2:40. Translation: POS is far more centralized, and corruptible, than POW, as the more stake you have in the network, the more "validation," and control, you can take in the network. It removes the competitive aspect of POW, and replaces it with "Payola."
The 51% attack for Proof of Stake was addressed incorrectly. You don’t need 51% of ALL currency you need 51% of all stakes. Actually, you might even need less if you game the selection process correctly. A successful attack on proof of stake can have malicious nodes validate their other malicious nodes, and if done long enough can sneak in malicious transactions before getting caught by which time it’s too late.
@@BeenDeliveredToday I think if every DOGE holder had a chance to find a 10k DOGE block there would be tons of individual nodes all over the world... I think it would be 100x more decentralized than mining
@@gimrexymcity4833 you can if you have 51% of the stake like the comment said it’s like 51% attack on PoW you just need 51% of the Mining power but in reality 51% aren’t worth it you ruin credibility of a coin and it becomes less worth it’s more worth using all that power in validating blocks
PoS centralizes more over time than PoW, because larger stakes get more rewards, therefore their size increases. PoW has its own downsides but it is more consensus proof.
You introduce a diminishing returns parameter that, after a certain threshold, your rewards in the stake pool diminish. Encouraging new people and even the stake pool itself to either set a new one up or find a new one thus incentivising decentralisation whilst disincentivizing centralisation. This is not a mark against PoS itself, only networks that don't have a work around.
Honestly? that's a good thing PoW gave the opportunity to the community to regulate their own currency and the only thing it caused was component hoarding, a chip shortage and irreversible damage to the environment Decentralized currencies are a nice idea but not at the expense of everyone else trying to live their lives
@@HardcoreDuce Introducing new pools helps decentralisation. Remember, you still need delegators (unless you are insanely rich, however, if the network is large enough, Bitcoin for PoW or Cardano for PoS, then the cost to perform an attack would be more than any benefit you'd gain), if the network and community see a whale gaining too much stake they can choose not to delegate to them in their current and/or next pool's. This already happens in essence over in POW systems, where let's say for example Ergo has a disclaimer on their Reddit asking users to switch to mining pools away from the number one for the sake of decentralisation. However, as far as I am aware, you don't get financially incentivised to switch like you would with the work around I mentioned for pos.
Major misconception here, proof of stake with a select few validators isn't more decentralized than a proof of work consensus algorithm. This is the reason bitcoin remains the most decentralized blockchain in the world still in 2021.
Uh no, It's because bitcoin started back early before the arise of Dapps. New Dapps now have more explosive popularity and growth in their infancy which proves my more potential than Bitcoin. But none the less both of them are decentralized.
The best and most simple explanation of the difference between Proof of work vs. Proof of stake that I have seen. Really great job Savjee, thank you so much!! :)
@@papaulogamerofc946 1. You are in a pool for Proof of stake. The pool consists of several other miners and as a whole you guys create the coin. 2. There is only 1 blockchain. 3. It's a mathematical equation that uses computing power. Whoever has higher computing power will have higher odds of winning - there is no definitive answer, only odds. 4. not sure if this is a question
I've been watching a bunch of videos explaining proof-of-stake, and this is the only one that actually contextualizes both PoS and PoW within how they interact with the blockchain.
Thank you for putting in the time and effort to create these videos. I’m basically a caveman when it comes to computer science and tech literacy and you’ve really helped clarify a lot of abstract and complicated topics on your channel. Liked and subscribed brother. Keep up the good work!
Proof of stake in Cardano is amazing: anyone gains: you can delegate your money to small pools which are incentivized over bigger ones. Every time a block is created the rewards are split across all participants, simply beautiful.
The centralization of mining pools isn't actual centralization because individual miners are free to leave any time. This is not a centralization threat. The power issue can't be viewed in a vacuum and one must consider the power that us not used because the transactions are happening elsewhere. There are places for both PoW and PoS. It's not as simple as this video makes it out to be.
Country Corner but miners would immediately disconnect from the pool. Miners get paid in coins/tokens. Do you think they are going to expend energy to mine something that would immediately become worthless upon a 51% attack? The only way a pool could pull this off is if they owned every miner that was connected and even then they would only be able to maybe falsify one block. The amount of work to falsify 2 or 3 especially in a network like bitcoin is impossible even if pools coordinated an attack.
Proof of stake will: 1. Encourage hoarding (and hinder spending). 2. Monopolistic behavior. Not sure I trust the masses anymore than I trust govt to limit monopolies and greed, which means, I see no answers to Bitcoin's "problems."
And what is going to stop people from pooling their bitcoin into a single account to recreate the same problems as mining pools? NOTHING There are no answers.
Great explanation, but nowadays, some coins allows staking pool. I agree that it may reduce electricity cost but doesn't necessarily solve the centralization problem from poolings.
As an ETH miner I'm stoked about it. This could bring in a lot of diversity into crypto and subsequent liquidity. PoW will have a place because gigahashes of computational power doesn't just disappear. The PoW miners will help close the gap that currently exists between ETH and other cryptos. Fee payouts then can be moved into stakes, offering something other than price action volatility, which exacerbates that problem, for further investment. When ETH goes to PoS we can hedge on the liquidity of that network.
Hehe, it's still random, just not completely random. Your odds are determined by your stake. But equal stake means equal chances, means random selection.
@@davidz2690 if your forced to add "not with uniform..." so it's not "completely random" as general public assumes it, with seems "fully random with uniform odds". Like in a game of dices, with strictly the same weight for each dice and for each face of each dice. A "fair" random. SimplyExplain is just better than you to explain clearly without acronyms or technical concepts ;)
The problem of Pools will not be fixed by a transition to PoS. You will need 32 ETH to become a validator. Which is $78000 (23.01.2022) right now. How many people have this amount of ETH in one wallet? Probably way less people than people owning a sub $78000 mining rig. Maybe in PoS having a lot of money is less beneficial than having a lot of money in PoW, however a lot of people will be excluded, if they don't join a pool.
I have a question. Cant validators come together and form 'validator' pools just like the mining pools we know about? Where fees are paid according to the security deposit that they have within
Yes it takes a lot less electricity but it isn’t more decentralized. There’s staking pools just like mining pools. And I believe its more centralized cause it’s easier to buy a lot of crypto then getting a lot of miners. Plebs have max 1-5 eth not 32...
For the first time, I feel like I understand PoS. Every other explanation I got for it never actually mentioned that validators were actually doing some calculation work when they were chosen. Thank you!
It seems that various currencies are using various approaches. One way is to multiply the stake of a validator by the time since its last validation. For example, if: V1 has a stake of 100 for 2 days = 200 V2 has a stake of 10 for 30 days = 300 Then everybody agree that V2 has to be the next validator. After the forging, V2 has a stake of 10 for 0 days. Another more "random" way might be to multiply the stake of a validator by an hash of (validator id + last block). This way a random node is selected, but everybody choose randomly the same validator.
