Vertcoin Settings/Configs (How To Mine VertCoin)

[sergi0]

Hi Folks !

I have a strange problem. I have 5 rigs of 4 sapphires 290x TRIX. Mining SCRYPT I get around 3.5Mhs per rig. I have 8gb of ram on each rig, 4 cards using USB risers, 60% fan and the temp never get more than 80c, and often in the 70C range. The rig are stable. Sometime ATI drivers crash, but autoreboot and no problem.  I wanted to use one of the rig to test Vertcoin... Here my problem begin.

I use Vertminer 0.5.3 and the following parameters :

Code:

vertminer.exe --scrypt-vert -o stratum+tcp://vtc.crypto-pool.fr:4001 -u xx -p xx

The goal was to see the generated parameters and use them. I did that for cgminer and it works well.

Here I only get 2 of the 4 cards mining and at a rate of 500hs. Not kilohash, just hash... I reboot 20 times in order to make multiple tests, but I never manage to have 4 cards mining, and worst, the 2 that are mining are not always the same !!! Plus, if I run cgminer on anything else, everything is working fine...

What is hapenning ? Is it a nasty plot to not allow me to mine Vertcoin ?

[1713442436]

1713442436

[1713442436]

1713442436

[1713442436]

1713442436

[1713442436]

1713442436

[1713442436]

1713442436

[giantkin]

vertminer.exe -o stratum+tcp://stratum.ecurie.io:3333 -u giantkin.c1 -p xo -d 1 --lookup-gap 2 --thread-concurrency 8193 -g 2 -I 13 -w 256 --auto-fan --gpu-fan 30-95 --temp-cutoff 90 --temp-overheat 85 --temp-target 80 --gpu-memclock 1500 --gpu-engine 1020 --gpu-powertune -20 --expiry 1 --scan-time 1 --queue 0
pause

3 280's

The above is working getting roughly 330-350 khs    usually.  however they drop to 100 each, then cycle back up. (temp hits 84 at the 350 range)   only 75ish for the 100 range.

Same setup but all one command the system would freeze after 1hr (to 4 hrs)  and need a hard reset.

this setup running an hr so far.


Why the freezing with all in one.

Why the drop in hash.   Im ok with variable speed etc, but it drops so far, seems kinda odd.


(my device 2 runs hotter than the other 2)
device 0 sometimes needs 2x to start up clean.   (this is the monitor device (tho, no monitor atm))

oh i am running 0.5.2   i noticed, so my next step is to update  the version.   

System 
Win 8.1
4gb ram
2 psu's  550, 850 (2 cards)     (maybe i should put all three on the 850)

Thanks
g

[Cal_42]

Settings for Sapphire reference 290 and Sapphire 290 Tri-X. Any advice would be appreciated.

Reference:

"device" : "1",
"intensity" : "13",
"worksize" : "512",
"lookup-gap" : "2",
"gpu-engine" : "947",
"gpu-memclock" : "1250",
"shaders" : "2560",
"thread-concurrency" : "10176",
"expiry" : "30",
"gpu-threads" : "1",
"scan-time" : "1"

Tri-X:

"device" : "2",
"vectors" : "1",
"intensity" : "13",
"worksize" : "512",
"lookup-gap" : "2",
"gpu-engine" : "1000",
"gpu-memclock" : "1300",
"shaders" : "2560",
"thread-concurrency" : "10176",
"expiry" : "30",
"gpu-threads" : "1",
"scan-time" : "1"

Bat file:

setx GPU_MAX_ALLOC_PERCENT 200
setx GPU_USE_SYNC_OBJECTS 1
COLOR 0A
vertminer.exe


I'm gettiing round 320kh and a WU of 290 on each card. I got the TC by multiplying shaders and subtracting 64. The reference runs at 74c and the Tri-X at 62c. High TC or intensity won't load the kernel and disables the cards. Overclocking seems to crash and I'm using vertminer0.5.3. No HW errors yet and hardly any rejects. Not sure where to go from here. I have 8GB RAM installed.

