I think non-ECDSA and non-SHA256/Scrypt POW are by far the best 2 features.

Are there any coins that use a different POW?

I made the thread as a spur of the moment thing, frankly i am adding the other coins as I think of them, i will edit and add them to OP and 2nd post.

I count at least 42 coins listed on the OP of https://bitcointalk.org/index.php?topic=134179.0 - and I dont know how many awful scamcoins were skipped.

There have been a handful of coins which attempted either an inflationary or an unlimited currency supply, i do not recall these offhand.  they were not met with success [well, none except BTC and LTC have thus far].

I suspect inflationary or unlimited currency coins will never succeed in the cyrptocoin world, the community generally hates these things.

Namecoin actually implemented a P2P domain system, another innovative and possibly useful design

Coins which have actually tried something fundamentally new and novel, for better or worse.  In order of their introduction.


Novel Proof-of-work:
Bitcoin -- 1/3/09 -- SHA2-256 originator and novel in nearly every regard.  Gold standard
Tenebrix -- 9/26/11 -- Scrypt originator. Was premined and thus died off.
Fairbrix -- 10/2/11 -- Scrypt.  Initial launch crippled by bad config.  Second launch attacked.  These problems led to its demise.
Litecoin -- 10/9/11 -- Scrypt  Current Gold standard for scrypt and altcoins.  Avoided problems of Tenebrix and Fairbrix
PPCoin -- 8/19/12 -- Hybrid proof-of-work/proof-of-stake
YAC -- 5/5/13 -- SHA3-512 instead of SHA2-256.  Cacha replaces Salsa.  scrypt(N,1,1), N increases over time.

Distributed P2P DNS
Namecoin -- Apr/2011

Coin redistribution to open source projects and developers
Devcoin -- 8/5/11 -- Redistributes 90% of block rewards to open source projects chosen by the developers of Devcoin(?)
Freicoin -- Dec/2012 -- 4.89% fee on any coins held per year to developers. 80% of block rewards to developers over 3 years.


Infinite Coin supply
Devcoin -- 8/5/11 -- Constant block reward.  Until this point all coins had a geometrically decreasing block reward with a finite ultimate sum
There may have been another coin prior to devcoin that did this, i will update if informed of one

Random block rewards
Junkcoin -- 5/4/13 -- Small % Chance to get 3x or 20x regular block reward

Ascending rewards to mitigate premining/early mining of the network:
Nibble -- 5/18/13
Digicoin -- 5/18/13


Multiple coins launched with a feature nearly concurrently are listed with their approximate dates

Bitcoin wiki list of altcoins
PPCoin's History of cyrptocurrencies on github

Let's brainstorm.

Litecoin added a legitimate feature and utility that bitcoin did not have -- a different proof of work, replacing SHA256 with Scrypt.  Allowing for defense against established hashing power in the SHA256 networks.

It also added faster block times, but there is no change to the fundamental security of the network by doing this.  Security and confirmation confidence is related only to the length of time since the transaction first was broadcast, and the amount of hashing power during said time.  Faster block times should not be considered a legitimate reason to use litecoin vs bitcoin.  It should also not be a legitimate reason to use a 'faster' coin over a slower one.

Since then, I have seen dozens, maybe even 100+ "altcoins" launched which have been absolutely embarassing and essentially transparent get rich quick schemes.


Things which are not new features and pretty much just certify that you're making a scamcoin to mine and get rich quick if these are the only things your coin changes

* Changing block time
* Changing difficulty
* Changing the # of blocks per retarget
* Changing reward
* Changing reward change [i.e. block rewarding halving/etc]
* Using Scrypt or SHA256
* Incentives for early adopters
* Not announcing your launch at least several days in advance
* Any mining before announcement
* Launching earlier than stated -- later is fine.
* Setting starting difficulty extremely low.  At this point there are way too many people with way too much has power ready to jump on every single new coin even if it's an obvious fucking scamcoin.  Set your difficulty so 10Mhash will take several hours to mine the genesis block


Some things which would be good to have in an alternate coin

* Replace ECDSA with a Lattice-based asymmetric algorithm.  All cyrptocoins currently are theoretically vulnerable to Quantum computers capable of implementing Shor's algorithm.  A lot of things are indeed broken by this, including systems used by governments, banks, everyone.  But only crytpocoins have direct ownership and control of the coins based ENTIRELY on the knowledge of the private key.  Using the hash of the public key for the address may offer some defense, but it is still a massive security breach.  It is generally believed that if these quantum computers exist, they are classified information in the hands of governments.

*Non-SHA256/Scrypt POW.  Scrypt was originally intended to be a CPU only POW, and it was. for a while.  Then someone figured out how to make cgminer work with it.  There was some controversy as to who did this first and if they abused it for a while to get a massive hash advantage.
Any new POW should either be fundamentally secure against this sort of thing (i.e. someone could figure out how to use GPU or FPGA/ASIC SHA256 hashers on it while everyone else is using CPUs) or several days/weeks prior to coin launch - release a cgminer so that everyone in the community can use GPUs.  However at this point, you might as well just use scrypt.  
On one hand, this may be hated because the vast majority of altcoiners are GPU miners who would howl if they couldn't use GPUs.  On the other hand -- you can use those idle CPUs on your GPU miners to mine said coin while your GPUs mine another coin. a _TRUE_ cpu-only POW would have great promise.

* Low reward at start to offset 'premining' and early hash power consolidation which leads to orphan-mania.  Based on how much hash power gets thrown at basically any coin when it launches, you probably cannot combat the orphan rate, but you can minimize the absurdity of how many coins  they can farm.  Also reduces problem of mining large numbers of coins early to bribe people with bounties unfairly.  Bounties should require honest work and hashrates to achieve, just like on a mature network like current BTC/LTC

* Replace the "retarget at X time" / "retarget at Y block" with a "retarget if average time between blocks for the last A blocks exceeds Z seconds".  This prevents a situation like FTC where the mining rate SKYROCKETS, difficulty lags slowly and then skyrockets, and then all that hashing power moves off, leaving a miniscule hashing rate to fight a massive difficulty and find noblocks.

* Different network paradigm.  One possible way might be to run the network by default, or even by requirement, on a network like Tor.  I believe some of this is built into bitcoin but i think it is only used to hide individual people by connecting through the network and using the exit nodes of tor.  A better method would be to do everything within the onion network, no exit nodes.

Generally Bad Items(tm) Which You Shouldn't Do

* Centralize and/or close source anything (Ripple).  Cyrptocurrency means P2P and open source

* Offering exchanges coin to put your coin on them.  Obviously this will be next to impossible to actually discover.  Exchange should pick up coins because said coins have become popular based on their own merit.  Hopefully merit based in technology, not because it's the flavor of the week.

* Jumping directly to the livenet.  There's a testnet for a reason.  Use it.  Completely skipping the testnet, or doing testnet things in secret, only makes you look like an incompetent scamcoin

--

I will add to this post as (if?) more ideas are proposed

With any luck, developers of new coins can use this as a reference or a guide to making a useful alternative cryptocurrency.

I've been playing with it but having to tweak the -E -Q settings as the average time between blocks varies so much is annoying

using -s 1 -E 10 -Q 0 atm

00:27:51

getnetworkhashps


00:27:51

2288605285

Litecoin network is at 18.7 Gh/sec

It has never been much above 20 Gh/sec

There is not >22.7 Gh/sec scrypt hashing power in existence.

The hashing power has already moved off, the network is lagging to update.

Watch 'getnetworkhashps' or 'getmininginfo'

it's falling off fast.

it is literally below litecoin now

Some day people will learn that MINER ATTENTION TO YOUR IMMATURE ALT-COIN = CRASH AND BURN

the solution is an altcoin that does not use mining.

good luck with that

up for me

Theoretically can you make the difficulty readjust if no block is mined in X seconds, or the average mining time for the last N blocks is Y seconds or longer?

Seems like it would work better against these massive movement of hashing power

HAHAHAAH

THE COIN HAS ALREADY BEEN DUMPED

0.007

MASSIVE HASH POWER MOVEMENT BACK TO LTC

IT'S ALREADY OVER

It's already over.   The big 100+ MH/sec people who took all the blocks and orphaned the fuck out of the blockchain have already cashed out

this is going to happen over and over until people learn what is going on.

literally the new feathercoin

Try block time of one plank time

sockpuppets are common, anyone who isn't an idiot at this will use them

nobody is going to touch a coin that someone premined a massive % of coins.  There's literally 5,6 maybe even SEVEN figures of profit on the line, the people premining need to keep the illusion that it's a coin "of the people" before they dump.

coins with very short block times demonstrate the ability to take almost 100% of the rewards with enough hashrate to orphan everyone else trying to mine

This is almost certainly intended behavior by their creators.  It allows them to premine while maintaining the illusion of "fairness".

Yes, even with smaller block rewards for early mining.

It is literally a race of seconds for the first person to throw several dozen Megahash at the network.

once you have it, you control it.  Everyone else will be orphaned.

Someone should make an altcoin based on one of the emerging asymmetric encryption algorithms thought to be secure against quantum computers (which implement shor's algorithim)

THAT would actually have some value.  THAT would be the most conservative & secure coin, even more than bitcoin

oh wait

that would require programming and mathematical and technical skill

BTC and LTC are literally the only coins with any vendors that use them at all

the others are 100% exchange, speculation, pump and dump, and mining.  or perhaps gambling within the coin itself or with other altcoins, aka basically an exchange with dice.

not to mention they have no features beyond BTC/LTC.  less block time = less security

you cannot have your cake and eat it too