@@harishkumargujudhur2234 No hashing does not mean PoW in PoS model. In PoW, you have to do tons of hashes on the same block again and again only modifying its nonce until the hash result matches a specific contract. It requires a lot of tries to find a valid nonce and these tries take time and energy. In case of PoS as explained above, you would only have to hash once per validator to determine the next validator. Easy task, even if there are many validators. No competition, no need for gigantic mining farms. To do a 51% attack on bitcoin you need half of the hashing computing power (that means only the 4 biggest bitcoin mining pools). To do a 51% on a PoS currency, you typically need 51% of the coins. Some people think that if you own 51% of a currency, you won't attack it because the currency will lose its value and you will lose all your investments. So PoS should be a lot less targeted by 51% attacks.
@@rt1517 Thank you very much. I'm currently leaning on PoW. So, feasibility-wise you mean PoS is less secure, but in terms of the desire to take the 51% attack potential, PoW has a higher competition. (excluding less popular PoW coins not targeted as Store of Value) Based on your explanation concerning the need for gigantic mining farms so far, it has a higher energy demand for attacks than PoS.
@@rt1517 Also, I would be really grateful if you would point out what other consensus algorithms use more than one hashing to check for a valid nonce. I'm much interested in non-identificational Proof-of-Weight. I don't know much about use of hashing other than in the form of basic OTP passwords.
PoS is less fair, because staking is a one-time investment. With PoW OTOH miners continuously have to pay their electricity bills. Banks are Dollar stakers who decide which transactions go through. Besides you are contradicting yourself: First you rightly say, that it is much easier to stake than it is to mine and later you wrongly claim, that it‘s harder to do a 51% attack with PoS. It clearly is harder to purchase mining equipment and set it up than to just buy enough stake in order to do double spend attacks.
This video advertise etherium secretly. It is a total misinformation. Most of this new coin are PoS. I think elites are behind this to stop Bitcoin. I think they also manipulate the GPU prices so that they halt miners from mining while advocating for etherium which is secretly centralized.
i am doing a research project on nfts, and this is by far the best video i've seen explaining PoS. Same goes for your other videos on blockchain and other stuff. #1 channel for actually understanding all this blockchain crap 10/10
I get the power argument here, but the whole point of crypto was to be decentralized, with proof of stake it is now pay to play and might as well be another fed.
Unfortunately, crypto has yet to be successfully decentralized. Bitcoin failed to achieve the goal, even if you ignore power consumption. So if PoS is even slightly decentralized, it'll be better
Great summary. I also run a channel that specialises in blockchain technology. For me Proof of Work is a subsidiary of Proof of Stake. Both require you to stake something; in the POW it is electricity and subsequently FIAT money. In POS, it is the cryptocurrency. Regarding the 51% attack, miner would in reality need somewhere around 65%, 51% would make the attack "possible" but still highly unlikely. Plus, once the network sees what you are doing, they will kick you out and reduce the damages that you have made. In reality, if you invest enough money to initiate a 51% attack, you soon realise that you could use this to make more money, if you use it to secure the network
The market trend can turn around very quickly. In fact, the indexes often switch from a bear market to a bull market when the news is at its worst and the mood of investors is at its lowest point. I read an article of people that grossed profits up to $150k during this crash, what are the best stocks to buy now or put on a watchlist?
stake = coins someone else owns = need permission to access those coins. it's not permissionless to participate in consensus and no reason to give up control.
His computer does the math in the background, I'm guessing. Since it's only a single hash (I think) it shouldn't require any processing power at all and should be complete in less than a second.
After going trough so many channels trying to explain me how proof of stake "work" and how I should invest exactly as I said in platforms they're probably sponsored by, it's really nice to find an educational video that goes trough enough to let you understand both the idea behind it and the risks. Thank you so much!
5:18 "encourages more people" ofc it does when they can lose their coin at any time. While PoW is expensive at the start it pays of itself shortly. PoS probably won't. It will be like bank deposites with investments of certain % increase. and again here, the ones that have the most in the network will have all the privileges all the time, meaning they will be chosen all the freakin time
Maybe I don't get it but the part at 4:06 seems wrong - I would happily approve a fraudulent transaction if I was colluding with someone who that transaction benefitted and who promised to pay me more than the value of the stake I lost approving a fraudulent transaction
thanks for the video! i'm wondering how does the network know that the validator approves a fraudulent block? since he/she's the only one doing the validation I suppose?
Everyone can have the data needed to create each block and forge the hash. So i guess the simplest way is that each elected first check the previous hash then hash the new one. Each validator is the validator of the previous. I guess the punishment is automatic.
@@philv3941 isn't it blockchain? That means everyone in the node will have the copy of transaction and if they notice some red flags they can raise them?
I've watched this video before, but only just paused at 2:04 to read the initial proof of stake idea by QuantumMechanic. I recommend everyone do this. It's an extremely enlightening post. What a visionary.
Thsats not a good thing if only people who own the coin can mine based on what they already own like who gets the first coins.. POS does not solve how to fairly distribute the initial coins only how to keep it going and thats like the main problem. If people can issue money for 0 energy how can they extract any energy without either breaking physics 2nd law or the fundamental economic arguments break down.
So PoS is basically: whoever has the most assets, makes the most assets. 🤔 So we're basically giving minting power to institutions who like to keep unproductive assets under their property/control, right? So we're basically reconstructing the very same centralised system we were trying to fight off in the first place? 🤔 Doesn't make any sense to me.
I agree it makes no sense POS doesnt even deal with how to distribute the initial coins. The leader of a dictatorship could print themselves all the coins in a POS framework and use POS to run it but the dictator gets all the money and control while the users have to verify the blocks or be dilluted and thats a very common way of using POS its really not secure or decentralized. Who cares about a 51% attack if the creators spawned all the coins for themselves for free thats basically an attack at the outset.
PoS is similar of some speculatory investing funds: It's a risky fund who people earns money by the high percentage of the investment, but you need to have a good amount of money into to generate a decent revenue. Many countries use this scheme to complement their lack of liquidity through normal economy activities, agreeing some aggressive foreign investors to try earn an "easier" money. IMHO, the actual and future versions of the Ethereum should enable both systems, because the PoW will start to be discussed again as a "more viable" currency generation at the time when stakers starts to monopolize the value of the coin.
What's stopping PoS pools from forming? It seems that for solo "validators" it wouldn't be worth the stake due to the very low chances of selection no? Will people not form pools that distribute rewards for a PoS network as well?
Ikr. Also with like ethereum 2.0 when it switches to a PoS consensus mechanism, you have to stake a whopping 32 ether! Who tf has that much ether? I certainly don‘t. So for small people like me, I would HAVE to join a staking pool to even stake my ether. It‘s not alot better than joining a mining pool (with my gaming rig with one single gpu) and getting my 3-5€ worth of ether every day... Also I would get alot less ether with proof of stake.
Exactly. Are you familiar with Steemit? Because that seems to be precisely what happened there... If people want to see how a POS system turns out, look at Steemit. Then look at Bitcoin. Which is superior should be apparent, imo. (Don't get me wrong--I have been a member of steemit for almost 9 months now, and use it daily. But I have seen how it has, imo, *worsened* over time, as opposed to *improve*.
