Just don't know where I am gonna end up............

[msattler]

9 years gone.......

Don't know where I am gonna end up with Seti............

The EVGA card's gotta go back on RMA, 'cuz they don't have any other answers as to whether there is a driver  problem or 'maybe' a problem with the card........
Such has been my lot in life lately.
The Setimods have given me 'the extended FU'  ban.......so I won't be able to discuss anything there for a couple of more weeks.............
And every day I get a little older and a little more tired of the BS.

After spending.....and this is a rather conservative estimate........$30,000.00 on Seti.......between toys to run it and the power to make it go......I am starting to falter in my enthusiasm.

I am now the #29 contributor in the world to the Seti project.......in terms of work done.

Wow.......that means a hell of a lot to me.........but maybe not so much to those who should care.

Maybe I am too proud of it.......some say users like me are just in it for the credits.   Well, FU too.......
What do you think got sooooooo many people involved and crunching in the first place?
Yes!!! I do it for the frickin' science!!  And have proclaimed so many times.......

But there has to be a point, at least in my view, where credit has to be given where due.

And I have received nothing but grief from the Setimods...........my language at times has been crude......but they could just delete the offending posts and move on instead of engaging me in argument. And then banning my ass for weeks on end.

Sorry if I tweaked some feathers here........

But the Seti forums are closed to me now.

If things do not change soon.
I am gonna have to completely reevaluate what I am doing, who I have been, and what I have wasted my life on for the last 9 years.

Sorry for the rant, but I had to tell SOMEBODY.....

Thanks for listening.........or not.

[Gecko_R7]

Mark, I moved this post to Discussion Forum 

You've always been known for your passion and dedication.  ANYONE that has crunched w/ you knows this much.
Passion is a double-edged sword however.

Yours is for the science and the personal accomplishment/challenge of "how much" of an impact can you make in your own way.
Not much different than the motivation of most current (& past) developers too. 

Seti is a self-motivating project in that HOPE & belief has to drive us more than the precedent of continuing success (like finding Primes).
10 years of searching "improvements" is a bit less emotionally tangible than a data-base of "definitive" non-natural & repeating signals for the effort. 

The project is also quite different than in 99' and IMO has a much-smaller "public face" and is far less interactive w/ user-base.
It's much more difficult for a Vol cruncher to "feel" a part-of and valued in the project today but this is more reflective of the current challenges and where the project actually is today.

We all know that Seti is just trying to continue operating w/ the minimal resources it has & a "shoe-string" budget would actually be an improvement over the continued financial challenges it faces.  It's therefore not difficult for me to understand that Berkeley's passion must be focussed on the science and operational priorities to maximize their paltry resources.

I believe there is ALOT of passion re: Seti, but passions are different between the project and the Vol cruncher base.
WHile the combined team of Project and Volunteers is working towards a goal, the motivations that drive us are different and not always easily understood, recognized or completely appreciated by folks on both sides of the coin.

Mark, there are only so many roles a Vol cruncher can fulfill. Not all roles are available to everyone, like coding/programming or having a Top 10 Power-Cruncher or Farm of hosts.  Some Mod boards, some "help" folks as constant northern stars to Noobs.  Some are just happy & content to hook-up their 5 year old P4 and contribute what they can.  You've been able to participate in a variety of ways & roles and on a larger scale. 

In the end, you kinda have to determine the best way you think you can fit that keeps the hobby fun for you.
If it's not fun any more, than why do it?  Life is too short. 
Maybe there are other projects that might get your motor revved-up more w/ a break from Seti for a bit? 

You can take or leave my comments.  Only trying to help. 

[msattler]

No problem, my friend.
I will keep searching.
The skies, and my soul, for the answers.

If my crunchers all go dark, it won't take but a few weeks or months for everybody to forget who 'he' was......
My successes, and my failures, will soon be forgotten.

I once thought I could make a difference.......
Now I have been banned on every forum from here to the Milky Way for speaking my mind.  Not much of a thing to be remembered by.

Build fire trucks each night, then come home and tend to the rigs......don't get much sleep.
Drink too much on the weekends, cry a lot.
Love my kitties. Pay the bills.  And go back to bed.
Restart cycle.
9 years. 

