poweramp tube matching

In a push-pull circuit (I'm guessing 85% of tube amps) if the pairs of push-pull tubes DON'T match, the weaker tube generally causes the stronger tube to wear faster, is what I've read.
.

Nsgarch, I think your right, but when each tube can be individually biased doesn't that overcome the problem you bring up? You not only set the initial bias but you can monitor it and adjust it VS getting a matched quad in an amp with auto bias which has no provision in the auto circuit to adjust the bias each tube automatically.

Its my understanding that 'matched' tubes don't necessarily wear evenly and at some point in time you will have the same problem of the tubes not being 'matched'. In an auto circuit you have no control, whereas when you can adjust each tube individually you can monitor and maintain a balance between the push and the pull tube(s).

Anyway thats how I understand it and it is why I prefer an amp which has manual bias and a bias pot for each power tube. Also being able to adjust bias settings gives you a bit of tone control at least on a scale of different IC's or cable. FWIW.

Newbee, I recently read an article on the very point you raised, namely

"when each tube can be individually biased doesn't that overcome the problem you bring up?"

And the author explained why it doesn't compensate for the tube's not matching. I'll try and remember where I saw it ;--)

Power tubes need to be matched for gain (transconductance or gm) as well as for current and both are equally important. Individual biasing does in effect match up tubes for current output, but it does not address gain. For that, you have to match the tubes as well. Also, even with individual biasing, it's better if a pair of tubes are in the same ball park for current ourput as well. My own rule of thumb, with individual biasing, is to match tubes within 3% for gm and 6% for current.

what about tubes in input stage, do they have to match?