I don't know how this problem can be solved generally, but I'll give an example. Orchid, which allows for decentralized VPN services, uses POS to select which node on the network will route a user's internet hops. Of course, a larger stake makes them more likely to be selected for a certain block. However, if the stake is so large that they are being awarded more blocks than they can possibly handle (being asked to provide more bandwidth to the network than they are able), then the extra staked amount is wasted. This gives an incentive not to stake more than what is needed. And also, it's important to note that pools DO exist very explicitly on most POS networks, as they allow you to "delegate" your staked coin to someone who has met the minimum amount to stake (these minimums are usually in the range of thousands of USD). I think a better permanent solution would be to introduce an artificial dis-economies of scale into the network, by having the chance of being selected for a block vary sub-linearly to the amount of stake on the network. For example, Validator_Chance = (Stake)^(.99) or something like this.
@@steviesevieria1868 Its understandable that the video creator would misspeak, however its also understandable why this would really confuse someone learning about this for the first time. He basically said the opposite of what he meant lol
Less. If you were to conduct every transaction on Wall Street with a PoW system it would use a phenomenal amount of electricity. With a normal transaction, you just pay the electricity to send a simple message to the bank. With a PoW transaction, hundreds of miners around the world compete to solve a puzzle to validate just one tiny transaction, consuming enormous amounts of electricity
@Eihoofd2 I don't think replying to a 2 year old comment will get you very far. And if you'd actually watched the video you'd see that a very concise explanation of Bitcoin was given. Most likely what he was referring to.
Exactly. You get 51% of the worlds currency, hello world dictator. You can now create or destroy anybody’s currency. You could create a fraudulent transaction from anyone in the world and just deposit it straight into your account. I think this risk is hugely understated.
Hi Savjee, thanks for the video, I was wondering if the following was true: In order to check if the transaction is correctly signed, you check the crypted private key. In order to check if the money has been double spent, you check the public key and see if the account has enough tokens on it. Thanks in advance for your reply!
Hmm no that's not true. To verify a transaction we check if it was signed with the private key that belongs to the public key (without knowing the private key). And to prevent double spending, we check the entire history of the wallet to see if it still has enough funds to make the payment.
The video was so clear and I really appreciated it... the only thing is not really clear to me is what really is a malicious transaction? What It effectively does to damage the network?
With proof-of-work, the energy you staked is lost forever, while the money you staked in proof-of-stake will be returned to you. Am I right? This makes proof-of-work more secure.
In PoS, even if you somehow manage to gather 51% of the total coins, as mentioned to be 79B $ worth of bitcoins as explained in the video, you are only ensuring that you get to be the validator for the next blocks. The block you build and propagate still has to be validated by all the other nodes in the network. Possessing 51% of total coins doesn't mean you also possess 51% of the nodes in the network, which is what you need to be able to successfully manipulate the blocks to your advantage
@@AwfulnewsFM everipedia took me my sum after an accurate depiction of one page and a non-accurate one for another. I provided images which were removed too. That was censorious, to begin with. The question: Have the amounts I initially had burned, or are they temporary? Also, the cash showed as zero after a great initial boost by a tipping node. As an aside, everipedia's efforts are laudable, but it's pretty flawed, even beginning from the four-pin code for the account, last I entered.
so who are these "Validators" that get selected and how can someone become one is it just by owning some crypto and just holding it? Edit: would coinbase count as a "validator" for me if i were in invest in 32 ETH and all i need to do is just hold?? i need to see more info on how this works
No. You have to STAKE the crypto within the blockchain. Honestly tho, proof of stake might be incredibly decentralized as first but if the whales were to stake all their coins they would literally control everything
The reason you cant spawn money with 0 energy and expect to get energy out is basic physics its so dangerous to say its secure when theres no mechanism whatsoever to distribute the first initial coins fairly so most POS coims are literally spawned from nothing for free to the benefot of the creator group very much not unlike airmiles
In the proof of stake. If the validator accept the fake transaction their security deposit will lost. that's the punishment. So no one will accept the fake transaction.
I am happy that you spread knowledge about this energy-intensive wasteful process. Unfortunately, the whole Western world does not seem to care about this environmentally damaging process while voting for green parties and fighting climate change. I am so happy I have discovered the alternatives exist. But they seem to be still marginal without much impact on the mainstream. I wish more people would act responsibly in accordance to their green/leftist/liberal politic preferences. And more media would highlight the problems, promotingenvironmentally friendlier options. How come banks, investment funds etc. are audited and vetted according their green performance but all this stops when the cryptos seem to be an antouchable topic. And the best alternative would be to connect scientific computations like Boinc or World Community Grid to cryptos. But nobody seems to care in the mainstream..
Yes well, then it's their coin basically and they can do as they see fit. Shareholders in a company that jointly have 51% can also make decisions over wether or not to do a stupid acquisition or to print more shares and doing an offering. It's very unlikely they will decide to do something obviously bad for the company because... they own a 51% stake in it!
secretly coordinate their stakes or else 49% loses faith and sell out and price drops and transaction fees go down unless coin is goverment issued/market accepted and is required for your daily survival/transactions and even if 51% coordinates to defraud you, you don't have alternatives .
fatjohn1408 unless it's a foreign government attack, it would just be written off as an expense of war. Governments spend billions on war, they could also spend billions taking down a country's currency network.
@@fatjohn1408 Right? They act like a 51% attack will be done by some sort of comic book villain. As if there was no incentive to validate & mint w/ 51% of all stake to obtain 51% of all rewards. Lol
Did anyone else find this video biased towards proof-of-stake / Eth? Some examples: Saying that bitcoin is more centralised through mining pools, meanwhile most people that stake Eth do so through pools like Aave? Bitcoin transactions are confirmed by nodes, not by miners. To stake ETH you need to have been pre-selected as a validator - doesn't this sound a bit centralised and not egalitarian? You need 32 ETH to be a validator if not going through a pool, rewarding only those who can afford to front that much capital with even more ETH. The fact that ETH literally has a CEO and foundation team / company that dictates and changes the rules and protocol regularly, by definition makes it centralised. Anyone else feel the same?
Technically, to keep the system secure, we require that the honest miners/validators are a majority over any malicious campaign (even when this majority is below 50%). Consider 3 pools of miners/validators: pool #1 represents honest nodes that make up 40% of the consensus resources, pools #2 and #3 represents dishonest nodes that each make up 30% of the consensus resources. Assuming that pools #2 and #3 are not colluding (due to different malicious goals, for example) and each of these pools generates a malicious chain of blocks, the chain generated by the honest pool (i.e., pool #1) has a majority of the consensus resources and will therefore create the longer blockchain that will be accepted by the network users.
It explain nothing Miners are there to make it impossible to change a previous block and make a valide parrallele modified blockchain. I don't get how PoS replace that mecanisme
A big problem is that, in PoW blockchains like Bitcoin, a lot of miners pool together which is potentially dangerous as in the future, if enough miners pooled together, they could do what is called a 51% attack since they have the majority of hashing power in the network (think of it as a vote). Also, miners themselves don't make it impossible to change a previous block. Blockchains are intrinsically immutable. I hope this helps.
