Messages

Would it be possible to add a choice of formats for the Elapsed Time, Time Left & Checkpoint tabs? 

01d,11:32:11 (02d,21:55:21) takes up more real estate than
35:32:11 (69:55:21) and provides no more useful information.
It would be nice to get the column widths under control a bit.

Where have I heard this before?  Oh yeah, I asked for it last month

1) ...
You can only set a check, when the BOINC or TThrottle column show a Yes.

I reviewed the Manual and found

Finding TThrottle and not BOINC (missing yes), indicates a missing remote_hosts.cfg or a missing entry in the file.

but I definitely have remote_hosts.cfg on all my boxes & as far as I know it's complete.  Certainly I can use BM to log onto any machine from any other machine using either IP or machine_name.  So what exactly does

a missing entry in the file

refer to? I'll be glad to tweak my remote_hosts.cfg(s) any way I need to

If this has been pointed out before, I apologize in advance.  I searched the FAQ for this topic but didn't find.  I also reviewed the manual.

In Find computers if you "Scan address range" then hit "Stop", it doesn't stop.  Also whether you Stop or not, it appears to scan well past the upper address limit you set, possibly to the end of the subnet defined at the router.  (I say that because the scan appears to take the same time whether I set an upper limit of 2 or 255).  And if I do set an upper limit of 2, known addresses beyond 2 are detected.

Meanwhile you're locked out because the Find computers daughter window holds focus over the main BT display.  I hate that   In fact I've used Process Explorer to kill BT because of it.  Few things tick me off more than my computer so busy doing something else that it won't respond to ME, its LORD & MASTER

1) I can't find anything wrong with it.
You can only set a check, when the BOINC or TThrottle column show a Yes.
2) The implementation of the checkpoints is... a bit strange.
Have to check that with AQUA. What project is running multicore?

1) Actually, I knew that...and (with the hosts already attached) I notice after a Find that the hosts have a Yes under Tthrottle but not under BOINC, which is showing a (correct) BOINC version number on the main Computers tab (also Status=Connected).  When I Remove one host and run Find again, same indications - with Yes under Tthrottle & not under BOINC, can't hold a checkmark.  This isn't a big problem for me, since I (a) have few hosts and (b) know my IP addresses.  But it's another one of those WTFs.  I'll check this out on one of my remote boxes and see what happens.  I just know that I've used this feature before to (re)attach machines, and without problem  

2) Only the "IQUANA" [1.14 D-Wave's Iterative QUANtum Algorithms : Multi-Threaded (mt1)] is multicore/multithreaded.  Actually according to their Server Status it's the only subproject generating work right now anyway.  BE WARNED:  IQUANA has a known tendency to randomly sequester 1 of n cores while the task is in state "Waiting," effectively idling that core, since it won't actually do anything with it.  It's recommended to Suspend Project on AQUA when not actually running the task.  Rumor on the boards is that it's (yet another) bug in the BOINC Scheduler/Scheduling Policy.  I've seen this behavior on both Dual and Quad cores, at random times.  Suspending either the task or project immediately releases the core for rescheduling.  BTW IQUANA is automatically assigned the Resource of n.00 CPUs (up to 8 ).  I believe there are command-line options to limit the number of cores.  Other than the sequestering problem, IQUANA 1.14 looks pretty stable.  I just couldn't resist the goal - developing algorithms for imaginary Quantum computers, like something out of Isaac Asimov  EDIT--In fact I'm reminded of the orignal Star Trek, where the old "duotronic" logic circuits had been superseded by Richard Daystrom's "multitronic" computer.  We may yet see the repeal of Moore's Law.

1) During my Tthrottle witch hunt (Op. cit.) I Removed my P8400 machine from BT.  When reattaching it, I noticed that the Computer>Find computers dialog seems to be broken.  After Find or Scan address range, you can't select anything from the list.  The checkmark appears for an instant then disappears.  I was able to manually add the computer with no problem by entering the IP address/password.  I rolled back to v0.69 & verified that the Computer>Find computers dialog list selection is working OK there.

