Genz-Benz Streamliner Part 5

[Eublet]

I'm censuring myself because there was a ground loop in these graphs, causing the readings to be off, particularly in the lows. I'll start a new thread and repost my new graphs once I'm comfortable with them.

 

[Scott McArron]

I've been using Bass at 9:00 myself. Sounds the most balanced around there. I push it to 10 if I want a little extra low end. Then I pump the 2.5 up to about 2:00. Fills in that little dip I hear and adds a nice bit of definition in the mix.

Thanks for the graphs! They look how I feel the S900 sounds to me.

 

[agedhorse]

Be sure to take these graphs in context of the vertical scale when comparing to any other graph. Changing the vertical scale can change the impression of the curve even though the data might be identical. For curves to be compared (visually) with any kind of accuracy, the horiz. and vert. scales must be the same.

 

[nostatic]

Fits my ear experience exactly - and why I cut bass and bump at 600hz especially relative to the Shuttle9.0 that I run pretty much dead flat. Nice work.

 

[Eublet]

Another quick note on the bass graph, if you compare the settings of Bass-10:30 with the bass set at noon, you'll notice that all that is really happening is that the roll off of the lows happens sooner. It's not really taking off much bass at the "knee" at all. It seems to have the effect of acting like a more aggressive HPF when you first start twisting it, which is a COOL thing. But again, if you think the head sounds to big at noon, only dropping the bass down to 10:30 might not give you the results you are looking for since much of the low-end content in the 50-80 hz appears to be unaffected when doing that. Cut some more if you want to smooth out the bottom a bit (reference the cuves Bass-10 and Bass-9)! Notice the reoccurring them here!

 

[nostatic]

Be sure to take these graphs in context of the vertical scale when comparing to any other graph. Changing the vertical scale can change the impression of the curve even though the data might be identical. For curves to be compared (visually) with any kind of accuracy, the horiz. and vert. scales must be the same.

Bah! Engineers...next you'll want controls and statistical analysis.

As my first chem prof told me - you need 3 points to draw a straight line. Though once in grad school a colleague of mine drew a parabola with only one data point - and it wasn't at the vertex

 

[Eublet]

Be sure to take these graphs in context of the vertical scale when comparing to any other graph. Changing the vertical scale can change the impression of the curve even though the data might be identical. For curves to be compared (visually) with any kind of accuracy, the horiz. and vert. scales must be the same.

This is a good point, and kind of what I was saying before. I think the shape of any SINGLE curve is pretty good, although the exact location of the frequencies between any two curves might not be as good as it should be. I think this might be true because, for example, the graph I did for the 2500hz setting seems to imply that boosting mids at 2.5k also cuts highs above 10k if you use the Y axis for reference. I don't think this is true at all by what I hear. I think the software package is picking a relative point for "0db" for each curve, and then plotting them both on the same graph. I don't want people to be too mislead by this.

 

[12bass]

If you're using RMAA, it uses a 1 KHz tone for setting 0 dB IIRC. It is also possible to increase the vertical resolution to get a more exaggerated response curve.

 

[Eublet]

If you're using RMAA, it uses a 1 KHz tone for setting 0 dB IIRC. It is also possible to increase the vertical resolution to get a more exaggerated response curve.

I did, and thanks for pointing me toward it. I tried to disable the normalizing feature, but it did little to help. I'm going to play with it some more, and I'll replace these links with the new graphs if I get it figured out.

Edit: Looking at the flat graph, it's impossible to tell what it is using for 0db. The more I look at these graphs, the more they bug me. Not the curves themselves, but the placement of any two on the same axis. This is unacceptable really. I have more work to do. LOL!

 

[Jazzdogg]

Kesslari, I've never heard a 60's jazz bass with flats "cut through" in the sense you seem to be using it. In the context of that clip however, I would describe the bass tone as being much more than supporting. You have note definition, pitch, and articulation. You can hear runs, double stops, staccato rhythms, slides, etc. I don't hear that bottom heavy, sub-driven supportive tone that seems to be so popular in today's PA mixes. It does however sound like a 60's jazz with flats, meaning there is no grindy top on it.

+1

I like the bass sound on that video - clear, articulate, authoritative, and ballsy without unnecessary gank, fret clatter, finger noise, or other annoying high-mid content, or so much deep bass that it overwhelms the on-stage mix.

 

[tombowlus]

Be sure to take these graphs in context of the vertical scale when comparing to any other graph. Changing the vertical scale can change the impression of the curve even though the data might be identical. For curves to be compared (visually) with any kind of accuracy, the horiz. and vert. scales must be the same.

+1 And if possible, narrow down your vertical range a bit. If you can establish a 0 dB point, setting the scale to +/- 5 to 10 dB would be more helpful.

