GeDoSaTo - Downsampling from Infinity and Beyond!

What the fuck is with all this fear-mongering?
I assume that when people hear the Word "Banned" they assume that means your entire account gets nuked permanently (which i must stress does not happen) .
I do can understand the fear of losing your Steam account that you have perhaps spend 1000's of money on (Even though that Fear is totally unfounded even if you got a ban which is worst case scenario ) so i can see where people are coming from .

But as some have already said and let me repeat myself : receiving a VAC ban does not mean you lose access to your account or all your games ,but merely lose access to the online play of games that Have VAC enabled .

In a way , getting VAC banned on Steam is actually less severe (depending if you play multiplayer or not) than getting Banned on Gaf. :p
 
I assume that when people hear the Word "Banned" they assume that means your entire account gets nuked permanently (which i must stress does not happen) .
I do can understand the fear of losing your Steam account that you have perhaps spend 1000's of money on (Even though that Fear is totally unfounded even if you got a ban which is worst case scenario ) so i can see where people are coming from .

But as some have already said and let me repeat myself : receiving a VAC ban does not mean you lose access to your account or all your games ,but merely lose access to the online play of games that Have VAC enabled .
In a way , getting banned on Steam is actually less severe (depending if you play multiplayer or not) than getting Banned on Gaf. :p
It's even narrower than that in a lot of cases. For most non-Valve games, you only lose access to online play in that specific game. Dark Souls II apparently blocked/blocks you for having any VAC bans at all, but they're treating that as a bug.

This has been mentioned quite often. Where is the "evidence" or records of it being 2 weeks in duration?
Valve themselves say that there can be days or weeks between a detected cheat and a VAC ban. It's to make it harder for people to work around cheat detection if they immediately know what gets them banned.

https://support.steampowered.com/kb_article.php?p_faqid=589
 
In other, more interesting news, I tested some games.

I installed Dollar Dash just to test the app on it. Another Unreal Engine game seems to work fine. Bunch of max quality 4K jpgs here.
Didn't test further because god what an awful game.

Games that I tested that allowed me to select the resolution, but didn't scale the image down so I ended up only seeing the center of 4K image:
Broken Age
Broken Sword 3
Aquaria

Has anyone managed to get any Unity engine games working? I tried Ittle Dew and Erica Reed, Ittle Dew didn't get past the unity launcher, Erica Reed got in the game, played the intro video and then crapped out.
 
So about VAC: Yes, you could be banned for using GeDoSaTo.

Personally, I consider that fairly unlikely though. GeDoSaTo makes no changes outside of its own dll, and most significantly it does not alter the game state in any way. I believe VAC is selective enough to know the difference.

But don't blame me if you do get banned - the possibility is not 0.

Games that I tested that allowed me to select the resolution, but didn't scale the image down so I ended up only seeing the center of 4K image:
Broken Age
Broken Sword 3
Aquariao
Interesting, I'll have to have a look at one of these some time, I never had that error patter so far.

Has anyone managed to get any Unity engine games working? I tried Ittle Dew and Erica Reed, Ittle Dew didn't get past the unity launcher, Erica Reed got in the game, played the intro video and then crapped out.
M&M X is Unity and works.
 
Huh, after I disabled the Gedosato AA to use only the inspector settings, as well as turning off the game DoF and AO for the tool and inspector to hopefully override respectively, I learn that I can run this rendering at 4k resolution at 60 fps now. I wonder if I had some bits fighting against each other.

I'm still pretty green at this tweaking thing, but I enjoy it a lot.
 
Huh, after I disabled the Gedosato AA to use only the inspector settings, as well as turning off the game DoF and AO for the tool and inspector to hopefully override respectively, I learn that I can run this rendering at 4k resolution at 60 fps now. I wonder if I had some bits fighting against each other.

I'm still pretty green at this tweaking thing, but I enjoy it a lot.
Wonderful to hear. Tweaking is actually oddly enjoyable. I always try and attain the perfect performance and IQ. It takes time, but gotta put all that hardware to work.
 
