StokeyBob 的帖子

101) 留言板 : Number crunching : Strange reset problem? (消息 251797)
发表于：22 Feb 2006 作者： Profile StokeyBob
Post:
I was thinking about the twin late results some.

One way I thought of that may make a bunch of machines lose the same work at the same time would be a couple of power failure's or drops in a row in a place like the San Francisco bay area here and you would see something like this.

I sort of think my work loss was by my wiring burning out. I've got a lot running on one circuit. I would think if it was truly overloaded I would see a breaker tripping though.
102) 留言板 : Number crunching : Strange problem with downloading too many work units (消息 251759)
发表于：22 Feb 2006 作者： Profile StokeyBob
Post:
Nice tip Lee Carre.

I've wanted go go back and read lost messages many times. I hope I can remember. I saved a copy of your message just in case.
103) 留言板 : Number crunching : Strange problem with downloading too many work units (消息 251697)
发表于：22 Feb 2006 作者： Profile StokeyBob
Post:
That's what will happen if you reset the project. If you want to reset the project on purpose you go to the projects tab in the BOINC manager, select the SETI@home "Project" tab and hit "Reset project".

It sounds like yours did it on its own.

Another way it could have happened is your machine crashed and rebooted. Sometimes that will corrupt one of the files and you will lose everything.

I've had it happen to my machines before. Don't worry about the work. It gets sent back out. The important thing is your machine is back up and running.

Welcome to the message boards Mikey.
104) 留言板 : Number crunching : Is there any way to calculate P90 years in Boinc? (消息 251692)
发表于：22 Feb 2006 作者： Profile StokeyBob
Post:
Here is a definition of sorts nearest I can figure;

Estimated Performance
GIMPS measures the rate at which a computer is able to determine the primality of a number in Pentium 90 CPU hours per day. The Pentium 90 was chosen as the benchmark machine and all statistics calculated are based on this standard. Based on this, a Pentium 60 would calculate at 66% the rate of that of a Pentium 90, and conversely a Pentium II 300 would calculate at 400% of the rate of a Pentium 90. Below is a table listing each of our labs, and the estimated rate at which they compute primality individually and as a whole

Plagiarized from here;

http://www.getty.net/gimps/
105) 留言板 : Number crunching : Slower Crunching Since 'Doze Updates? (消息 251680)
发表于：22 Feb 2006 作者： Profile StokeyBob
Post:
The Gas Giant,

I've been watching my "to completion" times in the BOINC manager and the two 3.2 machines are still OK but the 2.4 machine is still running with high times.

Just in case you were onto something I removed the last five updates to see what will happen. I know I am risking having wrath on Microsoft coming down on me. I already have the little balloon thing popping up.

When I removed the updates I had to do it in a certain order. Maybe sometimes they don't install right.?

Anyway, if I notice anything different I'll try to check back in here.
106) 留言板 : Number crunching : Strange reset problem? (消息 251338)
发表于：21 Feb 2006 作者： Profile StokeyBob
Post:
I had a similar experience around the same time. I rebooted/crashed and never got the data for a bunch of workunits. I find those red lines staring at me to be somewhat disconcerting.... here's my current page 'o errors.

I doubt that your crash had anything to do with the other people not returning results. I find that a fairly large number of the workunits that I return don't get all 4 results returned -- they tend to "concentrate" on my last page of results....


I wasn't so much thinking that my crash effected anyone else so much as I was thinking that what caused my crash may have caused others.

How else can all the duplicate work being received at the same time and not being returned be explained?

It goes beyond any odds or anything I can think of to explain it otherwise.

I see that you have a bunch of twin non-returns in your list also.

P.S. I could be just imagining things. I don't think I ever bothered to really look at the results of work units I lost. I suppose It could be common for there to be pairs. There are people that got them at the same time and succeeded. That is what counts.

P.S. P.S. Actually I think that what caused my crash is the power line to my computers finally burned all the way through and maintenance repaired it today. It never blew a fuse, it just burned the hot wire through. I hope the only crashes I get for a while are from playing games.
107) 留言板 : Number crunching : Strange reset problem? (消息 251234)
发表于：21 Feb 2006 作者： Profile StokeyBob
Post:
I had a machine crash or reset it self a couple of weeks ago. As I was looking through some of the results I trashed I noticed something odd. Work units that I received were also received by other machines at the same time and not returned in time. Here is an example of three but there are a bunch.

wuid=53273232

wuid=53168183

wuid=53760554

P.S. Sorry if the links fail. They are getting removed pretty fast.
108) 留言板 : Number crunching : So, where is credit then? (消息 251209)
发表于：21 Feb 2006 作者： Profile StokeyBob
Post:
Mine was strange. It didn't have any credit or work units it was just sort of a duplicate. I was able to delete it.

It looks like el nino is going to have more trouble than that getting things straightened out.
109) 留言板 : Number crunching : Is it just me?(hope not:) (消息 251205)
发表于：21 Feb 2006 作者： Profile StokeyBob
Post:
You could try setting your cache a little higher.

