MapTool 1.3 Final, patch04 (build 87)

[dorpond]

Odd.. I don't know how that would have changed from last build. I think we may have to wait for Azhrei to pipe in with all this. Thanks for the details Bone White

[aliasmask]

Hmm, just noticed the snap to grid changed. If I move a token slightly to the left or up, then it jumps to next square. This is something that use to happen, but was fixed.

edit: Also, it's kind of annoying that token label and move distance is below the FOW, but I'm not sure if having the token above the FOW be a good idea.. maybe only during the move?

OK, I just tested the token move - mine seems to be working fine with square grids. I did notice it a bit sensitive with hex grids. What grid are you noticing this issue with?

Also, regarding the Labels: Are you saying that we changed behavior with this version or that you are asking for a future request?

It's the square grid on the token layer. In b86, snap to grid would not move token until the center of token broke a boundary. Now, it has reverted to the upper left corner. So, if I want to move a token right, I have to move it 50px, but to move left I only need to move it 1px. Same thing with up vs down. So, the behavior when dragging the token make it jump to locations not expected.

This is best demonstrated by dropping a token on a map, taking off the snap to grid. Then move the token left or up .1, then put snap to grid back on. You'll see the token jump to next square.

As for labels, I don't really find that a big deal because we have the stat sheet popup. But the distance move can not be seen if the FOW is below your token. I don't usually play as a player, but when moving down a long corridor that was 5' wide and to the left, I couldn't see how far I was moving with the counter. I haven't played a player for awhile and noticed this before, but I know b68 had the label and move counter above FOW.

My thought is that the player name and move counter should be rendered above the FOW during a move and only for client. The other players shouldn't be able to see the label and move counter above the FOW, if that is possible. For the GM, since they have soft fog, it wouldn't matter what order the render is.

[dorpond]

What method are you guys using to launch? I download the zip and run mt via the batch file. My tokens are moving as expected. I'll have to test better later - bad time ATM.

[aliasmask]

What method are you guys using to launch? I download the zip and run mt via the batch file. My tokens are moving as expected. I'll have to test better later - bad time ATM.

I use the MapToolLauncher.exe, but it does the same thing with "Launch MapTool-1G-Memory.bat".

/// Java Report \\\

==== Java Information ====
Java Vendor....: Sun Microsystems Inc.
Java Home......: D:\Program Files\Java\jre6
Java VM Vendor.: Sun Microsystems Inc.
Java VM Version: 20.1-b02
Java Version...: 1.6.0_26
...............: Java(TM) SE Runtime Environment (build 1.6.0_26-b03)
...............: Java HotSpot(TM) Client VM (build 20.1-b02, mixed mode, sharing)


==== OS Information ====
OS Name........: Windows XP
OS Version.....: 5.1
OS Architecture: x86
Processor......: x86 Family 15 Model 6 Stepping 2, GenuineIntel
Number of Procs: 2

==== User Information ====
User Name: Chris
User Home: D:\Documents and Settings\Chris
User Dir.: E:\Downloads\System Info Java Report

==== Network Interfaces ====
Display Name..: MS TCP Loopback interface
Interface Name: lo
Address...: 127.0.0.1

Display Name..: Realtek RTL8139 Family PCI Fast Ethernet NIC - Packet Scheduler Miniport
Interface Name: eth0
Address...: 10.0.0.12

Host Address...: 10.0.0.12
Default Gateway: 10.0.0.1

==== Locale Information ====
Country.: United States
Language: English
Locale..: English (United States)
Variant.:

==== Display Information ====
Number of Displays: 1
Display 1: 1680x1050(32)

==== Internet Gateway Devices ====
Device Name.: Residential Gateway
Model Name..: RangeMax(TM) Wireless Router
Manufacturer: NETGEAR, Inc.
Model Number: WPN824v2
Model Desc..: RangeMax(TM) Wireless Router
Firmware....: Ambit OS/1.0 UPnP/1.0 AMBIT-UPNP/1.0

I just upgraded to 1.6.0_27 and it's the same thing.
I also re-downloaded the file and it's the same thing.

[aliasmask]

Good news is that I really do notice the increased speed when loading.

[Phergus]

The token movement issue is easiest to see if you zoom in to like 300% or so.

• Drop a token.
• Position your mouse pointer over the center of the token
• Left-click and hold on the token.
• Now slowly move up until the token snaps to the grid cell above.
• Without moving the mouse release the mouse button.
• Note where your mouse pointer is in relation to where it started.

Repeat for moving down as well as left vs right.

