Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - blippyp

Pages: 1 2 3 [4]
76
oh sorry, I'm in a bit of a hurry so kind of quickly read that (getting daughter ready for school also, so if I disappear, just give me a 30-40 mins and I'll be back).

edit your batch file - put that path at the end of the python line that calls the model

77
post your batch file in code blocks and your py file in code blocks so I can see them

78
your new script wants two arguments, the parameters for it is:
1) which model you want to use
2) the folder you want to start recursively start making images from

so for example:

mybatch.bat models/RRDB_ESRGAN_x4.pth H:\GUM\fieldsXCF

with this command it would use the rrdb_esrgan model and begin generating from images in anything in the H:\GUM\fieldsxcf folder (recursively), so if there's a folder in H:\GUM\fieldsxcf\testImages, then it will also generate images from that folder as well.

basically, if using the original files, just modify your batch file and put the path for your images in it after the model name, the line will look something similar to this:

python test.py models/RRDB_ESRGAN_x4.pth H:\GUM\fieldsXCF

79
@Satsuki Do you know if there's a way to configure ESRGAN so that subfolders are read as well as added to the results folder?
I've added an issue to the repo https://github.com/DmitryUlyanov/deep-image-prior/issues/43 but maybe I just didn't read and it's already possible?

I altered mine to do that, I modified this a bit, so take it with a grain of salt, but the beef and potatoes are there if you want to take a look:

Code: [Select]
import sys
import os.path
import glob
import cv2
import numpy as np
import torch
import architecture as arch
from subprocess import check_call
import fnmatch

model_path = sys.argv[1]
src_path= sys.argv[2]

if model_path == "models/RRDB_ESRGAN_x4.pth":
modelType = "ESRGAN"
elif model_path == "models/RRDB_PSNR_x4.pth":
modelType = "PSNR"
elif model_path == "models/Manga109Attempt.pth":
modelType = "MANGA"
elif model_path == "models/Interp_08.pth":
modelType = "INTERP"
else:
modelType = "UNKNOWN"

startingPath="results/"+modelType+"/"
workingPath="results/"+modelType+"/"

device = torch.device('cuda')  # if you want to run on CPU, change 'cuda' -> cpu
#device = torch.device('cpu')

test_img_folder = "LR/*"
test_img_folder=src_path

model = arch.RRDB_Net(3, 3, 64, 23, gc=32, upscale=4, norm_type=None, act_type='leakyrelu', \
                        mode='CNA', res_scale=1, upsample_mode='upconv')
model.load_state_dict(torch.load(model_path), strict=True)
model.eval()
for k, v in model.named_parameters():
    v.requires_grad = False
model = model.to(device)

print('Model path {:s}. \n'.format(model_path))

idx = 0

for root, dirnames, filenames in os.walk(test_img_folder):
for filename in fnmatch.filter(filenames, '*.png'):

dstPath="H:\\GUM\\esrgan\\results\\"+modelType+"\\"
dstDir=root.replace(src_path, dstPath)

path=root+"\\"+filename
dstFile=dstDir+"\\"+filename

if not os.path.isfile(dstFile):
if not os.path.exists(dstDir):
os.makedirs(dstDir)

idx += 1
print(root+"\\"+filename)
base = os.path.basename(filename)
img = cv2.imread(path, cv2.IMREAD_COLOR)
img = img * 1.0 / 255
img = torch.from_numpy(np.transpose(img[:, :, [2, 1, 0]], (2, 0, 1))).float()
img_LR = img.unsqueeze(0)
img_LR = img_LR.to(device)

output = model(img_LR).data.squeeze().float().cpu().clamp_(0, 1).numpy()
output = np.transpose(output[[2, 1, 0], :, :], (1, 2, 0))
output = (output * 255.0).round()
cv2.imwrite(dstDir+"\\"+filename, output)

80
WIP / Re: Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-06 18:56:47 »
Thanks LeonhartGR, I appreciate your faith! ;-P

Long story short, it looks like that faith isn't going unwarranted. I'm much more hopeful atm then a I was a couple of days ago. Unfortunately, I still can't do anything about those npc's in the game, but as for the rest of it, it's going well I think. I have completed merging my scripts for the most part and am applying a few tweaks still, but am now testing it to see if the screens produced work (and so far it looks like they will). Best part is that I think I've figured out how to auto-build the backgrounds as well. I don't think it will work for ALL the backgrounds, but for 90%'ish of them, I doubt there will be issues. The problem with that is I will need to verify by eye each screen, but there's only about 1K screens, so it's not an overly impossible task, just kind of a pain, but still FAR better than building them all by hand. The screens that don't work will simply need to be manually built, but I doubt there will be more than a dozen or two of them, hard to say atm.

81
WIP / Re: Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-05 14:41:57 »
My thoughts on the backgrounds which have NPCs around, the best thing would probably be for someone to extract the 3d models of those characters and manually put them into the scene.

I also agree that using the 3D models when possible is certainly the best method for screens like my previous post where I believe it's likely, but there are screens of npc's which I'm pretty sure have no models in the game like the one in this post. I'm pretty sure that having access to both the models as well as an actual artist is basically necessary, or at least an artist. They wouldn't even have to be super awesome, but good enough to redraw and color these characters in these scenes so they dont' look like a deformed mess. Another option I might have for the moment that would still kind of suck, but would very likely at least make them turn out better is not using the neural networking techniques on the characters and simply resizing them using normal methods and filtering them the same as everything else. They would be blockier than desired I'm sure and would still look out of place, but at least they'd kind of feel like they belong more and wouldn't appear horribly deformed. And in the future someone can always simply replace just those images as a side project.


How about you put the work you've already done into a repo, and let other's have a go?

If I actually do drop this project (which I haven't yet, just not expecting to release anything soon due to these problems, and am just letting you all know that), I'm not against posting what I've done if anyone is interested in using it as a base or to learn from or whatever. But I'd need a very large storage location. My project folder is already over 25GB's in size and will likely grow well past 30-40GB's before I finally 'toss in the towel'. I think Mega has a 50GB limit, so that might be an option assuming my source files don't grow past that. Either way, I'm sure there are options but I'm not ready to go that far just yet. I'm hoping someone will have a solution for the modeled characters, and if I looked hard enough and bugged enough people I could probably get some artist willing to help out somewhere online for the rest of the images.

I feel it's a shame to put the entire project on hold just because of these scenes.

I haven't given up, just don't see a solution at this point, and to complete this project properly I'm pretty sure I need to go over and catalogue many of the images, and there are a lot, almost 13,000 and many of the backgrounds (approximately 1K of them) actually have more than one associated with it and also needs to be documented. Initially, I felt this would be simpler, ie: extract the images, filter them, maybe apply some alpha's and be done with it, but this project is much bigger than I anticipated. I might even go so far as to dig into the Deling code to see how the creator of that is re-creating the backgrounds, whoever did that seems to have a pretty good handle on them, although even that isn't perfect. Who knows, in a month there might be a ESRGAN 2 released that will keep the alpha layering (that would be awesome), but somehow I doubt it.

Currently I'm just very discouraged, as I was hoping for something much more complete than what I'm currently looking at being able to provide, and I also thought I'd have it produced far sooner as well. I wasn't counting on my own filter process taking so long in the end either, I hadn't even considered it until I ran the AI filtering, but the math says it will take a long time.

I might just take a step back, take a deep breath and begin the long haul and instead produce the project in 'sections', like to start off, try to produce a demo area to check out like the garden and balamb that people can roam around in. That might generate interest and cause an artist or two to flock this way and be willing to help out with the screens I lack the skills to complete properly on my own, I don't believe there is a lot of them, and I'm not exactly a horrible artist either myself, so I suppose I could do them if need be, but even a 'hobbyist' artist would be far better than myself for such a task.

other's may have some ideas on how to solve the problem :)

Well, that was my primary reason for my previous post tbh, I was asking for help from anyone who might know how to do that.  :P

82
WIP / Re: Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-05 04:14:10 »
It saddens me deeply to say it, but I think I'm putting this project on hold. Possibly indefinitely, I may still find solutions, but for now, I'd say on hold with no expectation of completion at this point.

It's not that I can't complete it per se, so much as I won't be happy with the final product.

I will try to explain.

