It would be nice if BOINC worked as easily as classic.

It would also be nice if we got credit for all the work carried out instead of getting bombed out after 14 days, which is ridiculously too short a time.

It would be nice if we got credited for work completed within that time for that matter.

Why do so many people have so many problems with BOINC, is it perhaps that the system hasn't been very well thought through?

---

System works just fine run 3 p4 3.0s doing Hyper threading, in it for the science not the credits......

---

Semper Eadem
So long Paul, it has been a hell of a ride.

Park your ego's, fire up the computers, Science YES, Credits No.

It depends on the person. I have been running Boinc since last June. I have had no issues, and I have run it on everything from P1's to P4's. I have upgraded serveral versions, all without problems.

My take on it is some people are afraid of change. Or they do not take the time to find the information they need before they take the plunge.

I can understand some people might have issues with the software, but they have softwares that aren't compatible. Or they use wacky firewalls they do not understand how to use, or other blockers.

I hope people finally figure out what is wrong with whatever they have done to their installs/computers/whatever and can run Boinc as easily as many of us have already done.

---

My movie https://vimeo.com/manage/videos/502242

It would be nice if BOINC worked as easily as classic.

I don't know how Classic worked, but I have few programs more intuitive than Boinc running on my puter.

It would also be nice if we got credit for all the work carried out instead of getting bombed out after 14 days, which is ridiculously too short a time.

We get credits for useful work, not any work. 14 days deadline for 5h WUs is no way to short.

It would be nice if we got credited for work completed within that time for that matter.

As above: Only useful work is credited. Invalid results will not be awarded. That's absolutely fine imho.

Why do so many people have so many problems with BOINC, is it perhaps that the system hasn't been very well thought through?

I don't know of a lot of problems with Boinc other then not working exactly as Classic did, and that was not the goal of the project.
I know of some inconveniences, lot's of wishes on this list about (imho) minor issues, and real problems with some of the beta projects on it's way to Boinc.
But the 4 projects in my banner run fine.

---

Gruesse vom Saenger

For questions about Boinc look in the BOINC-Wiki

<blockquote>It would be nice if BOINC worked as easily as classic.</blockquote>
I think it does. Some people just seem to find it overly complex for some reason. What's so difficult? You attach to a project and let it run. Pretty simple.

<blockquote>It would also be nice if we got credit for all the work carried out instead of getting bombed out after 14 days, which is ridiculously too short a time.</blockquote>
It would be nice if BOINC didn't allow users to set their cache to 10 days. Deadlines are never an issue if you set your 'connect to' preference to a sane level. A large cache may work well if you are only attached to one project, but it's insane for multiple projects. The newer development versions of the CC are attempting to rectify this situation, and only allow you to download enough work that you can comfortably meet the deadlines.

<blockquote>It would be nice if we got credited for work completed within that time for that matter.</blockquote>
Sure it would. But I think the concept of actually verifying whether or not the science your machine produced was actually viable to the project is a good thing. S@H classic only cared that you returned a WU. (Not to mention it recycled WUs to keep users in work. I don't care to be given redundant work simply to keep me busy)

<blockquote>Why do so many people have so many problems with BOINC, is it perhaps that the system hasn't been very well thought through?</blockquote>
I think the system is fairly well thought out. But as with anything new, there are sure to be problems in the real world that didn't crop up on paper. Nothing of this scope has been tried before, so it's impossible to foresee every issue that might come up when it actually goes into service. You can't possibly set up every system configuration in your lab, nor can you account for the level of every users knowledge. Something that seems common sense for one person, may be an abstract concept to another. It's a work in progress basically. :)

---

I have never said the system doesn't run. Clearly it does, but it is the effectiveness that is highly dubious, along with some of the ridiculous assumptions.

I don't have an installation problem, as that is a case of run the file. Similarly there is no problem on the pc, but it is the way the whole thing works in relation to the core servers that is the problem.

It may take 5 hours to complete a wu IF the PC is doing nothing else. If it is doing other work or only switched for a few hours per day then clearly the available cpu cycles are going to be limited. If there is an automatic assumption that 14 days is a fixed deadline and anything you do within that period is scrapped with no recognition then the basic design parameters for the system were very badly flawed.

