Many KWSN got lost or fell down on the way to registration.
Over 70 of us did show for the start. Other teams seem a bit low also.

I love that pic!
Keep in mind that SETIzens can always register after WoW has started.
That's why I stopped my ridiculous countdown.
Plus, I lost my inspiration.

Maybe we'll see the Yip-Yips in another post.
Or maybe even Gonzo...since he thinks he's an alien! lol

T -2h29mins.

We're down about 200 participants compared to 2014 and 2015. I don't know if we will exceed any stats set in previous years this time. Of course, the ringer (Freezing) will help this year. I can't even imagine his crunchers farm. WOW!

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Stephen what does Boincstats say your output for yesterday was? Maybe you are still getting hit by the recent negative credits that SETI dumped on us over the weekend. That resolved itself on Sunday for us here in the States. Your output should be around 35k based on your stats here on SETI. Somethings wrong with your account reporting for the contest.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

With a RAC of ~40k, and 19k in 10hrs: that's about right
...if you didn't turn off network activity before the start of WoW (in order to keep your tasks local in "Uploading" state).

Both my rigs had a stash of 1,000 tasks and were offline for 4 days.
That's how I got in the top10 so fast. But one rig had already starved ~8hrs before the start.
And the other one crashed while being 200kms away and stayed down for ~30hrs.

So, you'll just keep going up the ranks slowly
...while I will slowly fall away down the 1st page to meet you somewhere on (my guess) the 2nd page.
(Unless Zalster's prediction SG WoW is an under-estimate:

21:47 Zalster: @Stubbles Be prepared for many more bunkers about half way and probably the last day Posted: Tuesday Aug. 16, 2016 at 4:58.

)

Both my rigs had a stash of 1,000 tasks and were offline for 4 days.
That's how I got in the top10 so fast. But one rig had already starved ~8hrs before the start.
And the other one crashed while being 200kms away and stayed down for ~30hrs.
21:47 Zalster: @Stubbles Be prepared for many more bunkers about half way and probably the last day Posted: Tuesday Aug. 16, 2016 at 4:58.

I wondered what was meant by 'bunkering'. Slick operation to get a high RAC at the beginning of the contest. I wondered how you made into the Top Ten. :^}

Stephen still should be reporting more than the 1900 credits per hour his stats page shows. Should be closer to 5K.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

why do you say that Stephen?

Have you looked at each computer's daily output today to see what they were?

Free DC or boincstats?

. . Because it is only showing 18,000 which is the output of this machine alone. According to 'My computers" listing on SETI this machine is 18,000, the Pent-D is 15,000 and the C2D is 5,600. Not seeing that on WOW. My total RAC is 39,000 (or thereabouts) for the whole farm.

[edit] . . I guess I do not know how to read the WOW stats correctly, It seemed to say that I had only completed 18,000 the previous day and to me that is this machine alone, but the difference and things did not make sense to me :(

Stephen what does Boincstats say your output for yesterday was? Maybe you are still getting hit by the recent negative credits that SETI dumped on us over the weekend. That resolved itself on Sunday for us here in the States. Your output should be around 35k based on your stats here on SETI. Somethings wrong with your account reporting for the contest.

. . I don't have a BOINCSTATS account to check with but yep SETI has my RAC between 38,000 and 39,000. I feel like I am letting team LEO down :(

. . On FreeDC it shows 41,093 for yesterday and 39,000 for today (which I take it are 24 hours behind)

why do you say that Stephen?

Have you looked at each computer's daily output today to see what they were?

Free DC or boincstats?

. . Because it is only showing 18,000 which is the output of this machine alone. According to 'My computers" listing on SETI this machine is 18,000, the Pent-D is 15,000 and the C2D is 5,600. Not seeing that on WOW. My total RAC is 39,000 (or thereabouts) for the whole farm.

so you are doing 18K an hour on that machine? or 18K a day?

With a RAC of ~40k, and 19k in 10hrs: that's about right
...if you didn't turn off network activity before the start of WoW (in order to keep your tasks local in "Uploading" state).

Both my rigs had a stash of 1,000 tasks and were offline for 4 days.
That's how I got in the top10 so fast. But one rig had already starved ~8hrs before the start.
And the other one crashed while being 200kms away and stayed down for ~30hrs.

So, you'll just keep going up the ranks slowly
...while I will slowly fall away down the 1st page to meet you somewhere on (my guess) the 2nd page.
(Unless Zalster's prediction SG WoW is an under-estimate:

21:47 Zalster: @Stubbles Be prepared for many more bunkers about half way and probably the last day Posted: Tuesday Aug. 16, 2016 at 4:58.

)

. . I didn't think of "buffering" my results and sending them up as one big bunch, very sneaky. I do buffer my results as a normal thing but only for a few hours at a time so I can monitor how the runtimes are looking (a few dozen jobs). But I guess Karma bit hard in your case ... :(

.

why do you say that Stephen?

Have you looked at each computer's daily output today to see what they were?