My process for filtering the images actually uses many layers upon layers, three of those layers are three different AI Neural Networking generated images that everyone is all in hype over these days. My computer spent the past 20'ish hours generating over 40K images so they're created and can be simply merged into my own filtering process at this point. However, to complete the final filtering would take my system well over 2 months according to my math, although there are shortcuts that can be taken that could easily decrease that time down to about a week and most people wouldn't even notice that those shortcuts were used. But would require me cataloging every single file into a database, which would probably take me a few days if not even a dedicated week (and honestly this probably needs to be done any way).

Although this isn't a comforting thought by itself, it's not the problem (there's still a couple others also, but again, they're not what's stopping me).

The heart of the problem, and why I'm putting this project on hold: There are many background images that include npc's standing around. They aren't modeled characters, but part of the background images. These npc's simply do not scale well and really look out of place, even with neural networking, in fact, neural networking makes them look FAR worse. We could simply choose to ignore these handful of screens, but we'd be dealing with the same issues overall that we have now. HD images mixed with low-res or uterly destroyed npc's standing around that look horrible. These graphics require an artists touch. Maybe I'm wrong, but the only way I can see to repair this is to have someone actually redraw them or use those HD models/textures that many of the characters are using in game. I don't know how to work with the models at all right now, so I can't simply position characters with the HD textures that have been made and 'rebuild the scene' with them and take a picture or however it would/should be done. The tools for this might exist and I just haven't noticed or come across them yet, but without doing this (or something similar) to bring these characters up-to-par, there's simply no point in going thru all the effort required to rebuild these screens, at least I don't think it's worth it. I could complete the project regardless and hope that someone would 'fix' these out of place characters at some point, but then their own graphics might not 'fit' with my own and feel out of place yet again, better, but still out of place.


So for now, until I can figure out how to correct the characters in the game to also do the HD upscaling justice, I'm putting this on the back-burner. I will likely keep trying to finish it when I get bored and dig into the character problem as much as I can, but so far I'm not aware of how to fix this.

There aren't many screens like this that I'm aware of, but it simply destroys all the efforts I've been doing. If anyone is aware of what I'm talking about and knows how to fix it and/or is willing to help please pm me. I've never been against working with others on this project, and at this point, it would certainly help speed things along even if the character issue could be resolved just having more people also generating the final images. To help with this task all you would need is GIMP with Python-fu and G'MIC installed. Now that I have a 'mass' storage of files to work with I'm also altering my script to build them all instead of just one at a time, so I'd have to complete that as well as merge a couple scripts into one. I will/could probably have that completed in a few days likely at most, it's what I'm currently working on.

Also if someone is familiar with the naming convention with Pupu for identifying the files, I believe there is a structure there that can be used to help assist with automating rebuilding the backgrounds, but still haven't quite figured it out - any information regarding that would be very helpful and could save me from having to go over each file and catalogue what type of file it is. The Deling program also autobuilds the files as well from what I just noticed and am now using that as a base for my backgrounds, but it requires manually opening each background and saving them, and even then, there are a few that need to be placed in a 'fix' folder for further review before running thru my filter.

Either way, I think I need a break for a bit from this project regardless. I'd love to see this completed and can now see what needs to be done for the most part. There are shortcuts, but I'd rather do it right and complete it with something I'd be happy to play with, otherwise, I think I'd rather just play the game as it was originally designed tbh.

I'm sure I'll keep working on this project regardless, but I want to do it right and slow and steady wins the race. I really would like to see this project at least come to fruition to the point where I could run around even at the start of the game with all updated graphics just to say I could see it. But those npc characters, like I said, imo, destroys any hope of a fully fleshed-out final product imo, it's not impossible to fix, I simply don't believe I have the artist skills to do it, or the modeling knowledge required to rebuild them.

83
WIP / Re: Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-03 21:20:34 »
Just my current thoughts and a fyi:

I finally got around to extracting all the field mapdata files. I ran them thru Pupu which returned 12,288 images in 925 folders. For a few days I kind of gave up hope on being able to automatically upscale all the images, but feel that I have another option after spending a couple of days going thru the last steps needed to rebuild the screens and add them to the game (the preview test videos I've been making). I am now going to start a database for all of these files and looking them over and cataloging them for which files are required to build 'the background', which files are lighting and other special effects and which files are animations for the backgrounds. As I go thru them I might discover another issue/type of image that will cause this not to work, but once I've completed organizing them all, I should be able to automatically upscale the images to produce a texture pack using whatever type of filtering/upscaling technique I wish to apply to each image.

Any of the textures which include characters though, as I've mentioned in the past, will look horrible, they simply don't upscale well. I'm hoping to find some way to fix this. I've noticed a few of the npc's have had their textures also upgraded by other people and don't have the old nasty textures attached to them, so if I can figure out how to use the models, I might be able to fix some of those, but some will likely need an actual artist to re-draw them. I can probably pull this off to at least be better than it's current state, but to get something that will 'look right' will probably need someone who can actually draw well, and not just 'kind of'.

My automated process is also designed to produce layered GIMP images of the final image so that if I have any desire to go back and alter a few images/layers used by hand after the process, I can easily alter any step used to get it to that point without rebuilding the entire texture pack again, so making minor edits like altering the characters on the screen or masking out some of the 'detail' added by some of the esrgan upscaling can easily be altered. I still certainly need to cleanup my script some more and have a few more changes to make to it, but have enough of it completed at this point that I'm absolutely confident that I can pull everything off, assuming I don't run into any more issues that I'm not already aware of (which still might happen).

84
WIP / Re: Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-02 19:38:45 »
Youtube Video - Preview 002

Altered a couple more screens, and as suggested by LeonhartGR, I also sharpened the graphics more in the Training Center.

85
WIP / Re: Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-02 07:26:38 »
Sample Video with a couple of screen updates

One of the screens unexpectedly had some of it's graphics overwritten by other textures which I clearly haven't altered yet.

86
WIP / Final Fantasy VIII - Graphical Update Mod (WIP)
« on: 2019-01-01 02:34:20 »
Graphical Update Mod (GUM)

Recently I started replaying FFVIII, which was one of my favorite old games. I downloaded it on Steam and, considering that I'm a programmer myself and absolutely love modding Skyrim, I was happy to realize that people were modding FFVIII as well.

Until now, I basically 'hi-jacked' the thread [FF8PC-Steam] Batch extraction of backgrounds but at this point feel it is time to start my own thread since my posts clearly no longer belong there. I have no idea how far I will take this project or if it will even get finished. I have completed a lot of work so far though and certainly feel it is doable and at least I can try to make all the graphics seamlessly work together with a new HD feel. I will begin with the backgrounds, but my hope is to redo all the in-game graphics so that it's all one package and just 'feels the same'.

As I'm sure you can tell from the title, I am currently calling this mod 'GUM' (I just came up with it, so it may/may not be temporary). The point of this thread is simply to keep people up to date with where I am and what I'm doing and also to hopefully get some feedback or even help from time to time from those more experienced with modding Final Fantasy games. I also consider it a 'diary' thread regarding the project's progress. So far, there have been a few people who have helped direct me in one way or another and I'd also like a thread like this to help keep track of those people so that I can add their names to a 'notable mention' list or 'contributors list', which may be helpful if this project ever actually gets completed.

A few people who currently come to mind are SaintBahamut, Yagami Light, and last but certainly not least - satsuki, who was kind enough to send me a rather lengthy and detailed pm regarding his own process.

Example comparision of the currently expected state of the images produced for this project:

Spoiler: show
Original Image


Project Image

87
For those interested, I have moved my discussion to it's own thread since my posts no longer seem to apply here.
Final Fantasy VIII - Graphical Update Mod (WIP)

88
Really liking this Manga filter so far, have only discovered a couple of images that will likely need to be 'tweaked' further imo. I have developed a process for mass processing at this point and am running it thru my initial testing. I still need to alter the process to also pull the alpha layers from the images as well and rejoin them since the Manga filter outright destroys those, and I still have many images to pull from the source data files, but definitely getting some where now. Even as is, I think this would make for a great background update.

I have processed 15 images so far, it didn't take that long tbh, once I have all the images collected and have the process in place to also preserve the alpha layers, I'm guessing I could process all the images in a day easily. I am currently running another 63 images thru the process and will be adding them to the album I'm posting now. I promised the family I'd take them to the theater, so when I get back I'll likely upload those images then. Although I'm showing at least 35 images have already been processed so I might be able to start uploading them before I leave even.

[EDIT: Those 63 images are uploading now]

Until then, here is the album:
https://imgur.com/a/mjVrfdV


89
awesome, thx, I'll look into that immediately :)

90
Ya, some images it looks good in, while others would need to be 'tuned down' on the effect. like that previous image I posted in the town, I think overall it looks quite well, but in zell's house it looks pretty bad (even considering reshade darkened it up more). But it was the easiest background to change that my current saves let me go to so I used it for an 'in-game' shot.

