Topics

If ever such column will be added in the Tasks tab, the displayed value could be the difference ("CPU time at last checkpoint" - "CPU time"). For strategic games like "do I shut down the client now, or should I wait a bit more..."
(BOINC client devs are not willing to add something like "Pause each running app after reaching a checkpoint".)

The BRs is again trying to mix WUs from different projects, at least when trying to make a copy of some files, for example:

Code Select Expand

23 July 2010 - 14:28:23 Unable to make copy: C:\ProgramData\BOINC\projects\setiathome.berkeley.edu\18no09aj.30466.7843.3.13.138 -> C:\Users\PeterS\AppData\Roaming\eFMer\BoincScheduler\capture\capture-wu.xml - Error: 2
23 July 2010 - 14:28:23 Systém nemôže nájsť zadaný súbor.

23 July 2010 - 14:28:23 wu18no09aj.30466.7843.3.13.138 - Read: -1.000000, Calculated: 157794699338661.000000, Ratio: -0.000000

Note that the 18no09aj.30466.7843.3.13.138 is a Seti Beta WU task on my host

More precisely, it tries to find 4 different Seti Beta WUs in the Seti project folder.

Again sending relevant data off-list...

My BoincMonitor v0.1 has been installed together with some older BT version (no idea whether the recent BT v0.65 installer still contained it). Could it add BM's shortcut into Start menu inside EFMER Boinc Tasks (where it actually and logically belongs) if it is (still planned to be) installed together with BT? Otherwise, I've possibly missed any download link for it?
Maybe BR's link could also go there in the future?

I've got a red ERROR: check the log!

Code Select Expand

---------- Debug begin
log_all_tasks = 1
rsc_fpops = 1
---------- Debug end
19 July 2010 - 19:40:22 BoincRescheduler V: 1.0

19 July 2010 - 19:47:37  true_angle_range: 0.44330727921088 beam_width: 0.0500000007 sample_rate: 9765.625 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
19 July 2010 - 19:47:37 wu12my10aa.26352.2112.14.10.25 - Read: 158430379595272.620000, Calculated: 158430379595272.000000, Ratio: 1.000000
19 July 2010 - 19:47:37  true_angle_range: 0.4431461863789 beam_width: 0.0500000007 sample_rate: 9765.625 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
19 July 2010 - 19:47:37 wu14my10aa.12824.14791.11.10.125 - Read: 158454586579697.500000, Calculated: 158454586579698.000000, Ratio: 1.000000
19 July 2010 - 19:47:37  true_angle_range: 0.44302007965294 beam_width: 0.0500000007 sample_rate: 9765.625 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
19 July 2010 - 19:47:37 wu15my10ab.28922.2112.6.10.177 - Read: 158473548581197.120000, Calculated: 158473548581197.000000, Ratio: 1.000000
19 July 2010 - 19:47:37  true_angle_range: 0.0097986224112693 beam_width: 0.0500000007 sample_rate: 9765.625 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
19 July 2010 - 19:47:37 wu01jn10aa.25560.80221.16.10.77 - Read: 160720000000000.000000, Calculated: 160720000000000.000000, Ratio: 1.000000
19 July 2010 - 19:47:37  true_angle_range: 0.42121945709895 beam_width: 0.0500000007 sample_rate: 9765.625 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
19 July 2010 - 19:47:37 wu15my10ab.7870.49530.8.10.103 - Read: 161922232675840.910000, Calculated: 161922232675841.000000, Ratio: 1.000000
19 July 2010 - 19:47:37 wu18no09aj.30466.7843.3.13.138 - Read: 161922232675840.910000, Calculated: 157794699338661.000000, Ratio: 1.026158
19 July 2010 - 19:47:37 wu18no09aj.30466.8661.3.13.235 - Read: 161922232675840.910000, Calculated: 157829454578104.000000, Ratio: 1.025932
19 July 2010 - 19:47:37 wu18no09aj.30466.9479.3.13.35 - Read: 161922232675840.910000, Calculated: 157835637345957.000000, Ratio: 1.025891
19 July 2010 - 19:47:37 wu18no09aj.30466.9479.3.13.208 - Read: 161922232675840.910000, Calculated: 157835637345957.000000, Ratio: 1.025891
19 July 2010 - 19:47:37 wu18no09aj.30466.10297.3.13.76 - Read: 161922232675840.910000, Calculated: 157829423360584.000000, Ratio: 1.025932
19 July 2010 - 19:47:37 wu18no09aj.30466.11524.3.13.76 - Read: 161922232675840.910000, Calculated: 157829557458494.000000, Ratio: 1.025931
19 July 2010 - 19:47:37 ERROR: Cpu and GPU countMore than one Plan Class: cuda23 ,cuda
19 July 2010 - 19:47:37 ERROR: Scan completed with an error.


