NXT :: descendant of Bitcoin - Updated Information

[wesleyh]

Bounty announcement

100'000 NXT will be paid for working JavaScript code that signs and verifies signatures using NRS algo.

- The licence must allow to use the code in any application
- Sign/verify speed must be not lower than 100 signatures per second on a 1 GHz CPU (1 core)
- All the code must be in a single non-obfuscated well-formatted .js file
- Input/output values must be strings like "8302504e4e57c6c65335289879c6915a273d3aae7bd086058e403fcd2bc18341"

The bounty is valid till the 20th of January, 2014 12:00:00 UTC. The complete code must be published in this thread.

Currently i can't see solution for speed up my code to achieve goal, so i decided release it in current state. Other conditions are met.

https://github.com/moadib/nxt_crypto_js

On my PC(core i5 2.6ghz) Java implementation from Nxt sources took approx 20ms for one signature.
My JS code on Google V8(chrome, nodejs) took approx 200ms.

The main problem is that JS doesn't support 64bit math, so script spend most time on emulating this.

CFB - any input on what this guy has done?

[Come-from-Beyond]

p.s. I hope he doesn't say now "Thank you for your frictionlesscoin competent opinion" 

[Come-from-Beyond]

CFB - any input on what this guy has done?

I didn't check if it works correctly. If noone shows code that works much faster then hoax will get part of the bounty.

[EmoneyRu]

Bounty announcement

100'000 NXT will be paid for working JavaScript code that signs and verifies signatures using NRS algo.

- The licence must allow to use the code in any application
- Sign/verify speed must be not lower than 100 signatures per second on a 1 GHz CPU (1 core)
- All the code must be in a single non-obfuscated well-formatted .js file
- Input/output values must be strings like "8302504e4e57c6c65335289879c6915a273d3aae7bd086058e403fcd2bc18341"

The bounty is valid till the 20th of January, 2014 12:00:00 UTC. The complete code must be published in this thread.

Nobody tried to convert cpp to js?

[EmoneyRu]

Bounty announcement

100'000 NXT will be paid for working JavaScript code that signs and verifies signatures using NRS algo.

- The licence must allow to use the code in any application
- Sign/verify speed must be not lower than 100 signatures per second on a 1 GHz CPU (1 core)
- All the code must be in a single non-obfuscated well-formatted .js file
- Input/output values must be strings like "8302504e4e57c6c65335289879c6915a273d3aae7bd086058e403fcd2bc18341"

The bounty is valid till the 20th of January, 2014 12:00:00 UTC. The complete code must be published in this thread.

Currently i can't see solution for speed up my code to achieve goal, so i decided release it in current state. Other conditions are met.

https://github.com/moadib/nxt_crypto_js

On my PC(core i5 2.6ghz) Java implementation from Nxt sources took approx 20ms for one signature.
My JS code on Google V8(chrome, nodejs) took approx 200ms.

The main problem is that JS doesn't support 64bit math, so script spend most time on emulating this.

Dependencies, included in script:

    Long class from Google Closure Library (Apache license)
    jssha256 by B. Poettering (GNU GPL)

Other conditions are met?

[Come-from-Beyond]

Nobody tried to convert cpp to js?

The code must be readable so ppl could make sure that it doesn't steal their passphrase.

[Come-from-Beyond]

Other conditions are met?

I didn't check it yet. It's not fast enough AFAIK.

[EmoneyRu]

Other conditions are met?

I didn't check it yet. It's not fast enough AFAIK.

I'm pointing to absolutely bad GPL license.

P.S. So what?

If I hardcode 136941 first blocks and run it at single node will my chain win? Why not?

[hoax]

Nobody tried to convert cpp to js?

Generated code approx 2x faster than my implementation, but not readable. And still not enough fast as required.

Dependencies, included in script:

    Long class from Google Closure Library (Apache license)
    jssha256 by B. Poettering (GNU GPL)

Other conditions are met?

If GPL license not suitable i can easily change it for another. Used only in crypto_sign and crypto_verify functions.

[EmoneyRu]

Generated code approx 2x faster than my implementation, but not readable. And still no enough fast as required.

Ok. Just curious. I have zero knowledge at web-developing

[opticalcarrier]

my update.html link tells me i can upgrade from 0.5.1 to 0.5.3 but the download button it gives me does nothing.

the link http://download.nxtcrypto.org/nxt-client-0.5.3.zip does appear to be valid though

has something broken in the update.html page?

