Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Regarding Cryptonight ASICs #1

Open
sergneo opened this issue Mar 25, 2018 · 20 comments
Open

Regarding Cryptonight ASICs #1

sergneo opened this issue Mar 25, 2018 · 20 comments

Comments

@sergneo
Copy link

sergneo commented Mar 25, 2018

https://www.reddit.com/r/sumokoin/comments/83w61w/regarding_cryptonight_asics/
will you update your version of xmrig with the upcoming fork where Monero and Sumokoin are going to update the algorithm, I dig SUMO I am relevant this coin for example.

@2010phenix
Copy link

xmrig already say that add support after Sumo move to cryptonight_heavy

@Foudge
Copy link
Owner

Foudge commented Mar 28, 2018

I will merge with xmrig branch and redo some optimization for Core2 if necessary.
The PoW algorithm update is scheduled for April 6. Not sure I'll be on time.

@sergneo
Copy link
Author

sergneo commented Mar 29, 2018

@Foudge I mining Sumokoin , you'll be able to enable PoW SUMO ? Date fork has not been determined, but promise after Monero.

@Foudge
Copy link
Owner

Foudge commented Mar 29, 2018

I don't know. If the difference is like between cryptonight and cryptonight-lite, it will be easy.
You have a link about the algorithm PoW evolution?

@sergneo
Copy link
Author

sergneo commented Mar 30, 2018

https://twitter.com/sumokoin/status/979728620125085696
Fork Block Height set at 117600 (on abt April 4th). Binaries will be released soon. Following the fork the light Qt GUI wallet will be released featuring remote node connection ability (it is ready for release). Web wallet is expected to be released soon as well.

xmrig#476 (comment)

@2010phenix
Copy link

2010phenix commented Mar 30, 2018

@ Foudge this is his(SUMO) new algo repo: https://github.com/curie-kief/pow_test/commits/master

and how work: https://github.com/curie-kief/cryptonote-heavy-design

@sergneo
Copy link
Author

sergneo commented Apr 1, 2018

SUMO Testnet pool:
http://92.53.67.18
Testnet wallets:
Sutoe6DuhQJVR2RWKDWtC3YWGaTVeemBZVUYizBirBzK6ecaock4LNyZD6QUYy9Fwvb5MSD6VLFGiHijRhU4AdbzCc2j355X8Yv

@sergneo
Copy link
Author

sergneo commented Apr 3, 2018

@Foudge https://github.com/xmrig/xmrig/releases/tag/v2.6.0-beta1
xmrig has just released a new version with Added Cryptonight-Heavy support for Sumokoin ASIC resistance fork.
You can now take the basis to update the optimization Core2 Duo, Quad as in your version of F4 ?

@sergneo
Copy link
Author

sergneo commented Apr 4, 2018

@sergneo
Copy link
Author

sergneo commented Apr 5, 2018

@Foudge fork is done, miners are working, can now optimize the code for core 2 duo ?

@sergneo
Copy link
Author

sergneo commented Apr 9, 2018

@Foudge where did you go?

@sergneo
Copy link
Author

sergneo commented Apr 15, 2018

@Foudge it seems the miner v2.6.0-beta2 is not very optimized for Core2 Duo Quad especially under Xeon.
cryptonight-heavy issues at E5440 43.5 H / s on 4 threads , on 3 threads 35 H/s. And this despite the fact that the cache L2 12Mb. Especially on a Quad with a 6Mb L2 on 3 threads gives more than 2. 22H/s vs 14H/s.

@sergneo
Copy link
Author

sergneo commented Apr 16, 2018

@Foudge But even more optimization is needed Monero v7

@sergneo
Copy link
Author

sergneo commented Apr 20, 2018

xmrig has already released a version https://github.com/xmrig/xmrig/releases/tag/v2.6.0-beta3 Oh. Please add to your version of pow Monero V7. Your version for no AES faster.

@Foudge
Copy link
Owner

Foudge commented Apr 20, 2018

I have a bad news. I stopped mining with my CPU and stopped develop this project. I started the merge but it's not functionnal and need work and time to do all the necessary.

But if you have a non-AES CPU, all is not lost. I discovered 3 weeks ago another miner : JCE Cryptonote CPU Miner
It is closed source, have fee, but it is really fast, and faster than my version of XMRig on this processors!
Besides it is up-to-date, support many algo variations and many coins and developper is very active.
Github : https://github.com/jceminer/cn_cpu_miner
Bitcointalk : https://bitcointalk.org/index.php?topic=3281187.0

I already used it, it works very well! I recommand it.

@Foudge
Copy link
Owner

Foudge commented Apr 20, 2018

In cryptomining, I will maybe develop algo rewriting on GPU (maybe an another XMRig fork). The idea is to use an API faster than OpenCL or CUDA : Vulkan
But how many time to control GPU programing and this new API? What is the gain to be hoped for? I don't know. If it's just 10% boost, no need to spend tens of hours on this project. So not sur at all if this project will born, but it's certain that development on CPU is over for me.

@sergneo
Copy link
Author

sergneo commented Apr 21, 2018

This is of course very sorry that you stopped work :( If you would have released at least one version after the fork would be good. JCE Cryptonote CPU Miner may be faster, but I have 100 computers are on the local network and do not have direct access to the Internet, ie this miner is not for me it does not work for me. The developer did something stupid. He doesn't have the http monitor, but I really need it.

@2010phenix
Copy link

JCE Cryptonote CPU Miner and not even start on old Windows x32

@sergneo
Copy link
Author

sergneo commented Apr 21, 2018

@2010phenix no http monitor, be sure to direct Internet access to the donation pool, 3% for the old CPU. Private code. Miner has no future, the developer will not work on it much. Reject share Minergate.

@sergneo
Copy link
Author

sergneo commented Apr 24, 2018

@Foudge please, could you combine your F4 version with the latest xmrig 2.5.3 version ? For your miner to work at least after the fork with Monero V7. Respond.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants