XMBC 2.18 Beta

x64 Replacement/Alternative to Microsoft's IntelliMouse application.
Forum rules
Please read the forum rules before posting for the first time.
The more information you can provide, the quicker and more accurately someone can help.
User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

XMBC 2.18 Beta

Post by phil » Sat Feb 24, 2018 4:24 pm

OK, Its been a while since I released 2.17 and a few issue have cropped up, especially some strange ones that I cant reproduce which is troublesome and hopefully this beta run can help isolate and sole those.

Now is the time to think about the future and what will be in XMBC 2.18 and so, with that in mind, here is my initial list. Note that the red items are the most complex and least likely to get looked at!

Also note that just because something is on the list, does not guarantee it will get added in 2.18!

Please keep this beta thread on topic. It is for announcements and reporting bugs in the current beta, NOT for requesting new features and/or asking about existing (non-beta) functionality!

The things on my to do list for 2.18 are:
  • 500 - Add ability/options to drag/size windows
  • 553 - Profile specific 'Enable Axis lock modifier keys' option not working.
  • 551 - Sticky Left Button X & Y axis not unlocking cursor.
  • 547 - Issue sending repeated complex simulated key sequence over a long period.
  • 549 - Issue with Button Held where the final action is not triggered after the previous action's time elapses.
  • 554 - Performance problem - causing laggy IO/System BEEPs. I cant reproduce this but can generally improve some areas where performance may be causing issues.[/color]
  • 556 - Issue with the default action being triggered after movement to scroll when it should not be. I am unable to reproduce this but will continue to try. May also be related somewhat to #554 above.
  • 548 - Improve Movement to Scroll acceleration and direction vectors based on Killy's input.
  • 552 - Add ability to toggle Movement to Scroll when a layer is active, so any movement when the layer is active, is translated into scroll.
  • 540 - Investigate problems with {PRESS} and {RELEASE} simkey tags, specially when the sequence is interrupted by releasing the button.
  • 566 - Modification to simkeys to action items only on button down, on button up or when repeating
  • Investigate the possibility of detecting more buttons (don't get your hopes up!)
NOTE: Green = Done, Red = Unlikely to be looked at in this beta!

Lets kick things off with 2.18 Beta 1 - coming soon with some of the above fixed.

Regards,
Phil
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Sat Feb 24, 2018 4:29 pm

OK so here is 2.18 Beta 1.

If you have check for beta versions enabled, you should get notified of a new version and prompted to update in the next day or so. This is the most efficient method (bandwidth wise) as the updates are only a fraction of the size of the full install. Otherwise, you can get the full installation beta HERE. Note that this link will always get you the latest beta version!

Changes since v2.17:
  • #555 - Log shows incorrect year in the timestamp.
  • #554 - Performance and laggy IO when using XMBC (some improvement - still investigating)
  • #553 - Profile specific 'Enable Axis lock modifier keys' option not working.
  • #551 - Fixed problem with Sticky Left Button X & Y axis not unlocking cursor.
  • #549 - Issue with button held - not triggering final action after timeout of previous action
  • #547 - Bug sending repeated complex simulated key sequence over a long period.
There is no new language template for 2.18 beta 1.

Any problems, PM me a copy of the log file (or post a snippet in a code block here).

Thanks,
Phil
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
injtsvetkov
Fanatic
Posts: 249
Joined: Mon Jun 06, 2016 8:51 am

Re: XMBC 2.18 Beta

Post by injtsvetkov » Sat Feb 24, 2018 10:21 pm

Well, well, well, what a pleasant surprise :)

Let's see who's up for 2.18 beta :P
HAMA Mirano
Windows 8.1 x64, Intel i5-3230M @ 2.60GHz, 4GB

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Sat Feb 24, 2018 10:31 pm

I thought it was about time, and the right thing to do before my software signing certificate expires next week (because last year it took a while to renew it - hopefully that wont be the case this year). Ive been sitting on a few of those fixes for some time and it was about time I let them loose, but more importantly, I'm trying to figure out why some are having issue 554 and 556 above while others aren't. A little wishful thinking that some potentially significant performance tweaks (CPU usage/hook load related) may help in these cases. We shall see!
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
injtsvetkov
Fanatic
Posts: 249
Joined: Mon Jun 06, 2016 8:51 am

Re: XMBC 2.18 Beta

Post by injtsvetkov » Sun Feb 25, 2018 10:52 am

Well indeed it sounds like there may be some relation between those two. Are the same people having both of them? If not, then you have to get those who have one of them to try reproducing the other (if possible) and confirm whether the root cause is common or not.
HAMA Mirano
Windows 8.1 x64, Intel i5-3230M @ 2.60GHz, 4GB

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Sun Feb 25, 2018 12:14 pm

Yes and no... One person has reported both issues but I don't think they are completely related.
I believe now I know why issue 556 is happening, at least in one case.... So I will look at that first and maybe beta 2 will be with us today :)
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Sun Feb 25, 2018 12:51 pm

OK so here is 2.18 Beta 2.

If you have check for beta versions enabled, you should get notified of a new version and prompted to update in the next day or so. This is the most efficient method (bandwidth wise) as the updates are only a fraction of the size of the full install. Otherwise, you can get the full installation beta HERE. Note that this link will always get you the latest beta version!

Changes since v2.18 Beta 1:
  • #556 - Issue with the default action being triggered after movement to scroll when it should not be..
There is no new language template for 2.18 beta 2.

Any problems, PM me a copy of the debug log file (or post a snippet in a code block here).

Thanks,
Phil
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
sukemaru
Dedicated
Posts: 91
Joined: Tue Oct 18, 2016 7:21 pm

Re: XMBC 2.18 Beta

Post by sukemaru » Sun Feb 25, 2018 2:15 pm

1.
v2.18 beta seems to have a small strange on initial startup. The first time when XMBC showes the setup screen, it seems to forget own language settings :?: -- English screen open (the tray icons menu is in English too), but the setting in the language tab is 'Japanese2' selected. I close the main window without any changes and reopen it, XMBC showes Japanese screen...

2.
It is not about the function of v2.18 Beta (and 2.17 stabled version), "User Guide PDF"s description is not updated about Sim Key tags {Layer:last}/{Layer:back}.
:arrow: ref. my past post in 'XMBC 2.17 Beta' thread

3.
phil wrote:
Sun Feb 25, 2018 12:51 pm
There is a new language template for 2.18 beta 1.
I could not find any changes in the new language template and GUI elements of XMBC v2.18 beta, even with XMBC Translation Tool. The new language template is not 'v2.18 beta1' but 'v2.17 beta16'.

Thanks,
suke

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Sun Feb 25, 2018 3:33 pm

1. Strange because I've not intentionally made any changes there, will see if others notice it (maybe its always done that?!).

2. Doh, forgot again... Will try and remember to fix it in 2.18 (Ive actually raised a bug report this time so it wont be so easy to forget).

3. Oppps thats supposed to say "there is no new ...."
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
Dolmatov
Member
Posts: 36
Joined: Tue Sep 19, 2017 9:19 pm

Re: XMBC 2.18 Beta

Post by Dolmatov » Wed Feb 28, 2018 10:41 pm

sukemaru, problem 1. Same have^ Run - open panel - english, reopen - russian.
In log no view any problem
XMouseButtonControl.zip
--
add. Force run in debug mode
Log-Debug.zip
I can not know for sure, but can the sequence of actions be confused?
The first language download after the settings are closed.
01-03-2018 02:55:20.770> MouseHookLLProc: Passing original mouse message 00000202 through....
01-03-2018 02:55:20.784> CMyMessageWnd::OnSetup - Setup dialog closed.
01-03-2018 02:55:20.784> Loading language file: 'Russian.xmbclp'
You do not have the required permissions to view the files attached to this post.

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Thu Mar 01, 2018 1:51 am

I think it always reloads the language when setup closes (in case you changed it). But it should be loading it beforehand too.
I must have broken it - I'll have a look and fix it, but I need to renew my signing certs before the next build so it wont be instant!

EDIT: Yes I broke it by mistake when changing something else!
I will fix it and build 2.18 Beta 3 as I have found my signing certificate is valid until the 3rd March (just in time!).
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Thu Mar 01, 2018 7:40 pm

OK so here is 2.18 Beta 3 with a few quick fixes before my signing certificate expires and I get a new one!

