Did you now??? That SETI-BOINC does not have extensive stats like SETI-classic, but that these stats are maintained by 3th party sites which can be found here.

BOINCstats for example offers details about your efforts and you can also compare your crunching power with other users that registered on the same day, have the same number of computers running, live in the same country, are a member of the same team and much more.

---

Join team BOINCstats

Did you know?

Projects don't always run smoothly, in fact some can be very unreliable. Should you want to ensure that you keep crunching numbers even when your favourite project is down, attach to another project or even a few projects and set their resource shares to 1 while putting the primary project resource share to 10000.

Did you also know?

With regards to projects being inaccessible, you can actually see whether the project can be accessed by an independant source. Point your web-browser to http://boincprojectstatus.ath.cx and look at the information for the project of your choosing.

---

The default values in the preferences were chosen for a reason. If it is at all possible for you to run with those default values, at least while getting started with BOINC, then you should avoid making any changes. When you _do_ make changes, it is a good idea to make them slowly - if you think you want a ten-day cache of work, for example (a really bad idea, IMHO...) then you might go from the default 0.1 to 0.5 for a day, then to 1, then to 2, then 4, then 6, then 8, then finally (if you haven't thought better of it by then) to 10.

Likewise on disk space - BOINC will not take up gigabytes of your space, but telling it that it can only have a few megabytes will probably mean that you don't get any work. It needs space _temporarily_ for things like downloading the science applications, and if you don't let it use that space, you won't be doing anything.

Oh - and don't hide your computers. There's no information available to others that would let them break in, but if you hide them, there's an awful lot of information we _won't_ be able to get when we are trying to help.

If in doubt about anything you see happening - read the Wiki _first_. Then see if anyone else here on the boards is having the same problem. (The project may be down for the normal Wednesday backup or something.) Then ask for help, and the volunteers here will do anything they can for you. Screaming and moaning about how awful everything is (when 99% of the time it's something YOU have done... like setting a 10-day cache...) and how "BOINC is broke", without giving us any helpful information, will just get you ignored. It really does work. It's different - it's complicated - and yes, the "official" documentation is, um, limited. But if you give it a chance and use it as it was designed to work... it works.

Credits don't show up instantly when the work unit is finished on your computer. There are many steps involved. First, the result data file is Uploaded. Then (at a LATER TIME) the fact that your computer has finished that result is Reported. Only when three people have reported a successful completion of a result is the Validator run for that Work Unit - until then it is "Pending". If two out of the three results match "strongly", then credit is (finally!) awarded. The highest and lowest claims are dropped, and the middle (or average of the remaining, if four or more people happen to have reported in before the validator runs) is awarded to all results that match at least "weakly". (If others come along later, and match, they'll get the same amount of credit as well.) Every WU at SETI is sent out to four computers initially. It may be sent out to others if those four have problems, miss deadlines, etc.

If you get zero credits for a result, but other people with the same WU have gotten credit, you need to first look at the status. Was it a computing or download/upload error? Even if it shows "success", you can then look at the text reported with the result itself. It may show "invalid", meaning your computer thought it was fine, but the validator says it was so far off the others that something is messed up. Or, it may actually tell you exactly what went wrong! All of us get occasional zero-credit results. Usually because something messed up at our end, our computer crashed, whatever. If you get several in a short period of time, that's when you need to worry.

The project web site has tons of information available on your work, that can help trace any problems. You may have to look at several pages to find it all. The project does NOT store all of this forever; after a couple of weeks, your results will start to be deleted from the web site. The credits remain, and the actual data you produced is stored in the Science Database, but there just isn't enough disk space to continue to show all the data on every result for years. If for whatever reason you have to know how many results you've done, and when, and so on, it's up to you to track that. There are 3rd-party programs available to do so, but you'll have to be running them _before_ the data goes away...

RAC (Recent Average Credit) is almost meaningless. A 10 or 20% change could just be a timing issue in when results were reported. About all it is good for is a general guideline on how much you're doing - a RAC of 1000 obviously means you're turning in more results per day than someone with a RAC of 50. If your RAC settles down after a couple of weeks of running to "somewhere between 175 and 225", and then one day you notice that it's at 30 - worry. But if one day it's at 200 and the next day it's at 175... chances are the day after that, it'll be at 225 for a while.

"RAC is the speedometer, Credit is the odometer" will be heard a lot. True... but this speedometer is like the one on my first bicycle - all over the place with every pump on the pedal. And it won't start to work even THAT well until your results are being validated at about the same speed you're turning them in, or in other words, a couple of weeks at least.

EDIT:: spelling, etc.

Open your mind, open your ears, ask questions, take part in the debate.

You are unlikely to get a helpful answer if your position is merely "Its different and therefore useless and I can't run it".

In the majority of cases you can, it just needs thought. Thousands here have already made the transition, and so can you.

---

All regular users, please post anything you think might speed the new users into a fuller understanding.

thanks
tony

The most important thig here @BOINC is:
DONT PANIC

---

Scorpions - Wind Of Change

Problems running Boinc?

