Bitcoin Forum
August 20, 2018, 01:59:48 AM *
News: Latest stable version of Bitcoin Core: 0.16.2  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 [18] 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 »
  Print  
Author Topic: [I0C] I0coin - The Best Choice In Digital Currency  (Read 64275 times)
brooksby
Full Member
***
Offline Offline

Activity: 199
Merit: 100


View Profile
December 31, 2015, 11:03:15 PM
 #341

(It seems like you are saying the development branch is what people currently get at the main URL?)

It is, and that is the same as just about any other software project out there (including Bitcoin).  If you want stable code, use the releases (tags).  If you want to stay on the latest code, use the main branch.  Note that I believe the branches to be mostly bugfree, but they are nevertheless no explicitly created and tested releases.
Did you write unit tests that pass on risky code? Bitcoin is not the same as yours because the code is tested and isnt put in a state where ppl can pull broken code especially if test coverage is above 95%... The releases happen after integration testing. So you should not recommend pullimg master or dev branch unless ots unit tested at the least

Isn't that exactly what I wrote?  I suggest to use the release tags, unless you know what you are doing and insist on pulling the branches (but I'm definitely not recommending doing that).  This is what I wrote.  Note that I actually do test the code before I push, but of course not as thoroughly as the release tags are tested.

... and again that's the way Bitcoin does it as well.
1534730388
Hero Member
*
Offline Offline

Posts: 1534730388

View Profile Personal Message (Offline)

Ignore
1534730388
Reply with quote  #2

1534730388
Report to moderator
1534730388
Hero Member
*
Offline Offline

Posts: 1534730388

View Profile Personal Message (Offline)

Ignore
1534730388
Reply with quote  #2

1534730388
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1534730388
Hero Member
*
Offline Offline

Posts: 1534730388

View Profile Personal Message (Offline)

Ignore
1534730388
Reply with quote  #2

1534730388
Report to moderator
1534730388
Hero Member
*
Offline Offline

Posts: 1534730388

View Profile Personal Message (Offline)

Ignore
1534730388
Reply with quote  #2

1534730388
Report to moderator
1534730388
Hero Member
*
Offline Offline

Posts: 1534730388

View Profile Personal Message (Offline)

Ignore
1534730388
Reply with quote  #2

1534730388
Report to moderator
brooksby
Full Member
***
Offline Offline

Activity: 199
Merit: 100


View Profile
December 31, 2015, 11:10:57 PM
 #342


As for the testing of release version 0.11.9, here are a few points to make.

First off, since I0coin software is over 99% the same as Bitcoin that brings along with it tremendous reliability and testing which was already done on Bitcoin. 

Second, the primary difference with Bitcoin is the merged mining.  That has been under live testing for a couple of months now with no problems reported.  We also have 100% of the mining now performed using 0.11.9.

Third, domob and I have done testing.  Some of my tests are listed below.

-whole 0.11.9 block chain downloaded from scratch
-connected to 51.254.131.226 (node hard coded in client) for block chain download
-connected to ybs6nrn4bpvckalt.onion (Tor address for 51.254.131.226) for block chain download
-pruned block chain to 600 MB
-loaded a 0.8.6 wallet with a 0.11.9 (pruned) client
-transferred I0C from 0.11.9 (pruned) to 0.8.6 client
-transferred I0C from 0.11.9 (pruned) to 0.11.9.9 (unpruned) client
-transferred I0C from 0.8.6 to 0.11.9 (unpruned)
-transferred I0C from 0.11.9 (unpruned) to 0.8.6 client
-set server=1, sent rpc commands and received correct responses
-txindex=1, successfully re-indexed block chain

brooksby
Full Member
***
Offline Offline

Activity: 199
Merit: 100


View Profile
January 02, 2016, 08:42:00 AM
 #343


I've been thinking about our I0coin logo for a while.  Moving forward I see some need to modernize... just a bit.  One of the issues that I0coin has had is confusion about whether the 'I0' is an 'I zero' or 'I' letter 'O'.  It became even more confusing when a rival coin, I/O coin, came out last year, which really is 'I' letter 'O'.  

Let's consider our existing logo.  Notice that the zero in 'I0' and the letter 'o' in 'coin' look identical.  This furthers the confusion.  So I'd like to fix that with a logo which clearly shows the difference.  As a secondary concern I think the existing logo font looks a bit out dated.  And to top every thing off, I do NOT have the master graphics file for the existing logo.  So there's really no way for me to create larger versions if need be.  

With all this in mind I've designed a new logo.  I've kept the color scheme to keep it looking as close to the existing logo as possible.  The coin image didn't change.  I've changed the font to look more modern, but it still has a similar look and feel.  Most importantly, the zero and letter 'o' are clearly different.  Compare them for yourselves.  I welcome feedback, both pro and con on the new design.

Existing logo:




New logo:



mmpool
Sr. Member
****
Offline Offline

Activity: 375
Merit: 250


View Profile WWW
January 03, 2016, 01:43:04 AM
 #344

With all this in mind I've designed a new logo.  I've kept the color scheme to keep it looking as close to the existing logo as possible.  The coin image didn't change.  I've changed the font to look more modern, but it still has a similar look and feel.  Most importantly, the zero and letter 'o' are clearly different.  Compare them for yourselves.  I welcome feedback, both pro and con on the new design.
I like the idea of making the logo more distinct. mmpool gets a lot of people confusing iocoin for i0coin. Is it possible to use a digit 0 with the diagonal line going through it to more solidly identify it as a number?

mmpool.org - BTC Pool - DGM - pays tx fees/vardiff/merge mining/tor
brooksby
Full Member
***
Offline Offline

Activity: 199
Merit: 100


View Profile
January 03, 2016, 07:50:27 PM
 #345

With all this in mind I've designed a new logo.  I've kept the color scheme to keep it looking as close to the existing logo as possible.  The coin image didn't change.  I've changed the font to look more modern, but it still has a similar look and feel.  Most importantly, the zero and letter 'o' are clearly different.  Compare them for yourselves.  I welcome feedback, both pro and con on the new design.
I like the idea of making the logo more distinct. mmpool gets a lot of people confusing iocoin for i0coin. Is it possible to use a digit 0 with the diagonal line going through it to more solidly identify it as a number?

I'll see what I can do.
sidhujag
Legendary
*
Online Online

Activity: 1918
Merit: 1002


View Profile
January 06, 2016, 11:57:19 PM
 #346

(It seems like you are saying the development branch is what people currently get at the main URL?)

It is, and that is the same as just about any other software project out there (including Bitcoin).  If you want stable code, use the releases (tags).  If you want to stay on the latest code, use the main branch.  Note that I believe the branches to be mostly bugfree, but they are nevertheless no explicitly created and tested releases.
Did you write unit tests that pass on risky code? Bitcoin is not the same as yours because the code is tested and isnt put in a state where ppl can pull broken code especially if test coverage is above 95%... The releases happen after integration testing. So you should not recommend pullimg master or dev branch unless ots unit tested at the least

Isn't that exactly what I wrote?  I suggest to use the release tags, unless you know what you are doing and insist on pulling the branches (but I'm definitely not recommending doing that).  This is what I wrote.  Note that I actually do test the code before I push, but of course not as thoroughly as the release tags are tested.
No because bitcoin will run the unit tests on development code nightly, the master will break if code pushed invalidates it and through peer review the unit tests are pushed along with daily commits to ensure reliability of master, one simply does not run a binary from master if the unit tests have failed.

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

★☆★Syscoin - Decentralized Marketplace and Multisig Platform
Pay with Bitcoin, ZCash and many more
For more visit Syscoin.org  ★☆★
domob
Legendary
*
Offline Offline

Activity: 998
Merit: 1007


View Profile WWW
January 07, 2016, 07:00:29 AM
 #347

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.

Use your Namecoin identity as OpenID: https://nameid.org/
Donations: 1domobKsPZ5cWk2kXssD8p8ES1qffGUCm | NMC: NCdomobcmcmVdxC5yxMitojQ4tvAtv99pY
BM-GtQnWM3vcdorfqpKXsmfHQ4rVYPG5pKS | GPG 0xA7330737
brooksby
Full Member
***
Offline Offline

Activity: 199
Merit: 100


View Profile
January 09, 2016, 11:10:24 PM
 #348

With all this in mind I've designed a new logo.  I've kept the color scheme to keep it looking as close to the existing logo as possible.  The coin image didn't change.  I've changed the font to look more modern, but it still has a similar look and feel.  Most importantly, the zero and letter 'o' are clearly different.  Compare them for yourselves.  I welcome feedback, both pro and con on the new design.
I like the idea of making the logo more distinct. mmpool gets a lot of people confusing iocoin for i0coin. Is it possible to use a digit 0 with the diagonal line going through it to more solidly identify it as a number?

After thinking about making the logo more distinct, I came to the conclusion this can't be done without updating the whole coin design.  Most of the time, across the numerous exchanges and information sites, the coin image will represent I0coin without the rest of the logo being present.

Below is my updated coin design.  In this new design I include both the numeric "0" with a slash and the written word "zero".  This should remove confusion about it being a zero or letter "o".  I dispensed with the "I".  It is not really needed, and it looked like an "X" when I put the slash through the "0".  As the coin is reduced in size the "zero" words become less visible, but the numeric "0" stays visible to quite small coin sizes.  I've provided various sizes for comparison.  The color of this design is more of a golden hue than the orange of the existing.  If the community likes the new design I'll also provide them for download on the website.  

Again, feedback is welcome.  This coin redesign is the first step.  Once it's acceptable I'll complete the rest of the logo.

New coin design for I0coin:






sidhujag
Legendary
*
Online Online

Activity: 1918
Merit: 1002


View Profile
January 10, 2016, 06:23:46 AM
 #349

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.
So its just a tweak of params? Whats different about this coin? Why the need to clone but not offer any credible substance over the trunk?

★☆★Syscoin - Decentralized Marketplace and Multisig Platform
Pay with Bitcoin, ZCash and many more
For more visit Syscoin.org  ★☆★
domob
Legendary
*
Offline Offline

Activity: 998
Merit: 1007


View Profile WWW
January 10, 2016, 12:34:58 PM
 #350

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.
So its just a tweak of params? Whats different about this coin? Why the need to clone but not offer any credible substance over the trunk?

Not sure what you mean here.  It is an alt-coin like 99,9% of the others out there - with its own blockchain and slightly changed parameters.  Apart from that, it is based heavily on Bitcoin's codebase plus merge-mining.

Use your Namecoin identity as OpenID: https://nameid.org/
Donations: 1domobKsPZ5cWk2kXssD8p8ES1qffGUCm | NMC: NCdomobcmcmVdxC5yxMitojQ4tvAtv99pY
BM-GtQnWM3vcdorfqpKXsmfHQ4rVYPG5pKS | GPG 0xA7330737
sidhujag
Legendary
*
Online Online

Activity: 1918
Merit: 1002


View Profile
January 10, 2016, 03:47:09 PM
 #351

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.
So its just a tweak of params? Whats different about this coin? Why the need to clone but not offer any credible substance over the trunk?

Not sure what you mean here.  It is an alt-coin like 99,9% of the others out there - with its own blockchain and slightly changed parameters.  Apart from that, it is based heavily on Bitcoin's codebase plus merge-mining.
The ones that have any attention are those that actually have innovative code on top of bitcoin.. Theres not many clones with just tweaked params.. Sorry but to me thats a waste of resources.. To each his own i guess

★☆★Syscoin - Decentralized Marketplace and Multisig Platform
Pay with Bitcoin, ZCash and many more
For more visit Syscoin.org  ★☆★
domob
Legendary
*
Offline Offline

Activity: 998
Merit: 1007


View Profile WWW
January 10, 2016, 04:26:59 PM
 #352

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.
So its just a tweak of params? Whats different about this coin? Why the need to clone but not offer any credible substance over the trunk?

