Save
Saving
  • JackIT

    For this test: I used a single 12 GPU 1080Ti rig on Windows split into two mining instances. Both mining instances were balanced to get as close to the same hash rate as possible. Each mining instance is running at the same time on the same machine on the same pool .

    Both instances have intensity set to 21, GPU target power set to 100%

    Static Pool Difficulty set to 74

    The instances/rig are managed by Awesome Miner, which restarts the miners for various reasons, based on rules I set.

    The plan is to do two rounds of testing, each round lasting approximately 10-12hrs, at the conclusion of each round I'll rotate the miner to a new instance, until each miner has had a chance to run on each instance. I'll also normalize the results to averaged blocks found per round.

    Miner tested:
    Enemy 1.08 - 1% dev fee


    Notes: This test was outside the norm of my previous tests, from the very beginning. Most of my previous test relied on three instance (4 GPUs each) of my 12 GPU rig, that I literally spent hours balancing the hash rate, so they're as equivalent as possible. In this test I decided to split the rig in half, and unfortunately I didn't save the GPU break-down for each instance, the last time I balanced a 2 instance setup. I spent about 30-45 mins balancing them, and seem to get it to a good equal point. As the results below show, I didn't do a good enough or long enough job of balancing the hash rates.

    However, the final result is valid (within some unknown margin of error) because I do two rounds of testing. In any event, I going back to the previous three instance configuration (which I thankfully have saved).

    Additionally, Round 1 was supposed to have concluded after ~10hrs, but I fell asleep watching TV :(



    Results:


    Round 1 - 264 Blocks Found - Duration: 1,020 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 259.73 RVN / Raw: 353.45 RVN - Enemy no OC Pool Link
    [Instance 2] Normalized: 273.78 RVN / Raw: 372.57 RVN - Enemy +250 mem Pool Link


    Round 2* - 124 Blocks Found - Duration: 647 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 401.00 RVN / Raw: 256.31 RVN - Enemy +250 mem Pool Link
    [Instance 2] Normalized: 409.06 RVN / Raw: 261.46 RVN - Enemy no OC Pool Link


    Normalized Average Results
    #1 - 334.40 RVN +250 mem
    #2 - 337.39 RVN no OC ( -0.90%)

    posted in RVN Mining read more
  • JackIT

    For this test: I used a single 12 GPU 1080Ti rig on Windows split into three mining instances. All three mining instances were balanced to get as close to the same hash rate as possible. Each mining instance is running at the same time on the same machine on the same pool .

    All three instances have intensity set to 21, GPU target power set to 90%, no overclocking.

    Static Pool Difficulty set to 50

    The instances/rig are managed by Awesome Miner, which restarts the miners for various reasons, based on rules I set.

    The plan is to do three rounds of testing, each round lasting approximately 10-12hrs, at the conclusion of each round I'll rotate the miner to a new instance, until each miner has had a chance to run on each instance. I'll also normalize the results to averaged blocks found per round.

    Miner tested:
    Enemy 1.08 - 1% dev fee
    Silent Miner v1.0.9 - 1% dev fee
    a1min3r Only human 1.3.3 Alpha - no dev fee


    NOTE: I didn't realize that "a1min3r Only human" was beta, (with some bugs to figure out), before I started this test and even though it did well against Silent Miner I think it's only fair to wait until it out of beta to fully test it.

    I cut round two testing to 7hrs, I'd been monitoring it periodically, and the gap between the two miners was fairly steady throughout the test. with the results from round 1, there was no reason to continue.