[NuSalvo]

Here's what I've gotten so far on a test rig, it's just 3 random cards (Sapphire 270x, Vapor 270x, AMD 7950) which usually does ~1.3 MH/s.

https://i.imgur.com/pqM9rJa.jpg

I can probably get a little more, but no HW errors makes me happy!  I'm glad I found this thread, I had WAAAYY too high TC's at first.  This is with 8193 TC on 2 cards and 22400 on the 7950. Seems like the 2 x 270x's are underperforming just a bit (they were ~470 e, I'd be happy with 200-210 vert mining)  Does anyone have recommendations for my setup?  Here's my config file

"intensity" : "13,13,20",
"vectors" : "1,1,1",
"worksize" : "256,256,256",
"kernel" : "scrypt,scrypt,scrypt",
"lookup-gap" : "2,2,2",
"thread-concurrency" : "8193,8193,22400",
"shaders" : "0,0,1596",
"gpu-engine" : "1150-1150,1000-1000,1100-1100",
"gpu-fan" : "0-85,0-85,0-85",
"gpu-memclock" : "1500,1500,1500",
"gpu-memdiff" : "0,0,0",
"gpu-powertune" : "10,10,10",
"gpu-vddc" : "0.000,0.000,0.000",
"temp-cutoff" : "90,90,90",
"temp-overheat" : "80,80,80",
"temp-target" : "70,70,70",
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "30",
"gpu-dyninterval" : "7",
"gpu-platform" : "0",
"gpu-threads" : "1",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "5",
"scrypt-vert" : true,
"temp-hysteresis" : "3",
"shares" : "0",
"kernel-path" : "/usr/local/bin",
"device" : "0-2"

My temp seems to be a bit lower mining this coin, is that normal?  Only like 5 C difference.

[casinocoin]

I see some people have their intensity above 12, I cannot get my rig to go above 12 without massive hw errors and no accepted.

Running 2 different rigs with 3 msi 280x ea
all cards can only hash around 200-300 where as with other scrypt i was up to 700-750 per card
I know it should be 1/2 of the total hashrate but mine is falling quite below that.

Any suggestions.
Both rigs have 16gb ram.
Running windows vertminer with the simple provided conf and some additional settings
Tried thread concur 8192 and 11200(works better and was previous setting)

[hero18688]

Is there someone runing 2*290 under win7 with 4gb ram.I can't go across threads larger than 14000.But with 8gb ram even 27400 is ok.

[ilic]

i'm mining Darkcoins at the moment but am thinking of switching to Vert for a while.  Can someone tell me the power consumption for mining vert compared to the usual scrypt coins?  DRK uses about 30% less power for me, i'm hoping Vert uses alot less than scrypt too.

[hero18688]

i'm mining Darkcoins at the moment but am thinking of switching to Vert for a while.  Can someone tell me the power consumption for mining vert compared to the usual scrypt coins?  DRK uses about 30% less power for me, i'm hoping Vert uses alot less than scrypt too.

Almost the same,vert is not power effective.

[gg234]

I am having 3 asus r9 280x cards and with the following settings one card is getting 16K/hash can you please let me know how to get proper hash on all 3 cards

vertminer.exe --scrypt-vert -o stratum+tcp://gns.pool.mn:3333 -u xxx -p xxxx -w 256,256,256 -I 13,13,13 -g 2 --lookup-gap 2,2,2 --gpu-powertune -20,-20,-20 --temp-overheat 85,85,85 --temp-cutoff 90,90,90 --gpu-memclock 1500,1500,1500 --gpu-engine 1000,1000,1000 --thread-concurrency 8191,8191,8191

Thanks

[casinocoin]

I am having 3 asus r9 280x cards and with the following settings one card is getting 16K/hash can you please let me know how to get proper hash on all 3 cards

vertminer.exe --scrypt-vert -o stratum+tcp://gns.pool.mn:3333 -u xxx -p xxxx -w 256,256,256 -I 13,13,13 -g 2 --lookup-gap 2,2,2 --gpu-powertune -20,-20,-20 --temp-overheat 85,85,85 --temp-cutoff 90,90,90 --gpu-memclock 1500,1500,1500 --gpu-engine 1000,1000,1000 --thread-concurrency 8191,8191,8191

Thanks

try thread concur 8192 or 11200, try changing I to 12 as well removing the memclock and engine settings(if you have afterburner etc running.)