If you have check for beta versions enabled, you should get notified of a new version and prompted to update in the next day or so. This is the most efficient method (bandwidth wise) as the updates are only a fraction of the size of the full install. Otherwise, you can get the full installation beta HERE. Note that this link will always get you the latest beta version!

Changes since v2.18 Beta 2:
  • #560 - Fixed issue in 2.18 Beta 1 & 2 causing language pack not to be loaded at start-up.
  • #558 - Fixed missing text in advanced settings tab.
There is no new language template for 2.18 beta 3.

Any problems, PM me a copy of the debug log file (or post a snippet in a code block here).

Thanks,
Phil
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
Kukurykus
Committed
Posts: 164
Joined: Sat Jul 02, 2016 1:15 pm

Re: XMBC 2.18 Beta

Post by Kukurykus » Fri Mar 02, 2018 4:10 am

Nice to see you're back to work on XMBC :) I waited long time to come with this post. Maybe it's good moment to make you busy this weekend with my XMBC nightmare I can't sleep at nights because of :)

I mentioned about it 3 months ago when I came with idea of new simulated keystroke, copy to clipboard between other things. You're fixing latency bugs recently so my problem should be close of your attempts.

This is Photoshop image:

Image

When I want to crop it I press c key and move mouse cursor over right upper corner. Now when I hold ALT and click with LMB (holding it as well) I may proportionally change (to outside/inside image) area to crop:

Image

Image

When I'm satisfied with changed area I want to crop I release LMB, then release ALT, and then I can press ENTER or Double Click in range of area to be cropped. Probably you did that not once too in your life :)

Because everything I do in Pthoshop I make simpler with XMBC I used simulated keystrokes also for this operation, especially that everyday I have to crop few hundreds images this way. Unfortunately I found there is bug (?) When I use ALT and LMB it always works. When I do it with XMBC it fails 3 / 10 times.


For XMBC Photoshop profile I set Button Chording for Right Mouse Button.
I fill 'Block / Delay original button action until chord is established'
I leave empty 'Unblock after' and 'Unblock when the mouse moves' boxes.
For RMB, so Default (No Chord) I set Simulated Keystrokes: {alt}{lmb}
I fill 'Block original mouse input' box and choose method 8: sticky hold.


This is very comfortable method to this operation (I didn't need it to anything else), but seems to be unuseful, that I feel like there wasn't this method at all, seeing so often unexpected results it gives :(

It's comfortable because I just click RIGHT mouse button on the corner of image when I want to crop and release this button. Now there's no pressure, I don't have to hold anything, just move mouse to in/outside of image. Normally I had to hold ALT together with LEFT mouse button, then release both. With simulated keystrokes when I set area to crop, eigth method let me press RIGHT mouse button again, and that is all.

Problem occurs not with second click to end operation, but with initial click to start it. for 7 of 10 cases it works but for 3 random it works with up to 1 second delay. (well at least I think it is like it if I'm not wrong)

By the Photoshop action/script I activate cropping area and then I can click on one of corners. What happens when it fails? Method eigth is activated, ALT as well, but LMB with often one second delay!

I have no idea why this delay happens, but I saw on this forums that injtsvetkov(?) had problems with simulated keystrokes where second command was precedeed by ALT key. Maybe it's somehow related.

How I know there is delay. Well I work fast and when I click on that corner I fastly release RMB and move mouse to inside of image to crop it then. When delay happens it doesn't click on corner but inside of image where I have cursor set at the moment of moving it (after I clicked with RMB on the corner 1 sec. earlier).

You can say that I do it too fast and first I move cursor then click so this way I could miss clicking on corner. But nah, I checked it many times. And when I click on corner & then slowly move mouse to inside I see that during first sec. (when bugs occurs) cropping area isn't reduced. It starts to be reduced (enlarged if I move to other direction) when about one second is passed. If bug happens and I move cursor a little faster, that cursor is before one second passed already outside of corner area then LMB click is made inside of image.

