Posts by Richard Haselgrove

1) Message boards : Cruncher's Corner : Parallella, Raspberry Pi, FPGA & All That Stuff (Message 132930)
Posted 4 days ago by Richard Haselgrove
And it should be available in a complete retail build next month (US release):

http://www.misco.co.uk/blog/news/02193/acer-launches-first-tegra-k1-powered-chromebook
http://www.techradar.com/news/mobile-computing/laptops/acer-chromebook-13-finds-luck-with-tegra-k1-processor-long-battery-life-1260731
2) Message boards : Technical News : Albert@Home down due to migration to Drupal (Message 132734)
Posted 14 days ago by Richard Haselgrove
I didn't see anybody else's profile, but I did see everyone's email address as a tooltip on mouseover the silhouette image by their posts.
3) Message boards : Problems and Bug Reports : change language (Message 132641)
Posted 18 days ago by Richard Haselgrove
On the Home Page under the header Community there is a option Language.

I think that just controls the language used to display this website - the question was about BOINC Manager.

In BOINC Manager, either simple or advanced view, choose the third item on the menu bar - 'Tools' in English, if you can recognise that in your current display language.

Click the second item on that menu - 'Options...' in English.

The first drop-down list on the Options dialog is the language selector. You'll need to restart the BOINC Manager for the new settings to be loaded.
4) Message boards : Technical News : Albert@Home down due to migration to Drupal (Message 132576)
Posted 22 days ago by Richard Haselgrove
Any news about Albert, did it get lost in the crossfire?

All I see there is You don't have permission to access /index.php on this server.

Details posted at http://boinc.berkeley.edu/dev/forum_thread.php?id=9517&postid=55135
5) Message boards : Problems and Bug Reports : Validate error - What this really means! (Message 132484)
Posted 25 days ago by Richard Haselgrove
Agreed, all GTX 6xx and 7xx cards are reported with an unknown speed and number of cores. The cards were manufactured after this app was built, and the app can't predict the future.
6) Message boards : Cruncher's Corner : a new (?) netgear router. (Message 132441)
Posted 27 days ago by Richard Haselgrove


Your picture is interesting, I have been only swapping the middle two pairs of wires, not every pair of wires. But as you said as long as both ends of the cable are the same the router will just send the signal down the wire to the next connection. I just did a speed test on my laptop and while it is running 'g' it is only getting 25/18. My desktop, which is hard wired, is getting 43/34, so my wireless IS slower. Time to run more wires I guess!!

What's more, it's a standard - specifically, TIA/EIA-568 - and I've always found it's better to follow standards if it costs no more than doing it any-which-way.

Specifically, Mike has illustrated T568A (as it's usually known): I've always found that to be commoner in telephone systems. Computer network wiring is more commonly made up to T568B - which is electrically the same, but with the white/orange - orange/white pair connected to the top pins 1 and 2, and the white/green - green/white pair connected to the symmetrical pins 3 and 6.

I too have done the sort of networking which starts with an electric drill, a 18" masonry bit, and a set of stepladders.....

... and I've still got the boxes of Cat 5 cable in my cellar to this day.

Note that Cat 5 cable can be bought in two forms: solid core and stranded. The stuff which comes in 1,000-foot boxes is almost always solid core: it's designed for permanent, fixed, infrastructure situations. It's designed for connection to patch panels or the faceplates on wall outlets, using an IDC 'punchdown' tool. That's the sort of setup I'd use for your project: run the stiff solid-core cable as far as the skirting board (baseboard? floor moulding? kick board?), and terminate it with a fixed socket.

Those little RJ45 plugs with the catch which breaks off are much easier to fit on the softer, more flexible, stranded cable: use that to make up short-length patch cables to connect your equipment to the wall socket. I used to make those up too, until I got bored and found that if you purchase wisely, you can buy them ready-made and certified for much the same price as the cost of the parts. Just don't buy them in high-street computer shops, where the mark-up can be horrendous.

Personally, I like the ones with an integral strain relief like this:



- doesn't get in the way of the latch for unplugging, but does stop the wires working loose in the connector. I usually break the latches off by tripping over the cable and ripping it forcefully out of the socket: no sleeve or boot is ever going to stop it breaking under those conditions.
7) Message boards : Wish List : Improved handling of low speed internet links (Message 132127)
Posted 40 days ago by Richard Haselgrove
Could it be changed to allow the user to set a limit of one file transfer at a time, until this limit is no longer desired?

That exists already. Using a cc_config.xml file (see Client configuration), you can set (separately)

<max_file_xfers>N</max_file_xfers>
Maximum number of simultaneous file transfers (default 8).
<max_file_xfers_per_project>N</max_file_xfers_per_project>
Maximum number of simultaneous file transfers per project (default 2).
8) Message boards : Problems and Bug Reports : Work Running 100% Of Time Despite Settings? (Message 132000)
Posted 46 days ago by Richard Haselgrove
Or, from the official ClientSchedOctTen document,

This will tend to get large (max-min) clumps of work for a single project, and variety will be lower than the current policy.
9) Message boards : Problems and Bug Reports : question about last checkpoint (Message 131829)
Posted 52 days ago by Richard Haselgrove
The last batch I saw like that checkpointed 11 times during the run, or after each 9.09% of progress.

With the 'recommended' 7.2.42 version of BOINC that you are using, the early stages of progress (before the first checkpoint) are not accurately recorded: they are simply an estimate based on time spent so far (earlier versions of BOINC displayed no progress at all during this phase, which was even worse).

If you can, leave the computer running until the progress% shows exactly 9.090% (it should then stop changing, until a second big jump to 18.181%). You should then be able to look up the CPU time at the first checkpoint: you would need to leave the computer running for at least that long in each session, to make progress with these tasks. If that's not possible, you might be better off choosing a different type of Einstein or BOINC work.
10) Message boards : Cruncher's Corner : BOINC weirdness again (Message 131743)
Posted 56 days ago by Richard Haselgrove
That may have been me. They are *started* in FIFO order.

There is an override, in which tasks are marked "high priority" in BOINC Manager: we tend to call it 'EDF' for Earliest Deadline First on project message boards. That should have kicked in here, but clearly didn't.


Next 10

Home · Your account · Message boards

This material is based upon work supported by the National Science Foundation (NSF) under Grants PHY-1104902, PHY-1104617 and PHY-1105572 and by the Max Planck Gesellschaft (MPG). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the investigators and do not necessarily reflect the views of the NSF or the MPG.

Copyright © 2014 Bruce Allen