Quote from: teknohog on April 21, 2018, 06:43:16 PM

On the plus side, at least they are admitting that the PoW they devised to be ASIC-resistant in the past isn't very resistant now

FYI, the current Monero team had nothing to do with devising the original algorithm.

I feel it kind of did okay in holding off not only ASICs but even domination of CPUs by GPUs for four years, which is more than can be said for most if not all other algorithms which have tried that. But in the end the designers made some incorrect assumptions and it proved to be broken.

This isn't really just a question of technology moving on, as CPUs (and GPUs) have also improved significantly so ASICs had and have a moving target. But still the algorithm just did not do what it was supposed to do.

Quote from: teknohog on April 21, 2018, 06:43:16 PM

multi-gigabyte memory requirements

Large memory requirements will not in and of themselves make it ASIC resistant. If just the amount of memory is the the main issue, then ASIC builders can just attach a lot of cheap external memory as they are doing with Ethereum ASICs. (Not intended as a review of MTP generally, just the comment about memory usage.)

Quote from: bbc.reporter on April 22, 2018, 01:36:09 AM

Yes, Fluffypony's justification is because Monero is still a small network. But then are they also planning for the longterm future of the project? Of course, and I reckon they should make determined efforts to deal with ASICs now than going for short term solutions in between.

But if they are doing it to buy time for their ultimate solution then good on them. They are one of the best development teams in the cryptospace for a reason.

I'm pretty sure FP has explained his reasoning being that ASICs need to get to the point where they are commoditized before they can be relied upon for a decentralized mining network. Size of the network and industry could be part of that. And yes he has said that he expects the best solution to be ASICs eventually, but not yet.