Okay Folks

I am back, I did not announce that I took a forced vacation as I had to move from one apartment to another... and from Sat (in June) to Thurs (in July) I had no Internet Connectivity. ARRGGGGGGGGHHHHHHHHHHHHHH

As I did a quick read there are many that are reporting things Astropulse that I do not see in the Seti Beta Forum. Josh is working to get this released and I hate to have to point him here to figure out what is "real." He had hoped to be able to take a Vacation this Summer...

I do see some things that are "BIT" Off Topic with respect to Astropulse... So when I created the Topic here wanted to insure that You had a chance to see some of the things that are about to happen. I do see others helping to explain what they saw how it affected them.

Thank You All

Al

---

Please consider a Donation to the Seti Project.

"TRUE" = Thermalright Ultra120 Extreme cooler.

Sorry didn't know "TRUE" referred to a HSF.

A quick review shows my oldest Pending to have reported about 40 days ago. You may well be right in your estimate of when I swapped the processor (and the MoBo - the RAM upgrade came a little later) but comparing crunch times on a range of sample WU's with the other rigs in
the vicinity convinces me that this is "home" for this setup.

Okay... why hide behind an older host? Why not start from scratch and eliminate any confusion?

My point is... here is another top 20 host... This host was just created in the past 24 hours. How can I believe the RAC that put the host in the top 20 when it was just created? So many ways to confuse the stats... Lets keep it simple and without a doubt.

I can't even merge the same host with itself running the same OS... I find it hard to believe the posted hosts in the top 20.

[My last OT post in this "AstroPulse" thread...]

Well it is getting very close

Astropulse release soon

You are welcome to go read.

Al

---

Please consider a Donation to the Seti Project.

Well it is getting very close

Astropulse release soon

You are welcome to go read.

Al

What would be the 'recommended host configuration', CPU & speed, amount off RAM, etc.

I'm astonished, why so many 'threads' are closed and why?
Hope, all 'pieces' fall into their, respective places?!

---

What would be the 'recommended host configuration', CPU & speed, amount off RAM, etc.

I'm astonished, why so many 'threads' are closed and why?
Hope, all 'pieces' fall into their, respective places?!

Hi Fred,
For your first question, I tried running a couple Astropulse wu's on my 1.3ghz P3 with 384mb ram. It ran. That's all I can say. Hehe. It was claiming about 600 hours or more to finish. One was canceled before it was finished due to an app version change (either that or the other hosts already returned theirs and mine was redundant, don't know which). I cancelled the other one when after arriving back at home after a lengthy stay out of town I found out a power failure had shut down my computer and it hadn't been crunching for over three weeks.
So as far as what "will" run it, I'd say any P3 with at least 256mb ram, but as far as running it quickly enough, I would say a minimum of 2-2.5 ghz and 512mb.

For your second question, all I can say is "don't ask". We are not allowed to discuss *any* moderator actions.

---

Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.

So as far as what "will" run it, I'd say any P3 with at least 256mb ram, but as far as running it quickly enough, I would say a minimum of 2-2.5 ghz and 512mb.

Is there a reason why the AP units need to finish quickly?

-Dave

---

Fred et al

What I was told as far as Seti Main, they are looking for a set of machines that can complete the "Result" in roughly 4 days (96 hours or 345600 seconds)
From the same conversation it was mentioned that an opt-in or opt-out "preferences set" was being considered.

So initially, if you have an app_info.xml file in place you will not see Astropulse.

What would be the 'recommended host configuration', CPU & speed, amount off RAM, etc.

My X2 6000 returns a result in roughly 80 hours.

Edit:
As I looked in the init_data.xml

<rsc_memory_bound>167772160.000000</rsc_memory_bound>

Astropulse is looking for 168 meg of RAM but running roughly 50 meg of RAM

---

Please consider a Donation to the Seti Project.

Hi Jim
If this is moderated off so be it.

I have been running Astropulse for some time over on Beta.

I would go along with It’s not so much on what it will run but how it is run. I would say ensure you don’t download too many WU’s, Keep your connection time at first to a minimum 0.1000 to 0.5000 days. At first ignore the completion times and see what your host will do. I also run seti@home and milkyway@home so these are shared hosts. Activity wise always run always connect.

The only time I have not been able to complete a Astropulse WU was when I set connection time to 10.0000 days instead of 1.0000 days you learn by your mistakes.

Some basic times at the moment on my hosts are below.

Q6600 2.4GHz running 24/7 between 44 hrs – 50 hrs or just over 2 days return time.
Q6600 2.4GHz running 12/7 between 44 hrs – 50 hrs or just over 4 days. Return time.

T2070 1.7GHz running 24/7 between 100hrs -120 hrs or just over a 5 days.

