Author Topic: T-Clock 2010 (download) (Read 1050609 times)

Stoic Joker · « **Reply #275 on:** March 30, 2011, 07:44 PM »

-AMMA (March 30, 2011, 08:55 AM)

Ok, here what appear for some different languages in older version:

http://img641.imageshack.us/img641/5183/engr.png

T-Clock 2010 (download)

Eng

http://img863.imageshack.us/img863/9218/greece.png

T-Clock 2010 (download)

Greece

And your version only show english locale(ie. the first one) no matter what other locale i set!
[/quote]

Okay, this is rather interesting. It seems that you have identified a paradigm shift in the project that I've never seen/noticed before. The original TClock, and TClock Lite both have the same behavior in that the Clock's regional language is independent of (and completely ignores) the system.

Where as Two_toNe & I (albeit unwittingly), Use/leverage the system's regional setting (ignoring its-own) to determine the clock's output (see below).

(Now if by chance that ain't actually Greek ...(I've not a clue)... I may be screwed)

So actually the Locale setting on the Time Format tab is a decorative artifact of a configuration that (perhaps was only for Win9x) is no-longer used.

Does that make sense? Or do you think there would be need to have the clocks textural region working independently of the (rest of the OS) Windows shell?

Obviously something must be done, as the non/dysfunctional control could be quite annoying. I'm thinking that (pull the control) shell integration/harmony is the (long term) less confusing option ... But you it seems do use/have a need for this feature. Yes?

I'll hold off on doing anything in either direction until I hear from you.

movrshakr · « **Reply #276 on:** April 01, 2011, 11:32 AM »

Sir, an uncertain observation...

I turned the option ON for 'monitor off on screen lock.' That worked--screen did go off when locked. But I had a strange phenomenon occur about 5 times over four days...

I'd be working along and the screen would go dark then come back on on a mouse move. Two of the times, it went into an off-on-off-on-off-on blink mode. The period was about 1.5 sec from off to off or on to on.

I have a widget for oneclick shutdown and was able during the ON blinks to eventually move the cursor over that and click a shutdown.

I do not know for sure that Tclock was causing this, but I strongly suspect it. Since turning the option back off yesterday, I have not had the situation reappear.

AMMA · « **Reply #277 on:** April 01, 2011, 03:00 PM »

-AMMA (March 30, 2011, 08:55 AM)
Ok, here what appear for some different languages in older version:
T-Clock 2010 (download)
Eng
T-Clock 2010 (download)
Greece

And your version only show english locale(ie. the first one) no matter what other locale i set!

Okay, this is rather interesting. It seems that you have identified a paradigm shift in the project that I've never seen/noticed before. The original TClock, and TClock Lite both have the same behavior in that the Clock's regional language is independent of (and completely ignores) the system.

Where as Two_toNe & I (albeit unwittingly), Use/leverage the system's regional setting (ignoring its-own) to determine the clock's output (see below).
(see attachment in previous post)(Now if by chance that ain't actually Greek ...(I've not a clue)... I may be screwed)

So actually the Locale setting on the Time Format tab is a decorative artifact of a configuration that (perhaps was only for Win9x) is no-longer used.

Does that make sense? Or do you think there would be need to have the clocks textural region working independently of the (rest of the OS) Windows shell?

Obviously something must be done, as the non/dysfunctional control could be quite annoying. I'm thinking that (pull the control) shell integration/harmony is the (long term) less confusing option ... But you it seems do use/have a need for this feature. Yes?

I'll hold off on doing anything in either direction until I hear from you.
-Stoic Joker (March 30, 2011, 07:44 PM)

When i started using Tclock i didn't have this feature in mind, but after using it i found it more convenient to have months and days names in my local language, since my most usage of these are in my local language and not in english.
Unlike my common usage for my PC; which is in English, so english local is needed for my system.
I hope you get my point.

Stoic Joker · « **Reply #278 on:** April 01, 2011, 06:17 PM »

When i started using Tclock i didn't have this feature in mind, but after using it i found it more convenient to have months and days names in my local language, since my most usage of these are in my local language and not in english.
Unlike my common usage for my PC; which is in English, so english local is needed for my system.
I hope you get my point.
-AMMA (April 01, 2011, 03:00 PM)

Okay, I think I can work with that. I'm corrently studying multiple versions of the code to see how who toggled what where to flip the default behavior. I'm thinking that the new default is probably best as a default behavior, for most folk ... But I may be able to add a checkbox to enable/disable the language window so both options/behaviors can be accessed.

