Is the network usable for mobile users?

Based on the data obtained from Zcash Mempool, it contains a volume of over 6000 transactions. Because presently, there is a complete absence of mobile wallets that calculate the fee in accordance with ZIP 317, the transactions initiated through these wallets experience prolonged delays, sometimes spanning hours. Only zcashd works properly. But one cannot expect everyone to run a full node for making a simple transaction.

This situation raises concerns about the overall usability. First there was spam and then there is DOS.

What, in your opinion, will be the impression formed by a new user?

4 Likes

I would also point out that the spammy transactions keep goin on a lower scale:
https://zcashblockexplorer.com/transactions/a4bd06d94ea60a7e5b7d6db979b382271d4bcd4b73bddf830cd80e85273d28f0

There are hundreds of transactions like that. In deshielding and shileding.

1 Like

Per the ZIP-317 standard, there is an unpaid action threshold that was included to address this specific issue - allowing for a grace period where the typical user-user transaction that happens with mobile wallets (and thus results in a much smaller total number of actions within a specific transaction) would still be mined in a given block. The goal was to reduce the overall high transaction load by increasing fees while still providing a buffer for mobile wallet users to be able to transact ZEC in a wallet that did not yet implement a proportional fees.

There is also another element that needs to be considered - the scalability of the chain going forward. Denial of Service implies the inability to access a site or service due to a high level of garbage traffic. Each transaction in the high-load periods is valid, accepted by the network and processed. If someone wanted (pre-317) to pay $10 or so a day to send a high volume of txs to the network, the network allowed it. Similarly, those high-action txs are being accepted with fees being paid.

zcashd wallet allows sending to multiple recipients, which is used in a few cases that I am aware of. Some of the attempts to filter “spam” transactions can result in legitimate large-action txs being filtered out from a recipient’s wallet, not a desired behavior either.

1 Like

It’d have been better if the deployment of ZIP 317 was coordinated with an upgrade to mobile wallets. It seems that it wasn’t the case, was it?

Zingo’s default transaction fee is large enough that users report no loss of performance.

3 Likes

@jelly5649 YWallet is also impacted. The tx fees is too small. It can result in the tx being delayed or even expired.
@nick The unpaid action threshold is currently being used by the spam transactions in the mempool. There are thousands of them. Some legitimate user may end up getting their transaction through but it is not guaranteed.

In theory, that is true. However, in practice, the threshold was chosen to be so high that it doesn’t really happen.

Only cases I know where users pre-splitting a large number of notes, and the ECC sending test coins to their own team.

4 Likes

Can confirm :zebra: , looking forward to much need updates on the mobile side. :eyes:

I am not aware of Zingo transactions being affected by the Mempool Sand Blast.

1 Like

The upcoming version of Ywallet will give you the ability to adjust the fees as you see fit.

10 Likes

The spam transactions consume exactly the entire unpaid action limit. If they consumed any more then none of them would get into a block but currently every transaction (because they all still pay 0.00001) shares in the unpaid action limit and so either a block will contain a single spam transaction or perhaps a number of smaller normal transactions. Typically normal 2-input transactions will go through in a reasonably short amount of time because of the way miners combine transactions in the template. A customizable fee will allow for those txs to (more) properly pay for their actions and thus will not be affected at all.

1 Like

Custom fees - a significant step on the path to a fee market.

2 Likes

It would be better obviously if they were properly determined but in the interim it should be beneficial

1 Like
4 Likes