More Stars ???

log in

Advanced search

Message boards : Number crunching : More Stars ???

Previous · 1 · 2 · 3 · 4 · 5 · Next
Author Message
zombie67 [MM]
           
Avatar
Send message
Joined: 30 Mar 10
Posts: 146
Credit: 6,953,555
RAC: 2,301
Total hours: 22,213,806
Message 1427 - Posted: 26 Jul 2013, 21:17:31 UTC - in response to Message 1423.

I won't add more stars. I will just increase Hours amount.
Probably, 10,000 - 25,000 - 50,000 and 100,000 hours.


Does this mean you are adding new colors for those lager amounts? Or does it mean you are reclassifying the current colors?
____________
Reno, NV
Team: SETI.USA

LanDroid
   
Send message
Joined: 30 Mar 13
Posts: 2
Credit: 98,786
RAC: 46
Total hours: 355,600
Message 1428 - Posted: 26 Jul 2013, 21:21:48 UTC

Suggest you include a level for 8760 hours = one year in each project.

Profile Conan
       
Avatar
Send message
Joined: 28 Mar 10
Posts: 399
Credit: 802,758
RAC: 58
Total hours: 1,773,603
Message 1429 - Posted: 27 Jul 2013, 2:06:21 UTC - in response to Message 1345.
Last modified: 27 Jul 2013, 2:07:05 UTC

I'm searching how I could add new badges.
A solution could be to increase badges size, so I could add more stars.
For example
:
20 apps || 40 apps || 60 apps || 80 apps || 100 apps | 120 apps | 140 apps | 160 apps | 180 apps



I could also increase credits amounts like this:

100 hours
250 hours
500 hours
1,000 hours
2,500 hours
5,000 hours
10,000 hours
25,000 hours
50,000 hours
100,000 hours
250,000 hours


G'Day WildWildWest,

I still like the idea you proposed above of adding more stars for the extra applications that we are doing.
Even little ol' me has more than 120 applications now and just another 4 applications to get past 100 hours to get my 5th star.
Yes it will take me a much longer time to get to all Sapphire but as Zombie has said he has enough applications and time for another 2 stars already (I think STEVE has as well).
Only thing with your suggestion maybe the colours for the higher point levels.

Conan
____________

Profile skgiven
       
Avatar
Send message
Joined: 7 Sep 10
Posts: 453
Credit: 935,250
RAC: 24
Total hours: 2,073,336
Message 1432 - Posted: 27 Jul 2013, 11:44:00 UTC - in response to Message 1429.
Last modified: 27 Jul 2013, 11:48:06 UTC

Instead of just basing stars on hours, you could base the levels on significant and progressive time periods (days, weeks, months...)

24h (day) - 20days total (20apps each for a day)
168h (week)
728h (month averaged)
2,912h (quarter/season)
5,824h (half a year)
8,736h (year)
17,520h (2years)
35,040h (4years)
87,600h (10years - decade)
219,000h (25years - quarter century)
438,000h (50years - half a century) - 1000years total (20apps for 50years)

This would make it easier to get the first two badges but harder after that. Both good things in my opinion; easy will attract people, harder will keep them attached to wuprop.
I don't really think that we need to go past 50years*20apps for 1000years, but I do think it's a long term realistic target for many people - even if they just wish to get one such star.

While moving the goalposts mid-game isn't usually a good idea, this is just a bit of fun, and it's only a few months in. The half year mark also roughly ties in with the present top star, 5000h, so the goalposts wouldn't move too much.

After you introduce higher badges you may want to reconsider adding more stars...

- Just a suggestion of course.
____________
=sticky tape=

Bok
         
Send message
Joined: 28 Mar 10
Posts: 89
Credit: 584,942
RAC: 201
Total hours: 2,124,656
Message 1433 - Posted: 27 Jul 2013, 13:16:20 UTC

I like skgiven's idea personally and it's not too far off the current levels.

I think I would try to keep the current levels at or below what they currently are though, so maybe just slide everything down one by making the bronze star fairly notional as 20 apps for 1hr

Think his math may be a little off though in places.. 5824 is NOT half of 8736 :)

If we take a year as 365days, that's 8760, and 1/2 is 4380, so if that was no sapphire, Steve would still have all sapphire (at least).