I have 6 Seti Beta tasks (18no09aj.30466.* - 5 cuda + 1 CPU), listed in the log; and additionally 6 Seti tasks (**my10**.* and 01nj10**.* - 5 cuda23 and 1 CPU), not (yet?) mentioned in the log.
Was the log output stopped prematurely? Did BR incorrectly mix tasks, belonging to different projects?

[MODIFIED:]

I've started the currently running BT v0.63 on 2.7.2010 at 19:11:18, after rebooting my machine. At around the same time I've started the BOINC 6.10.57 client, first messages (1, 2, 3, ...) displayed in BT are from 2.7.2010 19:11:14.

Code Select Expand


1 02.07.2010 19:11:14 Unrecognized tag in cc_config.xml: <heartbeat_debug>
2 02.07.2010 19:11:14 Unrecognized tag in cc_config.xml: <force_auth_off>
3 02.07.2010 19:11:15 Starting BOINC client version 6.10.57 for windows_x86_64
4 02.07.2010 19:11:15 Config: don't compute while VCBuildHelper.exe__ is running
5 02.07.2010 19:11:15 Config: don't compute while link.exe__ is running
6 02.07.2010 19:11:15 Config: GUI RPC allowed from:


On 7.7.2010 I've downloaded the 6.10.58 client, reinstalled it and after a maybe 1/2 hour delay I've started it at 11:12:18 and it still runs. BT was still running, I was observing the tasks'behavior prior to and after the installation (client alpha testing blah blah blah...).

Just now I've noticed that there is a large gap around the client's restart point, between messages Nr. 12280 and 12281:

Code Select Expand


12274 ralph@home 06.07.2010 3:42:14 [task_debug] result celldiv_zipA_2ozf_ProteinInterfaceDesign_2Jul2010_14839_4_0 checkpointed
12275 Quake-Catcher Network 06.07.2010 3:43:01 [sched_op_debug] Starting scheduler request
12276 Quake-Catcher Network 06.07.2010 3:43:01 Sending scheduler request: To send trickle-up message.
12277 Quake-Catcher Network 06.07.2010 3:43:01 Not reporting or requesting tasks
12278 Quake-Catcher Network 06.07.2010 3:43:01 [sched_op_debug] CPU work request: 0.00 seconds; 0.00 CPUs
12279 Quake-Catcher Network 06.07.2010 3:43:01 [sched_op_debug] NVIDIA GPU work request: 0.00 seconds; 0.00 GPUs
12280 Quake-Catcher Network 06.07.2010 3:43:05 Scheduler request completed
---gap--- missing messages ---
12281 climateprediction.net 09.07.2010 12:45:46 [task_debug] result hadsm3dhet2_k4g1_006615411_8 checkpointed
12282 Einstein@Home 09.07.2010 12:48:04 [task_debug] result p2030_53650_82622_0056_G62.31+00.80.N_6.dm_620_2 checkpointed
12283 Quake-Catcher Network 09.07.2010 12:48:59 [task_debug] result qcnk_sc300_sta200_024526_0 checkpointed
12284 WUProp@Home 09.07.2010 12:49:21 [task_debug] result wu_1274387826_252720_0 checkpointed

Could it be that BT somehow matched the IDs of the newer client's messages to the IDs of the older client? (I remember that each message contains its unique ID, which is started from 1 on the client start.)

