Jump to content
COMBATSIM Forum

Recommended Posts

Judging by the general look, lack of background colour and the 200x200 map, I'd say that Angola etc are TFX1 technology.

On the plus side is that the SSD files pointed at by the level files actually exist. The problem I think is the model enumeration via ssinfo.fn is not valid.

I really hope that's not the case as I don't see any alternative files looking like ssinfo.fn in the 'unknown files' category.

Link to post
Share on other sites
  • Replies 302
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

The funny thing is, I don’t see these errors in my 2016 screenshots. So either I broke one of the parsers or it’s problem with the files. (I switched to the extracted package you linked here; I don’t remember which extracted package or did.dat I used in 2016 …)

 

pal_dark is pretty romantic.

2020-05-03%20pal_dark.png

 

There is a set of palettes with lakes and ocean completely frozen … do you know whether it is used in-game?

  • pcd_cd   – day; frozen and clear; very bright colors
  • pcd_rn    – same
  • pcn_cd   – night; frozen and clear; very bright colors; blue horizon
  • pcn_rn    – same
  • pct_cd    – twilight; frozen and clear; very bright colors

2012(!) image as a reminder:

2012-09-29%20first%20EF2000%20terrain%20

Link to post
Share on other sites

The most recent link is to a fresh extraction of EF2000 V2.0's did.dat, so should be good.

I needed to that as I don't know the status of my older extracted folders.

 

If I just do a search for 'pcn_rn' for example, it turns up 'pn_rn.txt' in the lev folder, of which here's an extract:

; palettes
;ÄÄÄÄÄÄÄÄÄ

load_pal
pn_rn 1300

cloud_fade
1300 1400 1600 1700 0

load_pal
pcn_rn 1700

load_pal
pcn_rn 10000

load_pal
pcn_rn 20000

 

...and I'd guess it's used to help with the whiteout effect within a cloud...or something related to that.

Link to post
Share on other sites
2 hours ago, mikew said:

...and I'd guess it's used to help with the whiteout effect within a cloud...or something related to that.

That makes sense! Even more so considering the strange blue horizon!

 

2 hours ago, mikew said:

;ÄÄÄÄÄÄÄÄÄ

I guess the file encoding is OEM 850 :)

 

We call TAW’s hidden world Seaworld. Where does this name come from? I just noticed your EF2000 containing an overhead map called seaworld.lbm :)

Link to post
Share on other sites
6 hours ago, Krycztij said:

 

Just some experts trying to make the hidden/forgotten content more accessible :D

 

———

 

Well that's the part I understood.

 

Have you guys any recommendations wrt to programmes to open and edit the various extracted files, especially SSD and FAD types etc.? 

Link to post
Share on other sites
14 minutes ago, mikew said:

That's probably a remnant from TFX1 where Seaworld is one of the maps:

https://community.combatsim.com/topic/41326-back-to-tfx1/?do=findComment&comment=5186186

 

Oh yes, NOW I remember!

The Iceland map is just incredible. The many icel_* palettes do not work, but even with EF2000’s standard palette and all the graphics errors it’s a joy to fly there. Could have become a full-blown scenario.

 

 

There is something about the scale, though. I don’t believe Iceland’s mountains are 18k feet high. (And Angola’s doors shouldn’t be taller than a 737 either, for that matter.)

Link to post
Share on other sites
On 5/3/2020 at 4:13 PM, mikew said:

None of the terrain ssd files used in the default Norway map have out of range indices.

 

As I examine the files it is clear that some, or all, of the indices in the affected files  are either out of range, and / or erroneous. Could those files be for an alternate world? Would TFXplorer  try to load them anyway?

 

Looking at and comparing gtrond.ssd vs trond.ssd it is clear that gtrond's index values are 'erroneous' while tronds are correct. Further, if I compare them Block-by-Block they appear to be, basically, the same airfield super-shape, especially Block # 3 which is identical except for one entry. That makes them appear to be redundant. What does the g in gtrond designate? Is there a glide variance? If all the erroneous index containing files were culled from the ssd cohort would TFXplorer cease generating those errors?

Link to post
Share on other sites

Here’s the build with the GUI:
http://krishty.com/taw/tfxplorer/2020-05-05/TFXplorer.7z

 

It does come with lots of regressions, so it’s just for testing.

