If folks are interested in the old Monero PoW function (and, uh, the reason they changed it), I wrote up a thing about it a long time ago:

https://da-data.blogspot.com/2014/08/minting-money-with-mone...

The history of people trying to design GPU or ASIC-resistant proof-of-work functions is long and mostly unsuccessful. I haven't looked into RandomX; it's possible they've succeeded here (or possible that with the alt-coin market mining profitability tanking after Ethereum moved to proof-of-stake, it just wasn't worth it).

▲

tardedmeme 2 hours ago | parent | next [-]

RandomX is designed so that if you design a RandomX ASIC then you've designed a CPU. It writes and then executes random programs. To minimize the possible efficiency gains from matching the instruction set architecture, the same program is executed several thousand times, reducing the relative overhead of translating it to a different ISA.

▲

AureliusMA 3 hours ago | parent | prev | next [-]

There was a proposal on Ethereum that didn't succeed (progpow) since they were already in the late stage of transitionning to PoS. Ethereum did quite a good job at keeping asic advantage moderate (the speedup was 100% max - not orders of magnitude). RandomX is basically progpow that succeeded. You might be interested in Chia's Proof of Space and Time... and how it collapsed!

▲

hyc_symas an hour ago | parent | next [-]

ProgPow was ridiculously simple and would never have accomplished its goal. I covered it briefly in my Monerokon talk as well.

▲

idiotsecant an hour ago | parent | prev [-]

I don't know if PoW based approaches make much sense in the modern environment, anyhow -even very clever ones that provide ASIC resistance. Ethereum has been doing real proof of stake (and not delegated proof of stake which is both easier and terrible from a system safety perspective) for quite a while and it's seemingly cheap, effective, and robust.

▲

hyc_symas 39 minutes ago | parent [-]

PoW is useful in far more situations than PoS. A derivative of RandomX is now used to protect TOR too (Equi-X). https://github.com/tevador/equix/blob/master/devlog.md

	▲	idiotsecant 33 minutes ago \| parent [-]
		Yes I suppose it's difficult to stake something of value when the system you're securing is not stapled to a currency.

▲

ProllyInfamous an hour ago | parent | prev | next [-]

I partially heat my home by running the default Monero client on old Xeons (heat ejects near my desktoes). As I only mine when it's cold outside (using resistive heating), there is no actual net electricity cost. IMHO it's not "worth it" for an individual to buy equipment specifically to mine crypto... but if you already have an old machine AND you heat without a heatpump, it's a free hobby/heater.

----

To anybody else that is syncing a fresh monero blockchain copy (i.e. installing the official client), I recommend using the custom node flag ` --db-sync-mode safe ` — which is slower but corruption-avoiding — before node's initial bootup. Without safemode, any halt of the client will [most likely] corrupt the local blockchain (losing days of DL/verification).

Also, if you use an SSD for storing any blockchain (as recommended by monero team... but not by me), know that its lifespan will be greatly reduced from the constant IO/access. Personally, I recommend safemode (see above) on a 7200RPM spinner (HDDs effectively don't wear during IO/access).

----

What are your thoughts on running xmrig vs. the default getmonero.org client? Would you in general agree that monero remains ASIC-resistant?

▲

hyc_symas an hour ago | parent [-]

Just use a Linux laptop with a working battery so you never have to worry about power outages or other system crashes. In that case, you don't need safe sync mode, and you don't have to kill your SSD.

▲

ProllyInfamous an hour ago | parent [-]

Working battery ≠= avoiding system crashes | my local node has a UPS, and still Monero's client is dicey (Mac & Linux distros).

Particularly on its initial sync, Monero's daemon is flakeyAF.

If you (e.g.) don't allow `sync in background` (why is this not the default behavior?!), the official Monero client is notorious for locking up on wakeup. Once you kill the process, your local blockchain is [most likely] unusable.

Another reason to use safe-sync is (e.g.) if your system (Linux or whatnot) decides to update/restart during the several days it takes to sync-initially.

----

Just out of curiosity, why do you abuse an SSD so (safe-mode, or not)?

For SSD-diehards, I'd recomment getting a very large size because this'll last longer, presuming the drive self-levels.

	▲	hyc_symas 41 minutes ago \| parent [-]
		> Once you kill the process, your local blockchain is [most likely] unusable. Totally false. LMDB is perfectly crash-proof in that scenario and killing the process never damages the DB. The only thing that's not guaranteed is turning off syncs, in the face of an OS crash/power outage. If you don't sync, you're not abusing the SSD. If you run on Windows, the OS is too unstable to use without safe sync mode though.

▲

alcazar 3 hours ago | parent | prev | next [-]

This was a super interesting read, and it highlights exactly the strength of cryptocurrencies. They turn game theory in their favor, so egoistic players (I don't mean this in an offensive tone) contribute to making it stronger and safer for everyone else.

Thank you for sharing!

	▲	dgacmu 3 hours ago \| parent [-]
		They kinda do - I'll admit honestly that the final game I played in the cryptocurrency space I played solely to profit. (It was a minor, uh, **coin that didn't have a lot of redeeming value to start with). Though it turns out the incentives remained somewhat aligned: I ended up providing the developer with some security bug fixes to make sure someone couldn't mess with the cash cow. :) (To be clear: We were just optimizing mining; in the process of looking for ways to mine it faster, I found some security bugs and fixed them. We weren't exploiting the bugs, that crosses a line for me.)

▲

tomjen3 an hour ago | parent | prev [-]

They had to design a specialized verification function, which I imagine would be the easy way to break it.

The brilliant part of Bitcoin is that it uses very widely known crypto primitives - verification is the same as getting the right seed (you just happen to be told what the right seed is, rather than having to pay for it to be discovered).

	▲	hyc_symas 38 minutes ago \| parent [-]
		You must be on drugs. There is no separate specialized verification function. It's the same algorithm for verification as for mining.