It's been a long 9 years.

It makes me a bit sad to see what has become of 'my' Seti project.

I know they are strapped for resources.......both in terms of hardware and the people to manage it.
But it seems sometimes that they mismanage the resources they have.
I just took my hand at a Cuda installation.........and I see why it has been such a disaster.
The damned VLARkill should have been outlawed from the get-go.  You crunch what you are given, and that's that.  If you don't like the way it works, tough luck.  My rig is running that way for a short while now, because I have technical problems getting the card to run with the current drivers, and is in an experimental stage.........
But the card is getting yanked to RMA back to EVGA, and will not continue in that mode.

VLARkill is cherry picking, pure and simple.......
Unless it can automatically return the work within the host to the CPU and not send it back to Seti for some other 'sucker' to crunch, it bites the big one. And only contributes to current Seti woes.

If I get the card back and can bring it back up without trashing the work that Seti sends me, I will continue with it.  Otherwise, it goes into the trash bin.

As usual, these are my 'not so humble' opinions.....

[Geek@Play]

Mark.............

I have crunched along side you on Seti for many years now.  I consider your renting out your farm to the highest bidder your most creative and best contribution to date.  It shows dedication and commitment to the project that certainly goes unrewarded by the people at Berkeley.  No other person has even offered the same deal up for bid.

I too have been tireing lately of the lack of resources at Berkeley, and I cannot dodge the feeling that it will never improve.  The inability of the project to get even 1% of the crunchers to donate financially is especially dissapointing to me.

As for me, I am running down my cache and will be going on a vacation Monday.  It is doubtful that I will be able to upload the work even after it is finished crunching.  During my 10 days I will evaluate what SETI gives to me in return for my commitment to SETI and whether I should continue.

I suspect that I will continue on mainly because of the developers here at Lunatics.  These guy's here work hard to make our SETI experience a more meaningful and enjoyable one.  If you are looking to Berkeley for justification to crunch SETI you are looking in the wrong place!

Just my 2 cents worth for you...............

[msattler]

Let's not give up just yet....................

But when a project cannot solidly service the contributors it has, and then sends out mass emails begging for folks to 'come back' to the project........ya gotta wonder.........what are they thinking?

Get yer f'ing ducks in a row.......and THEN ask for more crunchers......sheeesh.

My persistence usually outlasts my anger..........so I suspect I shall be around until the last dog is hung.

But that does not stop me from wishing that things could be other than what they are now.

My rigs all have Seti work currently.....even the Cuda-bastardised one.

Some time this weekend, I gotta pull the Cuda card........and will probably trash the crappy Boinc installation and go back to an older one.........hope I can find the 'hidden' Boinc folders that it put on my rig......what a f'n joke.

Suppose that's what I get for living in the stone age........
I learned a long time ago......if it is working......don't F with it.......LOL.  The Boinc devs don't live by that credo.

[Jason G]

...
VLARkill is cherry picking, pure and simple.......
Unless it can automatically return the work within the host to the CPU and not send it back to Seti for some other 'sucker' to crunch, it bites the big one. And only contributes to current Seti woes.
...

  That's certainly one opinion that many might agree with, and one that has been tossed about here lately.  Mine happens to be that sending work that disables user machines for normal use with flawed application implementation, drives away users, potential donors and makes for a very poor overall picture.

What needs fixing is the application, and as soon as we can get to that without all the peripheral politics continually obstructing progress, then the sooner the issue is moot.  It sure as heck doesn't look like anyone else is going to be fixing it but us.

In any case, differences of opinion aside, all the best Mark, and good luck with getting things sorted.

Regards, Jason

[Raistmer]

VLARkill is cherry picking, pure and simple.......
Unless it can automatically return the work within the host to the CPU

Use Rebranding tool 1.9 if you have problems with that. It's pretty simple in use, even for novice users. And we talk about 9-year of experience here.

[msattler]

...
VLARkill is cherry picking, pure and simple.......
Unless it can automatically return the work within the host to the CPU and not send it back to Seti for some other 'sucker' to crunch, it bites the big one. And only contributes to current Seti woes.
...

