Topic: PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux) - page 307. (Read 784965 times)

I'll root for that. Everytime I need to do something with gpus I gotta translate them bloody 1-based indices.

not work -mvddc wnidows 7 PhoenixMiner 4.7c 4.6c . eror miner. in version 4.5 everything works.
setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100


PhoenixMiner.exe -amd    -mi 14 -coin ETH -mode 1 -wdog 0 -pool eu1.ethermine.org:4444 -worker CATx -wal 0x57119a8879088ae78442533c2e9e6649631004c6 -pass x  -clkernel 2  -gser 2 -proto 3 -mvddc 900 -mclock 2100 -gt 0 -rvram -1 

If the GPU has Samsung RAM then it will overclock really well which also goes for vega 56 with 64 bios flashed.
Flash the bios, increase mem and reduce core and subsequent voltages in overdriveNtool, use AMD Mem Tweaker to apply weak memory timings after that and run the miner. That's all I have to reach the highest possible hashrate in ETH and XMR.

seriously am I the only one who's confused by the difference in enums of this miner and the rest, like: win registry, amdmemtweak, odnt...

Anyone tried already new 19.11.2 AMD drivers? Miner supports them?

EDIT: I checked - new AMD 19.11.2 drivers have full PM support.

Thanks for the answer I will do it so.
Do you you have so good mem tweak for vega 56 with 64 bios and Vega 64 too on ETH ?

Definitely. Flash your Vega 56 with 64 air bios.
It'll help as it will increase the memory frequency of the card.
In 56 bios, I was able to hit 935 MHz but with 64 air bios flashed the memory frequency was stable at 1100MHz.
Any further and it was unstable.

@PhoenixMiner ?

Not sure I trust a modded Windows 10 OS. No offense but what's to prevent you from tossing in something malicious? (Trojan, Keylogger, Packetsniffer etc.)
I like the idea though.

Hello,
I have some vega 56 and 64 founder edition (ssamsung memory).
I'm trying to find some settings for them but it keep crashing.

I'm using this one on vega 56 : --rc 38 --rcdrd 12 --rcdwr 5 --rp 11 --faw 13 --RRDL 3 --rtp 4 --ref 15600
but i get only 41mh/s with crash.

Do I need to flash bios them with 64 bios ?

Is there any good specific timing for ETH on Vega 64 and 56 ?

Thanks

Thank you very much for this information, I did not know about this version before and i will definitely be using it for the fresh install. thanks again

Does -eres work for nvidia cards? Cos no matter what value I enter I always see same dag-size generated...

It's a good choice that you have decided to fresh install Windows.
Anyway, I'll recommend installing Windows 10 LTSC version as it is very stable, no feature updates so it won't go updating itself middle of the night and break something.
After installing, perform all the security updates, then disable automatic updates through group policy and start mining.
It'll run as stable as a Linux system.

Thank you for the suggestion, I tried it, disabled defender, ran program, no restart, same issue no change. I also went through disabled all other types of software and network related policies, tried again, no change. I guess ill bite the bullet and do a full wipe and reinstall of windows.

@PhoenixMiner

why are you enumerating the gpus starting from 1 and not from 0?
is there any way to change it so it will not be so "uniquely authentic" to the miner?

I recently have this problem, posting here because I use this miner, although its probably windows problem.

I start my miner by remote logging and manually starting miner. However, about month ago, win10 changed

behavior...when I shut down remote conection, rig machine goes to sleep after 1h exactly. I did all usual

sleep/hibernation stuff, even some advanced registry stuff I found on net, but nothing works


Anyone had similar problems and solved it?

new version can solve 1050ti DAG size problem?

   This seems like Windows Defender issue. Turn off Windows Defender completely (it can be done with gpedit.msc) and see if the issues go away.




   This is a long-standing problem with Nvidia 10x0 cards. As it started above certain size of DAG and gradually lowers the hashrate with each new epoch, it seems to be related to the TLB size (yes, TLB strikes again). At this point the only workaround is to increase the power limit (and thus indirectly the core clock) but whether it is worth it depends on your electricity prices. We are testing some theoretical fixes for this problem but the results are not very encouraging. Nvidia could probably fix this with driver update by increasing the page size as AMD did in the past but it is not very likely that they will bother to do it.

After AMD finally released Linux driver which can be used for Ethash mining with RX5700 cards, the release candidate of 4.8 is ready for download. You can download PhoenixMiner 4.8b from here:

(MEGA links are no longer active)

If you want to check the integrity of the downloaded file, please use the following hashes (you need the last file PhoenixMiner_NVRTC_Windows.zip only if you want to mine BCI with Nvdia cards under Windows):

The new features in this release are:

• Added support for the latest AMD drivers under Linux 19.30-934563 and for mining with RX5700 cards under Linux. Note that drivers are quite buggy and most of the overclocking options don't work, as well as most of the hardware monitoring options.
• Added a new flexible way for setting GPU-specific options with selectors. Example: -cclock 1-3:1090,4:1300 -mclock nvidia:+450,amd:2000,*:1900,gtx*1070*ti:+200 See bellow for more information
• Validated support for the latest AMD Windows drivers 19.11.1 (but they already work with 4.7c because there were no significant changes in the OpenCL driver)
• Other small improvements and fixes.

If you don't need or want any of these features, there is no need to upgrade yet.

About the AMD Linux drivers 19.30-934563:

These drivers finally fix the RX5700 TLB problem but they are not perfect:
- Most of the hardware monitoring options don't work for RX5700 cards - even the temperature is not reported correctly. Only the power usage, fan speed, and core voltage are reported.
- Almost all hardware control options don't work for RX5700 cards either with the notable exception of power limit, so the only way to avoid wasting power is to use -powlim -45 or something like this. The fan control options doesn't work either so you will be relying on auto-fan.

About the new selectors for the GPU-specific options:

This is best explained by example: -cclock *:1100,1-3:1090,4:1300 - here we are setting core clock to 1100 MHz
for all cards, except the cards 1 to 3, on which it is set to 1090 MHz, and card 4 to 1300 MHz.

The part before the colon : is the selector, which selects the GPUs for which the value after the colon is applied. The selector can be:
  - single GPU index: e.g. 5:1000 sets 1000 for the 5th GPU
  - range of GPU indexes: e.g 2-5:1200 sets 1200 for the GPUs 2,3,4, and 5
  - asterisk, which sets the value for all GPUs
  - label amd or nvidia: e.g. amd:1090 sets the value to 1090 for all AMD cards
  - arbitrary string that starts with letter and can contain letters, numbers and asterisks, which is matched against the GPU name as listed by PhoenixMiner.
    Example: gtx*1070:+500 will set value +500 for all cards which contain 'gtx' and '1070' in their names with anything
    between them. This will match 'Nvidia GeForce GTX 1070' but not 'Nvidia GeForce 1070'.
    
Note that if more than one selector matches given card, than only the last one counts.
Example: -cclock *:1100,1-4:1090,2:1300 will set card 2 to 1300; cards 1,2, and 4 to 1090;
and the rest of the cards to 1100 MHz core clock.

Anyone with the same issues? Or know a fix ?  (its getting annoying waking up or coming home to find no errors messages in windows or in the miner log, but the program had just turned off and will not restart.)