Sound Fair?

Stoic Joker · « **Reply #279 on:** April 01, 2011, 06:51 PM »

Sir, an uncertain observation...

I turned the option ON for 'monitor off on screen lock.' That worked--screen did go off when locked. But I had a strange phenomenon occur about 5 times over four days...
-movrshakr (April 01, 2011, 11:32 AM)

That sounds like a fairly good sampling size.

I'd be working along and the screen would go dark then come back on on a mouse move. Two of the times, it went into an off-on-off-on-off-on blink mode. The period was about 1.5 sec from off to off or on to on.

I have a widget for oneclick shutdown and was able during the ON blinks to eventually move the cursor over that and click a shutdown.

I do not know for sure that Tclock was causing this, but I strongly suspect it. Since turning the option back off yesterday, I have not had the situation reappear.
-movrshakr (April 01, 2011, 11:32 AM)

That shouldn't be possible (shouldn't...), the monitor off code only fires (once) in response to a Session Lock message that it can only receive from the system. during a Session Change event. Hence the session would have to be in the process of locking to fire the code. Relevant code shown below:

Code: C++ [Select]

case WM_WTSSESSION_CHANGE:
           switch(wParam) {
                case WTS_SESSION_LOCK:
                Sleep(500); // Eliminate user's interaction for 500 ms
                SendMessage(HWND_BROADCAST, WM_SYSCOMMAND,SC_MONITORPOWER, (LPARAM) 2);
                return 0;
           }

I'm running it on both of my Windows 7 machines, one I lock frequently, and one I never lock. So far I've seen no ill effects. Granted that doesn't auto-magically mean there isn't any ... But 2 out of 3 ain't my favorite odds.

When it went on/off/on/off did the power light on the monitor (turn orange) react accordingly? If it stayed green/blue/whatever it could be the florescent bulb starting to go in the monitor.

Video card overheating might cause this type of behavior also...but is unlikely without other behavioral indicators.

Leave the feature off for the next week and we'll see what it does. Then if nothing goes awry, re-enable the feature and we'll see if the behavior returns.

movrshakr · « **Reply #280 on:** April 01, 2011, 07:29 PM »

It is on a Vista Home Premium laptop, so no monitor light to see.

The 500ms in the code is interesting...the cycle time I saw could have been 1/2 sec for either on or off instead of 1-1/2 total as I said. I didn't measure it--just estimating. On and off time 'seemed' about equal, but that is not measured.

If you don't hear about it from anyone else, it may not be Tclock, or it may be some odd interaction with something on my machine and not others.

I will do as you request-- try in a week (assuming I remember to turn it back on in a week!). However, I don't think it started right when I set the option (I don't think...but frankly, I don't remember when I set that). Big help I am.

So, we shall experiment.

I don't read code much and certainly not C++; where is that WTS_SESSION_LOCK message (bit?) set, and what clears it? Could it still be true when unlock happens? But never mind--I was not consciously doing anything to cause the unlock (certainly not entering the password--it was not asked-for). That a clue? Wait! It actually was not locking, because it came back to logged-in screen, not to login password screen as would be normal if it really had locked. It was only screen off on, not lock-unlock.

Stoic Joker · « **Reply #281 on:** April 02, 2011, 09:41 AM »

It is on a Vista Home Premium laptop, so no monitor light to see.
-movrshakr (April 01, 2011, 07:29 PM)

Crap ... Unfortunately that lends credence to my florescent light bulb theory (could be the power board at the base of the screen assy).

The 500ms in the code is interesting...the cycle time I saw could have been 1/2 sec for either on or off instead of 1-1/2 total as I said. I didn't measure it--just estimating. On and off time 'seemed' about equal, but that is not measured.
-movrshakr (April 01, 2011, 07:29 PM)

The half second pause allows the user time to let go of the buttons, so the triggered action isn't inadvertently canceled.

If you don't hear about it from anyone else, it may not be Tclock, or it may be some odd interaction with something on my machine and not others.
-movrshakr (April 01, 2011, 07:29 PM)

Just because nobody else noticed, doesn't (conclusively) mean that it ain't broken (See AMMA's post/discussion

)

I will do as you request-- try in a week (assuming I remember to turn it back on in a week!). However, I don't think it started right when I set the option (I don't think...but frankly, I don't remember when I set that). Big help I am.

