Question 4: TIFF vs. JPG Image Quality — How the Formats Compare and the Hidden Trade-Off of TIFF Files

TIFF generally beats JPG on image quality because TIFF can store every pixel losslessly, while JPG discards detail to shrink files. That quality advantage comes with a steep trade-off: TIFF files are often 5 to 20 times larger than comparable JPGs, and the gap widens with resolution. For most casual viewing, a high-quality JPG is visually indistinguishable from TIFF. For editing, printing, or archiving, the difference matters.

Pixel Storage: What Each Format Actually Keeps

A digital image is a grid of pixels, and each pixel carries color values. TIFF is a container format, not a single compression scheme. It can store pixels uncompressed, or it can use lossless compression such as LZW or ZIP. In both cases, the decoded pixel values match the original exactly. Save a file as TIFF, reopen it, and every value returns bit for bit.

JPG uses lossy discrete cosine transform compression. The encoder divides the image into 8x8 pixel blocks, converts each block into frequency components, and quantizes those components. Quantization throws away small high-frequency differences the eye is less likely to notice. Decoding reconstructs an approximation, not the original. Each save-and-reopen cycle re-encodes the data, so quality degrades with repeated editing.

That block structure explains JPG artifacts. At high compression you see blocky patches in smooth gradients and ringing (halo-like ripples) around sharp edges. TIFF produces none of these because nothing is discarded. One plain-language way to hold it: TIFF keeps the full recipe, JPG keeps a good guess at the dish.

Quantified Context: Bit Depth, Channels, and File Size

The quality ceiling of any format is set by bit depth and channel count. A standard 8-bit JPG stores 256 levels per color channel, or about 16.7 million colors. TIFF supports 8-bit and 16-bit per channel, plus 32-bit floating point. A 16-bit TIFF holds 65,536 levels per channel. That headroom matters when you stretch tones in editing, because banding appears sooner in 8-bit data.

File size is where the trade-off becomes concrete. Take a 24-megapixel image (6000 x 4000 pixels, three channels at 8 bits). Uncompressed, it needs roughly 72 MB, and that is the baseline TIFF writers often use. The same image saved as a maximum-quality JPG typically lands near 10 to 15 MB. A moderate-quality JPG at settings around 80 drops to roughly 3 to 6 MB. Lossless-compressed TIFF can cut the 72 MB to somewhere in the 35 to 55 MB range depending on scene complexity.

Multiply that across a shoot. A 500-image session as 72 MB TIFFs consumes about 36 GB. The same session as 12 MB JPGs fits in about 6 GB. On a 1 TB drive, that is roughly 13,800 TIFFs versus 83,000 JPGs. The storage and transfer costs scale linearly with pixel count, so high-resolution work punishes TIFF hardest.

The 8x8 block quantization in JPG is why file size and quality move together. Lower quality settings coarsen the quantizer, so more coefficients round to zero and the file shrinks. TIFF's compression, by contrast, only removes statistical redundancy, never pixel information.

Edge Cases: Where the Comparison Breaks Down

The quality gap depends heavily on content. Busy scenes with fine texture, foliage, fabric, and noise compress poorly under JPG and show visible smearing at moderate settings. Smooth skies and solid backgrounds compress well but reveal banding and blocking in flat gradients, precisely where JPG quantization is most exposed.

Editing is the tougher edge case. Adjustments such as curves, exposure lifts, and color grading amplify whatever artifacts exist. A JPG that looks clean on screen can show block structure after a strong shadow lift. Each subsequent save recompresses, and by the third or fourth generation the damage compounds. TIFF edits are reversible in the sense that the pixel data survives intact, though the edits themselves must be saved as new files to preserve the original.

There is also a counterintuitive detail: a TIFF is not automatically higher quality than a JPG if the source was already JPG. Converting a JPG to TIFF preserves the existing artifacts and inflates file size without recovering lost detail. The format stores what it is given. Likewise, TIFF can hold CMYK, LAB, and alpha channels for print and compositing work, while baseline JPG supports RGB and grayscale only.

Misconceptions: What the Format Does Not Fix

A common belief holds that TIFF is inherently better for photographs. It is better for preservation and editing, but a well-encoded JPG at high quality is visually indistinguishable in normal viewing at typical screen sizes. Human contrast sensitivity falls off at high spatial frequencies, which is the same property JPG quantization exploits.

Another misconception is that JPG always loses quality every time you view it. Viewing and decoding do not alter the file. Only re-saving does.

Red flags worth watching:

Repeatedly opening and re-saving a JPG for edits, which stacks compression damage generation after generation. Assuming a TIFF from a camera is a raw file. It is processed RGB data, not sensor raw. Expecting lossless compression to be free. LZW and ZIP save space but slow down reading for large files. Using TIFF for web delivery, where browsers and many content systems handle it poorly or not at all. Trusting a "TIFF" label on a file that was converted from JPG, which cannot restore discarded data. Ignoring bit depth: an 8-bit TIFF has the same 256-level ceiling as an 8-bit JPG.

Timeline: The Lifecycle of an Image Through Each Format

The lesson from this time course is that the trade-off compounds. TIFF pays a storage cost up front and protects quality over time. JPG saves space up front and pays in quality the more you edit and re-save.

Practical Actions: Choosing the Right Format

Pick the format to match the job. For images you will edit and archive, save the master as TIFF, ideally 16-bit if the source supports it. For images you will view, share, or post to the web, export a high-quality JPG (settings near 90-95) from that master. The master stays intact while the JPG travels.

Concrete steps and thresholds:

Keep a TIFF master for anything you edit more than once, especially scans and composites. Export JPGs at quality 90-95 for web delivery; expect 8-12 MB at 24 MP. Use lossless TIFF compression (LZW or ZIP) to cut storage by roughly 30-50 percent with no quality change. Archive at 16-bit TIFF only when your workflow uses the extra headroom; it doubles file size versus 8-bit. Retain the original capture file; never overwrite it with an edited JPG. Budget storage at roughly 70-80 MB per 24 MP TIFF when planning drives.

Reconciling the thresholds: a 10-15 MB max-quality JPG is visually close to its TIFF source for display, yet the same JPG re-saved repeatedly drifts toward visible artifacts. The practical rule is to protect the master losslessly and compress only the delivery copy.

Limitations: Where Format Choice Stops Mattering

Format cannot rescue a bad capture. An out-of-focus or overexposed image will not improve by choosing TIFF. Dynamic range and sharpness are set at capture time, and no container format changes them. TIFF also does not define a universal color profile; color accuracy depends on embedded ICC profiles, which both TIFF and JPG can carry.

File sizes cited here are estimates based on standard pixel math and compression behavior, not guarantees for any specific camera or encoder. Actual sizes vary with scene content and software settings. The quality ranking itself is stable: lossless TIFF preserves exactly what JPG approximates.

For archival and production decisions in regulated or professional settings, consult your organization's documented image standards and a qualified imaging specialist, since requirements vary by industry and use case.