Topic: Bounty for Open-Sourced XMR/Cryptonight GPU Miner Bounties Thread - page 4. (Read 170943 times)

I got all excited thinking this was the nVidia version.... alas I have no AMD cards to try it on.

Any further news from djm34 about his efforts with the nVidia miner?

https://github.com/wolf9466/wolf-xmr-miner

can anyone determine whether the new version gets over 500 h/s? screenshots pls.

I can't speak for anyone else but my part of the original bounty is still good, so once some people with AMD mining gear have successfuly tested it for regular cryptonight (monero)  I will pay that out to wolf.

The AEON crowd has stepwise instructions to compile Wolf's recent miner in linux.

https://bitcointalksearch.org/topic/m.12668784

obviously, replace the aeon repository with the XMR repository.

None. djm34 hasn't really said much, unfortunately. Last thing he mentioned he said he was trying to figure out how to make it a solo miner, so at that point he was focused on milestone #1...

I see https://forum.getmonero.org/8/funding-required/2355/improved-nvidia-mining-software is fully funded! Good news! Considering you're the fella organising it, any news that you can share with us?

https://forum.getmonero.org/7/open-tasks/2400/open-source-amd-miner-by-wolf0

Wolf has taken up working on the AMD open source miner. I know there were bounty amounts for that specific task, so this FFS (forum funding system) thread can be used.

please go there and comment it up so it'll move to "funding required"

right on! pool or solo??

i REALLY LIKE THIS COIN BEEN MINING WITH MY CPU FOR LAST 4 WEEKS

Bumping the thread in the hopes of attracting more attention to the NVIDIA software development!

With the new software, we could see an increase in hash rate to push the nvidia hardware on par with claymore's AMD miner. This increased hashrate will help secure the network in general, and it will also take more of the hash away from the botnet / sysadmin operators. It's great these guys are securing the network, but at what cost?

furthermore, part of the arrangement with djm34 is to provide some key software architecture improvements independent of hashrate increase. So even if we don't see an increase in hashrate, we will definitely get the following:

1. the ability to solo mine with nvidia GPU. I'm hoping to test this with embedded GPUs as well. that would be awesome.

2. open source, well documented code so we'll hopefully see more development than we've seen with the existing code.

3. much friendlier software - will not require 50 lines of instructions to compile or get working in windows or linux.

It's official - the NVIDIA miner software funding thread is up and running

https://forum.getmonero.org/8/funding-required/2355/improved-nvidia-mining-software

if you want to contribute, just head to that page and click on the "contribute" button, you'll get the address and payment id.

alternatively, here is the address:


and the payment ID


and for your convenience, here's a transfer command with mixin 3 sending 200 xmr for your simplewallet

and we're back

https://forum.getmonero.org/7/open-tasks/2355/improved-nvidia-mining-software

This is from the Monero speculation thread, I didn't want to clutter it up any further, and moved it over here.

Quote from: QuantumQrack on Today at 08:51:47 AM
"Speaking of that, what happened to the mining software initiative?  And what happened to the post?"


GingerAle wrote:
"The initiative is still on, as far as I know. djm34 has been working on it, but he's a little discouraged by the ambiguous nature of the first milestone. I've found where he hangs out on freenode, so I'll try to see what the deal is.

Re the post : unfortunately it got lost / misplaced / sacrificed to the gods of the internet. Fluffypony is aware of the anomaly, but its unfortunately low on his priority list right now. In the meantime, I cloned the most final version of the pitch in the bitcointalk mirror of the open source mining software thread... which is... here:

https://bitcointalksearch.org/topic/m.12094899

if anyone can help make milestone #1 any better, that would help. I tried and tried to get djm34 to step up to the plate to detail his own milestones, prices, etc. But he wanted no part in that. So, I had a field day with it. FYI - milestone #1 is meant to replace an outright advance. djm34 wanted an advance for the work, but no one in the community wanted to pony up the 1000 - 1500 xmr. So, milestone #1 is kind of a secured funding regardless of hashrate increase. So djm34 would get something, and the monero infrastructure would get something (a cleaner, rebuilt open source nvidia mining package)

And I speculate bananas."

So, I wanted to reply to this.  To actually get funded and get the project into the funding stage and working stage, the programmer needs to put forth some information on what he will be doing, etc.  I don't think this is unreasonable.  So, my thoughts are that he/she should put forth the effort to edit the current goals/milestones that makes sense to him since he will be doing the work.  Its easy, the basics could probably be edited/written out in an hour or so.  If he is looking for 1500-2000 XMR up front before anything has been done....?  My inclination on that would be no.