2) Apparently the

-Fixed: Tasks: Checkpoint: When the resources are > 1, like 2 Cpu, divide the time by 2.

fix didn't work out so well.  Checkpoints of the AQUA multicore are now FUBAR.  The new (counter) is definitely tallying the number of checkpoints, but the time appears to be a derivative of the CPU time (sum of the cores) minus some unknown.  For example, with a CPU time of 02d,01:00:01, the Checkpoint value reads 01d,12:46:50.  The actual Checkpoint interval is <7 minutes.  Also the seconds still appear to be incrementing at (wallclock) * (number of cores)

So, simple divide by...bad idea.  Sorry I suggested it.  FYI the CPU tiime & CPU time at last checkpoint on the Task Properties sheet look accurate (considering the multicores of course) and would seem to provide the basis for a simple divide by to an accurate wallclock time since last checkpoint.  Go figure.

On my Win 7 X64 as user with administrator rights the driver just refuses to install properly. But installing it as administrator works just fine.

Which is why shortly after I got my first WinVista I learned when running any installer to "Run as administrator" even on an account with Administrator rights! 

[5/31/2009]

In device manager press action and select search for (changed) new devices.
(I use a different language version right now, so it may not be exactly right.)

Now it's there again.

Did that and yes it is   In the English version it's called "Scan for hardware changes".  Why can't they call it "Refresh," hotkeyed to <F5> like almost everything else in Windows   In fact, why can't it auto-refresh?

Press the start key and run sigverif  Select start.
When ready select advanced and select show log.

You should find cputemperature.sys       1-8-2010

OK, did that but got

cputemperature.sys       5/31/2009      None                Signed              N/A

I won't complain about the Δ date as long as it's running.

OK, all is good now, running with v2.1, communicating with BT.  (FYI I also forgot that when you wipe the Registry keys you have to reselect "Enable BoincTasks" in Preference...duh )  That's what I mean about stepping back from a problem.

Thanks for all the hand-holding.  However, see the BT 0.71 betatesting thread for new problem uncovered there

-----

P.S., Tthrottle 2.1 is properly displaying both CPU cores (still with nearly identical temp curves of course) on the P8400, which is where this whole thread started.  I noticed when installing 2.1 that the v2.00 one-time calibration test did not run, and in the Registry the Calibration folder is empty.

[Installer 32 Bit V1.2]

Well, I'm confused...

This means the driver is up and running:
Driver installed properly. Driver Version: 1.9

Yup, I saw that, and that's what I understood.  But, when I looked in Device Manager>Non-Plug and Play Drivers, cputemperature wasn't listed (as it is in my other WinVista_x64 host).  Also, when I looked in C:\Windows\System32\Drivers, cputemperature.sys wasn't there either.  In fact I did a Find File of the entire C:\Windows\*.* path, and didn't find it anywhere

So press remove again when TThrottle is shut down. It can take up to 4 seconds for TThrottle to shut down (the icon is the last thing to go).

Did that.

Go to the device manager and select show hidden devices.
In non plug an play.
cputemperature right mouse select uninstall.

Couldn't do that, it wasn't there to uninstall.

Go to C:\windows\system32\drivers and delete cputemperature.sys
Why it's called system32 on a 64 bit system.

See above.

Reboot the computer.

Delete the TThrottle64 folder. Make sure TThrottle is shut down and install V 2.1 again.

Did that.  (Also shut Kaspersky down completely during the install.)  The InstallDriver64.exe window flashes twice during the install, but too quickly to read.

You should see this:  Driver installed properly. Driver Version: 2.0
In the TThrottle log.