Not sure what you mean here.  It is an alt-coin like 99,9% of the others out there - with its own blockchain and slightly changed parameters.  Apart from that, it is based heavily on Bitcoin's codebase plus merge-mining.
The ones that have any attention are those that actually have innovative code on top of bitcoin.. Theres not many clones with just tweaked params.. Sorry but to me thats a waste of resources.. To each his own i guess

This is true, and I personally agree (that's why I work mostly on Huntercoin and Namecoin).  I was contracted to do the I0coin update, so that's what I'm doing here as well.

Use your Namecoin identity as OpenID: https://nameid.org/
Donations: 1domobKsPZ5cWk2kXssD8p8ES1qffGUCm | NMC: NCdomobcmcmVdxC5yxMitojQ4tvAtv99pY
BM-GtQnWM3vcdorfqpKXsmfHQ4rVYPG5pKS | GPG 0xA7330737
tinderbox
Member
**
Offline Offline

Activity: 103
Merit: 10


View Profile
January 10, 2016, 11:24:05 PM
 #353

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.
So its just a tweak of params? Whats different about this coin? Why the need to clone but not offer any credible substance over the trunk?


Most cryptos are either exact clones of another, or just slightly different.  I0coin's change in parameters from Bitcoin are small but gives it capabilities which improve on Bitcoin.

I0coin's features (many of which are improvements over Bitcoin):

> 6.7 times faster block times than Bitcoin so users can fully confirm txs in 9 minutes, not an hour like Bitcoin.

> 6.7 times the transaction bandwidth of Bitcoin even without block size increases.

> Faster mining rate early in I0coin's life means its 21 million coins are mostly mined already.  Near zero currency inflation means new I0coin generation doesn't devalue existing coins.

> Merged mining is more energy efficient because it shares mining resources.  This also offsets a reduced desire by miners to mine I0coin since it now has a small block reward.  I0coin has remained in the top 5% of cryptocurrencies when considering network hash power.

> Block chain compatibility with all of Bitcoin's block chain technologies.

> I0coin also serves to expand overall cryptocurrency bandwidth.  It's been well advertised that Bitcoin cannot even come close to fiat currency transaction volumes by itself.  Additional cryptocurrencies which are closely aligned with Bitcoin can share the bandwidth.

tinderbox
Member
**
Offline Offline

Activity: 103
Merit: 10


View Profile
January 10, 2016, 11:39:30 PM
 #354


Again, feedback is welcome.  This coin redesign is the first step.  Once it's acceptable I'll complete the rest of the logo.

New coin design for I0coin:




I'll give a thumbs up on the new design.  It improves on the old design and reduces the confusions you pointed out and definitely looks more metallic and realistic.  My vote is yes.
sidhujag
Legendary
*
Online Online

Activity: 1918
Merit: 1002


View Profile
January 11, 2016, 06:46:56 AM
 #355

Wow even the miner_tests pass or did you run the unit tests on i0coin? I see you haven't updated the alert key and the nonce's in miner_test... unless you have the same genesis block I don't think these will pass.

I did not run the tests for i0coin, but on the merge-mining branch of Namecoin from which i0coin is forked.  i0coin itself only changes the chain parameters and branding.  The regtests and unit tests for Namecoin (and the "just merge-mining" branch) are checked for every commit there.
So its just a tweak of params? Whats different about this coin? Why the need to clone but not offer any credible substance over the trunk?


Most cryptos are either exact clones of another, or just slightly different.  I0coin's change in parameters from Bitcoin are small but gives it capabilities which improve on Bitcoin.

I0coin's features (many of which are improvements over Bitcoin):

> 6.7 times faster block times than Bitcoin so users can fully confirm txs in 9 minutes, not an hour like Bitcoin.

> 6.7 times the transaction bandwidth of Bitcoin even without block size increases.

> Faster mining rate early in I0coin's life means its 21 million coins are mostly mined already.  Near zero currency inflation means new I0coin generation doesn't devalue existing coins.

> Merged mining is more energy efficient because it shares mining resources.  This also offsets a reduced desire by miners to mine I0coin since it now has a small block reward.  I0coin has remained in the top 5% of cryptocurrencies when considering network hash power.

> Block chain compatibility with all of Bitcoin's block chain technologies.

> I0coin also serves to expand overall cryptocurrency bandwidth.  It's been well advertised that Bitcoin cannot even come close to fiat currency transaction volumes by itself.  Additional cryptocurrencies which are closely aligned with Bitcoin can share the bandwidth.