Don't think it's ever a good PR idea to take anything away after the fact..
____________

Profile STE\/E
         
Avatar
Send message
Joined: 28 Mar 10
Posts: 452
Credit: 3,398,693
RAC: 323
Total hours: 17,560,249
Message 1434 - Posted: 27 Jul 2013, 14:11:08 UTC
Last modified: 27 Jul 2013, 14:23:11 UTC

Just make up my mind alright already ... ... Personally I don't like skgivin's Idea, I'm simply not going to run a bunch of piss a** poor paying Projects if they even pay any Credit at all for 2-5-10+ Years Time, in the 100 App's for my 5 Sapphire Star's I had probably 20-25 Projects that Paid me nothing for running them for 5000 Hr's. If it comes down to that I'll just run the other Badge Projects & not worry about WUProp Hr's.

Your have to make it something attainable for today's Hardware & not something Impossible to reach in a Lifetime ... IMO ... It took me 6 Month's to get the 5 Sapphire Stars, I can't even Imagine what it would take to get what skgiven's is Proposing, I won't live long enough to even get 20 App's up to the Higher Levels ...
____________
https://stats.free-dc.org/badgesbanner.php?cpid=13a87c3a303bcdca4ba0ed600daebb6b

Profile skgiven
       
Avatar
Send message
Joined: 7 Sep 10
Posts: 453
Credit: 935,250
RAC: 24
Total hours: 2,073,336
Message 1435 - Posted: 27 Jul 2013, 15:04:09 UTC - in response to Message 1432.
Last modified: 27 Jul 2013, 15:04:41 UTC

My effort at counting corrected,

24h (day) - 20days total (20apps each for a day)
168h (week)
730h (month averaged)
2,190h (quarter/season)
4,380h (half a year)
8,760h (year)
17,520h (2years)
35,040h (4years)
87,600h (10years - decade)
219,000h (25years - quarter century)
438,000h (50years - half a century) - 1000years total (20apps for 50years)

Adding additional colored stars won't help anyone with more than 5stars, but if we are not going to get more badges/stars I think the system needs to be a bit more difficult.
The above system is better aligned with some other badge systems, but it's just an idea.

STE\/E, you're right, it would take you ~8years just to get one top star (438,000h*20apps).

My suggestion could easily be changed so the badges are closer together,

24h (day) - 20days total (20apps each for a day)
168h (week)
730h (month averaged)
2,190h (quarter/season)
4,380h (half a year)
8,760h (1year)
17,520h (2years)
26,280h (3years)
35,040h (4years)
43,800h (5years)
Maybe do as Bok suggested or just drop the last colored star,
87,600h(10years - decade)

5 stars for 20*5years would take STE\/E ~4years, or another 3.5 from now.

Profile STE\/E
         
Avatar
Send message
Joined: 28 Mar 10
Posts: 452
Credit: 3,398,693
RAC: 323
Total hours: 17,560,249
Message 1438 - Posted: 27 Jul 2013, 22:26:53 UTC - in response to Message 1435.

STE\/E, you're right, it would take you ~8years just to get one top star (438,000h*20apps).


Wouldn't it be Double that for 20 App's ??? 10 App's would be 438,000 Hr's, so 16 Years to get there but more than likely 16+ Years ...
____________
https://stats.free-dc.org/badgesbanner.php?cpid=13a87c3a303bcdca4ba0ed600daebb6b

Profile Conan
       
Avatar
Send message
Joined: 28 Mar 10
Posts: 399
Credit: 802,758
RAC: 58
Total hours: 1,773,603
Message 1439 - Posted: 28 Jul 2013, 6:04:03 UTC
Last modified: 28 Jul 2013, 6:06:03 UTC

I believe that WildWildWest had a good proposal with his extra stars and extra hour levels.
However I would probably drop the 100,000 and 250,000 hours stars as just too hard, and keep from 50,000 down.

Getting 50,000 hours for any application would be a major achievement, so getting it for 20 applications is a very major undertaking.

As STEVE said though a lot of projects pay bugger all for their applications so maybe drop the 50,000 as well and leave it at the current level of 5,000 plus 10,000 and 25,000 hours and stop there.