The recent client's first messages (lines 1-6) are these, as seen by BOINC Manager:

Code Select Expand


07.07.2010 11:12:18 Unrecognized tag in cc_config.xml: <heartbeat_debug>
07.07.2010 11:12:18 Unrecognized tag in cc_config.xml: <force_auth_off>
07.07.2010 11:12:19 Starting BOINC client version 6.10.58 for windows_x86_64
07.07.2010 11:12:19 Config: don't compute while VCBuildHelper.exe__ is running
07.07.2010 11:12:19 Config: don't compute while link.exe__ is running
07.07.2010 11:12:19 Config: GUI RPC allowed from:


and a few messages around the end of the gap (lines 12278-12284):

Code Select Expand


09.07.2010 12:44:09 Quake-Catcher Network Scheduler request completed
09.07.2010 12:44:09 Quake-Catcher Network [sched_op_debug] Server version 609
09.07.2010 12:44:21 WUProp@Home [task_debug] result wu_1274387826_252720_0 checkpointed
09.07.2010 12:45:46 climateprediction.net [task_debug] result hadsm3dhet2_k4g1_006615411_8 checkpointed
09.07.2010 12:48:04 Einstein@Home [task_debug] result p2030_53650_82622_0056_G62.31+00.80.N_6.dm_620_2 checkpointed
09.07.2010 12:48:59 Quake-Catcher Network [task_debug] result qcnk_sc300_sta200_024526_0 checkpointed
09.07.2010 12:49:21 WUProp@Home [task_debug] result wu_1274387826_252720_0 checkpointed


Just for note, the most recent messages are not far from the break point:

Code Select Expand


12695 WUProp@Home 09.07.2010 14:15:48 Starting task wu_1274387826_255522_0 using data_collect version 132
12696 WUProp@Home 09.07.2010 14:15:50 [task_debug] result wu_1274387826_255522_0 checkpointed
12697 Quake-Catcher Network 09.07.2010 14:16:12 [task_debug] result qcnk_sc300_sta000_024948_0 checkpointed
12698 Einstein@Home 09.07.2010 14:18:30 [task_debug] result p2030_53650_82622_0056_G62.31+00.80.N_6.dm_620_2 checkpointed


Another note: out of curiosity, I've counted the messages, displayed in the BOINC Manager. The line numbers since the gap until now are exactly matching the BT's message Nr's!!

MODIFIED:
At around Approx. 30 minutes after the gap end time (possibly only a coincidence?), I've just unlocked my locked Windows session, the computer was running and was not sleeped or hibernated, just the session was locked since yesterday evening. I've been interacting with BT, just not looked exactly at the messages, so have no idea what was there...

In the BoincMonitor-02-07-2010.log I've noticed two occurrences of "Unable to open file: C:\ProgramData\BOINC\client_state.xml" at least 10 hours apart (overnight and at afternoon). For whatever reason (I have no idea, I'm running no antivirus or similar).

Does the BoincMonitor retry it and attempt to write again a bit later?

Code Select Expand

08 July 2010 - 12:33:20 BoincRescheduler V: 0.1
08 July 2010 - 12:33:48 ERROR: Copy state -----> Unable to make copy: C:\ProgramData\BOINC\client_state.xml -> C:\Users\{myself}\AppData\Roaming\eFMer\BoincScheduler\capture\capture-state.xmlError: 5
08 July 2010 - 12:33:48 ERROR: Scan completed with an error.
I guess the reason is it could not access C:\ProgramData\BOINC\ folder tree.
(BOINC is installed by the local admin. I've added {myself} (a plain user) to the boinc-admins group yesterday, but did not re-login since, so nearly all my running processes (including the Explorer shell and its child processes) do not yet belong to the boinc-admins group and are not yet granted access there. (Although they already belong to group and I wonder, why-the-hell is it not enough for them to get there inside Because boinc-users are allowed to open/read/execute the folder tree and files ))