Free DC or boincstats?

. . Because it is only showing 18,000 which is the output of this machine alone. According to 'My computers" listing on SETI this machine is 18,000, the Pent-D is 15,000 and the C2D is 5,600. Not seeing that on WOW. My total RAC is 39,000 (or thereabouts) for the whole farm.

so you are doing 18K an hour on that machine? or 18K a day?

. . That would be 18K a day :( . It is only a modest GTX 950 on an i5-6400.

. . So what exactly is "bunkering"?

You download as much work as you can across your computers, don't allow boinc to communicate back to seti and process the data and hold onto it until the event starts.

Once the event starts, you allow the computers to reconnect and dump all the results back to seti for instant credit on those waiting for validation. ie dropping a "bunker"

Then there is ghosting

Some of these people will download large amount of work units after the event starts and either hold onto those work units until after the event ends, or by means I won't go into, erase the work units off their computers so that those work units appear to be in progress and won't be validated for the event, so they time out weeks later and have to be resent to another person...

Now Stubbles, you know why people were upset about your 1000 work units caches. Imagine ghosting 100K work units just to keep other teams from collecting their credits.

. . That would be 18K a day :( . It is only a modest GTX 950 on an i5-6400.

I can't say for certain that it is only seeing 1 of your computers but remember each hourly update is only for those work units that got credit during the event.

Also the event started at midday of the reporting cycle (or somewhere around there) so you won't see a full days credit for your machines until the end of tomorrow.

Though you have me wondering if it is occurring like you speculate, guess we will need to keep an eye on it.

. . So what exactly is "bunkering"?

I think the best example up to now during WoW 2016 is:
Freezing collecting 3,000,000+ credits in just one hour!

You might want to read the WoW comms in the left hand side on any SG WoW webpages.
I have learned a lot today thanks to many, with special thanks to Zalster and Petri!

. . So what exactly is "bunkering"?

You download as much work as you can across your computers, don't allow boinc to communicate back to seti and process the data and hold onto it until the event starts.

Once the event starts, you allow the computers to reconnect and dump all the results back to seti for instant credit on those waiting for validation. ie dropping a "bunker"

Then there is ghosting

Some of these people will download large amount of work units after the event starts and either hold onto those work units until after the event ends, or by means I won't go into, erase the work units off their computers so that those work units appear to be in progress and won't be validated for the event, so they time out weeks later and have to be resent to another person...

Now Stubbles, you know why people were upset about your 1000 work units caches. Imagine ghosting 100K work units just to keep other teams from collecting their credits.

. . OK! The bunkering I can understand, sneaky but non destructive. The second practice is downright reprehensible. Sabotaging the project by deliberately wasting masses of WUs to gain an advantage in what is just a bit of fun? that sucks!

Though you have me wondering if it is occurring like you speculate, guess we will need to keep an eye on it.

Zalster,
Is there anything wrong with my calculations of Stephen's situation?

With a RAC of ~40k [edit]for your farm[/e], and 19k in 10hrs: that's about right
...if you didn't turn off network activity before the start of WoW (in order to keep your tasks local in "Uploading" state).

He's now at 21k after 13hrs of reporting.
It still looks good to me.
He should be above 40k after the Tuesday amintenance...likely around ~45k by my estimate.
Rob :-) ...going to Zzzzz

@Stubbles

Don't know, that is assuming a perfect world with all results coming back in quickly and no inconclusive or invalids.

World is seldom so accurate, lol

Best guesses are probably where we are.

Or in this case, a wait and see...

lol...

Laters

Now Stubbles, you know why people were upset about your 1000 work units caches.

I still don't... especially since I was testing a potential error with a specific scenario of a fast CPU and a slow GPU with Mr Kevvey's script,
I was testing to see if there was a limit before Mr Kevvy's script crashed (since it was slowing down to over 30secs with over 500 tasks that had optimized queues.
And in doing so, I discovered the 1,000 Boinc limit that no one ever mentionned.

The problem is that some assume and then use inappropriate words for a post or PMs.
To me "Hoarding" was only a grave clinical psychological disorder.
Even now, it's still an inappropriate term if Ghosting is implied.
Plus I was always clear that I would process my queues within 10 days.

Sorry, I got a small flashback.
As you can tell, I take issue with poor communication especially when others jump to judgemental negative conclusions and use abusive language.
I'm done hahaha

G'night,
RobG :-}

Though you have me wondering if it is occurring like you speculate, guess we will need to keep an eye on it.

Zalster,
Is there anything wrong with my calculations of Stephen's situation?

With a RAC of ~40k [edit]for your farm[/e], and 19k in 10hrs: that's about right
...if you didn't turn off network activity before the start of WoW (in order to keep your tasks local in "Uploading" state).