You get 25,000 hours up on 20+ applications and you are any projects dream volunteer.

Extra stars for extra applications I still think would be a good idea as there are well over 160 applications out there.
I have nearly gotten my 5th star with 100 applications, but I have run over 120 applications so far, it has all been done on CPUs, I have not run a single application involving GPUs.
Adding in the GPU applications and the Android applications that are out there and you can see that more stars will be needed.

So more stars (maybe up to 180 applications) and an extra two levels of 10,000 and 25,000 hours, will cover everyone for a long time.

The other thing you could do maybe is leave it at 5 stars and 100 applications, increase the levels a couple of rungs and then scan to see what are the most popular projects that people run.
But this stops the drive to run all available applications, then again do you really want to run them all? I don't know.

Conan
____________

Profile skgiven
       
Avatar
Send message
Joined: 7 Sep 10
Posts: 453
Credit: 935,250
RAC: 24
Total hours: 2,073,336
Message 1440 - Posted: 28 Jul 2013, 11:31:22 UTC - in response to Message 1439.

I'm a bit concerned that some people might want badges they can easily get. That's not what it's about. It's about setting targets big enough to accommodate everyone for the long run.
The existing badges are not challenging enough, otherwise some people wouldn't already have a full suite of 5 high value stars. So adding another two levels (10K and 25K) would be making the exact same mistake again.

If people have enough apps for 7 or more stars, they need to be accommodated (original 9star suggestion better than nothing), but that's a different issue from the star levels. The fact that people had 20*5000h inside a couple of months shows that the highest badge is far too easy.

Some people already have Very high numbers for some apps.
Below, Mumps [MM] has 7apps over 100,000h and one over 750,000h:

Project Application Running time (hours) Pending (hours)
NumberFields@home Get Decic Fields 768,465.70 623.10
sudoku sudoku@vtaiwan 473,647.85 24.12
NFS@Home 16e Lattice Sieve V5 333,260.82 3.58
NFS@Home 16e Lattice Sieve 208,755.75 1.52
Asteroids@home Period Search Application 182,939.27 0.00
boincsimap BOINCSIMAP simap application 120,028.05 0.33
Cosmology@Home CAMB 101,351.98 0.00
yoyo@home ecm 84,545.80 0.00
VGTU project@Home VGTU@Home application 63,187.42 0.00
physics@home Simple Cubic crystals cohrence sintering nano 53,299.08 0.00
World Community Grid The Clean Energy Project - Phase 2 50,990.65 0.00
correlizer Correlizer Applications 45,474.20 0.00
CAS@home Tsinghua Nano Tech Research 42,569.57 0.00
AlmereGrid Boinc Grid BioMedial Genome Correlations for Research and Healt Care 33,544.78 0.00
wanless2 Random-base WEP Factorization 31,464.00 0.00
Poem@Home POEM++ OpenCL version 26,734.60 8.55
NRG FlexAID Docking 15,839.92 0.00
NFS@Home 14e Lattice Sieve 15,397.12 0.00
Neurona@Home Neurona@home 12,299.58 0.00
Einstein@Home Binary Radio Pulsar Search (Arecibo) 8,831.12 0.00

While wuprop does not reflect Mumps contribution particularly well, he can clearly put in the cc_config and app_info hours so I doubt he would struggle to reach a 50K star.
The above 2.6Million hours evenly spread over 20apps is ~133,000h per app. In theory he could already have the second highest proposed badge, and probably hasn't been trying to spread resources since reaching the top badge.
Mumps may never reach 5stars, but if not it's because he focuses on a few projects, and makes large contributions towards them (that's his choice). If he reaches say 3 top badges but doesn't want to crunch for any other project perhaps he and others will stop running WUProp.

As well as facilitating those who wish to run hundreds of apps, the badge system needs to accommodate those who wish to concentrate on a few projects, and do a lot of work for them. Some people don't even want to crunch 5 projects never mind 20, or 100. At present WUProp's badge system does nothing for single project crunchers. Even large projects which have multiple apps probably don't have 20apps, and some crunchers just crunch one sub-project. I know some people who for personal reasons just crunch towards cancer research projects. It's a shame their contribution is not reflected.