Go to "YOUR ACCOUNTS" at the top of the page. Click on the blue "View or edit general preferences". At the bottom of the page click on the "Edit preferences". That will allow you to increase the spot that says "Connect to network about every". That will increase the amount of work that you will have on your machine ready to run. Maybe try one day or something small at first. You want it small enough that you get the work returned on time but big enough to get through an outage. I've heard that it is best to keep it on the low side for multiple projects.

To get it to take effect immediately you need to update the project under the BOINC managers "Project" tab.
110) 留言板 : Number crunching : This software messed up my pc! (消息 251196)
发表于：21 Feb 2006 作者： Profile StokeyBob
Post:
I've never heard of it doing that.

Where did you get the copy that you downloaded?
111) 留言板 : Number crunching : I can't believe current top compter list!!! (消息 251191)
发表于：21 Feb 2006 作者： Profile StokeyBob
Post:
I'm not sure about this but I think what is happening is if you had 10 of your 3.0Ghz Pentium computers on your computer list and then you merged them all together they would show up looking like some sort of super computer.
112) 留言板 : Number crunching : So, where is credit then? (消息 251170)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
It looks like some sort of error to me.

I had a mystery duplicate machine show up on my list the other day.

Maybe something weird is happening to the data since the merging thing got fixed.
113) 留言板 : Number crunching : whats with the stats ????? (消息 251165)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:

I'm alo wondering what it means on the stats as to the OVERTAKE coluums - make no sense to be - overtake WHAT OR WHO?


The way I've seen the term overtake used is it is referring to how many days until you overtake the next person, team or computer on the list.

On my BOINCstats world position a guy named "Roberto Biagiotti" would have overtaken me in one day. It looks like he will have to wait a little bit more.
114) 留言板 : Number crunching : Optimized problem (消息 251121)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
It doesn't look like he client erred that many. He's turning in good work now so if it does drop it will rack back up right away.

It can be a bummer to wipe out all of your work and then be trying to figure out what is wrong after you've been cut off. I've had it happen but it seems to get going again way before 24 hours is up.
115) 留言板 : Number crunching : whats with the stats ????? (消息 251116)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
BOINCstats doesn't have your credit. [BOINCstats] Willy is just a guy like you or me, except a lot cooler, because he set up a website on his own to report the information that SETI/BOINC sends out for statistics sites like his.

SETI hasn't sent the information out in couple of days. It happens at times, but the credits eventually turn up. Maybe in one big lump that will give you a record for your best five days list.

I've only got one day on the list that I haven't broken this year.

2005-12-15 8,933
2006-02-01 6,238
2006-01-07 5,031
2006-01-02 4,882
2006-01-04 4,830
116) 留言板 : Number crunching : Optimized problem (消息 251092)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
Terminal123,

Did you put an "app_info.xml" in with the optimized SETI application?

You may want to drop your connect to days down to something like .10 until you see good work coming through.

I am sort of partial to the instructions on the http://marisan.nl/seti/ web site. They should work for any of the optimized applications I believe. They are easy to read through anyway and will give you an idea of the steps to take.

Maybe you are trying an optimized application that doesn't match your computer.

This page on http://marisan.nl/seti/reference.htm has a list of processors and optimized applications that have been tried together. Maybe that may help to see if yours matches. They aren't the latest and greatest but the information may help. One thing though they ran rock solid.
117) 留言板 : Number crunching : Upgrading to Vista? (消息 250960)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
“An upgrade of the motherboard is considered to result in a “new personal computer” to which Microsoft® OEM operating system software cannot be transferred from another computer. If the motherboard is upgraded or replaced for reasons other than a defect, then a new computer has been created and the license of new operating system software is required.”

The reason Microsoft gave for this term is that “Microsoft needed to have one base component “left standing” that would still define that original PC. Since the motherboard contains the CPU and is the “heart and soul” of the PC, when the motherboard is replaced (for reasons other than defect) a new PC is essentially created.”


They sure seem to be missing an opportunity to have their operating system and your data being considered the heart and soul.

I suppose this means that once your OEM computer dies, it's dead. It still looks like the ones that you build yourself can live on forever.

This does seem a little like being forced to burn books because you've moved or gotten new glasses.
118) 留言板 : Number crunching : Impossible WU? (消息 250939)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
Did you notice that we now have color on the client errors and work units that are over?

http://setiathome.berkeley.edu/workunit.php?wuid=54886083


Niiice.
119) 留言板 : Number crunching : Slower Crunching Since 'Doze Updates? (消息 250853)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
If you are seeing the times climb then the "to completion" times in the BOINC manager will be moving up if you are using the 5.2.13 version. At least I'm thinking that's the way it works. I wrote my times down a while back.

3.2GHz was 55:18 and is now 56:27

3.2GHz was 45:49 and is now 49:36

2.4GHZ was 55:49 and is now 1:17:24 and is headed back down (just saw it change to 1:15:55).

I'm thinking we have just had some bigger than normal work units. I went through a bunch of results today and I noticed a couple of honkers.
120) 留言板 : Number crunching : New User (消息 250844)
发表于：20 Feb 2006 作者： Profile StokeyBob
Post:
Basically it is done by an average.


well more a median than an average 8)


They take out the high and the low and the the number in the middle is what is granted usually. On a rare occasion I think I have seen a average of the two middle numbers.


前 20 · 后面 20


 
©2020 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.