But the ERROR:5 message is not very helpful at all

After a "proper" relaunch, it produced a green result with log:

Code Select Expand

08 July 2010 - 12:52:57 BoincRescheduler V: 0.1

08 July 2010 - 12:53:23  true_angle_range:  -----> 0.39154643756078 beam_width: 0.0500000007 sample_rate: 155 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:23 wu -----> 10fe07ah.28266.17250.9.10.155 - Read: 107827855394641.870000, Calculated: 167233356108413.000000, Ratio: 1.550929
08 July 2010 - 12:53:23  true_angle_range:  -----> 0.01866315919762 beam_width: 0.0500000007 sample_rate: 69 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:23 wu -----> 04no09ab.7489.9065.9.10.69 - Read: 160720000000000.000000, Calculated: 160720000000000.000000, Ratio: 1.000000
08 July 2010 - 12:53:23  true_angle_range:  -----> 0.01866315919762 beam_width: 0.0500000007 sample_rate: 121 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:23 wu -----> 04no09ab.7489.9065.9.10.121 - Read: 160720000000000.000000, Calculated: 160720000000000.000000, Ratio: 1.000000
08 July 2010 - 12:53:23  true_angle_range:  -----> 0.01866315919762 beam_width: 0.0500000007 sample_rate: 75 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:23 wu -----> 04no09ab.7489.9065.9.10.75 - Read: 160720000000000.000000, Calculated: 160720000000000.000000, Ratio: 1.000000
08 July 2010 - 12:53:23  true_angle_range:  -----> 1.9850551776336 beam_width: 0.0500000007 sample_rate: 43 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:23 wu -----> 05no09ad.21268.10701.5.10.43 - Read: 98413285822852.625000, Calculated: 47560000000000.000000, Ratio: 0.483268
08 July 2010 - 12:53:24  true_angle_range:  -----> 1.4502373666909 beam_width: 0.0500000007 sample_rate: 51 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:24 wu -----> 04no09ab.21621.11928.10.10.51 - Read: 99266378211918.406000, Calculated: 47560000000000.000000, Ratio: 0.479115
08 July 2010 - 12:53:24  true_angle_range:  -----> 0.44813676185346 beam_width: 0.0500000007 sample_rate: 100 nsamples: 1048576 pot_min_slew: 0.00209999993 pot_max_slew: 0.0104999999 chirp_resolution: 0.1665
08 July 2010 - 12:53:24 wu -----> 05no09ah.17863.2112.4.10.100 - Read: 106346871916974.250000, Calculated: 157712748037692.000000, Ratio: 1.483003
08 July 2010 - 12:53:24 wu -----> 23mr10aa.15637.12746.4.13.156 - Read: 106346871916974.250000, Calculated: 159701421476571.000000, Ratio: 1.501703

08 July 2010 - 12:53:24 Scan completed -----------------------------
08 July 2010 - 12:53:24 Application setiathome_enhanced found, Version: 609 , Plann class: cuda23
08 July 2010 - 12:53:24 Application setiathome_enhanced found, Version: 603
08 July 2010 - 12:53:24 Application setiathome_enhanced found, Version: 608 , Plann class: cuda
08 July 2010 - 12:53:24 No results found matching: setiathome_enhanced
08 July 2010 - 12:53:24 Found: 0 Cpu - 0 Gpu (VLAR: 0 Cpu - 0 Gpu) (VHAR: 0 Cpu - 0 Gpu) (Unknown: 0 Cpu - 0 GPU)

A bug:

• "If GPU < then [ 0    ] move [ 0   ] units from CPU to GPU" - I guess it ought to be ..."GPU < [ 0    ] then move"...