I really like that manga109 image you posted though, that's a damn nice filter. How do I go about getting my hands on that? I'm not having much luck finding it so far with my initial searches. From what I can tell it's a dataset provided by a college/university or something like that and tbh, it's license might not really apply for using like this from what I've read so far. Do I have to fill out that application form or is there an easier way to use it?

With just a couple quick filters the pixelation on that image can be easily removed, it really requires very little effort to fine tune it, I'd like to see some of the other backgorunds done with it also (I consider this particular background to be one of the worst I've worked with so far tbh):


With just some very minor softening and resharpening of the image produces fantastic results, and that's just one attempt without even really testing our much I should soften or sharpen it up. I think we would be hard pressed to find something better than this to work with, I'd really like to get my hands on it.

91
Background comparison between the original graphics and Project Eden, Project AngelWing (with latest AllFIX Patch) and my own Upscaling techniques.

Original


Project Eden


Project AngelWIng (including latest AllFIX Patch)


Current blippyp Upscaling


It should also be noted that I also have reshade running in these shots, which tends to make my images darker.

Current blippyp Upscaling with initial skin texture updates as well to match backgrounds

92
Final stages of designing that filter was certainly a much bigger headache than I anticipated, but here are the results so far. The only thing missing from this process currently minus the ability to do a batch filter on many images (which from initial research probably won't be hard), is the initial merging of the ESRGAN and WAIFU2 files, which I use as the base for my filtering, but that's a simple process that I'm certain will cause no headaches and will be part of my process for turning this into a batch process. The real problem now should be finishing exporting the rest of the files and rebuilding the backgrounds that are broken apart, but as I mentioned earlier, I will be using the AngelWing mod as a base for which files to build from so it shouldn't be too hard to at least get a start on that once I've exported them. There are a few minor changes from this filtering than my manual edits also, since there were some steps I did in GIMP that don't actually have a filter for it, but GIMP itself was doing whatever it was doing and I had to try to mimic what I thought GIMP was doing to the images. An example of this would be performing a white balance on an image, so some of the steps have been slightly altered, but I think it's a fair representation, if not even better in some situations, for example my sharpening effect is more effective overall, and I think I prefer my filtered process more tbh.

Original


Filtered


Just as my previous manual edits and as I anticipated, the sky required some masking from the original WAIFU2 image in order to look good, I'm not sure how much of this type of manual work would be required from the images, but it's a relatively simple process, but if I have to do 500 of them then it could obviously take a while regardless. My plan is to save the masks I generate for each image, so when the process is completed, and if I decide to make/use a different filter at any point I can simply click on my batch process and all that sky work and other manual tweaks like that will be automatically done as well with the batch. I am also still anticipating that some of the images will require some other manuals edits, if not even have the filtering on them completely altered to look decent and match the rest of the images, but I won't know more about that until I start trying to actually produce the batch and see what the results are like when I go over each image separately.

93
I've been basically programming my entire life, and have used dozens of different languages, I promise you that all you need to get started is to pick up a basic programming book for dummies in whatever language you desire to learn. For the most part, they are all the same, but there are some pretty serious exceptions to that rule. Once you've learned the basics of one language though you will notice just how similar all the languages really are. They all store memory in variables and use conditions and loops to code. How you design a program/game in one language is still relatively the same way in another language, just the syntax is different in how you type it out.

You'll be hard pressed to find a good course on how to design games though. That just comes with experience. There are some references and books out there to learn how, but for the most part you're kind of on your own. You either get how to do it or you don't. Once you've been programming long enough you will realize there isn't much difference between making a game and an application or a command line tool anyway tbh.

It sounds like you're just starting out - So if I were you, I'd just go get yourself a 4 Dummies book or start reading up on one of those tutorial sites on whatever language you want to learn. It takes time, but can be very rewarding.

Good luck, hope that helps you out.