I thought VAC only banned you from that game, not all VAC enabled games. So if this gets you banned, then no more Counter-Strike too?
You'll be banned from games that use the same engine/executable thingie. If you get banned in Counter-Strike: Source, which is hl2.exe then you'll be banned from all hl2.exe games.

If I get VAC banned I won't care too much but that "VAC ban(s) on record" text on my profile is going to look pretty bad. Or we might have a MW2 situation and get some free games out of it. :)
 

Htown

STOP SHITTING ON MY MOTHER'S HEADSTONE
Certainly, but it doesn't answer the most important question, whether or not GeDoSaTo will get you a ban. I can understand why they wouldn't, though.
It will probably not get you banned.

On the off chance that you are banned, that ban essentially means they won't let you on the DS2 multiplayer servers.

Relax.
 
It will probably not get you banned.

On the off chance that you are banned, that ban essentially means they won't let you on the DS2 multiplayer servers.

Relax.
As someone who is on NG+++ on PS3, I decided that I don't really care if I get banned for using GeDoSaTo on PC. I am playing it again mostly for the graphics, so i am going YOLO on the VAC ban.

PS. Durante, praise the sun. Your third release fixed the stuttering issue I have been experiencing! 4K gaming baby!
 
I am very new to modding and am a bit lost. Can someone tell me what I need to enter to downsample from 1440p to 1080p in the ini file?

I see all this resolution width and height stuff and haven't a clue what I need to change things too.

I have a 660 so I cant go as extreme as the default but am unclear of all that I need to change.
 
I am very new to modding and am a bit lost. Can someone tell me what I need to enter to downsample from 1440p to 1080p in the ini file?

I see all this resolution width and height stuff and haven't a clue what I need to change things too.

I have a 660 so I cant go as extreme as the default but am unclear of all that I need to change.
In the .ini file look at the section that says rendering resolution and look at the two lines that aren't preceded by a hashmark. So change the first number to 2560 and the second to 1440. Then go down one section to the output resolution and make sure that the two lines that don't have a hashmark preceding them match your monitors resolution. Start the game up and in the main menu go to the video settings and choose the 1440p resolution. Now you're set and should be downsampling.
 
As others have reported, with alpha3 it seems SSAO and DoF aren't working (if I enable them via the in-game menu they certainly work).

However, more interestingly, when I set GeDoSaTo to downsample from 2560x1600 to 1680x1050, my framerate takes a pretty big hit. But if I use my normal method of downsampling (having used the registry trick from 12.8->13.1 catalyst drivers), my framerate is terrific, and it looks sublime.

Much more testing would be required to ensure I was using GeDoSaTo correctly in the first place (I'm sure alpha2 was working flawlessly - and there are many other settings that may have caused the framerate drops), however for now atleast, I've gone back to default Dark Souls 2.

Have a flashed 6950->6970, for reference.
 
Few more downsampling tests.

Sin Episodes seems to work for me, I saw on the list that it didn't before. But then again that list appears to have some other problems too, so maybe it's not reliable? I dunno. Maybe someone messed with the list or something.
Anyways, that source options menu is tiny, any larger resolutions and it gets a little crazy. Full max quality 4K jpgs here.
These games only show black screen after changing resolution. No crash and the game keeps running as the music keeps playing and I can hear the cursor moving in menus and such.
Dreamfall
Tomb Raider Anniversary
Tomb Raider Legend



Oh and hey Durante, even though the screens are sorted neatly in their own folders, would it be possible to put the exe name in the file name as well? Maybe even make it configurable.

Also, you said earlier that the eventual goal is to make this plugin based, so more games can have game specific settings, right? In the introductory blog post you also mention that "it uses a far more solid injection and interception method than my earlier efforts". So I'm curious, would there be any benefit at all to eventually "port" DPfix and DSfix to GeDoSaTo as plugins? Is the more solid injection and interception a "big deal" in that sense? Any other potential improvements, like new SMAA and SSAO code?