@@turbo6266 it's immuable because of the mining. If someone want to modifie a block, he have to calculate every hash or every block after the modified one, and the faster than all the other miners together, in order to submit a conforme modified blockchain. (Here is why there is this 51%). In other words, that's the mining who made the blockchain immuable. But I don't see that in PoS. I don't get how they make it immuable in pos
@@Merwip Well i guess it is technically is immutable because of mining, but that is because the transaction enters as an input, after which hashing occurs, resulting in an encrypted output as a checksum. iMMUTABILITY is the intrinsic nature that makes a blockchain what it is - safe. However, you said you could calculate every block back, which is impossible because the hash is so sensitive that if you change it just a bit, the encrypted outcome is totally different. It's almost impossible to calculate blocks back. Proof of stake is also immutable because the validators are selected randomly, and thus the node forges the input. Think of forging as a different term for mining, but for proof of stake instead of PoW. To conclude, proof of work makes nodes compete, proof of stake selects validators to solve, and try not to think about mining as the reason why bloickchains are immutable. They are immutable because the encryptive algorithm just makes it pretty hard to solve the encryptions. If some stuff, doesn't make sense, soz, i've only been looking at blockchain for like a week :P
Thank you, very clearly explained. I am rapidly coming to believe that we need far less wasteful means of achieving consensus networks, if DAPPs are to scale to where we'd all like them to be. Proof of Work feels wrong- it's energy inefficient by design. Is there a lucid argument in favour of continuing to produce and operate an increasing number of specialized devices running 24/7 in a mad, concurrent footrace for profit? This seems to be the opposite of sustainable development. I am excited about blockchain, not so much about (even more) rampant energy consumption in the name of greed (dressed as some kind of new utopian democracy). Would love to hear some encouraging words from somebody way smarter than I am about why crypto won't just accelerate the environmental problems we've already been perpetrating upon ourselves... Proof of Stake seems like a logical alternative. Are there other ideas aside from these two currently being explored?
The issue with POS is it doesnt at all solve how to distribute the first initial coins for fair work without a central actor just printing them for themselves thats why every POS coin seems to start with a presale , premine, Dev Tax ect where some group prints themselves free coins for nothing . The solution to the problem of POW and ASICs is the CPU mineable coins which assure much better distribution as anyone could mine with their home computer but also less massive mining farms and more mining on idle or recycled hardware. Its easy to say it would be nice to spawn currency without using energy but if that worked you could just print endless POS coins and have a free energy machine not so eith POW because the adjusting cost of energy and if we mine with useful computers we benefit from the network ifrastructure ( better computers for everyone ) One such coin is Monero
2:40. Translation: POS is far more centralized, and corruptible, than POW, as the more stake you have in the network, the more "validation," and control, you can take in the network. It removes the competitive aspect of POW, and replaces it with "Payola."
The 51% attack for Proof of Stake was addressed incorrectly. You don’t need 51% of ALL currency you need 51% of all stakes. Actually, you might even need less if you game the selection process correctly. A successful attack on proof of stake can have malicious nodes validate their other malicious nodes, and if done long enough can sneak in malicious transactions before getting caught by which time it’s too late.
He said proof of work makes the network more centralized because of mining pools. What do you think about this ?
ADA and WAVES both have over 75% of circulating supply staking and earning people money LOLS good luck buying 51% of staking supply lols
@@BeenDeliveredToday I think if every DOGE holder had a chance to find a 10k DOGE block there would be tons of individual nodes all over the world... I think it would be 100x more decentralized than mining
you cannot sneak in malicious transaction - all nodes validate forged blocks 🙄
@@gimrexymcity4833 you can if you have 51% of the stake like the comment said it’s like 51% attack on PoW you just need 51% of the Mining power but in reality 51% aren’t worth it you ruin credibility of a coin and it becomes less worth it’s more worth using all that power in validating blocks
PoS centralizes more over time than PoW, because larger stakes get more rewards, therefore their size increases. PoW has its own downsides but it is more consensus proof.
You introduce a diminishing returns parameter that, after a certain threshold, your rewards in the stake pool diminish. Encouraging new people and even the stake pool itself to either set a new one up or find a new one thus incentivising decentralisation whilst disincentivizing centralisation. This is not a mark against PoS itself, only networks that don't have a work around.
Honestly? that's a good thing
PoW gave the opportunity to the community to regulate their own currency and the only thing it caused was component hoarding, a chip shortage and irreversible damage to the environment
Decentralized currencies are a nice idea but not at the expense of everyone else trying to live their lives
The biggest downside of POW is literally destroying the environment and people still want it for 'decentralize' :/
@@N1rOx That doesn't make any sense. They can just set up a new node and continue...
@@HardcoreDuce Introducing new pools helps decentralisation. Remember, you still need delegators (unless you are insanely rich, however, if the network is large enough, Bitcoin for PoW or Cardano for PoS, then the cost to perform an attack would be more than any benefit you'd gain), if the network and community see a whale gaining too much stake they can choose not to delegate to them in their current and/or next pool's. This already happens in essence over in POW systems, where let's say for example Ergo has a disclaimer on their Reddit asking users to switch to mining pools away from the number one for the sake of decentralisation. However, as far as I am aware, you don't get financially incentivised to switch like you would with the work around I mentioned for pos.
Major misconception here, proof of stake with a select few validators isn't more decentralized than a proof of work consensus algorithm. This is the reason bitcoin remains the most decentralized blockchain in the world still in 2021.
I believe Cardano is more decentralized than Bitcoin now.
But aren't 70% of miners in china?
Uh no, It's because bitcoin started back early before the arise of Dapps. New Dapps now have more explosive popularity and growth in their infancy which proves my more potential than Bitcoin. But none the less both of them are decentralized.
Bitcoin is one of the least decentralized currencies out there.
@@AwfulnewsFM I'm sorry but either you haven't done your research or you don't understand the metric for calculating what decentralisation is.
I know how difficult it is to turn complex into simple. And I really appreciate the effort you put on and sharing it with us. Kudos to you!😀
The best and most simple explanation of the difference between Proof of work vs. Proof of stake that I have seen. Really great job Savjee, thank you so much!! :)
As always.
O
agreed!
I thought he said Xavier, but agree, good job. I’ll edit this with questions in short order...
@@papaulogamerofc946
1. You are in a pool for Proof of stake. The pool consists of several other miners and as a whole you guys create the coin.
2. There is only 1 blockchain.
3. It's a mathematical equation that uses computing power. Whoever has higher computing power will have higher odds of winning - there is no definitive answer, only odds.
4. not sure if this is a question
I've been watching a bunch of videos explaining proof-of-stake, and this is the only one that actually contextualizes both PoS and PoW within how they interact with the blockchain.
Thank you for putting in the time and effort to create these videos. I’m basically a caveman when it comes to computer science and tech literacy and you’ve really helped clarify a lot of abstract and complicated topics on your channel.