Yup, got that:

19 August 2010 - 15:56:37 Driver installed properly. Driver Version: 2.0

Setupact.log has more of the same

WdfCoInstaller: [08/19/2010 15:56.28.093] ReadComponents:  WdfSection for Driver Service cputemperature using KMDF lib version Major 0x1, minor 0x7

WdfCoInstaller: [08/19/2010 15:56.28.109] DIF_INSTALLDEVICE: Coinstaller version: 1.7.6001

WdfCoInstaller: [08/19/2010 15:56.28.109] DIF_INSTALLDEVICE: KMDF in-memory version: 1.7.6001

WdfCoInstaller: [08/19/2010 15:56.28.109] DIF_INSTALLDEVICE: KMDF on-disk version: 1.7.6001

WdfCoInstaller: [08/19/2010 15:56.28.125] Service Wdf01000 is running

WdfCoInstaller: [08/19/2010 15:56.28.125] DIF_INSTALLDEVICE: On-disk KMDF version is newer than or same as the version of the coinstaller. Update is not required

But cputemperature is still not in Device Manager>Non-Plug and Play Drivers, and still no log on to BT.  Like I said at the top, I'm confused.  It's like Tthrottle thinks the driver is properly mounted, but the rest of the system just isn't seeing it.  Also, when I shut down Tthrottle 2.1 and run InstallDriver64, I still get

Microsoft Windows Vista Home Premium Edition Service Pack 2 (build 6002), 64-bit
-------------------------------------------------------------------------
Installer 32 Bit V1.2
LoadLibrary(C) failed: 126
The WdfCoInstaller01007.dll library needs to be in same directory as the *.exe
ERROR: Driver not installed properly. (LoadDriver Dlg)

...and the .dll is right there for all to see.  And when I restart Tthrottle, the log now says

19 August 2010 - 16:40:47 OpenService failed!  Error = 1060 (start driver)
Unable to start driver. (is the driver installed?)
19 August 2010 - 16:40:47 (L01) ERROR: Driver NOT installed properly
Program version: 2.10 64Bit
Microsoft Windows Vista Home Premium Edition, 64-bit Service Pack 2 (build 6002)
Vista or higher detected
64 Bit OS detected (AMD or Intel)

I'm not a dummy, and I know the obvious culprit here is Kaspersky.  But he was completely shut down during the install, and in his Application Control I've set Tthrottle to be excluded from all monitoring, including "Child Processes."  Also, Tthrottle (v2.00 right now) is running on my other 32 & 64 bit hosts with the exact same KIS 2010   I believe it's time to step back from this problem for awhile

--SNIP--

Ok this is Vista 32 bit?
Just tried to install on Vista, works as it should.
As the V 1.82 isn't installing it's something different.
So remove the TThrottle installation folder and install TThrottle again.
Make sure TThrottle is closed, but it should otherwise you can't install.
Make sure you start the installer with the right mouse and select run as administrator. Even as you are logged in as administrator.
And a virus scanner may block a driver to install.

No...sorry, shouold have mentioned...this is Vista 64 bit...the problem in the previous thread back months ago was on my CoreDuo 32 bit machine.  That one has been fine since I changed my security package over to Kaspersky Internet Security, the same one I now run on all my hosts.  In fact it's running Tthrottle 2.00 without problem.

Will try your suggestions.  I always run installers of all software "Run as administrator," a trick I learned long ago.  The AV was actually the first thing I thought of, but it's exactly the same KIS 2010 I've been running on this machine for a year & Tthrottle has been coexisting with it fine.  No, something else has changed, and it was incident to the problem with v2.00 and rolling back to 1.82.

But I'm still puzzled by that error message - if the .dll is right there in the directory, why does the InstallDriver64.exe think it's not   Is one of the Registry keys maybe screwed up?

But this still puzzles me:

Installer 32 Bit V1.2
LoadLibrary(C) failed: 126

On a 64 bit system?

I wondered about that too, but I figured you just wrote the app as 32 bit because why would it need to be 64 as long as the driver is - still the name of the .exe invoking that install was definitely InstallDriver64.

Anyway, I uninstalled, deleted the Tthrottle directory, deleted the Registry's Tthrotle tree for good measure, rebooted, disabled Kaspersky, and reinstalled v1.82.  No change in symptom.  Cputemperature is not listed in Non-Plug and Play Drivers, and no log on to BT.

