# Should we change PoW algorithm?

**URL:** <https://forum.firo.org/t/should-we-change-pow-algorithm/477>\
**Category:** Firo Improvement Proposals\
**Created:** [November 26, 2019, 3:11pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477 "2019-11-26T15:11:51Z")\
**Posts on this page:** 20\
**Page:** 2

<div class="post-metadata">

**Author:** ![anon28269187](https://forum.firo.org/letter_avatar_proxy/v4/letter/a/f6c823/32.png) [@anon28269187](https://forum.firo.org/u/anon28269187)\
**Post date:** [January 14, 2020, 5:03pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/21 "2020-01-14T17:03:22Z")

</div>

My two cents for what they are worth.

I believe in Satoshis original vision 1 CPU, 1 vote. The egality of CPU mining is that almost everyone can do it, even the poor, who are most in need of crypto to escape their economic situation and to secure it.

CPU mining may not be a popular approach however it would allow anyone with a cellphone to mine Zcoin, provided that a mining app is created. Almost everyone has a cellphone, therefore those interested in mining, could do it on their device, not just on a computer as not everyone has one of those.

CPU mining would level the playing field for mining, no pools would be necessary as everyone can do it, it would reorg the entire space and make it much more attractive to people, imo.

Heck, i would love to mine Zcoin but joining a pool, using my computer connected 24/7 then having to constantly upgrade my GPU so i can get a few percent of a ZCoin is just not attractive. But if i could run on my phone in the background that would be easy. Another benefit is that it would create extra nodes and increase blockchain security.

My vote is for CPU mining only, no GPU, no ASICS. Keep it simple, keep it in the hands of the everyday person, keep it on your computer and on your phone. Zcoin should never become elitist and only for tech geeks.

---

<div class="post-metadata">

**Author:** ![ninjamirko](https://forum.firo.org/letter_avatar_proxy/v4/letter/n/fbc32d/32.png) [@ninjamirko](https://forum.firo.org/u/ninjamirko)\
**Post date:** [January 15, 2020, 5:35am UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/22 "2020-01-15T05:35:47Z")

</div>

Stay gpu-friendly. It could be ProgPow, it have similar efficiency on amd and nvidia so good decentralization in mining.

---

<div class="post-metadata">

**Author:** ![ctrlshift](https://forum.firo.org/letter_avatar_proxy/v4/letter/c/f14d63/32.png) [@ctrlshift](https://forum.firo.org/u/ctrlshift)\
**Post date:** [January 15, 2020, 3:10pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/23 "2020-01-15T15:10:13Z")

</div>

It seems fighting against ASICs is an ongoing battle. MTP was a good shot at that, but the bandwidth is a real issue. The question is what is the main focus of Zcoin - is it being the anonymous version of BTC or the evenly distributed coin. I used to be an ASIC hater, but maybe they make sense - if there is enough demand for the coin, the free market innovates and finds optimal solutions.  
Personally I would tweak MTP with minimal effort so it wouldn’t use so much network bandwidth and leave it. If someone develops FPGAs or ASICs - let them. About half of the coin supply is already mined and by the time someone develops them, it’s not such a big deal anymore.

---

<div class="post-metadata">

**Author:** ![anon28269187](https://forum.firo.org/letter_avatar_proxy/v4/letter/a/f6c823/32.png) [@anon28269187](https://forum.firo.org/u/anon28269187)\
**Post date:** [January 18, 2020, 3:00am UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/24 "2020-01-18T03:00:42Z")

</div>

How do you come to consensus on such an issue?

My preference is CPU mining only, but it is not in your list.

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 20, 2020, 3:54pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/25 "2020-01-20T15:54:30Z")

</div>

Do elaborate your thoughts on CPU mining!

Zcoin actually had experienced CPU mining before in our early days and it was a horrible experience when people used cheap cloud computing credits to mine and also botnets. Also coins that adopted RandomX such as Loki found that their hashrate was dominated by botnets, a problem so serious that they’re abandoning PoW completely and moving to PoS.

Due to the vast amounts of idle servers, cloud computing and botnets, CPU mining may not be as democratic as we thought.

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 20, 2020, 3:54pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/26 "2020-01-20T15:54:56Z")

</div>

Note that with ProgPOW the code already has been written so it’s not too difficult to plug it in. I believe a smaller coin called Bitcoin Interest did this.

