You are right. After my post, I tried to mine ETHW with TRM, which had about 6GB DAG, and it worked fine. I see at DAG generation "dual buffer mode":
GPU 0 Running kernels in dual buffer mode.
GPU 0 Starting DAG generation for epoch 618, 5.83 GB (dual buffer mode).

Kawpow algo run in "single buffer mode". I found command "--eth_dag_buf" to force buffer mode to Dual, but looks like it doesnt work under kawpow algo.

In usage.txt file I found, that older drivers not supported single buffer mode over 4GB. Unfortunately I really use very old driver (19.12.1), because all my cards works perfectly with it.
When I upgraded it to ANY newer, I had always some problems.

Can I somehow forced this Dual buffer mode on Kawpow algo??

I cant mine even Ravencoin anymore.
I have Rx580 8GB, RX5700 and Vega64.
6 VGA in rig, 64GB Virtual memory, but got error message:
"ERROR - failed to allocate ethash DAG buffer". For EVERY type of my VGAs above.
I can mine other coins easily with less, than 4GB Dag.

With Phoenixminer I can mine coins with over 4GB Dag.

Please.. Any idea???

Hello!
I cant mine Etc nor Eth with my Vega64. Trm v0.10.9-win Got message:

[2023-04-20 13:53:18] GPU 1 DAG allocation trying 8176.0 - 512.0 = 7664.0 MiB.
[2023-04-20 13:53:18] GPU 1 ERROR - failed to allocate ethash DAG buffer, tried down to 4088.0MiB.
[2023-04-20 13:53:18] GPU 1 thread exiting with error.

Under Phoenix miner I can mine Eth without problem.

I have enough Virtual memory, 60 GB (PM worked with 40GB, too).
I tried even to disable/enable my Gpu's, but nothing helped.

Hehe... Tricki... I tried this and I got about +2%, but the Phoenix started very-very slowly and comp frozed couple times for 2-3 seconds.
After about 2 minutes everything worked fine, and got this +2%, but looks like to me, that comp doesnt like this trick so much, and 2% not worth it.
But thanks anyway.

So sad..
I already use Phoenix, but Claymore15 work about 7-8% faster for me on other Ethash. Thats why I wanted to use Claymore to mine Ubiq too.
But if so, I must keep using Phoenix.
Thank You for answer.

Didnt see any answer. Can we mine Ubiq with Claymore? If yes, how?

But the funny is, that if I replaced config.json with the config what I used when I mined wow with xmrig_2.13, than the mining started.
Here is config from version 2.13:
{
    "algo": "cryptonight/wow",
    "api": {
        "port": 900,
        "access-token": null,
        "id": null,
        "worker-id": null,
        "ipv6": false,
        "restricted": true
    },
    "asm": true,
    "autosave": true,
    "av": 0,
    "background": false,
    "colors": true,
    "cpu-affinity": null,
    "cpu-priority": null,
    "donate-level": 1,
    "huge-pages": true,
    "hw-aes": null,
    "log-file": null,
    "max-cpu-usage": 100,
    "pools": [
        {
            "url": "*my_pool",
            "user": "*my_wallet",
            "pass": "6000",
            "rig-id": "*my_rigname",
            "nicehash": false,
            "keepalive": false,
            "variant": "wow",
            "enabled": true,
            "tls": false,
            "tls-fingerprint": null
        }
    ],
    "print-time": 60,
    "retries": 5,
    "retry-pause": 5,
    "safe": false,
    "threads": [
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        },
        {
            "low_power_mode": 1,
            "affine_to_cpu": false,
            "asm": true
        }
    ],
    "user-agent": null,
    "watch": false
}

And the new xmrig_2.15 config, where I replaced only the pool address and the wallet:

{
    "algo": "randomx/wow",
    "api": {
        "id": null,
        "worker-id": null
    },
    "http": {
        "enabled": false,
        "host": "127.0.0.1",
        "port": 0,
        "access-token": null,
        "restricted": true
    },
    "asm": true,
    "autosave": true,
    "av": 0,
    "background": false,
    "colors": true,
    "cpu-affinity": null,
    "cpu-priority": null,
    "donate-level": 1,
    "donate-over-proxy": 1,
    "huge-pages": true,
    "hw-aes": null,
    "log-file": null,
    "max-cpu-usage": 100,
    "pools": [
        {
            "url": "*my_pool",
            "user": "*my_wallet",
            "pass": "x",
            "rig-id": null,
            "nicehash": false,
            "keepalive": false,
            "variant": "rx/wow",
            "enabled": true,
            "tls": false,
            "tls-fingerprint": null,
            "daemon": true,
            "daemon-poll-interval": 1000
        }
    ],
    "print-time": 60,
    "retries": 5,
    "retry-pause": 5,
    "safe": false,
    "threads": [
        { "low_power_mode": 1, "affine_to_cpu": 0 },
    ],
    "user-agent": null,
    "syslog": false,
    "watch": true
}


What can be the problem? why the 2.15 version can work with "old" config and can not with original new?
(Of course I tried with 8 threads too, in this new config, but anyway the Hashrate is n/a)

Ok, Im stupid. I realized, that I had this problem, when I started the xmrig_2.15, with my previously config.json from 2.13 version. If I started 2.15, where I set up only "url" and "user" in config.json and changed nothing else, I got n/a H/s. I tried then with added 7 more thread (first used only this 1, what is in original config file), but same situation.
I downloaded then the RandomX-benchmark. When I started this program, I got error message: "Dataset allocation failed".
What should I do? Definitely, I must set up something in config.json, but I dont know what.

I will, but momentarily, Im not near the comp. I use TV, and the overclock is set up in Bios. I tried to lowered the clockspeed with F-Stream Tuning, but this also crashad the comp. I mean, an automatic looff (what I never seen before), and after login no matter what I do, the logoff happened again and again after few seconds.
Win10/64

Hello
My Ryzen 1700 could mine wow/r with 550H/s, at 3,8GHz boost. Same result for Monero, and it working now, too.
But, when I tried to mine this new WOW, the comp crashed after few minutes. The hash till crash was about 3800H/s.
Its good number I think, but why this constantly crash? What should I set up different, to work?

PS I see now, that somebody wrote, the Hashrate jump from 300-400 H/s to 3kH/s. Then, maybe my 3800 isnt so good instead of 550 stable. It should be 4500-5000.

TRM supported WOWnero mining?

Great news! Thanks. What miner do You offer? (rig with 6x Rx550 2G cards)
I mined CN8, but unfortunately, the Asics took controll over this algo again since January.

Hello
Can I mine Callisto with my RX 550 2GB cards?

Hello
Its maybe a stupid question, but can I mine anything with my RX550 2G cards? (Except cryptonight)
So far, I mine Monero, but last Months, the diff jumped up, and I cant be in plus.
My rig with 6 cards need only 250W from the wall, and the cost is only 0,05 usd/kW for me.
At Moneros 40-50B diff, that was ok, because the rig work 2500H/s on CryptonightV8 (2700 was on V7), but now, at 100B diff, I am on Zero.
Can my cards with only 2G memory mine anything else, with a profit or I must just sell them somehow?

Sorry, if its not the right thread for my question. If so, please show me the right thread. Thank You