Pages:
Author

Topic: [BBR] Boolberry GPU Miner Discussion - AMD & Nvidia - OpenCL & CUDA (Read 82997 times)

newbie
Activity: 5
Merit: 0
Здpaвcтвyйтe
ecть кapти paзниx пpoизвoдитeлeй(=paзниe биoc) нa Tahiti, пoлнoвec и oбpeзaнниe

пo мoeй cтaтиcтикe
78xx дaют 700-800h ecли нe гнaть cильнo, тoгдa oни в 80-85вaт "вxoдят"
290-e - 1500-1700h@140w
лoгичecки пpeдпoлoжить чтo тaити гдe-тo пocpeдинe 1100-1200
для cтoкa пoлнoвecнoгo 280x зaявляют 970h, чтo пoдтвepждaeт мoe пpeдпoлoжeниe
нo никaк нe пoлyчaeтcя бoлee 900 пoлyчить.. paзвe гнaть лютo a 1000h@160w - глyпocть
в итoгe пpoмoдeннaя 7870 дaeт пoчти тaк жe кaк тoжe пpoмoдeннaя 7970/280x, y кoтopoй кoнвeepoв - в пoлoвинy бoльшe

cиcтeмa - виндa 7/64,
дpoвa - лyчшими пoкaзaлиcь 16.7.3 xoтя пpoбoвaл и нa дpyгиx

пpoбoвaл "пoдpeзaть" пaмять видeляeмyю дo 2Гб - єтo ничeгo нe мeнялo
игpaлcя c дpyгими пapaмeтpaми.
Bпeчaтлeниe чтo peaльнo иcпoльзyeтcя тoлькo 2Гб oпepaтивки

в чeм мoя oшибкa?

PS: интepecнo, чтo клeймop 11.3 для 78xx и 79xx cтaвить пoчти идeнтичнyю hcnt + и 79-e впpaвдy бoльшe вceгo тaк дaют, чтo тoжe нeмнoгo нe лoгичнo, вeдь и кoнвeepoв +30% и пaмяти +50%
newbie
Activity: 5
Merit: 0
Hi, have some Tahiti card
cant take more then 900h.

Pitcairn 1Gb takes for me 710-730
Pitcairn 2Gb takes for me 740-780

Hawaii 4Gb - 1500-1700h

So.. Tahiti - 1100-1200 but only 820-890h

Win 7/64, 16.7.3 drivers.
Testing different drivers  - nothing change
Playing with BIT_PTR64, HEAP_SIZE and others params - nothing change.

Detail - with downgrade clBUFFERSIZE to 2100Mb hashrate practicaly identical.
Where I"m wrong?
member
Activity: 724
Merit: 49
Can you please try version 0.9.6? Made a topic here. And run it from console so we can see whats happening.
newbie
Activity: 52
Merit: 0
Found a bug with random crash when reconnect to pool, updated the link to version 0.9.4 in previous post.
0.9.5, N card does not work properly
0.9.4,is ok
What do you mean "N card"? Give some details please. I have changed only decoding JSON in 0.9.5 version, so it shouldn't do nothing with videocards part of code.
1060 1070 0.9.4 Working properly
0.9.5 Lightning exit
member
Activity: 724
Merit: 49
Found a bug with random crash when reconnect to pool, updated the link to version 0.9.4 in previous post.
0.9.5, N card does not work properly
0.9.4,is ok
What do you mean "N card"? Give some details please. I have changed only decoding JSON in 0.9.5 version, so it shouldn't do nothing with videocards part of code.
newbie
Activity: 52
Merit: 0
Found a bug with random crash when reconnect to pool, updated the link to version 0.9.4 in previous post.
0.9.5, N card does not work properly
0.9.4,is ok
newbie
Activity: 52
Merit: 0
WildRig 0.9.4 beta, new miner for Wild Keccak algo. Download

for those who are new to this miner:
- this miner provides better hashrate on poolside for NVIDIA cards(because of more accepted shares). AMD can get some improvement too, but less significant
- new interface
- API to check miner stats via web
- no limit for amount of GPU's

changes from previous beta:
- more comments for config.json(hope now it won't confuse people with configuration Smiley)
- additional parameter "print-full"(add it to your old config.json when upgrade) for printing hashrate for each videocard
- a bit more safe code
- now miner will redownload scratchpad if something goes too wrong with it
- fix for NVIDIA's, now instead of rare situations with multiple "Low difficulty share" hashrate will drop, but still better to solve more shares instead of reconnect Smiley
- no more rare situation with "Duplicate share" on fast rigs(testing is welcome, please report if it happens)
- changed default strategy(AMD users, please check if it works well)

Testing is welcome! Please report about any issue.

known issues:
- possible crash, see note below
- immediate crash when presss 'H' or using print-full set to true(will crash at first speed print). Very strange problem, occurs only if log-file specified, so be careful and test first. Will check that later
- fast enough rig can get 5 rejects one by one on low difficulty and start reconnecting, decrease intencity in this case if your pool doesn't support stat.diff. In next version I will provide some configuration about amount of rejects before reconnect