[gigajosh]

Please post your settings below and any tips to help new comers! Thanks

I'm using vertminer to mine Spaincoin (SPA) which has recently been changed from a Scrypt to a Scrypt-N coin.
I'm running vertminer under CGWatcher. I simply renamed vertminer.exe to cgminer.exe so that CGWatcher's Profile Manager recognises the executable.

Specs:
- Windows 7 Home Premium (64-bit)
- 3x Sapphire R9 290X
- AMD driver: 13.251 (aka: 13.11 beta 9.5 from Guru3D)
- AMD SDK v2.9
- vertminer-0.5.3 (download from this thread) (renamed to cgminer.exe as mentioned above so that it works under CGWatcher)
- I do not use a config file, I just set everything with command line arguments as follows:

--scrypt-vert --worksize 256 --thread-concurrency 14334 --auto-fan --auto-gpu --gpu-memclock 1500 --gpu-powertune 20 -I 13 --lookup-gap 2 -o {url}:{port} -u {username}.{worker} -p {password}

Performance:
With intensity set to 20 I was getting 475Kh/s per card, but then I realised I was getting loads of hardware errors and had a rubbish WU.
So I brought it right down to intensity 13, the highest I could go without getting hardware errors. At this rate unfortunately there are more profitable coins to mine.

NOTES:
1) TWEAK YOUR TC as follows: If you have mined scrypt coins successfully before, you need to half the TC you used for scrypt. If you still get a crash, use half your scrypt TC but subtract another 256.
E.g.
- If your original scrypt TC = 28008
- half this value = 14004 (maybe this crashes)
- if half didn't work, then try = 13748
When you get a successful TC:
- increase it by 32 and try again
- if the 32 increase was successful, keep increasing by 32 and trying again until you get a crash.
- then reduce by 2 and try again and repeat this until it works again
- now you will have optimal TC.

[WARLOCKZA]

Hey Guys,

I realise that registrations are close but it's worth an ask, I have 17Mh/s I's love to point to your pool, would it be possible for an invite

Thanks

[pifua]

With the zoo of 7950 cards like; MSI Twin Frozr III, Sapphire Dual-X and XFX with custom BIOS

Code:

Adapter  0    (BN=01, DN=00, PCIID=679A1002, SSID=E207174B)
    Asic Family        :  Tahiti         
    Flash Type         :  M25P10/c    (128 KB)
    Product Name       :  Tahiti B0 PRO C38101 GDDR5 3GB 500e/150m
    Bios Config File   :  207P0C0A.O47   
    Bios P/N           :  113-1E207200SA-T4
    Bios Version       :  015.018.000.000.000000
    Bios Date          :  03/30/12 06:02
    ROM Image Type     :  Legacy Image
    ROM Image Details  : 
        Image[0]: Size(65536 Bytes), Type(Legacy Image)
Adapter  1    (BN=04, DN=00, PCIID=679A1002, SSID=32211682)
    Asic Family        :  Tahiti         
    Flash Type         :  M25P20/c    (256 KB)
    Product Name       :  Tahiti B0 CR PRO C38602 GDDR5 3GB 500e/150m
    Bios Config File   :  795TXL13.BIN   
    Bios P/N           :  113-795ATDL&TNLV1
    Bios Version       :  015.024.000.001.000000
    Bios Date          :  05/19/13 21:08
    ROM Image Type     :  Legacy Image
    ROM Image Details  : 
        Image[0]: Size(65536 Bytes), Type(Legacy Image)
Adapter  2    (BN=05, DN=00, PCIID=679A1002, SSID=254C1458)
    Asic Family        :  Tahiti         
    Flash Type         :  W25X20      (256 KB)
    Product Name       :  GV-R795WF3-3GD/F43
    Bios Config File   :  R795W3GD.F43   
    Bios P/N           :  113-C3810100-X03
    Bios Version       :  015.026.000.000.000907
    Bios Date          :  10/24/12 08:55
    ROM Image Type     :  Hybrid Images
    ROM Image Details  : 
        Image[0]: Size(65536 Bytes), Type(Legacy Image)
        Image[1]: Size(65024 Bytes), Type(EFI Image)