 

[KJung]

This is a good point, and kind of what I was saying before. I think the shape of any SINGLE curve is pretty good, although the exact location of the frequencies between any two curves might not be as good as it should be. I think this might be true because, for example, the graph I did for the 2500hz setting seems to imply that boosting mids at 2.5k also cuts highs above 10k if you use the Y axis for reference. I don't think this is true at all by what I hear. I think the software package is picking a relative point for "0db" for each curve, and then plotting them both on the same graph. I don't want people to be too mislead by this.

Nice job. Exactly what I would have thought... relatively even through the mids with slight dip in the mid mids and no upper mid spike, but a BIG low end that is still pretty 'round' (i.e., if 'peaks' at the right place to not sound muddy or mushy... just big and fat and full). So, the ratio of mid/upper bass to mids is a bit higher than other heads (and quite similar to some all tube heads I've experiences.. that 'fat smooth' thing).

It also explains the magic of this head with the Bergantino AE410 (and, for a smoother tone, the HS cabs), which is a kind of yin yang of that plot... tight low end, aggressive upper mids and mid present in general. The Streamliner with the bass set at about 11 o'clock and the 600hz boosted just slightly to 1 o'clock sounds about as perfect as anything I've yet heard with that cab

 

[Eublet]

+1 And if possible, narrow down your vertical range a bit. If you can establish a 0 dB point, setting the scale to +/- 5 to 10 dB would be more helpful.

This I thought of doing, Tom, but I was concerned that many would perceive the slopes to be more exaggerated than what they are.

I do intend to work on the 0db reference thing, but I see no way thus far to accomplish that. If I get that figured out, I'll also see about changing the scale on the db axis also. Thanks for the input!

 

[faulknersj]

Thanks HvyMtlThunder. I think the Streamliner easily provides a 'balanced' tone, enough clarity in the upper mids to cut thru, and a nice cushion of big clean bottom end for the rest of the mix to sit on top of. And, those are 6 month old flats; fresh rounds can really zing thru this head if you want them to!

 

[tombowlus]

This I thought of doing, Tom, but I was concerned that many would perceive the slopes to be more exaggerated than what they are.

Obviously, you don't want to go overboard, but by the same token, if you use too broad of a scale, then you run the risk that many will perceive the slopes to be less significant than what they are.

FWIW, in our last issue, we used scales of +5 to -5 dB for the M3 Carbine, and +2 to -5 dB for the F500 (because the F500 didn't have any peaks above +2).

 

[adivin]

Why don't manufacturers publish this kind of data?

 

[ElMon]

Thanks HvyMtlThunder. I think the Streamliner easily provides a 'balanced' tone, enough clarity in the upper mids to cut thru, and a nice cushion of big clean bottom end for the rest of the mix to sit on top of. And, those are 6 month old flats; fresh rounds can really zing thru this head if you want them to!

+1, and your clips also highlight the magic that are D'Addario Chrome flats. I've never had a flatwound with as much definition/clarity while retaining the classic flat thump. Good job!

 

[agedhorse]

Typically, a scale of say +/- 20dB relative to a nominal zero (whatever the baseline datum ends up) gives a pretty good indication of the relative solpes to the overall response. If you get much finer than this you risk not being able to see the forest for the trees. The other thing that's just as important is whenever comparing graphs between different models or manufacturers, you MUST use the same scale or the comparison will be distorted.

 

[vision]

Why don't manufacturers publish this kind of data?

IMO, those of us on TB have a greater amount of appreciation for data like this and have more of an ability to understand what it means than most of the buying public. If manufacturers published data like this there would be alot of guys out there that don't really understand it taking it out of context and letting it affect buying decisions.

There's still a ton of people that for some reason think that wattage ratings on cabs have a correlation with how loud the cab is for example...and alot of guys that don't understand amp power ratings per 2 ohms vs 4 ohms when certain manufacturers publish 2 ohm ratings. You give the average Guitar Center customer graphs like these and there's no telling what he'll think it means!

 

[KJung]

Why don't manufacturers publish this kind of data?

IMO, because it causes more confusion than anything else. Without looking at the entire system, plots of one component are not very helpful. A 'flat' amp through a highly voiced cab with a scooped sounding bass could sound more voiced than a 'voiced' amp through a cabinet that yin yangs with it well. Plus there are about another zillion variables that define the true sound of an amp. Looking at a few, much less one, can be more misleading than helpful.

Looking at the 'scope' results of an amp without taking into acount the cabinet and bass, strings, technique, and in the case of the Streamliner, the impact of various settings of the input gain is more 'Talk Bass interesting minutia' than anything that would be that useful to a potential customer.

IMO!