---

<div class="post-metadata">

**Author:** ![greer](https://forum.firo.org/user_avatar/forum.firo.org/greer/32/408_2.png) [@greer](https://forum.firo.org/u/greer)\
**Post date:** [January 21, 2020, 5:48pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/27 "2020-01-21T17:48:52Z")

</div>

ProgPoW is an excellent choice. Just tune the parameters to something unique so that you’re not sharing the algorithm with Ethereum, Zano, Epic or anyone else using it.

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 23, 2020, 7:52am UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/28 "2020-01-23T07:52:16Z")

</div>

For transparency’s sake I am in touch with Kristy Leigh who created ProgPOW who will also be creating a [ZCS](https://zcs.zcoin.io) proposal on this.

ProgPOW has been controversial due to Kristy’s involvement and some of the other authors being anonymous mainly because she does work for FPGAs, ASICs and things like that so people are rather nervous as to her intentions. That being said some of the people making the largest noise against her are people who are involved in Ethash ASICs such as Linzhi who have a very clear economic incentive to not make ProgPOW a success as it will render their ASICs worthless.

However that being said, I believe we should judge the work based on facts and evidence and while I guess personality and intentions do play a role, what is more important is the actual product itself which has been audited by independent third parties ([https://medium.com/ethereum-cat-herders/progpow-audits-released-ed4973ebe073](https://medium.com/ethereum-cat-herders/progpow-audits-released-ed4973ebe073)) with both a software AND hardware audit which is rare in itself.

In a way Kristy’s involvement with hardware makes her one of the best suited people to create such an algorithm as well. MTP was designed by people who didn’t really have hardware experience and made wrong assumptions on the limitations of hardware development.

Efficient miners are already out and available. So far there’s been a proposed ASIC design by Linzhi which [claims a 1% efficiency increase](https://medium.com/@Linzhi/eip-1057-progpow-open-chip-design-for-only-1-cost-power-increase-eip-1057-progpow-d106d9baa6eb) so really that actually is a very good sign of ProgPOW’s anti-ASIC properties.

2Miners (who runs Zcoin’s largest pool) also wrote an excellent post introducing ProgPOW [https://2miners.com/blog/ethereum-progpow-explained/](https://2miners.com/blog/ethereum-progpow-explained/)

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 23, 2020, 7:56am UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/29 "2020-01-23T07:56:19Z")

</div>

Pinned globally on January 23, 2020, 7:56am UTC



---

<div class="post-metadata">

**Author:** ![greer](https://forum.firo.org/user_avatar/forum.firo.org/greer/32/408_2.png) [@greer](https://forum.firo.org/u/greer)\
**Post date:** [January 23, 2020, 6:11pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/30 "2020-01-23T18:11:57Z")

</div>

> [@reuben](#):
>
> claims a 1% efficiency increase

Linzhi’s post was flawed. See response [Responses – Medium](https://medium.com/p/d106d9baa6eb/responses/show)

---

<div class="post-metadata">

**Author:** ![cryptohype](https://forum.firo.org/letter_avatar_proxy/v4/letter/c/ba8739/32.png) [@cryptohype](https://forum.firo.org/u/cryptohype)\
**Post date:** [January 24, 2020, 1:12am UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/31 "2020-01-24T01:12:36Z")

</div>

progpow coming for zcoin. my dream coming true.

---

<div class="post-metadata">

**Author:** ![djsoyboi](https://forum.firo.org/letter_avatar_proxy/v4/letter/d/ba9def/32.png) [@djsoyboi](https://forum.firo.org/u/djsoyboi)\
**Post date:** [January 25, 2020, 8:51pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/32 "2020-01-25T20:51:26Z")

</div>

Switching to ProgPow is good for the miners in general. But my concern with algo change is it might create unwanted hard forks like zcoin reuben vision and zcoin cash. RandomX = botnet 101 so hell no. Other than that progpow is a decent choice

---

<div class="post-metadata">

**Author:** ![anon28269187](https://forum.firo.org/letter_avatar_proxy/v4/letter/a/f6c823/32.png) [@anon28269187](https://forum.firo.org/u/anon28269187)\
**Post date:** [January 27, 2020, 3:04pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/33 "2020-01-27T15:04:28Z")

</div>

Thanks Reuben.