I heave the following under BAMT (linux):

Code:

vertminer version 0.5.3 - Started: [2014-03-28 17:38:58]
--------------------------------------------------------------------------------
 (5s):879.5K (avg):877.4Kh/s | A:63710  R:0  HW:0  WU:807.0/m
 ST: 2  SS: 0  NB: 22  LW: 15258  GF: 0  RF: 0
 Connected to vtc.coinminerpool.com diff 135 with stratum as user vtc.w1
 Block: cb306335...  Diff:245  Started: [18:48:46]  Best share: 72.3K
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0: 73.0C 2145RPM  | 278.1K/277.2Kh/s | A:19138 R:0 HW:0 WU: 238.4/m I:13
 GPU 1: 76.0C 2621RPM  | 310.3K/309.3Kh/s | A:23277 R:0 HW:0 WU: 284.6/m I:13
 GPU 2: 73.0C 2170RPM  | 291.8K/291.3Kh/s | A:21295 R:0 HW:0 WU: 284.0/m I:13
--------------------------------------------------------------------------------

my config:

Code:

{
"pools" : [
        {
                "url" : "stratum+tcp://vtc.coinminerpool.com:3336",
                "user" : "vtc.w2",
                "pass" : "password"
        },
        {
                "url" : "stratum+tcp://vtc.coinminerpool.com:3306",
                "user" : "vtc.w3",
                "pass" : "password"
        }
]
,
"intensity" : "13",
"vectors" : "1",
"worksize" : "256",
"lookup-gap" : "2",
"thread-concurrency" : "8193",
"shaders" : "1792",
"gpu-engine" : "1000",
"gpu-fan" : "0-75",
"gpu-memclock" : "1490",
"gpu-memdiff" : "0",
"gpu-powertune" : "20",
"gpu-vddc" : "0.000",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "80",
"api-mcast-port" : "4028",
"api-port" : "4028",
"auto-fan" : true,
"expiry" : "120",
"gpu-dyninterval" : "7",
"gpu-platform" : "0",
"gpu-threads" : "1",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "1",
"scrypt-vert" : true,
"temp-hysteresis" : "3",
"shares" : "0",
"kernel-path" : "/usr/local/bin"
}

Give it a try

[kof91]

I've been poking around the settings, you do not need 4GB ram to run vertminer getting insane amount of hw error. I tried on my gaming rig with 16gb ram still getting hardware error with incorrect setting. imo the ram dun matter that much but you will still need a min of 2gb for scyrpt mining.

1. check if you are mining from a p2pool. most likely your card is trying to mine blocks with too low difficulty.

My worker is getting a lot of hardware errors

Because of the short blocktimes on the p2pool sharechain, your workers need to receive much lower difficulty work than they would in a traditional pool. They must finish their work and submit it before it goes stale - on the main blockchain, this would happen every time a block was generated (every 2.5 minutes on average), wheras on p2pool a new block on the sharechain happens every 10 seconds on average. Very low difficulty work often causes hardware errors. You can prevent this by telling p2pool the minimum difficulty your worker requires. You do this by appending a string to the end of your username (your Vertcoin address) in your miners startup command. The formula for calculating what to append is:

Vertcoin Address + 0.00000116 * Hash Rate in KH/s
So, if your hashrate was 350KH/s, you would calculate:
0.00000116 * 350 = 0.000406
and if your Vertcoin address was ViPBVm4sbXT38h9J23GkAWfNKiuPwVUYQX then you would use
ViPBVm4sbXT38h9J23GkAWfNKiuPwVUYQX+0.000406
as your p2pool username.