Liked and subscribed brother. Keep up the good work!
Proof of stake in Cardano is amazing: anyone gains: you can delegate your money to small pools which are incentivized over bigger ones.
Every time a block is created the rewards are split across all participants, simply beautiful.
Thanks for the very concise explanation!
The centralization of mining pools isn't actual centralization because individual miners are free to leave any time. This is not a centralization threat. The power issue can't be viewed in a vacuum and one must consider the power that us not used because the transactions are happening elsewhere. There are places for both PoW and PoS. It's not as simple as this video makes it out to be.
the power is in that if they work together they control over 51% of the system and falsify blocks
Country Corner but miners would immediately disconnect from the pool. Miners get paid in coins/tokens. Do you think they are going to expend energy to mine something that would immediately become worthless upon a 51% attack? The only way a pool could pull this off is if they owned every miner that was connected and even then they would only be able to maybe falsify one block. The amount of work to falsify 2 or 3 especially in a network like bitcoin is impossible even if pools coordinated an attack.
superb point people do not get it!
Hello Adam Shepherd: Interesting point. Could you please explain your idea in more detail or do you have some literature to support the same?
Which proof is better and why?
Proof of stake will:
1. Encourage hoarding (and hinder spending).
2. Monopolistic behavior.
Not sure I trust the masses anymore than I trust govt to limit monopolies and greed, which means, I see no answers to Bitcoin's "problems."
And what is going to stop people from pooling their bitcoin into a single account to recreate the same problems as mining pools?
NOTHING
There are no answers.
Great explanation, but nowadays, some coins allows staking pool. I agree that it may reduce electricity cost but doesn't necessarily solve the centralization problem from poolings.
Or the more philosophical issue of a handful of rich entities controlling the entire process in practice.
As an ETH miner I'm stoked about it. This could bring in a lot of diversity into crypto and subsequent liquidity. PoW will have a place because gigahashes of computational power doesn't just disappear. The PoW miners will help close the gap that currently exists between ETH and other cryptos. Fee payouts then can be moved into stakes, offering something other than price action volatility, which exacerbates that problem, for further investment. When ETH goes to PoS we can hedge on the liquidity of that network.
2:21 "One node is randomly chosen as a validator"
2:37 "Validators aren't chosen completely randomly"
Hehe, it's still random, just not completely random. Your odds are determined by your stake. But equal stake means equal chances, means random selection.
@@davidz2690 if your forced to add "not with uniform..." so it's not "completely random" as general public assumes it, with seems "fully random with uniform odds". Like in a game of dices, with strictly the same weight for each dice and for each face of each dice. A "fair" random.
SimplyExplain is just better than you to explain clearly without acronyms or technical concepts ;)
The problem of Pools will not be fixed by a transition to PoS. You will need 32 ETH to become a validator. Which is $78000 (23.01.2022) right now. How many people have this amount of ETH in one wallet? Probably way less people than people owning a sub $78000 mining rig. Maybe in PoS having a lot of money is less beneficial than having a lot of money in PoW, however a lot of people will be excluded, if they don't join a pool.
I have a question. Cant validators come together and form 'validator' pools just like the mining pools we know about? Where fees are paid according to the security deposit that they have within
I think some currencies have staking pools.. how each currency addresses the problems at 6:20 are unique
They can but there's less benefit from doing to then in proof of work.
Yes it takes a lot less electricity but it isn’t more decentralized. There’s staking pools just like mining pools. And I believe its more centralized cause it’s easier to buy a lot of crypto then getting a lot of miners. Plebs have max 1-5 eth not 32...
So true. How am I supposed to stake 32 ether? I barely have 0.1 ether
For the first time, I feel like I understand PoS. Every other explanation I got for it never actually mentioned that validators were actually doing some calculation work when they were chosen. Thank you!
Thanks for this explanation!
Perhaps I missed it, but how does the network actually randomly select a validator? They all have to acheive consensus on who the validator is, right?
It seems that various currencies are using various approaches.
One way is to multiply the stake of a validator by the time since its last validation.
For example, if:
V1 has a stake of 100 for 2 days = 200
V2 has a stake of 10 for 30 days = 300
Then everybody agree that V2 has to be the next validator.
After the forging, V2 has a stake of 10 for 0 days.
Another more "random" way might be to multiply the stake of a validator by an hash of (validator id + last block).
This way a random node is selected, but everybody choose randomly the same validator.
@@rt1517 Does hashing still mean PoW in the PoS model? What's the censorship potential by a majority?
@@harishkumargujudhur2234 No hashing does not mean PoW in PoS model. In PoW, you have to do tons of hashes on the same block again and again only modifying its nonce until the hash result matches a specific contract. It requires a lot of tries to find a valid nonce and these tries take time and energy.
In case of PoS as explained above, you would only have to hash once per validator to determine the next validator. Easy task, even if there are many validators. No competition, no need for gigantic mining farms.
To do a 51% attack on bitcoin you need half of the hashing computing power (that means only the 4 biggest bitcoin mining pools). To do a 51% on a PoS currency, you typically need 51% of the coins. Some people think that if you own 51% of a currency, you won't attack it because the currency will lose its value and you will lose all your investments. So PoS should be a lot less targeted by 51% attacks.
@@rt1517 Thank you very much.
I'm currently leaning on PoW. So, feasibility-wise you mean PoS is less secure, but in terms of the desire to take the 51% attack potential, PoW has a higher competition. (excluding less popular PoW coins not targeted as Store of Value) Based on your explanation concerning the need for gigantic mining farms so far, it has a higher energy demand for attacks than PoS.
@@rt1517 Also, I would be really grateful if you would point out what other consensus algorithms use more than one hashing to check for a valid nonce. I'm much interested in non-identificational Proof-of-Weight. I don't know much about use of hashing other than in the form of basic OTP passwords.
I bet the initial comments to QuantumMechanic were "That's a terrible idea and would never work.", as usual.
PoS is less fair, because staking is a one-time investment. With PoW OTOH miners continuously have to pay their electricity bills. Banks are Dollar stakers who decide which transactions go through. Besides you are contradicting yourself: First you rightly say, that it is much easier to stake than it is to mine and later you wrongly claim, that it‘s harder to do a 51% attack with PoS. It clearly is harder to purchase mining equipment and set it up than to just buy enough stake in order to do double spend attacks.
Thank you... This video is way off.
Proof of stake is more akin to our current Fiat system: centralized and favoring the rich
This video advertise etherium secretly. It is a total misinformation. Most of this new coin are PoS. I think elites are behind this to stop Bitcoin. I think they also manipulate the GPU prices so that they halt miners from mining while advocating for etherium which is secretly centralized.
@@diocre7446 indeed.
@@diocre7446 Ethereum is corporation issued currency.
i am doing a research project on nfts, and this is by far the best video i've seen explaining PoS. Same goes for your other videos on blockchain and other stuff.
#1 channel for actually understanding all this blockchain crap
10/10
I get the power argument here, but the whole point of crypto was to be decentralized, with proof of stake it is now pay to play and might as well be another fed.