I'm not one of the really big cruncher at WCG but I have still accumulated ~110years in total (almost 1000,000h), 13 projects over 2years (17,520h), 4 projects are over 10years (87,600h) and 2 are over 20years (175,200h). When HFCC restarts I intend to push on from my present 215,400h until the project stops. You might want to note that they use a badge system based on years, and people target that - crunch to reach a certain badge (albeit only up to 2years). 100,000h is meaningless to me as it's 11.4155251141. years.

Bottom line, whatever update is applied, the badge system needs to work for everyone, for the next 6years+ rather than 6months, it should accommodate those who concentrate on a few projects (and 20apps is quite a lot), and being retroactive is just looking over your shoulder. At this stage I think whole of the original idea or something very close to it might be the best solution for now. While I like my own ideas, and some other suggestions, I hope everyone lays off the flack when www implements his changes in original or modified form...

Good luck,
____________
=sticky tape=

Profile STE\/E
         
Avatar
Send message
Joined: 28 Mar 10
Posts: 452
Credit: 3,398,693
RAC: 323
Total hours: 17,560,249
Message 1441 - Posted: 28 Jul 2013, 12:15:41 UTC
Last modified: 28 Jul 2013, 12:20:34 UTC

I don't really care what the Levels are for the Badges, I'll most likely just work on 20 App's to get them as high as I can & to hell with the rest. The trouble with the Badge Projects is you reach the Goal Line & then they move it another 100 Yards away so it's a never ending task.

It took me about 6 Month's to get the 5 Sapphire Stars, so a 25K Level would take another 2 Years Minimum I figure, a 50k Level would be 4.5 to 5 Years away. That would be for me if I would choose to run that long.

For others with less Computers it would be a lot longer, for someone like Mumps yes it would be a lot shorter. But you have to think of Mumps as an Aberration lol ;) as he has access to many Computers. You can't set the Levels so High because of 1 Person, if you do then a lot of People will just get the Bottom few Star's and move on knowing they never have a chance for the next Level Star whatever that may be for that person. If you look at the Statistics there are 2 other People getting close to the 5 Sapphire Level, nobody else is really that close so it's not like everybody is going to get to that Level any time soon.

I don't want the Badge Levels to Low either, but I don't want them so high that their impossible to get except for someone like Mumps. I think a 10k 25k 50k either a 75k and or 100k level would suffice for now. That would keep most people busy for quit awhile. Even Mumps would have to work on getting to the 75k - 100k Levels for 100 App's. He could do it but it would take him awhile to do it so those levels should work for now & for the next 5-10 Years. The Project can always put in higher Levels as needed after that ... IMO
____________
https://stats.free-dc.org/badgesbanner.php?cpid=13a87c3a303bcdca4ba0ed600daebb6b

Profile Coleslaw
       
Avatar
Send message
Joined: 11 Apr 10
Posts: 179
Credit: 5,924,481
RAC: 3,280
Total hours: 11,595,889
Message 1446 - Posted: 28 Jul 2013, 20:18:32 UTC
Last modified: 28 Jul 2013, 20:19:15 UTC

You can however plan for multi-core/thread devices coming out each year that will provide people multitudes more in run time hours. So, basing the badges merely on todays resources would be rather stupid. WCG badges have been too easy to obtain for most crunchers for a few years. The biggest complaint has been lack of support for the large contributors. Why repeat that here? If you don't want to go after larger badges due to your quest for point based stats/rankings, then that is fine. Just don't suggest bad ideas because you like having the top award. This isn't hate posting, but more of a logical viewpoint with everyone in mind. I don't see a bunch of single threaded PC/single PC users complaining about how tough it is to get one badge here. The biggest complaint has been their own preference on what apps are valuable to run. By having more stars, that would reflect people like me who like to spread out their resources amongst many projects rather then just push for points so I can brag about being in the top X in the world.

