The Gridcoin mining experiment - What is going on with VGTU

in utopian-io •  last year

When I started the experiment I chose VGTU with a purpose. It's a very stable project and CPU only. Since September last year I never experienced any issues. The only interruption in the data gathering has been a power issue with my host at the end of Januari. What's happening now is really strange. If anyone has a clue please let me know.

About the experiment

If you want more background on the experiment I refer to these posts:

Change in tasks

Up until now task duration did not vary a lot. This is an example of last week (CPU time in 7th column):


And this is what it is looking like now:


As you can see quite a few tasks resulted in errors. This has never happened before. Well, not really never, I think maybe once or twice, but certainly not to this extent. A second observation is that the task that did complete successfully show a big increase in processing time. This is completely new so something must be going on. If this continues I'll try to contact the VGTU admin and request for some feedback.

RAC is rising despite erroneous tasks

This is completely unexpected. If the host is spending time crunching tasks that result in errors that processing time is lost and no points are awarded. On average less points should be rewarded. This of course depends on where in the task the error occurs. It might as well be at the very beginning which would have close to zero impact. None the less I certainly did not expect the RAC to rise because of this. You can see that the series are all over the place. RAC has never behaved like this before and points per second is obviously doing strange things as well.


Team RAC

Team RAC is skyrocketing. Although I don't show it in the charts I keep track of the share the top 10 users have in the team RAC. Back when I started the top 10 accounted for about 70%. Since dutch left that has gone down to about 60% and it remains stable. We can assume that the increase in RAC is quite uniformly spread. Even though the pools are increasing sharply there must be quite a few smaller crunchers rising at the same pace to compensate and keep the share stable. Most probably this recent rise is caused by crunchers moving away from greylisted projects to VGTU. As always GRC price influences the mining revenue most.


Top 10 users

This is an overview of the top 10 contributors on VGTU. The pools keep increasing their RAC at a steady pace. As mentioned before I like this a lot because let's not forget this is aiding scientific research. Not much new here though.



Posted on - Rewarding Open Source Contributors

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

jefpatat: You are not alone! I observe the same: There are more errors since a few days. VGTU has been rock-solid, but there are quite a few errors on my two computers since around 09th of March 2018.

Your observed RAC increase is because of the work-credit to computer time disparity since around the same time (7th of March). See below:


"Glad" it's not my host alone...

Leaving all other aspects the same credits earned vs CPU time should be constant and thus not impact RAC. RAC will be updated at larger intervals with larger credits. This might create some extra fluctuation but on average should not result in what is being observed here.

I signed up to VGTU only a couple of days ago and had more than 50% of my first batch end in computation errors. The errors all happen immediately after the tasks started so i'm not bothered by it too much.
The run time of tasks has been between 5 and 6 hours for me.


Thanks for the info.


The run time of tasks has been between 5 and 6 hours for me.

An update on work unit run times:
I have been getting longer work units this weekend ranging from 9 hours to 17 hours.

Thank you for the contribution. It has been approved.

You can contact us on Discord.

Is it possible that you have a corrupted OS or memory module? I had a streak of errors for a while with a GPU that was factory overclocked and was unstable.

This post was upvoted by Steemgridcoin with the aim of promoting discussions surrounding Gridcoin and science.

This service is free.

Hey @jefpatat I am @utopian-io. I have just upvoted you!


  • You have less than 500 followers. Just gave you a gift to help you succeed!
  • Seems like you contribute quite often. AMAZING!

Community-Driven Witness!

I am the first and only Steem Community-Driven Witness. Participate on Discord. Lets GROW TOGETHER!


Up-vote this comment to grow my power and help Open Source contributions like this one. Want to chat? Join me on Discord

Have you reached to VGTU?

I feel like there is a lot of data in the post, but little information.


I just had another erroneous task so I contacted them. The information is scattered over many posts I have made. I can't repeat the same info over and over. If you have any specific questions feel free to ask them.


Rosetta have had a few batches like this where the client software decides early on that a work unit is not worth processing and exits without creating an output file. BOINC Manager then reports a computation error because it doesn't find an output file to return to the server.
Perhaps there is a similar mechanic going on with VGTU in recent batches.