 Show Posts
|
|
Pages: [1] 2 »
|
Difficulty has doubled in just a day after the GPU miner has been released.
Difficulty has gone up 10 fold in about a day or two, does anyone know if the gpu miner is being sold by someone or if there's a github where I download it from? Bel is working on a beta GPU miner in the Discord server (check pins in #mining). It's still a bit buggy, but ask in the channel and we'll help you set it up. - brianmct |
|
|
|
Avermore miner v1.4.1: - Optimizations to Echo (+7%) for an overall 1% speedup in x16.
- Added code to automatically select AMD OpenCL platform by default. This means that setting --gpu-platform is no longer required for rigs with integrated graphics or Nvidia cards.
- More sane default parameters. Changed default xIntensity to 256, from a default intensity of 8.
https://github.com/brian112358/avermore-miner/releases/tag/v1.4.1 |
|
|
|
Ideas after you config file. And what GPU model? sgminer.exe -k x16r -o stratum+tcp://us.ravenminer.com:5678 -u myrvnwallet -p worker1 -X 256 (also tried -X 64 same result) This is a fresh Win10 install updated, redist installed, blockchain drivers XFX Core Black RX580 brand new out of the box with stock bios, installed in the pcie 16x slot... 1st. add --log-file log-x16r.txt after -X256 and look inside when starting miner. 2nd. the miner must compile a kernel for you GPU its may be more than 2-3 minutes up to 5. see log-x16r.txt 3rd. try without -X256. And why you use blockchain drivers? it's does not matter but why? new adrenalin drivers is good performance and more flexible with playing with voltage and overclocks(for me). Already did in my first post and this is what I got: [23:17:04] Started avermore 1.4.0 [23:17:04] * using Jansson 2.7 [23:17:18] Started avermore 1.4.0 [23:17:18] * using Jansson 2.7 [23:19:09] Started avermore 1.4.0 [23:19:09] * using Jansson 2.7 [23:19:13] Started avermore 1.4.0 [23:19:13] * using Jansson 2.7 Also tried new Adrenalin drivers, removed the X256... with "pause" as per example.bat gives me: Press any key to continue... if I hit any key just closes and without the "pause" simply closes down. Thank you for your time helping me on this Kaezar If you join the Ravencoin community Discord I can help you there. Much easier to provide tech support over chat than over BCT  |
|
|
|
What wrong with xevan for vega's GPU?
People trying to run xevan on Vega's were reporting hardware errors and no submitted shares. It's probably an issue with the new clang-based OpenCL compile stack for Vega cards miscompiling the kernels, but I don't have a Vega card of my own to test it on to try to fix the issue. |
|
|
|
Thanks, but that's not my question, i'd like to know the hashrate of RX 570 with an ETH bios.
6-8 mh/s thats like the speed of a 1060 for comparison 1070s get 11mhs and 1070ti like 15 these amd miners need a fully rebuilt and optimized kernel not just a copy paste you amd miners paying a fee to mine at the same speed as a 1060 need to stop and got mine xmr and buy raven if you have to lol An Rx 570 is comparable to a GTX 1060 as a lower end/mainstream gaming video card. A GTX 1070 is a whole market level higher, both for gaming and mining. heck no, not on pricing and the performacr on other algos, and what about the 8gb 580s 7 to 8 mhs on raven is total trash when it can get 32 mhs eth , 950 on xmr vs 24 max and 500 max for the 1060 6gb card which doesn’t eveb have and 8gb variant. the 1070 might be higer end but there is a reason 570 and 580s cost $100 more (8gb) than the 1060s. 32 cores vs 36 cores for the 570 580 comparison is not much of a difference. My 570s nitros havecthe same memory as the 580s and within 1% of the same hash rates given the same clocks or memory straps face it these coppied x16 kernels are trash , better to mine eth or xmr for now with kernels that were designed for Opencl This is somewhat of an unfair comparison. Nvidia x16r CUDA miners have had years of work put into them in optimizing their individual kernels, while sgminer has been more or less abandoned for the past few years. So, I'm basically starting from scratch in terms of optimizations for AMD. There's a lot of room to cover, since there are 16 different algorithms that need to be optimized, and simply optimizing one of them does not affect the average that much. The more optimistic way to look at this is that there's a lot more room for improvement for better x16r performance on AMD cards. I've improved the hashrate by around 20% from the next best x16 miner, and 10% from version 1.0 to 1.4. Even with the current performance of Avermore, Raven is one of the most profitable coins to mine on WhatToMine right now for AMD cards. Profitability will go up as the kernels get more optimized. If you don't like the miner, no one's forcing you to use it. But I would really appreciate it if you didn't insult / dismiss the dozens of hours of work I have put into making x16r and x16s mining better for AMD miners. |
|
|
|
Avermore version 1.4:- Optimizations to whirlpool (+65%), bmw (+50%), and echo (+3%) kernels for an overall 6% speedup in x16
|
|
|
|
Does this miner support older hawaii (390/290) and fiji (fury/nano) cards?
