Topic: [ANN] Memorycoin ★ Modern CPU PoW ★ COMPLETELY GPU/ASIC FREE ★ NO Free PoS - page 8. (Read 39215 times)

Hi, in2tactics!

I'll skip all non-constructive sentences and will try to be more constructive

Your are late dude, it's already MMC3 - look at sources I can change your large stake of mmc2 to mmc3 at good rate.

Oh, really? And how can you measure this percents? Please, provide us this great formula.

I agree. But, I don't understand what you mean in "coordinate our pools and properly bootstrap the sharechain" ?
I already have such node and I don't think that mmc blockchain has a lack of nodes.
About KillerByte's p2pool code - it's works and worked always. What do you want to change?
About grant block problem - this is still exist and can be eliminated (maybe I'm wrong) only by eliminating a "grant block" in sources. Moreover, it is unclear to whom these "grant blocks" should belong now?

I don't think so. Do you see any problems with wallet synchronization or lack of nodes? I can't see such problems.

Yam works without any hacks now. The only issue is a greed of his author, who hard-coded nodes (dead now) and closed YAM sources.
So, one should restart it every 8 hours or ~400 shares. A very simple "hack" with a scheduler. 

I do not claim to originality, but I want to remind you of the words of a great man - Linus Torvalds:
1. "Software is like sex; it's better when it's free."
2. "I don't ask for money. I don't ask for sexual favors. I don't ask for access to the hardware you design and sell. I just ask for the thing I gave you: source code that I can use myself."

I hope you will understand me and will not argue with me on this issue.

Cheers.

yes, correct.

@mycoinsstore  I am not trying to tell you what to do. You can use whatever you want. If you want to GPU mine, I would be more than happy to try and help you figure it out. I would like to note that I have been unable to locate the source code for any of the GPU miners as I believe all of them were closed source. Let me know if you have any of their source code.

@mycoinsstore  I believe what coaex is trying to say is that he never intended to definitively say GPU mining is no longer possible.

@coaex  Would my assertion be correct?

@coaex  Yes, I have been saying this all along. GPU mining stopped because it was no longer energy efficient when compared to CPU mining with YAM. You are correct, Kimoto Gravity Well (KGW) is still used. I found where Delinquency fixed the critical bug in the source code.

@mycoinsstore  To answer your first question, I have a large stake in MMC2. I have been mining since the beginning of 2014 and I have been following the project since mid-2013. I also currently represent ~40% of the MMC2 network hashrate and I can easily increase my hashrate by at least 300%. I think that should answer your remaining questions about my interest in MMC2 and whether or not I am here to argue.

Now, can we move on to a more productive conversation?

I would like to get p2pool working again, but we need to coordinate our pools and properly bootstrap the sharechain. I am prepared to host a node that will function as a bootstrap address. I think that I can fork KillerByte's p2pool code and make the necessary changes to get things working. However, one problem might still remain. Did Delinquency and KillerByte ever solve the grant block issue with regard to p2pool? I am still looking over both Delinquency's and KillerByte's work to determine if the problem still exists, but I have been unsuccessful so far in identifying if it was actually resolved.

I would also like to get the wallet updated. At a minimum, we need to update the seed nodes so that new folks can sync their blockchains without manually adding nodes. We will also need to compile new binaries for both Windows and OS X. I think that we need to put-off any discussion of updating the wallet to current Bitcoin sources.

The final issue we have is with YAM. After we get p2pool back on track, we need to contact yvg1900 and request that he update our p2pool addresses in his next update. I have a working hack that enables me to use YAM now, but I would prefer to get it properly fixed so that yvg1900 gets his 2%.

Any thoughts?

You may think that the intrinsic value of Memorycoin is way too high than current level (~1800 sat) because of these HPM performances of GPUs.
I thought the same in the past but it just went from 800 sat (where I joined the team) to 1800 sat. (while BTC has also increased ~85% in that period - so basically x4 increase in fiat)
I didn't accurately calculated but it seems that the profitability level should still be at least 10x from current level now since rapid mining phase is ended and only a small number of coins are being issued but you know, profitability (in other words, difficulty and coin supply) is not the only factor that affects the actual price. The original design was to issue 10 million coins in the first 2 years + 100,000 coins in each year with a hardfork. The coin supply was also problematic in FT's code, I guess delinquency had fixed that, as well.
So, in this case, what will happen? I simply don't know as there is no real developer around but what I know is that I don't consider making an effort for MMC at the moment unless a dedicated developer joins in.

oh guys, calm down. I just shared my experiences. I'd seen that MMC was GPU mineable before delinquency took over.
afterwards, he forked to v3 and implemented below. Afterwards, I haven't seen/heard any GPU miners. There were actually a few miners just to keep the blockchain healthy.