A wish:

• Maybe BR might briefly check whether BOINC client is running upon pressing the "Run" button?

There is one thing I wanted to ask a long time ago... observed on both Win XP SP3 and Win 7. When TTh is actively throttling BOINC applications, the process causes a lot of page faults. E.g. my currently not even 10 days running TTh has already had 1770 mil. page faults (2100/sec over the whole time, but has been throttling just a part of the time), I can often see up to 200 000 page faults during a second or two, followed by a silent pause.

I believe these are not the "hard page faults", caused by the swapping to/from HDD because of the lack of RAM, but "soft page faults", possibly caused by accessing memory protection boundaries, some types of system calls, etc. Hard to tell without any possibility to see symbolic call stacks. But the net effect is the application uses maybe 3% of its CPU life in user mode and 97% in the system calls. 1-2 CPU hours daily.

Any way to find out, what exactly is causing this? Could solving this allow lowering TTh's CPU footprint?

BTW, a long time ago I've already reported the Process Explorer's and Process Hacker's dev teams, that their applications have issues with such pagefaulting programs, because when their counters overflow 31 bits, the values are either displayed negative, or incorrectly positive, or even the broken displaying routines get blocked

[this]

• In History, while it is possible to get finished tasks belonging to a particular project together using the sorting, it is necessary to break existing sorting and fix it afterwards. A functionality similar to "Messages: - Right click menu, show only selected project" would be useful.
• For existing implementation in Messages: What will happen if the number of projects will exceed the number of lines, which will fit on-screen in the selection popup menu? Will the list be scrollable? If not (and anyway, for user-friendliness, because it is a context menu and it is necessary to look for the current line's project name somewhere deep down in the list), It'd be good to have one more entry between [ All ] and the [ List of projecs ], which would be named like [ Selected project ], meaning "Show only this project". Or directly the selected line's project name, sorted into the list's beginning. Maybe add one more separation line under it?
• For History, in addition to "Show only this project", also a "Show only this application" entry. (A list of applications would be probably overkill, although I do remember BoincView was doing it this way.)

[AND]

I have 2 computers defined (localhost+extern). While starting, BT was trying to connect to them, Computers/Status for the extern one was showing "Not connected, Connected". After a while (and possibly a couple of unsuccessful connection retries - the host is in a different network) the text changed to "Not connected" AND "1.80" appeared in its TThrottle column!
Note that the machine does not run TTh and is not reachable at all (switched off and in a different city part), so it was a very very wild guess by BT
I guess the connection status should be something like "Not connected, Connecting..."?

On the Programs tab, TTh lists "matching programs (processes)" and sorts them into Cpu: and Gpu: categories. According to what key? Observed CPU usage?

Because neither of the following apps is a GPGPU app (as you possibly know), just a (little-)CPU(-consumption) applications or wrappers (observing now on TTh 1.76):

Code Select Expand

Gpu: data_collect_1.25_windows_intelx86__nci.exe, PID: 12696, Thready: 3
Gpu: famous_6.07_windows_intelx86.exe, PID: 13920, Thready: 3
Gpu: famous_6.07_windows_intelx86.exe, PID: 8992, Thready: 3
Gpu: qcn_5.62_windows_intelx86__nci.exe, PID: 13404, Thready: 6   (added later)
Gpu: hadsm3_6.07_windows_intelx86.exe, PID: 7372, Thready: 3
Gpu: hadsm3_6.07_windows_intelx86.exe, PID: 7004, Thready: 3

I was little surprised not to see the QCN app here, some week ago (observed on TTh 1.74) was showing also:

Code Select Expand

Gpu: qcn_5.62_windows_intelx86__nci.exe, PID: 4504, Threads: 4

I can see it is marked by two $$ signs in the BOINC tab, as
  qcn.stanford.edu_sensor = $$#qcn_5.62_windows_intelx86__nci
apparently because of two matching exclude rules (qcn_ + __nci). After removing one of them, one $ sign disappeared and "Gpu: qcn_5.62_windows_intelx86__nci.exe" appeared in the list.
Weird..... 