Following the successful and timely completion of a wu, with 2 days spare, and a further 7 in its cache, it downloaded a further 40! Yet there is no way of deleting the ones that are going to be a waste of time and resources. And this is after over a year of the software being live! I manually deleted the old deadline files, yet the system decided to download replacements and retain the imminently out of date deadline.

The purpose of BOINC (and classic etc) is to make better use of spare processing capacity not waste it pointlessly.

BOINC would work much better if there was some thought into the how the whole thing works holistically and after all this time if someone actually fixed the illogical problems and ommissions.

This has nothing to do with not wanting to change or any other unhelpful comment, I installed the app 'out of the box'. I left all the settings 'as installed', yet I have seen a whole host of problems that are mirrored by hundreds if not thousands of other people, and after having considerable classic experience where all these issues complained about are easy to fix or resolve even if manually BOINC seems to be a hostage to fortune. It is pretty obvious that the complexity of BOINC (translating 27.36 credits to an average of 0.08 among other things) was done because it could be, and not because it needed to be. The system is now suffering because of this unnecessary and expensive complexity. The system holistically includes us, the client pcs, the networks, the internet, the core servers amojng other things and not just the pcs and servers. It really needs a re-think. Unfortunately such views are exacerbated by the project teams silence on the systemic issues, responses being left to well-intentioned volunteers. Instead of defending what has been done the effort would be better spent correcting the numerous reported problems.

Apart from anything else I have 5 almost out of date wus, that won't delete and a further 40 with a deadline of a further 13 days. I didn't set any cache the system did. I didn't download all these, the system did. If there was a delete or re-order option I would use it. At least with classic the wus were visible and could be reordered or deleted within the add-on programmes, they could be controlled and cache size set, extremely easily and simply. BOINC can do none of this, and it is supposedly an improvement! This 'improvement' seems to be on a par with Longhorns wonderful new Red Screen of Death.

---

It may take 5 hours to complete a wu IF the PC is doing nothing else. If it is doing other work or only switched for a few hours per day then clearly the available cpu cycles are going to be limited.

Given time Boinc will realize that time your PC is switched on and running and will download units accordingly.

Following the successful and timely completion of a wu, with 2 days spare, and a further 7 in its cache, it downloaded a further 40!

Go to your account web page, general preferences and set "Connect to network about every" to 1 or less. This will reduce the amount of wu downloaded.

Yet there is no way of deleting the ones that are going to be a waste of time and resources.

There is a way to abort wu, but is not yet implemented in Boinc Manager, but you can do it via a third party application that a lot of people use instead of BoincManger, which is BoincView that can be downloaded from here: http://boincview.amanheis.de/

---

There is a way to abort wu, but is not yet implemented in Boinc Manager, but you can do it via a third party application that a lot of people use instead of BoincManger, which is BoincView that can be downloaded from here: http://boincview.amanheis.de/

It is implemented in the versiion of BOINC that is being tested for release (currently 4.42).

---

BOINC WIKI

Hey, Cole!
You should read some stuff about BOINC before complaining too much. And try to understand what it really doesl. It is not SETI@home 2, it is BOINC. Try to get it.
And by the way, try to understand more than that:
Just an example:
The Team-Link in your own profile points to 'http://setiweb.ssl.berkeley.edu/create_account_form.php?teamid=3571'
And that's the page to create a new profile, good man!
Are the BOINC people responsible for that, too???

think before talk!

---

[b]phees solutions[b][br]
[img border="0" src="http://boinc.mundayweb.com/one/stats.php?userID=1410" /]
[img border="0" src="http://boinc.mundayweb.com/one/teamStats.php?userID=1410&amp;prj=1" /]

I sympathize with you, and I agree with most of what you are saying.

The BOINC core client is now at version 4.43 and it's the buggiest version yet. I am running BOINC on 7 computers. When I upgraded them, 2 of them lost all of their work and could not communicate with the BOINC servers. I had to detach and re-attach all of the projects on those two machines. That's certainly not helping the science at all.

I have been running BOINC for 10 months, and there has been over 30 versions of the core client in that time. Way too many versions. Way too many changes. Way too many problems.

Classic Seti coupled with SetiQueue did almost everything. The only things BOINC has allowed me to do is run more than one project, and to disable the screensaver. Both of these things are useful, but BOINC has been infinately more frustrating, technically, than Classic Seti + SetiQueue.

--- bt