Intel expects cell phones to have 48 cores within the next 5-10 years. Since projects support Android now, it would be safe to say that racking up runtime will only continue to become simpler.
____________

Sellyme
 
Send message
Joined: 21 Jul 13
Posts: 55
Credit: 163,067
RAC: 0
Total hours: 324,976
Message 1447 - Posted: 28 Jul 2013, 21:05:06 UTC
Last modified: 28 Jul 2013, 21:14:28 UTC

I figured I'd chime in on my suggestions for badges. I have a massive interest in statistics and data, and have devised many systems like this for various software, games, and competitions in the past, so I figured I might as well see what I can do with this.

For the record, in this post I assume "month" to mean "30 days", and "year" to mean "365 days". I do not account for February, leap years, or any other silly astonomical phenomenon, as in the context of computing, it hardly really matters.

(Also, 30 days is a much rounder figure for a month than the actual average of 30 days and 10 hours)

Current levels:
100 hours (4 days, 4 hours)
250 hours (1 week, 3 days, 10 hours)
500 hours (2 weeks, 6 days, 20 hours)
1,000 hours (1 month, 11 days, 16 hours)
2,500 hours (3 months, 14 days, 4 hours)
5,000 hours (6 months, 28 days, 8 hours)

1 star: 20 apps
2 stars: 40 apps
3 stars: 60 apps
4 stars: 80 apps
5 stars: 100 apps

Okay, now here's my suggestion:

Introduction of a new, easy-to-get badge.

I like skgiven's idea of making the badges reflect nice rounded time periods like a month as opposed to 1,000 hours, so I think we should have a badge for 1 day of computation. However, I don't think that the bronze star should be moved from 4 days 4 hours to 1 day, that's way too large of a change. As such, I propose this badge for anyone who has 20 apps exceeding 1 day (24 hours) of computation:



Excuse the horrible 5-second Paint.NET job, but basically, a badge with no stars. This won't "upgrade" along 40/60/80/100 apps like the stars, it's basically just something for new people to latch on to instead of having to wait nearly a month running 24/7 on a quad-core CPU with optimal application management to get the firstbit of recognition. With a quad-core CPU, this badge would take about a week to achieve.

A new star

I agree entirely with Roadrunner on that forcing people to run alpha/unstable projects is a bad idea. Extending to 180 apps is, at this point, definitely not a good thing to do. When there actually are 180 stable applications, then sure, but in the meantime it's best to expand the time scale before the applications scale. That being said, there are still enough stable applications to include a new star for 120, so I don't see why we shouldn't do that. A number of users would already have this badge, zombie67 [MM] would already have it as Ruby! It's thus clear that it wouldn't force people to run projects that are unstable or things they just don't want to run, as people are already running them anyway, and receiving no recognition for it.

As for an extention to a 7th star, I wouldn't be too upset about that, but that's getting into murky territory with alpha projects, so I'd recommend not doing it at this point. For the record, zombie67 [MM] would have a silver 7th star, and would only be 41 hours on a single Einstein@Home app away from gold. As far as I'm aware, no user would have an eighth star in even BRONZE, let alone Sapphire, so that's definitely a bad idea.

So, in summary, a 6th star will reward a number of users who are doing what's necessary for it already, a 7th star will only be applicable to 3 or 4 users and would be really getting close to being just plain annoying to get, and anything beyond that is stupid (for now).

Re-defining the hours requirements

Earlier I mentioned that I liked skgiven's "round" numbers. Here's my adjustment of his figures:

No star, but badge: 1 day (24 hours)
84 hours (half a week, 3.5 days)
168 hours (1 week)
504 hours (3 weeks) or 336 hours (2 weeks, half a month)
720 hours (1 month)
2,160 hours (3 months, one season)
4,320 hours (6 months, half a year)

With the exception of the 504 hours option for the gold badge, they would all be slightly lower than the current requirements, so no-one would lose a badge, addressing Bok's concerns.

Personally I prefer the 336 hours figure for the gold star over 504. Whilst 504 is a lot closer (only being four hours above, way too little to be bothered complaining about a lost badge), and 336 hours is nearly an entire week shorter, the progression of 1 week > 2 weeks > 1 months is a lot smoother and more aesthetically pleasing. Having going from a gold to ruby star being easier than from silver to gold seems a bit silly.

Extending the stars

For the main piece, extended stars! Again, I'm going for slight modifications of skgiven's figures.

