# Gravity Wallet

**URL:** https://forum.hypercore.one/t/gravity-wallet/270
**Category:** Development
**Created:** [December 10, 2023, 7:32am UTC](https://forum.hypercore.one/t/gravity-wallet/270 "2023-12-10T07:32:15Z")
**Posts on this page:** 1
**Showing post:** 24

<div class="post-metadata">

### Author: ![vilkris](https://forum.hypercore.one/user_avatar/forum.hypercore.one/vilkris/32/47_2.png) [@vilkris](https://forum.hypercore.one/u/vilkris)
#### Post date: [December 28, 2023, 11:33am UTC](https://forum.hypercore.one/t/gravity-wallet/270/24 "2023-12-28T11:33:09Z")

</div>

> [@Nostromo](#):
>
> I would estimate that the PoW calculations on the web app would take something like 10 - 20x longer than in Syrius. So maybe a transaction that would require 30 seconds of hashing with Syrius might take 7 minutes with the web app.

You also need to take [light mode and fast mode](https://forum.hypercore.one/t/plasma-pow-with-randomx/224#downsides-and-potential-issues-with-randomx-2) into consideration. Fast mode requires at least 2 GiB RAM which is not possible in a browser. There’s around a 5x difference between the modes based on my testing. If Syrius is running in fast mode, then the difference in speed becomes 50-100x, assuming your estimation of 10-20x is correct.

I would start with a PoC to prove that RandomX plasma PoW can be computed within a reasonable time in a browser before basing any decisions on that assumption. I doubt it will be fast enough to be a viable option and not worth the additional complexity it introduces.

---

_[View the full topic](https://forum.hypercore.one/t/gravity-wallet/270)._