And I fully realize that this is probably ways of in grand scheme of things anyway, but I figured I'd ask anyway.
 
So the only visual effect setting in-game that should be turned on is anti-aliasing, correct? I thought I read that has to be enabled in order for GeDoSaTo to work.
Correct, you must leave the in-game anti-aliasing turned on. Everything else should be turned off in-game if you use GeDoSaTo's equivalent feature.
 
I have been using tools that work in a similar fashion as GeDoSaTo for years in VAC enabled games and I have never been VAC banned, I would be extremely surprised if anyone got banned for it, it would be super weird.

The tools I use BTW are WidescreenFixer and Flawless widescreen, they fix/modify the FOV of the games they support to play it in triple screen resolution.

I have more than 500 hours of Modern Warfare 2 using Widescreen Fixer to play Multiplayer in triple screen and have never been banned.
 

Htown

STOP SHITTING ON MY MOTHER'S HEADSTONE
As others have reported, with alpha3 it seems SSAO and DoF aren't working (if I enable them via the in-game menu they certainly work).

However, more interestingly, when I set GeDoSaTo to downsample from 2560x1600 to 1680x1050, my framerate takes a pretty big hit. But if I use my normal method of downsampling (having used the registry trick from 12.8->13.1 catalyst drivers), my framerate is terrific, and it looks sublime.

Much more testing would be required to ensure I was using GeDoSaTo correctly in the first place (I'm sure alpha2 was working flawlessly - and there are many other settings that may have caused the framerate drops), however for now atleast, I've gone back to default Dark Souls 2.

Have a flashed 6950->6970, for reference.
yeah 0.3 definitely breaks all AO, and the problem is not in the new VSSAO2.fx file

Geforce 560Ti here

can't even really use 0.2's version of working SSAO because it leaves these weird faint dark horizontal lines across the screen in addition to doing what it's supposed to do
 
asked a few times in the Dark Souls II thread already, nobody seems to know:

should I disable SSAO in GeDoSaTo if I use HBAO+ with the help of Nvidia inspector?
 
So I'm curious, would there be any benefit at all to eventually "port" DPfix and DSfix to GeDoSaTo as plugins?
Yes, there would besome minor benefits, but not nearly enough to justify the huge amount of work for almost the same end result.

Is the more solid injection and interception a "big deal" in that sense?
Not really, the old method works for those two games. It's just much more important for a generic tool.

I have been using tools that work in a similar fashion as GeDoSaTo for years in VAC enabled games and I have never been VAC banned, I would be extremely surprised if anyone got banned for it, it would be super weird.

The tools I use BTW are WidescreenFixer and Flawless widescreen, they fix/modify the FOV of the games they support to play it in triple screen resolution.

I have more than 500 hours of Modern Warfare 2 using Widescreen Fixer to play Multiplayer in triple screen and have never been banned.
That's good to know.

asked a few times in the Dark Souls II thread already, nobody seems to know:

should I disable SSAO in GeDoSaTo if I use HBAO+ with the help of Nvidia inspector?
Yes.
 
Doesn't seem to work with Child of Light at the mo despite playing around with all the settings :(

Though I'm sure Durante is going to pick that game up at some point in time.

Using regular old bog standard driver downsampling really brings out a lot in the artwork and in the "field" your character is in 3D so really cleans up the aliasing and that's just at 3200x1800.
 
That's really surprising, it works perfectly with Rayman Legends and I thought they use the same engine :/

I will pick up Child of Light by Friday, may have a look.
 
Yeah Legends and Origins are both perfect and I went back and tested them without altering anything to my .ini file and the resolution showed up in both of them but not CoL.

Might be me being silly at my end :/
 
Hahaha yeah tried both and didn't show anything apart from what I have set through Nvidia Control Panel. It shows all my other custom resolutions so why it doesn't pick up GeDoSaTo is beyond my capability. Definitely a DX9 game as well.

Game still looks gorgeous at 3200x1800 though (although there's a black fade in battle transition effect that doesn't scale higher than 1920 in width which is a bit jarring) so I'm sure you have more important things on your plate.
 