So, we shall experiment.
-movrshakr (April 01, 2011, 07:29 PM)

Thank you. The list of things I can't remember could fill a library ... *Shrug* ...It's part of being human.

I don't read code much and certainly not C++; where is that WTS_SESSION_LOCK message (bit?) set, and what clears it? Could it still be true when unlock happens? But never mind--I was not consciously doing anything to cause the unlock (certainly not entering the password--it was not asked-for). That a clue? Wait! It actually was not locking, because it came back to logged-in screen, not to login password screen as would be normal if it really had locked. It was only screen off on, not lock-unlock.
-movrshakr (April 01, 2011, 07:29 PM)

The Clock registers itself with the system to receive Login Session Changes, and then responds to them by firing the monitor off code. And yes you are correct about the clue, the fact that no login was required is important.

Two things (tests) that could help the diagnosis, if/when the behavior reoccurs:
1. View the monitor while shining a flashlight (or other light source) on the screen at an angle to see if the desktop is still there but very faint. Usually have to get really close (8-10") and toggle something like the Start Menu to see it. This is typical behavior if the screen's back-light goes out.

2. Try connecting an external monitor (Might require mode switching the video output), to see if it gives you back the desktop.

movrshakr · « **Reply #282 on:** April 02, 2011, 12:09 PM »

Just for education, when "monitor off" is commanded (by either Tclock's call or something else) are you saying it is just backlight being turned off and the screen data is really still there?

Your no. 1 will be hard to do as it is only off for a moment and actually there seems to be a "fade-to-black" transition going on, although very fast, as it goes into black. But it comes back almost instantly, so seeing what is faintly still there during the off time, if anything, will be iffy.

Stoic Joker · « **Reply #283 on:** April 02, 2011, 04:31 PM »

Just for education, when "monitor off" is commanded (by either Tclock's call or something else) are you saying it is just back light being turned off and the screen data is really still there?
-movrshakr (April 02, 2011, 12:09 PM)

No, the exact opposite is true. If T-Clock is guilty, then the monitor should shutdown completely (and logoff [caused by source unknown]). If the Backlight bulb is faulty, then the ghost image of the desktop should still be visible.

Your no. 1 will be hard to do as it is only off for a moment and actually there seems to be a "fade-to-black" transition going on, although very fast, as it goes into black. But it comes back almost instantly, so seeing what is faintly still there during the off time, if anything, will be iffy.
-movrshakr (April 02, 2011, 12:09 PM)

I was/am shooting for either an incident (of black-out) where T-Clock's feature isn't enabled. Or a longer term episode like the one you mentioned having to shutdown to get out of.

movrshakr · « **Reply #284 on:** April 02, 2011, 07:38 PM »

...I was/am shooting for either an incident (of black-out) where T-Clock's feature isn't enabled. Or a longer term episode like the one you mentioned having to shutdown to get out of.
-Stoic Joker (April 02, 2011, 04:31 PM)

I have had no incidents occur when Tcolock was not enabled.

The time I had to do the shutdown was when it was in one of the off-on blink manifestations. The ones where it completely went to black and stayed there, I was able to get back in a normal way...though I do not remember now whether a log in was required for those. Those happened first; the last 2 or 3 were the blinking kind.

ajp_anton · « **Reply #285 on:** April 04, 2011, 10:29 AM »

Bug:

When using a negative width in the "Clock Text" tab, next time the hour changes the width will appear as if the width setting was double the amount I have chosen.
For example, if I set the width to -32, the next hour (or if I just change the system time manually) the width will go to and remain at -64 (even if it's impossible to set manually).
It's not really a problem as I can just set it to -16 to get -32, but thought I'd point it out.

If it matters (haven't really tested every possibility...), my time format is "yyyy-mm-dd hh:nn:ss".

Stoic Joker · « **Reply #286 on:** April 04, 2011, 11:38 AM »

...I was/am shooting for either an incident (of black-out) where T-Clock's feature isn't enabled. Or a longer term episode like the one you mentioned having to shutdown to get out of.
-Stoic Joker (April 02, 2011, 04:31 PM)
I have had no incidents occur when Tcolock was not enabled.

The time I had to do the shutdown was when it was in one of the off-on blink manifestations. The ones where it completely went to black and stayed there, I was able to get back in a normal way...though I do not remember now whether a log in was required for those. Those happened first; the last 2 or 3 were the blinking kind.
-movrshakr (April 02, 2011, 07:38 PM)

Guess I need to do a bit of digging them ... Let me know how the test progresses.