I tried it with 2 different mouses, and it's the same. Tomorrow when I'm at work I'll send you DEBUG log when it behaves correctly and when not. We'll see if there will be any difference. If not what then? :)
HAMA Roma, Rapoo 3920P
Windows 10 x64, Intel i5-4670K @ 3.40GHz, 8GB,
Intel(R) HD Graphics 4600, Intel SSD 179 GB HDD

User avatar
phil
Site Admin
Posts: 5399
Joined: Sun Apr 06, 2003 11:12 pm

Re: XMBC 2.18 Beta

Post by phil » Fri Mar 02, 2018 10:09 am

Can you increase the "delay between simulated keystrokes" in advanced settings and see if it makes it more reliable - I found yesterday this helped with another issue ({LMB}{CTRL}{ENTER}) in explorer to open folder in a new window... It seems XMBC is now sending things too quickly and windows/the app is not recognizing the input correctly (in that case anyway). I upped it to 10ms from 4 and it make a huge difference (it worked every time as apposed to one in 10 or more!).

Thanks,
Phil
--[ Phil ]--
--[ Administrator & XMBC Author ]--
Logitech G9, Logitech MX518, Microsoft Intellimouse, Trust 16341 BT Mouse
Windows 10 x64, Intel i5-6600k, Asus 170-A, 16GB DDR4,
nVidia GeForce GTX 970, 2xEvo 850 (RAID 0), 2x2TB WD Black (RAID1)

User avatar
Kukurykus
Committed
Posts: 164
Joined: Sat Jul 02, 2016 1:15 pm

Re: XMBC 2.18 Beta

Post by Kukurykus » Fri Mar 02, 2018 4:39 pm

I tired it months ago and now as well. Unfortunately it's not the cause of problem. Of course it helped me a lot with other simulated keystrokes I used, when 'Delay between simulated keystrokes' was too low/high. I didn't need to change that value for more than this, and other and maybe some one more simulated keys I use, however I use them really many. So the optimal for my work I found is 7ms. Increasing it may affect too slow performance while decreasing that some things don't work sometimes (actually often) as should.

As I say I tried now many values, like 10 and higher, but that doesn't made any difference. Problem exists.

I am attaching two good DEBUG logs and one bad, only extracted parts, from moment it starts to it ends.




Two parts when there is no problem differes but a little. When you look into code (I separated lines that differes to make them more radable) and compare them you will see that in first good code, this part:

Code: Select all

Obtained lock for 'SendInput::ProcessQueue' after 0ms.
CSendInput::ProcessQueue: Sending 1 INPUTS
is before:

Code: Select all

Translating Right Button Up Key state 0x0: Action 128 (Button Chording) becomes Action 28 [Simulated Keys '{alt}{lmb}' (sticky hold)]
Key Hold Sticky: Ignore button up
while in second good code it's after.




In case of bad behaviour there are more differences, like this line at beginning (of correct behaviour):

Code: Select all

Unable to enter critical section for lock 'SendInput::ProcessQueue' (TID: 00001BD0) because it is already locked by 'SendInput::SendInputDelay' - Waiting...
is replaced to this:

Code: Select all

CSendInput::ProcessQueue: Sending 1 INPUTS



Additionally this part:

Code: Select all

Obtained lock for 'SendInput::ProcessQueue' after 0ms.
CSendInput::ProcessQueue: Sending 1 INPUTS
that was in case of good parts either before/after:

Code: Select all

Translating Right Button Up Key state 0x0: Action 128 (Button Chording) becomes Action 28 [Simulated Keys '{alt}{lmb}' (sticky hold)]
Key Hold Sticky: Ignore button up
doesn't occur at all. So above code is neither preceeded nor followed by that part of code.




There is one more difference. When then we have this part of code in good and bad ones:

Code: Select all

ProcessButton: Chord active; Releasing global chord
MouseHookLLProc: Removing message 0x0205 from mouse hook queue as we have overridden it.
Ignoring key injected by X-Mouse Button Control
CSendInput::ProcessQueue: Sleeping for 7ms
only in case of bad behaviour there is additionall line of code that isn't present in good ones:

Code: Select all

CSendInput::ProcessQueue: Finished sending queue... (Flushed = 0)
You do not have the required permissions to view the files attached to this post.
Last edited by Kukurykus on Fri Mar 02, 2018 4:54 pm, edited 2 times in total.
HAMA Roma, Rapoo 3920P
Windows 10 x64, Intel i5-4670K @ 3.40GHz, 8GB,
Intel(R) HD Graphics 4600, Intel SSD 179 GB HDD

Post Reply