8,760 hours (1 year)
17,520 hours (2 years)
43,800 hours (5 years
87,600 hours (10 years, a decade)
219,000 hours (25 years)

I agree with Ste\/E's concerns about latter badges being impossible to get, again using zombie67's stats, he would only have the first badge in that extended series, and only just, with his 20th application at 8,833.57 hours. I don't know Ste\/e's stats (as much as I'd love to), as he hasn't made them public, but I assume he wouldn't be much higher up. As such, I would completely support the 25 year star not being included. Not only is it at a level that not even the highest-ranking contributors on WUProp would achieve (even for just 20 apps) for nearly six years, but it's also kind of ugly.

It's predecessor, the decade-long Sapphire-on-Ruby badge, might also be dropped: a quick estimate says that it will take Ste\/e around 5 years to get that, and he has already expressed distate with going for the plain Sapphire for the 5th star. I doubt that anyone else would ever be able to get that for any more than maybe the first two, or in zombie67's case, maybe the third, too.

So where does this leave us? These are my suggestions.

New badge, no stars, for 20 apps at 24 hours
6 stars, 20/40/60/80/100/120 apps
Hours adjusted to represent real figures (days/months/years, etc) instead of hours
All adjustments made so no-one loses their badge.
3-5 new colours of star added, continuing the already-existing progression.
Realistic goals, but still very difficult ones. Nothing impossible.

Thoughts?

EDIT: Whoops, Coleslaw posted as I was writing this!

Coleslaw, I don't think that multi-core devices are going to make much of a difference; yes, in 5-10 years the production may double, but even then, we have to wait a decade for that to start to take effect. WUProp currently has no Android app, so we can completely ignore mobile devices at the moment, and desktop machines simply aren't getting any higher number of cores/threads at a rate big enough to make a true impact on what is and isn't actually possible to achieve.

Regarding "Well you don't have to get the badge!", no. That is an absolutely horrid point. I understand the reasoning behind it, but it's just wrong. An old friend of mine, Greg McClanahan, was (and still is, as far as I know) the rewards system manager for Kongregate, one of the largest gaming sites on the internet. He has created thousands of badges on thousands of different games, and he has always refuted the "Well you don't have to get them!" mentality. If achieving something isn't just difficult, but also annoying and painful, and you're rewarding that, you're basically just saying that you want to see people suffer. The entire point of badge systems in distributed computing is to make people more likely to continue helping your project out. If you make the user experience worse with badges, then you're achieving the exact opposite of what you're setting out to do.

Regarding how hard it is for light-weights like myself to get badges, I don't complain about it because it's difficult, not annoying. If I had to run projects that crashed 9 times out of 10 to get some badges, then I'd be complaining about it, and that's exactly what Ste\/e is doing.

As for having more stars for people like you, you wouldn't have any more stars, so including just a single extra star is easily enough of an increase in the badges to keep you occupied for a couple of years.

Profile STE\/E
         
Avatar
Send message
Joined: 28 Mar 10
Posts: 452
Credit: 3,398,693
RAC: 323
Total hours: 17,560,249
Message 1449 - Posted: 28 Jul 2013, 22:18:42 UTC - in response to Message 1446.
Last modified: 28 Jul 2013, 22:20:17 UTC

By having more stars, that would reflect people like me who like to spread out their resources amongst many projects rather then just push for points so I can brag about being in the top X in the world.


I didn't know I was Bragging, I can't seem to Post anything in any Forum without someone suggesting I'm either Bragging or Complaining ... Bye

Sellyme
 
Send message
Joined: 21 Jul 13
Posts: 55
Credit: 163,067
RAC: 0
Total hours: 324,976
Message 1450 - Posted: 28 Jul 2013, 22:25:05 UTC - in response to Message 1449.

Considering that the alternative to bragging was "to spread out [your] resources amongst many projects" I don't think he was saying you were bragging. I figure he was more talking about people who exclusively run a single project that gives the most points.

Profile Conan
       
Avatar
Send message
Joined: 28 Mar 10
Posts: 399
Credit: 802,758
RAC: 58
Total hours: 1,773,603
Message 1452 - Posted: 29 Jul 2013, 12:18:32 UTC - in response to Message 1449.
Last modified: 29 Jul 2013, 12:19:42 UTC

By having more stars, that would reflect people like me who like to spread out their resources amongst many projects rather then just push for points so I can brag about being in the top X in the world.