Bug:

When using a negative width in the "Clock Text" tab, next time the hour changes the width will appear as if the width setting was double the amount I have chosen.
For example, if I set the width to -32, the next hour (or if I just change the system time manually) the width will go to and remain at -64 (even if it's impossible to set manually).
It's not really a problem as I can just set it to -16 to get -32, but thought I'd point it out.

If it matters (haven't really tested every possibility...), my time format is "yyyy-mm-dd hh:nn:ss".
-ajp_anton (April 04, 2011, 10:29 AM)

Just so I can duplicate this properly (first try failed), you are saying that the setting in the properties dialog is doubling (which has a corresponding ill effect on the clock's display output), correct?

And would I be correct in assuming that you are using the latest released build (95) of T-Clock?

ajp_anton · « **Reply #287 on:** April 04, 2011, 01:12 PM »

Bug:

When using a negative width in the "Clock Text" tab, next time the hour changes the width will appear as if the width setting was double the amount I have chosen.
For example, if I set the width to -32, the next hour (or if I just change the system time manually) the width will go to and remain at -64 (even if it's impossible to set manually).
It's not really a problem as I can just set it to -16 to get -32, but thought I'd point it out.

If it matters (haven't really tested every possibility...), my time format is "yyyy-mm-dd hh:nn:ss".
-ajp_anton (April 04, 2011, 10:29 AM)

Just so I can duplicate this properly (first try failed), you are saying that the setting in the properties dialog is doubling (which has a corresponding ill effect on the clock's display output), correct?

And would I be correct in assuming that you are using the latest released build (95) of T-Clock?
-Stoic Joker (April 04, 2011, 11:38 AM)

Latest build, yes.

In properties, if I want a width of -x, I'd just set it to -x, right? This works for a while as expected.
However, the next time the hour digit changes (which I can also do manually by changing the system time), the clock layout will suddenly display as if the width was set to -2x. When going to properties, the width is still set to -x, and if I click "Apply" (have to change something to activate the button), the clock will jump back to -x and stay there until the hour digit changes again.
My workaround is to set the width to -x/2, and change the system timer by one hour in any direction (to get my -x width) and back again (for correct time). The width setting in the properties will keep showing -x/2, so every time I change any setting, the width will be set to -x/2 again until the hour changes.

Sorry for the long post, it's a bit complicated and didn't find an easy way to explain it =).

edit: attempt at a short explanation:
If width is negative: Every time the hour digit changes, it sets the width to double whatever the width setting is, without actually changing the setting. To fight back I could "apply" the settings again every hour, or just "give up" and set the width to half of what I want.

paul_ray · « **Reply #288 on:** April 06, 2011, 11:48 AM »

I don't know if there are plans to use the "GetTickCount64" function instead of the "GetTickCount" function in getting the uptime but it would be nice.

The new function does not have the 49.7 day limit.
It is supported starting with Vista.
(It does not require 64 bit OS).

http://msdn.microsof...724411(v=VS.85).aspx

paul_ray · « **Reply #289 on:** April 06, 2011, 11:51 AM »

Are there plans to use Environment variables for the display?
I would like to use the "COMPUTERNAME" variable in my display format, so I don't have to hard code the machine name.

This makes it easier for support when I need to know the computer name.

Stoic Joker · « **Reply #290 on:** April 07, 2011, 07:05 AM »

I don't know if there are plans to use the "GetTickCount64" function instead of the "GetTickCount" function in getting the uptime but it would be nice.

The new function does not have the 49.7 day limit.
It is supported starting with Vista.
(It does not require 64 bit OS).

http://msdn.microsof...724411(v=VS.85).aspx
-paul_ray (April 06, 2011, 11:48 AM)

Hm... (there are now) That is an interesting option. Actually I had been planning to try using the other method that leverages the info stored in HKEY_PERFORMANCE_DATA ... But this sounds much easier/simpler/quicker - Not to mention a 585 million year wrap/limit should be sufficient for most folks...

Thank you!

Are there plans to use Environment variables for the display?
I would like to use the "COMPUTERNAME" variable in my display format, so I don't have to hard code the machine name.

