Future of the Formula BOINC

[scole of TSBT]

While a year long event based on the Marathon points is a competition I don't think it mirrors F1 really other than F1 is season long but not year long. I do think being year long make some feel not as inclined to complete because it is a year long commitment. Yeah, I know I said that already. What might make more crunchers interested, with or without their entire team being on board is individual standings as well as team standings. That is a feature of F1.

I do like the idea of helping smaller projects. One way to limit the amount of bunkering would be to have 23 (or however many are on the schedule that year) 1, 2 or 3 day Grand Prix events and have them back to back, overlapping 2+ day events like the Pentathlon. That way there isn't a lot of time to bunker between events. That might help smaller projects like ODLK/ODLK1 hand an event. Having it all occur within a 30 day window would also allow teams to give FB an all out effort and still participate other events throughout the year.

Just some thoughts.

[Yeti]

For our team FB is more than only make smaller projects better known.

Most of us are crunchers since 10, 15 or even 20 years now and FB helps us to have a lot of fun under the year. You are not only running your projects stupidly day by day, but have to be flexible and Switch to the Sprint-Project and to make optimal decisions what and how to crunch this project.

I know several people that have upgraded their hardware to become more powerfull in races. And this always is not only for races but for all projects.

So, Races are an important part for making crunching more attractiv.

So, I would prefer if FB goes on in 2022.

What I would like to see better in future:

1) Better consideration of other races like Penta or others. If Penta is running it isn't attractiv to take in a FB-Sprint at this time

2) More / better switch between CPU- and GPU- oriented projects. Actual the last sprints have always been GPU-Races.

Yeti (Rechenkraft)

[blahblahblah]

Unban team gridcoin? If Mercedes was banned from F1 for winning too much, I too would lose interest

[UBT - Timbo]

Unban team gridcoin? If Mercedes was banned from F1 for winning too much, I too would lose interest

I think the issue is that Gridcoin (the team) has grown spectacularly fast, on the basis of the rewards that it offers to it's members.

And the vast majority of BOINC teams cannot offer such an incentive to increase the numbers of active members.

So, Gridcoin has undermined every other team, by way of financial inducement, and this is unfair, as no other team can afford to do this.

If I were a millionaire, I could easily undermine Gridcoin and every other BOINC team, by setting up a team and rewarding members with any reward of my choosing - whether it is money, or chocolates or airline tickets etc. Would that be "fair" and reasonable? Yes, it would be my money to give away as I want, but it would still upset most other teams, who would see their own team membership being drained away, just because of rewards.

And what would be the point of doing that?

The fact is that GRC "needed" a huge number of members in order that it's entire crypto eco-system had the computing power in order to sustain itself and create/establish/maintain its own blockchain. And by latching onto an existing user base of thousands of BOINC members allowed it to do this.

So, it is impossible to achieve a level-playing field in BOINC, while Gridcoin (the team) continues to offer rewards, even if members no longer "need" to join their BOINC team.


As for your F1 analogy - it doesn't really work: Over the last 20-30 years, various teams have become "prominent" in the sport for short periods of time: Tag/McLaren, Honda/Williams, Ferrari (with Schumacher), then Red Bull, and then Brawn/Mercedes. And in each case, teams are rewarded for their efforts with large financial benefits, which they can choose to invest in any area they want - and that helps to keep the competition between teams as new ideas come forward.

And the FIA change the rules every so often to ban certain designs that help make one manufacturers cars go faster (in the old days it was "ground effect", then we saw the dominance of "turbo engines", and various "aero" designs, and most recently the changes to the underfloor, which is reducing Mercedes advantage). And these FIA changes help to level the grid in terms of each teams lap times and endurance.

And please bear in mind that the difference in lap times between the 1st and last F1 driver in most qualifying laps is a second or two...so, there is almost a level playing field in F1 - witness Ocon at the last race, fending off Vettel...though they both had help due to the crashes/retirements at the beginning...even so, there is little to choose between most of the cars.

But I digress !!

regards
Tim

[Cougarpelardou]

+1

[Cougarpelardou]

Hello.

A suggested evolution of Formula Boinc for 2022.

Go from Formula 1 to Biatlhon which is much more "green".

The competition will include two separate rankings:
* a marathon ranking, keeping the current formula,
* a sprint ranking, keeping the current formula.

We will of course add a general ranking, based on the cumulative points obtained in the two events (usually 25 points for the first, 18 for the second ...).

Sorry for my English.
Have a nice week end.

[UBT - Timbo]

Hi all

Personally speaking I would like to see the following tweaks and changes made to Formula BOINC, either next year or perhaps introduced over the next year or two.:

1) The Marathon and Sprints are kept but with some subtle changes:

The Marathon points are awarded to the Teams only, based on their cumulative BOINC credits awarded from all their members over the course of the year (much as it is now).

So, the Marathon becomes a league table only for Teams.

However, FB points from Sprints are only awarded to individual members, on each day of the Sprint Weekend (ie Friday, Saturday and Sunday) which will help prevent some of the bunkering.

And the Sprint is only for members based on their daily credit earned during each Sprint (So, say 25 points are given to highest credits from a member on the Friday, another 25 points given to highest member credits on 2nd day and same again for 3rd day. And 18 FB pts for 2nd place, 15 for 3rd place etc).

At the end of the year, the Sprint league winner will be the individual who has earned the most FB points across the approximate 20 Sprints, over the 60 or so days of competition.

Team rewards (based on FB points earned over the year) and Individual rewards (based on FB points earned on each day of the Sprint) would be on offer.


2) All members must register individually (using their CPUID) and they must nominate a team (which may not be their own) that they will crunch for all season long (with no changes allowed).

This ensures that no team relies on inactive members (who know nothing about FB, esp if they have a high number of members) and it prevents swapping teams for political or tactical advantage.


3) The Sprint weekends follow the F1 schedule, but we could also introduce another "Challenge" for women members (as frankly I think there are too many men involved in BOINC as it stands) and perhaps even an E-Challenge where only specific "environmentally friendly" projects are "allowed" in said league (such as those that support NCI tasks?).


4) Every Team nominates a single member to act on the Sprint Committee, so that a wider range of opinions are catered for when choosing projects for Sprints - and with each League choosing it's own Sprint project individually. This helps reduce the load on project servers over any given Sprint weekend.


5) Promotion and demotion from each League is widened...so, maybe keep each League constrained to 25 teams (even introduce a League 4, if more teams want to join in. And at the end of each season say 6 or 8 teams are promoted/relegated...which will hopefully encourage more crunching so that teams work harder to avoid demotion or to gain promotion. Weaker, less active teams will be demoted or expelled if NO points are earned in the year.


regards
Tim

[Cougarpelardou]

Hello.
For sprints, it's a good idea to be able to pick a project multiple times a year.
Shouldn't a limit be set at 2 per year for the same project, so that several projects can benefit from FB sprints?

Sorry for my Google English

[UBT - Timbo]

Hello.
For sprints, it's a good idea to be able to pick a project multiple times a year.
Shouldn't a limit be set at 2 per year for the same project, so that several projects can benefit from FB sprints?

Sorry for my Google English

Hiya

Personally I don't think a limit should be set for one simple reason:

Sometimes, there is a limited choice of projects that have sufficient tasks available, in order for a Sprint to be a success (and we all know what happens when a project is chosen that cannot cope with a Sprint, in terms of tasks or bottlenecks).

And although a project has not been chosen for a 3rd time this year, it may be necessary to choose one, if the number of "suitable and online" projects is limited, due to external factors.

But I do agree with you about the principle that many different projects should be accommodated within each "season" so that many projects can benefit from our collective CPU/GPU resources.

regards
Tim

[Cougarpelardou]

If Rosetta or WCG are selected for one of the last 3 FB sprints, one of these two projects will have benefited from 15% of the FB sprints. I do not find this fair for the BOINC projects community.

This is my personal opinion. We are moving away from the original spirit of the FB.