 gpus it should show those on the list (in my experience anyway).
I had an issue where I had to rerun the aticonfig command once I added more GPUs or it would not recognize them. Mine does not allow me to mine on the CPU however, so I don't know any more than that currently
if i write: cgminer -n, i get: [2013-04-07 15:14:09] CL Platform 0 vendor: Advanced Micro Devices, Inc. [2013-04-07 15:14:09] CL Platform 0 name: AMD Accelerated Parallel Processing [2013-04-07 15:14:09] CL Platform 0 version: OpenCL 1.1 AMD-APP-SDK-v2.4 (595.10) [2013-04-07 15:14:09] Platform 0 devices: 2 [2013-04-07 15:14:09] 0 Cypress [2013-04-07 15:14:09] 1 Cypress [2013-04-07 15:14:09] 2 GPU devices max detected I think that the gpus are correctly configured. I will try to update the driver. I think i had to reinstall the driver after I booted the system with all cards plugged in. The driver would not auto activate all cards. So if you install the drivers with 1 card plugged in, then restart and plug all of them in, I had trouble getting it to show them all. Then i reinstalled the drivers while the system had all cards plugged in and it worked after I re-ran: sudo aticonfig --adapter=all -f --initiall Thanks, but If I want to mine BTC everything works fine, I have a doubt....I set the memory GPU clock to 300mhz, but I've seen that a lot of users set an high memory speed (around 1200 for a 5870), my question is: do the LTC mining use also the GPU memory?
|
|
|
|
chrcoe01
|
|
April 09, 2013, 01:47:40 PM |
|
yes, i believe it does because when you lower the gpu memory very low, the hashrate suffers
|
"You may delay, but time will not, and lost time is never found again." -Benjamin Franklin
|
|
|
chrcoe01
|
|
April 09, 2013, 01:51:24 PM Last edit: April 09, 2013, 02:34:49 PM by chrcoe01 |
|
I think I had them disabled already, but just for kicks I enabled one of the buttons on one card .. it destroyed the hashrate and now cgminer reports it as "sick" whatever that means as an update, before i tried to push that blue button, i upgraded to 13.10 drivers and it got me about 5 more kh so I was at 525kh for each card... now i'm wrestling to get all 4 working again ... don't push that button! Since the button just enables a second BIOS, I’d try powering the whole machine off completely, then disable the buttons and power it back on. That should put you right back to where you were. i did this, then when it came back up from rebooting, bam! cgminer only saw 2 cards again. so I went into my bios (asus rampage iv) which has a GPU post menu to see what GPUs the bios sees. It seems that some times when I boot, it only sees 2, other times it sees 3, i've only had it see all 4 about twice. i searched through the manual over and over to no avail. There's an 'ez_plug' which is supposedly extra power for the pci-e slots, but i thought this is not needed if each card receives its own power (each from its own rail which is how i have it setup). Maybe the bios needs the extra juice to just turn on all 4 slots at once. I'll see about how to get an extra plug in there, maybe that's it. (wishful thinking lol) edit: did some research on this and it turns out that you are supposed to plug in the ez_plug when running 4 GPUs which is an additional 6pin PCIe connector, which I seem to be lacking on my silverstone st1500... with 4 gpus, it takes up all the pcie connections, so I am going to have to use one of those dual molex to 6 pin adapters to get the extra power. I guess it allows extra power to the x16 only slot to be able to turn on when there's 3 other gpus plugged into pcie slots 2,3 and 4... oh well I guess i'll settle for 2 gpus mining until i can get to the store for that connector :/ in other news, the powerboost button, I turned it off on that one card, and have rebooted/reseated it several times and cgminer just keeps reporting it as being sick, then dead after 15 minutes. It was working flawlessly (at the same crappy 525kh/s) prior to pushing that button. I think since I have time, I will exchange it for one of the gigabyte versions that FullLife mentioned since he was able to get ~720 kh with it. If that works with similar settings etc, I will probably exchange all three for that card.