    Results:


    Round 1 - 144 Blocks Found - Duration: 651 minutes (2 restarts for Only Human)

    [Instance 1] Normalized: 138.40 RVN / Raw: 177.15 RVN - Enemy Pool Link
    [Instance 2] Only Human Pool Link
    [Instance 3] Normalized: 129.02 RVN / Raw: 165.15 RVN - Silent Miner Pool Link


    Round 2* - 81 Blocks Found - Duration: 420 minutes (no restarts)

    [Instance 1] Normalized: 118.96 RVN / Raw: 85.65 RVN - Silent Miner Pool Link
    [Instance 2] Only Human
    [Instance 3] Normalized: 129.75 RVN / Raw: 93.42 RVN - Enemy Pool Link


    Normalized Average Results

    #1 - 134.07 RVN - Enemy
    #2 - 123.99 RVN - Silent Miner [ -8.13% ]

    Raw Average Results

    #1 - 135.29 RVN - Enemy
    #2 - 125.40 RVN - Silent Miner [ -7.88% ]

    posted in RVN Mining read more
  • JackIT

    I have not tested Silent Miner, do you have a link?

    posted in RVN Mining read more
  • JackIT

    For this test: I used a single 12 GPU 1080Ti rig on Windows split into three mining instances. All three mining instances were balanced to get as close to the same hash rate as possible. Each mining instance is running at the same time on the same machine on the same pool .

    The instances/rig are managed by Awesome Miner, which restarts the miners for various reasons, based on rules I set.

    Static pool diff set to 66, miner intensity set to 21

    The plan is to do three rounds of testing, each round lasting approximately 8-12hrs, at the conclusion of each round I'll rotate the miner to a new instance, until each miner has had a chance to run on each instance. I'll also normalize the results to averaged blocks found per round.


    Miner details:
    Enemy 1.08 - 1% dev fee


    Round 1 - 131 Blocks Found - Duration: 720 minutes (one restart for "80% TDP / +250 mem")

    [Instance 1] Normalized: 147.40 RVN / Raw: 155.72 RVN - (80% TDP / No overclock) Pool Link
    [Instance 2] Normalized: 142.50 RVN / Raw: 150.54 RVN - (80% TDP / +250 mem) Pool Link
    [Instance 3] Normalized: 152.37 RVN / Raw: 160.97 RVN - (100% TDP / +250 mem) Pool Link

    • Notes: I'm not sure why, but (80% TDP / +250 mem) generally had the largest deviation between miner reported hash rate and pool reported hash. Miner reported hash rate for all three instance tended to be very close, with less than a 10% spread (typically ~5%) from high/low.

    Round 2 - 112 Blocks Found - Duration: 671 minutes (No Restarts)

    [Instance 1] Normalized: 157.09 RVN / Raw: 141.89 RVN - (100% TDP / +250 mem) Pool Link
    [Instance 2] Normalized: 146.90 RVN / Raw: 132.68 RVN - (80% TDP / No overclock) Pool Link
    [Instance 3] Normalized: 145.39 RVN / Raw: 131.32 RVN - (80% TDP / +250 mem) Pool Link


    Round 3 - 129 Blocks Found - Duration: 633 minutes (No Restarts)

    [Instance 1] Normalized: 139.23 RVN / Raw: 144.84 RVN - (80% TDP / +250 mem) Pool Link
    [Instance 2] Normalized: 143.31 RVN / Raw: 149.09 RVN - (100% TDP / +250 mem) Pool Link
    [Instance 3] Normalized: 133.23 RVN / Raw: 138.60 RVN - (80% TDP / No overclock) Pool Link



    Normalized Average Results

    #1 - 150.92 RVN (100% TDP / +250 mem)
    #2 - 142.51 RVN (80% TDP / No overclock) [ -5.90% ]
    #3 - 142.37 RVN (80% TDP / +250 mem) [ -6.01% ]


    Raw Average Results

    #1 - 150.65 RVN (100% TDP / +250 mem)
    #2 - 142.33 RVN (80% TDP / No overclock) [ -5.85% ]
    #3 - 142.23 RVN (80% TDP / +250 mem) [ -5.92% ]

    posted in RVN Mining read more
  • JackIT

    For this test: I used a single 12 GPU 1080Ti rig on Windows split into three mining instances. All three mining instances were balanced to get as close to the same hash rate as possible. Each mining instance is running at the same time on the same machine on the same pool .

    All three instances have intensity set to 20, GPU target power set to 80%, fans at 75%, no overclocking.

    The instances/rig are managed by Awesome Miner, which restarts the miners for various reasons, based on rules I set.

    The plan is to do three rounds of testing, each round lasting approximately 10-12hrs, at the conclusion of each round I'll rotate the miner to a new instance, until each miner has had a chance to run on each instance. I'll also normalize the results to averaged blocks found per round.

    Miner tested:
    Suprminer 1.6 - no dev fee
    Enemy 1.08 - 1% dev fee
    Ravencoin Miner v2.5 - no dev fee


    Results:


    Round 1 - 98 Blocks Found - Duration: 511 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 100.96 RVN / Raw: 100.62 RVN - Suprminer Pool Link
    [Instance 2] Normalized: 115.46 RVN / Raw: 115.07 RVN - Enemy Pool Link
    [Instance 3] Normalized: 105.67 RVN / Raw: 105.31 RVN - RavenCoin Pool Link


    Round 2 - 100 Blocks Found - Duration: 598 minutes (Suprminer = 1 restart, accepted shares not increasing)

    [Instance 1] Normalized: 106.65 RVN / Raw: 108.46 RVN - RavenCoin Pool Link
    [Instance 2] Normalized: 113.26 RVN / Raw: 115.18 RVN - Suprminer Pool Link
    [Instance 3] Normalized: 114.44 RVN / Raw: 116.38 RVN - Enemy Pool Link


    Round 3 - 97 Blocks Found - Duration: 566 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 118.62 RVN / Raw: 117.01 RVN - Enemy Pool Link
    [Instance 2] Normalized: 109.91 RVN / Raw: 108.42 RVN - RavenCoin Pool Link
    [Instance 3] Normalized: 106.68 RVN / Raw: 105.23 RVN - Suprminer Pool Link


    Normalized Average Results

    #1 - 116.17 RVN Enemy
    #2 - 107.41 RVN RavenCoin [ -8.16% ]
    #3 - 106.97 RVN Suprminer [ -8.60% ]

    Raw Average Results

    #1 - 116.15 RVN Enemy
    #2 - 107.40 RVN RavenCoin [ -8.15% ]
    #3 - 107.01 RVN Suprminer [ -8.54% ]

    posted in RVN Mining read more
  • JackIT

    TEST IN PROGRESS

    For this test: I used a single 12 GPU 1080Ti rig on Windows split into three mining instances. All three mining instances were balanced to get as close to the same hash rate as possible. Each mining instance is running at the same time on the same machine on the same pool .

    All three instances have intensity set to 20, GPU target power set to 80%, fans at 75%, no overclocking.

    The instances/rig are managed by Awesome Miner, which restarts the miners for various reasons, based on rules I set.

    The plan is to do three rounds of testing, each round lasting approximately 10-12hrs, at the conclusion of each round I'll rotate the miner to a new instance, until each miner has had a chance to run on each instance. I'll also normalize the results to averaged blocks found per round.

    Miner tested:
    Suprminer 1.6 - no dev fee
    Ravencoin Miner v2.5 - dev fee disabled
    PoolParty 1.3 - dev fee disabled


    Results:


    Round 1 - 127 Blocks Found - Duration: 671 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 110.08 RVN / Raw: 125.57 RVN | Suprminer Pool Link
    [Instance 2] Normalized: 110.10 RVN / Raw: 125.59 RVN | RavenCoin Pool Link
    [Instance 3] Normalized: 103.98 RVN / Raw: 118.61 RVN | PoolParty Pool Link


    Round 2* - 103 Blocks Found - Duration: 684 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 103.29 RVN / Raw: 95.56 RVN | PoolParty Pool Link
    [Instance 2] Normalized: 103.48 RVN / Raw: 95.73 RVN | Suprminer Pool Link
    [Instance 3] Normalized: 105.70 RVN / Raw: 97.79 RVN | RavenCoin Pool Link


    Round 3 - 104 Blocks Found - Duration: 691 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 109.73 RVN / Raw: 102.50 RVN | RavenCoin Pool Link
    [Instance 2] Normalized: 105.23 RVN / Raw: 98.30 RVN | PoolParty Pool Link
    [Instance 3] Normalized: 107.39 RVN / Raw: 100.32 RVN | Suprminer Pool Link


    Normalized Average Results

    #1 - 108.51 RVN RavenCoin
    #2 - 106.98 RVN Suprminer
    #3 - 104.17 RVN PoolParty

    Raw Average Results

    #1 - 108.63 RVN RavenCoin
    #2 - 107.21 RVN Suprminer
    #3 - 104.16 RVN PoolParty

    posted in RVN Mining read more
  • JackIT

    @revdete I don't use afterburner. I use Nvidia Inspector

    posted in RVN Mining read more
  • JackIT

    I decided to test various intensity settings for Suprminer 1.6, eliminating as many variables as possible.

    For this test: I used a single 12 GPU 1080Ti rig on Windows split into three mining instances. All three mining instances were balanced to get as close to the same hash rate as possible. Each mining instance is running at the same time on the same machine on the same pool .

    All three instances have GPU target power set to 80%, fans at 75%, no overclocking.

    The instances/rig are managed by Awesome Miner, which restarts the miners for various reasons, based on rules I set.

    The plan is to do three rounds of testing, each round lasting approximately 12hrs, at the conclusion of each round I'll rotate the miner to a new instance, until each miner has had a chance to run on each instance. I'll also normalize the results to averaged blocks foundperround.

    Miner tested:
    Suprminer 1.6 - no dev fee


    Conclusion/Commentary:


    Going from an intensity setting of 19 to 20, yields a 10% improvement, a notable difference. The gain from 20 to 21 is much lower, less than 2%.

    I've also anecdotally noticed that some GPUs can handle handle an intensity setting of 21 better than others. What I mean is that it seemed that the spread of MH/s (High/Low) on GPUs with an intensity of 21 was greater than on any other intensity setting. For example on an instance running -i 21, the slowest card might be at 14MH/s and the fastest card would be at 20MH/s. Keep in mind that these are all 1080Ti's

    Seems to me, that there might be some more gains when optimizing intensity on a per GPU basis.

    For what it's worth, I run my 1080/1080Ti's at -i 21 and my 1070Ti's at -i 20. If I get bored or have too much free time, I might experiment with testing optimizing intensity settings on the 1080's or even on a per GPU basis.. but that would take a lot of time.


    Results:


    Round 1 - 146 Blocks Found - Duration: 720 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 102.96 RVN / Raw: 104.15 RVN | -i 19 Pool Link
    [Instance 2] Normalized: 109.83 RVN / Raw: 111.10 RVN | -i 20 Pool Link
    [Instance 3] Normalized: 110.55 RVN / Raw: 111.83 RVN | -i 21 Pool Link


    Round 2 - 128 Blocks Found - Duration: 684 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 137.80RVN / Raw: 122.21 RVN | -i 21 Pool Link
    [Instance 2] Normalized: 122.27 RVN / Raw: 108.43 RVN | -i 19 Pool Link
    [Instance 3] Normalized: 133.00 RVN / Raw: 117.95 RVN | -i 20 Pool Link


    Round 3 - 159 Blocks Found - Duration: 621 minutes (zero restarts or downtime)

    [Instance 1] Normalized: 104.63 RVN / Raw: 115.26 RVN | -i 20 Pool Link
    [Instance 2] Normalized: 104.19 RVN / Raw: 114.78 RVN | -i 21 Pool Link
    [Instance 3] Normalized: 86.07 RVN / Raw: 94.82 RVN | -i 19 Pool Link


    Normalized Average Results

    #1 - 117.52 RVN -i 21
    #2 - 115.82 RVN -i 20
    #3 - 103.77 RVN -i 19

    Raw Average Results

    #1 - 116.27 RVN -i 21
    #2 - 114.77 RVN -i 20
    #3 - 102.47 RVN -i 19

    posted in RVN Mining read more