Burst is busted?



  • Few days back I was reading this article at coindesk.com. The creator of Bit-Torrent, @bramcohen was 'proposing' the Proof of Space because Proof of Work was a waste. In the article there wasn't any mention of Burstcoin, but there was a mention of a new white paper of SpaceMint, by one of the guy for the Proof of Space white paper.

    I've read the SpaceMint whitepaper, in there it was said that

    In terms of efficiency, a major issue with Burstcoin is that a constant fraction (0.024%) of dedicated disk space must be read every time a block is mined. In contrast, SpaceMint requires accessing only logarithmically many blocks in the size of dedicated space. For instance, at 1TB, Burstcoin reads 24GB, while SpaceMint reads 24 MB. Finally, verification in Burstcoin is also problematic: a miner has to hash over 8 million blocks to verify another miner’s claim.

    As a complete noob that only started mining a month ago, I'm confused on this technical details. So my question is to the Developer, please explain this to me like I'm five because I'm like mining in the pool, then learned that there's more profitable pool to mine I just switched pool. My question is: Is that 'inefficiency' a real problem?

    aBustt

    I don't know if Burstcoin is a bust or not, since I'm still new. But, from my short experience in trying other alt-coin, Burst is the easiest to obtain and traded with other top 10 cryptos on the exchange. IMHO, someone should enlighten Bram on Burst.


  • admin

    @headsink His numbers are totally wrong. During mining the miner reads 1 of 4096 scoops in each nonce. So, for 1TB, divide by 4096 and you'll find it actually reads 268MB per TB. If it was really reading 24GB/TB then how could I possibly be mining 50TB in less than 1 minute?

    I'm pretty sure the 8 million block comment is totally incorrect, but I'll let someone more familiar with the internal details comment.



  • I think he's going to try and build his own hdd mining coin. Competition is a good thing. I think there will be more coins being developed for hdd mining. We'll see



  • I can answer the question about verification of other miners blocks, again the author's numbers are completely inaccurate. When a miner submits a block for a deadline the single nonce in submitted with the account ID number, the combination of nonce and accountID can generate a complete nonce by performing 4096 SHA calculations and a single XOR on the final result, we then look at the single SCOOP (64 bytes of data) as identified by the block chain to verify the claim. The work realized to validate a single claim is nearly nothing. It only becomes slightly problematic when a pool has a high deadline for submission times and single miner may submit 2 to 5 nonces and there could be potentially hundreds of miners on a single pool, therefore the server has the work load of calculating possibly thousands of nonces for verificationas quickly as possible, if the pool can't verify it, the block will not be submitted to other wallets/pools.

    -IceBurst



  • I wouldn't trust or listen to Bram Cohen at all, he tries to come off like he knows what he is talking about but he doesn't know anything. He used to say bitcoin was a scam and a ponzi now he's pro bitcoin and started bashing on altcoins. I can't find the stupid posts he made before about bitcoin but here is one where he is saying mining is a waste and that people can't solve it back in 2015. Based on that video you can say bitcoin is still producing "waste" with its mining and that other forms of mining doesn't work too because of "waste" and issues like centralization. Guess bitcoin is done for lets move onto proof of stake since there is no mining "waste." I think it's wrong to say proof of work is a waste it's necessary for security and decentralization compared to other options.

    Burstcoin mining is really efficient on electricity/cpu/gpu you won't even notice the miner running. It's basically free money compared to some of the other coins on the market. The market cap for burst been raising steadily and will continue to raise because the Asset exchange is providing some real service that is useful and can gain money from outside sources. I don't think burst's proof of space will make it big but the community, market and the AE's use case will make it big.

    https://www.reddit.com/r/Bitcoin/comments/3d05fh/bram_cohen_removing_the_waste_from/



  • Thanks @socalguy for the video, it answered my doubts of SpaceMint. I agree competition for hdd mining is good, but I can't see that even in the near future....

    "There's no code here, this is a theoretical construct. I'm not comfortable building this yet because the underline primitives are not quite there yet..." -

    I guess the people that have been involved with Burst are more knowledgeable because at least they've done the real physical experiment. Thanks @IceBurst & @haitch for the very detailed technical answer which would take some time for me to understand.

    The Reddit post provided by @hah convince me that I can't take Bram Cohen seriously.


