Miner- Optiminer/Zcash GPU miner v1.7!

[BaldEagle]

i see
was the miner dropping gpu?
it did that for me, that why i moved to claymore and ever since i had not one issue

[dlehenky]

Yes, it was. I’ve found it to be more stable on the fglrx 15.12 drivers, which means Ubuntu 14.04.3 (no higher than that, otherwise you end up with amdgpu-pro drivers). I also run one miner per GPU, which helps. I’ve been using -i = 3 for intensity, which helps stability but doesn’t reduce hash rate, for me.

[ayleph]

@cryptomined

I never saw any actual license in the file itself or online. so if I want I can and will redistribute it.next time include the actual license in the readme

You’re kidding, right? The .tgz files all contained a LICENSE file which stated the following.

Modification or disassembling of the program is disallowed.
Distribution to 3rd parties is not permitted.

It doesn’t get much more explicit than that.

[dallase]

@dlehenky

I also run one miner per GPU, which helps.

Okay, so instead of getting 1 workload and distributing between 6 GPUs, you are getting 6 workloads and making all your GPUs compete for solves?

This sounds very inefficient to me. Tell me I’m wrong?

I was under the impression that anyone running multi-rig farms used proxies, otherwise you take 10% hit easy. Running like this your 6 independant GPUs are like a 6 node farm.

[dlehenky]

It appears you don’t quite get how it works There’s only one workload for a block/transaction. If you have 1 optiminer running 6 GPUs, each GPU is getting exactly the same work - there’s is no “6 workloads”. Whether you run one optiminer per GPU or one optiminer per 6-card rig, the GPUs are doing exactly the same work. There is no difference.

[dlehenky]

Also, why do you think there’s some advantage to a multi-rig farm using a proxy through a single stratum connection, versus each rig or each GPU having it’s own stratum connection? The advantage is the stratum protocol (vs. “get-work”), not the number of stratum connections you use. Btw, the proxy approach introduces a single point of failure (it goes down, the whole farm goes down); it’s not a great idea for a farm. You do have to go that route (proxy) if the farm is solo mining, due to the current limitations of the zcashd client, but definitely not on a pool.

[dallase]

@dlehenky

It appears you don’t quite get how it works

Apparently I shouldnt believe everything I read Thanks for explaining. Cheers!

[cryptomined]

ah ok i see it now, your right

[agis]

Do you know if there is a Zcash proxy available?

[dlehenky]

Hi. Yes, there is, although the binary build has a 2.5% devfee. There is source, so … Just Google “Zcash Solo Mining Proxy” - you’ll find it on github. It’s Sammy007’s work, ported from Ethereum (it used to be free). HTH.

[voxelot]

@agis

Do you know if there is a Zcash proxy available?

Using eXtremal zcash pool is the same thing in many mays, fairly easy to set up and maintain if you use it locally. And it’s free.

GitHub - eXtremal-ik7/pool_frontend_zcash: ZCash mining pool/solo mining suite

Set up your own private stratum pool

Important: this will work with only zcash 1.0.3 So, you will have to use GitHub - zcash/zcash at v1.0.3 when downloading zcash I thought I’d share with you all what I have learned in building a privet (or if you want public) zec pool. I pursued this in hope that maybe if can build my own farm, I will have it work on my own private pool, instead of using a public pool. The core of the pool (and 90% of what I will write) is from eXtremal-ik7 work Zcash Pool Frontend, but I had a lot of help fr…

[dallase]

@dlehenky

It appears you don’t quite get how it works There’s only one workload for a block/transaction.

So does that cause 6x 16GB page files for virtual memory? Because I only have 64GB SSD. I would have to run only 3 then?

optiminer -d 0 -d 1
optiminer -d 2 -d 3
optiminer -d 4 -d 5

[dlehenky]

First, I don’t do Windows, so I can’t speak to the virtual memory usage. I run 6 GPUs on Ubuntu 14.04 linux with 4 GB of RAM, and it never touches VM (no swap space is being used). Yes, you can run them as you have in your post, with 2 GPUs per miner; I‘ve run them that way just fine.

[dallase]

@dlehenky

First, I don’t do Windows, so I can’t speak to the virtual memory usage.

I’m on Ubuntu 16.04, and just launched 6 of them w/ 8GB ram, 64GB SSD. Still have 4GB physical free mem and all swap, and 29GB free disk.

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
12105 root 20 0 16.715g 223920 63884 S 4.0 2.8 1:05.77 optiminer-zcash
12149 root 20 0 16.721g 230164 63976 S 4.0 2.8 1:05.26 optiminer-zcash
12094 root 20 0 16.714g 225424 63848 S 3.7 2.8 1:05.85 optiminer-zcash
12116 root 20 0 16.721g 230156 64004 S 3.7 2.8 1:05.97 optiminer-zcash
12127 root 20 0 16.715g 223472 63896 S 3.7 2.8 1:05.36 optiminer-zcash
12138 root 20 0 16.721g 232204 63944 S 3.7 2.9 1:05.32 optiminer-zcash

Memory…

KiB Mem : 8123484 total, 4469304 free, 2492044 used, 1162136 buff/cache
KiB Swap: 8267772 total, 8267772 free, 0 used. 5167860 avail Mem

Disk…

$ df -h | egrep ‘sda1|Used’
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 51G 21G 29G 42% /

[dlehenky]

Great! So, for future reference, the 16 GB virtual memory recommendations you see for some (e.g. Claymore) miners is specific to Windows; it does not apply to linux for any miner that I’m aware of.

[agis]

@dlehenky, @voxelot
Thank you guys

[Dzonkarlo]

Thanks for coming back Optiminer. I really liked your software and look forward to any improvements you come up with

[agis]

Now that pools have started supporting secure (ssl) stratum connections, the protection process must have become easier.
Can you give us an estimate on when to expect the new version?

[Optiminer]

Public version 1.0.1 just released!

New features:

• Fully async network I/O.
• New logging library that supports max log file size and log rotation.
• New --nodevfee for people that do not like supporting devs.
• zcash-stratum over tls for encrypted connection to pool

Released Linux first as last Windows release did not get much usage.

Next release will focus on increasing the hash rate again.

[Dzonkarlo]

Thank you Optiminer. This is great news. Unfortunately I can’t remotely change from Windows to Linux at the moment. So I will have to wait until I get access to my rigs again or you release a windows binary to give it a go. Once again thank you from all of us.