Consolidated Litecoin Mining Guide for 5xxx, 6xxx, and 7xxx GPUs

[hi]

I have two cards 7700, 7850.  
Run cgminer once and one card gets 30khs and the other gets 280khs...how do i make the other card boost up. My cgminer bat file contains --scrypt --intensity 13 --thread-concurrency 7000,8192

any help appreciated.

EDIT:  After looking at temps and settings on afterburner, I realized that my 7700 is pulling the 255kh/s and the 7850 is only getting 30kh/s ?  This makes no sense?

[zulzedd]

Can i ask for help? Here is my problem:
https://bitcointalk.org/index.php?topic=167821.msg1747320#msg1747320

[zulzedd]

I have two cards 7700, 7850.  
Run cgminer once and one card gets 30khs and the other gets 280khs...how do i make the other card boost up. My cgminer bat file contains --scrypt --intensity 13 --thread-concurrency 7000,8192

any help appreciated.

EDIT:  After looking at temps and settings on afterburner, I realized that my 7700 is pulling the 255kh/s and the 7850 is only getting 30kh/s ?  This makes no sense?

My freshly bought 7850 with cgminer on intensity 13 getting me ~300kHash. Here, on forum, i read, that normal thread-concurrency value for 7850 is ~64*246, but cgminer crashes on 8500 or more. So i want use reaper for LTC mining, but got troubles (see prev. message).

[chrcoe01]

i have 4 7970s and for the life of me can't seem to get anywhere near what other say it can get...  the highest I've gotten was 500kh/board if I'm lucky.

i've been using these settings to get 500 (cgminer):

--shaders 2048
-I 13 #any higher and it drops like a rock, any lower, slowly drops until around 10 where it only does around 50-70k
--thread-concurrency 8192 # i've been messing with this and anything but 8192 seems to drop it by about 5kh/s
--gpu-memclock 1700
--gpu-engine 970

I've messed with the clock speeds over and over but on my memclock, anything over 1700 seems to lock up after about 10-15 minutes, so I started messing with the engine to bring it into the .57-.6 ratio range and this is how I finally got it to go to 500kh.

i'm at a loss as to what to try next, if i bump the clocks up any more it will be unstable it seems.

I haven't messed with the voltage, but i wouldn't know what to do with the without destroying the cards.  every guide i've looked at mentions that the voltage shouldn't have to be adjusted to get around 700k

anyone have any ideas what to try next?

[KerryH]

Hey guys,

I wanted to get some input from the more seasoned veterans than myself but I have a mining rig that consists of:
2 x 5870 (~430kh/s ea fine tuned)
1 x 5850 (around 372kh/s fine tuned)
and I just got in a 7850

Now everything was running great w/ cgminer 2.11.3 and drivers 13.2 w/ SDK 2.5 manually installed.

The issue is that I received the 7xxx series card and I reformatted the hard drive to install the current drivers and SDK for it.

Now with just the 7850 in place, I am able to get good performance out of it (around 380kh/s with overclocking) but when i try to have all the cards running at the same time the driver is constantly crashing after some time causing cgminer to crash etc.  I've tried lowering the overclocks and intensity without success.

The same rig has been 48++hr stable with just the 5xxx cards in place and running I 19.

Are there some tips to mixing generations like this?

[qbits]

hmmmm

I have a smalish problem:
- i use ubuntu 12.10+ati 13.3+cgminer 2.11.3 video drivers on 2 rigs

first rig has 6990 it only runs at speeds 610/1250, if I try to use any higher gpu clock than 610 (say 700) system will simply shutdown in 10 min or so. some of you will say that I use shitty 850W PSU from "ace"  and you might be right. when it runs on 610/1250 it seems stable, no hw errors and an average of 650 kH/s, which is not much but seems fine. Here is the script I use for mining:

Code:

export DISPLAY=:0
export GPU_USE_SYNC_OBJECTS=1
export GPU_MAX_ALLOC_PERCENT=100
aticonfig --od-setclocks=610,1250 --adapter=all
aticonfig --od-getclocks --adapter=all
./cgminer --scrypt --api-listen --api-network -c cgminer.conf --thread-concurrency 8192 --lookup-gap 2 --worksize 256 -I 18 -g 1 -o stratum+tcp://notroll.in:3333 -u x -p y

second rig has 2x5830+5850+A4 cpu with some 6xxx type thing in it. it runs stable (for months) mining btc at 600/667 (a4) 920/600 (5830) and 910/600 (5850) with hash rates of: 28 (a4) 282 (5830) and 365 (5850). no world records there, but I'm happy mining with a stable rig that needs virtually no maintanance. LTC is a whole different matter. What ever I tried to do I would only mine for 10-15 minutes at most. When it did it was stable with no HW errors, temp was 75°C, speeds were: 25 (a4), 220 (5830) and 300 (5850) it drew no more power out of the socket as it did when it was mining btc. but after 10 minutes or so it would shut down. this rig however has high quality 1000W Corsair PSU. I Cannot get it to work stable! Here is the script

Code:

export DISPLAY=:0
export GPU_USE_SYNC_OBJECTS=1
export GPU_MAX_ALLOC_PERCENT=100
aticonfig --od-setclocks=615,1200 --adapter=1
aticonfig --od-setclocks=615,1200 --adapter=2
aticonfig --od-setclocks=615,1125 --adapter=3
aticonfig --od-getclocks --adapter=all
./cgminer --scrypt --api-listen --api-network -c cgminer.conf --thread-concurrency 3072,5760,5760,5760 --lookup-gap 2 --worksize 256 -I 18 -g 1 -o stratum+tcp://notroll.in:3333 -u x -p y

it is interesting that in both cases computer would not start at the press of the power button. I had to disconnect power and reaply it and then restart the pc. this would indicate that indeed some sort of protection mechanism inside PSU triggered the shutdown. but what?

Has anyone seen anything like this??? Where are these shut-downs coming from?
Could this be a cpu scaling (cpu changing freq) related thing?
Should I test the rig card by card?
How come there is no problem mining btc even at 920/1000 on all cards?
Should I reduce GPU_MAX_ALLOC_PERCENT? How about GPU_USE_SYNC_OBJECTS?

p.s. third rig i tried this on uses windows, 2x5770 and 450W!!!! Corsair PSU and is able to mine ltc without a single shutdown!!!

[qbits]

first rig has 6990 ....

first rig just solved my first LTC block share score was 16.2M... I should have mined solo

see the screemshot. notice the absence of hw errors? low rejects? best share 16M?

Code:

 cgminer version 2.11.3 - Started: [2013-04-05 20:01:37]
--------------------------------------------------------------------------------
 (5s):694.8K (avg):644.1Kh/s | Q:520  A:2503  R:30  HW:0  E:481%  U:8.9/m
 ST: 2  SS: 40  DW: 1022  NB: 138  LW: 2251  GF: 7  RF: 0  WU: 588.6
 Connected to notroll.in diff 64 with stratum as user XXX
 Block: 5d22ae5f789723cb...  Diff:10M  Started: [00:39:07]  Best share: 16.2M
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0:  74.0C 3723RPM | 329.4K/322.1Kh/s | A:1245 R:14 HW:0 U: 4.42/m I:18
 GPU 1:  74.5C  81%    | 329.6K/322.0Kh/s | A:1258 R:16 HW:0 U: 4.47/m I:18
--------------------------------------------------------------------------------