This makes it easier for support when I need to know the computer name.
-paul_ray (April 06, 2011, 11:51 AM)

That's definitely an enterprise (only) level feature (me thinks). Are you from the company planing to do a T-Clock roll-out? Computer name shouldn't be difficult to grab/add, but what other environmental info bits (if any) might be good to toss in with it?

paul_ray · « **Reply #291 on:** April 07, 2011, 12:27 PM »

That's definitely an enterprise (only) level feature (me thinks). Are you from the company planing to do a T-Clock roll-out? Computer name shouldn't be difficult to grab/add, but what other environmental info bits (if any) might be good to toss in with it?

You are correct about enterprise but I also can use this at home. It is quicker to get the machine name vs going to the command line or to properties of "computer".

The only variables I can think about using are "COMPUTERNAME" and "USERNAME".

Stoic Joker · « **Reply #292 on:** April 07, 2011, 05:39 PM »

Okay, I'm currently working on figuring out how the alternate language feature got broken. But when that's done I'll take a stab at doing the computer name option. I was thinking that the computer name & uptime combo might make a good default option for the tooltip popup that shows when you hover the mouse over the clock.

movrshakr · « **Reply #293 on:** April 07, 2011, 05:44 PM »

... I was thinking that the computer name & uptime combo might make a good default option for the tooltip popup that shows when you hover the mouse over the clock.
-Stoic Joker (April 07, 2011, 05:39 PM)

Or in options, have a place where user can select what he want to show on hover?

Stoic Joker · « **Reply #294 on:** April 08, 2011, 09:07 AM »

... I was thinking that the computer name & uptime combo might make a good default option for the tooltip popup that shows when you hover the mouse over the clock.
-Stoic Joker (April 07, 2011, 05:39 PM)

Or in options, have a place where user can select what he want to show on hover?
-movrshakr (April 07, 2011, 05:44 PM)

Already does that, bottom of mouse tab.

movrshakr · « **Reply #295 on:** April 08, 2011, 09:11 AM »

Already does that, bottom of mouse tab.
-Stoic Joker link=topic=21944.msg244730#msg244730

Then ???.
So. the current discussion is about ADDITIONAL variables that can be put in there?

Guess I'll have to go look at what can be put there now.

LATER- can the font in the balloon be changed? I need it bigger/darker.

Stoic Joker · « **Reply #296 on:** April 08, 2011, 10:56 AM »

Already does that, bottom of mouse tab.
-Stoic Joker link=topic=21944.msg244730#msg244730

Then ???.
So. the current discussion is about ADDITIONAL variables that can be put in there?
-movrshakr (April 08, 2011, 09:11 AM)

Yes.

Guess I'll have to go look at what can be put there now.
-movrshakr (April 08, 2011, 09:11 AM)

All the same format specifiers options as the main clock.

LATER- can the font in the balloon be changed? I need it bigger/darker.
-movrshakr (April 08, 2011, 09:11 AM)

Currently no, but I'll think about it (my to-do list is getting a bit out of hand...

).

movrshakr · « **Reply #297 on:** April 08, 2011, 12:39 PM »

date=1302271902]
LATER- can the font in the balloon be changed? I need it bigger/darker.
-movrshakr (April 08, 2011, 09:11 AM)

Currently no, but I'll think about it (my to-do list is getting a bit out of hand... ).
-Stoic Joker (April 08, 2011, 10:56 AM)

Dang those users; always wanting more and more!

Stoic Joker · « **Reply #298 on:** April 08, 2011, 01:01 PM »

date=1302271902]
LATER- can the font in the balloon be changed? I need it bigger/darker.
-movrshakr (April 08, 2011, 09:11 AM)

Currently no, but I'll think about it (my to-do list is getting a bit out of hand... ).
-Stoic Joker (April 08, 2011, 10:56 AM)

Dang those users; always wanting more and more!
-movrshakr (April 08, 2011, 12:39 PM)

Meh, It's ok ... Gotta keep me off the streets somehow ... And I'm too young for Bingo...

movrshakr · « **Reply #299 on:** April 09, 2011, 01:46 PM »

Sir, regarding the previous posts about the "Locking workstation turns off monitor" strangeness I discussed...

since then, as stated, I had that checked OFF, not turned on. Have had no issues.

This morning I turned it back on. My FIRST attempt to go into lock, it did lock, but instantly, the blink monitor on and off began. I was able to log back in (the on/off of monitor seemed to have no effect on typing the password in). The blinking continued after back into logged in mode. I did a reboot, and once in, unchecked the option.

This may be something that is happening only in MY particular configuration, but there is no doubt in my mind that it is due to the option being ON in T-Clock.