|
"You may delay, but time will not, and lost time is never found again." -Benjamin Franklin
|
|
|
wowfreaks22
Newbie
Offline
Activity: 41
Merit: 0
|
|
April 09, 2013, 07:02:58 PM |
|
7950 vapor-x with 950/1250 at 0.950volt, 170watt, 572 k/h
think new gigabyte cards are voltage locked, not sure if just up or both
|
|
|
|
pekv2
|
|
April 09, 2013, 08:39:29 PM |
|
7950 vapor-x with 950/1250 at 0.950volt, 170watt, 572 k/h
think new gigabyte cards are voltage locked, not sure if just up or both
Nice, I mine @ 170watts for 400 khash/s. 5850. Nice electric cost with a big hash rate boost. Cannot wait for my 7950, thinking about picking up a http://www.newegg.com/Product/Product.aspx?Item=N82E16814202030
|
|
|
|
jhd
|
|
April 09, 2013, 09:22:17 PM |
|
Hello, i have questions about 5970 cards. Is it possible to use 3 cards (6 gpu) under windows? Dummy plugs are mandatories ?? How can i undervolt these cards to run 3 cards on 1000w psu. Thanx in advance
|
|
|
|
Flying Hellfish
Legendary
Offline
Activity: 1764
Merit: 1756
Verified Bernie Bro - Feel The Bern!
|
|
April 10, 2013, 01:04:05 AM |
|
Newb question from a guy that is not very technical, sorry in advance if the questions I ask are silly.
I just bought a 7950 for fooling around with mining LTC. I had installed the GUI Miner Scrypt (tyvm TacoTime) as I'm not really confident with CGMiner and couldn't even get mining with Reaper.
After installing the card and latest AMD drivers I loaded up GUIMiner and connected to Give-me-ltc pool (stratum server) using the pre-configured 7950 high settings (below).
worksize 256 vectors 1 aggression 20 threads_per_gpu 1 gpu_thread_concurrency 21712
I managed to get ~550 kh/s for a couple of hours (still a little low from what I read in this thread?). I had to re-start the machine so I shut the miner off and re-started, when I opened GUIMiner and connected back to the pool I got connection error and hardware errors. The only way I can get back mining is to set the pre-configured settings to 7950 low and its starts mining but I can only get to ~300kh/s now. if I increase the settings back to the 7950 high settings above while mining it continues to run and only increase's my HR by about 10-15kh/s.
I did try to run cgminer directly but as I am green I can't see where to make all the changes to the different settings to match the above. What is also interesting to me is that CGMiner reports my HR avg as ~550 mh/s but my pool dashboard shows me only at ~7kh/s
rig is a little old rough specs are: intel core 2 duo CPU 4G Ram OS Vista (eww ya I know but its an older machine I had around not being used so I figured I would throw a 7950 in and screw around mining LTC...)
Is there any advice one of you grizzled veterans/smart people could throw a newbie's way, please?
|
|
|
|
IlbiStarz
|
|
April 10, 2013, 01:12:43 AM |
|
I have a 6870 and 5830 in one rig. When I use CGMiner both of them large amounts of HW errors, about 20-30%. Neither of them are overclocked...does anyone know why this is happening?
|
|
|
|
rocks
Legendary
Offline
Activity: 1153
Merit: 1000
|
|
April 10, 2013, 05:09:00 AM |
|
I have a 6870 and 5830 in one rig. When I use CGMiner both of them large amounts of HW errors, about 20-30%. Neither of them are overclocked...does anyone know why this is happening?
I had this exact same rate when my thread_concurrency was too low, bumping up the TC to 4x the number of shaders fixed it.
|
|
|
|
Starlightbreaker
Legendary
Offline
Activity: 1764
Merit: 1006
|
|
April 10, 2013, 06:13:04 AM |
|
finally got my 3x7950.
....voltage locked.
FFFFFFFFFUUUUUUUU
3 cards pulling 1kw = omfg.
|
|
|
|
|
Starlightbreaker
Legendary
Offline
Activity: 1764
Merit: 1006
|
|
April 10, 2013, 03:17:49 PM |
|
HIS IceQ 7950, lol
shit, might have to sell and replace it with sapphire ones.
|
|
|
|
chrcoe01
|
|
April 10, 2013, 05:00:27 PM |
|
anyone have any luck with sapphire's trixx program? https://www.sapphireselectclub.com/ssc/trixx/trixx.aspxit looks like it should be able to allow us to change the voltage on the vapor-x GHz 7970's (among others) there's a list at the bottom of that link which tells which features are supported for which cards.
|
"You may delay, but time will not, and lost time is never found again." -Benjamin Franklin
|
|
|
mufa23
Legendary
Offline
Activity: 1022
Merit: 1001
I'd fight Gandhi.
|
|
April 10, 2013, 08:19:33 PM |
|
do the LTC mining use also the GPU memory?