Initial log entries from Tthrottle as follows:

19 August 2010 - 10:20:44 Driver installed properly. Driver Version: 1.9

Program version: 1.82 64Bit
Microsoft Windows Vista Home Premium Edition, 64-bit Service Pack 2 (build 6002)
Vista or higher detected
64 Bit OS detected (AMD or Intel)

Language: User: 1033 ENU ,System: 1033 ENU

Vendor ID: GenuineIntel
Vendor: INTEL
HighestIntegerValue: 0000000A - Processor Signature: 00010676
Misc. info: 00020800
Feature Flags1 0008E3FD
Feature Flags2 BFEBFBFF

Processor: Intel(R) Core(TM)2 Duo CPU     P8400  @ 2.26GHz               
Processor: Family: 6h, Model: 17, Stepping: 6
Processor: Type: 8400, Stepping: C0, Cores: 2
Processor: Max Case Temperature: 72.4 °C, Max Power 65.0W
Processor: Speed: 3 GHz, Busspeed: 1333 MHz
Processor: tSpec: SLAPL, Technology: 45 nm, Cache: 6 MB
Processor: VoltageLow: 0.8 V, VoltageHigh: 1.4 V

Core Temperature: 55 °C, Raw Data: 882D0000
882d0000,882d0000,882d0000,882d0000
882d0000,882d0000,882d0000,882d0000,01111111
882c0000,882c0000,882c0000,882c0000
882c0000,882c0000,882c0000,882c0000,01111111
882c0000,882c0000,882c0000,882c0000
882c0000,882c0000,882c0000,882c0000,01111111

BOINC:
aqua.dwavesys.com
boinc.bio.wzw.tum.de_boincsimap
climateprediction.net
www.primegrid.com
www.rechenkraft.net_yoyo
www.worldcommunitygrid.org

You can help by reading www.efmer.eu/boinc/faq.html How can I help!
Select the send EMail button,or copy everything in this logging window and mail it to me!
boinc@efmer.eu. We use this information to improve this product.

19 August 2010 - 10:21:26 Number of matching Programs (Processes): 3
Cpu: famous_um_6.11_windows_intelx86.exe, PID: 2604, Threads: 4
Cpu: famous_6.11_windows_intelx86.exe, PID: 2236, Threads: 3
Cpu: iquana_1.14_windows_x86_64__mt1.exe, PID: 3036, Threads: 4
---------------------------------------------------------------------------------------------------------------------   3 0 0 0
19 August 2010 - 10:21:58 Number of matching Programs (Processes): 3
Cpu: famous_um_6.11_windows_intelx86.exe, PID: 2604, Threads: 4
Cpu: famous_6.11_windows_intelx86.exe, PID: 2236, Threads: 3
Cpu: iquana_1.14_windows_x86_64__mt1.exe, PID: 3036, Threads: 4
---------------------------------------------------------------------------------------------------------------------   3 0 0 0

I ran InstallDriver64.exe/uninstall and it said:

Microsoft Windows Vista Home Premium Edition Service Pack 2 (build 6002), 64-bit
-------------------------------------------------------------------------
Installer 32 Bit V1.2
Driver removed!

(Isn't that supposed to mean that the driver was running )  I ran InstallDriver64.exe/install and it said:

Microsoft Windows Vista Home Premium Edition Service Pack 2 (build 6002), 64-bit
-------------------------------------------------------------------------
Installer 32 Bit V1.2
LoadLibrary(C) failed: 126
The WdfCoInstaller01007.dll library needs to be in same directory as the *.exe
ERROR: Driver not installed properly. (LoadDriver Dlg)

As Yogi Berra used to say, "it's deja vu all over again."

