How to Share House of the Dragon Screenshots Without Spoiling the Episode 6 Death

Crop out the kill shot, keep your reaction, and save the file at quality 80 or above — that's the whole problem and the whole fix for sharing House of the Dragon Season 3, Episode 6 screenshots right now. The episode dropped July 26 on HBO Max and opens with the death of the character the press has been calling "the realm's greatest warrior," all within the first ten minutes. Every group chat in the country currently has someone who hasn't watched yet, and someone else who's already screenshotted the moment. Handling both halves of that — the spoiler and the compression — takes about ninety seconds if you know what to crop and what quality setting to pick.

The spoiler problem is a framing problem

The instinct after a scene like that is to screenshot the whole frame and fire it off with a caption. Don't. The death happens early enough in the episode that plenty of people watching on their own schedule haven't gotten there yet, and a full-frame screenshot in a group chat notification preview is enough to ruin it before anyone even opens the message.

The fix isn't to avoid posting — it's to post less frame. Two options, depending on what you're trying to share:

• You want to share your reaction, not the scene. Crop tight on your own face-in-camera reaction, a text overlay, or a neutral shot from earlier in the episode. Nobody needs to see the actual moment to understand you're losing it over it.
• You want to share the scene, but strip the identifying detail. This is where cropping alone sometimes isn't enough, because the spoiler might be baked into the composition — a character's face, a specific object, a wide shot where the who is unmistakable even from body language.

For the second case, don't reach for a blur. A Gaussian blur that looks opaque to the human eye is still, in a lot of cases, partially reconstructable — there are tools built specifically to reverse light-to-moderate blurs, and it's not worth the risk over something people are this invested in. A solid color block does the job properly: the pixels underneath are gone, not obscured. Optimage's watermark tool lets you place an opaque rectangle over exactly the part of the frame you want hidden, in the browser, with no account and no software install. If the spoiler is confined to one corner or edge of the shot instead, cropping it out entirely is the cleaner option — the person or object just isn't in the image anymore, so there's nothing to reconstruct.

The general playbook here — when to crop versus when to block — comes up with any photo where you need to hide part of the frame without editing software, and it's covered in more detail in Blur Faces in Photos Online for Free.

Then there's the compression problem, which is separate

Even once you've solved for spoilers, there's a second issue that has nothing to do with plot: House of the Dragon's cinematography is dark by design, and dark, high-contrast footage is exactly what standard compression handles worst. Episode 6 leans heavily on torchlit stone interiors, night exteriors, and dragon-fire lighting that throws harsh, localized brightness against deep shadow. That's a gorgeous look on a calibrated TV. It's brutal on a JPEG encoder set to default settings.

Here's the mechanism, in short: JPEG compresses images in 8x8 blocks and rounds off small brightness differences it treats as negligible detail. In a normally lit photo, that rounding is invisible. In a scene where most of the frame sits in near-black shadow with a narrow band of torchlight cutting through it, those "negligible" differences are the actual texture — the grain of stone, the falloff of firelight, the shape of a face half in shadow. Round them away and you get flat, blocky patches instead of a gradient, plus visible banding where a smooth dark-to-black transition turns into distinct stepped tones.

Social platforms make this worse before you even get involved. Discord, X, and group messaging apps all apply their own compression on top of whatever you upload, usually tuned for average daylight photos rather than moody prestige-TV lighting. A screenshot that already looks slightly rough gets a second pass of degradation on the way to your friend's screen.

The setting that actually fixes it

Don't rely on a platform's default upload compression for this kind of shot. Compress it yourself first, at a quality level high enough to keep the shadow gradients intact:

Format	Quality setting	Result on dark torchlit scenes
JPEG, below 75	Default social auto-compression	Visible blocking and banding in shadow areas
JPEG, 80–90	Manual, before upload	Gradients mostly hold, moderate file size
WebP, 80–90	Manual, before upload	Cleaner gradients than JPEG at a smaller file size
PNG	Lossless	No banding at all, but files run several times larger for no real benefit here

The practical rule: stay above roughly quality 75 for anything from a night scene or torchlit interior, and treat 85 as the safer default rather than the ceiling. If the platform you're posting to accepts WebP — most do now, including Discord and X — prefer it over JPEG at the same quality number. WebP's prediction-based encoding is simply better at holding onto subtle tonal steps in near-black regions, which is precisely the content Episode 6 is full of. Optimage's compress tool lets you set the exact quality level and output format before you upload anywhere, rather than trusting whatever the destination app decides to do to your file.

This is the same underlying issue that shows up with any prestige show shot dark and moody rather than bright and flat — the mechanics are covered in more depth in Silo Season 3 and Why Underground Screenshots Look Like Mud, and a similar night-scene compression problem comes up in Star Trek: Strange New Worlds Season 4 for its own darker sequences.

Put both fixes together

The full workflow for an Episode 6 screenshot you actually want to post, in order:

1. Take the screenshot at full resolution — don't crop or compress inside your screenshot tool, do both steps separately afterward.
2. Decide what needs to be hidden: crop it out if it's at the edge of the frame, block it with a solid fill if it's central to the shot.
3. Compress what's left at quality 80 or higher, in WebP if your platform supports it, JPEG if it doesn't.
4. Post it, confident the shadows still look like shadows and not a smear of grey blocks.

None of this takes long once you know the two failure modes — spoiling someone who hasn't caught up, and mangling a scene that was supposed to look moody rather than muddy. Crop handles the first when the spoiler sits at the frame's edge, watermark handles it when a solid block is the safer call, and compress handles the shadow-detail problem regardless of which screenshot you end up sharing.