  • there is less error output because a major overhaul is in progress
  • it does not work with packed EF2000 did.dat (already fixed for the next version)
  • on Windows 7, some buttons don’t show correctly unless you also install the Segoe MDL2 Assets font from https://aka.ms/SegoeFonts
  • not tested with Windows XP
  • there are no more explosions due to overhaul
  • the bullets are invisible due to overhaul
  • if the game stops receiving input after a focus change, open the menu and close it

How to install

  1. extract to any folder (it doesn’t have to be a TFX PROGRAM folder any more)
  2. open Regedit.
  3. create a key HKEY_CURRENT_USER\SOFTWARE\TFX to UAW\TFX paths
  4. add a new string value:
    • name: 1
    • data: path to your TAW/PROGRAM (without quotes)
  5. add another string value
    • name: 2
    • data: path to your EF2000/PROGRAM (without quotes)
  6. Upon start, choose Scenarios and you should see something in the list.
Link to post
Share on other sites
18 hours ago, mikew said:

The most recent link is to a fresh extraction of EF2000 V2.0's did.dat, so should be good.

I needed to that as I don't know the status of my older extracted folders.

 

The file tm/marsh.tm seems damaged (it’s not 48 KiB, but 47.8). Its modification date is also a few years too early. Could that be an extraction error?

Link to post
Share on other sites

Strange. I did have trouble with tm/marsh.tm which is why I replaced it with an earlier file which I thought was the right size.

EF2000 wouldn't run with that file corrupt, so I don't know exactly what the situation is. It'll have to wait until after work...as will trying out your new version of TFXplorer. This would have been a good day to 'work' from home. :)

 

The reason I had trouble with that file is that it's the last to be decompressed. From my Python script, I call a DLL to decompress each file and return a result.

The DLL handles the file reading and writing, so once the last file is decompressed, the DLL returns OK and my script exits. Could it be that any ongoing file process is killed by Windows if the process ordering it isn't running anymore and I end up with a corrupt file?

Link to post
Share on other sites
20 minutes ago, mikew said:

The reason I had trouble with that file is that it's the last to be decompressed. From my Python script, I call a DLL to decompress each file and return a result.

The DLL handles the file reading and writing, so once the last file is decompressed, the DLL returns OK and my script exits. Could it be that any ongoing file process is killed by Windows if the process ordering it isn't running anymore and I end up with a corrupt file?

 

Depends on how it’s implemented and how the script exits. Windows’ file buffers are flushed in any case. User-mode file buffers (like buffered stdio in C/C++ programs) are flushed when the DLL is unloaded, due to it receiving a notification.

It would only be an explanation if the file handling used hand-written file buffering or if the process got terminated the hard way (i.e. via TerminateProcess()). Can’t tell given the information I have.

Link to post
Share on other sites
16 hours ago, Scorpion82 said:

Have you guys any recommendations wrt to programmes to open and edit the various extracted files, especially SSD and FAD types etc.?

Not really. The best we've done for SSD files is to convert them to text form for editing. What to edit is another problem though...

Link to post
Share on other sites

The decompression part is basically this:


from ctypes import *
lib=cdll.LoadLibrary('dera.dll')
while e < num_files:
    if fileinfo[e][6]=='RA':
        sres=99
        filnam='known\\'+fileinfo[e][1]+fileinfo[e][2]+fileinfo[e][3]      
        print ("Decompressing : "+filnam)   
        bfilnam=bytes(filnam.encode("utf-8"))
        s1=c_char_p(bfilnam)
        s2=c_char_p(bfilnam)
        sres=lib.dera(s1,s2)           
        if sres!=0:
            print ("Error processing "+filnam+" code: "+str(sres))
    e+=1 

Pretty crude, but it mostly works.

I could put a loop to waste time at the end if I can't find out the proper way to do it. :)

Link to post
Share on other sites
16 hours ago, DrKevDog said:

Looking at and comparing gtrond.ssd vs trond.ssd it is clear that gtrond's index values are 'erroneous' while tronds are correct. Further, if I compare them Block-by-Block they appear to be, basically, the same airfield super-shape, especially Block # 3 which is identical except for one entry. That makes them appear to be redundant. What does the g in gtrond designate? Is there a glide variance? If all the erroneous index containing files were culled from the ssd cohort would TFXplorer cease generating those errors?

There may not be any big conspiracy. If gtrond worked at all, it must have been with a different ssinfo.fn. My theory is that Block 5 (and maybe Block 2) are more useful to a design tool, as in game it doesn't seem to be used. Likely wrong of course. :)

Link to post
Share on other sites
1 hour ago, mikew said:

The decompression part is basically this:

*snip*

Pretty crude, but it mostly works.

I could put a loop to waste time at the end if I can't find out the proper way to do it. :)

Looks fine to me. If you use LoadLibrary(), maybe there is also a FreeLibrary() call to do a controlled unloading?

Link to post
Share on other sites

Weird. I just couldn't get it working in batch mode. If I create a script to just decompress that one file, it works fine.

Anyway, here's the link to the fixed TFX2 dataset:

https://app.box.com/s/k2h83wm20cqs594l49ezjyq6f1a88v9q

 

Now playing around with your new TFXplorer...