He's now at 21k after 13hrs of reporting.
It still looks good to me.
He should be above 40k after the Tuesday amintenance...likely around ~45k by my estimate.
Rob :-) ...going to Zzzzz

. . I am now convinced you are right. But when I first looked at the stats it said "yesterday 18,000" and yesterday according to FreeDC I did either 41,000 or 39,000 for the farm (depending on how far behind FreeDC is) so I mistakenly presumed it was only looking at the GTX950 machine. But I guess it is just the way the WOW monitoring software handles the stats. The difference column still baffles me, but the totals are now making sense.

The problem is that some assume and then use inappropriate words for a post or PMs.
To me "Hoarding" was only a grave clinical psychological disorder.
Even now, it's still an inappropriate term if Ghosting is implied.
Plus I was always clear that I would process my queues within 10 days.

. . I agree, accusing someone of hoarding WUs implies the intention of retaining them long term or doing nothing with them. When the clear intention is to process them and return results as quickly as possible then that is false and a little abusive.

. . Even the way the accusation was couched .. "I denounce you as a hoarder" sounded like something from a witch trial and made me wonder if the accuser was making a joke, even if only of himself. :)

The problem is that some assume and then use inappropriate words for a post or PMs.
To me "Hoarding" was only a grave clinical psychological disorder.
Even now, it's still an inappropriate term if Ghosting is implied.
Plus I was always clear that I would process my queues within 10 days.

. . I agree, accusing someone of hoarding WUs implies the intention of retaining them long term or doing nothing with them. When the clear intention is to process them and return results as quickly as possible then that is false and a little abusive.
. . Even the way the accusation was couched .. [***quote removed***]
sounded like something from a witch trial and made me wonder if the accuser was making a joke, even if only of himself. :)

I totally agree Stephen. Thank you. It means a lot to me to have someone finally paraphrase what I've been saying.
(Just keep in mind, I wasn't trying to "return results as quickly as possible" since I was doing tests. Just with a turnaround of 5 days or so.)

I am aware that I could have probably done a better job of defending myself.
I've just never been involved in such a situation, and I was very taken aback by the agressive, hostile and insulting words.
And when you don't know the protagonist's style, in my mind that's why the mods are there.
If someone's posted over a thousand posts, I can't imagine that I was the 1st to experience his wrath.

I don't know if you noticed that I remove the quote you included in your post.
The only reason is because someone could have an issue with it and I didn't want my reply to be removed ...since it could:

"... Message board postings are subject to the following posting rules:
...
- Do not posts comments related to specific moderator actions. The moderators may be contacted at seti_moderators@ssl.berkeley.edu
- Your posts may be deleted if it contains material from another post which has been deleted. You should avoid including material from a post you expect will be deleted.
"

Hopefully there will be some openess within S@h to update the rules to include PMs. That's my only goal now.
Cheers,
Rob :-)

He's now at 21k after 13hrs of reporting.
It still looks good to me.
He should be above 40k after the Tuesday amintenance...likely around ~45k by my estimate.
Rob :-) ...going to Zzzzz

. . I am now convinced you are right. But when I first looked at the stats it said "yesterday 18,000" and yesterday according to FreeDC I did either 41,000 or 39,000 for the farm (depending on how far behind FreeDC is) so I mistakenly presumed it was only looking at the GTX950 machine. But I guess it is just the way the WOW monitoring software handles the stats. The difference column still baffles me, but the totals are now making sense.

BoincStats, FreeDC and others use the daily dump found in: https://setiathome.berkeley.edu/stats/
and when those files are not updated (or are updated incorrectly) then we experience that like we did this week.
The 1st bad update: the data was likely not updated since, it reported +0 incremental credits.
The 2nd bad update: the data was likely corrupted since there was a negative incremental updates...which is impossible, like a Ferris Bueller Day Off moment trying to make the Ferrari's odometer go backwards.
I don't know about FreeDC, but BoincStats host data is always 20-44hrs old (and the user stats are alsways 1 to ~26hrs old) unless something goes wrong.

SG WoW uses hourly updates. They likely have the permission from the S@h staff to query the server(s), (but it might be slightly diff from what Shaggie does with his scripts),
This likely puts a much higher load on the S@h HTTP server (if SG WoW doesn't have a direct VPN connection to query the server database with SQL statements).
SG WoW is a great time to test the S@h infrastructure (hardware and software at the same time).
I was looking at the Server Status Page yesterday during the 1st few hours of WoW.
The field "Results received in last hour" went up from the normal 70-75k load, to almost 130k when Freezing dumped his tasks to cash in 3 million credits.

Keep in mind: the project staff is condoning "bunkering".
Had they issued a statment like: "We are for the SG WoW event but we are concerned that the servers might not survive the added stress"
then I wouldn't have dumped my 1700 tasks during the first two hours of WoW. I wouldn't even have turned off my Boinc network activity prior to WoW.

Hope this makes sense in trying to understand the bigger picture.
I might be off on some details though and I invite constructive feedback...since that's a great way for everyone to keep on learning about the incredible Boinc architecture and how this great project continues to use is and adapt (to) it.

Cheers and happy SG WoW event day-2 (starting soon),
RobG :-)