I Agree with Richard, bunkering is not a good option on this last days.
IMHO What is needed to do is a way to send the resends only to those hosts who are still actively crunching and returning their work in the last few days. Maybe a week or something similar. And if possible in a small batch of files only. Something like 10 WU max per host.
The results in the field are already distributed, but a lot will be expired and if they will be sended again on a non crunching hosts (IE a host with AV problem, bunker, etc) will be a long wait until it reaches the dateline again.
And BTW the dateline must be reduced to the minimum possible to expedite the crunch (by make the host entering in panic mode in case it's runs multiprojects). Something radical like 3 days for a GPU WU and 5-7 days for the CPU WU.
My 0.02

. . Again +1

Stephen

:)

Yeah, I have gotten 6 tasks since April 1st. I have one host still connected to SETI. When this host communicates with SETI the scheduler always does a 30 minute back off. I have seen where the RTS shows a dew tasks. I manually have BOINC communicate and I get nothing for the next 30 minutes. The next time I look at the servers, there are zero tasks RTS.

Siran

Are you sure, I checked mine this morning, according to the "all tasks" list I had downloaded, processed returned and had validated 18 tasks in a single batch less than an hour previously.

When I rechecked about 2 hours later, THREE hours after the original download they had already been removed from my lists.

Resends are getting processed and deleted at a very fast rate.

---

Kevin

watching the mind melting rage of those who think this is cheating, and do not even understand the significance of the number 1337, is great entertainment for my morning coffee.

i can say with certainty, the GPU number you are seeing is a label only, and in no way affects how many tasks he's getting.

Hi Ian,

Sorry to disagree with you here, but I do. Case in point, my current host:
When I look at my host list it show [2] NVIDIA GPUs. Guess what, I HAVE 2 NVIDIA GPUs in this host. Now, SETI upped the amount of tasks per device. I was getting 150 CPU tasks and 300 GPU tasks. Before they increased it Before I added the second GPU I was getting 150. Let me whip out my calculator... 1337 x 150 = 200550 and 70000 / 150 = 467 (rounded up 1). So I disagree and say that he got more tasks because he cheated with the spoofing. And the GPU number does have a bearing on the number of GPU tasks one gets. Since SETI was putting out less tasks over the past month or more, he got a lions share of what they did put out to be sent. If he hadn't cheated, my host could probably still be crunching and not just 1 per day or whatever it is now since 4/1. I have gotten 6 since April 1st, he still has over 70K! If he hadn't raised the number from 64 to 1337, he would not have as many today.

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

Hi Juan,
I would not agree with that statement. That would not be fair to those of us that do NOT spoof (cheat) as Ville Saari does. He has over 70K STILL IN PROGRESS! Why allow him to get ALL the resends and not let any through to anyone else. I still have BOINC set to get anything I can from SETI, be it GPU or CPU tasks. If only active hosts are allowed to get resends, than as I said, it is not fair to those of us that do not cheat.
[edit] And, you cannot tell me that Ville Saari is not cheating by spoofing. One of his PCs is said to have 1337 GPUs!!!!!!!!!! He is CHEATING!!!!!!! [/edit]
[edit2] The last time I looked at his PC list that PC said it had 64 GPUs. He increased the spoofing variable value (or whatever it is called in the software) just so he could get the lions share of any tasks ready to send. [/edit2]
Have a great day! :)
Siran

. . I think there are some translation issues here. I think what Juan was trying to say is that hosts that are still active and contacting the servers regularly AND returning any work assigned to them promptly should be given the resends, rather than hosts that are sitting on large numbers of tasks and not returning them. I agree that still doesn't preclude hosts that have massive cached numbers if they are still returning work regularly but it is hard to cover all cases. Hence Richards call to those individuals to tend to their rigs. It is a moot point anyway because I do not foresee the Berkeley guys making such changes even at this stage of the project.

Stephen

:(

. . I think there are some translation issues here. I think what Juan was trying to say is that hosts that are still active and contacting the servers regularly AND returning any work assigned to them promptly should be given the resends, rather than hosts that are sitting on large numbers of tasks and not returning them. I agree that still doesn't preclude hosts that have massive cached numbers if they are still returning work regularly but it is hard to cover all cases. Hence Richards call to those individuals to tend to their rigs. It is a moot point anyway because I do not foresee the Berkeley guys making such changes even at this stage of the project.

