I understand why the switch between the 2. What I think will be confussing, not that it makes much difference but is fun/interesting to see is:
With Seti Classic you can see that you have completed (say) 500 units and someone else has completed 2500 units. You can see a ratio or difference and understand it.
With Seti/Boinc if you have 500 "credits" and someone has 2500 "credits" there is no way to distinguish it because you get different amounts for each project. Maybe it's just me...but i'd like to be able to see a normal ratio like you can with Seti Classic. Any thoughts out there?

---

[img][/img]

> I understand why the switch between the 2. What I think will be confussing,
> not that it makes much difference but is fun/interesting to see is:
> With Seti Classic you can see that you have completed (say) 500 units and
> someone else has completed 2500 units. You can see a ratio or difference and
> understand it.
> With Seti/Boinc if you have 500 "credits" and someone has 2500 "credits" there
> is no way to distinguish it because you get different amounts for each
> project. Maybe it's just me...but i'd like to be able to see a normal ratio
> like you can with Seti Classic. Any thoughts out there?
>

The BOINC version of Seti@home has no correlation to Einstein@home. Two different sciences; therefore comparing SETI to Einstein to CPDN to Predictor is not valid.

Compare SETI Credits to SETI Credits.

BOINC is only a vehicle to run the underlying @home project. SETI@HOME is still running a SETI client.

As for different credits awarded for different WU's it seems to me that a flat rate applied to every WU no matter how small or how big is not really a fair way to evaluate the actual work done. Why should you get the same credit for a small WU that I get for a Large WU or I get the same credit for a small WU that you get for a large WU? The credit awarded is based on several factors that make the BOINC evaluation of work done fair.

It's different and soon you will get use to change..

---

I'd rather speak my mind because it hurts too much to bite my tongue.

American Spirit BBQ Proudly Serving those that courageously defend freedom.

> I understand why the switch between the 2. What I think will be confussing,
> not that it makes much difference but is fun/interesting to see is:
> With Seti Classic you can see that you have completed (say) 500 units and
> someone else has completed 2500 units. You can see a ratio or difference and
> understand it.
> With Seti/Boinc if you have 500 "credits" and someone has 2500 "credits" there
> is no way to distinguish it because you get different amounts for each
> project. Maybe it's just me...but i'd like to be able to see a normal ratio
> like you can with Seti Classic. Any thoughts out there?
>

As the size of the different WUs is absolutely not comparable, a simple WU count will show you nothing if you really want to compare between projects.

For my machine the (approx.) avarage time for 1WU:
Seti: ~3h
Predictor: ~1,5h
Einstein: ~9h
CPDN: ~800h
Lattice: ~10-20h (Alpha, still very unstable ;)

The credit system brings them all to something derived from the real stuff, the flops. For information regarding Credits look here!

---

Gruesse vom Saenger

For questions about Boinc look in the BOINC-Wiki

This is all I meant....and it's no big deal because I don't care if i'm in the top 1% or bottom 1%.
Seti@home gives you credits based on the amount of work (that's fine). But you can't look at that credit "figure" and understand it. You would have to see a chart or a rank or something for that number (the credit #) to mean anything to you. The number may be big or small but it has no real meaning without a reference as to how many units you have done.
Seti@home Classic gives "work units" completed. That's a number you can understand.
See my point?

---

[img][/img]

> The number may be big or small but it has no real meaning without a
> reference as to how many units you have done.
> Seti@home Classic gives "work units" completed. That's a number you can
> understand.
> See my point?

I see your point, but I disagree.

A seti WU can keep my puter busy for anything between 1min and 4h, depending on the amount of noise/signals in it. If I just count them, it says nothing about the amount of work I've done.
The only real measurement would be a real flop-count, but that would be far overdone. So the selected solution was this:

claimed credit = ([whetstone]+[dhrystone]) * wu_cpu_time_in_sec / 1728000

And that's fine with me ;)

> I see your point, but I disagree.
>
> A seti WU can keep my puter busy for anything between 1min and 4h, depending
> on the amount of noise/signals in it. If I just count them, it says nothing
> about the amount of work I've done.
> The only real measurement would be a real flop-count, but that
> would be far overdone. So the selected solution was this:
>
> claimed credit = ([whetstone]+[dhrystone]) * wu_cpu_time_in_sec / 1728000
>
> And that's fine with me ;)

A couple months ago you would have been happy to get all the 1min WU's you could to make your rank and units completed look high. Now you want to be given credit for the amount of work it takes your computer to complete WU's. That's fine by me. The claimed credits however are now in the millions and soon (i guess) billions. Those numbers will be big and mean nothing without reference. Let's see a credit for time as well as the "data units completed" so our minds can understand the figures. The credit unit may start to look like: example 246 to the 10th power....means nothing to the mind.