An important point to consider is that Monero, your primary competitor has moved to CPU mining, specifically to the RandmomX algorithm.

> [@reuben](#):
>
> a horrible experience when people used cheap cloud computing credits to mine and also botnets. Also coins that adopted RandomX

Considering that Monero has changed to RandomX and is facing the same potential issue with mining bots, but, still went ahead with the fork and change, they would have given all due consideration to this issue, however the positives still outweighed any potential negatives, which is why they proceeded.

Potential botnet mining were not considered a negative to stop a change to CPU mining on RandomX.

> [@reuben](#):
>
> their hashrate was dominated by botnets, a problem so serious that they’re abandoning PoW completely and moving to PoS.

I think this misses the point. Whether it a genuine miner like is mining Zcoin on his computer (or phone) or whether it is a cloudserver or botnet, the point is that the network is being secured through mining and that it stops a potential 51% attack. The same problem exists with ASIC mining or even GPU mining, large mining pools or entities are a threat in terms of attack, however, CPU mining is more egalitarian and permits anyone with a computer to mine, thereby securing the network more.

We should note what happened with the Monero network when they switched to CPU mining.

 ![Screenshot from 2020-01-27 08-55-49](https://forum.firo.org/uploads/default/original/1X/52551be6f6dd525ac7a41c5c9666dcf24d1e2f70.png)

(source:[CPU Mining Dominates Monero Since RandomX Upgrade - MinerUpdate](https://minerupdate.com/news/miner-insights/cpu-mining-dominates-monero-since-randomx-upgrade))

My point:

- CPU mining is egalitarian; it permits anyone with a computer to mine Zcoin whereas having ASICS or GPU mining limits mining to people from developed countries with a substantial monetary base.

- Monero switched to CPU mining using RandomX specifically because of the ASIC problem and inequality. Monero is much larger than Zcoin and have engaged in extensive research before making such a dramatic decision, however the consensus after all the reviews, the pros and cons of RandomX, they still went ahead and made the switch

- Lastly, Zcoin switching to CPU only mining using RandomX could be of massive benefit to everyone, as Monero miners can also mine Zcoin, thereby further securing the network.

Forget price, and think long term mining viability and security of the Zcoin blockchain and ask yourself, how many people in this world can afford a spec’d out GPU mining rig or ASIC and how many people can afford a simple computer to mine Zcoin.

---

<div class="post-metadata">

**Author:** ![anon28269187](https://forum.firo.org/letter_avatar_proxy/v4/letter/a/f6c823/32.png) [@anon28269187](https://forum.firo.org/u/anon28269187)\
**Post date:** [January 27, 2020, 3:07pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/34 "2020-01-27T15:07:01Z")

</div>

Monero switched to CPU mining using RandomX.

Do you think that the Monero devs did not consider this issue? Of course they did, but they went ahead anyway because the pros outweigh the cons.

Botnet mining isn’t that big an issue otherwise Monero would not have switched. Consider that.

---

<div class="post-metadata">

**Author:** ![anon28269187](https://forum.firo.org/letter_avatar_proxy/v4/letter/a/f6c823/32.png) [@anon28269187](https://forum.firo.org/u/anon28269187)\
**Post date:** [January 27, 2020, 3:13pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/35 "2020-01-27T15:13:48Z")

</div>

One article that you and the devs need to read regarding RandomX is this:

> **[Monero: ‘Size of mining bots on RandomX will be smaller’](https://eng.ambcrypto.com/monero-size-of-mining-bots-on-randomx-will-be-smaller/)**
>
> Monero recently upgraded its network, keeping in line with its constant efforts to steer clear of ASIC centralization. One of the major network upgrades included the shift of the mining algorithm from CryptoNightR to RandomX. The new algorithm not...

The summary being, " Subsequently, Jethro Grassie, developer and contributor of [Monero], stated that since there was a requirement of [2 GB RAM](https://eng.ambcrypto.com/monero-the-efficiency-of-the-inefficient-randomx/) it would be “pretty difficult to hide” the mining process. He further stated that there hasn’t been “any evidence to suggest that [mining malware and botnets are] significant, adding “I don’t see anything with RandomXthat suggests that number is going to grow."

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 27, 2020, 3:26pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/36 "2020-01-27T15:26:50Z")

</div>

Monero devs were not in favor of RandomX. It was the community that wanted it.