 [2013-04-06 00:37:49] Accepted 1a10327d Diff 74/64 GPU 1
 [2013-04-06 00:37:49] Stratum from pool 0 requested work restart
 [2013-04-06 00:37:53] Accepted be027246 Diff 75/64 GPU 0
 [2013-04-06 00:37:54] Accepted b681e587 Diff 70/64 GPU 1
 [2013-04-06 00:37:55] Accepted ee775292 Diff 86/64 GPU 1
 [2013-04-06 00:37:57] Accepted 911aa621 Diff 299/64 GPU 1
 [2013-04-06 00:37:58] Accepted a01c946b Diff 812/64 GPU 1
 [2013-04-06 00:38:03] Accepted d11bdf8f Diff 95/64 GPU 1
 [2013-04-06 00:38:14] Accepted e1669a74 Diff 781/64 GPU 0
 [2013-04-06 00:38:14] Accepted be408c19 Diff 72/64 GPU 1
 [2013-04-06 00:38:20] Accepted 250a616e Diff 453K/64 GPU 0
 [2013-04-06 00:38:25] Accepted fa37daae Diff 142/64 GPU 1
 [2013-04-06 00:38:26] Accepted 9556063d Diff 142/64 GPU 1
 [2013-04-06 00:38:27] Accepted 8aac9b24 Diff 87/64 GPU 0
 [2013-04-06 00:38:36] Accepted 80bc3b37 Diff 3.74K/64 GPU 0
 [2013-04-06 00:38:37] Accepted 35e5add5 Diff 549/64 GPU 1
 [2013-04-06 00:39:03] Accepted 454f7a49 Diff 92/64 GPU 0
 [2013-04-06 00:39:04] Accepted 654f7802 Diff 99/64 GPU 0
 [2013-04-06 00:39:07] Stratum from pool 0 detected new block
 [2013-04-06 00:39:15] Accepted 7bb3ae78 Diff 102/64 GPU 1
 [2013-04-06 00:39:22] Accepted ad3707be Diff 65/64 GPU 0
 [2013-04-06 00:39:26] Accepted d3d6082a Diff 231/64 GPU 0
 [2013-04-06 00:39:26] Accepted e36fe328 Diff 79/64 GPU 1
 [2013-04-06 00:39:32] Accepted 59b3288a Diff 71/64 GPU 0
 [2013-04-06 00:39:32] Accepted fb54a843 Diff 173/64 GPU 1
 [2013-04-06 00:39:34] Accepted 825d8fea Diff 101/64 GPU 0
 [2013-04-06 00:40:04] Accepted baf8d570 Diff 95/64 GPU 1
 [2013-04-06 00:40:05] Accepted 07ae8df0 Diff 112/64 GPU 1
 [2013-04-06 00:40:30] Accepted 90cfacba Diff 117/64 GPU 0

[meebs]

So has anyone here ever proved that LTC mining uses more electricity?

I find that hard to believe.. I dont know about anyone else.. but here is how things tend to be with my 7950's (gigabyte triple fan)

on a given day.. mining BTC at 600mhash/s (1150mhz) the card will say run at oh.. 75C...

Switch to LTC at 630 khash/s (1050/1400) and the card now runs at 67C (exact same machine, fan speed, environmental temps etc)

is there something that I am missing that would cause power to increase for some other reason?

Also.. using the EXACT same settings (copy pasted, so i know they are identical)... 2 of my cards (same machine) get 570 and the other 3 get 630.. *unamused*.. this is in CGminer by the way.

[creativex]

Sure it's easy to prove. Just look at the killawatt. BTC mining is essentially unaffected by memory speed so make that as low as possible and the power consumption of your GPUs drops.

Too much is made of the difference quite often IMO, but then my electricity rates are pretty reasonable.

[gyverlb]

So has anyone here ever proved that LTC mining uses more electricity?

I find that hard to believe.. I dont know about anyone else.. but here is how things tend to be with my 7950's (gigabyte triple fan)

on a given day.. mining BTC at 600mhash/s (1150mhz) the card will say run at oh.. 75C...

Switch to LTC at 630 khash/s (1050/1400) and the card now runs at 67C (exact same machine, fan speed, environmental temps etc)

is there something that I am missing that would cause power to increase for some other reason?

