tube specs

[Richard]

Short answer and conscious Iīm opening a can of worms.

They contradict themselves, they show HUGE differences in their red grey yellow coloured bar lows mids highs yet they trace the preamp, where difference should be even more marked, since in every case both triodes in a bottle are of the same brand by definition and graphs clearly show all are parallel, meaning the camel hump frequency response you see comes from the circuit parts and values, NOT from tubes, which only show some gain difference , which by the way is expected and acknowledged.

So those poetic descriptions,such as: crisp/bright/thick/harmonic_texture/shimmer/punchy/articulate/presence/lean/sparkle/etc. are just that, poetic words.

The main real difference is gain/transconductance , not frequency response.

Capacitance difference is minute and definitely swamped by the feet of wire they used in the test rig.

They donīt explain how the experiment was made, how many people participated, what does each of them say he thinks he heard , how many disagreed with others and so on.

Oh well.

They go into it a bit, I don't know if it would satisfy scientific rigor but it seems reasonably thorough. The bar graphs are subjective and relative but not proportional, more like a ranking than a quantitative measurement.

[loudthud]

If frequency response was the only component of tone, a preamp with opamps (solid state with lots of feedback) would sound the same as a preamp with tubes. I hope most people here would agree this is flatly not true. It is the non-linearity of tubes that gives color to the tone. Opamps produce very little non-linearity until the signal reaches clipping. The gain, point where they self bias and clipping behavior of tubes also flavor the tone. It is the variations of these imperfections that makes one tube sound different than another.

[J M Fahey]

True, but that was not the point mentioned or analysed in that tube comparison page; they stuck to low/mid/high balance exclusively ... and graphs showed such balance did not change, at all.

There were important and very measurable gain differences, but amount of gain difference was the same throughout the whole Guitar spectrumand then some.

What you mention is very important, and deserves its own thread.

That said, nickb DID overdrive and analyze different tubes, with very interesting results, and found significative differences in that respect.

[SoulFetish]

I had a stab at it. It's far from perfect and if I were to do it again I'd do it differently, but it is what it is. Skip down to >>part 2<< for the overdrive bit

Damn it, Nick! That's a great page, and a great article. Dude, between that, your GUI for the UTracer, and the interactive curves (my personal favorite), you're crushing it with putting some really usefull info and utility on your site. Teach me the ways of the force so I can be a jedi like you.
So, here's a question: have you experimented at all in developing a Utracer GUI on Linux? The reason I ask is I'm surrounded by raspberry pi and they are too damn useful and cheap not to put into a project. So I was thinking about building a stand alone utracer build into a pelican case; raspberry pi and HD display integrated right into the enclosure running the interface. I would like to have a truely reliable way of matching and testing tubes for designing and service purposes, and.. well I gotta' see the curves man. The curves tell me so much. What are your thoughts about the idea? More work than it needs to be, or is it a worthy tool in the end?

[nickb]

Thank you ( insert slightly embarrassed emoji here).

I have done a uTracer uGUI for linux but I hadn't updated it for some time so it lagged behind the Win version. At some point I see the link for the linux version got dropped off the site too. I didn't do that deliberately - least i don't remember doing it.

I know nothing about Raspberry Pi (-es other than eating them). If you want a version email me and we'll work something out.

[Chuck H]

I just want to say that it's great to see the real world differences being discussed instead of the usual "there is no difference" argument that the most technically inclined often use. There is indeed no difference, all things being equal. Which they never are. Justin was good enough to cover it so I won't bother, but, indeed, to make two tubes actually sound the same the operating conditions have to adjusted for each tube. And this is why the real world difference in tone between tubes is typically small enough to be ignored. With a few notable exceptions...

No offence to other "old timers" here, but in the golden age of tubes I think it was more common for tubes, brand to brand, to conform to some standards and therefor perform more similarly. But I do think current production tubes are a bit less standardized. That is, actual mu, capacitances, current characteristics and hard vs. soft cutoff character are more variable today. As noted, even these differences are more relevant when a tube is clipping. But isn't that just what we do to tubes in guitar amps? So shouldn't evaluation of a tubes "tone" WRT guitar amps ALWAYS include this consideration? Just sayin'.

So kudos to all the answers so far that are well beyond my technical abilities to understand completely. I'm following along to glean any education I can. But damn if there aren't tonal differences other than gain between tubes with a simple swap in the socket. So before that get's poo poo'd as false because "I" can't prove it on a technical battlefield I wanted to say that any limited technical argument proving tubes DON'T sound different MUST be just that, limited. Because they certainly DO sound different beyond just gain.