Posts by robsmith

1) Message boards : The Lounge : The Seti is Down Cafe (Message 92540)
Posted 5 hours ago by robsmith
Cheers - mine's a double one of them over there.....
2) Message boards : Questions and problems : What happened to "requested" and "granted" credits? (Message 92516)
Posted 3 days ago by robsmith
Credit award, both policy and magnitude, is not something done across the board by BOINC but is very project specific so why not ask over at SETI?
3) Message boards : The Lounge : The Seti is Down Cafe (Message 92488)
Posted 7 days ago by robsmith
Now that would so (in)convenient, but I see where you are coming from - there would be some who would be looking for a suitable sharp implement.....
4) Message boards : BOINC Manager : Boinc Manager won't run on Ubuntu 18.04 after update (Message 92486)
Posted 7 days ago by robsmith
Linux can take an age to boot after updates and filed boots....
In the past I've found the reciting the incantation "If you don't behave yourself you stupid lump of silicone I'll blow Linux away and install Windows CE" loudly and repeated a dozen times has worked ;-)
5) Message boards : The Lounge : The Seti is Down Cafe (Message 92485)
Posted 7 days ago by robsmith
.....and how many semi-retired computers are brought out of their rest-homes to serve the cause once more.
6) Message boards : The Lounge : Grumbles and Glory (Message 92468)
Posted 8 days ago by robsmith
Now I know where my additional weight came from ;-)
7) Message boards : The Lounge : The Seti is Down Cafe (Message 92439)
Posted 11 days ago by robsmith
Or Henry VIII's??
8) Message boards : The Lounge : The Seti is Down Cafe (Message 92412)
Posted 13 days ago by robsmith
Own up....
Who stole it yesterday????
9) Message boards : Questions and problems : Computing suspended even though laptop is plugged in (Message 92300)
Posted 26 days ago by robsmith
Two possible things.
One of my laptops has an "over eager" charger plugged in detector - it will report the charger not being plugged in even when it is (and it is charging!). My solution is to unplug the charger and plug it back in.
The other is that some change has been made in the sensor/charger system (within Windows) that reports "charger not plugged in" when actually it means "charger plugged in, but not charging".
10) Message boards : The Lounge : The Seti is Down Cafe (Message 92272)
Posted 28 days ago by robsmith
And now down - just as I was about to hit the reply button......
11) Message boards : The Lounge : Grumbles and Glory (Message 92253)
Posted 21 Jul 2019 by robsmith
Glory - just about finished downloading the >20k photos from my road trip. Most will go into a set of time-lapse videos. Some may even get posted....
12) Message boards : Questions and problems : rosetta@home tasks don't download (Message 92062)
Posted 5 Jul 2019 by robsmith
Assuming you have permission to run BOINC & BOINC-based applications on what would appear to be a "company owned and managed" computer you will have to ask for the Rossetta servers to be white listed on the corporate firewall and any associated security applications that may be running. It is not unknown for firewalls to allow an "initial sniff" of data through from an "unknown" source, then to pull the drawbridge hard and fast once a lot of data is being transferred.
13) Message boards : The Lounge : Grumbles and Glory (Message 92021)
Posted 2 Jul 2019 by robsmith
...or those PM "random strangers" with that self same question (all in the hope of getting the answer they want, not the one that will fix their problem)
14) Message boards : The Lounge : Seti -- questions (Message 91990)
Posted 30 Jun 2019 by robsmith
GPUUG are one of he most consistent donors to SETI@Home, and for that I am grateful.
The best place to see what you can do to help SETI@Home is to ask the question on the SETI@Home boards.
15) Message boards : The Lounge : Seti -- questions (Message 91988)
Posted 30 Jun 2019 by robsmith
A few answers:
1 - The BLP gets its data from a number of sources. The biggest amounts are from the Green Bank Telescope in the USA and the Parkes Telescope in Australia. As far as I'm aware they do not use the big telescope in China or the Arecibo telescope as neither is very good when it comes to collecting data from a single point in the sky as neither is "fully steerable". Also it collects some data from various optical telescopes for "Optical SETI".

2 - Undoubtedly as the instruments get more sensitive and analysis techniques improve more solar systems will be located.

3 - If you are asking about SETI@Home, and how many FLOPS per hour per user then the best place to ask the question is on the SETI@Home forum - this forum is for non-project specific questions.

4 - The 64 GTX1080 GPUs are used to gather the data from the telescope receivers. An individual is probably more than welcome to provide the money for another receiver interface computer, but the cost isn't only the $$$ for the GPUs, but also all the RF hardware, the software required to run it all, and the data interfaces to those performing the analysis.
Errr - where do you get the "40% of the processing we do" from. If one is talking about SETI@Home then 64 GPUs is a very small fraction of the total processing power of that project.
What do you mean by "Can we have faith in that?"?

David Anderson is, as far as I'm aware, doing OK, he's involved in a number of projects based at the UCB Space Sciences Lab.

And, as ever, money is tight for the majority of science projects.
16) Message boards : The Lounge : Grumbles and Glory (Message 91983)
Posted 29 Jun 2019 by robsmith
It hasn't rained at Glastonbury this year......yet, but there is still time ;-)
17) Message boards : Questions and problems : green screen death when using my GPU no message computer then crashes. (Message 91958)
Posted 27 Jun 2019 by robsmith
Byron - the last time I saw this sort of thing was a GPU that was suffering from old-age. Given that you K4200 is probably about 4 or five years old it may be heading for the Great GPU Graveyard :-(
18) Message boards : GPUs : I can not run calculations on the GPU. (Message 91947)
Posted 27 Jun 2019 by robsmith
The clue is in this line:
Running as a daemon (GPU computing disabled)

The Windows drivers used when running as a daemon do not (by default) support the use of an nVida GPU for computational use. There may be third-party drivers that do
The only solution I can think of (and admit that this may not be a route open to you) is to re-install BOINC in non-service mode, but then it will only run under the installing user account.
19) Message boards : Questions and problems : Control the tasks you have (Message 91942)
Posted 26 Jun 2019 by robsmith
Hi - there are a few ways to achieve what you are asking....

1a - You will have to "play" with your cache settings, these easist are accessed from your SETI account home - reduce the number of days cache and the additional days cache from the defaults - it will take some attempts to get it right, and give it at least double the cache size to settle down.

1b - This is different to the cache setting. In the advanced view of the BOINC GUI go to options, then "computing preferences", this will bring up a window with a number of tabs, one of which is "computing", on that tab you will see a line something like "use at most x% of CPUs" - reduce x% from its default of 100% to the fraction percentage that gives you the number of core you want to use (each core in use equates to one task being run concurrently) - for example if you have a four core machine and want to run two tasks at a time set x=50%.

2a - Not very easy, but you can set daily schedules for both computing and network activity using the "computing" and "network activity" tabs (above).
One trick is to setting a start time of 20:00 on day one, and an end time of 08:00 on day two will allow activity between 8pm and 8am, and nothing in between - which can be very useful for those with very expensive day-time power, or very high daytime temperatures....
If you use the "download" size limit it is worth knowing that a "normal" task is just over 700kb in size, and an AstroPulse task is about 7Mb. For both the returned file is somewhere between 20 and 35kb in size.

2b - If you happen to be looking when a task finishes you will see the icon blink for a few seconds. But generally using either the fully automatic, or scheduled automatic will look after things without you having to worry too much.

3 - There are a whole load of logging flags, again accessed via the advanced view options, but this time the "event log options", I think the only ones you want set is "task" (but the transfer one can be useful when looking to see what has come and gone...)

I hope this helps - as you can see there are lots of options, and I've only scraped the surface with the ones that don't need you to manually edit files (which can be a very risky operation)
20) Message boards : Projects : CPDN Project going offline tomorrow Wednesday 29th May (15:30 UK time) (Message 91718)
Posted 5 Jun 2019 by robsmith
Well - it looks as if it partially back on its feet after a week's slumber.
All the uploads that were stacked up have finished, but still some stuff to report.

Next 20

Copyright © 2019 University of California. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.