Stephen

:(

Thanks Stephen you get the meaning. yes i have some problems with the translations, you all know i'm not a native english speaker.

About the GPU number... It's just a simple TAG as posted by Ian, nothing else, to prove this point please tell me any number (integer greater than 0 of course and under 32000) and i will switch mine for you to see nothing changes on my host. So please stop with this conspiracy theory.

---

Agreed, they should cancel on the server side all tasks still in the field and re-send them with much smaller deadlines. It would mean lost computation for the slowest of slow hosts (and the cheaters who consider they deserve more than the rest of us), but it would allow the last pending tasks to complete much faster. Adding a drastic limit per host is also better indeed, and should always have been added on top of the limit for CPU work and per GPU.

. . Well not ALL tasks in the field. but maybe all those older than a week or thereabouts. So all hosts with tasks less than a week old would still have a day or 2 to clear them. As for an overall limit on task numbers, believe it or not but there is such a limit in place. But obviously some have found a way to circumvent it. S@H is not perfect :( But certainly forcing resends on all older tasks would expedite the process of cleaning up the project.

Stephen

. .

Yeah, I have gotten 6 tasks since April 1st. I have one host still connected to SETI. When this host communicates with SETI the scheduler always does a 30 minute back off. I have seen where the RTS shows a dew tasks. I manually have BOINC communicate and I get nothing for the next 30 minutes. The next time I look at the servers, there are zero tasks RTS.

Siran

Are you sure, I checked mine this morning, according to the "all tasks" list I had downloaded, processed returned and had validated 18 tasks in a single batch less than an hour previously.

When I rechecked about 2 hours later, THREE hours after the original download they had already been removed from my lists.

Resends are getting processed and deleted at a very fast rate.

Hi Kevin,

Yep, I got 1 yesterday, 2 on the 11th, 1 on the 9th and 2 on the 3rd. 6 since the 1st. I even decided to go for CPU tasks instead of just GPU since I was having problems with Rosetta constantly restarting it's tasks and always running in high priority. An administrator said I should just set NNT and do something else since it will take weeks to months for a fix for the app to work properly and set checkpoints. Others were having that same problem. So I added CPU tasks here in hopes to get more tasks.

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

Agreed, they should cancel on the server side all tasks still in the field and re-send them with much smaller deadlines. It would mean lost computation for the slowest of slow hosts (and the cheaters who consider they deserve more than the rest of us), but it would allow the last pending tasks to complete much faster. Adding a drastic limit per host is also better indeed, and should always have been added on top of the limit for CPU work and per GPU.

+1

And spoofing up to 1337 GPUs!!!!!!!!!!, to be able to get as many tasks he wants, that' just ridicolous, and extreme CHEATING!!

. . And with the very low rate of resends available also meaningless. Even if he was reporting 2 million GPUs, he will not receive more tasks than are available, and that is small numbers indeed. Getting new work at this point is more like winning the lottery as has been remarked on several occasions. So one does have to wonder why he persists with such a pointless tactic at this time.

Stephen

< shrug >

Now it is crunching the WU with dateline of May 18 04:11:11 AM EST.
So it will report them more than a month in advance of the dateline!
So what is the problem with that?
I'm still waiting someone show me where this "hidden rule" who said a host can't has more than xxxx WU is write so i could read it.

One point of view is that the rules changed on 2 Mar 2020, when we were all given 4 weeks' notice that the project was switching from raw data processing to analysis. The extended "anyone can participate here - even the slowest tortoise can join the fun" deadlines became meaningless from that moment.

I, too, ran my computers up to the wire, and a little bit longer when the final tapes couldn't be split because the database was full. But my last remaining first-run task was reported complete at 2 Apr 2020, 20:01:50 UTC. I would have expected that we would all be participating in the clean-up by now, but instead we're debating (yet again) our freedoms to bend and interpret the rules however we like.