Same host T2070 running as a laptop 100 hrs – 120 hrs anything up to 14 days to 18 days.

Michael

---

Dave

From what was stated, there is no real way to reduce the size of the data file to make the Application effectively run faster. With Older Machines taking weeks it would drive pending credit nuts.

So choices boiled down to only give out Astropulse to machines that can return it in xx.xx hours or trickles.

Trickles requires changes to the Server Side and the Application.
Machines "fast enough" can be done by creation of a table of "certain machine types" and then the Scheduler be ammended to check for a machine type...

Is there a reason why the AP units need to finish quickly?

-Dave

---

Please consider a Donation to the Seti Project.

So choices boiled down to only give out Astropulse to machines that can return it in xx.xx hours or trickles.

It seems like the reverse is happening to me.I get all 6.02s on my quad, at 3420 and aps on a pentium d 920 at 3150. Both machines run well! It looks like the APs will be about 60~80 hrs.

Voyager

Recently (luck of the draw) I recieved two Astropulse "resends" when there was no new AP workunits to work on. So without trying to sort which direction to face in the morning or which of the gods (mythology/computer) to pray to... The Scheduler decided what to send.

The point is that the Scheduler will shortly be responsible for deciding which machine to send Astropulse to. That is not easy to decide.
In any case unless you put an app_info.xml that states Astropulse only it will still be hit or miss. You have a Human and two computers that are all trying to decide what will happen.

I do have to note that Intel is ahead as far as Astropluse is concerned.

So choices boiled down to only give out Astropulse to machines that can return it in xx.xx hours or trickles.

It seems like the reverse is happening to me.I get all 6.02s on my quad, at 3420 and aps on a pentium d 920 at 3150. Both machines run well! It looks like the APs will be about 60~80 hrs.

---

Please consider a Donation to the Seti Project.

Pappa, is there any progress on app_info.xml for those of us who run optimised apps, to also be able to run AP?

80-100 hours isn't too bad considering. Orbit wu have been coming in at 303+ hours on my Q6600's recently.

---

BOINC blog

Mark

There is a thread in Astropulse app_info.xml file for AP where Urs has posted a complete AP portion of the app_info.xml. What is left is the release of AP here

Pappa, is there any progress on app_info.xml for those of us who run optimised apps, to also be able to run AP?

80-100 hours isn't too bad considering. Orbit wu have been coming in at 303+ hours on my Q6600's recently.

---

Please consider a Donation to the Seti Project.

Pappa, is there any progress on app_info.xml for those of us who run optimised apps, to also be able to run AP?

80-100 hours isn't too bad considering. Orbit wu have been coming in at 303+ hours on my Q6600's recently.

On a your Q6600 expect AP times of about 40 hours.

Mark

There is a thread in Astropulse app_info.xml file for AP where Urs has posted a complete AP portion of the app_info.xml. What is left is the release of AP here

Pappa, is there any progress on app_info.xml for those of us who run optimised apps, to also be able to run AP?

80-100 hours isn't too bad considering. Orbit wu have been coming in at 303+ hours on my Q6600's recently.

There's another related issue, of course. When an app_info.xml is in place, the files it lists are also supposed to be there. So to stay with optimized setiathome_enhanced and add AstroPulse requires downloading the AstroPulse files as well as adding the section to app_info.xml.

It ought to be possible to add the file URLs in the <file_info> parts of the app_info.xml once we know the correct URLs, then BOINC would fetch. Or someone could host archives of the file sets for each platform which could be manually downloaded and extracted. I don't know which approach would be best, though obviously the archives would help to limit Berkeley's server load during the rollout.

                                                                Joe

Starting to get APs on my quad.. There going at just under 30hrs. Credit looks like it pretty much should be.( my boinc stats have dropped sence switching from AKV8 to beta but all seems well. IMO should launch ok!!

A comment recently seen was that Astropulse could be in Seti Main as soon as Monday the 21st of July

---

Please consider a Donation to the Seti Project.

Also seen on the web (IE Goggle/ search Astropulse/message boards /not logged in will have to find my password again) as Bionc is playing up on connection anything up to five minutes to Projects. Any ideas on how to correct this apart from reinstalling..

Is there any chance of us mere mortals or (rabble) term of endearment I hope. “ the general public” over here being told in advance of these changes. ( who, what, when, where and why?).

I understand that change over has been mooted by Matt as” soon”, and I did read some where this morning maybe soon = Monday? no Guarantees’?

Will we wake up one morning with hundreds of short wu’s.

(96hr 4day) work units what does this mean.

How close a tolerance is being given 1% 5% or 10 hours 15 hours. Or is it 96:00:00 in 96:00:01 out.

What would be the Deadline no one in their right mind is going to say a week surely at present a 30 day deadline would take care of even the 166 hour hosts.

What hosts will and won’t be able to obtain them. Maybe a list?

Machines such as T2080 1.7GHz 100 hour turn around may be in while some faster 2GHZ and 3 GHz 166 hour machines are out.

Also long and short AP wu,s What is this.

As I have pointed out below 783008 Change T2070 to T2080 those who wish to run them should be given the chance.

A limit on units could initially be imposed say a max of 8 for a quad which is not unreasonable for a dual. Even on a T2080 I think with a bit of self management could complete in 30 days. (Okay possibly running 24/7 and NNT but less damage is done to the host running continuously then switching on/off every night.)

---

Also seen on the web (IE Goggle/ search Astropulse/message boards /not logged in will have to find my password again) as Bionc is playing up on connection anything up to five minutes to Projects. Any ideas on how to correct this apart from reinstalling..

Is there any chance of us mere mortals or (rabble) term of endearment I hope. “ the general public” over here being told in advance of these changes. ( who, what, when, where and why?).

I understand that change over has been mooted by Matt as” soon”, and I did read some where this morning maybe soon = Monday? no Guarantees’?

Will we wake up one morning with hundreds of short wu’s.

(96hr 4day) work units what does this mean.

How close a tolerance is being given 1% 5% or 10 hours 15 hours. Or is it 96:00:00 in 96:00:01 out.

What would be the Deadline no one in their right mind is going to say a week surely at present a 30 day deadline would take care of even the 166 hour hosts.

What hosts will and won’t be able to obtain them. Maybe a list?

Machines such as T2080 1.7GHz 100 hour turn around may be in while some faster 2GHZ and 3 GHz 166 hour machines are out.

Also long and short AP wu,s What is this.

As I have pointed out below 783008 Change T2070 to T2080 those who wish to run them should be given the chance.

A limit on units could initially be imposed say a max of 8 for a quad which is not unreasonable for a dual. Even on a T2080 I think with a bit of self management could complete in 30 days. (Okay possibly running 24/7 and NNT but less damage is done to the host running continuously then switching on/off every night.)

Too many questions, to which we 'the rabble' don't have the answers to.

There is a rumour that AP units will be restricted to those hosts that can do a unit in 96 hrs, deadline ????? reading between line I guess short.
How they will determine if computer is fast enough and on Seti sufficient hours I do not know, but don't think it is a simple as some have suggested.

On a desktop core2's at 3GHz they take less than 40 hrs, with present ver. Pappa is saying ~80 hrs on his AMD X2 6000+. A T7xxx @1.7GHz is reported as taking just over the 96hrs suggested.

Also seen on the web (IE Goggle/ search Astropulse/message boards /not logged in will have to find my password again) as Bionc is playing up on connection anything up to five minutes to Projects. Any ideas on how to correct this apart from reinstalling..

Is there any chance of us mere mortals or (rabble) term of endearment I hope. “ the general public” over here being told in advance of these changes. ( who, what, when, where and why?).

I understand that change over has been mooted by Matt as” soon”, and I did read some where this morning maybe soon = Monday? no Guarantees’?

Will we wake up one morning with hundreds of short wu’s.

(96hr 4day) work units what does this mean.

How close a tolerance is being given 1% 5% or 10 hours 15 hours. Or is it 96:00:00 in 96:00:01 out.

What would be the Deadline no one in their right mind is going to say a week surely at present a 30 day deadline would take care of even the 166 hour hosts.

What hosts will and won’t be able to obtain them. Maybe a list?

Machines such as T2080 1.7GHz 100 hour turn around may be in while some faster 2GHZ and 3 GHz 166 hour machines are out.

Also long and short AP wu,s What is this.

As I have pointed out below 783008 Change T2070 to T2080 those who wish to run them should be given the chance.

A limit on units could initially be imposed say a max of 8 for a quad which is not unreasonable for a dual. Even on a T2080 I think with a bit of self management could complete in 30 days. (Okay possibly running 24/7 and NNT but less damage is done to the host running continuously then switching on/off every night.)

Too many questions, to which we 'the rabble' don't have the answers to.

There is a rumour that AP units will be restricted to those hosts that can do a unit in 96 hrs, deadline ????? reading between line I guess short.
How they will determine if computer is fast enough and on Seti sufficient hours I do not know, but don't think it is a simple as some have suggested.

On a desktop core2's at 3GHz they take less than 40 hrs, with present ver. Pappa is saying ~80 hrs on his AMD X2 6000+. A T7xxx @1.7GHz is reported as taking just over the 96hrs suggested.

And my G4 Mac (1 gHz) does them in 370 hours...so any old Macs are out also...

---

Clk2HlpSetiCty:::PayIt4ward