Unfortunately, crypto has yet to be successfully decentralized. Bitcoin failed to achieve the goal, even if you ignore power consumption. So if PoS is even slightly decentralized, it'll be better
Great summary. I also run a channel that specialises in blockchain technology.
For me Proof of Work is a subsidiary of Proof of Stake.
Both require you to stake something; in the POW it is electricity and subsequently FIAT money.
In POS, it is the cryptocurrency.
Regarding the 51% attack, miner would in reality need somewhere around 65%, 51% would make the attack "possible" but still highly unlikely. Plus, once the network sees what you are doing, they will kick you out and reduce the damages that you have made. In reality, if you invest enough money to initiate a 51% attack, you soon realise that you could use this to make more money, if you use it to secure the network
This is exactly the logic of the US government. We've gone full circle here.
The market trend can turn around very quickly. In fact, the indexes often switch from a bear market to a bull market when the news is at its worst and the mood of investors is at its lowest point. I read an article of people that grossed profits up to $150k during this crash, what are the best stocks to buy now or put on a watchlist?
Buddy, you are a real genius and natural born teacher. Thank you for all your movies!!
Your explanations are to the point! Thanks a lot, Sajvee, I appreciate you! Keep up the good work 🙏
ḭnvest in Ƀḭt¢oin trading with my administrative Valenzia Andrew...Add up on...T E L E G R A M...+ 1 3 2 5 7 3 2 0 1 9 1.
stake = coins someone else owns = need permission to access those coins. it's not permissionless to participate in consensus and no reason to give up control.
How does the chosen validator validates the transactions?
His computer does the math in the background, I'm guessing. Since it's only a single hash (I think) it shouldn't require any processing power at all and should be complete in less than a second.
@@TwistZero wtf! that makes no sense
@@fineartpottamus9020 how so?
By far the best straightforward explanation 👍🏻👍🏻👏👏
Awesome explanation, clear and to the point. Came here after learning about Ethereum 2.
Same
After going trough so many channels trying to explain me how proof of stake "work" and how I should invest exactly as I said in platforms they're probably sponsored by, it's really nice to find an educational video that goes trough enough to let you understand both the idea behind it and the risks.
Thank you so much!
Crystal clarity: you gotta love it when you watch it! Thank you!
Little problem is that Proof of Stake is completely subjective and theirfore invalid.
Agreed. It is very easy to understand
Watching this video when it was first launched gives me goose bumbs... because i can understand better now compared to 4 years back
5:18 "encourages more people" ofc it does when they can lose their coin at any time. While PoW is expensive at the start it pays of itself shortly. PoS probably won't. It will be like bank deposites with investments of certain % increase. and again here, the ones that have the most in the network will have all the privileges all the time, meaning they will be chosen all the freakin time
exactly
you know guys crypto isn't only about the mining ..
Maybe I don't get it but the part at 4:06 seems wrong - I would happily approve a fraudulent transaction if I was colluding with someone who that transaction benefitted and who promised to pay me more than the value of the stake I lost approving a fraudulent transaction
thanks for the video! i'm wondering how does the network know that the validator approves a fraudulent block? since he/she's the only one doing the validation I suppose?
Everyone can have the data needed to create each block and forge the hash.
So i guess the simplest way is that each elected first check the previous hash then hash the new one.
Each validator is the validator of the previous. I guess the punishment is automatic.
@@philv3941 isn't it blockchain? That means everyone in the node will have the copy of transaction and if they notice some red flags they can raise them?
I've watched this video before, but only just paused at 2:04 to read the initial proof of stake idea by QuantumMechanic.
I recommend everyone do this. It's an extremely enlightening post. What a visionary.
your videos are so well structured, keep them coming
holy shit. Thank you! finally someone who explained why PoS is less power.. put simply there is a restriction to who can actually do the mining.
Thsats not a good thing if only people who own the coin can mine based on what they already own like who gets the first coins.. POS does not solve how to fairly distribute the initial coins only how to keep it going and thats like the main problem. If people can issue money for 0 energy how can they extract any energy without either breaking physics 2nd law or the fundamental economic arguments break down.
@@bitcoinizakaya9579 wtf do you think you can extract energy from bitcoin or something?
@@AwfulnewsFM One can sell their crypto to heat and power a house which is indeed converting money into energy
So PoS is basically: whoever has the most assets, makes the most assets. 🤔
So we're basically giving minting power to institutions who like to keep unproductive assets under their property/control, right?
So we're basically reconstructing the very same centralised system we were trying to fight off in the first place? 🤔
Doesn't make any sense to me.
could not agree more
I agree it makes no sense POS doesnt even deal with how to distribute the initial coins. The leader of a dictatorship could print themselves all the coins in a POS framework and use POS to run it but the dictator gets all the money and control while the users have to verify the blocks or be dilluted and thats a very common way of using POS its really not secure or decentralized. Who cares about a 51% attack if the creators spawned all the coins for themselves for free thats basically an attack at the outset.
@@bitcoinizakaya9579 exactly, and a 51% attack becomes harder and more costly when there are more people mining right?
What do you think of PoS in comparison to PoW?
PoS is similar of some speculatory investing funds: It's a risky fund who people earns money by the high percentage of the investment, but you need to have a good amount of money into to generate a decent revenue. Many countries use this scheme to complement their lack of liquidity through normal economy activities, agreeing some aggressive foreign investors to try earn an "easier" money. IMHO, the actual and future versions of the Ethereum should enable both systems, because the PoW will start to be discussed again as a "more viable" currency generation at the time when stakers starts to monopolize the value of the coin.
Could you explain the improvements that are implemented by Elrond in their "Secure Proof of Stake" mechanism?
So what is the first coin from proof of stake? There is no one has the coin, who is going to be selected?
What's stopping PoS pools from forming? It seems that for solo "validators" it wouldn't be worth the stake due to the very low chances of selection no?
Will people not form pools that distribute rewards for a PoS network as well?
Ikr. Also with like ethereum 2.0 when it switches to a PoS consensus mechanism, you have to stake a whopping 32 ether! Who tf has that much ether? I certainly don‘t. So for small people like me, I would HAVE to join a staking pool to even stake my ether. It‘s not alot better than joining a mining pool (with my gaming rig with one single gpu) and getting my 3-5€ worth of ether every day... Also I would get alot less ether with proof of stake.
Very good explanation
What's to stop the creation stake pools similar to mining pools?
Exactly. Are you familiar with Steemit? Because that seems to be precisely what happened there... If people want to see how a POS system turns out, look at Steemit. Then look at Bitcoin. Which is superior should be apparent, imo. (Don't get me wrong--I have been a member of steemit for almost 9 months now, and use it daily. But I have seen how it has, imo, *worsened* over time, as opposed to *improve*.
You need a balance of bitcoin's decentralization but the ability for pools, companies, teams to speed up velocity based on real world work.