Please let's just agree to differ, finish whatever work we've got (without worrying how we got it), and wish Eric and David the best of luck with their analysis, report-writing, and publishing.

So, then 64 is also a tag, and all the extra WU's for a spoofed "64" client is just a conspiracy, and the thousands of "extra" WU's on such a host does not exist??

Please not put wrong words on my mouth, i just say in my case, the 1999 is a simple tag. Not related to the number of WU/GPU's or the way the host works. Nothing more or less. I just changed my to 2020 for you to see. If you go to E@H and Rossetta you will see the tag there is still 1999

---

Hi Juan,
I would not agree with that statement. That would not be fair to those of us that do NOT spoof (cheat) as Ville Saari does. He has over 70K STILL IN PROGRESS! Why allow him to get ALL the resends and not let any through to anyone else. I still have BOINC set to get anything I can from SETI, be it GPU or CPU tasks. If only active hosts are allowed to get resends, than as I said, it is not fair to those of us that do not cheat.
[edit] And, you cannot tell me that Ville Saari is not cheating by spoofing. One of his PCs is said to have 1337 GPUs!!!!!!!!!! He is CHEATING!!!!!!! [/edit]
[edit2] The last time I looked at his PC list that PC said it had 64 GPUs. He increased the spoofing variable value (or whatever it is called in the software) just so he could get the lions share of any tasks ready to send. [/edit2]
Have a great day! :)
Siran

. . I think there are some translation issues here. I think what Juan was trying to say is that hosts that are still active and contacting the servers regularly AND returning any work assigned to them promptly should be given the resends, rather than hosts that are sitting on large numbers of tasks and not returning them. I agree that still doesn't preclude hosts that have massive cached numbers if they are still returning work regularly but it is hard to cover all cases. Hence Richards call to those individuals to tend to their rigs. It is a moot point anyway because I do not foresee the Berkeley guys making such changes even at this stage of the project.

Stephen

:(

Hi Stephen,

Yep, Juan explained what he meant and I agree with him. :)

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

watching the mind melting rage of those who think this is cheating, and do not even understand the significance of the number 1337, is great entertainment for my morning coffee.
i can say with certainty, the GPU number you are seeing is a label only, and in no way affects how many tasks he's getting.

. . Puts hand up ...

. . I'll take a wild guess, Copernicus?

Stephen

? ?

My host is not actively crunching.

this is the second time he's made some kind of claim about his system not working on SETI when his task list clearly shows otherwise.
it takes just the smallest amount of effort to verify a claim like this, especially when it can be verified by anyone who can see your list of tasks. just look, it's not that hard.

. . OK, again I think this time you are being too literal. I am sure he was simply concerned that since he is not getting regular work to crunch he would not be included in Juan's definition of 'active cruncher', nothing more. I think we can all move past that confusion ...

Stephen

1337 = leet = elite

it's like internet-culture-pseudo-hacker lingo.

. . . D'OH!

Stephen

. . Not my world ...

Even better, and for no effort on behalf of the servers, he just detaches those two computers then deletes his SETI account. Doing so would show that he at least had some grain of thought for the rest of the SETI community.

. . Or perhaps if he simply set NNT and let it go at that ??

Stephen

? ?

watching the mind melting rage of those who think this is cheating, and do not even understand the significance of the number 1337, is great entertainment for my morning coffee.

i can say with certainty, the GPU number you are seeing is a label only, and in no way affects how many tasks he's getting.

Hi Ian,

Sorry to disagree with you here, but I do. Case in point, my current host:
When I look at my host list it show [2] NVIDIA GPUs. Guess what, I HAVE 2 NVIDIA GPUs in this host. Now, SETI upped the amount of tasks per device. I was getting 150 CPU tasks and 300 GPU tasks. Before they increased it Before I added the second GPU I was getting 150. Let me whip out my calculator... 1337 x 150 = 200550 and 70000 / 150 = 467 (rounded up 1). So I disagree and say that he got more tasks because he cheated with the spoofing. And the GPU number does have a bearing on the number of GPU tasks one gets. Since SETI was putting out less tasks over the past month or more, he got a lions share of what they did put out to be sent. If he hadn't cheated, my host could probably still be crunching and not just 1 per day or whatever it is now since 4/1. I have gotten 6 since April 1st, he still has over 70K! If he hadn't raised the number from 64 to 1337, he would not have as many today.