I've had people run Avermore on both Hawaii and Fiji cards successfully, so they should both work. You might need to play around with the worksize (-w, default 64) and xIntensity (-X) a little for best performance. See the Optimizing Hashrate for tips on tweaking these parameters. |
|
|
|
Avermore version 1.3: - Significant optimizations to Skein (+80%), Blake, (+65%), and Keccak (+40%), for an overall 3.5-4% increase in hashrate.
- Dev pool switching is now much less disruptive (no longer reinitializes GPU threads). This should fix an issue where GPUs would occasionally go DEAD upon dev pool switch.
|
|
|
|
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. |
|
|
|
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. |
|
|
|
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. |
|
|
|
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. |
|
|
|
Hey everyone, I just made a new release: Avermore v1.2Made some major improvements to the Hamsi kernel (+35%), with a 4% improvement in the overall average hashrate. I'm now getting 7.8-8 MH/s average on my test rig's RX580 (up from 7.5 MH/s). As always, feedback and reporting of issues is highly appreciated. |
|
|
|
hi well i have a problem evrey few houres when the miner swich from my pool to dev pool the miner just get stuck and wont recontect to my pool allwayes saying trying to reconect but it wont so i need to close the miner and start it agian .can you please fix it . beside that it does make 10% more  I have the same issue, gets stuck after switching to dev pool, unusable in current state... stops every couple hours. Most people using it don't seem to have this issue; seems like it might be a problem with certain pools / network configurations. Trying to gather some more data to debug the issue; if you could provide any of the following info, it would be super useful: - Pool you're mining on - Logs with the pool switching issue (Run the miner with "-L log.txt" and DM me the log file) - OS |
|
|
|
Does anyone have suggested OC or power limit settings for RX 580 cards?
I personally run 1400 core, stock mem/BIOS, full power on my RX 580. Although, I haven't done much hardware tuning, since this is just on my test / benchmarking rig, which I prefer to have consistent results on for testing optimizations. Replied on the Github issue. |
|
|
|
if anyone shares Vega .bat settings , would be glad to appraise.
Try starting with the following configuration for Vega (there's some more tips for tuning your settings here): set GPU_FORCE_64BIT_PTR=0
set GPU_USE_SYNC_OBJECTS=1
set GPU_MAX_ALLOC_PERCENT=100
set GPU_SINGLE_ALLOC_PERCENT=100
set GPU_MAX_HEAP_SIZE=100
:: Set the pool, username, and port here. All three arguments are required.
sgminer.exe -k x16r -o stratum+tcp://pool.here:port -u user -p password -g 2 -w 64 -X 320
Let me know how it goes Awesome, will give it a try later today.
There are no AMD miners for X17, but it seems similar to X16R with 1 more algorithm. Is it possible to port the optimizations into a separate X17 miner/kernel?
That sounds doable; I'll look into it. |
|
|
|
Thanks brian really glad to see an optimized AMD miner for Raven. I'm dying to get this loaded onto Hive OS as dimafern is currently only using your sgminer fork version which is very slow for AMD cards - can you confirm that version will now cease being supported and effectively replaced with Avermore?
Yes; my old sgminer fork was made for the RVN AMD bounty, hence the lack of optimization. I will continue my optimization work for AMD cards on Avermore. Unfortunately, Avermore is not yet on HiveOS, but I made a wiki guide for how to replace my old sgminer fork manually with Avermore; it should be pretty straightforward. Multiple people have already requested that Avermore be added to HiveOS, and it's on the HiveOS miner wish list; it should be added soon. |
|
|
|
I'll be the first to admit that my original sgminer for x16r / x16s wasn't very well optimized. It was made in a hurry for the RVN AMD miner bounty, so it was rushed. Anyways, I've finally gotten my hands on a RX580 card and had the time to make optimizations. Introducing... the Avermore miner. Not only is the devfee only 1% compared to aerisminer's 2%, it's also faster. Aeris is claiming 5-6.5 MH/s on RX570/580 and 11-12 MH/s on Vega? Well, my miner does 7.2-7.5 MH/s on my stock RX580, and others report 13-14 MH/s on Vega. So, a whole 15-25% faster than aerisminer. Avermore also works with x16r, x16s, and xevan, and is faster than any current AMD miner out there for all three of these algos. And the best part? Avermore doesn't blatantly violate the sgminer GPL license. For those of you who are unaware, the GPL license (which sgminer is released under) requires that "programs distributed as pre-compiled binaries be accompanied by a copy of the source code, a written offer to distribute the source code via the same mechanism as the pre-compiled binary, or the written offer to obtain the source code that the user got when they received the pre-compiled binary under the GPL." [source]. Aeris is clearly violating this license, as they clearly admit that it is "based on sgminer-gm-5.5.5". If you have any respect whatsoever for the time and effort developers put into developing the miners that makes GPU mining possible, you should not support Aeris' blatantly illegal ripoff of others' work.May you please create a thread ? Made an ANN thread here. I had previously announced it on the RVN ANN thread, but this is better. Thanks for the suggestion. |
|
|
|
Hi everyone, This is my first time making a post here, but some of you may know me as the developer of Nevermore miner for x16r / x16s, and creator of one of the first sgminer forks for x16r. I recently had a generous person donate a RX580 to me, and thus was able to give the AMD side some love by making some x16r optimizations for AMD. After around a week of work, I've made numerous optimizations to the x16r / x16s kernels, and with preliminary testing I'm getting around +20-25% over ace's sgminer (the previous fastest x16r miner). x16r / x16s average hashrates: RX580: 8.3 MH/s Vega56: 13-14 MH/s Vega64: 14-15 MH/s xevan hashrates (still in beta; doesn't work with all cards yet): RX580: 1.8-2.1 MH/s, depending on OC Vega: N/A (HW errors) As with Nevermore (which is for Nvidia cards), Avermore (for AMD) has a 1% devfee to support continued work on the miner. I look forward to hearing your feedback about this new release. Avermore releases (Download latest release v1.4.1 here)Getting Started guideRecommended bat file to start with (see Optimizing Hashrate guide for more tips on tuning your system): set GPU_FORCE_64BIT_PTR=0
set GPU_USE_SYNC_OBJECTS=1
set GPU_MAX_ALLOC_PERCENT=100
set GPU_SINGLE_ALLOC_PERCENT=100
set GPU_MAX_HEAP_SIZE=100
:: Set the pool, username, and port here. All three arguments are required.
sgminer.exe -k x16r -o stratum+tcp://pool.here:port -u user -p password -g 2
Note, if you have a system with more than 6 cards, then you'll have to remove the "-g 2" option since more than 13 mining threads is currently unsupported. Alternatively, you can try splitting the cards between two instances of Avermore by using the -d flag, to circumvent the 13-thread limit. For example, to split a 12-card rig, run one instance with "-d 0,1,2,3,4,5" and another with "-d 6,7,8,9,10,11". Other useful links:Source codeGithub wiki pagesTroubleshooting guideOptimizing Hashrate guideHiveOS manual installation (to upgrade to latest release before it gets pushed by HiveOS) smOS installation |
|
|
|
|