  That's certainly one opinion that many might agree with, and one that has been tossed about here lately.  Mine happens to be that sending work that disables user machines for normal use with flawed application implementation, drives away users, potential donors and makes for a very poor overall picture.

What needs fixing is the application, and as soon as we can get to that without all the peripheral politics continually obstructing progress, then the sooner the issue is moot.  It sure as heck doesn't look like anyone else is going to be fixing it but us.

In any case, differences of opinion aside, all the best Mark, and good luck with getting things sorted.

Regards, Jason

Like I think you said............
It's not the work that is flawed..........it's how it is worked on.

You folks here have done sooooooooooooooooo much more than Seti could ever give you credit for.
And I don't mean to diss Raistmer's efforts.........it's just that the work-around is a bad thing for the project.

The only solution is to find a way to crunch it properly.........which I know you are working your hearts out on it.

Although in the real world 'catch and release' is a good thing......in Seti crunching it poisons the pond.

[msattler]

VLARkill is cherry picking, pure and simple.......
Unless it can automatically return the work within the host to the CPU

Use Rebranding tool 1.9 if you have problems with that. It's pretty simple in use, even for novice users. And we talk about 9-year of experience here.

If and when I get the Cuda card up and running with the current drivers I will try that, or the non-VLARkill app that is available.
Your work is not unappreciated, Raistmer........you have undoubtedly spent countless hours on it.

I just dislike the premise that you can toss out what you don't like and take the rest.
That has never been the Seti way.

[Jason G]

When all's said and done Mark, rest assured that none of us 'like' it, and I personally will be happiest when the necessity for any kind of workaround is gone, as that means our fixes can be introduced to stock.  Until then, there are thankfully some very effective workarounds that take some end-user effort.  Let's hope they're enough to cover the gap until it can be fixed properly, without all this bickering.

 

[Geek@Play]

I am currently running the MB_6.08_CUDA_V12_noKill_FPLim2048.exe app in cunjunction with the 1.9 rebranding that runs 2 times each day.  This setup works fine.

[msattler]

I am currently running the MB_6.08_CUDA_V12_noKill_FPLim2048.exe app in cunjunction with the 1.9 rebranding that runs 2 times each day.  This setup works fine.

From what I understand.....and correct me if I am wrong.......
I cannot run that app right now because my friggin' Cuda card won't run on anything more current than the 182.50 drivers.......
That's why it's going back to EVGA for replacement.

The 6.6.36 Boinc is going back for replacement also........but that is another whole set of issues.

[Gecko_R7]

VLARkill is a matter of perspective which is one of the reasons it's debated.
It can have a positive impact on production by shifting WUs to a CPU host that can process more efficiently than a GPU that gets bogged -down by it.  End result is better use of Vol resources for MORE crunching in the end.  Totally valid use as long as there's sufficient project bandwidth available to handle the number of hosts using it.  I think the general consensus here is the project's current bandwidth issues are related more to a convergence of several other factors and not primarily or even significantly as the result of VLARkill.

The other side of the debate is that it does benefit someone motivated by greater reward via cherry-picking.
However, while the visible symptom looks similar on the surface, in this case there's a valid medical reason.   (fubar stock CUDA application, since project doesn't discriminate WU distribution by processor type).

This type of WU selection is different than if it occurred in the context of normal MB crunching via CPU application.
In that case, a person wanting ONLY mid-AR WUs to optimize their credit/hr ratio & forcing greater bandwidth use so others can complete less rewarding AR WUs would be justifiably wrong!  

I personally prefer rebranding as a better bandaid, BUT also accept VLARkill as a valid & acceptable (not ideal) workaround as well.

 

[Jason G]

From what I understand.....and correct me if I am wrong.......
I cannot run that app right now because my friggin' Cuda card won't run on anything more current than the 182.50 drivers.......
That's why it's going back to EVGA for replacement.

  Quite right.  nVidia chipset based cards should work with nVidia Drivers.  These things rarely come in 'ones', so you'll likely get a revised version with obvious physical changes, or maybe even a model up.  Only problem is they'll dump the entire existing stock on the market here in Oz, as per usual 

The 6.6.36 Boinc is going back for replacement also........but that is another whole set of issues.

 

[Jason G]

... as a better bandaid....

  That's it! ... More bandaids!