Have a great day! :)

Siran

I've seen the code in use, you haven't. So you don't know what you're talking about.

the number is just a tag and has no bearing on what number of tasks he's getting. that's handled by a different process in the code.

the OLD method to get more tasks was GPU spoofing, and was very rudimentary and acheivible by anyone who is able to copy/paste data in their coproc_info.xml file (Windows or Linux) and even has youtube videos explaining the process.

but that's not what Ville is doing. he's using a totally different and more sophisticated/elegant method.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Thanks Stephen you get the meaning. yes i have some problems with the translations, you all know i'm not a native english speaker.
About the GPU number... It's just a simple TAG as posted by Ian, nothing else, to prove this point please tell me any number (integer greater than 0 of course and under 32000) and i will switch mine for you to see nothing changes on my host. So please stop with this conspiracy theory.

So, then 64 is also a tag, and all the extra WU's for a spoofed "64" client is just a conspiracy, and the thousands of "extra" WU's on such a host does not exist??
Geeze, how the cheaters try.......
Nah, the spoofers should have been banned immeditely when it was discovered that they areCHEATING.
Basta!!!

. . Dude! Walk outside and roll around in the snow for a while (if there is any left), you need to cool off ...

Stephen
Sheesh ...

Yeah, I have gotten 6 tasks since April 1st. I have one host still connected to SETI. When this host communicates with SETI the scheduler always does a 30 minute back off. I have seen where the RTS shows a dew tasks. I manually have BOINC communicate and I get nothing for the next 30 minutes. The next time I look at the servers, there are zero tasks RTS.

Siran

Are you sure, I checked mine this morning, according to the "all tasks" list I had downloaded, processed returned and had validated 18 tasks in a single batch less than an hour previously.

When I rechecked about 2 hours later, THREE hours after the original download they had already been removed from my lists.

Resends are getting processed and deleted at a very fast rate.

Hi Kevin,

Yep, I got 1 yesterday, 2 on the 11th, 1 on the 9th and 2 on the 3rd. 6 since the 1st. I even decided to go for CPU tasks instead of just GPU since I was having problems with Rosetta constantly restarting it's tasks and always running in high priority. An administrator said I should just set NNT and do something else since it will take weeks to months for a fix for the app to work properly and set checkpoints. Others were having that same problem. So I added CPU tasks here in hopes to get more tasks.

Have a great day! :)

Siran

you can't base history on what you can see NOW. we already went through this several days ago. just look back in your own messages. you got like 50 something tasks on April 3rd, but they have mostly been purged from the task list already so you can't see them anymore. but that doesn't mean they were never there.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Hi Kevin,
Yep, I got 1 yesterday, 2 on the 11th, 1 on the 9th and 2 on the 3rd. 6 since the 1st. I even decided to go for CPU tasks instead of just GPU since I was having problems with Rosetta constantly restarting it's tasks and always running in high priority. An administrator said I should just set NNT and do something else since it will take weeks to months for a fix for the app to work properly and set checkpoints. Others were having that same problem. So I added CPU tasks here in hopes to get more tasks.
Have a great day! :)
Siran

. . Yeah, I have also had issues with Rosetta, it does not seem to play nice with E@H so I also stuck it on hold and went to WCG for the CPU work. It's been pretty seamless since then.

Stephen

:(

watching the mind melting rage of those who think this is cheating, and do not even understand the significance of the number 1337, is great entertainment for my morning coffee.
i can say with certainty, the GPU number you are seeing is a label only, and in no way affects how many tasks he's getting.

. . Puts hand up ...

. . I'll take a wild guess, Copernicus?

Stephen

? ?

Ah well, I could have said something about cheating, and Obsessive Crunching Disorder, but I won't, this time...... :-)

But I will say to the spoofers: Spare us from hackers lingo please....

don'7 b3 m4d jus7 cuz U Rn7 1337 enuf

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours