December 2015 "Fastest Crypto" Bake-Off (topic locked) - page 3.

Fuserleer

legendary

Activity: 1050

Merit: 1016

Quote from: Hangman01 on October 09, 2015, 09:25:17 AM

Ah ok. The op was never serious about testing then. If he wanted then he could test geist geld today.

Sounds like a lame attempt to get some free btc

Edit. Op should do an ipo, give me money thingy

Then tell the geist geld devs, or ask the community to chip in. 0.2BTC is fair IMO, costs of hardware to rent/buy to do the tests, electricity use, time spent setting the tests up, managing them etc...

If you dont trust the OP, then have him make a multi-sig wallet with all the developers as co-signers.

onemorexmr

sr. member

Activity: 252

Merit: 251

Quote from: Hangman01 on October 09, 2015, 09:25:17 AM

Ah ok. The op was never serious about testing then. If he wanted then he could test geist geld today.

Sounds like a lame attempt to get some free btc

Edit. Op should do an ipo, give me money thingy

OP said he wants 0.2BTC für a test. IMHO That amount is ok for the kind of setup he described and the transactions fees he has to pay.

Hangman01

newbie

Activity: 23

Merit: 0

Ah ok. The op was never serious about testing then. If he wanted then he could test geist geld today.

Sounds like a lame attempt to get some free btc

Edit. Op should do an ipo, give me money thingy

hughbt

hero member

Activity: 613

Merit: 501

Quote from: Hangman01 on October 09, 2015, 09:10:24 AM

So this thread is really a bitshares, vanilla, emunie circle jerk.

Op grow so balls, the original idea of a speed test is good, your execution of it sucks. Why not start some testing already.

It's all about money. He is not going to start testing untill there will be some contestants. But it's pretty obvious it is going to be a circle jerk, there is nothing bad about it. Some cryptos are obviously faster than the other ones, that's why there will only be few contestants(if any, but that's because of entry fee). On the other hand this is probably a lot of work so there should be a reward for the tester.

Quote

0.2 BTC Entry Fee and a pay-back BTC address. (The winner gets back his/her entry fee.) Pay to 1C6nd36KSm6sqAhjEFqd91nNpJkq4sVEUL. PM me links to the entry items and the BTC timestamp of payment.

Hangman01

newbie

Activity: 23

Merit: 0

Lets see, reading through the thread I am mistaken. There is a forth entrant, though ignored.

Geist geld. I don't know what it is but it looks old.

Edit. If people dont like what i'm saying then win this account from me.

https://bitcointalksearch.org/topic/m.12636241

Fuserleer

legendary

Activity: 1050

Merit: 1016

Quote from: Hangman01 on October 09, 2015, 09:10:24 AM

So this thread is really a bitshares, vanilla, emunie circle jerk.

Op grow so balls, the original idea of a speed test is good, your execution of it sucks. Why not start some testing already.

I'm happy for OP to do some testing of eMunie next week with the next beta.

As for the circle jerk, get some other cryptos to step up, then it won't be will it? Or the OP could start testing released cryptos that have made speed claims I guess.

Hangman01

newbie

Activity: 23

Merit: 0

So this thread is really a bitshares, vanilla, emunie circle jerk.

Op grow so balls, the original idea of a speed test is good, your execution of it sucks. Why not start some testing already.

Fuserleer

legendary

Activity: 1050

Merit: 1016

Quote from: wingspan on October 09, 2015, 08:05:08 AM

So, in eMunie, instead of nodes being blocks, you have nodes being 1 transaction only. Instead of a chain of nodes, you have a tree. Still a ledger. Still tracks publicly the order of transactions. Still allows state recovery just like blockchains. OK, sounds intriguing. Apples-to-Apples testing still possible.

Yeah pretty much, its kind of a hybrid.

There is a balance sheet ledger too that runs along side, but that has to verify against the current transaction ledger (and vice versa).