Yes RAM power usage. You can underclock your RAM with BTC (although not much on 6xxx and 7xxx hardware) and it's almost not used instead of LTC which is very RAM intensive

[tacotime]

hmmmm

I have a smalish problem:
- i use ubuntu 12.10+ati 13.3+cgminer 2.11.3 video drivers on 2 rigs

first rig has 6990 it only runs at speeds 610/1250, if I try to use any higher gpu clock than 610 (say 700) system will simply shutdown in 10 min or so. some of you will say that I use shitty 850W PSU from "ace"  and you might be right. when it runs on 610/1250 it seems stable, no hw errors and an average of 650 kH/s, which is not much but seems fine. Here is the script I use for mining:

Code:

export DISPLAY=:0
export GPU_USE_SYNC_OBJECTS=1
export GPU_MAX_ALLOC_PERCENT=100
aticonfig --od-setclocks=610,1250 --adapter=all
aticonfig --od-getclocks --adapter=all
./cgminer --scrypt --api-listen --api-network -c cgminer.conf --thread-concurrency 8192 --lookup-gap 2 --worksize 256 -I 18 -g 1 -o stratum+tcp://notroll.in:3333 -u x -p y

second rig has 2x5830+5850+A4 cpu with some 6xxx type thing in it. it runs stable (for months) mining btc at 600/667 (a4) 920/600 (5830) and 910/600 (5850) with hash rates of: 28 (a4) 282 (5830) and 365 (5850). no world records there, but I'm happy mining with a stable rig that needs virtually no maintanance. LTC is a whole different matter. What ever I tried to do I would only mine for 10-15 minutes at most. When it did it was stable with no HW errors, temp was 75°C, speeds were: 25 (a4), 220 (5830) and 300 (5850) it drew no more power out of the socket as it did when it was mining btc. but after 10 minutes or so it would shut down. this rig however has high quality 1000W Corsair PSU. I Cannot get it to work stable! Here is the script

Code:

export DISPLAY=:0
export GPU_USE_SYNC_OBJECTS=1
export GPU_MAX_ALLOC_PERCENT=100
aticonfig --od-setclocks=615,1200 --adapter=1
aticonfig --od-setclocks=615,1200 --adapter=2
aticonfig --od-setclocks=615,1125 --adapter=3
aticonfig --od-getclocks --adapter=all
./cgminer --scrypt --api-listen --api-network -c cgminer.conf --thread-concurrency 3072,5760,5760,5760 --lookup-gap 2 --worksize 256 -I 18 -g 1 -o stratum+tcp://notroll.in:3333 -u x -p y

it is interesting that in both cases computer would not start at the press of the power button. I had to disconnect power and reaply it and then restart the pc. this would indicate that indeed some sort of protection mechanism inside PSU triggered the shutdown. but what?

Has anyone seen anything like this??? Where are these shut-downs coming from?
Could this be a cpu scaling (cpu changing freq) related thing?
Should I test the rig card by card?
How come there is no problem mining btc even at 920/1000 on all cards?
Should I reduce GPU_MAX_ALLOC_PERCENT? How about GPU_USE_SYNC_OBJECTS?

p.s. third rig i tried this on uses windows, 2x5770 and 450W!!!! Corsair PSU and is able to mine ltc without a single shutdown!!!

Corsair HX1000 is 2x 12v rails with 480W on each rail instead of a single rail.  Try moving around your power cables, as you probably are drawing more than 480W from one of the rails.

[qbits]

Corsair HX1000 is 2x 12v rails with 480W on each rail instead of a single rail.  Try moving around your power cables, as you probably are drawing more than 480W from one of the rails.

i did not realize that. is there a simple way to test that? when you say "move your cables" do you mean:
- move cables between 4 available blue connectors thai pcie power cables attach to?
- move some cables from blue pcie connectors to general black ones and use converters to turn 4 pin peripherial connectors to pcie power connector?

corsair manual for this psu http://www.corsair.com/en/media/cms/manual/HX1000_English.pdf does not say which connector on psu attaches to which rail.

[FullLife]

So after reading through this whole thead, I thought I'd contribute my setup and config.