This is an exact clone of namecoin (minus name service) with tweaked params.. Most coins have atleast something to offer on top of tweaked parameters.

6.7 more transaction bandwidth? Well this means more bloat and probably cheaper dust attacks, perhaps a candidate for a unit test because i bet by simply "tweaking parameters" you will degrade a security aspect od the underlying mechanics which are all assuming certain thresholds.. This is the problem with most devs in coins they simply do not understand software enough to fully comprehend what they are actually doing but just a get rich quick by trying to fool others that hey ive offered a betted alternative which is untested and probably full of hidden bugs of which are the worse kind.

★☆★Syscoin - Decentralized Marketplace and Multisig Platform
Pay with Bitcoin, ZCash and many more
For more visit Syscoin.org  ★☆★
tinderbox
Member
**
Offline Offline

Activity: 103
Merit: 10


View Profile
January 12, 2016, 04:21:04 AM
 #356

6.7 more transaction bandwidth? Well this means more bloat and probably cheaper dust attacks, perhaps a candidate for a unit test because i bet by simply "tweaking parameters" you will degrade a security aspect od the underlying mechanics which are all assuming certain thresholds.. This is the problem with most devs in coins they simply do not understand software enough to fully comprehend what they are actually doing but just a get rich quick by trying to fool others that hey ive offered a betted alternative which is untested and probably full of hidden bugs of which are the worse kind.


If you're referring to fast block times being less secure it's no less secure than someone paying a vendor in Bitcoin and waiting only for the transaction to show up in the receiver's wallet (no confirmations).  Vendors can't wait 10 minutes for every customer, let alone an hour.  Users can set the number of confirmations to what ever makes them comfortable.  I would argue that one quick I0coin confirmation is better than no Bitcoin confirmations since you know at least one miner has validated the transaction.

As for bloat, yes, the block chain grows faster, but many other coins have similar or faster block times.  But with the recent introduction of pruning this has little affect on the average I0coin user.  Users set the size of the block chain to just about any size they want. 

Parameters are generally a balance between opposing, but equally important features.  Security vs. tx times.  Block rewards vs currency inflation.  I think it's useful for different coins to have different parameter settings to demonstrate how they work in real world scenarios.


sidhujag
Legendary
*
Online Online

Activity: 1918
Merit: 1002


View Profile
January 12, 2016, 05:26:21 AM
 #357

6.7 more transaction bandwidth? Well this means more bloat and probably cheaper dust attacks, perhaps a candidate for a unit test because i bet by simply "tweaking parameters" you will degrade a security aspect od the underlying mechanics which are all assuming certain thresholds.. This is the problem with most devs in coins they simply do not understand software enough to fully comprehend what they are actually doing but just a get rich quick by trying to fool others that hey ive offered a betted alternative which is untested and probably full of hidden bugs of which are the worse kind.


If you're referring to fast block times being less secure it's no less secure than someone paying a vendor in Bitcoin and waiting only for the transaction to show up in the receiver's wallet (no confirmations).  Vendors can't wait 10 minutes for every customer, let alone an hour.  Users can set the number of confirmations to what ever makes them comfortable.  I would argue that one quick I0coin confirmation is better than no Bitcoin confirmations since you know at least one miner has validated the transaction.

As for bloat, yes, the block chain grows faster, but many other coins have similar or faster block times.  But with the recent introduction of pruning this has little affect on the average I0coin user.  Users set the size of the block chain to just about any size they want. 

Parameters are generally a balance between opposing, but equally important features.  Security vs. tx times.  Block rewards vs currency inflation.  I think it's useful for different coins to have different parameter settings to demonstrate how they work in real world scenarios.