94
Considering I've only setup ESRGAN once myself and the fact that I'm hardly a Python expert, I'm far from the person to explain/make a tutorial on how to get it working. In fact, despite the fact that the filter I'm making right now for these images is being made with Python, I've never actually used it before now so I honestly have very little to offer on the subject. What I can tell you though from what I noticed while setting it up for myself, is that it was rather confusing and which version of Python you have installed will drastically alter how you install the necessary libraries for it and which commands/methods you use to install them to get ESRGAN working to begin with. Initially I tried to install it using Python 3.7 and ran into many issues trying with that version. I'm sure someone more familiar with Python can easily explain how to install it properly with that, but for me it was difficult and gave up quickly.

In the end I used 2.7. There were three libraries I needed to install, but can't remember which exactly, but they'll easily come up as you google on how to install each library. Begin with the instructions on the ESRGAN page and from there it will make sense. I used what I believed to be package/library manager for lack for a better description called Pip. Using that, I installed all the libraries necessary to use ESRGAN. While googling on how to install each library for Python 2.7 and using Pip, you should come across some pip commands that you run from the command line that automatically downloads and installs them. I used the same process to install all three libraries, just with slightly different commands referring to each library. Once they were all installed, I simply ran ESRGAN as explained on the developer's page.

Wish I could help you more than that, but that's the general 'how to' on how I installed it myself, hope that helps and hopefully I didn't leave anything out.

95
For anyone who may actually be interested, which may be no one since no one seems to be showing any real interest, but figured I'd post this as an update regardless. I have been continuing my efforts, albeit somewhat slower on developing a mass process for filtering these images for what I consider to be an improved HD quality over WAIFU2 filtering. I still have plenty of images to export from the original data files using Unmass/Pupu, however I have begun developing a filter for automatically processing all the images. I still have to look into turning my filter into a process for mass batch processing, but have certainly made some headway so far and expect the base filter to be done shortly after the holidays. My current filter only requires a few more steps to be completed before adopting it to a mass process and ready to apply on all the images once I've determined which images need to be made, which I'll likely just use the AngelWing mod as a base to start with. Below are two images, the original image from the game at 4x it's original size without any modifications, and the second is my current expectations for my current filtering once it is completed. I have removed a lot of the filtering I was applying previously in hopes of removing some of the line art that was appearing, I think it looks better, but might alter the process again before finally doing a mass filter job. I certainly prefer this one over the last, but feel some of the detail may be missing that could be applied.

Original


Current Automatic Filtering Process


Up until this post, all the editing I've done has been manual, but this process is automatic and therefore can be easily applied to thousands of images requiring only the time needed for my system to process these images for them to be completed once they are collected and the filter I'm making is completed.

96
I am having a ton of fun with this, thanks  ;D

Waifu2


My Edit


97
And if anyone is interested what im actually using.
https://github.com/xinntao/ESRGAN

oh that's awesome, thx! I will certainly give that a go! :)

98
Actually, it is in the game, it's been a while since I fully played it though, so I just verified it in the data files to be sure. I believe it's when you're Laguna and living with what's her name about halfway thru the game.

As for the rest of your comments however, I already explained all of that in one way or another in my first post. The images used at the bottom is the game itself, one using AngelWing and the other with my own altered background, which also required being rebuilt using multiple files like you're suggesting, so I do understand what you mean but it's more time consuming than hard to do. Resizing them all individually is a very bad idea though, as that won't give a HD effect that we're all after. The hard part is determining which files should be included, but joining them is a simple matter, just do it with the originals before resizing them, export the alpha layer of the originals and resize those as well, then re-apply them (altered if need be, which is a very likely possibility as far as images like the one I posted, which should have the characters completely redone in them) before saving the final files for Pupu to import.

99
I'm realizing that posts on this board are few and far between  :-o
Figured I'd make another edit, much happier with the way this one turned out though.

Waifu Version


SaintBahamut Version


Our Combined Efforts


Considering I'm still doing this by hand and haven't wrapped this into a one-click filter yet (that would be awesome) so there might be some differences, but afaik I used the exact same process for this, however, as a final filter I applied a 'sharpening effect' to the enitre image, which really cut down on the blurryness my image added, overall I think this looks quite good and is far better then the waifu version by far.

Here's the imgur where my pictures are getting posted if anyone else is interested, plus it's easier/faster to update this then wait for a moderator to accept my posts and causes less clutter here, I'll likely keep updating it with pictures until I get bored or get a response from SaintBahamut.

https://imgur.com/a/UzpMdew

100
First off, I apologize for the length of this message.
tldr; I want to help, MY proposed image example will be at the end.