The setupact.log said:

WdfCoInstaller: [08/19/2010 10:19.47.713] ReadComponents:  WdfSection for Driver Service cputemperature using KMDF lib version Major 0x1, minor 0x7

WdfCoInstaller: [08/19/2010 10:19.47.743] DIF_INSTALLDEVICE: Coinstaller version: 1.7.6001

WdfCoInstaller: [08/19/2010 10:19.47.746] DIF_INSTALLDEVICE: KMDF in-memory version: 1.7.6001

WdfCoInstaller: [08/19/2010 10:19.47.773] DIF_INSTALLDEVICE: KMDF on-disk version: 1.7.6001

WdfCoInstaller: [08/19/2010 10:19.47.789] Service Wdf01000 is running

WdfCoInstaller: [08/19/2010 10:19.47.792] DIF_INSTALLDEVICE: On-disk KMDF version is newer than or same as the version of the coinstaller. Update is not required

Back to you Fred

--SNIP--

Ok this is Vista 32 bit?
Just tried to install on Vista, works as it should.
As the V 1.82 isn't installing it's something different.
So remove the TThrottle installation folder and install TThrottle again.
Make sure TThrottle is closed, but it should otherwise you can't install.
Make sure you start the installer with the right mouse and select run as administrator. Even as you are logged in as administrator.
And a virus scanner may block a driver to install.

No...sorry, shouold have mentioned...this is Vista 64 bit...the problem in the previous thread back months ago was on my CoreDuo 32 bit machine.  That one has been fine since I changed my security package over to Kaspersky Internet Security, the same one I now run on all my hosts.  In fact it's running Tthrottle 2.00 without problem.

Will try your suggestions.  I always run installers of all software "Run as administrator," a trick I learned long ago.  The AV was actually the first thing I thought of, but it's exactly the same KIS 2010 I've been running on this machine for a year & Tthrottle has been coexisting with it fine.  No, something else has changed, and it was incident to the problem with v2.00 and rolling back to 1.82.

But I'm still puzzled by that error message - if the .dll is right there in the directory, why does the InstallDriver64.exe think it's not   Is one of the Registry keys maybe screwed up?

I made some small changes. So anyone with a 2 core Intel CPU try this version:

http://www.efmer.com/download/boinc/tthrottle/TThrottle_2_0_0_1.zip

With regedit.exe you have to remove the folder: HKEY_CURRENT_USER\Software\eFMer\TThrottlexx\calibration and only the calibration folder.
Start this new version and check how many temperature sensors are detected.

I haven't had a chance to try the 2.00.1 .exe yet, due to a more urgent problem: after rolling back to v1.82, the cputemperature.sys no longer wants to install

I went back to the Lost connection to CPU thread and reviewed that info, so you don't need to type all that stuff up again    I verified that cputemperature is definitely absent from Device Mgr.  I tried the InstallDriver64 utility, used Admin privileges, did the Uninstall first, ran the Install & got a very strange error message:

Installer 32 Bit V1.2
LoadLibrary(C) failed: 126
The WdfCoInstaller01007.dll library needs to be in same directory as the *.exe
ERROR: Driver not installed properly. (LoadDriver Dlg)

But the .dll (file version 1.7.6001.180000) is right there for all to see   I tried another "clean" install of v1.82 but no change in symptom.  The good news is that this is my console machine, so I can see the temps on the OSD (also it runs pretty cool, only 55~65C).  The bad news is there won't be any throttling if needed, and no log onto BT.  This is the CoreDuo x64 with WinVista SP2.  Tthrottle has been running fine for months, beginning with v1.73.

Anyway, I'll be staying away from the 2.00.x until this is resolved.  I'm open to suggestions.

Much better

Now on to testing.

2) I mostly leave that filter on, as there are thousands of WU in it.

I do too.  But I do get single entries for projects like GPUGRID & AQUA, and that's how I noticed.

[(0) 00:00:00]

1) Looks like sorting on History has gone haywire.  Whatever primary sort column is selected, it is always sorting by Project name instead.  Secondary & tertiary sortpoints have no effect either.  That alone makes 0.70 dead on arrival for me, so I'll be rolling back to 0.69  As always, prioritize this how you see fit, but for me it's a fatal flaw.

2) I did notice that "Ready to start"  tasks that are not combined by the filter are now displaying a (0) 00:00:00 Checkpoint - probably as a result of

-Fixed: Tasks: Checkpoint: Show the time when a filtered item is only one.

