My3gger said:
My tests also show that tubes inside chasis needs hours for warming up everything to max temperature, while it takes maybe half hour when they are on metal plate while doing prototypes. I don't think this is reason we couldn't hear any real differences between new EH and known good NOS tubes because tube itself warms up very fast compared to the rest. I like to run regulators at max 70 degrees C, transformers cool, only some tubes are warmer than this. So gear should last a very long time, regarding sound heat doesn't influence the rest of the circuit much after about half hour, it shouldn't drift so much that could be heard. At least it seems logical, might be wrong.
What Emrr said is absolutely true, IMO. It's so much cheaper for companies to slap it all on a PCB. Regarding power, my limited experience is that each torroid has a different magnetic field and I've learned not to bolt them down until I have the circuit running and can move them around to minimize hum. I've come to prefer a PT in an can, bolted on the outside. One other reason I can think for tubes being in the enclosure is that it creates fewer problems in shipping gear/warranty costs.
Regarding heat, it would seem as if the tube heats up within half an hour and everything is stable but in practice it doesn't always work out that way. I deal in tubes and have heard endless comments from mix engineers as to their tube gear sounding different over the first few hours of use. Granted, we're talking minutia but for a professional who is who is splicing takes, making comparisons, etc, it can be a clear difference. Some get so irritated by it that they might leave the gear on all week during a mix session. Usually, the comments are related to specific pieces of gear and most of them happen to have the tubes in the enclosure.
In my own observations, and in a compilation of observations from others, the general consensus seems to be that tubes tend to go through most of their "warm-up" over the first 90min, regardless of being inside a chassis or out. But, when inside, it's slowly bringing up the temp of other components as well. In modern gear with good parts, this difference is smaller than in vintage gear or modern gear using vintage parts, where the values can certainly drift and change the operating points and thus sound. I don't think I've heard anyone mention that they needed more than 3 hours for the gear to stabilize but some are pretty insistent that they can can hear the tone change over those first 3 hours and with their credentials in the recording field, I'm not going to question them.
I've always heard it said that solid state gear does the same though I've not personally found it to be the case (not that it doesn't happen in SS gear, just that I've never noticed).
As to the price of the EF804, I believe you. With tubes there are wild variances in price so it's completely feasible that you saw higher prices a few years ago, depending where you were looking. There is a lot of B-stock out there now so places like ebay tend to fetch lower prices because the trust factor isn't there due to the high number of scammers and poor quality tubes being sold as good.
Regarding why you didn't hear a difference in modern vs known good NOS, there could be a lot of reasons. I get a lot of emails that are very passionate about the differences in modern and vintage EF86s but I don't feel that it's just a vintage vs modern thing because there are also a lot of low-quality NOS EF86s out there. In fact, most remaining NOS EF86s are not much better, if any better, than current production. Companies like Siemens, RCA, Amperex, etc even put their names on those lower quality tubes once they stopped producing them in their own factories (in fact, RCA, Tung Sol, and GE never even made their own EF86 but they sold a a variety of imports under their names). So, it can get very confusing to those who are not intimately familiar with all the various productions. Even sellers get confused and sell them with incorrect descriptions. This is why sellers that know what they're doing and have the good stuff tend to make sure they get a good price for it. It's also unfortunate because buyers often get their first experience with NOS tubes in the form of an RFT or other moderate quality tube and end up thinking all NOS tubes are overpriced/under-performing. Not saying this is what you experienced, I just have a tendency to ramble on when it comes to tubes.
Regarding your EH codes, there's a good chance that's true. It's usually easier to produce batches when it's not a high-volume tube. For example, in the 60's, Phillips would often have one of their Euro factories produce a large volume of moderately popular designs for one year and distribute them to the other brands under the Phillips umbrella (Amperex, Valvo, Mullard, Miniwatt) during their downtime for that specific tube.