SaintBahamut, I sure hope you're still around, cause I'd really like to work together on this if you're willing. Like you, I also decided to play this game recently again and was pleasantly surprised to see that mods existed for it (I only ever played it on playstation before, but purchased it recently on steam). And just like you, I was pretty discouraged with the backgrounds among a few other things, but at the moment, mostly the backgrounds. I'm no artist or graphic designer, but I do play with GIMP quite a bit for fun, rebuilding images from scratch without using the original in the final product. I also decided to begin a project to redo these backgrounds as well, but have only been working on this now for a couple of days. It took me about a day to figure out how to extract the backgrounds and the crazy but rather ingenious process required to actually mod the textures due to the efforts of Tonberry (forgot the original designer's name), but two thumbs up for his determination on getting that accomplished.

I certainly won't claim to be very familiar with this entire process yet, so I'm going to explain things as I see it just in case I'm mistaken.

Currently, there are two major mods for textures as far as I'm aware (Eden and AngelWing), I played the game up until just before the part when you first fight Seifer and Edea so far with the AngelWing textures, but only just installed the Eden textures. As far as I understand it, Eden is apparently better, but the hashes are messed up so people just install Eden and use the hashes from AngelWing. I haven't begun playing again to really see what kind of change this has made, but apparently its what people are doing. From what I can tell (and am guessing) both of these mods basically just applied Waifu2 on all the images, I'm guessing Eden did a better job of it and/or also has some other custom edits involved.

Given the images I saw you post, I took one of the ones I thought people probably weren't liking, but also see the potential as to why you've posted this. I don't really understand what you're doing with your images entirely, but there is certainly more detail there than Waifu2 is providing. So I decided to play with one of your images using the process I've been using and I'd like to see if you also think what I'm suggesting has potential.

Before I go further I'll tell you a little about myself and what I've noticed so far from this post. For starters, the background images you're using aren't accurate, and I'll explain that more in a bit, but I promise you won't like what I have to say. I'm currently an unemployed programmer, have been using computers heavily for well over 30 years now, so I'm not a newb by any means, but I am clearly new to modding FFVIII. I have no official graphical design or artist training, but consider myself pretty good at it as far as a hobbyist is concerned despite my lack of training with virtually no artistic skill. So for me, this is a 'fun hobby', I don't claim to be an expert at this.

With that said, from what I've seen so far, I can tell you this: Altering the backgrounds for this game isn't a simple matter of pulling the textures out of the game and running a filter on them. If you haven't gotten a good look at them yet, when you do, you will soon realize that you actually have to 'build' some of these backgrounds using the textures provided, and some of them actually have 2 or more actual backgrounds that need to created and altered. This is due to the animation effects that happen in the game. An example of what I'm talking about is in the archive that was sent to you with the backgrounds. Some of those backgrounds have a map slapped in the center of the screen and clearly can't be used, and that's just one example, they get much trickier than that.

Now having said that, I'm well aware that we would need to build these backgrounds and am willing to do this, I will post an example of one of these images that I've already done as well, I still don't have any clue how many of these backgrounds need to be generated (I'm still extracting the files myself with Unmass, I really wish there was a better process for this, but nothing I've tried so far works, I have downloaded that Deling program posted up above but received the same error you did. I'm sure this would be rather simple to fix, but I'm almost done extracting them with Unmass at this point, so don't see the point in trying to solve another solution that isn't a simple download/install/click and be done with it process).

Okay, okay, enough babbling, I'll now explain/demonstrate how I think I can help you with a little 'how I got here' to begin with.

So the original game's textures are something like 320x200'ish or whatever. Tonberry wants us to enlarge them four times bigger making the average image approx 1280x896, some of the images have large widths, but are always 896 pixels tall as far as I can tell so far. The original graphics for this game are quite nice at the original resolution, the trick is getting them bigger with so little to work with however.

Waifu2 will produce an image like:


Here is your image you posted:


This is a 'current' example of what I planned on doing with my backgrounds using this same image. Personally, I think this is much better than the Waifu2 produced image on it's own, but preference varies. This isn't what I'm suggesting at this time, but what I'm currently going with.