Thank you for the instructions it is much appreciated.

ah ok. Thank you. I was just following the thread in order to get news regarding the new miner )

I assume you're talking about tsivs miner. I've found you have to end up having all of the below nonsense for it to work - probably some library dependency. So I could post my binary, but i've copied binaries from one linux box to another and it didn't work (forget why), so I just ended up following these from start to finish.

step 0: set up your miner box to run headless via SSH. This is the best way. You might be able to pull off using your video card and mining at the same time, but I wouldn't recommend it.

First, install an nvidia driver. I've found 331 works the best for a 750 ti.

http://www.noobslab.com/2014/12/use-nvidia-graphics-drivers-in.html

Then, go through this nightmare to install the CUDA nonsense UP TO THE INSTALL CUDA MINER PART.

https://d3adbra1n.wordpress.com/2014/05/03/cuda-miner-installation-on-a-fresh-ubuntu-14-04-lts/

Tip: when the long ass agreement thing pops up, you can hit ctrl-c to skip straight to typing accept. I didn't know this for a long time and would just put a weight on my down arrow to scroll through the thing. Or whatever the scroll was.

When you get to step 31, do this instead:

git clone git://github.com/tsiv/ccminer-cryptonight

that'll plop it in /ccminer-cryptonight/

then pick it up from step 35.

A reboot is required to get your video driver working (from my experience)

Could someone please release a compiled version of this miner for Linux. Or point me towards instructions for Linux noobs that explains how to compile this.

fluffypony recently was doing some server maintenance

is there actually a reason why the thread disapppeared from the official forum at getmonero.org?

Okay, with the help of chocolatebar, palaxender, and myagui, these milestones are the current working set:

Primary Milestones:
1) Overall software architecture (1000-1500 XMR):

A) Open source software that is nice, clean, well written and documented code that will be easy to maintain and work on; a pleasure for others to jump in and contribute going forward.

B) Cross-platform compatible. e.g. works on apple, linux, microsoft. Binaries available for all platforms when code is released.

C) Parameters can be completely tuned. The software should be designed so that miners can change every value with a config file to tweak settings for their particular hardware. This will possible extend the useful life of the software.

D) Limited dependencies for compiling, and required dependencies must be available via package managers. 100 xmr will be deducted for each dependency required, max 500.

E) Works on as much available hardware as possible. All Maxwell architecture is a must.

F) Works as a solo miner - is able to communicate directly with the monero wallet software.

2) Mining performance (up to 3500 XMR)

A) Hash rates reach the price point equivalent of Claymore's AMD GPU miner. Currently the highest reported hashrates are 745 h/s for the Sapphire R9 290X, which at the time of writing is a $350 GPU. The NVIDIA pricepoint equivalent is the GTX 970. Existing hashrate of a GTX 970 is ~400 h/s.

The exponential function would be payout = (hashrate increase)^(1/29.3) , which would produce an exponential curve to exactly 3587 XMR for 745 h/s. I don't think this curve is attractive, because a hashrate of +100 would only reward ~11 xmr. + 200 = 113, +300 = 1208,

Okay, I came up with a nice looking curve.

payout = ( 1.2 ^ ((hashrate increase) / 17.7) ) * 100

but hashrate increase >= 50.

so the minimum is 169 XMR

    100 h/s, ~280 XMR

    200, 784 XMR

    250, 1313 XMR

etc. If you can come up with a better curve, please do! My excel skills got boring.

Secondary Milestones

(these will only be funded if the primary milestone funding levels are met)
3) GUI - 200 XMR

A GUI for all available platforms.
4) Longterm support - 1000 XMR (or however much is leftover after primary milestones)

20 XMR for a pull request that results in > 10 % increase in singular hardware (one card)
100 XMR for a pull request that results in > 10% increase in at least 5 cards
100 XMR for adding compatibility with new hardware (e.g., the code didn't work on newly release hardware at all, the pull request makes it work)

This still isn't set in stone. Please provide input.

The rationale for all this is over here:
https://forum.getmonero.org/7/open-tasks/323/open-source-gpu-mining-software-hashrate-increase-fund-bounty-hire-a-dev-whatever

Also, claymore never responded to my question re: how much it would cost to make his miner open source.

edit #1: include solo mining function in first milestone, reworded sentence using tunable so it doesn't get flagged by spellchecker, because apparently tunable isn't a word.