Open the images tab of almost any mod page and you’ll find the same thing: ten glamour shots of the modded game and, if you’re lucky, one vanilla screenshot taken somewhere else on a different day. People scroll past it. Not because the mod is bad, but because two stills in different scenes prove nothing, and the users deciding whether to hit download have learned to tell showcase from evidence.
An interactive before and after does what the gallery can’t: both versions of the exact same scene in one frame, with a handle the viewer drags to flip between them. The difference stops needing a caption. Making one takes about a minute with imgi, free, no signup. Getting the two screenshots right takes a little longer, and it’s the part that decides whether the comparison convinces anyone.
The capture is the whole game
A comparison is only as honest as its two frames. Everything else is upload mechanics.
Same camera, same instant, or as close as you can get. In Bethesda games (Skyrim, Fallout, Oblivion Remastered), open the console and use tfc 1 for a free camera with frozen time, then tm to hide the HUD. Line up the shot, screenshot, toggle your mod, screenshot again without touching the camera. In games with a photo mode, like Cyberpunk 2077, the pause inside photo mode does the freezing for you. No console, no photo mode? Find an interior with no NPCs and no weather, stand still, and don’t breathe on the mouse between shots.
Watch the wandering variables. Clouds move, godrays flicker, NPCs walk into frame. In Skyrim you can force a specific weather from the console (fw plus a weather ID) so both shots get the same sky. Time of day matters double for lighting mods: a comparison where vanilla is noon and modded is dusk isn’t a comparison, it’s a mood board.
PNG, not JPEG. JPEG compression eats fine grain, dithering, and sharpening, often the exact things your mod adds. If you shoot through Steam, turn on “Save an uncompressed copy” in its screenshot settings. Same resolution for both shots, obviously, and native res so nothing gets scaled.
Name the files immediately. vanilla_gate.png and modded_gate.png. Twenty minutes of capturing later, “ScreenShot47.png” tells you nothing, and publishing a comparison with the sides swapped is the most embarrassing bug a mod page can have.
Turn the pair into a slider
- Open imgi.co and drop both files in.
- Drag the handle to check alignment. If the frames drifted, recapture; a two-pixel shift reads as blur.
- Label the sides. Vanilla and your mod’s name beat Before and After, since a viewer three screenshots deep into your page has stopped reading captions.
- Publish. You get a permanent link and an embed code.
Here’s one running, go drag it:
That’s from Shadows Reworked by Tewix, an Oblivion Remastered shadow overhaul, the kind of mod that’s nearly impossible to sell with stills because the whole point lives in the transition.
Getting it in front of modders
Different platforms take different formats, and this is where most people give up one step too early.
Nexus. Mod descriptions don’t run iframes, so a live slider can’t sit inline. Two moves work: paste the comparison link with a one-line pitch (“drag the slider, same scene, all settings identical”), and drop a GIF export of the same comparison into the description so there’s something moving on the page itself. A pinned comment with the link catches the people who never read descriptions.
Reddit. Most subreddits only take images and video in posts. Export the comparison as MP4 (smoother and smaller than GIF) and post that, with the interactive link in a comment. The video gets the upvotes, the link gets the people who want to inspect.
Discord. GIF or MP4 plays inline. The link unfurls with a preview card too, so posting both costs nothing.
Your own site or a wiki. This is where the iframe embed shines: paste one line of HTML and the slider runs on the page.
One scene is a claim, five scenes is a case
If your mod touches more than one thing, one comparison undersells it and ten separate links annoy everyone. Two better shapes:
A single comparison can hold up to ten images, with a dropdown that swaps which pair is under the slider. That suits one scene rendered under five different configs, like preset intensities or performance tiers.
An album groups several separate comparisons, each with its own labels, under one link. That suits five different scenes: town at noon, cave, night exterior, a face, a failure case. Yes, a failure case. If your mod costs frames or looks off in fog, showing it yourself buys more trust than the comment section finding it for you, and it will.
For preset authors specifically, ReShade can capture the before and after of the same frame automatically; the ReShade screenshot guide covers the ini setting. And if your mod’s pitch involves performance claims, pair the comparison with the method in the graphics settings guide so the FPS numbers hold up too.
Where mod comparisons usually go wrong
- Different scene in each shot. The classic, and it reads as hiding something even when it isn’t.
- An ENB or ReShade left on in the “vanilla” shot. Check your overlays before calling a frame clean.
- Comparing across game versions or different load orders, so half the change isn’t even your mod.
- Cropping or downscaling one side. For texture and model mods especially, shoot close enough that the detail you changed fills the frame, and keep both files full size.
- Labeling with internal preset names (“v3-test-B”) that mean nothing to a stranger.
None of these are hard to avoid. You just have to decide the comparison exists to prove something, and a comparison that proves something is what turns a scroller into an installer.
The quiet payoff
A comparison link keeps working after you post it. The same URL goes in your Nexus description, your release Reddit post, your Discord announcement, and the changelog when v2.0 improves the same scene. If the images ever need swapping, the link doesn’t change. Your mod page stops being a gallery you re-upload every version and starts being one link you maintain.
Grab two screenshots of the same scene and see what your mod looks like with a handle on it.