I'm curious about this as well. I'm thinking about buying a bunch of 7850's, but didn't know if I should get the 2GB or 1GB model. The 2GB ones are typically $30 more.
|
Positive rep with: pekv2, AzN1337c0d3r, Vince Torres, underworld07, Chimsley, omegaaf, Bogart, Gleason, SuperTramp, John K. and guitarplinker
|
|
|
pox67
Newbie
Offline
Activity: 37
Merit: 0
|
|
April 10, 2013, 09:11:03 PM |
|
finally got my 3x7950.
....voltage locked.
FFFFFFFFFUUUUUUUU
3 cards pulling 1kw = omfg.
Brutal, I have 4 7970's in two different computers (2 per computer) and they are pulling 1050watts for both rigs. Advantage of being able to get VDDC down to 1.025-1.030. Cards are Sapphire Dual-X OC and MSI OC Boost. I was fiddling around some more last night and found something interesting. On the Sapphire cards the best I have been able to get is 950/1425 for 650KHs. I am OK with that as they are drawing little power doing it. I was pushing the memory up and down trying to get a little more out of them and found nothing better up to 1700mhz on the RAM. On a whim I tried 1220mhz RAM and got exactly 650KHs, weird. That seems to be the best so left them running at 950/1220, wish I could drop the RAM voltage below 1.5v... This is using CGminer with 2048 shaders, 8192 Thread Concurency and - g2 and intensity 13.
|
|
|
|
Anvi
Newbie
Offline
Activity: 40
Merit: 0
|
|
April 10, 2013, 10:05:44 PM |
|
Hello, i have questions about 5970 cards. Is it possible to use 3 cards (6 gpu) under windows? Dummy plugs are mandatories ?? How can i undervolt these cards to run 3 cards on 1000w psu. Thanx in advance I don't see a problem with it software wise, but u'll most likely need stronger PSU or add 2nd PSU to the system. Does your PSU have enough PCIe 6+2 cables? Dummy plugs aren't required anymore (as of Catalyst 12.8 if I recall correctly) Use MSI Afterburner to control voltage etc.
|
|
|
|
northcape
Member
Offline
Activity: 98
Merit: 10
|
|
April 10, 2013, 11:18:28 PM |
|
finally got my 3x7950.
....voltage locked.
FFFFFFFFFUUUUUUUU
3 cards pulling 1kw = omfg.
Yikes, I’d definitely send them back. My rig with four Sapphire 7950s pulls 940W and that’s using an i5 730 at default clock/voltage and two different PSUs.
|
|
|
|
gfoot
Newbie
Offline
Activity: 14
Merit: 0
|
|
April 11, 2013, 03:38:04 AM Last edit: April 12, 2013, 10:27:06 AM by gfoot |
|
I have a few questions regarding LTC mining in general:
1) 7970 GHz edition: my card comes with factory clocks of 1050/1500; using various thread concurrency settings and intensity settings, I think the best I ever manage is a measly 450khash. I have also had issues with getting GUIminer-scrypt to work properly on this machine -- I think the best I got was maybe 300-350khash. If someone has some suggestions, I'm all ears!
Try --thread-concurrency 3072 -g 9 -I 13 --gpu-powertune 20, and possibly --worksize 256Edit: terrible advice - although it reports a higher hash rate, it just generates garbage (never gets a share). Do try using the --powertune option though, and possibly --thread-concurrency 8192 -g 5 or thereabouts - which does seem to give me a slight edge over large concurrencies and -g 1.
|
|
|
|
Tejsei3
Newbie
Offline
Activity: 36
Merit: 0
|
|
April 11, 2013, 04:50:50 AM |
|
My current best setup is 925/1250 at 0.962V for 560 kH/s at 170W or 3.29 kH/s/W, after many hours of systematically mapping gpu clock (
3.3 kh/J is great man!! Have you done any work with 58xx cards? Cheers! Just the 7950 but if I do I'll share it here. I think I mostly got lucky just lowering the voltage at the stock gpu clock and memclock. I looked all over for more efficient combinations and nothing was as efficient as 925/1250/0.962V. I have four Sapphire 7950s and can confirm that these are by far the best settings I’ve found so far, thank you for sharing them. I run four cards at 925/1250/0.950, although that 0.950 is set via --vddc in cgminer, and GPU-Z shows ~0.996V for all GPUs. So I guess they are actually not undervolted at all. I could probably squeeze some more efficiency out of the system by setting each card to its minimum stable voltage, but mapping the correct cards between cgminer, GPU-Z and Trixx/Afterburner is too much of a hassle. Good to know. I got two more Sapphire 7950 100352-4L's which ARE voltage unlocked, and they also appear to thrive at 925/1250/0.975, 545kH/s, 188W, 2.90 kH/J.
|
|
|
|
|