One Logo, Two Destinations: Resolution, Format and Colour Mode
A regional museum asks you to prepare one logo lock-up for two destinations: a 300 mm × 900 mm vertical pull-up banner that the print shop will run at 150 ppi, and the museum website, where the image sits in a 400-pixel-wide header. The only approved logo file you have been given is a PNG measuring 600 × 1800 pixels, exported years ago in sRGB. The banner works out to roughly 11.8 × 35.4 inches. The print shop has warned that 'low-resolution artwork will be rejected at pre-flight', and the web team reports that the current header image takes 1.4 seconds to appear on mobile connections. The director wants to know why you cannot simply 'make the file bigger' in an image editor and be finished.
1. Calculate the effective resolution of the existing PNG at banner size and state, with reasoning, whether it will pass the print shop's pre-flight.
Read the worked answer
The banner is 300 mm wide = 300 ÷ 25.4 ≈ 11.81 in, and 900 mm = 35.43 in.
Required pixels at 150 ppi: 11.81 × 150 ≈ 1772 px wide and 35.43 × 150 ≈ 5314 px tall.
The supplied PNG gives 600 ÷ 11.81 ≈ 50.8 ppi and 1800 ÷ 35.43 ≈ 50.8 ppi - roughly one third of the requirement, consistently in both directions.
So it fails on two counts: the pixel count is short, and resampling to 1772 × 5314 does not add detail. Interpolation averages neighbouring pixels, so the edges of letterforms go soft and the anti-aliased edge picks up a halo. Pre-flight software reads pixel dimensions, not intentions, so the upsampled file is still rejected.
The correct move is to re-supply the artwork: request the original vector logo (AI/EPS/SVG) from the brand owner, or, if none exists, redraw it as paths and export at final size.
2. Explain why a vector master would change both workflows, and how you would export the web header from it.
Read the worked answer
A vector logo is stored as mathematical paths and fills, so it has no fixed pixel dimensions. The print shop's RIP renders it at whatever the device needs (often 1200–2400 dpi for line work), and the same file scales down to a 400 px header without loss.
That removes the resolution argument entirely: the banner and the header become two exports from one master rather than two versions of a raster file.
- Banner: place the vector or supply a PDF/EPS at final size with 3 mm bleed and convert the brand colour to an agreed CMYK build (or a spot ink).
- Web: export a PNG at 400 px, plus an 800 px version for high-density screens. Keep sRGB.
A flat logo compresses to a few kilobytes as PNG-8 or SVG, which addresses the 1.4-second load far better than a 600 × 1800 photograph-style PNG. Judgement point: do not scale the low-res PNG down and assume the softness will not show - set the header from the vector export instead.
3. The print shop will convert the sRGB artwork to CMYK. Explain what can go wrong and how you would manage the conversion.
Read the worked answer
The PNG is tagged sRGB, a display-referred space; a four-colour press needs CMYK. Two mistakes to avoid are leaving it RGB so the RIP converts blindly, and converting with the wrong profile.
Process: ask the printer for their ICC profile, then convert (not assign) to CMYK in a single step from the RGB master, keeping a copy of that master. Converting is one-way and lossy; CMYK → RGB → CMYK again shifts the brand colour a little each time.
Judgement point: check the brand colour against a printed CMYK swatch. Saturated blues and greens commonly fall outside CMYK gamut, so the banner will look duller than the screen. If that colour is the brand's whole identity, ask whether a spot ink or Pantone match is possible on the banner run rather than a four-colour build.
Soft-proof with the printer's profile and black-point compensation before sending, and keep the web export in sRGB so the screen version stays as intended.