NOTE: please, create a bat-file like below and use it instead of direct start of wildrig.exe, I was reported about one crash, so use it for avoiding zero mining
Code:
:loop
wildrig.exe
ping 127.0.0.1 -n 5 > nul
goto loop
How much is the speed of 1060 and 1070 and 580?
member
Activity: 724
Merit: 49
Found a bug with random crash when reconnect to pool, updated the link to version 0.9.4 in previous post.
member
Activity: 724
Merit: 49
sample of configuration for 4 videocards. If you have some troubles - start miner from console and check what it writes. Probably you need to change opencl-platform
Code: (json)
{
    "background": false,
    "colors": true,
    "donate-level": 2,     // default donation level is 2%, you can increase it if you want :D
    "log-file": null,      // set it to "log.txt" if needed
    "print-time": 60,      // print time for hashrate statistics
    "print-full": false,   // print hashrate for each videocard
    "retries": 1000,
    "retry-pause": 10,     // time before reconnect
    "opencl-platform": 1,  // if you have no integrated GPU - change it to 0, or try 2 in some cases with AMD/NVIDIA in one PC
    "scratchpad-url": "http://bbr.miningclub.info/scratchpad.bin",  // where miner can download scratchpad
    "scratchpad-file": "scratchpad.bin", // where to save it(including new file name), use double \\ if you want to set full path(Win only)
    "strategy": 0,         // 0 - standard, 1 - "I'm Feeling Lucky"

    "threads":
[
         // if you have more than one videocard, just copy section below more times and increase index for every new videocard
        {
            "index": 0,
            "intensity": 256,       // good point to start is 256, decent cards can use 1024, try different values. But don't set it high if hashrate is not increasing
            "worksize": 0,          // don't touch it if you don't know what is it :)
            "affine_to_cpu": false,
        },
        {
            "index": 1,
            "intensity": 256,
            "worksize": 0,
            "affine_to_cpu": false,
        },
        {
            "index": 2,
            "intensity": 256,
            "worksize": 0,
            "affine_to_cpu": false,
        },
        {
            "index": 3,
            "intensity": 256,
            "worksize": 0,
            "affine_to_cpu": false,
        },
],

    "pools":
    [
        // Currently no support for failover strategy using more than one pool
        {
            "url": "bbr.miningclub.info:7777", // URL of mining server
            "user": "1Ak5pgtx3bUhNN2aR9Y3mBB5QjWaCER8scFwKEoghMmLSHdzwrC8pzSLpVjdoJj3SZXcN1eV46kGxM7DatYMoYpw5GbG23o",
            "pass": "4706e751c7dd95f90076ca1d7b50babaf6ccbde29303bc0dc7e448ae732b0e03",
        },
    ],

    // same as in XMRig
    "api":
{
        "port": 0,
        "access-token": null,
        "worker-id": null
    }
}
newbie
Activity: 103
Merit: 0
You can configure an example
member
Activity: 724
Merit: 49
new miner WildRig released. Made a topic here.
jr. member
Activity: 176
Merit: 2
Somebody help me, pls. I have a mining rig with 13 GPU Zotac 1060 6Gb. The mining software only support 8 GPU, how can I use another tools to mine with 13 GPU. TKS so much!

Same problem here...

you can try -t 13 for 13 GPU.
newbie
Activity: 15
Merit: 0
Somebody help me, pls. I have a mining rig with 13 GPU Zotac 1060 6Gb. The mining software only support 8 GPU, how can I use another tools to mine with 13 GPU. TKS so much!

Same problem here...
member
Activity: 130
Merit: 11
Hey guys!

I just want to report my, kinda weird, hashrates across my 3 rigs

1st rig
4x 390x ~1800kH/s per card, total ~7200 kH/s
settings -i 22
win7 x64
crimson 16.7.3 drivers


2nd rig
4x fury + 2x 390 ~1100 kH/s per card furys and 390s too
settings -i 22
win10 x64
crimson 16.3.2 drivers

3rd rig
fury x + 3x 390 same results as 2nd rig ~1100 kH's across all cards
settings -i 22
win10 x64
crimson 16.3.2

seems like i get higher hashrates per card on the first rig which has win7 but slightly newer drivers,

what hashrates do you guys get with your 390s and furys? is it the os or the driver problem?
meanwhile I'll try to find a solution and let you know...
newbie
Activity: 1
Merit: 0
Somebody help me, pls. I have a mining rig with 13 GPU Zotac 1060 6Gb. The mining software only support 8 GPU, how can I use another tools to mine with 13 GPU. TKS so much!
newbie
Activity: 89
Merit: 0
This provides a faster hash rate which are available on the the stratum pool.
newbie
Activity: 64
Merit: 0
The transfers on the host devices are sometimes slow then it will not be the best idea.
newbie
Activity: 69
Merit: 0
I do not really feel like dealing with that scratchpad non sense otherwise would have helped.
newbie
Activity: 89
Merit: 0
When this GPU miner is made, use the copy plus paste from sphlib, but I avoid it.
Pages:
Jump to: