[ANN][RIC] Riecoin: constellations POW *CPU* HARD FORK successful, world record

[wakawakalj]

Sentiment poll -

I have the linux version of the next iteration of my riecoin miner.  It'll substantially reduce memory use on many platforms, and is a bit faster, and a better long-term building block.

It doesn't work on windows and I'm having a heck of a time debugging it.

Is there any sentiment towards releasing it where it only works on Linux and letting more people stare at what's wrong with windows vs. kicking at it longer and trying to fix it to have both win/linux ready at the same time?

Well, I'm primarily a windows user - both of the boxes I have mining are running windows. I know what *nix is good for, and it's good for lots of things. I frequently hate windows with the fir-y passion of ten thousand burning suns (less now that I simply re-image my boot drive every 4 months). But unfortunately windows has the software I need for my career.

I would really like to see a new version of the windows miner, especially if the *nix variant is already out. However, if it's a debugging/windows being a whiny bitch issue, I don't see any major problem with releasing the working one first. Anyone who pouts about it not working on windows is free to contribute to the codebase, I'm sure.

[1714973278]

1714973278

[1714973278]

1714973278

[1714973278]

1714973278

[1714973278]

1714973278

[notsoshifty]

I'm not an expert but I think Python 2 and 3 are different languages, not compatible.
You have to use the correct one or port the code

It looks like 'yield from' is a 3.3 thing (fairglu is using 3.2):

https://docs.python.org/3/whatsnew/3.3.html#pep-380

Fairglu - any chance of upgrading? Otherwise, converting 'yield from X' to 'for Y in X:  //   yield Y' might work.

[northranger79510]

Any update this week Gatra?

[surfer43]

Sentiment poll -

I have the linux version of the next iteration of my riecoin miner.  It'll substantially reduce memory use on many platforms, and is a bit faster, and a better long-term building block.

It doesn't work on windows and I'm having a heck of a time debugging it.

Is there any sentiment towards releasing it where it only works on Linux and letting more people stare at what's wrong with windows vs. kicking at it longer and trying to fix it to have both win/linux ready at the same time?

I say release it. All you have to do if you’re running windows is download an iso and run it off a flash drive / cd. Unless of course you need to use Windows while mining..

[philpescy]

so good the coin is.

[gatra]

Hi people!

A small update:

I'm currently testing the new client. It comes with some speed improvements, but it's a scary thing since a bug there can cause the coin to halt. I'm being extra careful.

The Riecoin Foundation is working in many projects. Thanks!

Difficulty seems stable with a very small tendency upwards. Still expecting it to take off and break records any time.
We are almost finished creating the basis from where to grow, don't give up on RIC!


Thanks and regards,
Gatra

[istudy92]

Sentiment poll -

I have the linux version of the next iteration of my riecoin miner.  It'll substantially reduce memory use on many platforms, and is a bit faster, and a better long-term building block.

It doesn't work on windows and I'm having a heck of a time debugging it.

Is there any sentiment towards releasing it where it only works on Linux and letting more people stare at what's wrong with windows vs. kicking at it longer and trying to fix it to have both win/linux ready at the same time?

'


I am a linux and windows user.

I personally like using windows as its just my very nature in terms of simplicity.  Like northranger stated waiting would be better as haters are going to hate if they can not use the OS they are comfy with.

Fairness is key to success.

[notsoshifty]

I'm currently testing the new client. It comes with some speed improvements, but it's a scary thing since a bug there can cause the coin to halt. I'm being extra careful.

EDIT: I over-reacted

[notsoshifty]

I have the linux version of the next iteration of my riecoin miner.  It'll substantially reduce memory use on many platforms, and is a bit faster, and a better long-term building block.

It doesn't work on windows and I'm having a heck of a time debugging it.

Is there any sentiment towards releasing it where it only works on Linux and letting more people stare at what's wrong with windows vs. kicking at it longer and trying to fix it to have both win/linux ready at the same time?

If it is just "a bit" faster (how much?) and doesn't give Linux miners a huge advantage, I'd vote to release it and get a wider audience debugging it on Windows, and so get the Windows version out sooner. I'm not a miner though.

[fairglu]

I have all transactions ID that funds were sent into if that is what you are asking.

I was also asking in case they happened after the block the explorer was stuck on.

As a quick fix, you can ignore them. Later you can accumulate and report RIC on nonstandard tx

Ok, as current workaround those transactions will be considered as having sent their amount to the "nonstandard" address, that'll make it easy to check if there are many of those or if this was just an isolated case (and thus a probable bug or experiment)

The explorer is now catching up with a fully synchronized client, so it shouldn't be long until it catches up.

It looks like 'yield from' is a 3.3 thing (fairglu is using 3.2):

https://docs.python.org/3/whatsnew/3.3.html#pep-380

Fairglu - any chance of upgrading? Otherwise, converting 'yield from X' to 'for Y in X:  //   yield Y' might work.

Urg, had overlooked 3.2.3 vs 3.3 :/
Not sure about 3.3, this is a Debian 7, and I'm wary of straying from the main update & security channels on a production server.

There are multiple yields in the function, and while I'm not fluent in Python, if Python's yield works like in other languages, it might be more complicated to unravel them. Best solution would probably be an API in the client though.

[notsoshifty]

Urg, had overlooked 3.2.3 vs 3.3 :/
Not sure about 3.3, this is a Debian 7, and I'm wary of straying from the main update & security channels on a production server.

There are multiple yields in the function, and while I'm not fluent in Python, if Python's yield works like in other languages, it might be more complicated to unravel them. Best solution would probably be an API in the client though.

Makes sense about not straying from the official updates! I use Fedora20 and 3.3.2 is the default. I used to use CentOS, and they always used much older (but perhaps more tried & tested) versions.

Can you try this one?

https://gist.github.com/notsoshifty/11027154#file-check_proof_of_work_nss-py

Feel free to do a diff to check I'm not trying to steal your coinz first

Also, make sure you do 'view raw' or clone the repo or something, as I had a few copy/paste issues with spaces etc initially. It is Python, after all

[fairglu]

Ok, the blockchain explorer is now fully synchronized!

• The nonstandard transaction was an isolated case (search for "nonstandard" in the explorer to see it or check if other appear in the future)
• The multisig address that was mentioned previously is now present, it was in the blocks after the nonstandard one.
• I've added the two active pools I could find (YPool & UpCPU), though extraction stats are not very interesting right now as YPool just dominates with 95%.

Feel free to test and check things to see if everything else is okay!

[northranger79510]

Ok, the blockchain explorer is now fully synchronized!

• The nonstandard transaction was an isolated case (search for "nonstandard" in the explorer to see it or check if other appear in the future)
• The multisig address that was mentioned previously is now present, it was in the blocks after the nonstandard one.
• I've added the two active pools I could find (YPool & UpCPU), though extraction stats are not very interesting right now as YPool just dominates with 95%.

Feel free to test and check things to see if everything else is okay!

Wow! This is excellent news. The multisig does work for the foundation which is great because the community can check whether money is withdrawn or not . Added recommendation: https://darmstadt.goxadidi.dk/stats/rich_list.html for Rich List.

I think we have a winner for the bounty guys. Another community member is working on a block explorer also and I think I will offer him a bounty too to have 1+ block explorer.

[fairglu]

Can you try this one?
https://gist.github.com/notsoshifty/11027154#file-check_proof_of_work_nss-py

Great, seems to work alright! I'll see about using it in the explorer (will take a tiny bitsy of dev to be efficient, I'll just hack something dirty first)

Would there be an interest in publishing a whole database of found prime constellations?
I mean something like a raw format, like a big text file with all the primes, that might be more practical for math/research purposes.

[northranger79510]

By the way, if you guys take a look at the foundation's address on the block explorer, it shows we withdrew about 2RIC. That was for testing purposes to test if the multi-sig works.

[fairglu]

Added recommendation: https://darmstadt.goxadidi.dk/stats/rich_list.html for Rich List.

Rich list is recomputed at least once every 6 hours currently, plus when I manually ask it - which I just did

The amounts seem to agree AFAICT, only difference is my explorer lists the last changes (ie. including deposits), while your list shows last withdrawal.

[northranger79510]

Added recommendation: https://darmstadt.goxadidi.dk/stats/rich_list.html for Rich List.

Rich list is recomputed at least once every 6 hours currently, plus when I manually ask it - which I just did

The amounts seem to agree AFAICT, only difference is my explorer lists the last changes (ie. including deposits), while your list shows last withdrawal.

Ahh I see. Well since your block explorer seems to have everything for Riecoin addresses you get the bounty. Please post your address on here and it will be sent in 1-2 days.

[fairglu]

Further updated the RIC explorer to leverage notsoshifty's code and display the prime.

To avoid cluttering things, only the base prime (n+0) of the constellation is shown.

I'll make all the primes available as a simple zipped text file later on (ie. when the db is ready).

Ahh I see. Well since your block explorer seems to have everything for Riecoin addresses you get the bounty. Please post your address on here and it will be sent in 1-2 days.

Thanks   RL9Ldk7EKHBanArEmYznLdGjTZHvazUXjb

[dga]

Further updated the RIC explorer to leverage notsoshifty's code and display the prime.

To avoid cluttering things, only the base prime (n+0) of the constellation is shown.

I'll make all the primes available as a simple zipped text file later on (ie. when the db is ready).

Ahh I see. Well since your block explorer seems to have everything for Riecoin addresses you get the bounty. Please post your address on here and it will be sent in 1-2 days.

Thanks   RL9Ldk7EKHBanArEmYznLdGjTZHvazUXjb

Bonus 500 RIC bounty sent also.  Thanks for doing this!

[notsoshifty]

Further updated the RIC explorer to leverage notsoshifty's code and display the prime.

To avoid cluttering things, only the base prime (n+0) of the constellation is shown.

I'll make all the primes available as a simple zipped text file later on (ie. when the db is ready).

Great stuff

(Full credit of course to ysangkok's for the original code to extract the prime numbers from the block, which I made a couple of small amendments to; ysangkok if you're reading this perhaps you can check if the Python 3.2 version changes are correct?)