EDIT:...and now I see that TFXplorer's console is showing missing files that I should have extracted so I need another attempt. :(

Link to post
Share on other sites
47 minutes ago, mikew said:

Weird. I just couldn't get it working in batch mode. If I create a script to just decompress that one file, it works fine.

Anyway, here's the link to the fixed TFX2 dataset:

https://app.box.com/s/k2h83wm20cqs594l49ezjyq6f1a88v9q

 

Works great! Even more files than before ❤️

 

47 minutes ago, mikew said:

EDIT:...and now I see that TFXplorer's console is showing missing files that I should have extracted so I need another attempt. :(

 

I don’t know. It shows missing files for TAW as well, even with the vanilla did.dat … it should be fine :D

Link to post
Share on other sites

Well TFXplorer is complaining it can't find a bunch of files, and indeed these 4 seem to be missing from my extracted set:

tm\ground.tm
tm\boomy.tm
tm\cross.tm
tm\bildtex.tm

but the others in the list don't seem to exist.

 

If I use the real EF2000 did.dat, it doesn't complain that anything is missing...but there are no EF2000 scenarios.

Link to post
Share on other sites
Just now, mikew said:

If I use the real EF2000 did.dat, it doesn't complain that anything is missing...but there are no EF2000 scenarios.

 

Yep, I noticed this as well (but sadly after uploading everything). It’ll be fixed in the next version.

Link to post
Share on other sites

It's great being able to see all the EF2000 palettes in action with an easy selection process. :thumbsup:

 

I realize that TFXplorer's TFX2 rendering isn't quite as sophisticated as for TFX3, but do you know the reason that textures seem to swap as the view angle of a terrain tile changes?

This can be seen a bit with EF2000, but the effect is amplified with the hidden levels like Iceland.

Link to post
Share on other sites
7 minutes ago, mikew said:

It's great being able to see all the EF2000 palettes in action with an easy selection process. :thumbsup:

 

I realize that TFXplorer's TFX2 rendering isn't quite as sophisticated as for TFX3, but do you know the reason that textures seem to swap as the view angle of a terrain tile changes?

This can be seen a bit with EF2000, but the effect is amplified with the hidden levels like Iceland.

 

I suspect that some textures are not found and these mess up my texture cache. The problem did get better a lot after you gave me a proper marsh.tm. It’ll take me some time to nail this down, and I’d like to bundle it with the strange effects of daytime on city tiles. Going to do this after more construction work on the engine.

Link to post
Share on other sites

Slight problem in that new TFXplorer now crashes on startup. The only difference to when it was working is that I've moved the PC to another location and am using a wireless keyboard/mouse instead of wired.

There's about 3000 lines of log, and I've removed most of them starting with 'button':



TFXplorer
by Pjotr Krycztij 2020-05-05 05:49

special thanks to:
    mikew
    DrKevDog
    Home Fries
    Wombat1940

thanks to
    Benjamin Haisch (RA decompression)
    ... and the COMBATSIM.COM community


 found HID " "
  slider is z
  axis FA0A (0010): 0-255
  axis FE01 (0011): 0-3
  axis FE01 (0012): 0-3
  button FA10 (0000)
  button FA11 (0001)
  button FA12 (0002)
  button FA13 (0003)
  button FA14 (0004)
  button FA15 (0005)
  button FA16 (0006)
  button FA17 (0007)
  button FA18 (0008)
  button FA19 (0009)
  button FA1A (000A)
  button FA1B (000B)
  button FA1C (000C)
  button FA1D (000D)
  button FA1E (000E)
  button FA1F (000F)
found HID " "
  skipping HID "\\?\HID#VID_045E&PID_0745&MI_02&Col03#7&3f908cc&0&0002#{4d1e55b2-f16f-11cf-88cb-001111000030}": no axes
found HID " "
  slider is z
  axis FF05 (0502): 0-31
  axis FF02 (0602): 0-255
  button 0C00 (0000)
  button 0C01 (0001)
  button 0C02 (0002)
  button 0C03 (0003)
  button 0C04 (0004)
  button 0C05 (0005)
  button 0C06 (0006)
  button 0C07 (0007)
....
  button FDFB (05FD)
  button FDFC (05FE)
  button FDFD (05FF)
  button FDFE (0600)
  button FDFF (0601)
found HID " "
  slider is z
  axis 0E38 (0000): -32767-32767
  axis FF25 (0500): 0-4
  axis FF26 (0501): 0-65535
  button 0C00 (0001)
  button 0C01 (0002)
  button 0C02 (0003)
  button 0C03 (0004)
  ....
  button 0FFE (03FF)
  button 0FFF (0400)
  button FD01 (0401)
  ....
  button FDFD (04FD)
  button FDFE (04FE)
  button FDFF (04FF)


ERROR: no game.cfg found; using defaults

I get the same HID log with an older version of TFXplorer, but that starts up OK, so it's strange.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...