In the channeled implementation, its not really a chain nor a tree but more of a mid-way point between the two (trains or chees? lol). The chains are allowed to fork out, but only on branches one link long. Transactions are assigned a "transactional period", subsequent transactions must link to only one transaction in the previous transactional period as per a set requirements.

wingspan

newbie

Activity: 54

Merit: 0

So, in eMunie, instead of nodes being blocks, you have nodes being 1 transaction only. Instead of a chain of nodes, you have tree-like channels. Still a ledger. Still tracks publicly the order of transactions. Still allows state recovery just like blockchains. OK, sounds intriguing. Apples-to-Apples testing still possible. Anyone disagree?

Fuserleer

legendary

Activity: 1050

Merit: 1016

OK, I'm going to clear something up, as r0ach's dis-information and assumptions about how our ledger is structured even before I've finished writing the public docs to explain it seem to be causing some confusion.

Legend: Circle = Genesis, Box = Transaction/Block

First we had a good old block chain:

We scrapped block chains in Aug of 2013 for what we called "block trees"

They were themselves scrapped in Dec 2014 for "transaction channels"

As you can see in all cases, every transaction links to a previous one, in the same manner that a block does in a chain. Previous transactions can easily be audited and are public.

These diagrams are from documentation I'm currently working on regarding our ledger design that will explain all the concepts and will put this "eMunie doesn't keep any historic information" argument to bed if this post doesn't.

patmast3r

hero member

Activity: 980

Merit: 1001

Quote from: monsterer on October 09, 2015, 02:42:46 AM

Quote from: Fuserleer on October 08, 2015, 08:55:56 PM

Why? If both provide a public record, are secure, provide consensus, what exactly is the difference?

The big difference between having a chain and having no chain is that there is no recognisable ordering of transactions in a chainless system, the 'current state' is all that matters. With a chain, you can see the ordering, which makes validating the state a lot easier.

However, I believe emunie does have a chain, so this probably isn't relevant?

Is there any such system in crypto land that really doesn't have any sort of chain ?
Does a ledger (and I beliebe that's what emunie is using) not count as a chain ?

monsterer

legendary

Activity: 1008

Merit: 1007

Quote from: Fuserleer on October 08, 2015, 08:55:56 PM

Why? If both provide a public record, are secure, provide consensus, what exactly is the difference?

The big difference between having a chain and having no chain is that there is no recognisable ordering of transactions in a chainless system, the 'current state' is all that matters. With a chain, you can see the ordering, which makes validating the state a lot easier.

However, I believe emunie does have a chain, so this probably isn't relevant?

Fuserleer

legendary

Activity: 1050

Merit: 1016

Quote from: r0ach on October 08, 2015, 08:16:00 PM

Hughes should be required to actually release it first before any of this discussion happens.

Who made you law maker around here? The same should be said for other cryptos too then, as theres much discussion regarding various aspects of those before they are released, but I don't see you complaining there.

Quote from: r0ach on October 08, 2015, 08:16:00 PM

Like I said earlier, ledger/balance sheet vs blockchains are not really an apples to apples comparison either.

Why? If both provide a public record, are secure, provide consensus, what exactly is the difference?

Quote from: r0ach on October 08, 2015, 08:16:00 PM

The thing also has to exist in the wild for a while so people can diagnose if the system is functional in the real world. It will then take months or years for people to discover all the miscellaneous vulnerabilities such as Bitcoin faced like faking shares, selfish mining, etc.

Since there is no real finite resource at play here (stake or PoW hash power), the simple act of modifying the client might open big holes.

Oh come on, like this doesn't apply to all software, crypto or not. The presence of POW, POS, POI, or whatever else is completely irrelevant!

r0ach

legendary

Activity: 1260

Merit: 1000

I'm not sure why there's so much rush to test Emunie vs other cryptos. Hughes should be required to actually release it first before any of this discussion happens. Like I said earlier, ledger/balance sheet vs blockchains are not really an apples to apples comparison either. The thing also has to exist in the wild for a while so people can diagnose if the system is functional in the real world. It will then take months or years for people to discover all the miscellaneous vulnerabilities such as Bitcoin faced like faking shares, selfish mining, etc.

