These are the same polyester with one difference in the chemistry, and that difference is the entire reason one of them is on every desktop printer and the other is a specialist material.
Unmodified PET is semi-crystalline. As it cools, chains fold into ordered crystalline regions, and crystalline polymer is denser than amorphous polymer — so the part contracts sharply, unevenly and over a longer period than the print takes. Adding cyclohexanedimethanol to the backbone disrupts that folding. The chains cannot pack, the material stays amorphous, and the result is PETG: a polyester that cools predictably.
What crystallisation does to a dimension
PET's stored shrinkage range is 0.4 to 1.2%; PETG's is 0.2 to 0.6%.
Applied to a 100 mm part at the top of PET's range, it finishes at 98.8 mm and needs a 1.2146 mm correction to land on size. The stored note adds the part that makes it worse: annealed PET parts move again, because heating a semi-crystalline polymer above its glass transition lets crystallisation continue. So the figure to compensate for is not one figure at all: it is the as-printed contraction plus whatever crystallisation adds later, and only the first half of that is something the shrinkage calculator can give you. The second half has to be measured on a part that has already been through the oven.
A material whose dimensions depend on its thermal history after the print is a material you cannot design fits in. PETG's stored weakness list does not mention dimensional predictability at all; PET's names it first.
And it is the denser polyester
Unmodified PET is the heaviest of this family at 1.32 to 1.40 g/cm³, against PETG's 1.23 to 1.29. Across a kilogram of 1.75 mm filament that is 327.36 m for PETG and 301.27 m for PET — 26.09 m less, an 8.661% density increase costing 7.971% of the length. The length and weight converter will re-run it against whatever figure your supplier publishes.
So the same model is heavier in PET, and PET's $25–45 per kilogram already sits above PETG's $18–30. On cost per finished part the gap is wider than the shelf price shows.
Drying is not optional for PET, and the reason is different
Both materials absorb moisture, but the consequence is not the same.
Wet PETG prints badly — popping, stringing, a hazy surface — and drying it restores the material. Wet PET hydrolyses in the melt, which cuts the polymer chains and produces a part that is permanently weaker. The stored schedule reflects the seriousness: 70 °C for 6 to 8 hours, against PETG's 65 °C for 4 to 6, and the note calls it non-negotiable.
This is the distinction what moisture does to filament is built around, and PET is the clearest example of it on the site. You cannot fix a hydrolysed part by drying the spool afterwards.
The rest of the gap
- Nozzle temperature. PET wants 250–275 °C against PETG's 230–250. Not impossible, but it rules out some machines and narrows the working window on others.
- Bed adhesion. PET needs a hot bed and an adhesive because the crystallisation shrink pulls corners hard. PETG's problem is the opposite — it sticks to smooth PEI too well.
- Difficulty. PET is rated 4 here, PETG 2, and the stored note on PET names the narrow process window as the reason beginners struggle.
- Strength. PET is genuinely stronger, at 50–70 MPa against 45–55, and slightly more heat resistant at 70–85 °C against 68–80. This is the honest case for it.
Where PCTG fits into this family
There is a third member and it is worth placing, because the three of them form a clear progression rather than a set of alternatives.
PET is the unmodified polyester: strongest, densest, hardest to print. PETG replaces some of the backbone with glycol to suppress crystallisation. PCTG uses a different and more thorough modification, which buys the ability to deform under stress without whitening — the property PETG against PCTG is built around.
Read the stored densities in order — 1.32–1.40, then 1.23–1.29, then 1.22–1.27 — and the pattern is visible: each modification disrupts the packing a little more, and each step down in density is a step down in crystallinity and a step up in printability. The chemistry and the ease of use are the same story told twice.
When PET is worth it
Two situations, both real.
Recycled-bottle filament projects. Drinks bottles are unmodified PET, and anyone extruding filament from them is printing PET whether they intended to or not. Understanding why the parts shrink unpredictably is most of what makes that hobby workable, and the answer is crystallisation rather than a bad extrusion.
Chemical resistance requirements that PETG does not meet. PET's crystalline structure gives it better resistance to some solvents than the amorphous copolymer, and for a container or a fitting in a specific chemical environment that can be the deciding property.
Outside those, PETG does the same job at lower temperature, lower price, higher predictability and a fraction of the effort.
The general lesson worth keeping
The PET-to-PETG story is the clearest illustration on this site of a rule that applies far beyond these two materials: a semi-crystalline polymer shrinks much more than an amorphous one, and its shrinkage depends on how fast it cooled.
That is why polypropylene and nylon warp badly and PETG and ABS do not, why chamber temperature matters so much more for some materials than others, and why annealing changes dimensions in PLA — another polyester that can crystallise if you let it. Learn the distinction once and half the materials on this site stop being surprising.