I've been doing some more work on the VSSAO shader for this (if anyone is interested).

I still can't figure out why some are reporting that the latest vssao version is causing them massive performance drops over the previous. I've been benchmarking both and I've the same performance with both of them.

Anyway, I've been experimenting with using depth as the base for the sampling radius, instead of hard coded values (eg: 0.2). So far, it's turned out quite nicely, at no extra performance cost, that I've noticed.

Some comparison screens below (raw AO output to clearly show differences.)

Previous Sampling


New Sampling

Previous Sampling


New Sampling
 
Very nice work Asmodean. I did get a noticeable framedrop when I tried your VSSAO a while ago on a R9 280X. Have you uploaded this version yet? I'd love to give it a try and see if it (still) has impact on the performance.
 
[Asmodean];109963582 said:
I've been doing some more work on the VSSAO shader for this (if anyone is interested).

I still can't figure out why some are reporting that the latest vssao version is causing them massive performance drops over the previous. I've been benchmarking both and I've the same performance with both of them.

Anyway, I've been experimenting with using depth as the base for the sampling radius, instead of hard coded values (eg: 0.2). So far, it's turned out quite nicely, at no extra performance cost, that I've noticed.

Some comparison screens below (raw AO output to clearly show differences.)
Damn!

I hope i can get the GeDoSaTo AO to work, if that's the result.

So far i'm stuck using the basic game AO, and it's pretty shitty, almost thinking about simply disabling it.
 
I can upload it for testing a bit later. I've quadruple-checked it for performance, and unless there's black magic going on somewhere, on some people's systems. I simply can't see why it's causing such an fps hit for some lol.
On my gtx 680 with latest beta drivers - I've 2% less GPU useage with my latest version, over the GeDoSaTo v0.2 VSSAO(the version people are saying has better performance for them).

Unless I'm missing something, I'm stumped >.<

Edit: oh yeah, Durante - it would seem that it doesn't need the skybox y-vec cutoff anymore since you worked on getting AO output in the base HDR pass. I just kept in there anyway to be safe, as it does no harm. The 0.3 changes work excellently my myself so far by the way. I'm still kind of o_O that they have no shadow buffer in some areas.
 
There seem to be quite a few people reporting that VSSAO doesn't work at all with 0.3 - is this widespread or is it just affecting a few people? What's the commonality? Anyone with an Nvidia card find it's not working? I'd quite like to have VSSAO working in my game, having to make do with the default at the minute. There's not really a good way to find this out other than a load of boring 'works for me, my specs are.../doesn't work for me, my specs are' type posts.

So I'll start:

Doesn't work for me, my specs are: ASUS Radeon 6950 on the latest 14.4 WHQL drivers, 16:10 monitor (1680 x1050).
 
I have a "way out there" idea about the issue. Zenball, or anyone suffering from it, can you try running the game at 1280x720 fullscreen and see if you get AO?

[Asmodean];109963582 said:
I've been doing some more work on the VSSAO shader for this (if anyone is interested).
I achieved asimilar effect by fixing (or at least improving) the linearization of the depth.

[Asmodean];109971190 said:
I can upload it for testing a bit later. I've quadruple-checked it for performance, and unless there's black magic going on somewhere, on some people's systems. I simply can't see why it's causing such an fps hit for some lol.
On my gtx 680 with latest beta drivers - I've 2% less GPU useage with my latest version, over the GeDoSaTo v0.2 VSSAO(the version people are saying has better performance for them).

Unless I'm missing something, I'm stumped >.<

Edit: oh yeah, Durante - it would seem that it doesn't need the skybox y-vec cutoff anymore since you worked on getting AO output in the base HDR pass. I just kept in there anyway to be safe, as it does no harm. The 0.3 changes work excellently my myself so far by the way. I'm still kind of o_O that they have no shadow buffer in some areas.
I think keeping it is the best choice, it should actually speed things up (skipping all that sampling for sky box pixels).