A tale of two tRNAs--how a 'clock' with a single 'alarm time' (EF-Tu) allows the ribosome to discriminate between good and bad matches

Consider two tRNAs, each pairing with a glycine 'GGG' codon.
At top, a perfectly fitting glycine-bearing tRNA. At bottom, a 2/3 matching arginine tRNA. Let's track the two through 4 different slices of time, numbered in orange above
Time 1: Each tRNA settles in. Each is accompanied by gray EF-Tu, which 'covers up' the amino acid that is on board, and each EF-Tu is in its starting state with a bound GTP molecule

Time 2: Because EF-Tu is an inefficient GTPase (GTP breaking enzyme), at time 2 it will likely still not have succeeded in breaking off the gamma (outside) phosphate. Because of its reasonably strong matching, the GLYCINE (CCC) tRNA is in good shape. On the other hand, the poorlymatching ARGININE (CCU) tRNA is being buffeted by the 'winds' of Brownian motion, and is likely to fall off at any moment!

Time 3: The EF-Tu of the glycine tRNA finally breaks its GTP down to GDP. Alas, the arginine tRNA has been ripped away from the mRNA by Brownian motion, its EF-Tu clock stops, and the codon is available for some other tRNA to wander in and perhaps have better luck

Time 4: Having changed shape because the gamma phosphate has broken off and left, the EF-Tu that was bound to the glycine tRNA has structurally 'lost interest' and wanders away. The glycine amino acid is now exposed! The ribosome can now 'use' the amino group of the glycine tRNA to 'attack' the polypeptide chain that is on the OTHER tRNA in the ribosome (the one in the 'P' or peptidyl site; see textbook or other reference on translation).