All false positives.

When the red (straight) lines for temperature limits are drawn over the temperature values' texts at the graph left side, they can under circumstances render them unreadable, e.g. the values 60º and 80º are now stroked through, the 6 is not distinguishable from an 8 at all.
Is it possible to draw the lines _under_ the temperature texts?

[boinc_users]

Is the word boinc_users any secret password or "s"-word or something else unwanted here at this forum? Note that I've actually broken this word with empty italics section, but as soon as I write the complete word anywhere in the text or in the Subject, neither message Preview nor message Post works - I get an empty browser window. (Tried with both Google Chrome and Mozilla SeaMonkey.)

Peter

When accessing an already open Graph window, using either Graph tab / Larger graph button or toolbar icon's Graphs menu item, if the Graph window is underneath other windows or minimized, it will stay so. The user will notice something went wrong, have to take attention, look for the Graph window's icon and elevate/restore it.

Let's raise the graph window in such situation automatically.

My unstarted PrimeGrid Sophie Germail Prime Search (LLR) 5.1 task LLR_SGS_42720474_1 is currently in "20001-state:" (according to BT), BoincManager shows just Ready to start.

What does it mean?

client_state.xml contains just

Code Select Expand

<file_info>
   <name>LLR_SGS_42720474</name>
   <nbytes>42.000000</nbytes>
   <max_nbytes>0.000000</max_nbytes>
   <status>1</status>
   <url>http://www.primegrid.com/download/workunit_llr_sgs.php?id=42720474</url>
</file_info>
<file_info>
   <name>LLR_SGS_42720474_1_0</name>
   <nbytes>0.000000</nbytes>
   <max_nbytes>331072.000000</max_nbytes>
   <generated_locally/>
   <status>0</status>
   <upload_when_present/>
   <url>http://www.primegrid.com/cgi/file_upload_handler</url>
</file_info>
<workunit>
   <name>LLR_SGS_42720474</name>
   <app_name>llrTPS</app_name>
   <version_num>511</version_num>
   <rsc_fpops_est>1700000000000.000000</rsc_fpops_est>
   <rsc_fpops_bound>10000000000000000.000000</rsc_fpops_bound>
   <rsc_memory_bound>100000000.000000</rsc_memory_bound>
   <rsc_disk_bound>10000000.000000</rsc_disk_bound>
   <file_ref>
       <file_name>LLR_SGS_42720474</file_name>
       <open_name>llr.in</open_name>
   </file_ref>
</workunit>
<result>
   <name>LLR_SGS_42720474_1</name>
   <final_cpu_time>0.000000</final_cpu_time>
   <final_elapsed_time>0.000000</final_elapsed_time>
   <exit_status>0</exit_status>
   <state>2</state>
   <platform>windows_intelx86</platform>
   <version_num>511</version_num>
   <wu_name>LLR_SGS_42720474</wu_name>
   <report_deadline>1264765033.000000</report_deadline>
   <received_time>1264160235.823425</received_time>
   <file_ref>
       <file_name>LLR_SGS_42720474_1_0</file_name>
       <open_name>llr.out</open_name>
   </file_ref>
</result>

IIRC I've already seen such uncommon state in the past.