I don't know how this problem can be solved generally, but I'll give an example. Orchid, which allows for decentralized VPN services, uses POS to select which node on the network will route a user's internet hops. Of course, a larger stake makes them more likely to be selected for a certain block. However, if the stake is so large that they are being awarded more blocks than they can possibly handle (being asked to provide more bandwidth to the network than they are able), then the extra staked amount is wasted. This gives an incentive not to stake more than what is needed. And also, it's important to note that pools DO exist very explicitly on most POS networks, as they allow you to "delegate" your staked coin to someone who has met the minimum amount to stake (these minimums are usually in the range of thousands of USD). I think a better permanent solution would be to introduce an artificial dis-economies of scale into the network, by having the chance of being selected for a block vary sub-linearly to the amount of stake on the network. For example, Validator_Chance = (Stake)^(.99) or something like this.
This video needs a community notes section.
2:16 i thought proof of STAKE (not work) is where one node is randomly chosen to validate the next block?
He misspoke one time, give it a break
@@steviesevieria1868 Its understandable that the video creator would misspeak, however its also understandable why this would really confuse someone learning about this for the first time. He basically said the opposite of what he meant lol
At 2'17 he says Proof of Work instead of Proof of Stake in case anyone is confused.
I think after that he spoke POW instead of POS in most of the places
How much energy do all the bank buildings and branches use ? Not to mention the huge power used on Wall Street for example.
Less. If you were to conduct every transaction on Wall Street with a PoW system it would use a phenomenal amount of electricity. With a normal transaction, you just pay the electricity to send a simple message to the bank. With a PoW transaction, hundreds of miners around the world compete to solve a puzzle to validate just one tiny transaction, consuming enormous amounts of electricity
The best and most clear video about pos. Thank you very much!
One of the best explanations ever on bitcoin..thank you so much.
@Eihoofd2 I don't think replying to a 2 year old comment will get you very far. And if you'd actually watched the video you'd see that a very concise explanation of Bitcoin was given. Most likely what he was referring to.
It very poorly explains Bitcoin, nobody that uses Bitcoin wants proof of stake.
Anyone interested read the whitepaper by A. Poelstra Published 2015: "Distributed Consensus from Proof of Stake is Impossible".
6:08 But that's _nothing,_ if the reward is control over, say, world's reserve currency.
Exactly. You get 51% of the worlds currency, hello world dictator. You can now create or destroy anybody’s currency. You could create a fraudulent transaction from anyone in the world and just deposit it straight into your account. I think this risk is hugely understated.
People as in real dictators are actually trying this slowly but surely
You definitely earned my subscription. You have very good technical knowledge.
How do you know if you are validating a fraudulent transaction? How can miners/validators avoid that kind of scenario? Cheers. Great video.
They check if a transaction was correctly signed (with the matching key) and if money hasn't already been spent (double spending)
And you can't safeguard yourself against fraud I.e money laundering. Even harder with the new masternode coins with built in washing facility.
Hi Savjee, thanks for the video, I was wondering if the following was true: In order to check if the transaction is correctly signed, you check the crypted private key. In order to check if the money has been double spent, you check the public key and see if the account has enough tokens on it. Thanks in advance for your reply!
Hmm no that's not true. To verify a transaction we check if it was signed with the private key that belongs to the public key (without knowing the private key). And to prevent double spending, we check the entire history of the wallet to see if it still has enough funds to make the payment.
No, it's completely automated
Well and clearly explained in a short period of time with good visuals, thanks!!
In POW every new block will be having hash of previous, what about POS? how it ensures security?
The only difference its its permissioned based on stake. Which is inferior to open to all equally.
The video was so clear and I really appreciated it... the only thing is not really clear to me is what really is a malicious transaction? What It effectively does to damage the network?
With proof-of-work, the energy you staked is lost forever, while the money you staked in proof-of-stake will be returned to you. Am I right? This makes proof-of-work more secure.
In PoS, even if you somehow manage to gather 51% of the total coins, as mentioned to be 79B $ worth of bitcoins as explained in the video, you are only ensuring that you get to be the validator for the next blocks. The block you build and propagate still has to be validated by all the other nodes in the network. Possessing 51% of total coins doesn't mean you also possess 51% of the nodes in the network, which is what you need to be able to successfully manipulate the blocks to your advantage
Except that the stake in POS is temporary while the cost of mining equipment is permanent making that system more secure and indelible.
The PoW one?
Stake lost in PoS is burned and is lost forever.
@@AwfulnewsFM everipedia took me my sum after an accurate depiction of one page and a non-accurate one for another. I provided images which were removed too. That was censorious, to begin with.
The question: Have the amounts I initially had burned, or are they temporary? Also, the cash showed as zero after a great initial boost by a tipping node.
As an aside, everipedia's efforts are laudable, but it's pretty flawed, even beginning from the four-pin code for the account, last I entered.
Staking pools exist, much like mining pools. Meaning the centralization could be equal to proof of work
I really enjoy it, well done man!
While you've done the opposite of _sell_ me on the idea of Proof of Stake, you did explain it simply as your channel name promised, so thank you.
so who are these "Validators" that get selected and how can someone become one is it just by owning some crypto and just holding it?
Edit: would coinbase count as a "validator" for me if i were in invest in 32 ETH and all i need to do is just hold??
i need to see more info on how this works
No. You have to STAKE the crypto within the blockchain. Honestly tho, proof of stake might be incredibly decentralized as first but if the whales were to stake all their coins they would literally control everything
"Erhereum is actively developping PoS" 3 years laters : not yet
Can a super rich government just purchase 51% of the market cap? 100b for the US Fed is a pocket change
@Rene Media Ahhhh... I see
Thank you. If POS is really as secure then it's definitely the way to go vs POW consuming massive amounts of energy and ever increasing mining costs.
The reason you cant spawn money with 0 energy and expect to get energy out is basic physics its so dangerous to say its secure when theres no mechanism whatsoever to distribute the first initial coins fairly so most POS coims are literally spawned from nothing for free to the benefot of the creator group very much not unlike airmiles
After ETH has switched to PoS, only Bitcoin and Litecoin will remain as the classics. Litecoin is the most efficient of the two
Proof of stake trends towards centralisation.
Proof of work trends towards decentralisation.
Great video. One question, how does the network actually detect fraudulent validators and punish them? What's the mechanism there? Thanks.
In the proof of stake. If the validator accept the fake transaction their security deposit will lost. that's the punishment. So no one will accept the fake transaction.
@@mixwood1130 unless they have large shorts in place on a futures market
@@mixwood1130 the original question was asking how the network detects a fraudulent transaction or double spend
@@bitcoinizakaya9579 Have a great day!
@@SidelongSuggestive have a nice day!
how can letting someone buy their way into validating a block be less centralizing than the current method ?
Proof-Of-Use removes 51% attack.
What's that?
@@uchihai_a_h4871 A protocol i invented but i was unfortunately shutdown by larger parties.
