Critical Process

[gamethrower]

What is the downsides of using "BSoD if killed"? Would be nice to have this in miner.

[UnamSanctam]

Well, means you can't uninstall it mainly.

[gamethrower]

Is this the only reason?

[UnamSanctam]

Problem is also that it will BSoD if the program crashes or just exits normally as well, if the program closes in any way it will BSoD.

[gamethrower]

Even when shutting down the PC?

[UnamSanctam]

Don't think so, but if you input the wrong connection settings (or lose internet connection) in the miner then it will eventually close (tries for 5 minutes I think or something), and then you would BSoD since it will close down.

[gamethrower]

Yeah, now I see that this would be a little bit rude to have such feature :rofl:

[UnamSanctam]

Haha yeah, I did experiment a bit with making the process a debugging process which means you can't close the process easily, I did have some success but have to test more. I'll also be releasing a new version of the XMR and ETH miners today or tomorrow mainly focused on decreasing detections.

[gamethrower]

How do I modify the code to make the miner start only after reboot?

[UnamSanctam]

Remove this line: https://github.com/UnamSanctam/SilentXMRMiner/blob/31992cf2d9f9dbecde1592f8c31668eca1d64b83/SilentXMRMiner/Resources/Program.cs#L100

[gamethrower]

Thanks

[gamethrower]

Can you explain in detail what this xmrig arguments do? --cpu-memory-pool=1 --randomx-no-rdmsr

[UnamSanctam]

The --cpu-memory-pool is how many 2MB pages it should do for the persistent memory pool, though in the new version you can remove that since it's automatic. --randomx-no-rdmsr stops it from reverting the MSR register values when the miner exits.

[gamethrower]

in the new version you can remove that since it's automatic

Do you mean the new version of xmrig or your miner builder?

[UnamSanctam]

In the new version of XMRig (that my miner also uses).

[gamethrower]

Is this the "ok" amount of shares rejected with "Block expired" reason or I should reduce latency?

[UnamSanctam]

That's a 0.5% invalid share amount so it's alright. But of course if you can reduce the latency (to reduce the amount of expired block shares) that would be better, you can also try other pools since some pools have things to circumvent that.

[gamethrower]

Is it even worth to implement latency checker in miner and have a few servers for eu, asia and so on?

[UnamSanctam]

Not really, it's such a small amount that it doesn't really matter but if you still want to lower it then you can just change pool until you find one which works better.

[gamethrower]

It's not about pool but server that hosts proxy. I'm currently using EU based VPS and half of my workers are asian, so their latency sometimes reaches 500-1000 ms.

[UnamSanctam]

Yes but that also depends on the pool. Depending on the pool I can get between 5% stale shares or 0% stale shares.

[gamethrower]

I'm mining to nanopool. They are holding my XMR below withdraw limit as a hostage, so I can't switch to other pool :sob:

[UnamSanctam]

Haha yeah I know that feeling. After you get those you can try some other pool and see if you get any difference.

[gamethrower]

When I test miner on my PC (3 threads) with 70% CPU load it uses 1 threads instead of 2. Before I launch it CPU usage doesn't go above 10%, so I guess 2 threads should be available. Why it uses 33% instead of 66%?

[UnamSanctam]

Could be that only 1 core is idle.

[gamethrower]

How can I check what cores are idle? Win7 task manager doesn't show that. Here's what it shows when I start the miner:

[UnamSanctam]

Use something like Process Hacker instead to see the CPU usage, it looks like all your cores are currently being used there at least.

[gamethrower]

With miner off:

[UnamSanctam]

And with the miner on? One core could also be reserved so only 2 might be available, and then of those two only one could be idle which would lead it to only mine on 1 core (since only one is free).

[gamethrower]

Is this the default max-threads-hint behavior or you made some changes in xmrig?

[UnamSanctam]

The default max-threads-hint behavior, I have not made any changes to it (since none is required).

[gamethrower]

The problem is it doesn't use more CPU in Idle Mode because of this. It would be nice if miner used -t at least for Idle Mode.

[UnamSanctam]

Well that's only because of your bad CPU, on a real computer it will work fine, for me I have Idle mode on 80% and it uses 6 out of my 8 cores which is correct.

[gamethrower]

How do I know if MSR works for me? When I launch xmrig it doesn't say anything.

[UnamSanctam]

If you mean the normal xmrig then it will say whether it set the MSR registers correctly or if it didn't have enough permissions.

[gamethrower]

It doesn't say anything about MSR even if I delete the sys file.

[UnamSanctam]

Take a screenshot.

[gamethrower]

[UnamSanctam]

Because you don't have enough RAM to store the RandomX dataset, but it does have permissions as you can see at the top though it doesn't matter since you don't have enough RAM.

[gamethrower]

I'm just trying to figure out something. If I drop the WinRing in some location and then change the path here to that location, will it find the driver?

[UnamSanctam]

Yes, that's what I do in my miner.

[gamethrower]

After a few days of running proxy it started to reject the connections. Debug info doesn't say anything. When I restarted the proxy amount of upstream dropped to one and amount of miners increased. I was able to mine to the proxy again. What was that and why upstreams are increasing over time even tho I don't have a lot of miners online(less than 256)?

[UnamSanctam]

Because that's how xmrig-proxy works, over time some of those upstream slots will be filled up due to the nature of TCP connections That's no problem with the miners but just how xmrig-proxy works by design. What are you hosting it on? If you are hosting it on Linux then you might have to increase the open files limit.

[gamethrower]

I have checked the default limit and it's 1024. Doesn't it mean 1024 constant connections?

[UnamSanctam]

Technically yes, but since it's TCP connections and not UDP connections then over time not really. You should increase the limits.

[gamethrower]

Oh, that explains it. Thanks!

[gamethrower]

Will the next Monero hard fork break all miners?

[UnamSanctam]

All hard-forks that change the algorithm do that, since they change the algorithm, unless they hard-fork to an algorithm that the miner already has.

[UnamSanctam]

But it doesn't really matter if you use "Remote Configuration" since you can just mine any of the hundreds of other cryptocurrencies supported by the miner.

[gamethrower]

So if RandomX algo doesn't change everything should be fine. And Monero devs have no reason to change it, right?

[UnamSanctam]

Yes as long as the algorithm stays it will work, and will also work if they ever change it if the algorithm already exists in my miner. But yes I do not think there is any reason to change it at the moment, even more cryptocurrencies have also started using variations of RandomX as well.