Thanks; good to know that. At the first look, pretty the cpuminer style; guess it will be a quick job. Hope that works the way I think that straight. |
|
|
|
# Two CPU miners are available now and already done testing locally; both ported from M7 minichain miners: XCN and 1GH's work. # Cudaminer (1GH) is ported too, while this is only my "best effort" at this time, never tested, because I don't have any Nvidia cards available for testing. (I have a mac which has Nvidia GeForce 9400M, but I stopped after few tries, not successful and no much time to research further; anyone has experience on that please help me a guideline). Though it has spent me quite a lot time, I pretty learned from the way of revising 1GH's cpuminer in order to adapt to XMG, and hence here the Cudaminer. Compiling is doing pretty fine in my LM machine. Based on my experience my impression is AMD miner could be probably done too (correct me if I am wrong; also not much time to spend on this any more, will get back). # djm34's ccminer is yet to be done since source code is not available. As you can see, I've pretty much picked suggestions from the community. I even didn't expect the Cudaminer to work, but now seems like we made it (need to confirm though). So please leave any suggestions; I'll keep that in mind. Some figures about the distribution of mining rewards (PoW and PoS-II interest) will be issued after releasing testnet. I am open to suggestions; this will also be our "last" chance to modify before formal launch. For me, these figures are quite important in order to suppress network diff. I hope the 2nd testnet launch would occur within two days (or likely more) from now.
|
|
|
|
You are certainly a good developer!  Keep up the good work! Nice to know, and hopefully be experienced over time.  |
|
|
|
This is going to be the best coin ever. I love how you are making it in testnet first. Real innovation here. Quality developers.
Thanks for the note. It was a good decision that we delayed the launch so now we can test it out fully. Though have been doing coin stuffs a while, I am not a serious c++ coder (not a real CS person); but sort of coding using fortran usually, my real work is in the scientific research field, kinda "curious" sometimes and keen to "investigation/thinking style".  (English is not my native language sorry for any misleading in my words if there is any) |
|
|
|
Updates:
Standalone CPU miner is ready for the 2nd test; also try to manage the data format revision in JSON API call so that all M7 miners can be immediately available to XMG. That way cuda and ccminer may work too. Stay tuned.
what do you mean "manage" the data format ? So you are changing everything ? I was wrong, no way of changing there, unless we revise the block header, so I am now porting the M7 miners into XMG. |
|
|
|
|
I can see someone is still running the old testnet wallet (Ip: **.80.**.25*). Could you please quite the wallet so we can have an easy up?
|
|
|
|
I can test a small amount of mining, if valid, I can be a lot E3 mining, electricity is expensive
Thanks, will release the 2nd testnet wallets ASAP. |
|
|
|
Test again. Mining invalid?
We had a test in the testnet few days ago which turned out to be an issue in the network sync. The problem is about the PoS-II mining. The PoS-II block resists to sync somehow. To be safe, we will run public tests one more time (hopefully no more in the future) after I run more tests in my vps nodes. And I wish I could manage to get cuda and ccminer tests too. |
|
|
|
|
Updates:
Standalone CPU miner is ready for the 2nd test; also try to manage the data format revision in JSON API call so that all M7 miners can be immediately available to XMG (turned out that only possible if we change the block header). That way cuda and ccminer may work too. (will port the miners then, kinda "guess" woring on cuda since I don't have Nvidia available for test) Stay tuned.
|
|
|
|
Work on the coin, no updates yet; back on work tomorrow. |
|
|
|
I didn't try that, but I guess no (need confirmation though) |
|
|
|
|
Thanks all for reporting the issue. This is something I've seen before. My prior solution seems mitigate the issue until now; I will use a different approach to get around the PoS block stuck and make a fix. You can shut down the testnet wallet for now. Once I am done correction, we'll need another run of test.
|
|
|
|
Interesting. Will look into helping with this. Looks good so far.
Take a look at this: https://bitcointalk.org/index.php?topic=735170.msg8453963#msg8453963I hope this coin can attract community's interests. BTW, I've seen people are now very concern who is the developer behind the coin, looks like we've been through a hard time with many scams out there (Just image people who can do anon can make many coins, the 2nd, 3rd coin ... just catch people's eyes simply because of the anon, cann't figure why people are keen to this). If you all want to assure who I am, I could give more details as I can in the following days (since someone suggested the PoD previously). |
|
|
|
Mined 44 blocks, 1080 coins, the only problem right at the end:  Don't know what happened at 00:09, looks like two orphaned blocks. Turning it off until tomorrow. The one with 10 rewards is PoW mining, we can wait for a longer time to see confirmation or not. 0.649x is PoS mining actually. |
|
|
|
dev, there are no tools GUP AMD's launch?
This will be the second priority (Nvidia could be a more probable one), not sure when we could get into that, I have to say hardly possible to have it on the launch. |
|
|
|
oh my gosh, this was a big mistake I've made; I am surprised people in prior posts didn't take this mistake, obviously they did the debug or just gave up at 1st try . I should directly copy from my working conf.  (did the correction btw) It's important that we've found it. You should update the OP, maybe put some info about downloads and basic config at the bottom of that post to be easy to find it. I'm willing to test Nvidia GPU miner on testnet when it's available. Will ccMiner work with M7 and Magi? https://bitcointalk.org/index.php?topic=167229.0Let me know if I could help with testing something else. Thanks for this. I've updated the OP. I will update the miner (cpu) first; unfortunately gpuminer won't be available immediately (code tweaking to adapt to magi is needed; will get back on this once everything else is accomplished). The implementation of M7 in Magi has proven to be much different from Cryptonote minichain since Magi is standard blockchain based. Currently I am trying to locate potential issues affecting the network sync problem, will update what I am getting and the next step towards the launch. Any updates on the PoW mining and PoS rewards will be a great help. |
|
|
|
oh my gosh, this was a big mistake I've made; I am surprised people in prior posts didn't take this mistake, obviously they did the debug or just gave up at 1st try . I should directly copy from my working conf. (did the correction btw) |
|
|
|
Moved the executable to the same folder just to be sure, the same error: itod@ubuntu:~/.magi$ ls -al
total 5296
drwxrwxr-x 3 itod itod 4096 Aug 20 16:54 .
drwxr-xr-x 55 itod itod 4096 Aug 20 16:45 ..
-rw------- 1 itod itod 16384 Aug 20 16:46 blkindex.dat
drwx------ 2 itod itod 4096 Aug 20 16:45 database
-rw------- 1 itod itod 0 Aug 20 16:45 db.log
-rw------- 1 itod itod 6335 Aug 20 16:55 debug.log
-rw------- 1 itod itod 0 Aug 20 16:45 .lock
-rw-rw-rw- 1 itod itod 91 Aug 20 16:51 magi.conf
-rwxrwxr-x 1 itod itod 5377128 Aug 20 16:32 ubuntu_14_magi-qt_v1.0.0.1
This is strange; the debug.log and the contents in the folder all indicate the wallet is not running on the testnet (a folder testnet2 should be in ./magi once it does). Only the conf file matters basically. From my side your settings are quite okay. We may try indicating the specific conf, with option (try qt and the daemon too (magid)): -conf=/home/itod/.magi/magi.conf. If you still have problems, I can setup a ubuntu vm (away from the machine I compiled qt) and try it out here. Let me know. Thx. |
|
|
|
Good to know Now I get this: ./ubuntu_14_magi-qt_v1.0.0.1
ubuntu_14_magi-qt_v1.0.0.1: src/main.cpp:2660: bool LoadBlockIndex(bool): Assertion `block.GetHash() == (!fTestNet ? hashGenesisBlock : hashGenesisBlockTestNet)' failed.
Aborted (core dumped)
Contents of magi.conf: tesetnet = 1
daemon=1
server=1
rpcallowip=127.0.0.1
rpcuser=rpcuser
rpcpassword=randompassword
having the magi.conf in ~/.magi, the wallet should be good to go. Can you make sure tesetnet=1 is set and magi.conf in the right folder? |
|
|
|
Tested on Ubuntu 14.04 64bit, I get this: ./ubuntu_14_magi-qt_v1.0.0.1: error while loading shared libraries: libdb_cxx-5.1.so: cannot open shared object file: No such file or directory
Any quick fix for that? Edit2: solved: sudo apt-get install libdb5.1++ Good to know |
|
|
|
|
Show Posts
