How do I find out about AR

Go to your computer page accessible from your account page at the main Website page, select "Results" and select the result you want from your results in the left-hand column. Halfway down the page of the individual result the angle range of that workunit will show. I see you're using Simon's cruncher just like I am. This may not work with those who are using the default cruncher. Angle ranges in the 0.22 degree area are very rare. That's the first one that I've seen that close to 0.226. My guess as to the reason for this is that the Arecibo receiver mechanism is either stationary (and stars drift by about 0.356 to 0.446 degrees in 107 seconds), it's tracking the stars (and stars drift less than 0.113 degree in 107 seconds) or it's slewing (and the stars drift by up to 15 degrees in 107 seconds). Probably there has been little effort spent in half-tracking the stars to produce a 0.226. But I can't quite explain the 0.6s, 0.8s, etc. Guess it's just a short movement of the apparatus.

---

98.90

Workunit 95241960
Created 17 Oct 2006 3:27:31 UTC
Sent 17 Oct 2006 14:56:04 UTC
Received 18 Oct 2006 1:39:01 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 2645708
Report deadline 5 Nov 2006 17:16:04 UTC
CPU time 19734.265625
stderr out

<core_client_version>5.4.11</core_client_version>
<stderr_txt>
Optimized Windows SETI@Home Enhanced application
Version info: Windows SSE2 32-bit V5.15 'Chicken Good!' (R-1.3|+freq|xW)
Compiled by Simon Zadra (KWSN - Chicken of Angnor) - Member of the Knights who say Ni! (http://www.kwsn.net)
Download Updates at: http://www.zadra.org/seti_enhanced/

CPU real speed: 2496 MHz

Work Unit Info
True angle range: 0.083472

Flopcounter: 25504536222870.453000

Spike count: 0
Pulse count: 4
Triplet count: 6
Gaussian count: 0
</stderr_txt>

Validate state Valid
Claimed credit 98.8891161419167
Granted credit 98.8956130651039
application version 5.15

Result

---

98 credits for those VLARs (0.000 to 0.112) seems to be a fluke. Mine have been about 58 to 59 credits. But maybe not all VLARs yield the same credit. Still, that's funny. I did see an unusual one today, a 70-credit one. It was just above the VLAR range, maybe 0.113.

---

How do I find out about AR

Go to your computer page accessible from your account page at the main Website page, select "Results" and select the result you want from your results in the left-hand column. Halfway down the page of the individual result the angle range of that workunit will show. I see you're using Simon's cruncher just like I am. This may not work with those who are using the default cruncher. Angle ranges in the 0.22 degree area are very rare. That's the first one that I've seen that close to 0.226. My guess as to the reason for this is that the Arecibo receiver mechanism is either stationary (and stars drift by about 0.356 to 0.446 degrees in 107 seconds), it's tracking the stars (and stars drift less than 0.113 degree in 107 seconds) or it's slewing (and the stars drift by up to 15 degrees in 107 seconds). Probably there has been little effort spent in half-tracking the stars to produce a 0.226. But I can't quite explain the 0.6s, 0.8s, etc. Guess it's just a short movement of the apparatus.

thanks found it

---

the universe is to big to just have life on one planet

I'll have to eat my words about the 98-credit fluke. I did a 93.97 (AR .088046) and an 88.18 (AR .09398). However the prediction data (on which I made a graph several months ago) is way off. All the 0.000 to 0.110 were supposed to take the same time to crunch, as well as all the 1.14 and up. Nothing could be further from the truth.

---

Hey why not, sounds like fun.

Ok here goes.

My highest credit for a WU so far is...drum roll please....

HostID  WU_id       Res_id      Angle       Date        Secs            Flops       Claimed Granted
2739880 95257904    397355306   0.133037    10/20/2006 4:45 15,686.97   2.3075E+13  89.47   89.47

Well, thats not quite true. I tried changing flops total value to be stored in a 64 bit integer, but compiler bugs made calculating it go wrong sometimes. Very wrong...toward the high end...
My actual highest claim (faulty) is .
2739880 94697091 394961666 0.434598 ... 18,612,225.34

I have since corrected the <edit>64 bit integer problem I introduced into the </edit> code ;)

I have since corrected the code ;)

So why waste time ...post it on the dev list as soon as possible to get it checked into the source ;-)

---

Join BOINC United now!

