# Neptune needs UPoW if it wants to survive **URL:** \ **Category:** General\ **Created:** [September 13, 2026, 7:24am UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391 "2026-09-13T07:24:44Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![neptunenigga64](https://talk.neptune.cash/letter_avatar_proxy/v4/letter/n/85e7bf/32.png) [@neptunenigga64](https://talk.neptune.cash/u/neptunenigga64)\ **Post date:** [September 13, 2026, 7:24am UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391/1 "2026-09-13T07:24:44Z")
When bitcoin was launched, it launched with two major innovations: decentralized trustless money and also a mechanism that’s able to attract large amounts of compute. Recently there’s been major strides in useful proof of work that have been gaining a lot of traction, notably: PRL [https://pearlresearch.ai/](https://pearlresearch.ai/) found a way to do PoW by doing MatMul meaning PoW no longer compete with AI, their chain can be merge mined with useful proof of work AI data centres. NOCK [https://nockchain.org/](https://nockchain.org/) they made ZKPoW where miners generate zero knowledge stark proofs to secure the network, and the roadmap is for these proofs to verify arbitrary computation, you could use them to help neptune even. The end goal is for nockchain to be the chain that attracts large amounts of compute and verifies all the work for the world to use. I actually believe NOCK to be a massive threat to neptune as it is basically the same thing as neptune, built using zk starks, programmable, scalable, private money for the world, except with the addition of UPoW which will act as a flywheel for it. Miners produce ZK proofs for all of humanity to use which helps the chain function. I think UPoW is the last key missing piece for neptune’s roadmap that the team should strongly consider adding. The end goal of money was never just human → human transactions. Money is ultimately a coordination layer for resources, and compute is becoming the most important resource on earth. The chain that attracts the most compute becomes the marketplace for it wins. With your knowledge of ZK-Starks, you could build this. I think tritonVM is a competitive product, Nockchain also has some issues during the launch with some people saying it wasn’t that fair. I think NPT being completely fairly launched can ultimately win if you take this approach. ---
**Author:** ![nptbtc](https://talk.neptune.cash/letter_avatar_proxy/v4/letter/n/d6d6ee/32.png) [@nptbtc](https://talk.neptune.cash/u/nptbtc)\ **Post date:** [September 18, 2026, 12:39pm UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391/2 "2026-09-18T12:39:53Z")
**Case 1: You want to perform certain “genuinely useful” ZK work (i.e. work that’s not just finding zeroes that are useless other than to secure a cryptocurrency).** Wouldn’t you want to do this as directly and as cheaply as possible? (Why would you want to bother with securing some cryptocurrency?) **Case 2: You want to secure a cryptocurrency and earn its mining rewards.** Why would you want to perform some other so-called “useful” computation? (We already assumed it’s not useful _for you_ - you couldn’t care less about pet projects X, Y, and Z.) Clarifications on the above would help me understand these things better. ---
**Author:** ![neptunenigga64](https://talk.neptune.cash/letter_avatar_proxy/v4/letter/n/85e7bf/32.png) [@neptunenigga64](https://talk.neptune.cash/u/neptunenigga64)\ **Post date:** [September 19, 2026, 6:08am UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391/3 "2026-09-19T06:08:59Z")
Case 1: It is cheaper to do it whilst mining the cryptocurrencies. There is no real overhead, it is basically the same price to do ur work, u r just mining with the work you are already doing earning extra. So basically every single AI data centre will just be mining PRL whereas NPT/BTC will be competing with them. Obviously PRL is going to gather more compute as it isn’t competing with data centres for hardware. Case 2: If you do computations that can be used for useful work, but don’t do useful work with it you will just earn less than those that do useful work and sell it whilst mining. The more interesting thing is the long term vision though. Whatever cryptocurrencies is able to attract the most compute will eventually become the most valuable as it’ll be money backed by that compute. A simple privacy coin like NPT is great, but is 100% going to fall short if it doesn’t have UPoW. ---
**Author:** ![neptunenigga64](https://talk.neptune.cash/letter_avatar_proxy/v4/letter/n/85e7bf/32.png) [@neptunenigga64](https://talk.neptune.cash/u/neptunenigga64)\ **Post date:** [September 24, 2026, 3:44pm UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391/4 "2026-09-24T15:44:03Z")
I mean I think for NPT. ZK-Pow (what NOCK built) works perfectly right? Anyone doing ZKPs for neptune can also be securing the chain and earning tokens. ---
**Author:** ![nptbtc](https://talk.neptune.cash/letter_avatar_proxy/v4/letter/n/d6d6ee/32.png) [@nptbtc](https://talk.neptune.cash/u/nptbtc)\ **Post date:** [September 26, 2026, 11:25am UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391/5 "2026-09-26T11:25:16Z")
> [@neptunenigga64](#): > > So basically every single AI data centre will just be mining PRL The coin’s hashing power would then be coming mostly from the biggest AI hardware centers. For **decentralization** , we should ask: Someone finds an energy source in the middle of nowhere. Is it just as feasible to setup AI hardware there as it is BTC/NPT miners? Case 2 is still unanswered for me. I should have phrased it like this: You have capital to invest (anywhere on the planet). Why would you necessarily earn more return on capital by investing in “useful” work (plus mining PRL/NOCK), rather than investing in hardware that mines BTC (SHA-256 hashes) or mines NPT (Tip5 hashes)? (The point being, even if these AI hardware centers have far more power, BTC/NPT could still end up having _more than enough_ power in the opinion of most savers.) ---
**Author:** ![nptbtc](https://talk.neptune.cash/letter_avatar_proxy/v4/letter/n/d6d6ee/32.png) [@nptbtc](https://talk.neptune.cash/u/nptbtc)\ **Post date:** [September 26, 2026, 11:30am UTC](https://talk.neptune.cash/t/neptune-needs-upow-if-it-wants-to-survive/391/7 "2026-09-26T11:30:23Z")
Also, I’m not sure how much STARK/succinctness rely on the Tip5 hash function. (Do you think we will ever see Tip5 ASICs, @aszepieniec ?)