Topic: lolMiner 1.64: Fastest Kaspa miner in market - page 72. (Read 155997 times)

You are right... one of the best drivers is 20.4.2, why I say this version, because it is also the last one that works perfectly with MSI Afterburner and also with OverDriveNTool... that is really helps to configure voltage of the memory cards.

Let's enjoy in 20min the autofork to ETCHASH

You quoted the instructions for lol....in your message.....

So go to AMD site and download newest driver for Polaris cards (your RX 480 are polaris cards)

https://www.amd.com/en/support

The scroll up through thread and see link from lol for 1.16....
It is literally 5 posts before this one....

https://github.com/Lolliedieb/lolMiner-releases/releases/tag/1.16

[/b]

Which ones are they? I don't understand which ones it refers to. Do you have a link to download?

Well the instructions you quoted say Blockchain and Newer drivers.  Win 7 probably want to go to win 10.  Current version lolminer 1.16.

My operating system:
Win 10 x64
GPUS: AMD Rx 480 GPU
What video driver should I install? Which version?
Which version of lolMiner is recommended?

I have another operating system:
Win 7 x64
GPUS: AMD RX480
What video driver should I install? Which version?
Which version of lolMiner is recommended?

Sorry for not good english

Quick info for upcoming ETC hardfork (in few hours):

lolMiner 1.16 (and the versions before) is ready to support the fork that will happen on block 11 700 000.
To automatically switch over to the new scheme use --algo ETCHASH when you run the miner manually.

If you are using a mining OS like HiveOS, please select ETC or ETCHASH as parameter in your flight sheet.
In case you can not change it or are uncertain you alternatively can add --enable-ecip1099 to your extra user configuration parameters.

Note that ETCHASH = ETHASH + ECIP 1099, so the result will be the same and you can safely add  --enable-ecip1099 to any working configuration without damaging it.


All of this also applies when you mine ETC+ZIL! In this case just configure for ETC mining - the miner will recognize the ZIL mining and switch to ETHASH for that time and switch back to ETCHASH once Zil mining is over.
Have fun with the fork.



--------------




I answered about this one few pages earlier. Octupus is extremely compute heavy, in particular multiplication of two 32 bit numbers to a 64 bit one. While Turing and Ampere can do this operation using only one clock cycle, the AMDs need 4 (professional cards & Vii) to 8 clock cycles. Also Turning and Ampere can do easier operations in parallel. Therefore the expected performance on AMD is rather ... horrible. I did some tests but was way away from the two modern Nvidia numbers here. It might be that big navi can do better, but the full instruction set for this architecture is not yet released. Waiting for that if this allows some quicker approach.

Thank you for the quick answer, sounds promising!

And another question, the new Octopus algo is promising at nVidia, is there a plan to implement these in your miner for AMD?

Which driver version you are on and have you checked other values for --keepfree and --4G-alloc-size like described in the 4G_Windows_Readme.txt?



Well wrt SAM I am not sure yet what effects that has. On ethash you see the impact of the cache when building the DAG (way faster then on RX 5700 although bandwidth did not grow that much) and when mining ZIl, cause that uses a 1G dag that fits to over 12% into the cache - there the card is about 10% faster then for the 4G dag. For other algos I still try to understand how to use it ideally. Generally the card already is the fastest Equihash card among all AMDs (reaching 90+ sol/s on BTG) and its 2nd fastest on Beam. I got good hope to profit from the cache in Beam and Grin C32.

Is there a chance of a speed up of the RX6000 Series Cards because the new "Infinity Cache" or "Smart Access Memory"?

Can you put a screenshot or the GPUZ Information to see if that could help to find the problem

It is Windows 10, not Linux.

That is an interesting one. Usually should not happen... unless... is this Linux and if yes: which kernel version do you use please (you can find out with "uname -r")

What could be the reason for this message?
"GPU 1: Allocation of DAG failed with error: Memory object allocation failure GPU 1: will be stopped."

I tried to use lolMiner 1.15 and 1.16 - same error.
GPU 1 is a RX 5700 XT 8GB from PowerColor.

Can you give some details about the problem? e.g. cards, driver, error messages - the more details we have the quicker we get it going

thank you so much for your help

One of the best drivers now is 20.30, the 20.40 is having some problems on som Linux versions.

For mining in Windows, one of the best is 20.4.2.

Which are the best drivers to use with lolminer 1.16 for rx570 4gb?

i am not able to mine etc+zil wit lolminer someone can helps? thanks

Is there any new about the 6800 x 2 GPU?...

With the limit of bandwith that numbers doesn't seem normal... let's see

It might be, because my miner has no pauses between kernel calls ... it has two threads to keep the card busy ... Will need to think about this.



Can you share more info e.g. driver and operation system? I am running my 580 4G with still 15 mh/s on ETC epoch 388 in Linux with 20.40 drivers.



Wrt this 6800 stuff: it is indeed strange using the same card 2x. Why this works and if we can replicate this to get the speedup on other systems: not yet sure. Still awaiting an own 6800 for tests, the card to test the miner was a remote one where a friend allowed me to check it out - and there it was only shown as one card.