Topic: [ANN][GAP] Gapcoin - Prime Gap Search - New Math Algo - CPU / GPU - Zero Premine - page 41. (Read 287669 times)

Hi, Mr. Higgins

I got around that little error by ./configure --without-miniupnpc

But got this error chainparams.cpp

https://github.com/bitcoin/bitcoin/commit/a2b04ddfe6452e7d6274f4096bf3f2aee695a6d9

I applied this patch to get around the Chainparams.cpp error which worked but has created issues with database errors ect.  and wallet.dat incompatibility.

I am currently downloading the blockchain with a auto-generated wallet.dat file however, so I got it "working" on Debian.

Will keep testing it......

The Gapcoin codebase remains unchanged from its Jan 2015 state (https://github.com/gapcoin/gapcoin/commits/v0.9.2-gap). Whilst the (then-contemporary) OS X and Windows binaries continue to function on today's modern distros, that isn't true of the Linux world, you will run into errors if you try to build using a modern Linux distro because, as you have discovered, some Linux libraries often introduce backwards-incompatibility.

fwiw, I wouldn't expect to be able to build a Linux codebase that's been unmaintained for years without first addressing the backwards-incompatibilities, especially if the dependencies include boost, Qt and openssl.

The miniupnpc library is repeatedly not backwards-compatible, see https://github.com/slimcoin-project/Slimcoin/blob/master/src/net.cpp#L873 for the version switches I added to Slimcoin, it should be just a copynpasta solution.

In other circumstances, I'd create a helpful Pull Request but in this instance it'd be a waste of my effort/time as it seems fairly clear that the dev has made a deliberate choice not to actively maintain the Gapcoin github repos.

HTH,

Cheers

Graham

Hi, Mr. Higgins

I was trying to build for the first time using Debian. 

I was not aware of the PoWUtils.h error until now but you have it backwards I figured it out, but not sure as the real problem I was having was Debian updated OpenSSL to 1.1.0-dev after I switched to 1.0.2-dev.  That fixed most of it. 

But many many errors, I never see using Trisquel, so I guess it could be any program I just assumed it was more boost errors (not my specialty) (forced conscription). 

But I thought it would build up until this error it fails on Debian.  Hope you can figure it out. 

CXX      net.o
net.cpp: In function ‘void ThreadMapPort()’:
net.cpp:1087:68: error: invalid conversion from ‘int*’ to ‘unsigned char’ [-fpermissive]
     devlist = upnpDiscover(2000, multicastif, minissdpdpath, 0, 0, &error);
                                                                    ^~~~~~
net.cpp:1087:74: error: too few arguments to function ‘UPNPDev* upnpDiscover(int, const char*, const char*, int, int, unsigned char, int*)’
     devlist = upnpDiscover(2000, multicastif, minissdpdpath, 0, 0, &error);
                                                                          ^
In file included from net.cpp:24:0:
/usr/include/miniupnpc/miniupnpc.h:63:1: note: declared here
 upnpDiscover(int delay, const char * multicastif,
 ^~~~~~~~~~~~
In file included from util.h:15:0,
                 from net.h:18,
                 from net.cpp:10:
tinyformat.h: In static member function ‘static const char* tinyformat::detail::FormatIterator::streamStateFromFormat(std::ostream&, unsigned int&, const char*, int, int)’:
tinyformat.h:784:21: warning: this statement may fall through [-Wimplicit-fallthrough=]
             out.setf(std::ios::uppercase);
             ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
tinyformat.h:785:9: note: here
         case 'x': case 'p':
         ^~~~
tinyformat.h:790:21: warning: this statement may fall through [-Wimplicit-fallthrough=]
             out.setf(std::ios::uppercase);
             ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
tinyformat.h:791:9: note: here
         case 'e':
         ^~~~
tinyformat.h:796:21: warning: this statement may fall through [-Wimplicit-fallthrough=]
             out.setf(std::ios::uppercase);
             ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
tinyformat.h:797:9: note: here
         case 'f':
         ^~~~
tinyformat.h:801:21: warning: this statement may fall through [-Wimplicit-fallthrough=]
             out.setf(std::ios::uppercase);
             ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
tinyformat.h:802:9: note: here
         case 'g':
         ^~~~
Makefile:869: recipe for target 'net.o' failed
make[3]: *** [net.o] Error 1

I seem to be inordinately lucky then because (coincidentally) just last night I successfully compiled the git repos checkout on Mint 18.2 (after correcting for the known problems in rpcrawtransaction.cpp and PoWUils.h). Changes to the Linux distro packages means I need to add CPPFLAGS=-fPIC CXXFLAGS=-fPIC to the make command but other than that, it seemed to be a standard autotools build.

Am I missing something?

Cheers

Graham

---

--- a/src/PoWUtils.h
+++ b/src/PoWUtils.h
@@ -296,6 +296,7 @@ class PoWUtils {
     static const uint64_t log_150_48 = 0x502b8fea053a6LL;
 
     /* 1.0 / 2^47 (for debugging) (not 48 bytes because rounding errors could count as failure) */
-    static constexpr double accuracy = 7.105427357601002e-15;
+    static const double accuracy = 7.105427357601002e-15;
 
     /**

---

--- a/src/rpcrawtransaction.cpp
+++ b/src/rpcrawtransaction.cpp
@@ -296,7 +296,7 @@ Value listunspent(const Array& params, bool fHelp)
             CTxDestination address;
             if (ExtractDestination(pk, address))
             {
-                const CScriptID& hash = boost::get(address);
+                const CScriptID& hash = boost::get(address);
                 CScript redeemScript;
                 if (pwalletMain->GetCScript(hash, redeemScript))
                     entry.push_back(Pair("redeemScript", HexStr(redeemScript.begin(), redeemScript.end())));

---

Considering Gapcoin fails to build on every single operating system except Trisquel I would assume the value is appropriate.  I don't mind but even Trisquel will upgrade past Boost 1.58 and OpenSSL 1.0.2.

There is not a single distribution with updated software in which Gapcoin will build now.  You can update outdated OpenSSL using a different repository until OpenSSL no longer maintains 1.0.2, but it is extremely difficult to downgrade Boost 1.62 to Boost 1.58 to get it to build.

I'll bite: Uses Functional Mathematical Optimization of Prime Number Gaps to secure the Public Ledger.

Cheers

Graham

Realize that this blockchain has been live and fully functional since 2014. As with Primecoin, the on-chain math solutions are produced with every block. This is a crypto-currency, just like Bitcoin, it's still mine-able and you can trade with it if you want to ...

This is one of the best project of the year based on decent concept. Thisproject have great future.

Trying to finalize more concise blurb for an FAQ etc.,

Not a Token | Not an ICO | No Pre-sale | No Pre-mine | Unique blockchain

Useful PoW | Functional Mathematics | Logarithmic Coin Issuance

...

What does Gapcoin's PoW do that Bitcoin's PoW doesn't do?

... Functional Mathematical Optimization (of Prime Number Gaps) on a Blockchain? ...

- https://en.wikipedia.org/wiki/Functional_(mathematics)
- https://en.wikipedia.org/wiki/Mathematical_optimization

Mathematicians. Is this statement correct? Cheers!

No, that only offers -t threads, too crude to use as a CPU usage control lever.

I looked at: https://github.com/gapcoin/GapMiner

Cheers

Graham

OK thanks for that. Does that also apply to the standalone linux miner I recall ?

EDIT : source code for the miner seems to have been removed from github ?

Unfortunately, that codebase is authored in Javascript and would likely prove a challenging integration (I confess I've spared myself the chore of actually checking). OTOH, Dropcoin (https://bitbucket.org/nscrypto/dropcoin.git), (using a Core 0.10 base vs Gapcoin's Core 0.9 base, i.e. close but still non-trivial) has a "percentage of available hashpower" slider in a GUI "mining" tab which might suffice but the feature is incomplete because there is no corresponding parameter for the RPC-API setgenerate call. Rectifying this might be feasible/practicable.

Cheers

Graham

- Bitcoin in the Holy Bible - The Book of Daniel
- https://bitcointalksearch.org/topic/bitcoin-in-the-holy-bible-the-book-of-daniel-167960
- https://en.wikipedia.org/wiki/Bible_code

       

Nice one!

Something which I think is really essential for any new CPU mining software builds (and the wallet) would be the ability to set a % of CPU mining. My current work-around for this is to use cpulimit;


Run gapcoind;



You then need to start a screen instance for top and run top;



This will display all of your systems process ID's - PID's.

You then need to find the process ID for gapcoind. It's most probably near the top of the list.

Then hold and press crtl + A + D to exit the screen session.

Note that if you want to re-enter the same screen session later to check % CPU mining is working then you need to do;


NOT as this will just start another new screen session! You only need to do this after you have re-started your server!

Back at the prompt execute;


Where PID is your gapcoind numbered process ID and 64 (example) is the maximum CPU usage for gapcoind.

Note : I wouldn't recommend going much below 64% CPU usage or much above 74% CPU usage per single core instance - depending on what else your server is running.

Hit enter twice, first to confirm the process ID and second to action the CPU limit into effect.

Now we can start mining with gapcoind;




etc.,


You can now confirm that the mining process (gapcoind) is using the set CPU limit by checking top again.


Then hold and press crtl + A + D to exit the screen session.

To exit the SSH connection with the server it's;

and enter.

...

I will link to this post from https://gapcoin.club/linux-gapcoind-setup.php and also add more 'free' cloud server company options for mining in this regard.

Note that the server company may still open a ticket for 'high CPU usage / abuse' - you might explain to them that you are running a node for a mathematics / scientific research project and tell them you can further restrict the CPU usage, if necessary.

The flag for constant CPU usage varies. However, if your total system use remains on average below 85% to 90% then you will probably go undetected as a 'good egg' in this regard.

Good luck!

@BitcoinFX

Here's some of my optimal 400-500 shift sieves.  Check the pps on these guys don't think you'll be disappointed  

I spent some time sieve searching/optimizing, think these are near the top end of CRT's throughput capability.  The 512 sieve I've set a couple of records with already, got some ~28-30 merits.


--------------
./gapminer-cpu -o 127.0.0.1 -p 31397 -u gapcoin -x password -t 8 --shift 479 --crt crt/crt-22m-480s-dif22084c.txt --fermat-threads 6 --sieve-primes 13000 -q

|== ChineseSet ==|
n_primes:     70
size:         11058
n_candidates: 769
offset:       1476535438052344808249114914435996661778930746178207307948586337728699041748900 69973599252984037935841114047624491809544987485395054755413948
--------------

--------------
./gapminer-cpu -o 127.0.0.1 -p 31397 -u gapcoin -x password -t 8 --shift 511 --crt crt/crt-22m-512s-dif21588k.txt --fermat-threads 6 --sieve-primes 13000 -q

|== ChineseSet ==|
n_primes:     74
size:         11319
n_candidates: 766
offset:       1336758545212663910491770481671492155581631117386004561640004944472923534481900 192446905129909515783553630929900340855652983761167499030099751977773980
--------------

Thanks pdazzl. I will certainly give it a go.

I also notice a drop in the number of blocks found using the CRT miner, probably around that percentage. One wonders if anything in the standalone rpc miner is a miss? Anyhow, I'm still finding the 896 shift to be optimal for my set-up.

It must be j0nn9 who is again updating the records list at gapcoin.org - I don't have access there.

Good good!

Does anyone know who is actively updating the records list?  http://gapcoin.org/primegaps-length.php

It appears to be getting updated maybe once a month.  I've been targeting records in the larger shift ranges, 400-600 and sometimes close to 1024.  Would be nice to see it get updated more often.

If anyone wants to go after the 1020-1024 range here is my best customized sieve file.  Strange thing is gapcoin seems to not record about 15-20% of found blocks for these much higher shifts for some reason, not sure if it's a bug or something else.  Anyways if you're more interested in setting a bunch of new records than getting more coins and have the cpus available then this should do you well.

./gapminer-cpu -o 127.0.0.1 -p 31397 -u gapcoin -x password -t 8 --shift 1021 --crt crt/crt-22m-1024s-dif21514b.txt --fermat-threads 6 --sieve-primes 23000 -q

|== ChineseSet ==|
n_primes:     130
size:         18863
n_candidates: 1039
offset:       4626081985258152152387146716381104236401743138963348264823465389447497680671113 1850105279036924603869257530367004081634142637510553687550996988149393864286408 0741632575037157518785072850028569785866506191383414969297175673797072420119844 8403639569393650350154418384633294148378069691014056008325991106608

Totally. Don't panic!  

Now that the Gapcoin network infrastructure is more stable, I'm going to start website promotion for gapcoin.club , across multiple traffic exchange platforms. This will hopefully grow interest in the project organically. Various traffic exchanges also accept web banners 468x60 , so in regards to your response above I've been working on a simple tag line which tries to be more new user friendly than  "... The largest decentralized prime gap search ..." , which definitely makes most people glaze over when you mention it to them! So, ...

Get Gapcoin | Real Crypto | Real Math | Real Commodity Money

- https://en.wikipedia.org/wiki/Commodity_money

I did start with ...

Get Gapcoin | Real Crypto | Real Math | Real Currency

Any thoughts ?

I'll post when advertising is in full flow (probably starting this week). Already sending 5k - 10k+ unique hits per day via auto exchanges to help build web rank. This advertising is 'free' as I have a good number of TE credits built-up over the last years.  

I understand your point but the real answer is that as long as people are scratching for 'money', rather than commodities, it is hard to redefine 'currency' as a commodity and distinguish between nonproductive currencies like fiat, bitcoin etc and productive currencies like math coins, for most people.

As for addnodes https://chainz.cryptoid.info/gap/#!network always has current nodes e.g.
addnode=104.207.158.149
addnode=104.236.168.125
addnode=106.226.77.104
addnode=108.170.1.134
addnode=109.245.35.136
addnode=113.195.30.114
addnode=117.163.224.99
addnode=117.163.227.164
addnode=118.212.39.226
addnode=138.197.159.202
addnode=138.68.172.174
addnode=139.18.122.9
addnode=139.59.103.137
addnode=139.59.57.170
addnode=148.251.16.180
addnode=153.126.210.34
addnode=172.104.85.43
addnode=172.104.88.43
addnode=182.84.19.90
addnode=188.134.72.213
addnode=193.233.60.184
addnode=201.190.235.165
addnode=212.19.17.213
addnode=217.175.119.125
addnode=217.8.92.48
addnode=23.100.8.94
addnode=37.139.15.251
addnode=45.32.246.15
addnode=45.32.36.228
addnode=45.63.84.31
addnode=45.76.140.157
addnode=62.193.158.46
addnode=73.114.131.4
addnode=77.37.192.38
addnode=82.79.34.116
addnode=88.91.78.109
addnode=91.152.121.223
addnode=95.134.24.99
addnode=95.134.51.27
active within the last 24 hrs

Added standardised config's for running Gapcoin on clearnet and Gapcoin on Tor, respectively.



If you still cannot connect to the Gapcoin network you may need to try an alternative port or configure P2P port 31469 on your firewall and/or router.