The project was annonced in 2017, but if I remeber well it only started in fall 2018.
I joined WCG in February 2018, and I remember very well that at that time it wasn't running.

That's about 18 month, and I am pretty sure they they are keeping an eye on the quality of the work units since the beginning. It is just that it might not be that easy to understand what happens with these batch, and to address the problem.

It is correct. The project has been launched in late August 2017. I remember very well about it, I was giving a training in Copenhagen at those days.
I agree as well regarding the 3 years old software which is relatively bad performing since the beginning. I wrote multiple comments about this - especially during the first months after the project launch - since I feel that this bad performance is not fair for the contributors.
It is the reason why I do mostly avoid to contribute to this project.
I do not know what the reason is for the results below the expectations:
- Is it a question of science design?
- Is it a computational question?
If it is the latter, it would be fine to take the opportunity to improve the computation performance as well.
Cheers,
Yves

We just wrapped up a monthly call with the research team.

1. The project's average run time for most of this year has been ~150 CPU years per day. The researchers are finding this to be the "sweet spot" for giving them a good amount of data to analyze at a manageable pace.
2. They wrote a short project update, which we published after our last monthly call.
3. The research team's first paper is still being reviewed for possible publication, and they're continuing to work on additional papers as well.

Current status of work units:

Available for Download:  3,437 batches
In Progress:  7,472 batches (12,399,105 work units)
Completed:  294,485 batches total - 7,011 batches in the last 30 days - average of 233 batches per day
Estimated backlog: 14 days

Since the launch of OPN1, MIN's runtime has fallen well below the "sweet spot" you mention. Yesterday's runtime was 85 CPU years, with the stats not logging a statistic over 100 years since May 15th. How will this affect the researchers, if their data resources continue to dry up, as volunteers divert CPU time to OPN1?

thanks for the quick reply. Just wanted to ask - given how I've seen the stats for all other WCG drop projects since OPN1 launched. (The exception generally being FAAH2, ARP and SCC, obviously)

Keep up the good work.