I didn't know I was Bragging, I can't seem to Post anything in any Forum without someone suggesting I'm either Bragging or Complaining ... Bye


I wouldn't say you were bragging either, more like "Phew, after a bloody lot of hard work, I have made it".

Just support from me, mate.

Conan
____________

Profile Coleslaw
       
Avatar
Send message
Joined: 11 Apr 10
Posts: 179
Credit: 5,924,481
RAC: 3,280
Total hours: 11,595,889
Message 1457 - Posted: 29 Jul 2013, 20:37:11 UTC
Last modified: 29 Jul 2013, 20:39:45 UTC

No STEVE\/E I'm not saying you are bragging at all. We just have a difference in opinion for direction of badges.

Sellyme, what do you mean WUProp doesn't have an app for Android? I've been running it on ~4 phones for months.

http://wuprop.boinc-af.org/show_host_detail.php?hostid=46378
http://wuprop.boinc-af.org/show_host_detail.php?hostid=53828
http://wuprop.boinc-af.org/show_host_detail.php?hostid=49745
http://wuprop.boinc-af.org/show_host_detail.php?hostid=56540

They just don't have an x86 Android app out yet since there are no Android version of BOINC for x86 just yet. However, it should be available in the future according to Rom. http://boinc.berkeley.edu/dev/forum_thread.php?id=8518
____________

Sellyme
 
Send message
Joined: 21 Jul 13
Posts: 55
Credit: 163,067
RAC: 0
Total hours: 324,976
Message 1459 - Posted: 29 Jul 2013, 21:11:10 UTC - in response to Message 1457.


Sellyme, what do you mean WUProp doesn't have an app for Android? I've been running it on ~4 phones for months.


Via NativeBOINC, I assume? Because that has been around for far longer than WUProp has had badges, so continuous advances in phone hardware would already be showing effect on the ease of badge achieval, and as such a simple consistent linear rise in hours is to be expected (and is exactly what I factored for). Phones are not going to go from 4 cores to 48 overnight.

And besides, even if they somehow did, the badges can just be extended again. As long as you leave room for expansion (like I did in my suggestion; with additional stars, and at least two additional star colours as options), then future-proofing doesn't really need to be done, it's not like badges can't just be added on again in a couple of years.

Profile Coleslaw
       
Avatar
Send message
Joined: 11 Apr 10
Posts: 179
Credit: 5,924,481
RAC: 3,280
Total hours: 11,595,889
Message 1461 - Posted: 30 Jul 2013, 3:12:23 UTC

You can also run WUProp with Berkeley's version of BOINC. So, you are right that I was using nativeBOINC. My 3 older phones have Android 2.2.2, so can't run Berkeley's version.

You are right that the phones wont have 48 cores over night. That is why Intel was forecasting 5-10 years. Don't forget the various Raspberry Pi devices that are also now capable of Android. WCG alone already has over 6000 of the ARM processor devices attaching since they released their app. You will find that many of the devices being added are quad core these days. Look through the stats on nativeBOINC.

So, it really isn't that hard to add a lot of run time. You just have to be committed to doing it.
____________

Sellyme
 
Send message
Joined: 21 Jul 13
Posts: 55
Credit: 163,067
RAC: 0
Total hours: 324,976
Message 1463 - Posted: 30 Jul 2013, 9:51:32 UTC - in response to Message 1461.


So, it really isn't that hard to add a lot of run time. You just have to be committed to doing it.


The thing is that 5 years for 120 different applications (which is what I recommended as the new top-limit, with an easy extension up to 25 years for 140 applications if/when it's necessary) is slightly more than "a lot" of run time. Admittedly, I don't have the thousands of devices that some of you guys have, but I really doubt any of you would rack up 600 years of run time any week soon, let alone the 3,500+ years necessary for the extensions.

Previous · 1 · 2 · 3 · 4 · 5 · Next
Post to thread

Message boards : Number crunching : More Stars ???


Home | My Account | Message Boards | Results


Copyright © 2019 Sebastien