Since there is no real finite resource at play here (stake or PoW hash power), the simple act of modifying the client might open big holes.

wingspan

newbie

Activity: 54

Merit: 0

even though there are two more months, if people would like to have me test early and then re-test later, I'd be happy to test your crypto exactly as I would in December - and I'll publish those early results if you like how you do. That might encourage others to keep tuning their system...knowing what they are up against.

wingspan

newbie

Activity: 54

Merit: 0

Quote from: StanLarimer on October 07, 2015, 11:24:38 PM

...
Hopefully somebody can set up such a test using our code base.
Unfortunately we're launching two real-time blockchains this month (BitShares and PeerTracks' MUSE) so we're a bit busy.
But we'll be keeping tabs on how this goes.
Thanks
Stan

I will postpone the October testing by 2 months - so those needing time can still participate. Thanks - good luck preparing your crypto. I'll update the OP, now, to reflect the delay.

Fuserleer

legendary

Activity: 1050

Merit: 1016

Quote from: ?? on ??

it's all good... While we were busy fighting over which free public crypto has the biggest pecker (and distracting our brothers), we all became criminals.. nice knowing ya

http://www.zerohedge.com/news/2015-10-08/how-tpp-could-lead-worldwide-internet-censorship

The fact that you have time to spend on "bad grammer" means that you are indeed in the wealthiest 1% of all humans on earth. Enjoy that internet while it lasts..

For what its worth your comment warmed my heart

wingspan

newbie

Activity: 54

Merit: 0

I thought it was sarcasm. I really don't know you. When you said: "e-munie - faster than the fastest crypto on the planet!" I figured the misspelling of eMunie and the hyperbole was a sign you didn't like it. That's all.

wingspan

newbie

Activity: 54

Merit: 0

Quote from: ?? on ??

...

Besides the sarcasm, any thoughts on easy ways to make my test fairer? I think I'll add a step where I try to measure peak performance - maybe after the three 100-minute tests assuming all nodes are still in sync. Also, I am thinking I should run the network after the third test and after the peak test for 24 hours - or until the network breaks (3 of 20 nodes fall out of sync for more than an hour) - to test that capacity. Anything else I should add to the main 100-minute tests?

BTW, I'm "wingspan" on the eMunie forum...if you wish to see my history of posts there, and pic. I live in the Pacific timezone not too far from Seattle.

Fuserleer

legendary

Activity: 1050

Merit: 1016

Quote from: bitcoin carpenter on October 08, 2015, 10:35:08 AM

So who is winning so far?

Could we get a tally of claimed tps
And a tally of tested tps for projects?

The bake-off hasn't started yet, so there are no "official" results from the OP.

We had a beta test about a week ago where we achieved a 2400 tps peak - https://bitcointalksearch.org/topic/emunie-the-fastest-crypto-currency-1191535 Theres another beta scheduled soon, and amongst other things we'll be doing further load testing with high throughput, we should see 5000 tps peaks and 500 tps sustained easily.

Bitshares have done some stress testing too over the last week or so, IIRC they achieved a peak of ~1200+ tps but ran into some issues they need to solve - https://bitsharestalk.org/index.php/topic,18684.0.html - I've no idea what Bitshares can reliably sustain though, as they've dialed the test net back to concentrate on functionality testing.

Other than that, there isn't much else that has done stress tests in a verifiable environment. Lots of claims of x thousand tps from developers and project supporters, but nothing to back up these claims really other than hear say and developer lab testing.

Topic: December 2015 "Fastest Crypto" Bake-Off (topic locked) - page 3. (Read 5720 times)