On the down and right movements your mouse pointer should end up in roughly centered over the token in its new position. For up and left your mouse pointer will still be in the grid cell it started in and only moved a few pixels.

[Azhrei]

The token movement issue is easiest to see if you zoom in to like 300% or so.

This is the result of Phergus' patch to fix hex movement, I believe. (Sorry, Phil. I noticed it right after that patch went in.)

I rewrote the movement code awhile back and it got tricky because the mouse pointer coordinates need to be adjusted based on the grid type: square grids use the upper left corner and hex grids use the center. But figuring out what grid is being used is a hassle when you're inside the mouse event handling code (find the current map renderer, find the map, find the grid for the map, determine if the grid is snap-to-grid, determine if the current token is snap-to-grid, apply the offset if necessary by asking the grid object what the offset should be, and so on). Add to it that some function calls need to know the grid cell coord and others need the map coord...

The end result is that the code that was there handled square grids okay but was off for hex grids. Changing the hex grid to work broke the square grid. The correct way to handle this is to redo the entire mouse event system so that it (a) uses grid cells on gridded maps with snap-to-grid tokens, and (b) uses map coordinates everywhere else. I started on that, but the tentacles of that code are far-reaching.

I'll look at Phergus' patch and see if there's some other way to handle it that might be more effective.

[wolph42]

Ok with this version done I thought of doing a heavy benchmark. I've have a library builder which builds HUGE json objects and also build cached strings which can be loaded into a frame (HTML), so it loads fasts. These are mainly json operations and some lib calls. Here are the results of the last 6 versions (B82 upto B87). I tested B87 at the beginning and at the end of this.
I killed most of my background processes while running the benchmarks.

Here are the results:

Benchmark.jpg (50.32 KiB) Viewed 2095 times

I leave the conclusions to the reader.

Here some snippets of code that are repeatly run

Building jsons

Code: Select all

[h:newWeaponName = "Godwyn-Deaz Pattern Bolter"]
[h:newWeaponObject = json.set("{}", "Category","Ranged", "Group","Bolt", "Class","Basic", "Range",90, "SingleShot",1, "SemiBurst",2, "FullBurst",0, "WpnDamDice",1, "WpnDamDiceType",10, "WpnDamMod",5, "DamType","Xplosive", "Penetration",4,"Clip",0,"ClipSize",30, "WpnAttMod",0,"NumClips",0, "ClipType","-","RldTime","1 Full", "numHands",2, "Special","Reliable, Tearing","SpecialText","", "Weight",6, "Cost","1000", "Availability","Very Rare", "Page",114, "Source","DH-BoM")][h:BasicWeaponsTmp = json.set(BasicWeaponsTmp,newWeaponName,newWeaponObject)]
[h:allWeaponsTmp = json.set(allWeaponsTmp,newWeaponName,newWeaponObject)]
 

Building forms

Code: Select all

[h:'<!-- createWpns -->']

[h:strWpns = arg(0)]
[h:Facing=""]
[h:WeaponLibrary = getLibProperty(strWpns, "Lib:Compendium")]
[h:WeaponList = listSort(json.fields(WeaponLibrary), "A+")]
[h:me = "%{me}"]

<html>
<head>
<link rel='stylesheet' type='text/css' href='myCSS@[r:getMacroLocation()]'></link>
</head>
<body>
<form id="pickWeapon" method="json" action="[r:macroLinkText('ShowPickedWeapons@Lib:DH', 'all',"",me)]">
<table id=strWpns width="100%" bgcolor="white">
<tr>
    <td class="head" valign="bottom"><b>Select</b></td>
    <td class="head" valign="bottom"><b>Weapon</b></td>
    <td class="head" valign="bottom"><b>Class</b></td>
    <td class="head" valign="bottom"><b>Group</b></td>
    <td class="head" valign="bottom"><b>Range</b></td>
    <td class="head" valign="bottom"><b>RoF</b></td>
    <td class="head" valign="bottom"><b>Damage</b></td>
    <td class="head" valign="bottom"><b>2H</b></td>
    <td class="head" valign="bottom"><b>Special</b></td>
    <td class="head" valign="bottom"><b>Rld</b></td>
    <td class="head" valign="bottom"><b>Clip</b></td>
    <td class="head" valign="bottom"><b>Wt</b></td>
    <td class="head" valign="bottom"><b>Cost</b></td>
    <td class="head" valign="bottom"><b>Avail</b></td>
    <td class="head" valign="bottom"><b>Source</b></td>
    <td class="head" valign="bottom"><b>Facing</b></td>
</tr>
[h:cssclass='oddRow']

[foreach(Weapon,WeaponList,""),CODE:
{
    [H: varsFromStrProp(json.toStrProp(json.get(WeaponLibrary, Weapon)))]

    [h: RoF        = strformat("%{SingleShot}/%{SemiBurst}/%{FullBurst}")]
    [h: Damage    = strformat("%{WpnDamDice}d%{WpnDamDiceType}+%{WpnDamMod} %s (%{Penetration} Pen)", substring(DamType,0,1))]
    [h: Source    = strformat("%{source}(%{page})")]

    [h: colorClass = strformat("%s%s",substring(Source,0,2),if(cssclass=='oddRow', 'odd', 'even'))]
    
    <tr class='[r:cssclass]'>
        <td valign="top"><input type="checkbox" name="[r:Weapon]" '[r:"%{"+Weapon+"}"]'></input></td>
        <td valign="top" class = "[r:colorClass]"><b>[r:Weapon]</b></td>
        <td valign="top">[r:Class]</td>
        <td valign="top">[r:Group]</td>
        <td valign="top">[r:Range]</td>
        <td valign="top">[r:RoF]</td>
        <td valign="top">[r:string(Damage)]</td>
        <td valign="top">[r:numHands]</td>
        <td valign="top">[r:Special]</td>
        <td valign="top">[r:RldTime]</td>
        <td valign="top">[r:ClipSize]</td>
        <td valign="top">[r:Weight]</td>
        <td valign="top">[r:Cost]</td>
        <td valign="top">[r:Availability]</td>
        <td valign="top">[r:Source]</td>
        <td valign="top">[r:Facing]</td>
    </tr>
    [h: cssclass = if(cssclass=="oddRow", "evenRow", "oddRow")]
    }]
</table>

<input type="hidden" name="token" value='[r:me]'></input>
<input type="hidden" name="pickedWpnType" value='[r:strWpns]'></input>
<input type="submit" value="Choose Weapons"></input>
</form>
</body>
</html>
 

[Phergus]

This is the result of Phergus' patch to fix hex movement, I believe. (Sorry, Phil. I noticed it right after that patch went in.)

No biggie. All my patch did was revert out the changes that really screwed up the hex movement.

[Natha]

The new font is great.
Easier to read but more compact.

[Azhrei]

That must be a side effect of the TinyLAF update -- I didn't change anything with regard to the theming except update the library. I also noticed the font change (or thought I did!) but it looked better to me too so I didn't bother to look into it. (If it had looked worse, then I would have!)

One key thing about TinyLAF: the control panel that allows you to change the theme defaults is no longer part of the base library. It was split off by the library author into a separate JAR which is not part of the MT download. I was thinking of adding a Help menu item re: changing the theme that gives them a short description of how it works and then a button that takes them to the TinyLAF home page where they can find the control panel. What do y'all think? Since the control panel functionality used to be there and isn't any more, it seems like there should be some help for people trying to find it again...

[mfrizzell]

Well my group and I used B87 in our game last night and it worked flawlessly. Only had one issue when I thought VBL was broken. All of my players were seeing all of my creatures no matter where on the map they were even when behind or in the middle of VBL. I thought VBL was broken but discovered that when I loaded my campaign into B87 it set vision to off and FOW to off. So all I had to do was turn them back on and all was well with my world again.
Great job.

[neofax]

So far have played two 4 hr sessions and the only problem I have is with the new theme. Previously the OK button was defaulted to be selected so when I hit enter on the keyboard, OK was depressed. Now, I have to tab through a whole input tab to select OK to hit enter. The reason I have to do this is because input dialogs pop under the windows task bar/jump bar if they are larger than the vertical height of the screen. Is there any way to see if the upstream library has a fix to find out how many pixels the task bar is and cut out these pixels when determining view size?

[aliasmask]

Here's a new weird development. My router UPnP doesn't work with b87. I went through the motions of resetting and all that junk, but it's not working. I loaded b86 no problem. I tried using the MapToolLauncher and the Bat files to no avail. I'm going to check my router for an upgrade and I'll let you know how it goes. I can still connect via lan though, and I'm sure if I set up port forwarding that would work too, but I like my UPnP.

Image1.jpg (14.09 KiB) Viewed 2001 times

edit: I just checked router firmware and it's the newest version. I did just update my java recently to 1.6 27 (from 26). But I don't think that's it since b86 works fine.

/// JAVA INFO \\\

==== Java Information ====
Java Vendor....: Sun Microsystems Inc.
Java Home......: D:\Program Files\Java\jre6
Java VM Vendor.: Sun Microsystems Inc.
Java VM Version: 20.2-b06
Java Version...: 1.6.0_27
...............: Java(TM) SE Runtime Environment (build 1.6.0_27-b07)
...............: Java HotSpot(TM) Client VM (build 20.2-b06, mixed mode, sharing)


==== OS Information ====
OS Name........: Windows XP
OS Version.....: 5.1
OS Architecture: x86
Processor......: x86 Family 15 Model 6 Stepping 2, GenuineIntel
Number of Procs: 2

==== User Information ====
User Name: Chris
User Home: D:\Documents and Settings\Chris
User Dir.: E:\Downloads\System Info Java Report

==== Network Interfaces ====
Display Name..: MS TCP Loopback interface
Interface Name: lo
Address...: 127.0.0.1

Display Name..: Realtek RTL8139 Family PCI Fast Ethernet NIC - Packet Scheduler Miniport
Interface Name: eth0
Address...: 10.0.0.12

Host Address...: 10.0.0.12
Default Gateway: 10.0.0.1

==== Locale Information ====
Country.: United States
Language: English
Locale..: English (United States)
Variant.:

==== Display Information ====
Number of Displays: 1
Display 1: 1680x1050(32)

==== Internet Gateway Devices ====
Device Name.: Residential Gateway
Model Name..: RangeMax(TM) Wireless Router
Manufacturer: NETGEAR, Inc.
Model Number: WPN824v2
Model Desc..: RangeMax(TM) Wireless Router
Firmware....: Ambit OS/1.0 UPnP/1.0 AMBIT-UPNP/1.0

Can anyone else confirm the same problem?

edit: Btw, I did test port forwarding and that works fine. I started 2 servers. One in b86 using UPnP and one in b87 using port forwarding (on different port of course). Trying to use UPnP with b87 still fails for me. So, I'm guessing something changed in the last update. Let me know if there something you want me to do or test.

edit: Just to verify it wasn't the java change, I removed the current version and reloaded 1.6 26. Same problem.

[Azhrei]

I thought VBL was broken but discovered that when I loaded my campaign into B87 it set vision to off and FOW to off.

I just loaded an old campaign in b87 and the FOW and Vision settings were correct for every map in the campaign. There were 7 that had FOW, 6 that had Vision!=Off, and a total of 9 maps (some with FOW also had Vision, hence the overlapping numbers).

Can you load your old campaign in b86 and see if the FOW and Vision were actually set correctly on the map that had the wrong settings? The campaign loading code did have a bug fixed regarding the loading of images, but it shouldn't have affected the FOW or Vision settings of the map at all...

Previously the OK button was defaulted to be selected so when I hit enter on the keyboard, OK was depressed. Now, I have to tab through a whole input tab to select OK to hit enter.

Hm. I did see a note in the TinyLAF Release Notes that the handling of the default button had changed slightly. Let me take a look -- I think it's a checkbox in the TinyLAF control panel now... If so then it can probably be changed by the program as well.

The reason I have to do this is because input dialogs pop under the windows task bar/jump bar if they are larger than the vertical height of the screen. Is there any way to see if the upstream library has a fix to find out how many pixels the task bar is and cut out these pixels when determining view size?

The input function simply creates a dialog and populates it. The JDialog class that it uses is part of the Java library so it would require an update to Java to fix this at the source. However, I'll look and see if there's a simply way to constrain the window width/height prior to displaying the dialog window (calling setSize() might work, but I don't know if the JDialog class resizes the window internally prior to displaying it; I'll try it).

My router UPnP doesn't work with b87.
[...]
Can anyone else confirm the same problem?

Won't be me. Apple broke the UPNP library by changing the behaviour of the networking layer in their Java build so that socket() always causes an exception now. I haven't been able to use the UPNP stuff for about 10 months now.

edit: Just to verify it wasn't the java change, I removed the current version and reloaded 1.6 26. Same problem.

Very interesting. There was no update made of the UPNP library! I just checked and it's been the same library since SVN 4000! (We're currently at revision 5805.)

Looking through the SVN history I don't see any changes mentioned in the UPnPUtil code since Sept of 2010. It's possible something that called those functions was changed...

Please try the following:

1. Clear out any existing log.txt file under .maptool.
2. Use the logging.xml file in the same directory to turn on DEBUG level logging for UPnPUtil
3. Start b87 and use Help > Gather Debug Information. Since your post indicates that it found the router, that should cause logging information in log.txt that shows what it looks like when things are working.
4. Now start a server with UPNP checked.
5. Post the contents of the log.txt

There should be both working and failing entries in the log file and hopefully I can trace through them to determine where the problem is. Thanks for posting this, am.