BIP 101 was proposed in June 2015 for activation in January 2016. Under its rules, the blocksize now would be 154MB.

BIP 100 was proposed a few days earlier for activation in January 2017. It would have allowed miners to choose a new limit every 2016 blocks.

Had that been adopted, and if miners had consistently voted for the largest increase allowed, the blocksize now would be 14GB.

BIP 103 was proposed in July 2015 for activation in January 2017. It proposed a 17% increase per year over almost 50 years. Under its rules, the blocksize now would be 3.36MB.

Historically, Bitcoin Core limited the size of the blocks that it would create. That limit was raised from ~250kB to 750kB in the 0.9 release in March 2014

(As this wasn't a consensus rule, pools could choose to override it. Eligius pool did so, mining blocks of up to 1MB in size despite the ~250kB default)

Had BIP 103's 17% growth rate been applied the a 250kB base instead, the blocksize now would be 1.35MB.

One thing to note about all these proposals was that unlike segwit, they didn't simply provide extra space for signature data, but broadly allowed more transactions per block.

That means that, unlike segwit, those changes would increases the rate at which the utxo set can grow, both in the average case due to people making more transactions, and in the worst case given attackers or hostile miners.

In the 12 months from June 2013, the utxo set grew from ~6GB to ~11GB. Increasing that rate by a factor of 35% (250kB+17%pa) would have brought that up to about ~13GB. BIP 103 (1MB+17%pa) would have brought it up to ~21GB.

BIP 101 would have potentially increased that to ~550GB, and BIP 100 would have potentially increased it to somewhere above 8TB.

In practice, in 2023, the average blocksize was 1.7MB. In 2022, the average blocksize was 1.2MB. When segwit was proposed, we predicted that once it was widely adopted, blocks would be in the range of 1.6MB to 2MB.

One thing that people revisiting the blocksize wars seem to gloss over is that it wasn't about "lets just change 1MB to something slightly higher" -- the numbers originally proposed were much higher, more closely matching the strategy eventually adopted by BCH and BSV.

I don't think it makes sense to talk about blocksize increases for Bitcoin today: monetary transactions are only filling a fraction of available block space, inscriptions are supposedly more valuable the less block space there is, and utxo set growth rate is already concerning.

One of eligius's big blocks (followed by a 500kB block by BTC Guild, and 250kB blocks clearing out the rest of the backlog by other pools)

statoshi's stats on utxo set size