I'm not a coder, nor a code-analyst. Maybe, Memorycoin just changed into a real CPU/RAM-dependant coin after AES-CBC implementation.
I also remember that difficulty readjustment has changed from Kimoto Gravity Well to Dark Gravity Wave but I'm not sure of that, I might be mistaken.
So far I remember, I was getting 10-11 HPM with i5-2500k 3.3Ghz (~125W) while a strong GPU in 2015 was getting 20-22 HPM (~300W). Maybe that would be reason why GPU miners were not around. Settings I remember - yam-miner sandy bridge - hugepages on

edit: oh, that DGW has never been implemented, sorry. I should have confused it with something else.

Why not debating? Or you think that this forum is only for your "facts"?
You can claim anything you want. If you claim that it is possible, prove it.
Let me decide what more efficient for me, ok?

So, you mean to say that coaex is a liar? He confirmed that Delinquency did GPU mining impossible.
He's an old, respected member of the forum. I believe him, and you're not? Do you have any evidence that he told a lie? So, who spreading FUD here?

I'm an mmc bagholder, and who you are?
I offered to community to run mmc pool (if someone needs this), and what is your interest in this thread?
And the main question - are you interested in the future of mmc or just come to argue?

What is the point of debating GPU mining for MMC2? I previously stated only that GPU mining is possible, not that it should be done. CPU mining with YAM is far more efficient hashes/W than GPU mining ever was for MMC2. That is an actual fact.

I have looked through the source code and I am saying that your claims are incorrect. Show me in the source code where I am wrong or stop spread FUD.

I am positive that Delinquency is not a liar. Show me in the source code where I am wrong. Correlation does not imply causation!

I was there too and I remember well what did and what did not work. There are other possible reasons for things to not be working correctly. Without seeing your error messages and knowing your setup, it is too hard to conjecture.

Hi, in2tactics !

Unfortunately I have no time to go so deep into the mmc sources, but I believe to coaex and my own eyes.
GPU mining stopped working exactly after one of the Delinquency mandatory mmc code and wallet updates.

And this Delinquency's post is not true:
Pools discontinued accepting shares for Memorycoin AFTER changes to mmc code and wallet, not before.
Ask yourself - why should pools kill GPU mining of mmc? They had a good fees from this mining, because of a big amount of GPU miners.

Also, GPU miners supported not only pool mining, but solo mining too. I mined mmc with my 2 R9 270x 2GB cards, using mc2miner and other miners (that supports GPU and CPU mining). But now, even solo GPU and CPU mining not working. Only yam miner working now. The question is why?  

Btw: I remember time when GPU miners worked even with p2pool mmc pools. I have a working code of such p2pool, but it no longer accept shares from GPU miners.

I spent the last few days going over all of the commits and merges made to the source code by Delinquency on GitHub. I went as far back as December 2013 and I found no changes to the algorithm or more specifically no changes to the implementation of momentum. I did find changes to the protocol with respect to difficulty re-targeting and fork handling. Delinquency also added copyright notices and changed bitcoin to memorycoin in all variables, functions, and file names. Additionally, I found significant updates to the Qt interface.

Have you even gone through the source code?

https://github.com/memorycoin/memorycoin
https://github.com/Delinquency/memorycoin

I also found the following post from Delinquency where he failed to confirm that GPU mining is no longer possible. He only stated that the pools supporting GPU mining for MMC2 discontinued it.

Now back to my original post...

I think that the facts I have previously mentioned vindicate my original statement.

Additionally, I would like to add the following as further evidence concerning YAM's performance.

Catch it - last wallet+full synced blockchain - https://mega.nz/#!Dc9hUJiQ!HDRDpah_FiF_3bbweTHdIjdoAPMrtTHIJ7-yflB4lfI
Only add your own memorycoin.conf file.

