[ANN] Avermore miner - optimized sgminer (AMD) fork for x16r/x16s/xevan(beta)

[brianmct]

strange, but I actually get lower rates with 1.2 on RX480 8G with Uber 3.1 mem timings
(running with  -g 2 -w 64 -X 256 --benchmark)

X16r    Avermore-1.1    8.4 mh
X16r    Avermore-1.2    7.743 mh

X16s    Avermore-1.1    8.366 mh
X16s    Avermore-1.2    7.711 mh

Interesting. Can you compare the Hamsi kernel hashrates between 1.1 and 1.2? (run with --benchmark-sequence=BBBBBBBBBBBBBBBB). It's possible some of the changes I made caused regressions in cards other than the RX580.

[1714746458]

1714746458

[1714746458]

1714746458

[1714746458]

1714746458

[1714746458]

1714746458

[1714746458]

1714746458

[4ward]

strange, but I actually get lower rates with 1.2 on RX480 8G with Uber 3.1 mem timings
(running with  -g 2 -w 64 -X 256 --benchmark)

X16r    Avermore-1.1    8.4 mh
X16r    Avermore-1.2    7.743 mh

X16s    Avermore-1.1    8.366 mh
X16s    Avermore-1.2    7.711 mh

Interesting. Can you compare the Hamsi kernel hashrates between 1.1 and 1.2? (run with --benchmark-sequence=BBBBBBBBBBBBBBBB). It's possible some of the changes I made caused regressions in cards other than the RX580.

exits immediately when I use --benchmark-sequence=anything (even 0123456789ABCDEF) in both 1.1 and 1.2

edit:
I also noticed that I'm hitting the thermal card power limit and it starts throttling
in 1.1 the GPU clock gets throttled to ~1200Mhz and temperature stays around 80c, so it can probably reach even higher speeds with better cooling
in 1.2 the GPU clock is ~1300Mhz and temperature is ~75c

580 is basically the same card, with slightly higher clocks, so should perform similarly

[brianmct]

strange, but I actually get lower rates with 1.2 on RX480 8G with Uber 3.1 mem timings
(running with  -g 2 -w 64 -X 256 --benchmark)

X16r    Avermore-1.1    8.4 mh
X16r    Avermore-1.2    7.743 mh

X16s    Avermore-1.1    8.366 mh
X16s    Avermore-1.2    7.711 mh

Interesting. Can you compare the Hamsi kernel hashrates between 1.1 and 1.2? (run with --benchmark-sequence=BBBBBBBBBBBBBBBB). It's possible some of the changes I made caused regressions in cards other than the RX580.

exits immediately when I use --benchmark-sequence=anything (even 0123456789ABCDEF) in both 1.1 and 1.2

edit:
I also noticed that I'm hitting the thermal card power limit and it starts throttling
in 1.1 the GPU clock gets throttled to ~1200Mhz and temperature stays around 80c, so it can probably reach even higher speeds with better cooling
in 1.2 the GPU clock is ~1300Mhz and temperature is ~75c

580 is basically the same card, with slightly higher clocks, so should perform similarly

I'll take a look into why --benchmark-sequence isn't working on Windows.

I did some testing on my test rig, and the results are weird.

The Hamsi kernel, when benchmarked by itself (i.e. repeated 16 times), is indeed 35% faster on 1.2 compared to 1.1 (4.2 MHs -> 5.7 MHs). Also, when I benchmark the rest of the algos (0123456789ACDEF), the hashrates are the same between the two versions (as expected, since the only change was with the Hamsi kernel).

However, when I test running all of the algos (0123456789ABCDEF), v1.1 is indeed slightly faster than v1.2 (7.9 MHs vs 7.7 MHs). I'm a bit stumped right now on why this would be the case. So anyways; if you're getting better results on v1.1, then stick with that version for now. I'll see if I can get a patch to v1.2 to address this issue.

All tests were done with -g 2 -w 64 -X 64, by the way. Found this to be the optimal xIntensity on my RX580. When running with -X 256, v1.2 did beat out v1.1 (7.1 MHs vs 6.8 MHs), but both were slower than with -X 64.

[pigseye]

Hi Brian,
Great miner. 

Question, What does the -w 64 command do?  Should we be adjusting this or do you have guidelines on how to adjust it for max hash rate?

Thanks

[4ward]

Hi Brian,
Great miner. 

Question, What does the -w 64 command do?  Should we be adjusting this or do you have guidelines on how to adjust it for max hash rate?

Thanks

its worksize, and and it might give you different results with different numbers (can try 128 / 256), but in general 64 usually works best for most algos on amd

[pigseye]

Hi Brian,
Great miner. 

Question, What does the -w 64 command do?  Should we be adjusting this or do you have guidelines on how to adjust it for max hash rate?

Thanks

its worksize, and and it might give you different results with different numbers (can try 128 / 256), but in general 64 usually works best for most algos on amd

Thanks for the tips!

[pigseye]

Hi All,
Another question.  I seem to get the highest and most consistent hash rate from 1.0.  My hashrate for 1.1 is significantly lower than 1.0.

I've got a 12 card system with 6-580s, 3-570s, 2-480s, and a R9290.

I obviously do not use -g 2 because of the 12 cards but have tried different intensities from 64 to 512 and 1.0 comes out on top.

Any explanation?

Thanks!

[brianmct]

Hi All,
Another question.  I seem to get the highest and most consistent hash rate from 1.0.  My hashrate for 1.1 is significantly lower than 1.0.

I've got a 12 card system with 6-580s, 3-570s, 2-480s, and a R9290.

I obviously do not use -g 2 because of the 12 cards but have tried different intensities from 64 to 512 and 1.0 comes out on top.

Any explanation?

Thanks!

1.1 and 1.0 should be identical when it comes to x16r/x16s; I just added xevan support for RX cards in 1.1.

Make sure when you're comparing hashrates you either average for a few hours, or use the --benchmark flag. X16R hashrates will vary a lot depending on the current block's hash sequence.


If you have 12 cards on the same rig, you can try splitting up the cards into two instances of Avermore (using the -d flag) to get around the 13-thread limit.

On the first instance of Avermore, run with -d 0,1,2,3,4,5 and on the second, run -d 6,7,8,9,10,11. This should allow you to run the miner with -g 2, which should get you better hashrates.

Let me know how it goes.

[CryptoDocker]

brianmct any chance you can give us the commands to manually update Avermore in Hive OS to the newer version? Would like to help test it but im terrible in linux

[brianmct]

brianmct any chance you can give us the commands to manually update Avermore in Hive OS to the newer version? Would like to help test it but im terrible in linux

I wrote up a guide for how to manually update Avermore in HiveOS here.

[CryptoDocker]

brianmct any chance you can give us the commands to manually update Avermore in Hive OS to the newer version? Would like to help test it but im terrible in linux

I wrote up a guide for how to manually update Avermore in HiveOS here.

Your the best! Thanks for that

[tome3000]

hi nice pepole 
can someone help set a fail over for this ?  . tried the folowing with no luck :

my bat  : sgminer.exe --gpu-platform=2 --config failover.conf  -w 64 -g 2 -X 256 --no-submit-stale
my conf :
 {
 "pools": [
   {
     "name": "Raven minepool",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://minepool.com:3636",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   },
   {
     "name": "Raven Yiimp",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://yiimp.eu:3666",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   },
   {
     "name": "Mining Panda",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://miningpanda.site:3666",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   }
 ],
 "api-port": "4028",
 "failover-only": true
}

[pigseye]

Hi All,
Another question.  I seem to get the highest and most consistent hash rate from 1.0.  My hashrate for 1.1 is significantly lower than 1.0.

I've got a 12 card system with 6-580s, 3-570s, 2-480s, and a R9290.

I obviously do not use -g 2 because of the 12 cards but have tried different intensities from 64 to 512 and 1.0 comes out on top.

Any explanation?

Thanks!

1.1 and 1.0 should be identical when it comes to x16r/x16s; I just added xevan support for RX cards in 1.1.

Make sure when you're comparing hashrates you either average for a few hours, or use the --benchmark flag. X16R hashrates will vary a lot depending on the current block's hash sequence.


If you have 12 cards on the same rig, you can try splitting up the cards into two instances of Avermore (using the -d flag) to get around the 13-thread limit.

On the first instance of Avermore, run with -d 0,1,2,3,4,5 and on the second, run -d 6,7,8,9,10,11. This should allow you to run the miner with -g 2, which should get you better hashrates.

Let me know how it goes.

This is a great idea.  Duh!  I could have thought of that. 

I'll give it a try tonight and get back to you.  Thanks!

[hammuh]

hi nice pepole 
can someone help set a fail over for this ?  . tried the folowing with no luck :

my bat  : sgminer.exe --gpu-platform=2 --config failover.conf  -w 64 -g 2 -X 256 --no-submit-stale
my conf :
 {
 "pools": [
   {
     "name": "Raven minepool",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://minepool.com:3636",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   },
   {
     "name": "Raven Yiimp",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://yiimp.eu:3666",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   },
   {
     "name": "Mining Panda",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://miningpanda.site:3666",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   }
 ],
 "api-port": "4028",
 "failover-only": true
}

Maybe try a password like "c=RVN"?

[LovellaSai]

I am sincerely looking forward to using this as my new thing and it’ll be a good thing.

[Bwtmn7]

Brian do you think you could optimize for Fiji because im only getting 5.5MH/s with OC at 1150 /500 no bios mod

[wtfonly16]

only 5.5mh/s on 570.. two threads tried 64,96,128,256,512,768,896,1024 various restarts still same hashrate...

sometimes all dip to 3.9 for no reason at all...

[tome3000]

hi nice pepole 
can someone help set a fail over for this ?  . tried the folowing with no luck :

my bat  : sgminer.exe --gpu-platform=2 --config failover.conf  -w 64 -g 2 -X 256 --no-submit-stale
my conf :
 {
 "pools": [
   {
     "name": "Raven minepool",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://minepool.com:3636",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   },
   {
     "name": "Raven Yiimp",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://yiimp.eu:3666",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   },
   {
     "name": "Mining Panda",
     "nfactor": "10",
     "algorithm": "x16r",
     "quota": "1;stratum+tcp://miningpanda.site:3666",
     "user": "RFBMCANQkBpd6KRiTFizuf59TCZNUv8GbC.VegaFrontier",
     "pass":
   }
 ],
 "api-port": "4028",
 "failover-only": true
}

Maybe try a password like "c=RVN"?

.

no  still does not work  any other idea?

[MoNTE48]

Did a fork of this miner https://github.com/KL0nLutiy/sgminer-gm-x16r, must be 1-2% faster

Hi, friend. I use Ubuntu (HiveOS).
You can compile your miner for linux? Big thank you!

[pubmongrelT]

You can set up this at regular AMDs if you have the 18.3.4 drives in your system.