  • admin

    Bram Cohen is describing Burstcoin and its advantages in almost every aspect. In 2017.

    One feature he describes is very interesting and I see no reason why it can't be implemented in Burstcoin Software:

    Writing plots in the free sectors of your hard-drive (e.g. C: Drive) but also allow the Operating System to overwrite plots by still seeing the plots as free sectors afterwards. So the user and the operating system is not stressed having a full disk all the time.

    I had a similar idea over a year ago and called it "dynamic plotting". The Win Client could always leave 10% of the HDD space "in use" free for the user. If the users deletes some big files plots get slowly rewritten in X GBs chunks and if a user needs more space plots get deleted in X GBs chunks. Not as elegant as he describes but a very similar outcome for the end user. @Lexicon @wombat

    I don't get his point at the moment why Burst is busted for him. Looks like it has to be busted because he is talking in front of audiences of the next big thing. For him the solution is "too simple"? Time (timestamp) plays an important role in the Burst Protocol too, as we all know. The numbers in the "Proof of Space" paper about Burst are incorrect. Time to send these guys a mail to read the wikipedia article. ;) (In defense: In the time of the paper being written there was not much technical documentation about Burst available. Just the code.)

    After Vitalik Buterin we have now the second famous person calling this project flawed or busted. Soon we can make a video like this one:



  • @daWallet that guys a joke lol if burst was busted surely someone would be mining all the coins faster than 3 jews with 8 coupons.

    the solution is simple and because its simple it makes it genius. running a hashing function 4096 times is the main brunt of the work. to work this out depending on the language its written in and the amount of processing power thats put onto it makes it incredibly hard to pull any sort of speed out of your processor. however reading the prewritten plots that have already gone through htis routine would be much much faster as you dont have to do the hashing function 4096 times.

    thats why you cant mine on cpu or gpu efficiently.. im not saying it cannot be done.. as i myself have written a burst cpu miner however to can hardly put out the amount of nonce's the plotter does. and could probably do 50gb an hour.

    the stupid thing is though all this information was already out there. just these guys would rather bash it at face value rather than actually look and do any sort of market research


  • admin

    I love the article because it states all the reasons why I am supporting this coin as much as I can:

    "Even ignoring questions of waste, power-intensive mining has inarguably also led to geographic concentrations of the industry. Because miners receive the greatest returns where energy is cheaper, areas like Iceland, western China and Washington State have become the dominant epicenters."

    You can buy HDDs and USB sticks everywhere in every town in the world.

    "The search for a system that would encourage the opposite of these qualities – generic equipment, non-wasteful work and low power consumption."

    Check.

    "Under a proof-of-space system, miners allocate a certain amount of their unused disk space to the network, with the probability of successfully mining a block being proportional to the amount of space allocated divided by the total capacity of the network."

    "To briefly summarize, miners add new transactions into a block by combining them with values from a lookup of data stored in a space they have allocated for mining."

    "One notable consequence of using storage as a proof, is that it could democratize the mining process."

    "Another upshot of proof of space is that there is no incentive for any actors to try and gain mining share by spending large sums on storage space up front."

    This is 100% Burstcoin.

    "This creates an inherent attack where someone with a large amount of resources can re-mine since genesis – go back to the beginning and make a whole new chain, giving themselves all the rewards."
    "In order to prevent this, Cohen's idea is to use proof of time in combination with proof of space, specifically, proofs derived from a dedicated time server on which some time will elapse in between each period of mining."

    Here he disqualifies himself imo or I don't get the point. You can't go back to genesis with a lot of space and mine everything to be up to date to the other chain. Burstcoin has a 4 minute block time and uses every miner as a time server. This is exactly what Burst is doing - so we are also Proof of time. lol.

    @Lexicon Instead him coming here to improve the algo, he wants his own more complex solution with his name tag on it and maybe a nice bag of coins as an early adopter.

    Burstcoin seems to stay the underdog coin for the crypto people for some more time. (Gpu)Miners don't want it, ASIC industry doesn't want it and some famous people in crypto space dismiss it without even scratching the surface.

    Proof of Underdog.



  • lollerskates


Log in to reply
 

Looks like your connection to Burst - Efficient HDD Mining was lost, please wait while we try to reconnect.