I have since corrected the code ;)

So why waste time ...post it on the dev list as soon as possible to get it checked into the source ;-)

Sorry Crunc3r to get excited...but my phrasing was imprecise "I have corrected the bug I introduced by making it a 64 bit integer, and changed it back to a double value". I haven't created a "fix" yet for VLAR low/high credit problem.

I have since corrected the code ;)

So why waste time ...post it on the dev list as soon as possible to get it checked into the source ;-)

Sorry Crunc3r to get excited...but my phrasing was imprecise "I have corrected the bug I introduced by making it a 64 bit integer, and changed it back to a double value". I haven't created a "fix" yet for VLAR low/high credit problem.

Sorry for the misundertanding, thought that you hand a fix to the VLAR/Credit BUG ...

Sorry

---

Join BOINC United now!

I'll have to eat my words about the 98-credit fluke. I did a 93.97 (AR .088046) and an 88.18 (AR .09398). However the prediction data (on which I made a graph several months ago) is way off. All the 0.000 to 0.110 were supposed to take the same time to crunch, as well as all the 1.14 and up. Nothing could be further from the truth.

Quite true. The splitter estimates from which those tables were derived change formula at half the min_slew rather than one beam_width. They forecast a flat line from ar=0 to ar~=0.113 but that should cut off at 0.083 (the beam_width for the Arecibo line feed). The formula above that point is not remotely related to reality either. The formulas for 0.226 and above aren't too terrible.

The issue around 0.083 is that Pulse and Triplet finding are done over 1 beam_width of data. At 0.084 they have to do just about twice the work as at 0.083. The credit claim doesn't double because FFTs, chirping, and Spike finding don't change significantly at that divide.

                                                          Joe

Check this out
http://setiathome.berkeley.edu/workunit.php?wuid=101504813

claimed credit = 58.87
granted credit = 103.60

how this happened ?

---

Check this out
http://setiathome.berkeley.edu/workunit.php?wuid=101504813

claimed credit = 58.87
granted credit = 103.60

how this happened ?

Two computers in the quorum running really old versions of BOINC - 4.19 and 4.25

These versions don't report the number of calculations SETI has done on the workunit - just the time taken, and the benchmark speed of the computer. The amount of credit claimed by these programs can vary widely - sometimes too low, sometimes (as here) too high.

Don't worry about it - it's their problem, not yours!

Two computers in the quorum running really old versions of BOINC - 4.19 and 4.25

These versions don't report the number of calculations SETI has done on the workunit - just the time taken, and the benchmark speed of the computer. The amount of credit claimed by these programs can vary widely - sometimes too low, sometimes (as here) too high.

Don't worry about it - it's their problem, not yours!

Thanks for the info Richard.

---

My highest new credit

---

My highest new credit

Hmmmm, did you notice the app said the CPU was running at -103.222 Ghz for that run?

Now that's a PD that overclocks like you read about! :-)

Alinator

My current highest wu is 80.01 http://setiathome.berkeley.edu/workunit.php?wuid=101006903

---

Pretty soon you'll probably get a VLAR (which have an angle range of from 0.000 ti 0.112 degree) unit. Some, but not all, of these give 90 or so credits.

---

96 and it takes my system 24hrs to do it.

---

96 and it takes my system 24hrs to do it.

Perhaps better stated as "...it takes half of my system 21hrs to do it."

                                                       Joe

Simons latest does well on VLARs (with my PD950s). They take longer but they have more credits. VLARs yield more credits per hour (on my PD950s). The units that have around 0.7 to 1.12 angle range don't take quite as long to do but have much fewer credits. Their credits-per-hour yield is low, often only 1/2 that of the VLARs. With other setups Simon's latest will behave differently.

---