---

[img][/img]

> A couple months ago you would have been happy to get all the 1min WU's you
> could to make your rank and units completed look high. Now you want to be
> given credit for the amount of work it takes your computer to complete WU's.
> That's fine by me. The claimed credits however are now in the millions and
> soon (i guess) billions. Those numbers will be big and mean nothing without
> reference. Let's see a credit for time as well as the "data units completed"
> so our minds can understand the figures. The credit unit may start to look
> like: example 246 to the 10th power....means nothing to the mind.
>

Personally I fail to see all the brew ha ha over this issue. Your Credits or Points and $1.79 will get you a venti coffee of the day at Starbucks. So who cares if they award one point or 10-25 credits for each work unit finished.

Bragging rights are all they are worth and this time next week I will be bragging I went over 50K SETI credits... :)

---

I'd rather speak my mind because it hurts too much to bite my tongue.

American Spirit BBQ Proudly Serving those that courageously defend freedom.

> A couple months ago you would have been happy to get all the 1min WU's you
> could to make your rank and units completed look high. Now you want to be
> given credit for the amount of work it takes your computer to complete WU's.
> That's fine by me. The claimed credits however are now in the millions and
> soon (i guess) billions. Those numbers will be big and mean nothing without
> reference. Let's see a credit for time as well as the "data units completed"
> so our minds can understand the figures. The credit unit may start to look
> like: example 246 to the 10th power....means nothing to the mind.
>

What is the reference of such a random thing as 'WU'?
And why is there such an obsession with the dumb WU-count with some?

A WU is som random piece of data, crunched for some random time, with more or less results as an output. No connection to any credible count-ability.

What can be counted is the time it takes to crunch the WUs, that's the CPU-time, and that's what everybody donates to the project(s).
IMHO there are two ways to compare this donation:
1. Just count the time, and the older the machine, the higher your donation.
2. Count the work done (i.e. flops) in a managable manner, and the newer the machines, the higher your donation.

I prefer the second solution.

And as CW said, for all your credits and couple of cash you can get yourself a decent coffee in the next Café ;)

> Bragging rights are all they are worth and this time next week I will be
> bragging I went over 50K SETI credits... :)

Congrats CW,I just went over 100K today.

---

> Personally I fail to see all the brew ha ha over this issue. Your Credits or
> Points and $1.79 will get you a venti coffee of the day at Starbucks. So who
> cares if they award one point or 10-25 credits for each work unit finished.
>
> Bragging rights are all they are worth and this time next week I will be
> bragging I went over 50K SETI credits... :)

I personally never look. I just look at processing times and CPU usage on occasion to make sure my machine is still running efficiently.

> I personally never look. I just look at processing times and CPU usage on
> occasion to make sure my machine is still running efficiently.
>

Just my point.. If my machine running efficently I don't rightly care

---

I'd rather speak my mind because it hurts too much to bite my tongue.

American Spirit BBQ Proudly Serving those that courageously defend freedom.

For someone who doesn't care...you sure make sure that your stats are always posted with your comments. You care! YOU REALLY Care!!

Just my point...

> I personally never look. I just look at processing times and CPU usage on
> occasion to make sure my machine is still running efficiently.
>

Just my point.. If my machine running efficently I don't rightly care

Randy said "With Seti Classic you can see that you have completed (say) 500 units and someone else has completed 2500 units. You can see a ratio or difference and understand it. With Seti/Boinc if you have 500 "credits" and someone has 2500 "credits" there is no way to distinguish it because you get different amounts for each project."

That is true HOWEVER even with Classic if you crunch 2500 units and you don't know how much I crunch what good is it? It means nothing unless you have something to compare it to, in your question.
The ratio is still there it is just figured differently. Before it waws done with a unit to time comparison, for non cheaters that was a decent comparison for yourself! BUT to compare it to someone else is like comparing apples to bananas since you have no idea if he is cheating. With Boinc you are comparing apples to apples, because to cheat in Boinc is very difficult at present. I was going to put impossible but never say never.
My stats right now are:
SETI@home member since 17 Dec 1999
Total credit 375,478.58
Recent average credit 1,995.41
SETI@home classic workunits as of 15 March 2005 52,617
SETI@home classic CPU time as of 15 March 2005 343,214 hours

Does that mean your 2500 is wimpy? ABSOLUTELY NOT!!!! It means you are doing the same thing I am, finding something we believe in and sticking with it.
It does mean I have more resources put towards the project than you do, but that is ALL it means!!!

---

> For someone who doesn't care...you sure make sure that your stats are always
> posted with your comments. You care! YOU REALLY Care!!
>
> Just my point...

Bragging rights!!

---

I'd rather speak my mind because it hurts too much to bite my tongue.

American Spirit BBQ Proudly Serving those that courageously defend freedom.