5 ms·

> Why not have a central bank currency that can be traded on the blockchain, especially since converting it to real money will still entail KYC? Because litera

> Why not have a central bank currency that can be traded on the blockchain, especially since converting it to real money will still entail KYC? Because literally the only point is to avoid the existing banking system and you can do that with a postures database with much less cpu involved.

algo_lover 1y ago
But with multiple parties involved, who has the rights to read and write to the postgres instance? How do we make sure transactions were not forged? How do we know data at rest is not being tampered with? Blockchain solves that. Newer blockchain protocols especially an L1 is much faster, easier on the environment, and provides all the immutability, transparency, and traceability benefits.
- topranks 1y ago
  We need to trust those running the system. Societies cannot function without trusted intermediaries, in finance and many other things. If we are in a democracy then the government regulates such organisations and should punish those who do not comply. Blockchain doesn’t scale as a replacement so the point is moot.
- oblio 1y ago
  You know you can just use regular cryptography to validate data, right? Also, you always have to trust someone, in this case Stripe. Regarding L1 blockchains, how exactly do they solve the speed problem for a distributed global database that needs to be replicated everywhere for the security guarantees to actually work? What do they forgo out of https://en.m.wikipedia.org/wiki/CAP_theorem https://en.m.wikipedia.org/wiki/CAP_theorem ?
  - packetlost 1y ago
    Pretty much always A. In systems like this, it's better to deny transactions than allow inconsistencies.
  - jekrb 1y ago
    non-potatoe hardware and elbow grease in the software https://github.com/anza-xyz/agave https://github.com/anza-xyz/agave
    - oblio 1y ago
      Is there a white paper for what would be a revolutionary discovery in the field of software?
DennisP 1y ago
There's not that much CPU involved. Most of the stablecoins are on Ethereum, and I think the rest are on other proof-of-stake platforms, not Bitcoin.
- jakewins 1y ago
  Ethereum is able to process something like 150 transactions per second, using about 1,000,000 validator machines. Postgres running on a single Raspberry Pi is something like 200 TPC-B read/write transactions per second. Saying Ethereum “is not using very much CPU” is baffling to me. It is the state-of-the-art in this regard, and it uses something like six orders of magnitude more CPU than a normal database running a ledger workload?
  - ricericerice 1y ago
    there's 1,000,000 validators (defined as a public key), but you can run multiple validators per machine. Most estimates that crawl the p2p network to index nodes comes out at like ~20,000 machines doesn't invalidate your point but it at least shaves off a few orders of magnitude and a single PG node is not a fair comparison, we're talking 100% uptime global networks. Visa does about 70,000 transactions per second - how many servers do you think they run across their infra?
    - wredcoll 1y ago
      > Visa does about 70,000 transactions per second - how many servers do you think they run across their infra? So uh, how do we scale etherum's 150tps to that?
      - DennisP 1y ago
        ZK rollups which greatly compress data on chain without losing security guarantees, combined with temporary storage using data sampling so each node only has to store a small portion of it. The zk rollups are live and support significantly more than 150tps today, and the data sampling goes live in November. There's a lot more work to be done but that puts the major pieces in place.
sunshine-o 1y ago
> Because literally the only point is to avoid the existing banking system and you can do that with a postures database with much less cpu involved. Ethereum is actually very low resource intensive nowadays. You can run a validator node on a RPI, a full sync node on a Intel N100 minipc with a big fast SSD and the "light clients" can probably run on something very small. I have seen banks having to bring semi-trailers full of diesel generators to plug them to their mainframe because the current requirements were too high for the grid during big batch jobs.
- topranks 1y ago
  Just because they ditch the proof-of-work doesn’t make it efficient. The blockchain structure, the validation mechanism etc are still a very inefficient way to do general compute or database type functions.
- ChadNauseam 1y ago
  I like crypto (I'm formerly in the industry), but that's not quite a fair comparison. 1. Running a validator is inexpensive in terms of compute, but there are 1,000,000 validators or something, which adds up to a lot of CPU usage. Of course, I think it's insanely awesome that you can run some code on Ethereum and it'll be replicated on 1,000,000 independently-operated machines, but it's not a very CPU-efficient strategy. 2. Banks doing those batch jobs probably had much higher TPS than ethereum.
  - sunshine-o 1y ago
    > Banks doing those batch jobs probably had much higher TPS than ethereum. Yes the platform running in most banks, usually built on what we call "mainframes", is still mind blowing and with incredible performance. Also just one of those CPU is about the price of a house... Also the requirements I cited is for running an Ethereum mainnet "Layer 1" node. And most "TPS" happens on the layer 2s anyway. So it is hard to compare technically. But one thing for sure is becoming an active participant in the Ethereum mainnet has a very low barrier. They got rid of the whole intensive "Proof of work" part about 5 years ago. For a full sync node the waste is more at the bandwidth and disk levels.