I also need windows version of memorycoin wallet. Can you upload it to help us? Thank you very much

I agree.
But another problem is a huge amount of ?premine? Or I wonder what is it in this 5 wallets:
1. MVgvtTcrX7A7pW96su5ipfGMReE1f5DzKS - 4,060,000.001 MMC 
2. MJXcKaNuJMesxAn1vpC2gdHtcF5UEa46H5 - 371,315.73 MMC
3. M9iSxGQaqCjnnhSEmAQdJjPcaQFH9Nbk4N - 236,481.67666435 MMC
4. MFWwzCpZMQHttNUEqFmGxG3mYeKqUNadqm - 229,023.6917106 MMC
5. MMC1337wCzx1gFt29htCzEBVQ21pKRn4Mb - 204,385.14813958 MMC

It's a half of all mmc!

i am sure you are correct although i am personally not an expert (just a noob) on coding gpu miners

still there should be nothing stopping some individual with the proper skills to fork the code of mmc2
and develop their own gpu miner

I cannot imagine it would be a radical jume in efficient mining but with closed source one never knows....

Honestly I was quite interested in mmc2 at the time it was launched although I normally don't like asic resistant algorithms that use (some people say 'waste') a lot of memory.......

i kind of lost interest when I knew I was solo mining or mining at a few hashes per minute . basically a rubberband slingshot and some others had gunpowder !!

my work on other projects leaves me no time to volunteer time to develop for mmc2

my involvement was for my own learning only still mmc2 imo introduced many good features like the voting ect.... that while others discussed in concept , never deployed them to an experimental coin like mmc2

gotta give credit where due

Believe me, mmc code was changed and GPU mining was blocked. You can see this changes in mmc sources.

I know there was a code change when MemoryCoin Beta and ProtoShares transitioned to MemoryCoin 2.0 (MMC2). After Delinquency took over MMC2, I remember a hard fork sometime in the second half of 2014, but I did not realize that was a major algorithm change. I also remember dwarfpool.com and 1gh.com delisting MMC2 around that same time-frame.

It was never my understanding that GPU mining was dead due to code changes, but I could be wrong. I just remember the GPU mining discussion dying when yvg1900 released YAM with MMC2 support.

This is a great advantage of this coin :Momentum (64K XOR AES-CBC x 50) with SHA512 Generation
Security high level.

Code was changed. This is a fact.
Exactly after that all major pools (mmcpool.com and 1gh.com) delisted mmc.
And people lost interest to mmc, because ask yourself - how many people has Xeon servers?
Also, yam-miner main problem is a proprietary closed-source code with hardcoded (dead now) pools.

Yes, but this supply reduction, GPU mining impossibility and CPU mining difficulty leaded to nearly death of mmc
Solo mining possible only on low difficulty. If it raises to 0.00000010, for example, it will be impossible and not economically viable to people to mine mmc.
So, why new people should be interested in mmc?

Delinquency did that and I still think it's brilliant. By the way, Memorycoin was originally designed to be GPU/ASIC resistant but it simply failed in that part.
GPUs along with botnets hit this coin badly and made it depreciated. Therefore, FreeTrade abandoned, Delinquency took over and designed the coin impossible to be mined by GPUs/ASICs.
As the coin supply is reduced dramatically couple of months ago (again based on original design), all the coins in the market should have high intrinsic value. In other words, you have a great chance to resurrect the coin now as the base is stable and you guys can make a difference and make it appreciated.

The coding for MMC2 was not changed. If I remember my early months of 2014 correctly, the Radeon HD 7970 or Radeon R9 280X produced ~20HPM at around 350W. The i7-4770K initially produced ~3HPM at less than 90W. When yvg1900 released YAM, the i7-4770K jumped to ~13HPM. GPU miners simply could not compete with the power efficiency of CPU miners when they are between 2 and 3 times more efficient. AES-NI FTW!!! On a similar note, it is normal to see a dual socket Xeon servers pushing 40 HPM when using YAM. The problem right now is that all the default pools are down and YAM stops mining when it is unable to submit shares for the developer.

Yes.
But some of this clever guys (I don't know who exactly), was able to modify original mmc code to make it impossible (now) mine mmc with GPU.