Fractal Talk

[State of Epicicity]

Not to forget that just because those capacitors or electronics behaved within tolerance when new, doesn't mean they would behave that way 30 years later. Some of these amps are no longer made or were of a certain generation. The 2203 and 2204 come to mind. I think the lifespan advertised by vendors for electrolytic capacitors is 15 years. Not to say that Fractal doesn't test everything on their end, but for those that own the products and are trying to do a comparison, have they maintained their amps?

Cliff has written on the Fractal forum that they do get their amps serviced regularly. I'd hate to see that maintenance bill!

 

[State of Epicicity]

Not to forget that just because those capacitors or electronics behaved within tolerance when new, doesn't mean they would behave that way 30 years later. Some of these amps are no longer made or were of a certain generation. The 2203 and 2204 come to mind. I think the lifespan advertised by vendors for electrolytic capacitors is 15 years. Not to say that Fractal doesn't test everything on their end, but for those that own the products and are trying to do a comparison, have they maintained their amps?

Here's where Cliff talks about the amp servicing:

Post in thread 'Axe-Fx III Firmware 25.00 Public Beta (Cygnus X-3)'
https://forum.fractalaudio.com/thre...00-public-beta-cygnus-x-3.202659/post-2542066

And here's the important part:

"Now, our reference amps are always serviced before matching. Any resistors that are out of tolerance are replaced. Capacitors are replaced as needed and new tubes installed."

 

[DLC86]

Depends on the mfg. and the parts they spec but using a Marshall BMT tone stack as an example:

3 - variable resistors
3 - resistors
4 - capacitors

Assuming a +/- 20% tolerance:

R1 is 33K ohm with an acceptable variance of 26.4 to 39.6 - a range of 12.2K
C1 is 470pF with an acceptable variance of 376 to 564 - a range of 118pF

Now multiply every possibility by this:

No way slight tweaks to some variable resistors is going to compensate for all the range possibilities.
Back in the 80s there was this one JCM800 head that all the hair bands were trying to steal because
it just happened to end up with a combo of part values that sounded better than the other 800s.

If you guys can hear/feel differences with the most recent updates you can bet your house that you
could hear the differences between various production amps if given an optimum AB testing setup.

A 20% tolerance is common for pots but surely not for metal film resistors which is usually 1% for the cheapest ones, color coding only gets to 10% so I assume that's the highest for old carbon ones.
Never checked about caps but by a quick googling it appears 10% is common for electrolytics, oil and ceramic, other types are usually lower.

But what you said is certainly still valid. Several years ago I built two identical clones of a Hiwatt DR504 and, despite being built exactly with the same components, same set of tubes (the first EL34s I mounted on them were all coming from a matched quartet) and biased identically with the same multimeter, they never sounded really the same, I always had to set their tonestacks slightly differently to try to match them, but it was an impossible task regardless, they were still different.
Not a huge difference but definitely hearable.

PS: and the same always happened even when building identical pedals, despite them being much more simple circuits than amps.

 

[Achilles]

color coding only gets to 10% so I assume that's the highest for old carbon ones.

It goes lower.

 

[DLC86]

It goes lower.

That's what I said

 

[la szum]

Used the new firmware with the band last night. Since we all know one another
so well, and are jaded old men, compliments don't get handed out too often, but
I received them both from the bassist (who plays on the same side of the stage
as I do), and the other guitarist last night on tones and playing.

Conincidence??? Probably.

 

[djd100]

With all of the production tolerances involved - and the fact no two amps of the same model sound EXACTLY the same anyway - is there a point where an amp gets signed off on by Fractal as being complete, i.e.; no more tweaking to that particular amp model with future updates?

No, as Cliff over time is improving the modeling of various circuits and components which improve various aspects of the sound, and by "improve" I mean getting the modeling to sound closer in general to the tube circuits and components it's trying to emulate under a infinite number of conditions.

At times the improvements are subtle, other times not so much. For the user then, the only updated algorithms that matter are the ones that a specific user can hear and/or feel, noting that the various users all have varying critical listening skills, and varying actual tube amp experience, while noting again that tube amp experiences in general are unique to each individual user (human hearing).

So, as always, use the FW Rev that sounds best to you or has the features you need, keeping in mind that Cliff has a high level of critical listening skills along with a high level of tube amp experience, and no doubt hears things that the majority of users can't, which is why he continues to strive to improve what has already been the most accurate component modeling guitar amp modeler for a long time.

 

[Stone]

Used the new firmware with the band last night. Since we all know one another
so well, and are jaded old men,compliments don't get handed out too often, but
I received them both from the bassist (who plays on the same side of the stage
as I do) and the other guitarist last night on tones and playing.

Conincidence??? Probably.

So they all looked at each other and -->

 

[Whizzinby]

Hasn't that particular amp been nailed and put to bed by now?

Not if we have anything to say about it!

 

[Orvillain]

 

[dean701]

Here's where Cliff talks about the amp servicing:

Post in thread 'Axe-Fx III Firmware 25.00 Public Beta (Cygnus X-3)'
https://forum.fractalaudio.com/thre...00-public-beta-cygnus-x-3.202659/post-2542066

And here's the important part:

"Now, our reference amps are always serviced before matching. Any resistors that are out of tolerance are replaced. Capacitors are replaced as needed and new tubes installed."

that's exactly to what I was referring. The fractal parts have been serviced but when owners of amps are comparing to Fractal, and they complain that their amp sounds different than the Fractal offering, have they done the same maintenance to their own amps?

 

[unix-guy]

I had a chance to spend a little more time with my FM9 with FW8 on it yesterday. I am quite impressed. I kept thinking that I don't remember this thing ever sounding this good and having this much feel to it. I had honestly forgot that I put the new firmware on it last week before I went out of town. To me, that is a testament as to how this new firmware has improved the models.

FM9 is FW7... FM3 is FW8

 

[GTR37]

FM9 is FW7... FM3 is FW8

ahh Tomato Tomoto, Potato frittata. .same dif

 

[MirrorProfiles]

Surely the improved modelling is more to do Cliff’s ability to measure the real circuits, and model complex non-linear behaviours in real time, and finding ways to avoid cutting corners while being as accurate as possible.

I guess there’d also be some refinement based on circuit variations, but fixing, correcting incorrect values etc.

 

[aflynt]

I was playing around with FW8 on my FM3 this morning trying out different amps. I dialed in the JS410 Lead Orange, and wow! I'm not a shred guy at all, but that thing sounds fantastic. Almost makes me want to join a metal band.

-Aaron

 

[Warmart]

FM9 is FW7... FM3 is FW8

So you're telling me that the FM3 has more updates than the FM9!

 

[JiveTurkey]

The FAS Vibe is spectacular

 

[Boudoir Guitar]

Yeah whenever I dabble with multiple digital units I get stuck in a A/Bing hell. This doesn't happen with real amps for some reason and I'm able to recognize those serve their own purpose. I wish I could view a digital device similar to that of an amp head but alas

Even if I'm not in A/Bing hell, I find myself not being all that efficient on either unit as my brain takes time to get back into the workflow of the other unit.

 

[Achilles]

The FAS Vibe is spectacular

If you're ever looking for a killer pedal version try this:

VIBE MACHINE V-3 - DryBell

An exceptional sounding Vibe, a bit different from the previous versions with its own unique sound, character and depth.

drybell.com

I went on a mad search for the best vibe in a stomp and this one was the winner.

 

[JiveTurkey]

If you're ever looking for a killer pedal version try this:

VIBE MACHINE V-3 - DryBell

An exceptional sounding Vibe, a bit different from the previous versions with its own unique sound, character and depth.

drybell.com

I went on a mad search for the best vibe in a stomp and this one was the winner.

I know this particular journey is a quest and a half (at the very least). I'll stick with the digital on this one