At one point Monero proponents said that having botnets secure is a good thing…because it meant the botnets weren’t doing other nefarious activity. I think this is a horrible argument as botnets can always be redeployed to do other things.

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 27, 2020, 3:41pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/37 "2020-01-27T15:41:36Z")

</div>

Thanks for the time you spent replying to this it is greatly appreciated. I however do wish to reply to some points you make.

51% attacks are more likely the more centralized hashrate is so there is an issue if it’s controlled by botnets/a small group or entity. However we hope to get rid of this risk via chainlocks. But it’s still not good if the majority of the coins are going to a few people.

Zcoin has had experience with CPU mining algorithms early in our history and in fact it wrecked us badly. Loki also has had a similar experience with RandomX. Monero’s network is a lot larger so it may not be as pronounced.

Many experts believe a RandomX ASIC is possible even though it is more challenging. Looking at the audits they more focus on the cryptography and the implemenations when compared to ProgPOW there is someone with hardware experience that analyzed how to go about building a ProgPOW ASIC and where efficiencies can be gained.

Hashrate ‘doubled’ means nothing. It’s always about the relative number of individuals and devices securing it. When you change algorithm you can’t compare hashrate.

---

<div class="post-metadata">

**Author:** ![reuben](https://forum.firo.org/user_avatar/forum.firo.org/reuben/32/1171_2.png) [@reuben](https://forum.firo.org/u/reuben)\
**Post date:** [January 27, 2020, 3:52pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/38 "2020-01-27T15:52:47Z")

</div>

Note our implementation of MTP is even harder then since it uses 4 gb of RAM as well. Our personal opinion is that 2 gb of RAM nowadays is way too small.

---

<div class="post-metadata">

**Author:** ![anon28269187](https://forum.firo.org/letter_avatar_proxy/v4/letter/a/f6c823/32.png) [@anon28269187](https://forum.firo.org/u/anon28269187)\
**Post date:** [January 27, 2020, 3:56pm UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/39 "2020-01-27T15:56:25Z")

</div>

> [@reuben](#):
>
> But it’s still not good if the majority of the coins are going to a few people.
> 
> Zcoin has had experience with CPU mining algorithms early in our history and in fact it wrecked us badly. Loki also has had a similar experience with RandomX. Monero’s network is a lot larger so it may not be as pronounced.

I understand. Once burned, twice shy, however, my main point was that CPU mining is egalitarian and permits essentially everyone who has a computer to mine, even those people from developing countries like Sudan, Ethiopia, etc, where there is relative poverty compared to Thailand, but some of those people may have a computer but not a GPU rig, those people could mine whereas if Zcoin is GPU focused, they are denied.

Every mining solution has problems, whether ASIC, GPU, CPU or a combination, but you must consider how to benefit the most people from your blockchain.

Adendum to the point of the Monero community choosing to switch to RandomX, the lead dev of Monero Alexander Blair said,

> “I like the fact that miners that are smaller have a chance to do mining now, with large [GPU]rigs becoming the norm […] building a huge mining farm, that kind of takes away from the idea of a coin that is for the people in my mind and centralizes for the people who have the money, which is the same exact argument you go with ASIC.”

> **[Monero’s RandomX is a ‘very interesting hashing algorithm’](https://eng.ambcrypto.com/moneros-randomx-is-a-very-interesting-hashing-algorithm/)**
>
> Monero, the world’s leading fungible cryptocurrency, has always taken a different path than the rest \[…\]

---

<div class="post-metadata">

**Author:** ![ctrlshift](https://forum.firo.org/letter_avatar_proxy/v4/letter/c/f14d63/32.png) [@ctrlshift](https://forum.firo.org/u/ctrlshift)\
**Post date:** [January 30, 2020, 10:34am UTC](https://forum.firo.org/t/should-we-change-pow-algorithm/477/40 "2020-01-30T10:34:18Z")

</div>

Please explain why egalitarianism is so important and why it should be the focus of Zcoin? Such a small project can only have one main focus which for Zcoin should be stable anonymity, not utopian decentralized mining. MTP was a good attempt against ASICs and I tihnk it should be optimized a bit and then focus on privacy and speed.

[Previous page](https://forum.firo.org/t/should-we-change-pow-algorithm/477.md?page=1)

[Next page](https://forum.firo.org/t/should-we-change-pow-algorithm/477.md?page=3)