You don't want to set this value too high, or you will increase your stale shares - it sets the minimum difficulty for the shares you receive, and if you set it too high you will receive a large percentage of shares that are too difficulty for your worker to complete before the next block is generated, so the work will go stale. Setting this parameter is a balancing act between submitting work fast enough to avoid stales, and obtaining work that is difficult enough that it doesn't produce HW errors, or consume unnecessary bandwidth (if you have a very low minimum difficulty, your worker has to ask the pool for work too often, consuming extra bandwidth unnecessarily).

Here are some suggested values for various hashrates:
KHash/s     Difficulty

50          +0.00005821
100         +0.00011641
250         +0.00029103
500         +0.00058207
750         +0.00087310
1000        +0.00116414

please refer to the following guide for details
https://vtcpool.co.uk/p2pool-guide.html
2. to find the right setting balance btw concurrency thread and intensity. thread-concurrency works in a 64 block
for example if you put 8256 you will mine at concurrency block of 8256-64 = 8192
in order to mine at thread-concurrency  of 8256 set your thread-concurrency at anything  8256 + 0~64 it will automatically round down to the next nearest 64 block.
you can check the file generated to see which thread-concurrency it is running at by checking your folder for "scrypt140121Tahitiglg2tc8256w256l4.bin"
3. imo thread-concurrency uses ram on you gpu, the higher you set , the more gpu ram will be used to some point miner will tell you that you do not have enough ram
4. higher thread-concurrency doesn't seems to affect much on the overall kh/s but intensity does. you will need to find a thread-concurrency that allows you to go for highest intensity without hardware error. unfortunately at this point of time, from what I read, same card may not work with same thread-concurrency / intensity combination. you will need to manually poke the right value for your card.

my config setting is
"intensity" : "18",
"worksize" : "256",
"lookup-gap" : "2",
"gpu-threads" : "1",
"expiry" : "30",
"scan-time" : "5",
"thread-concurrency" : "6000",
"temp-cutoff" : "88"

5850 sapphire running at 830/875mhz oc via trixx, higher ram/gpu speed will affect the kh/s but to some point the increase in ram/gpu speed will not benefit much as compared to heat increase. find the balance to get the most out with least heat for a longer lasting hardware.

lastly a screenshot of my 2gb pc running without hardware error. happy tuning and share your finding here.
https://i.imgur.com/hi2a8Hk.png

[bitcoinsguru]

Can anybody help me to mine Vertcoin with CPU, which miner should I use..

[Thirtybird]

You've got the general concepts correct - let me fill in the details for you.

Thread-Concurrency : for scrypt, this ONLY affects the size of the OpenCL buffer on the GPU.  It creates a buffer in size = TC*128/(1024/Lookup-Gap) (roughly, for LG=1, 2, 4 & 8 - there's other considerations for the other LG's).  TC of 8192 and Lookup Gap of 2 = 512 MB buffer - this is the most common setting I see people using for regular scrypt coins.

Lookup-Gap : as this setting goes up, this increases the effective size of the memory available to openCL hashing threads at the cost of some GPU cycles.  A LG of 2 doubles the buffer size, a LG of 4 quadruples it.  In-between settings give in-between values, etc.  The same setting of TC=8192 and LG=2 gives an effective memory size of 1024MB

Hash Size - based on N.  Vertcoin is currently at N=11 and requires 256KB of memory per hash.

AMD GPU shaders (this one is based on observation) : AMD cards running mining software can get 4 * the number of shaders executing simultaneously.  Each one needs its own memory allocation.

Combine all of this.  

Take the number of shaders on your card, multiply by 4.  Then multiply by 256 KB.  This is how large your effective memory needs to be.  With the TC 8192 and LG 2, the effective memory of 1024 MB means that if your card has 1024 shaders, you're fine as this is the exact amount needed!  If you have any more shaders (say even an R9 270 with 1280 shaders), you need to allocate a higher buffer size or you will get HW errors.  With 1280 Shaders, you need effective memory of 1280 MB which can be accomplished by a TC of 10240/LG 2.  