algo_lover 1y ago

But with multiple parties involved, who has the rights to read and write to the postgres instance? How do we make sure transactions were not forged? How do we know data at rest is not being tampered with? Blockchain solves that. Newer blockchain protocols especially an L1 is much faster, easier on the environment, and provides all the immutability, transparency, and traceability benefits.

topranks 1y ago
We need to trust those running the system. Societies cannot function without trusted intermediaries, in finance and many other things. If we are in a democracy then the government regulates such organisations and should punish those who do not comply. Blockchain doesn’t scale as a replacement so the point is moot.
oblio 1y ago
You know you can just use regular cryptography to validate data, right? Also, you always have to trust someone, in this case Stripe. Regarding L1 blockchains, how exactly do they solve the speed problem for a distributed global database that needs to be replicated everywhere for the security guarantees to actually work? What do they forgo out of https://en.m.wikipedia.org/wiki/CAP_theorem https://en.m.wikipedia.org/wiki/CAP_theorem ?
- packetlost 1y ago
  Pretty much always A. In systems like this, it's better to deny transactions than allow inconsistencies.
- jekrb 1y ago
  non-potatoe hardware and elbow grease in the software https://github.com/anza-xyz/agave https://github.com/anza-xyz/agave
  - oblio 1y ago
    Is there a white paper for what would be a revolutionary discovery in the field of software?

DennisP 1y ago

There's not that much CPU involved. Most of the stablecoins are on Ethereum, and I think the rest are on other proof-of-stake platforms, not Bitcoin.

jakewins 1y ago
Ethereum is able to process something like 150 transactions per second, using about 1,000,000 validator machines. Postgres running on a single Raspberry Pi is something like 200 TPC-B read/write transactions per second. Saying Ethereum “is not using very much CPU” is baffling to me. It is the state-of-the-art in this regard, and it uses something like six orders of magnitude more CPU than a normal database running a ledger workload?
- ricericerice 1y ago
  there's 1,000,000 validators (defined as a public key), but you can run multiple validators per machine. Most estimates that crawl the p2p network to index nodes comes out at like ~20,000 machines doesn't invalidate your point but it at least shaves off a few orders of magnitude and a single PG node is not a fair comparison, we're talking 100% uptime global networks. Visa does about 70,000 transactions per second - how many servers do you think they run across their infra?
  - wredcoll 1y ago
    > Visa does about 70,000 transactions per second - how many servers do you think they run across their infra? So uh, how do we scale etherum's 150tps to that?
    - DennisP 1y ago
      ZK rollups which greatly compress data on chain without losing security guarantees, combined with temporary storage using data sampling so each node only has to store a small portion of it. The zk rollups are live and support significantly more than 150tps today, and the data sampling goes live in November. There's a lot more work to be done but that puts the major pieces in place.

sunshine-o 1y ago

> Because literally the only point is to avoid the existing banking system and you can do that with a postures database with much less cpu involved. Ethereum is actually very low resource intensive nowadays. You can run a validator node on a RPI, a full sync node on a Intel N100 minipc with a big fast SSD and the "light clients" can probably run on something very small. I have seen banks having to bring semi-trailers full of diesel generators to plug them to their mainframe because the current requirements were too high for the grid during big batch jobs.

topranks 1y ago
Just because they ditch the proof-of-work doesn’t make it efficient. The blockchain structure, the validation mechanism etc are still a very inefficient way to do general compute or database type functions.
ChadNauseam 1y ago
I like crypto (I'm formerly in the industry), but that's not quite a fair comparison. 1. Running a validator is inexpensive in terms of compute, but there are 1,000,000 validators or something, which adds up to a lot of CPU usage. Of course, I think it's insanely awesome that you can run some code on Ethereum and it'll be replicated on 1,000,000 independently-operated machines, but it's not a very CPU-efficient strategy. 2. Banks doing those batch jobs probably had much higher TPS than ethereum.
- sunshine-o 1y ago
  > Banks doing those batch jobs probably had much higher TPS than ethereum. Yes the platform running in most banks, usually built on what we call "mainframes", is still mind blowing and with incredible performance. Also just one of those CPU is about the price of a house... Also the requirements I cited is for running an Ethereum mainnet "Layer 1" node. And most "TPS" happens on the layer 2s anyway. So it is hard to compare technically. But one thing for sure is becoming an active participant in the Ethereum mainnet has a very low barrier. They got rid of the whole intensive "Proof of work" part about 5 years ago. For a full sync node the waste is more at the bandwidth and disk levels.