Topic: The state of crypto - The only serious thread on the subforum - page 7. (Read 19483 times)

What about network latency?

https://en.wikipedia.org/wiki/Latency_%28engineering%29#Packet-switched_networks

What if nodes are at opposite ends of the network and cant be located close to each other?

https://en.wikipedia.org/wiki/Speed_of_light

What is the actual envisioned confirmation speed of a regular transaction operating in normal network domains?

If the transaction rate can scale up sufficiently the HFT would be eaten by transaction fees before they could consume the entire network bandwidth. Also HFT operates on the principle of having the closest connection to the trading exchange, thus I assume they will run their own node in my system and scale up to what ever they need using as many servers as they need.

I envision millisecond confirmation speed.

For that one I will believe it when I see it.  For a different topic, Smooth and I were talking about decentralized exchanges and I brought up that HFT (high frequency trading) is a form of PoW itself in practice that demonstrates the end game for Bitcoin PoW.  You're fighting over a finite resource (skimming profits) where even the speed of light matters.  Parties are expending resources to position themselves to capture this supply, and eventually it's a winner take all monopoly or something extremely close to it...

Smooth argued that an equities DEX would be useless because you would get front run by the HFT.  The Bitshares 2.0 dex will probably have 3 second blocks at launch, which will lower down to 1 second without a hard fork.  At 1 second blocks you can't trade as fast as the HFT systems, but I would argue there's a large chance the decentralized exchange would gain more credibility over the HFT systems over time, and the HFT might just be left with two bots trading against each other on a system nobody uses.

Node specification?

You're talking about each node processing between 10,000-100,000 transactions per second if the network is perfectly partitioned in some manner, and these partitions are 100% independent of each other....which I find hard to believe if I'm honest.

1000 - 10,000

What kind of network size do you require to achieve 100M TX/s?

Oh in fact my proposed design is expected to scale to a 100 million transactions per second easily. Even Bitshares doesn't scale to 100s of DEX running simultaneously. Again you are going to want to know the details, and again my answer will be wait until I finish the implementation, then I will release all the details and code into the open source.

I am specifically targeting to have the best real-time transaction architecture among other improvements.

Okay that is enough about my vaporware. You raised the issue otherwise I wouldn't have mentioned it.

Let's instead focus on the linked post you quoted.

An example of a private chain is a decentralized forum or ebay market place. It may or may not be private, but at least we can say it is orthogonal to the currency block chain.

One issue is who pays for and provides the resources of mining and storage of the decentralized chain. You've got to have an open market for that otherwise network effects don't scale. You don't see TCP/IP run by a corporation that issues shares.

Bitshares can't possibly succeed. Sorry I hope you don't get angry at me, but I am telling you frankly that in my opinion Daniel's weakness has always been that he doesn't understand scaling, end-to-end principle, and other scaling principles and economics. He is a talented coder I guess. And apparently several talented coders in his community. I wish him the best.

I thought it was hilarious because he is right in saying all cryptocurrencies are basically Rube Goldberg machines.

Your statement is correct depending on the context, as is mine.

Most people however that read vaporware will consider the position of "it doesn't exist", as opposed to the truth in this case of "it isn't ready for general release"


Another egomanic, whose comments I'm not going to acknowledge.

There is a saying "The loudest in the room is the weakest" and its never more true than it is when applied to loud developers (of which there is no shortage around here)

Blockchain 3.0
I don't know if anyone has finalized the metrics concerning what "Blockchain 3.0" is yet, but this guy's post + my response to it cover pretty much what it will be and what metrics are important for running such a thing:

https://bitsharestalk.org/index.php/topic,18434.0.html

Things like Bitcoin and PoW are so dead in the water it's not even funny to me.  Whatever dream system you're trying to build, Anonymint, I don't think you're designing it around the necessity of having to run a fast global dex + all these auxiliary features on top of it.


In other news:
For Kelsey, in honor of Litecoin making an appearance on "Buttcoin":

https://www.reddit.com/r/Buttcoin/comments/3kr27t/litebutters_cant_figure_out_why_the_rats_are/

Without getting into a semantics argument, let's just say for the purposes of this thread, the term vaporware applies to anything not being traded on exchanges or being actively, natively accepted in commerce.  Emunie can't be evaluated until released into the wild because of Realsolid's half correct, half i-don't-know analysis below:

[20:57] <@Realsolid> no one is going to understand emunie to ever work on it besides the dev
[20:57] <@Realsolid> bitcoin struggled and it was much simpler
[20:57] <@Realsolid> bitcoin , for what it needs to do, is a rube goldberg machine
[20:57] <@Realsolid> emunie is like rube goldberg to the square
[20:58] <@Realsolid> it means it will be vulnerable, overly complex systems always are, bitcoin was very vulnerable early on
[20:58] <@Realsolid> im one of the best coders in the world and even i sometimes make mistakes, even when developing ultra efficient killing machines
[20:58] <@Realsolid> let alone a rube goldberg ^ 2
[21:01] <@Realsolid> im one of the best coding, best debugging, best looking motherfuckers you ever seen

There is some good information there for sure.  Thank you for the right up its appreciated.

Well I guess I must agree. Thanks for making that point.

Edit: let me add that for as long as the individual miners can't decide what they put in the block for their mining shares, then my statement that the entropy of proof-of-work is unbounded needs to be corrected to bounded, which is the same as for proof-of-stake.

i've certainly read this thread, how about less arrogance more input, i mean you spin the same shit ad nauseam, but you're not really saying anything.

stop baffling on about me being troll here with litecoin, hell i attack and troll litecoin more then most coins, there's no sentimental attachment to litecoin or any coin here.

i've said many a times the only thing good to come out of crypto so far is the idea.

yes its self evident that litecoin todate is the best we've got, and if you asked my why, and to my reply you wouldn't listen just troll me saying i'm a litecoin troll.

I disagree that eMunie's is "vapourware".

1. It is function in the client now
2. Anyone can test the client out
3. The database of the client can be queried for the consensus components.
4. More documentation going up this coming week
5. Finally, I can provide source code relating to the consensus should anyone be willing to be legally bound not to leak it.

That to me isn't "vapourware".

Come on smooth. You've been a serious programmer long enough to observe that demand for bandwidth and CPU fills up any new supply. If the new standard for bandwidth is 1000X greater, then users will be doing 1000X more microtransactions.

And you've been a serious programmer long enough to become very acquainted with the very important factor that big O algorithmic complexity drives your big wins in terms of performance and other attributes that hinge on it.

These are the sort of intuitive priorities that are just no brainer to any senior level programmer.