• After adding a separate thread to the Find Computers dialog, it is finally obvious what's going (or sleeping ) on - please add one more edit field with a timeout. (I believe 3-5 seconds (for a very remote hosts, and 1-2 seconds for a LAN) should be mostly sufficient to keep the user a warmy feeling that the search really seems to be progressing.) (Maybe add more working threads for the client search, relative to the timeout length? Like, 3-4 threads for a 10 second timeout.)
• From the log: "19 január 2010 - 10:40:13 Language ---- User: 1051 (SKY), System: 1051 (SKY), Selected: 6 (DAN) Used: 1051 (SKY)" - what does the "Selected: 6 (DAN)" mean? I have currently the "System" selected.
• Sometimes my BT is minimised for a couple of hours or even days. After opening the GUI, usually a very outdated information is kept displayed. I can see that BT is counting down while waiting for the next refresh moment, then communicates the new data with BOINC client and redraws what is displayed. Could the redraw be immediately done after reopening a minimized GUI? I guess that the up-to-date info is continuously available. Even the taskbar's Gadget is continuously showing an up-to-date info (a side-effect of being refreshed after Show only over icon for ... seconds is over ), while the main GUI may be still just counting down for the next update.

Unrelated wish:

• It would be nice if the overall Refresh rate could be overriden for any (usually remote) computer separately in the Computers window - some Refresh column?

I'm slowly beginning with Slovak (Slovensky, 1052, SKY) translation.

But: I've immediately faced one problem: code page. Pure ANSI is for Slovak language not sufficient.
UTF8 or ...? Texts like "Zvyšný čas" ("Time Left") displays like "ZvyšnĂ˝ Ä..." (like if it were displayed as written in Eastern Europe (1250) code page).

     

OK, I've written the file in Central European (1250) codepage - now BT displays it just fine.
So, Fred, you can expect the Slovak translation in a couple of weeks 

When 2 single-threaded apps are running on my computer, the Tasks' "CPU %" column show they use around 93% of CPU. I suppose it means of a real/virtual core.

When a 2-thread multithreaded Aqua is running, the "CPU %" column shows it uses 100% of CPU. I suppose the CPU usage was calculated at around 170-180% and was clipped to 100%.

Fred or we should agree on whether, in such case, the column should display some 89% (of the whole CPU in the computer) or 178% (of one typical core). I'd prefer and heartly recommend making it user-selectable because I know each one prefers the other way

Similar to the Tasks and Projects, the Computers window could also have a context menu for the (selected) computers (and/or a toolbar) with various useful commands, like:

• snooze/suspend/resume/basedonprefs CPU activity
• snooze/suspend/resume/basedonprefs GPU activity
• suspend/resume networking
• reset STD debts
• reset LTD debts
• reset GPU debts
• shut down client
• run benchmarks
• read config file
• read local prefs file
• etc.

Currently the Computers window is just a list of computers the user can connect to (and open "their" window). Maybe, for clarification and fast overview, one more column could state, whether there is an active communication ongoing with any particular computer? (Note that there are/were folks with manymany dozens of computers, controlled by BoincView.)

Another possibility, which would IMHO better fit into current concept: I think that currently you can open a new "Combined [...][....]" window for any group of computers. Each window has then its own set of (one to n) computers and you can display their Projects, Tasks, Messages, etc. Let's add a Computers button into the Selection bar and Computers entry in the Show menu. When pressed, such window would list computers maintained by this window (and which of them are (or were recently) online) and their lines could support context menus with mentioned commands. I guess that the Operation toolbar could also get 1-2 most important options added...

A related bug:

• It is not possible to add computers to / remove from a selection using <Ctrl>+<mouseclick>.
  The <Shift>+<click>, <Shift>+<Up/Down cursor keys> and <Ctrl>+<Up/Down cursor keys>+<Space> selections work correctly.

It often happens that some columns are a bit narrow and trailing parts of message / application / task / project / date / time / etc. string are "truncat..." To be able see the hidden text, it is necessary to enlarge (and then narrow back) the column.

• When hovering mouse cursor over such text, a tool tip containing the whole text could be displayed.
• Such tooltip is usally left-aligned with the original text. But if the tool tip would be displayed through the right display border (or left border for any right-to-left reading locale? OK kidding now), it's position should be shifted to the left as much as necessary to display the whole tool tip on-screen. (A good example of ignoring such situation is Boinc Manager - it has no problem to display tooltips obscured by the screen border.)