Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
World Community Grid Forums
Category: Completed Research Forum: OpenZika Thread: OpenZika - Crunching Chart |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 74
|
Author |
|
dango
Senior Cruncher Joined: Jul 27, 2009 Post Count: 307 Status: Offline Project Badges: |
Now we're down at 1% according to Research status. More batches generated/predicted ? ooo nice to see it on 1% :) also CEP2 went down from 99% |
||
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges: |
CEP2 is down, 'cause of the re-release of science after successful BETAs!
----------------------------------------OZ...well, techs, admins: can we have a total batch number? is it still 19026 or has it changed? ;) |
||
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
What part of 'initial' batch set do you not understand, click? How can the scientists possibly tell when 'how to progress' experiment input is based on previous experiment output, barely 2 weeks into the project? So, a number was plugged to go to an 'maybe' completion time, year(s) in the future, just as I've done in this threads OP and other crunching charts OPs, and not the first time pointed out to you!
----------------------------------------[Edit 1 times, last edit by SekeRob* at Jun 1, 2016 8:10:26 AM] |
||
|
Mathilde2006
Senior Cruncher Germany Joined: Sep 30, 2006 Post Count: 269 Status: Offline Project Badges: |
a.) it's Klik- not click
----------------------------------------b.) calm down c.) he just was asking, what's behind the change from 17 to 2%. You must learn, to be friendly- even, when people ask same questions ever and ever again. |
||
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
'over and over' you might mean... eventually the friendly then says goodbye in less circumvent words ;o)
----------------------------------------On topic, new Result record yesterday... 483K. Fresh chart is up. [Edit 1 times, last edit by SekeRob* at Jun 1, 2016 10:59:55 AM] |
||
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges: |
What part of 'initial' batch set do you not understand, click? How can the scientists possibly tell when 'how to progress' experiment input is based on previous experiment output, barely 2 weeks into the project? So, a number was plugged to go to an 'maybe' completion time, year(s) in the future, just as I've done in this threads OP and other crunching charts OPs, and not the first time pointed out to you! well, I still trust your data charts more than their estimates...talked about that more often here on forum, not going to repeat it... even asked techs to change their way of calculation the ECD...which clearly doesn't work - apparently! even saying them how to do it & not look only last week of crunching as a based calc... remember when we had issues with OET & short WUs? btw, your post are saying that we passed 400K results per day & keeping it stronger & stronger...so why the % have dropped to 1% & now are on 2%?! so asking them about the total number of batches & if it has been changes is viable question for me? also, note that % have been going up from a day to a day...& now they dropped...so what's going on, again? btw, I'm thinking about making a 7 Android crunching machine, connected with this: - D-Link USB Hub DUB-H7, 7 ports - Asonic USB Hub N-UH3702, 7 ports so wandering is OZ is going to last, to be viable investment?! |
||
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3294 Status: Offline Project Badges: |
What part of 'initial' batch set do you not understand, click? How can the scientists possibly tell when 'how to progress' experiment input is based on previous experiment output, barely 2 weeks into the project? So, a number was plugged to go to an 'maybe' completion time, year(s) in the future, just as I've done in this threads OP and other crunching charts OPs, and not the first time pointed out to you! well, I still trust your data charts more than their estimates...talked about that more often here on forum, not going to repeat it... even asked techs to change their way of calculation the ECD...which clearly doesn't work - apparently! even saying them how to do it & not look only last week of crunching as a based calc... remember when we had issues with OET & short WUs? btw, your post are saying that we passed 400K results per day & keeping it stronger & stronger...so why the % have dropped to 1% & now are on 2%?! so asking them about the total number of batches & if it has been changes is viable question for me? also, note that % have been going up from a day to a day...& now they dropped...so what's going on, again? btw, I'm thinking about making a 7 Android crunching machine, connected with this: - D-Link USB Hub DUB-H7, 7 ports - Asonic USB Hub N-UH3702, 7 ports so wandering is OZ is going to last, to be viable investment?! It probably decreased to 2% because more work as added to the pipeline (doesn't necessarily mean more work beyond those 19026 batches). I think... AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W AMD Ryzen 7 7730U 8C/16T 3.0 GHz |
||
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Downloaded one of batch 7253 right at noon of 19026 makes it 38.1% of known [purely linear, when above 12013 the batches are 1/10th the size]. On a 2 year timely, 2% what it shows now [this morning still 1%, dango saw that too yesterday, after having fast-tracked towards 20%] is now exactly [very coincidentally] matching the WCGPI number. Guess the Research page only shows the integer part, so 2.9999% would still show as 2%... it would have to as else 99.5% would print as 100% on the timeline, but then we've seen that too whilst work was still flowing ;O].
----------------------------------------BTW, seem to recollect that the term 'loaded' was used, so we could be at 100% whilst there's weeks of work right before the feeder/transitioner gauge. :) (Reminds me of the Julie Christie/Donald Sutherland movie... 'Don't look now') [Edit 1 times, last edit by SekeRob* at Jun 1, 2016 1:45:20 PM] |
||
|
3A4scLiRhJVcdT2K9q9kQNxzxYJ9
Advanced Cruncher Joined: Nov 16, 2009 Post Count: 72 Status: Offline |
btw, I'm thinking about making a 7 Android crunching machine, connected with this: - D-Link USB Hub DUB-H7, 7 ports - Asonic USB Hub N-UH3702, 7 ports so wandering is OZ is going to last, to be viable investment?! KliK, I suggest, you don't(!!) go with this kind of device because the specs say: *** Supplied Charging Current Standard Mode: All seven USB ports- 0.5 A per port 2 Fast-Charge Mode: 5 standard USB ports- 0.5 A per port 2 2 USB fast charging ports - 1.2 A per port 2 *** This is not going to to be enough to keep more than two 4core Android device on fire (on the mentioned 1.2amp ports!). Of course, if you have old dual-core androids, this might be Ok... but I guess that's not going to be your purpose. Trust me Klik, you want to be going with something like this: Amazon.de -> B015HCLIZ6. Because this device offers 10x 2.4amp it have enough power to go with any common octo-core. Sadly it does not offer USB-Data... so you'll need to work with wifi... which does quite well for me. I ended up ordering this device: Amazon.UK -> B00MQZGOM0 The huge AC-DC converter does at the very most barely even get lukewarm, so this is to be recommended. (beware of the power plug edition according to your country) Any under-powered USB-hub will likely either not going to have all your Androids run at 100% or the converter might get really hot and die soon. (avoid fire hazard, get "serious stuff") As for your question about it being worth the invest... I'd say "yes". Because even if WCG at some point might not have any Android subproject going on (let's say, Zika and OET come to an end) there will still be other (whitelisted) projects from other Boinc-Projects to be crunched with/on Android. The real risk will be: How long will there be project support for the Android version that you will have installed... for now I would say, Android 4.4.2/.4 still works best. But devices now sell with 5.0/5.1 or 6.0, so if dev's at some point choose to switch to newer Android, you might end up dry. So far, Android development has been TOO fast for dev's to port crunching to any new Android edition, but with dropping Android 4.2/4.4 sales, I expect this to change at some point. Only future will tell. Do not invest much into your Android devices, if they go "obsolete" the "investment damage" will be limited... (but anyway, still way more efficient than most common x64 units... and way, way cheaper... depending on what you buy and where you buy..... ah yes, and *if* you go with smartphones/tablets, make sure batteries are easily replaceable... batteries die after 1yr of 24/7 connection) |
||
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
/tot
|
||
|
|