Rig:
i3-3225
Gigabyte G1.Sniper 3 mobo (in hindsight, I should have bought a different board, this one is overkill)
8GB DDR3
(4) Gigabyte 7970s link (in hindsight, I should have bought either vanilla 7970s or 7950s, these are not worth the extra $$)
1200W Coolermaster Silent Pro Gold ps (I can't say enough about this power supply, it's excellent)
250GB Samsung 840 SSD
(2) WD hard drives (500GB, 2TB)
Windows 7 Ultimate SP1 64-bit

Setup:
So I had all 4 cards running in my case, but with them being all crammed together, it was way too hot.  So I pulled everything out and sat it on a rack.  I have 2 of the cards plugged into the board, a 3rd card plugged in via 16x riser cable, but couldn't get the 4th card to work with another 16x riser cable. So, I ordered a 1x to 16x riser cable and hopefully this will solve my problem.

In the meantime, I'm running 3 cards and getting pretty nice hashrates.  These are my settings and results using cgminer 2.11.3 and AMD Catalyst 13.1 drivers:

--intensity 13 -g 2  715/710 = ~1500Kh/s
--intensity 13 -g 2  925/1000 = ~1950Kh/s
--intensity 13 -g 2  1000/1206 = ~2100Kh/s
--intensity 13 -g 2  1022/1244 = ~2150Kh/s

Other settings:
--thread-concurrency 8192 -w 256

I've literally spent most of a entire week playing around with settings and trying to find the sweet spot and I feel like I'm pretty much there.  As you can see, I actually underclocked my cards to get these hashrates.  Stock clocks are 1100 core/1500 memory.  Unfortunately the default voltage is about 1.2 volts on my cards and its locked (@#&($$^(@&$!!), otherwise, I'd be able to save 100-150W easily by undervolting.



I'm seriously thinking about getting rid of these cards and getting 7950s or 7970s that don't have the voltage locked.  Having the voltage locked is such a huge drawback, I'm just wasting too much power right now.

You sir kick ASS!!!!! I tried forever to get my damn gigabyte card over 700 with no luck.  I used your settings and poof, 720 easy!

Thanks so much!

Glad I could help .  Btw, switching from that 16x-->16x riser to a 1x-->16x riser fixed that problem I was having with the 4th card.  Just in case anyone runs into the same problem.

[dsky]

Don`t know if this has been asked before:

I use HD5770, 6770 and 7770 cards in my computers - every computer with 1 card.
I get ~220 khash/s from each card with cgminer.
But the system becomes so laggy that you could not work with it.

Lowering the intensity will lower the khash to 20...

Isn`t it possible to mine ltc with a machine where you can work with while its mining?

[peacefulmind]

Taco can you take a look at my problem quoted below?

Okay - need some help on this anomaly.

First rig - Sapphire 7970 standard edition with Vapor X, used settings suggested in Taco's thread - out of the box 650k/H per second.

Second rig - two (2x) Sapphire 7970 "GHZ Editions", used exact same settings in Taco's thread - out of the box - only about 350 k/H per second.

I have spent the last week in "newbie purgatory" so have been unable to discuss this issue.  Finally on the second rig, I was able to get 500k/H per second/card by putting aggression back to 13, and overclocking the memory to 1680Mhz.

Can anyone tell me WHY do two of the same brand/manufacturer, same model, have such huge variance in performance?

Why would the two "Ghz editions" VASTLY under-perform the standard one?

Does anyone have any suggestions experience with a Sapphire "GHZ Edition" that could explain the variance?  The standard one was plug and play @ 650k.   The better ones have taken a week of tweaking to even reach 500k/H, and NONE of the recommended settings worked.  I know it is not bad cards because they both perform the same - what is different about the "GHZ editions" that is hurting them so much, and made them so finicky?

First rig is Q6600 65nm, 4GB DDR2, PCI-E 2.0, Win7 Pro 64, Sapphire 7970 - 650kH/s

Second rig is 22nm Quad i5 Ivy Bridge, 8GB DDR3, PCI-E 3.0, Win7 Pro 64, 2 x Sapphire 7970 "GHZ Editions" - 990kH/s (both cards)

I followed all instructions in this thread.  The second rig would only do about 300 per card until I overclocked the RAM on both cards to 1700Mhz and dialed back aggression to 13.  Any variance from that setting and I get LESS than 500kH per card.

