Topic: Ixcoin TODO - page 85. (Read 631747 times)

Ok i have a few things to say here

First GR, I took out your pre release flag OUT after Ahmed told me the code was complete when I built the 9.3 wallet. After testing and testing on several machines for hours I releaesd the windows version only. Testing continued after I compiled the 9.3 Windows Client now available for download on Thomas's ix.org site.  I pushed the release to Vlad's github since I don't have collaberator privliges to Ahmed's and Thomas seems to want to keep his on the old code base although soon I plan to reach out to him or do a gitpull to ask him to merge the new code if no one else does.  I certainly did not want to create a new one there are too many iX coin githubs as it is. In my eyes the only two that are of any value are the original by Thomas and the most recent 9.3.  (the 8.x github iX for windows would not compile for me until I forked it and rewrote a few things including simple copy paste items indicating it was based off of i0 coin.)


Those asking for linux builds are making me wonder.  I never ran a pool but I run cloud nodes for (I won't say how many coins for various reasons , abe block crawlers , all sorts of things on ec2 / dig. o ect...   / I never worked close with an exchange either.


If I can imagine myself in the position of the person running the pool or exchange I would NEVER  want a precompiled binary for nix from anyone.  I would want to build it myself. Anyone who has ever compiled this client knows that compiling it on windows is about 1000x harder than linux.  Since we do not use the gitan build system of bitcoin with a consensus of 3 identical builds I would find it odd that any large pool would trust anyone, even someone with my level of integrity  to make their daemon for them.

I agree with Ahmed that there is only so much testing that someone like he and I can do with whatever small hash we control. (i know i have almost nothing)

Mmpool I would be very interested to know more technical details about exactly the issues you had and what code bases of all coin clients you were using at the time also the details of merged setup.....eloipool ?  , (if i recall p2pool merged mining was p2pool only on btc and you were basically solo for the auxuliary chains.  Also curious which version and details of the level db on the bitcoin, namecoin, i0 , dev and iX since having some 'mix and match' in my mind (nothing i tested so i could be incorrect) could cause some conflicts.

A few of you brought up a valid point. The heavy lifting (coding) has been done by people who are not being paid but basically are doing this because we want to do it. (And Vlad who does not code but paid out of his pocket hired guns)

So honestly .............how many 'normal end users' need or want a linux build daemon to send and receive coins ect.... ?

If I have time I may make one. I don't like wheezy and hate using it. If I do it likely will be ubuntu. I can do basic tests but until someone solves a block on main net (sorry I don't think testnet is reliable for iX or other coins for that matter---- the main net is where tests really pass or fail, yes if it fails on testnet you don't try on main net but I have seen lot's of things work on coins testnets perfect until you put on main net) we really won't know will we ?   Also Ahmed has a valid point.  It would be unwise and unreasonable to ask cex or anyone to test it with the full blast.  Much better to try out on a small portion of the network.  Even if no blocks are solved if there are crashes, timing issues , ect... we will know we need tweaks.

If these are detected instead of a gitpull if you don't have the fix open an issue. Github is also useful for discussion between devs and others close to the code. Another advantage is you see almost no idiots on github.

To those asking for mac clients I purchased a mac book pro a few months ago to build wallets for another project I am working on.
I built some wallets but they have issues and I won't release something I am not confident in so I put it to the side.
Recently I got the mac book pro out again and am having another crack at making mac wallets. 

(After all if I am known for anything it is making good wallets)


When I have time I need to finish my other project first and will be a good primer to start to try to build mac wallet for iX 9.3.

Please don't expect this right away. Eventually I will get to it.


iX coin has ten minute block target times


Finally while I am not as ambitious as others who want to continue to develop iX to keep up with the furious pace that the btc dev team is pushing commits to github (9.3 is working perfect for me, see my prior posts about the number of tests conducted)there is one item I think would be a very good idea to merge if possible.

It is this in bitcoin https://github.com/bitcoin/bitcoin/pull/2340#event-210292499https://github.com/bitcoin/bitcoin/pull/2340#event-210292499  and has to do with fees and future security.... for those who understand these things the write up on github is quite clear why this is even more important for iX than for bitcoin since iX is mined out.

It is a 'protection' for when bitcoin block awards are less than fees or there are no block awards it will prevent 'mischief' and fortify the codebase as to make sure future changes do not make it more difficult or impossible to add this later in case someone sees it as an issue.

Would like the opinion of Ahmed, Gr , and even Mmpool about merging this.  A bit tricky since the original gitpull was done a long time ago and the core changed a lot since that time.

Love all.

Well the conf times still seem around 10 minutes per block. 

You can say goodbye to your IxCoin blocks and hello to very long conf times....  Lulz

@mmpool i do certainly understand what you are saying, however what do people expect me to do? the ixcoin testnet is non-existent. The original source did not contain a valid testnet, and creating one is a job in itself since a lot of the IXCoin code in particular has no seperation of code for mainnet and testnet which breaks a lot of things on testnet.

At best the only thing i can do is test when i get my pool functioning properly (HDD ran out of space on my last test server *doh*) without it i cant do any testing whatsoever on the mining end which truly makes things impossible for me. This is what most people fail to understand. Only few pools have the capability to test this, thats you and cex.io

I do wish this wasnt the case but it sadly is.

Ahmed

@Ahmed @Vlad  

Sure can understand your frustration, and perhaps my leaving the pre-release flag so that the Ixcoin Core wallet builds a pre-release was a mistake, it's easy to change in src.  It sure would be nice to see cex.io step forward with additional development funds, but equally as important, test results.  Other than a couple, very nice donations (much thanks), which I've still got at my donation address, the work done to-date has been personally rewarding, not based on financial gain.  Seems the same for everyone here, from what I can tell, has all been done for FREE, or personal cost.   However, until bounty targets have been met, there is only so much one can do, to show progress.  As a group, think we've done a great deal, if you count the combined hours.  

With comments like mmpool's just posted, I've gotta say....didn't know that, lets see if we can get it fixed, and then get even more feedback, after a new build is made.
However when referencing your message link, and the source being discussed, I find that it's https://github.com/FrictionlessCoin/iXcoin/pull/1
your talking about, in other words, it references the v8 client.  So I'm kinda confused as to wither or not your talking about the v9.3 ixcoin having this problem.  I don't think it does.
Please feel free to pm me here with your details, or post the specifics related to your v9 build tests, that would be helpful.  These things do need testing, and I've been of the opinion, we need more feedback, before the new wallet upgrade is shipped out in a released state.

@mmpool - If you can help test, Is the Source on github ok, or do you need a particular binary, if so which platform.  

General observation:
Please correct me if I'm wrong, but from what I can tell, there are a growing number of v9.3 clients out there on the network now, and working very well for the users.  RPC calls to the daemon or -qt server, for use in various operations also seem to also be working very well.  If this is not the case, please help me catch up on the backlog of discussions and point me to it, so we can get it fixed.  Hard to do, if you don't know about it.


WIll try to do as much as I can, off the top of my head... which means it could be wrong, however best shot:

The default RPC port is 8338

Min Tx fee, never really got worked on by me, pretty much standard bitcoin v0.9.3 code I think.  After our discussions months ago, seemed consensuses was to leave it as bitcoin v9 had done, allot of stuff changed between v8 and v9 there.   We can certainly work on hard coding something specific now, if that is what is wanted, I argued it should be enforced back then, although now we're really starting to taste the reason why we must move towards an enforceable minimum.
 
Estimated block time, hmm got me there, would have to look that up.  The block target time is something I should know.

the daemon file is called ixcoind.exe on windows and simply ixcoind on linux

Linux versions I now build, have a configuration directory path of /home//.ixcoin
...in which all the blockchain, wallet and ixcoin.conf (the config file) settings can be found.

On Windows versions, the config files are called the same name (ixcoin.conf), but the path to them varies, depending on which flavor of windex your running.

We don't have a v9 version for the Mac, that I know of.  And my latest attempts to build them here just failed last night, even with the latest version of bitcoin v0.10.99 depends.  Think that without an upgrade on my debian OS, the machine can't find dependencies new enough.   It will have to be done by someone else for now, I've gotta load a new OS, on another hard disk, before trying again, takes time.   Been working hard on getting darwin to build, but it's kinda complicated to build on a Mac for a Mac, and I'm trying to do it from a linux machine.