[joefox]

this is very usefull.

http://wiki.nxtcrypto.org/wiki/Nxt_API

who compiled this? can't find a donation account to support this work.

This was me.  The original page was built on the old wiki site, and then ported over to the new site (which is why the original version looks like it comes from Israel).

I have some formatting work to do to make inputs and outputs more reasonable (at NxtChg's request) but anyone can contribute.  To fight spam, you have to register and let me know you want in (instructions are here) but after that you can edit as needed.

There is page on web.xml too: http://wiki.nxtcrypto.org/wiki/Web_xml_configuration_file

I will also have a page built for APIv2 (the low-level API) as that one is constructed.

[EmoneyRu]

the link http://download.nxtcrypto.org/nxt-client-0.5.3.zip does appear to be valid though

It seems valid

[joefox]

Re-posting my Public Service Announcement:

If you are new to Nxt and don't want to read through 845 pages of forum posts, come to the wiki:
- Learn about Nxt
- Get started
- get basic questions answered in the FAQ and common problems pages
- Learn how to set up a public node on a VPS, using laowai80's steps
- read up on the Nxt high-level API, which you can use to build apps, games, clients, utilities, and more

Also remember there are great forums with lots of info and support for beginners. Many of the best Nxt wiki articles are based on information originally contributed here:
- The nxtcrypto.org forum
- The nextcoin.org forum

I comb these forums every day and pull tidbits to update the wiki with, so that you don't have to.  

This thread is long and daunting for new folks.  Don't get frustrated -- come to the places where new folks are welcomed and pampered!

[EmoneyRu]

How about peer cleaning? Now I have 800 active peers. It is too much for my little node

[wesleyh]

my update.html link tells me i can upgrade from 0.5.1 to 0.5.3 but the download button it gives me does nothing.

the link http://download.nxtcrypto.org/nxt-client-0.5.3.zip does appear to be valid though

has something broken in the update.html page?

Are you using the http:// or https:// page?

Use the http page, since the download itself is over http:// (at the moment). Https pages do not allow downloading of http documents. This will be fixed in the future when the files will also be available over https.

[notsoshifty]

P.S. So what?

If I hardcode 136941 first blocks and run it at single node will my chain win? Why not?

My guess is:

 - You can't change the genesis block, as it's hardcoded into the general release and future blocks wouldn't be accepted at all
 - You would need to know the pass phrases of a sufficient number of the 72 original accounts to forge the remaining blocks with at least the same total difficulty as the current chain
 - Anyway, the release client has a restriction that it will never switch onto a fork >720 blocks old (not sure about this one?)

[ferment]

Bounty announcement

100'000 NXT will be paid for working JavaScript code that signs and verifies signatures using NRS algo.

- Sign/verify speed must be not lower than 100 signatures per second on a 1 GHz CPU (1 core)

Nobody tried to convert cpp to js?

Same problem that one of my guys ran into too.

The main problem is that JS doesn't support 64bit math, so script spend most time on emulating this.

[utopianfuture]

I recall someone advising NXT marketing efforts to be put on hold until most of the client and network stability issues are resolved and I agree. Until those issues are resolved, though, I believe that all the marketing savvy NXT folks should start making a BIG STINK out of the fact that NOBODY has been able to find a flaw yet in the other topic!  Who knows? ...this marketing stunt could continue all the way till April 3!!!  

Personally, I have reached a Dead End in my efforts (although I will continue looking and bugging the smarter people) and there has also been admittance to not being able to find any more bugs (not to mention flaws) by another NXT friend that has provided much more valuable input than me along the way.

The closest anyone has gotten to the flaws, so far, was asking CfB a question that he could not answer because it would reveal the flaw, but that doesn't necessarily mean that the person asking the question had the specifics in mind!  

Many people say so but it is probably me one of the most vocal. However, there would be AWESOME AWESOME promotional stunts when the nexrn's client is ready and the asset exchange is on the Testnet. By that time, I hope we also have a stable network ready for some serious stuffs.

I am kind of feel bored these days because I cannot really read codes. So there aren't much to do except contacting some prospective investors and leisurely browsing this thread and some forums 

[EmoneyRu]

I am kind of feel bored these days because I cannot really read codes. So there aren't much to do except contacting some prospective investors and leisurely browsing this thread and some forums  

Me 2