The current Argon2 draft RFC, I think, provides a good, reasonably brief answers to this question. TL;DR: most people will indeed want to use Argon2id and not the "pure" variants.
The introduction summarizes the issues quite well:
Argon2 has one primary variant: Argon2id, and two supplementary variants: Argon2d and Argon2i. Argon2d uses data-depending memory access, which makes it suitable for cryptocurrencies and proof-of-work applications with no threats from side-channel timing attacks. Argon2i uses data-independent memory access, which is preferred for password hashing and password-based key derivation. Argon2id works as Argon2i for the first half of the first iteration over the memory, and as Argon2d for the rest, thus providing both side-channel attack protection and brute-force cost savings due to time-memory tradeoffs. Argon2i makes more passes over the memory to protect from tradeoff attacks.
Basically, there are two kinds of attack considered:
- Side-channel timing attacks.
- Time-memory tradeoff (TMTO) attacks.
Argon2d is vulnerable to timing attacks, but offers the best resistance to TMTO (section 9.2):
The best tradeoff attack on $t$-pass Argon2d is the ranking tradeoff attack, which reduces the time-area product by the factor of 1.33.
Argon2i is invulnerable to side-channel timing attacks, but is weaker against TMTO:
The best attacks on the 1-pass and 2-pass Argon2i is the low-storage attack described in [CBS16], which reduces the time-area product (using the peak memory value) by the factor of 5. The best attack on 3-pass and more Argon2i is [AB16] with reduction factor being a function of memory size and the number of passes. For 1 GiB of memory: 3 for 3 passes, 2.5 for 4 passes, 2 for 6 passes. The reduction factor grows by about 0.5 with every doubling the memory size. To completely prevent time-space tradeoffs from [AB16], number $t$ of passes must exceed binary logarithm of memory minus 26.
Argon2id sits in between:
The best tradeoff attack on 1-pass Argon2id is the combined low-storage attack (for the first half of the memory) and the ranking attack (for the second half), which bring together the factor of about 2.1.
Since the Argon2 runtime grows linearly with the number of passes, Argon2i with six passes is about six times as slow as Argon2id with one pass.
The draft RFC also states (without explanation) that Argon2id resists timing attacks, and recommends it as the primary variant, and explicitly telling you (section 4, "Parameter choice") that if you don't know which to use, you should pick Argon2id. You should probably wait until the RFC is approved to fully commit to anything, but in the meantime if you're writing some early code it makes sense to bias things in favor of Argon2id.
