Post by mentalmachine on Dec 12, 2020 18:38:03 GMT 10
Hi all, very new here so please forgive me if this should be in a new thread, however I thought this was an acceptable thread given a few others seem to have the same issue as me. In short, I am having trouble applying the toolset to my files without causing the game to hang and seemingly crash (the game’s memory usage per Task Manager stops dead) when entering the area corresponding to the `r101.udas.lfs`file I am editing.
This includes simply taking the `r101.udas.lfs` (and renaming the original to ensure it does not get loading by default e.g. rename to `r101.udas.lfs.bak`), extracting via the `LFS Extractor`, unpacking the produced `r101.udas` file via `RE4UHD_UDAS_Tool` and then simply repacking again (literally running extract and repack in sequence) via `RE4UHD_UDAS_Tool` and copying the rebuilt `r101.udas` file back to the folder that contained the original `r101.udas.lfs` file.
I am testing the editing of the udas file by starting a new game, proceeding through the first area as per normal, and then upon press E/A to enter the r101 area/Village proper, the game shows the door transition screen affect, but then freezes on the frame before the proper `loading` screen appears, e.g. it simply hangs forever on the zoom effect. I should stress I have not attempted this in other areas, but given I wish to modify the `r101.udas` file at some stage, seems critical to get the most basic case working.
OS is Win7 (I know ...), and the game is the latest Steam version. The toolset I am using is freshly downloaded from the current link on Page 1, however I do have a few older versions of the Toolset that all produce the same results.
For completeness I am still hitting the issue even after running a `Verify Integrity of Game Files` via Steam, and following the same procedure outline previously *without* applying the 4GB patch, nor installing the `UltimateTrainerCompanionByRaz0r`, nor using a v1.0.6 exe (am running the default Steam exe so I believe v1.0.10), so as clean installation as possible (I did try previously with all of these installed with the same hanging/crashing outcome).
For some firmer details, here is the size of the `r101.udas` files before and after extracting/repacking (using the `dir` command via `cmd`): Before: 10,049,280 Bytes After: 10,028,704 Bytes
And the MD5 checksum via the `Certutil` tool (e.g. `CertUtil -hashfile <file> MD5): Before: b4 f2 79 4e c9 6c f7 cd e6 b3 65 79 b6 c0 bf aa After: ed 97 94 2f a4 ef c6 3c a1 78 eb 38 46 94 4c 8e
Without access to the source code and knowing what the expected udas file structure or MD5 is meant to be, I am not sure what is next - if I have left anything unclear please let me know, I should state that the clean installation plays normally and I can access the Village proper without hanging.
Greetings! Thank you for a very well written query (I really appreciate teh detail you put into this, as it makes things easier).
I think everything you are doing is perfectly fine. I suspect there is a problem with the UDAS tool as several users have reported that the files arent working after unpacking/repacking.
Could you please upload the tool you are using and provide a link here. I want to observe the file in HEX. I suspect there is something wrong and looking at the HEX will show me if there are any differences with the working tool I have now.
Last Edit: Dec 12, 2020 21:02:01 GMT 10 by Mr.Curious
Are the toolsets versioned tagged in someway I am not aware of (beyond the usual `-?` or `-h/-Help` cli args)? I have several existing versions of the toolset (all cause the same issue/produce the same repacked UDAS file), however I cannot strongly say when I obtained them which would make diagnosing when the issue was introduced harder on your/SoP's end.
Thanks for doing this. The tool you sent is the exact same file (byte for byte) that I have. I tested this tool last night by:
extracting the r101.udas.lfs with the RUN.bat file for the QuickBMS tool for removal of the LFS compression. Bat file contents are as follows:
echo for %%a in (*.lfs) do quickbms.exe RE4_UHD_LFS_Extract.bms %%a del *.lfs
After that I made 2 copies of the untouched r101.udas file and unpacked/repacked one of them, then compared both files in a hex editor. The files were identical. I also downloaded the original UDAS tool from the Resident Evil 4 Ultimate HD Model Importer/Exporter Toolset and compared it with the one in the 2018 distribution and see that it is exactly the same file byte for byte.
So at this point we have two users using the exact same tool getting different results. I can only assume that this must be a Windows issue, or that the LFS tool is interfering.
Could you please use the bat file and remove the LFS from the r101.udas.lfs file and send the decompressed version for me to look at. I need to look at all steps of the process here. Make sure not to use the UDAS tool on this file, just remove the LFS compression.
As per your query about versions. I did not include release versions of any kind. I simply updated the download links with new files when something was needed (like adding new tools, or fixing a file etc). I have reached out to Son of Persia (the creator of the these tools) in hopes that we get to the bottom of this, as there have been several users with the same issue.
Last Edit: Dec 13, 2020 6:35:19 GMT 10 by Mr.Curious
In addition, I went ahead and was curious if anything at the binary level of the two files was so out of place that it was noticeable to someone like me who isn't as familiar with the UDAS file structure - links below are screenshots of side by side comparisons of the two files in HxD (been a while since I played with binary data, so forgive me if my terms are wrong).
In the pics, left is the original UDAS and the right is my repacked UDAS.
First link is the start of the files, where the header is clearly reversed as mentioned by someone else previously, however you can also see there is nearly 20 Bytes of missing (I assume) padding before the start of data e.g in the original can see '0x25' value at the 0x420 position whereas in my 'repack' version it is at the 0x400 position.
