How to make a vertical review video comparing several games

Last updated 2026-08-16 · 6 min read

What is the short version?

A comparison reel is a listicle with a strong opinion. The research is quick, the script is short, and the reason these take an evening is the assembly: a title card, footage and a caption for every game, all timed to a voiceover you have to record first.

1. Go to Vizard Agent and name the titles you want compared.
2. Say what you are comparing them on and what your verdict is.
3. Say the length and the feed.

What do you need before you start?

The titles and the angle. Vizard Agent researches each game and writes the script, so the thing it needs from you is the take — the ranking, the framing, the reason a viewer should care which of these three you preferred.

• The titles. Two is a comparison; three or four is a listicle.
• The angle. Graphics, story, value, hype versus reality. Pick one.
• Your verdict. A reel with no opinion is a press release.
• Your previous reels. In the same project, so Vizard Agent can match the look you already have.
• The length. Sixty to seventy-five seconds is the working range for this format.

What do you type into Vizard Agent?

Name the games and the terms of comparison. Vizard Agent does the research itself, so the sentence that changes the result most is the one that says what you actually think — everything else it can find out.

Prompt

Variants worth knowing:

• Rank them explicitly. "Worst to best" is a stronger structure than a neutral comparison.
• Use your own gameplay. Upload your captures and Vizard Agent uses them before sourcing anything.
• A hook up front. Ask for the verdict in the first three seconds; it is the convention that works on these feeds.

What does Vizard Agent actually do?

Here is the order Vizard Agent worked in on a real comparison reel. It starts with your back catalogue rather than with the games — because a channel's reels have to look like each other before they have to look good.

1. Loads the listicle playbook and your project ledger.
2. Reviews your past finished reels and researches the three titles.
3. Downloads those past reels and checks the asset prices before planning.
4. Probes them and extracts sample frames.
5. Tiles the reference frames and looks at them to read your established style.
6. Finds the official trailers and checks the voice and music options.
7. Starts the voiceover and music while searching stock, rather than waiting on each in turn.
8. Generates the additional visuals and the title graphics.
9. Transcribes its own narration for word timings and builds the contact sheets for review.

Step 5 is the one that makes this a channel rather than a video. Vizard Agent looked at your published reels as a single sheet of frames and matched their typography and pacing, which is the part a fresh template always gets wrong.

What does the result look like?

From the run this page is written from, probed on the delivered file: 1080x1920, H.264, 30fps, 74.9 seconds, AAC audio. Vertical and seventy-five seconds, narrated, with title graphics and captions timed to the narration and footage for each of the three titles.

Ask for a shorter cut for a different feed and Vizard Agent tightens the script rather than speeding up the read.

Seventy-five seconds across three games is twenty-five seconds each, which is one clear point per title and no room for a second. That constraint is what makes the angle you choose matter more than the research behind it.

It is not instant, and this category was not separately measured. Comparable work runs a median of 28 to 38 minutes end to end. Across all projects the median cost by tier is Flash 47, Pro 55, Max 242, Ultra 263 credits.

When does this not work well?

Game footage belongs to somebody. Vizard Agent sources and generates what it can, and the material that makes these reels look best — actual gameplay of an unreleased title — is the material with the most complicated rules around it.

• Trailer and gameplay footage is owned. Commentary and review have latitude in some places and not others. Know where you stand.
• Generated game visuals are approximations. They look like the genre rather than like the game. Say if that is not acceptable.
• Unreleased titles have no gameplay to show. A reel about three upcoming games is necessarily built from trailers and generated shots.
• Your own captures beat everything. If you have played it, upload it.
• The research is a starting point. Vizard Agent grounds the script in what it finds, and release dates and specifications change. Check anything you assert.

How do you fix a result that came back wrong?

Name the game or the section. Vizard Agent keeps the script, the narration, the word timings, the graphics and the footage separately, so changing your verdict on one title re-reads that section and re-renders its card without touching the other two.

• "Move that one to first place." A reorder; the graphics follow.
• "That fact is out of date." One line rewritten and re-read.
• "Use my own gameplay for the second game." A footage swap at the same timing.

How does Vizard Agent compare to doing it yourself?

By hand this is research, a script, a recording, three sets of footage, three title cards and a caption pass — repeated every week, on a channel that needs several of these a month before the format does anything for it at all. Each individual step is easy and the repetition is what kills the channel.

Common questions

Will it match my channel's look? Yes, if your earlier reels are in the same project. Vizard Agent reviews them as a set before designing anything.

Can it use my own gameplay footage? Yes, and it prefers it. Upload the captures alongside the brief.

Does it research the games? Yes. Vizard Agent grounds the script in what it finds, and you should check anything time-sensitive.

How many games can one reel cover? Three or four in seventy-five seconds. More than that and none of them get a real point made.

Can it put my verdict at the start? Yes, and it is worth asking for. The first three seconds decide whether the rest is watched.

Can I get a longer YouTube version? Yes. Vizard Agent expands the script rather than slowing the same one down.

Where does the footage come from? Stock, generated shots and your own captures. Vizard Agent uses yours first whenever you have them.

Can it add a call to action at the end? Yes. Say the wording and Vizard Agent sets it as a closing card in your channel's style.

Can it compare only two games? Yes, and a head-to-head gives each title more room than a list of four does.