I've built both orange 86 and g9 units. 'Better' etc is very subjective and each person may have different preferences. How do you define 'better' ? Wider freq response? Lower noise floor? Lower THD or IMD ?
Or the complete opposite? Is 'better' more characterful? And so on. And on.
I found both units to be low noise floor, full freq response etc, so they are both of high enough quality for serious work.
My preference is for the Orange 86 - I found it has more 'character' than the G9. I thought this would be because the Orange 86 would have more harmonic distortion, but I measured the opposite, using single tone tests.
I found the G9 had quite a bit more overall distortion, even though it sounds very clean. The Orange 86 had less overall distortion and quite a different mix of even/odd harmonics. But it sounded more typically 'tube like' ie. distortion. It too was clean but driving it harder gives more character as the amount of negative feedback is reduced. (also true of the G9 - it also uses a negative feedback attenuator)
Perhaps because my Orange 86 has less gain (maybe 40dB or so max) than the G9, I tend to push it harder, giving more THD and so being more pleasing to my guitar-biased ears. But certainly under test I found the G9 had more THD. Surprisingly so.
It could be the transformer choices too - Lundahl ins, Carnhill outs on the G9 and Haufe/UTC ins and Edcor outs on the Orange 86. There is some small difference between the Carnhill and Edcor. Nothing detectable by me between the Lundahl and Haufe (except 6x price!). The UTC has less wide bandwidth but not so much.
I would expect other's experience may well differ from mine.
My opinion would be that the G9 is a fine unit but not so different in sound from solid state or discrete units. The Orange86 was much closer to what I expected a tube unit to sound like.
In other words, why go to the trouble of doing a tube unit if it sounds very similar to a cheaper, easier solid state or discrete unit? For that reason, I think the Orange86 is a must have, the G9 doesn't see so much use.
The main thing imho is generalizations about preamps 'better', 'worse' and so on are not very useful. Much less, comparing mythical (to most people) units to diy builds.
All the options are probably very good - the best one for an individual is a combination of things, like
- price
- audio performance
- likelyhood of completing a good quality build
- usability
- character
You make a choice and go with that. Then make another choice and go with that. And so on.
Eventually you forget the original question