After seeing your image and the detail on it, I decided to run it thru the process I'm using (basically a bunch of filters and layers), hoping that it would produce a better image despite all the noise some of your images were producing, but the results weren't as expected, and certainly wouldn't want to use this either.


As someone suggested previous (sorry, don't remember your name), but I did try combining your image with a Waifu2 layer. It looks better, but overall not by much, so I decided to join your image I made with my image, both run through the filtering/layering process I've been using:


Personally, I think this looks pretty damn good. It adds the details that your images are clearly providing but Waifu2 is absolutely lacking, plus the filtering that I like with my own version.

Each image has to be taken with a grain of salt. This image isn't the best, but it's not bad and with a little effort I can come up with all kinds of variations on how to make this look differently, so there's no need to insist on keeping with the process I'm currently using. I honestly wouldn't be surprised if after a time I came up with a much better process and wanted to just redo all the images, but for now, I think this looks pretty awesome considering it was designed from a 400x224 image.

Take into account that each image can have some TLC applied, in this example, I didn't really like the sky much, but preferred the sky in the Waifu2 image much more, so I masked it into the image, which required some 'manual' work, but it wasn't hard and literally only took a couple of minutes. THIS is the final image I want you to see and consider. There's room to grow here, but I think this looks great (well, for someone who isn't an artist and compared to the Waifu2 image most of us are currently using anyway).


So after all this nonsense, what am I suggesting? I like your images, I want them. From what I understand you're using some commercial filter that I don't have, at the very least, I would love to hear what it is and how you're using it. If you're interested I'd like to discuss this further on Discord or something, create a mega account we can use to combine our efforts, build all the screens we can, and if successful simply provide it as a texture package.

Initially, the hard part as far as I can tell so far is figuring out how to generate all the proper images, which isn't impossible but isn't super easy either. Once we have those, I don't know what you need to do to the images, but I'd like a copy of every image so I can edit them as well. Currently, I'm manually applying the filters and layers, but once a process is decided on, I can turn that entire process into it's own 'filter' in GIMP with a little effort, but even if there's problems with that and I have to do them all by hand, I'm willing to do that, or would at least love to try.

Now I'm gonna make you mad though, but feel it needs to be said. Here is an example of the original game using AngelWing:


This is a pretty complicated background. For one, it's a duplicate (it's during the time when you're about to confront Seifer and Edea for the first time), I haven't verified it yet, but there is another background in the data files for this area and it has a moon and the lights are brighter, I'm assuming it's for any other time other than this moment in the game. See all those people there? They all have images for them as well, and they're really nasty, and I'm guessing there is a lot of this that needs to be fixed up. They've been clearly cleaned up I think, but not by much, there isn't much to go on, which is a huge problem as far as making this game look 'complete'.

I rebuilt this background as well, I'll show you what will happen with it, it looks good I think for the most part, but I would go back and darken the image a bit and/or remove some of the filtering I applied for some of the areas (the bushes mostly, maybe the sidewalks).


There are three 'kinds' of textures here for the npc's standing around, some are horribly blurry, that was due to lazyness on my part, I modified the background and simply ran the rest of the images thru a simple resize in GIMP (didn't even use Waifu2, it takes a while), they're obvious, at the very least, we can make them look like they did in AngelWing though. Look at Squal and Irvine though, those characters are done really well (much better than I can do). I also ran one of those images thru my filter though which you can see on the guards and the two people in the lower left corner the best. IMO, this looks better than what was done by AngelWing, but they still look absolutely horrible now with an updated background.

I don't really have much of a point with these last two images, except that once this can of worms is opened (redoing the backgrounds), all those little details (sprites/animations) in the game really begin to stand out. I suppose the point I'm making here is that doing the backgrounds is only a step, maybe it's all we'll do if you're willing, but those other details will also have to be looked at and will drive people nuts (I may even make an attempt at actually redrawing them myself afterwards) just so all the graphics look and feel the same in the game, might even have to degrade the images of the characters like squal a bit just to make them match the rest of the game.

Anyways, this is a really long message. So I'll stop for now and hope to hear from you in some way or any feedback from anyone would be cool even. Even if SaintBahamut doesn't want to work with me on this, I'm going to continue this myself either way, and with a little time and effort, who knows what I'll come up with. But as it stands, I would really like to use SaintBahamut's images as well, I just think they look much better combined with mine.

Pages: 1 2 3 [4]