SIX ORIGINAL PNG EXPORTS · OCTOBER 15, 2026
Roblox Shirt Template: Does Logo Source Resolution Change Size?
In this editor test, doubling or quadrupling source resolution did not change the logo's saved width when its proportion inside the PNG stayed the same. At 45% Image size, a 50-pixel mark inside a 150-pixel PNG, a 100-pixel mark inside a 300-pixel PNG, and a 200-pixel mark inside a 600-pixel PNG each exported 84 pixels wide on the Shirt front. Check visible-art proportion and the downloaded PNG before increasing resolution just to fix size.
Measured on this editor's flat PNG export with centered, hard-edged squares. The result is not a universal resolution or image-quality rule.
What changed in the test
We made three original transparent square PNGs. Each held a centered opaque red square taking exactly one-third of the source width and height: 50 of 150 pixels, 100 of 300, and 200 of 600. We imported each file afresh, set Image size to 45% and then 60% in separate runs, dragged the image from canvas center to the same TORSO FRONT center, and decoded red pixel bounds in the Download PNG output. All six Shirt exports were 585 × 559 pixels.
Roblox's Classic clothing documentation specifies a 128 × 128 FRONT region for classic shirts. In this editor's 585 × 559 template, the visible front guide spans x 128–255 and y 191–318. The measurements below use red pixels with alpha of at least 128; they describe the saved file, not a Studio rig.
| Source PNG | Visible source mark | Image size | Saved red bounds | Width / front margin |
|---|---|---|---|---|
| 150 × 150 | 50 × 50 | 45% | x 150–233; y 213–296 | 84 px / 22 px each side |
| 300 × 300 | 100 × 100 | 45% | x 150–233; y 213–296 | 84 px / 22 px each side |
| 600 × 600 | 200 × 200 | 45% | x 150–233; y 213–296 | 84 px / 22 px each side |
| 150 × 150 | 50 × 50 | 60% | x 136–247; y 199–310 | 112 px / 8 px each side |
| 300 × 300 | 100 × 100 | 60% | x 136–247; y 199–310 | 112 px / 8 px each side |
| 600 × 600 | 200 × 200 | 60% | x 136–247; y 199–310 | 112 px / 8 px each side |
The 45% samples had identical 84 × 84 bounds and 7,056 red pixels. At 60%, all three had identical 112 × 112 bounds; the 150-pixel source produced 12,540 threshold-matching red pixels while the other two produced 12,544. That four-pixel difference is a reminder that matching outer bounds does not mean every interior pixel is identical after resampling.
The six bars show measured exported widths. The row labels are source-canvas widths, not logo widths.
What this means for your logo
The source file's pixel count and the portion occupied by visible art are separate facts. Here, every mark occupied 1 ÷ 3 of its PNG width, so changing only the resolution left its exported outer size unchanged at a fixed slider setting. In contrast, our tight-versus-padded test changed the visible-art fraction and measured an 84-pixel versus 28-pixel result at the same 15% setting. That is why trimming unnecessary transparent canvas can solve an undersized-logo problem that merely exporting a larger file does not.
For the centered squares here, front-panel margin is (128 − saved width) ÷ 2. The 45% result leaves (128 − 84) ÷ 2 = 22 pixels per side; 60% leaves (128 − 112) ÷ 2 = 8. A larger source PNG alone did not buy extra margin or occupy more of the front panel. This calculation applies to the measured exported bounds and guide, not to perceived size on every avatar.
Repeat the check with your own artwork
- Find the visible bounds. Open your logo PNG against a checkerboard. Note both the full canvas dimensions and the smallest box containing the art, including any shadow you intend to keep.
- Compare proportions. Divide visible-art width by source-canvas width. A 50/150 and 200/600 mark both occupy one-third; a 100/100 tight crop occupies the entire width. This gives a more useful starting clue than pixel count alone.
- Import and place once. Use the editor's Image upload and Image size controls, then drag the artwork into TORSO FRONT. Keep the guide visible while placing it.
- Inspect the download. Download PNG and check the actual visible width and edge position. If the mark is too small, trim accidental transparency or adjust Image size; if it approaches the edge, reduce the setting or recenter the source.
- Test on an avatar. Follow Roblox's Studio procedure: create a Block Avatar rig, add a Shirt object, and assign the uploaded texture to ShirtTemplate before relying on the 3D appearance.
Limit: We tested three centered, hard-edged red squares with the same one-third art fraction, two slider settings, and a controlled browser pointer drag. We did not test photographic logos, fine text, shadow quality, other aspect ratios, physical touch, Studio rendering, or Marketplace acceptance. Higher source resolution may change edge detail even when measured outer bounds match.
Sources and reproducibility
- Roblox Creator Hub: Classic clothing — primary source for the 128 × 128 FRONT region and Studio test procedure.
- Avatar Shirt Maker editor — first-party October 15 six-export import, drag, download, and pixel-decoding test.
The project audit retains the exact test script and result JSON. The six-row comparison, proportional graphic, margin arithmetic, fraction comparison, and repeatable decision steps were independently produced for this article.