I would REALLY like to get the potential out of these supposedly "better" Sapphire 7970 Ghz editions.

[Cablez]

i have 4 7970s and for the life of me can't seem to get anywhere near what other say it can get...  the highest I've gotten was 500kh/board if I'm lucky.

i've been using these settings to get 500 (cgminer):

--shaders 2048
-I 13 #any higher and it drops like a rock, any lower, slowly drops until around 10 where it only does around 50-70k
--thread-concurrency 8192 # i've been messing with this and anything but 8192 seems to drop it by about 5kh/s
--gpu-memclock 1700
--gpu-engine 970

I've messed with the clock speeds over and over but on my memclock, anything over 1700 seems to lock up after about 10-15 minutes, so I started messing with the engine to bring it into the .57-.6 ratio range and this is how I finally got it to go to 500kh.

i'm at a loss as to what to try next, if i bump the clocks up any more it will be unstable it seems.

I haven't messed with the voltage, but i wouldn't know what to do with the without destroying the cards.  every guide i've looked at mentions that the voltage shouldn't have to be adjusted to get around 700k

anyone have any ideas what to try next?

You are not alone with this. It seems there are certain 7970s out there that do not follow the trends that others have laid out.  Can you give some details of your setup, OS, card brand, asic quality, driver and sdk version etc. Maybe we can find some common ground to figure out why they under perform.

[qiuness]

Hello.
I was wondering which hashrate is the correct one when considering:

1. cgminer says: 402/398.4
2. notroll says 461-475

Which one is it then?
Running stock Gigabyte 5970

[pox67]

You are not alone with this. It seems there are certain 7970s out there that do not follow the trends that others have laid out.  Can you give some details of your setup, OS, card brand, asic quality, driver and sdk version etc. Maybe we can find some common ground to figure out why they under perform.

Yes I was surprised about this as well.

I had 2 x Sapphire OC 7970's that do 650KHs at 950/1425. They sit at under 70c as well and at 1.025volts which is great as it saves power.

Next I got 2 MSI 7970 OC with the reference cooler. I went straight to 950/1425 thinking that would work but they got under 400KHs, WTF!
After playing around the best I could get was about 612KHs at 909/1575 and 1.030 volts. I am thinking the MSI cards might need their RAM clocked even higher but Afterburner stops at 1575MHZ.

Using CGMiner with the following start line:
cgminer.exe --scrypt --config my.conf --shaders 2048 --thread-concurrency 8192 -I 13 -g 2 -w 256

Has anyone tried CGMiner with higher thread concurrencies for 7970's?

[Cablez]

I have tried it with thread concurrencies up to 22400 and found no better result then ~610kh/s.  My guess is it is a bum card.  I am thinking of flashing the bios to a ghz edition and see if that changes it.

[Tejsei3]

Looking for help to improve kH/s/W.

Here's my setup:
MSI Z68MA-ED55 (B3) motherboard
Intel G530 processor
4GB DDR3 1066 RAM
7950 Sapphire 100352-2L (purchased March 2013, voltage unlocked!)
16X-16X riser cable
Corsair TX750W power supply
AMD 13.1 drivers
AMD SDK 2.8
cgminer 2.11.3 (-I 20, --tc 22400 -w 256 -g 1 -l 1)
Sapphire Trixx

Best settings: 925/1250/.962V 560kH/s 3.29 kH/s/W.

Given high electricity cost my goal is to maximize kH/s/W.

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 (850-1250), memclock (1250-1750), and gpu v (0.900-1.250). I've had the rig up to 740 kH/s stable (1225/1750/1.250V) but at 345W or 2.14 kH/s/W the efficiency just isn't there (would be great for a free power situation!).

Looking for advice from the experts on what to do next to improve my efficiency.

Thanks,

Tejsei