Second link was me curious to see how much data had been truncated off the end given the file size difference, and you can see from the final offset on the right vs the left just how much has been snipped off. Interestingly, it seems that only those 20 Bytes I mentioned from before are the only missing Bytes (though this doesn't mean a heap of other Bytes have been flipped or written incorrectly, etc) given the start of the 'D8 3F 4A' Bytes (position 0x99028D - 0x99028E on the left and 0x99026D - 0x99026E on the right). But there is clearly something severe going on, as the right ends at offset 0x990690 while the left ends at offset 0x9956F0.
The above is probably overkill, but beyond sharing the files and without access to source code I am not sure what else I can do to help out - most of my dev experience lately is more in the Java/Python space and has been quite a while since I did anything this low level (and even then that was Embedded C development).
Post by Mr.Curious on Dec 13, 2020 12:32:00 GMT 10
ok thank you for taking the time for a good comprehensive reply. To make sure there are no errors I took the liberty of analyzing the files you sent with those that I am using. Here are the results:
My Version Here is my original r101.udas with the header with no edit. Only LFS tool:
20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 00 00 00 00 00 A8 98 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00 00 00 00 00 AC 98 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 FF FF FF FF
I repack, file is exactly the same
Your Versions Here is your original r101.udas with the header with no edit. Only LFS tool:
File: r101.udas.orig(link) CA B6 BE 20 CA B6 BE 20 CA B6 BE 20 CA B6 BE 20 CA B6 BE 20 CA B6 BE 20 CA B6 BE 20 CA B6 BE 20 00 00 00 00 40 AE 98 00 00 00 00 00 20 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00 00 00 00 A0 02 99 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 FF FF FF FF
As you can see above, the Endian order of the header has been swapped and there is a variance near the end.
File: r101.udas.repack (link) 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 20 BE B6 CA 00 00 00 00 40 AE 98 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00 00 00 00 40 B2 98 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 FF FF FF FF
As you can see in this version, the Endian order has been restored, however there is variance between this file, your original, and my original file.
Possible Scenarios for the issue Upon doing some more tests, if I repack your original file, it changes from the original, (but as seen above this does not happen with my files), so this leads me to believe that:
1 - the original file version that we both start with (with LFS extension) is different. I think you said you have 1.1.0 whereas I am using 1.0.6.
2 - Our LFS tool is different, or is acting differently.
So I think the first thing we should do is to compare the original compressed r101.ufas.lfs files and see what is going on. If they do not vary, then we will compare the LFS tool, but for now please just send over your r101.ufas.lfs file.
As suspected it seems that the file is being changed by
I freshly downloaded/installed RE4HD from Steam the other day, so whatever the latest version is on Steam is what I have, which I had thought was v1.0.10 (as I had thought that was the latest version and hence the version on Steam), though as stupid as this sounds I am having a bit of trouble confirming the version on my disk - Steam update page mentions v1.0.6 as the last update, and via the properties window on the exe it is 'apparently' `File Version: 220.127.116.11` ...
Post by Mr.Curious on Dec 13, 2020 15:34:06 GMT 10
You know, as a preemptive measure you could simply get the RE4HD project UDAS files and work with those. THose should be the baseline for all stage mods anyways as ALbert has made some changes to the pack files and tweaks to some of the rooms.
Post by mentalmachine on Dec 13, 2020 19:01:33 GMT 10
Okay, this is getting weird.
If I take your r101.udas.lfs (for simplicity just going to prefix yours with 'mc.' when discussing it), Run LFS Extractor to get mc.r101.udas, can load that in game just fine. If I THEN take mc.r101.udas, unpack and repack with the UDAS tool and apply it, I get the same hanging issues. The file also looks like the same byte manipulation has been applied to it from a quick once over.
I repeated this twice just to make sure I somehow didn't use the wrong files.
For completeness have attached the files below, first is mc.r101.udas.lfs simply LFS extracted and the second is mc.r101.udas extracted and repack via UDAS tool:
Post by Mr.Curious on Dec 13, 2020 19:54:29 GMT 10
So you have:
the same file as I do the same tool as I do but your tool changes the file while mine doesn't.
The only thing I can imagine now is that it could be the LFS tool. Please upload all the files that are in the LFS folder in ZIP file. I will compare all the files to the ones I have on my machine. If everything is identical again, then we are in for a big mystery.
blackboar101: help me complete the mod~there are obstacles i cant handle myself!!
Aug 6, 2023 9:28:27 GMT 10
joaobrazil: I just found a rabbit hole
Aug 10, 2023 2:49:06 GMT 10
zorlin3321: the forums not dead, it just doesn't have that many people because this site competes with nexusmods in search engines and nexusmods will beat you every time. not really fair but this place provides exclusive content that's not on nexusmods so i visit here
Aug 11, 2023 9:48:51 GMT 10
Drax888: Olá a todo. Pessoal alguém aí sabe como fazer Mods de Scolp para o Resident 4 UBSOFT por que eu Tenho um Mod de Rifle Semi Altomático que não tem Sclpe e gostaria de poder adicionar uma Scolpe para ela, aguem sabe como fazer?
Aug 20, 2023 19:40:17 GMT 10
joaobrazil: Hello drax
Aug 23, 2023 3:37:05 GMT 10
NuKe5369: Its definitely been very quiet yeah
Aug 24, 2023 0:30:26 GMT 10
maxfabian: Chris and Sheva's frames all fire continuously
Remodeled items possessed
All weapons possessed
Changed Chris and Sheva's knives
Unlock All Chapters and Professional Mode
Have All BSAA Emblems
Aug 24, 2023 4:34:17 GMT 10
maxfabian: Help pls want give but can add name coolkid2d-
Aug 24, 2023 4:34:56 GMT 10
NuKe5369: Anyone that takes commissions ? If so pls msg me on here so we can talk
Aug 25, 2023 5:37:49 GMT 10