Did you know there is a goldmine of information inside the tab "Messages" of Boinc Manager.

If you go there and want to post relevant messages, left click on one message and adding other messages to the selection by using "Ctrl-left mouse click", use the button "Copy selected messages" and then in your post use "Ctrl-V" to paste these messages. This will help us to resolve your problem more quickly.

Want BOINC to crunch a little faster? If your computer is one you leave and come back to or use for many things other than net surfing, disable the Internet connection and activate the BOINC Manager's 'Disable BOINC Network Access' option. When you return to your computer or need the network again, restart your computer's Internet connection and uncheck the disable option in the manager, and let BOINC contact anything it needs to. It might not immediately seem like much but over time more work will be done. The same applies for any program (especially under Windows) that is running in the background that you might not need. Less running programs mean more CPU computational power for BOINC.

(Tony, good thread.)

---

There is a search, for these forums. It is just one level up, from where one might expect it to be.

If you click on "Message boards" above you will go to:

http://setiathome.berkeley.edu/forum_index.php

which has a spot to search.

---

Did you know that under "your account" there are "forum preferences". Under "Forum Preferences" there is a "filter" section that you can adjust and a box so you may enter the "user id" of a user and have all the posts from that user removed from your view (or kids view).

At the bottom of every post there is a "rate this post" + - boxes. When 5 people rate a post with the +, then a Green Exclaimation point appears in the post and in the whole thread. When 25 users rate a post with the - selection, it will automatically be filtered out of your view. The 5 and 25 are the defaults, but can be changed by you in forum preferences.


[edit]to find out the "user ID" of any user, just point your cursor over a user name, and look at the bottom bar of your browser. [/edit]

[second edit]I guess I'm learning to. When I first used an Overclockix (linux/knoppix) CD I noticed that when I used "mozilla/firefox" browser that there is a text link near the rating boxes that allowed users to report a post as "Offensive". I had never seen one with MSIE (microsoft internet explorer), but after writing this post, I noticed that MSIE now shows a X in a red box next to the ratings boxes, and by pointing the cursor over the X it shows it to be a "report this [edit]post[/edit] as offensive" link.

[edit]to find out the "user ID" of any user, just point your cursor over a user name, and look at the bottom bar of your browser. [/edit]

Or, look under the name or avatar of the user, for the line that starts "ID:"

MJ

---

Dial up users,
did you know that the more posts and graphic in a thread the longer it takes to load. In the upper [edit]right[/edit] corner you can change your setting to "most recent posts first". Then when you select a thread and it starts loading that you can click "stop" on your browser and it will display the newer posts and not finish downloading all the old ones. This tip may improve you fora participation experience.

Title = Hints, cheats, and advice for New users.

I looked through this entire thread and have not found one way to cheat!

NO Cheats!!

Wonderful concept!

---

Boinc....Boinc....Boinc....Boinc....

Once a work unit has been validated (for projects that use validation) and the results recorded in the database, old work units are deleted. For most projects, results for individual work units are available for a couple weeks or so after results are recorded. Older results are deleted.

BUT--only the old work units are deleted. The credits for crunching stay with your account.

edit: minor wording tweak, spelling goof fixed.

---

If you think education is expensive, try ignorance.

I hope everyone will add to this to improve the efficiency of these boards. these boards have moderattors, but it's best if we as a group set the rules and live by them.

Some suggestions. They're my personal opinion; if you disagree, you're certainly entitled but I'm not interested in debating them. They're not official rules or even official suggestions--just an opinion from one member, take 'em for what they're worth.

1) Before posting a question, take a quick look through previous threads to see if someone else is dealing with the same thing. (You don't have to read every post from the beginning of time, but scrolling down the page may be helpful.) Likewise, the Wiki is your friend.

. . .

OK, that's enough nanny behavior out of me (see rule 2 above). ;-)

Cheers,

Brian

Also, you can search the posts in the fora for terms that relate to your question. There is a good chance someone is discussing what you are wondering about already.

Try to keep an eye on your computer list for any new machines that pop up that are duplicates. They need to be merged before they get credit.

It happens pretty regularly after you remove BOINC and install a new version.

---

If you return units succesfully, but they are not validated and not granted credit, then it could be a stability problem on the host. Check that it is not overheating and/or overclocked. If your motherboard supplier did not provide a temerature monitoring utility, then there are several free ones available, such as;

• PCWizard2005
  
• Everest Home edition
  

which give you lots of info about your computer.

To test for stability, other than using Seti etc, there are programs like;

• Prime95 Torture test
  
• SuperPI, do 4M minimum
  
• Memtest86, boots from floopy or cd and takes time, over 30mins on P4 2.53.
  

if any of these programs fail to run then you can be sure that the BOINC crunching that you do will produce no good scientific results and probably more importantly for you no credits.

---

Did You Know;

That this thread is pointless...

...and this post will vanish when your American censors read it. God Bless America, land of freedom and a lesson to us all.

---

...and this post will vanish when your American censors read it. God Bless America, land of freedom and a lesson to us all.

Actually I'm from Denmark =)