I've just sieved two c179s with parameters similar to the consensus from this thread.

The first c179 used parameters

Code:

tasks.I = 15
tasks.qmin = 20000000
tasks.lim0 = 95000000
tasks.lim1 = 135000000
tasks.lpb0 = 31
tasks.lpb1 = 32
tasks.sieve.lambda0 = 1.88
tasks.sieve.mfb0 = 58
tasks.sieve.mfb1 = 90
tasks.sieve.ncurves0 = 20
tasks.sieve.ncurves1 = 13

Poly score was 1.122e-13.

64.9M CPU-seconds of sieving for 321M raw relations, 224M unique.

TD=110 produced a 14.8M matrix.

The second c179 (almost a c180) used the same parameters except with lambda0 removed.

Poly score was 1.000e-13.

64.4M CPU-seconds of sieving for 313M raw relations, 218M unique.

TD=110 produced a 15.2M matrix.

So the second c179 only sieved about 2% slower despite having an 11% worse poly. Looks like no lambdas might be the way to go here.

For the purposes of a hypothetical c180 parameters file, 300M relations would probably be a good target - I sieved a bit more to make the matrices nicer.