I am happy that you spread knowledge about this energy-intensive wasteful process. Unfortunately, the whole Western world does not seem to care about this environmentally damaging process while voting for green parties and fighting climate change. I am so happy I have discovered the alternatives exist. But they seem to be still marginal without much impact on the mainstream. I wish more people would act responsibly in accordance to their green/leftist/liberal politic preferences. And more media would highlight the problems, promotingenvironmentally friendlier options. How come banks, investment funds etc. are audited and vetted according their green performance but all this stops when the cryptos seem to be an antouchable topic. And the best alternative would be to connect scientific computations like Boinc or World Community Grid to cryptos. But nobody seems to care in the mainstream..
Imagine when an explanation from 3 years ago explains something happening this year better than explanations from this year.
4:19 it’ll be released for the persons own use. So it’ll still be the persons. Just to clarify
I think the important thing to consider is that without mining the GPUs would still be using power by playing games.
This is horribly wrong haha
Wow! Defi/Blockchain explained in a way I understood!! Good work man! Gonna check out your other vids
Proof of Stake: stakeholders can pool their stakes to achieve 51%.
Yes well, then it's their coin basically and they can do as they see fit. Shareholders in a company that jointly have 51% can also make decisions over wether or not to do a stupid acquisition or to print more shares and doing an offering. It's very unlikely they will decide to do something obviously bad for the company because... they own a 51% stake in it!
secretly coordinate their stakes or else 49% loses faith and sell out and price drops and transaction fees go down unless coin is goverment issued/market accepted and is required for your daily survival/transactions and even if 51% coordinates to defraud you, you don't have alternatives .
The velocity of the coin will drop if a small group can only validate mine it.
fatjohn1408 unless it's a foreign government attack, it would just be written off as an expense of war. Governments spend billions on war, they could also spend billions taking down a country's currency network.
@@fatjohn1408 Right? They act like a 51% attack will be done by some sort of comic book villain. As if there was no incentive to validate & mint w/ 51% of all stake to obtain 51% of all rewards. Lol
My biggest concern with PoS is the nodes getting concentrated on big cloud hosts like AWS because of the penalty associated with downtime.
Very good point.
As a Hungarian I wish that miners would power Hungary :( Those mining rigs would be very power-Hungary :D
Lol
Oh.. Dear... Lord...
Nice
Simple no nonsense explanation! Thank you!
That was so brilliant. You are so smart. I am so happy you shared this.
Loved this. Shared with my mother, an investor.
i feel like i have clicked on this notification before...
Sorry about that! The original video has a mistake which I wanted to fix ;)
Did anyone else find this video biased towards proof-of-stake / Eth?
Some examples:
Saying that bitcoin is more centralised through mining pools, meanwhile most people that stake Eth do so through pools like Aave?
Bitcoin transactions are confirmed by nodes, not by miners.
To stake ETH you need to have been pre-selected as a validator - doesn't this sound a bit centralised and not egalitarian?
You need 32 ETH to be a validator if not going through a pool, rewarding only those who can afford to front that much capital with even more ETH.
The fact that ETH literally has a CEO and foundation team / company that dictates and changes the rules and protocol regularly, by definition makes it centralised.
Anyone else feel the same?
Technically, to keep the system secure, we require that the honest miners/validators are a majority over any malicious campaign (even when this majority is below 50%).
Consider 3 pools of miners/validators: pool #1 represents honest nodes that make up 40% of the consensus resources, pools #2 and #3 represents dishonest nodes that each make up 30% of the consensus resources. Assuming that pools #2 and #3 are not colluding (due to different malicious goals, for example) and each of these pools generates a malicious chain of blocks, the chain generated by the honest pool (i.e., pool #1) has a majority of the consensus resources and will therefore create the longer blockchain that will be accepted by the network users.
I do not understand that part about staking being more decentralized. Wouldn't there just be staking pools instead of mining pools?
It explain nothing
Miners are there to make it impossible to change a previous block and make a valide parrallele modified blockchain.
I don't get how PoS replace that mecanisme
A big problem is that, in PoW blockchains like Bitcoin, a lot of miners pool together which is potentially dangerous as in the future, if enough miners pooled together, they could do what is called a 51% attack since they have the majority of hashing power in the network (think of it as a vote). Also, miners themselves don't make it impossible to change a previous block. Blockchains are intrinsically immutable.
I hope this helps.
@@turbo6266 it's immuable because of the mining. If someone want to modifie a block, he have to calculate every hash or every block after the modified one, and the faster than all the other miners together, in order to submit a conforme modified blockchain. (Here is why there is this 51%). In other words, that's the mining who made the blockchain immuable.
But I don't see that in PoS. I don't get how they make it immuable in pos
@@Merwip Well i guess it is technically is immutable because of mining, but that is because the transaction enters as an input, after which hashing occurs, resulting in an encrypted output as a checksum. iMMUTABILITY is the intrinsic nature that makes a blockchain what it is - safe. However, you said you could calculate every block back, which is impossible because the hash is so sensitive that if you change it just a bit, the encrypted outcome is totally different. It's almost impossible to calculate blocks back.
Proof of stake is also immutable because the validators are selected randomly, and thus the node forges the input. Think of forging as a different term for mining, but for proof of stake instead of PoW.
To conclude, proof of work makes nodes compete, proof of stake selects validators to solve, and try not to think about mining as the reason why bloickchains are immutable. They are immutable because the encryptive algorithm just makes it pretty hard to solve the encryptions.
If some stuff, doesn't make sense, soz, i've only been looking at blockchain for like a week :P
hey Xavier, really good one. as linguists say, CBS, Clear Brief and Sincere!
very good explanation. thank you
00:50 "this puzzle is solved by minors."
me: "0_o well that's oddly specif... oh... right... carry on"
You can still have 'mining pool situations' with proof-of-stake. People can get funding from the crowd. It's like buying company shares
thats why Cardano is the future ( or something similar)
All in ADA
2:28 for proof of stake
Thank you, very clearly explained. I am rapidly coming to believe that we need far less wasteful means of achieving consensus networks, if DAPPs are to scale to where we'd all like them to be. Proof of Work feels wrong- it's energy inefficient by design. Is there a lucid argument in favour of continuing to produce and operate an increasing number of specialized devices running 24/7 in a mad, concurrent footrace for profit? This seems to be the opposite of sustainable development. I am excited about blockchain, not so much about (even more) rampant energy consumption in the name of greed (dressed as some kind of new utopian democracy).
Would love to hear some encouraging words from somebody way smarter than I am about why crypto won't just accelerate the environmental problems we've already been perpetrating upon ourselves... Proof of Stake seems like a logical alternative. Are there other ideas aside from these two currently being explored?
The issue with POS is it doesnt at all solve how to distribute the first initial coins for fair work without a central actor just printing them for themselves thats why every POS coin seems to start with a presale , premine, Dev Tax ect where some group prints themselves free coins for nothing . The solution to the problem of POW and ASICs is the CPU mineable coins which assure much better distribution as anyone could mine with their home computer but also less massive mining farms and more mining on idle or recycled hardware. Its easy to say it would be nice to spawn currency without using energy but if that worked you could just print endless POS coins and have a free energy machine not so eith POW because the adjusting cost of energy and if we mine with useful computers we benefit from the network ifrastructure ( better computers for everyone ) One such coin is Monero