Taking this to the extreme - the 290X has 2816 shaders, to keep the LG at 2, you need a TC of 22528 (1,408 MB, effective size of 2816).  At this level, you may have difficulty allocating the whole buffer size in one thread if your system ram is only 2GB - The buffer gets allocated in system ram before being passed to the OpenCL kernel for execution.

If you think you've got the hang of it, try tuning your card to mine YACoin, where each hash requires 4MB of memory

[SpeedDemon13]

Seems like the most important setting in Scrypt-N is the TC setting and running the core clock at default or a little lower. If some how someone can get Scrypt-N to lower power consumption and keep the same hash rate, that would be a goldmine....

[kof91]

You've got the general concepts correct - let me fill in the details for you.

Thread-Concurrency : for scrypt, this ONLY affects the size of the OpenCL buffer on the GPU.  It creates a buffer in size = TC*128/(1024/Lookup-Gap) (roughly, for LG=1, 2, 4 & 8 - there's other considerations for the other LG's).  TC of 8192 and Lookup Gap of 2 = 512 MB buffer - this is the most common setting I see people using for regular scrypt coins.

Lookup-Gap : as this setting goes up, this increases the effective size of the memory available to openCL hashing threads at the cost of some GPU cycles.  A LG of 2 doubles the buffer size, a LG of 4 quadruples it.  In-between settings give in-between values, etc.  The same setting of TC=8192 and LG=2 gives an effective memory size of 1024MB

Hash Size - based on N.  Vertcoin is currently at N=11 and requires 256KB of memory per hash.

Hash Size is variable according to difficulty?

AMD GPU shaders (this one is based on observation) : AMD cards running mining software can get 4 * the number of shaders executing simultaneously.  Each one needs its own memory allocation.

Combine all of this.  

Take the number of shaders on your card, multiply by 4.  Then multiply by 256 KB.  This is how large your effective memory needs to be.  With the TC 8192 and LG 2, the effective memory of 1024 MB means that if your card has 1024 shaders, you're fine as this is the exact amount needed!  If you have any more shaders (say even an R9 270 with 1280 shaders), you need to allocate a higher buffer size or you will get HW errors.  With 1280 Shaders, you need effective memory of 1280 MB which can be accomplished by a TC of 10240/LG 2.  

Taking this to the extreme - the 290X has 2816 shaders, to keep the LG at 2, you need a TC of 22528 (1,408 MB, effective size of 2816).  At this level, you may have difficulty allocating the whole buffer size in one thread if your system ram is only 2GB - The buffer gets allocated in system ram before being passed to the OpenCL kernel for execution.

I am looking at my memory usage, actually it does not used up to maximum allocation as calculated fluctuating btw 700~1000mb. 5850 has 1440 shaders (1440MB). this value might be the theoretical max required.
probably can lower the tc or limit number of shader to enable the card to run with lower system memory but still may not achieve stability with hw error which is another problem all together. strangely I seen ppl report with same card but cannot use same setting on both card to achieve stability.

If you think you've got the hang of it, try tuning your card to mine YACoin, where each hash requires 4MB of memory

will try once I have a better rig, currently running on old rig which is still running on ddr2, originally had 4GB ram but one stick failed recently crippling it to 2gb only

Great info, thanks!

[Thirtybird]

Hash Size is variable according to difficulty?

No, the memory required to complete one hash is variable based on N.  N is currently 2048 aka NFactor=10, taking 256KB GPU memory per hash for now.
N is scheduled to change to 11 on  2/25/2016 where N will be 4096 and it will take 512 KB of GPU memory to complete each hash.

I said 11 in my previous post, but that's not correct - Vertcoin/nscrypt kernels N factor is off by 1 from scrypt for the same memory requirement and I got that wrong in my last post.

[Ed4252]

I have a radeon HD6850 and I'm only getting around 11 khash for my gpu while normally I get 170 khash running litecoin with intensity 13.

What could be wrong with my vertcoin settings?