@steelhouse - Agree with you there, was surprised to find out how complex the transaction fee code had gotten in v9.3, planning to once again attack that section of code and hopefully produce results that enforce a minimum, but not break what they are doing already, for other aspects of the calculations.   As there was so much work to do getting the upgrade to this point, was burden enough, without trying to modify the Tx fee calculations as well.  It will be revisited.  As of yet though, I do not see any consensuses on what that should be.

Ixcoin amazed me, by passing the 'last coin' mined date, with success in all our client versions, now we're entering a new phase.  Mining has risen, and is about to fall hard into economic bad times.  There will be pockets of opportunity though, and for Ixcoin we must learn to survive on Tx fees only, a true test for any cryptocurrency, and one that could become a shining beacon of hope.  Not only for those miners in the years to come, but also the users of it, a true currency, and method of transferring wealth around the planet, far easier than gold.   I'm confident we will succeed.

GR

We did set a minimum fee with the latest update but it was mainly to prevent blockchain bloat like what happened to Bitcoin early on.

Good for you Ahmed.

You have done more than enough.

I'm sure CEX.io can afford a couple Bitcoins to do the rest.

It's a bit risky for them to test on mainnet since they could destabilize their bitcoin mining. If the ixcoin code crashes or causes a problem when a bitcoin block is found then they are out of pocket for the block. When I ran the last ixcoin client update on the pool it hit issues with ixcoin crashing when certain P2P messages occurred for example.

Does IXC have a mandatory transaction fee?  If not I suggest a permanent 0.01 or 0.001 fee to prevent spam attacks.  The I0C coin logo looks real good.  If people don't know the history I0c and IXC are bed brothers.  I0C was created since some did not like the idea of creating a premine and bounties.  There are some old IXC logos that were pretty nice.

Man.... this is a really brutal bear market!

Well, we will just have to wait and see when CEX.IO will start testing the latest IXC.

Thats it, i'm done for now. When and IF i get the time i'll complete the testing of the IXCoin code on my own pool after ive got it set back up. Until then it's going to take a back seat. Sick of being taken for a ride.

Ahmed

There corrected that sack of shit for you

big difference. We're asking IXC to TEST some code for usage on their pool. there's a big difference. No one has asked them to write a single line of code. If they dont get it to work they can come back to me with logs and i can fix it.

You dont seem to notice that.

It's unfair for CEX to have to do any work for the coin they swore to support.

CEX only mined 500,000 bitcoins last year.

It's unfair to ask them to pay for anything.

I, while unemployed, spent more money on updating IXC code than CEX.

Is that fair?

Jumbley, you're the most obvious CEX troll I've seen here yet.

I am very very close to bringing in a partner that will allow me to fire CEX.io and all their trolls.

It's tested as far as i can do so. It's completely unfair of you people to expect me to to do something i cannot with what little capability i have.....
CEX.IO CAN test the coind on main-net very-very easily and can report back to anyone here with any issues. They have the hashpower and capability to do so. Whereas i do not.


Why do i even bother with you all. Me, GR and cinnamon have done all we can to get everything working as best as we can, asking little in return while dirtbags like cex.io and co can't be bothered to give something a quick test which is very, very easy for them to do rather than stringing this out. And on top of that we get comments like i've quoted above. You think things are fair on us? Slaving away for nothing trying to get something to work which you cant?

Update:


CEX informed me today that they are not able to run/test the 9.3 update.

I asked for more details but haven't heard back yet.

That was a good find jumbley. Multidex would be amazing to go onto, it would add a decentralized exchange and lots of free marketing.

Currently we're looking for popular coins compatible with the Bitcoin RPC.





And look, ixc is popular!! One of the cool kids..




Hello everyone! To allow the SuperNET (www.supernet.org, MultiGateWay, InstanDEX, ...) to support this coin I would like to kindly request from you the following data:

Default RPC port
Minimum transaction fee
Estimated block time in seconds
daemon file
Exact spelling of the conf file with path for
Windows
Mac
Linux

My server ran out of HDD space before i was able to succesfully test a block. I know for sure it's not working on testnet. On mainnet i think it may do. With testnet i was having an issue where it complained that the block time was too far in the future

Will follow your progress Vlad,

...just so far out of the loop, it's hard to comment on, have not followed the politics much, been buried in the details.

Wanted to thank the member joining our translation team.  Italian done well, will be a big PLUS!