Those should probably be suppressed like

-Changed: Tasks: Don't show the checkpoint when a task is: Ready to report, uploading or in error.

3) The new checkpoint warnings via config.xml appear to be working OK so far EXCEPT - as I predicted, the XML parser won't accept "WORLD COMMUNITY GRID" with the embedded spaces  It also wouldn't take "WORLD_COMMUNITY_GRID".  No error or anything, it just ignored that project section altogether.  I didn't play around with other combinations, just truncated the project value to "WORLD" and it's fine.

That's all I had time to look at before rolling back.

I installed v2.00β over v1.82 (which was running fine) on a Core Duo x64, WinVista SP2 host.  During the update, 2 unusual conditions:

• BOINC had AQUA IQUANA in progress, a multicore task running across both CPU cores
• BOINC was auto-suspended (time of day) for 10 minutes to allow disk defragging by Diskeeper

After the install completed, the "one time test" ran to identify processors.  After that Tthrottle only sees one CPU core (core 0).

I tried uninstall/reinstall v2.00, no change in symptom.  I tried rolling back to v1.92 (it sees both cores) and (re)installing v2.00 over it.  The one time test did not run again, and v2.00 again does not see core 1.  V2.00 installed on 2 other hosts (a 64 bit quad and a 32 bit dual) and all cores are recognized on those machines.  So the one time test appears to be the problem.

I need a workaround to force rerun of that test with both cores working on separate tasks.  Some way to do a completely "clean" install, edit the Registry, whatever.  And the one time test may need some work.

Ok go to HKEY_CURRENT_USER\Software\eFMer\TThrottle(64)
Give me the values in the calibration folder.
Remove the calibration folder and restart TThrottle.

Very strange.  The calibration folder values are as follows:

Windows Registry Editor Version 5.00

[HKEY_CURRENT_USER\Software\eFMer\TThrottle64\calibration]
"cores"=dword:00000002
"core_delta"=dword:00000010

And as I paste that in here, I'm looking at Tthrottle displaying a single core, both on the Temperature tab and on the OSD.  So I deleted the calibration folder, restarted Tthrottle and the one time test ran for a very long time - over 15 minutes.  I had to leave with it still running and when I came back, v2.00 was again displaying a single core.  The rewritten calibration folder is as above.

So for now I'll be rolling back to v1.82 on this machine until we get this figured out.  How about a complete removal of all Tthrottle related stuff for a truly clean install?  (BTW the processor is Intel Core2 Duo P8400@2.26Ghz.)

OK, so I deleted the entire HKEY_CURRENT_USER\Software\eFMer\TThrottle(64) tree (actually exported to a .reg file for safekeeping), uninstalled v2.00 and installed v1.82 clean.  It recognizes both cores.  With 1.82 the calibration folder is present but empty, and of course the calibration test did not run.  But I guess you know that.  I again installed v2.00 as an update over v1.82.  To be safe this time I shut down both the BOINC client and BoincTasks.  v2.00 installed, the calibration test ran (just 2~3 minutes) and again Tthrottle sees only Core 0.  The calibration folder is the same, except for some odd reason the keys are written in reverse order - but the values are the same.  Maybe the new test just doesn't like the P8400 processor.

Anyway, I manually edited the core_delta key from 10₁₆ to 1₁₆, which matches the value on the other hosts that are displaying correct number of cores.  That fixed the problem - v2.00 now displays core 0 & 1.

I'll leave it to you to figure out where the calibration test is going wrong.  Happy hunting.

EDIT--I should mention that I've noticed on this machine that the temperature curves are always very close together: not identical, but close.  If I idle one core, the temp on both cores drops and they still maintain about the same curve.  I'm guessing that in the P8400 package  the temperature sensors are physically very close together.  This differs greatly from my other CoreDuo, where each core has a very different temperature curve, usually by several degrees.  Maybe in the P8400 your test is having a tough time telling the 2 cores apart