You dont get it, your not a developer with any kind of experience sorry. Anyways i hope the dev here raises the quality of this project by properly assessing the ramifications of the changes

★☆★Syscoin - Decentralized Marketplace and Multisig Platform
Pay with Bitcoin, ZCash and many more
For more visit Syscoin.org  ★☆★
tinderbox
Member
**
Offline Offline

Activity: 103
Merit: 10


View Profile
January 13, 2016, 03:42:58 PM
 #358

You dont get it, your not a developer with any kind of experience sorry. Anyways i hope the dev here raises the quality of this project by properly assessing the ramifications of the changes

You sure like being vague don't you.  Give us some specifics on parameter issues.

sidhujag
Legendary
*
Online Online

Activity: 1918
Merit: 1002


View Profile
January 13, 2016, 06:31:36 PM
 #359

You dont get it, your not a developer with any kind of experience sorry. Anyways i hope the dev here raises the quality of this project by properly assessing the ramifications of the changes

You sure like being vague don't you.  Give us some specifics on parameter issues.



Here is an example for you. It may or may not apply to this project but applies to one that I'm working on.

If you modify the block time from say 10 minutes to 1 minute you must assess bloat. The max size of a tx is 100k and max size of 1M per block. Logically you shoudl consider reducing max tx size of 10k and max block size of 100k, however with max tx size of 10k you can't fit a standard tx with anything above 50 or so txins so you need to keep that at 100k. Now you have the ability for a single tx to fill up an entire block (which may or may not open up other security ramifications in the system as a black box). Another thing to consider is the amount of blocks to keep by default with bitcoin's IBD push, since by default its set to a certain "size" in MB and changing max block size will affect this change because it seems optimized for this "size". It is set to 288 by default but if you reduce max block to 100k then you would want to set this to 2880 to match the same amount of data stored before pruning.

If you keep block size to 1M then you have 10x more bloat than bitcoin and that may invalidate any hope for it to survive under the logical pretense that storage efficiency will outweigh the network bandwidth constraints that would exist with 10x more bloat (this is because bandwidth costs more than storage in a blockchain network).

THis is just an example of 1 single parameter changing, now multiply that with x number of ramifications for every parameter now you have atleast 1000x extra checks to do by making something like 5 changes to the consensus model (that is an over-exaggeration but you get what I mean now?)

★☆★Syscoin - Decentralized Marketplace and Multisig Platform
Pay with Bitcoin, ZCash and many more
For more visit Syscoin.org  ★☆★
tinderbox
Member
**
Offline Offline

Activity: 103
Merit: 10


View Profile
January 14, 2016, 04:22:53 AM
 #360

You dont get it, your not a developer with any kind of experience sorry. Anyways i hope the dev here raises the quality of this project by properly assessing the ramifications of the changes

You sure like being vague don't you.  Give us some specifics on parameter issues.



Here is an example for you. It may or may not apply to this project but applies to one that I'm working on.

If you modify the block time from say 10 minutes to 1 minute you must assess bloat. The max size of a tx is 100k and max size of 1M per block. Logically you shoudl consider reducing max tx size of 10k and max block size of 100k, however with max tx size of 10k you can't fit a standard tx with anything above 50 or so txins so you need to keep that at 100k. Now you have the ability for a single tx to fill up an entire block (which may or may not open up other security ramifications in the system as a black box). Another thing to consider is the amount of blocks to keep by default with bitcoin's IBD push, since by default its set to a certain "size" in MB and changing max block size will affect this change because it seems optimized for this "size". It is set to 288 by default but if you reduce max block to 100k then you would want to set this to 2880 to match the same amount of data stored before pruning.

If you keep block size to 1M then you have 10x more bloat than bitcoin and that may invalidate any hope for it to survive under the logical pretense that storage efficiency will outweigh the network bandwidth constraints that would exist with 10x more bloat (this is because bandwidth costs more than storage in a blockchain network).

THis is just an example of 1 single parameter changing, now multiply that with x number of ramifications for every parameter now you have atleast 1000x extra checks to do by making something like 5 changes to the consensus model (that is an over-exaggeration but you get what I mean now?)

Point taken.

It wouldn't hurt to do a review to ensure parameters are optomized.  And optomized parameters today may not be in a year.  They may need to evolve based on a number of factors.  Average transactions per block, average block size, etc.

I'd expect it's a continuous, iterative process.

Thanks for the heads up.


domob, brooksby,
Do you know of any questionable parameter settings?


Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 [18] 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!