S
Solarflares
Guest
Bugger that.
That would be like using an information Highway to send pics of your last meal.
Where/when did "modern" guitar tone begin?
- Thread starter misterwhizzy
- Start date
That would be like using an information Highway to send pics of your last meal.
Chistopher
malapterurus electricus tonewood instigator
It's actually very simple. It's kind of like how you can put a picture on a screen using a bunch of small RGB LEDs, except you are making a circuit using a bunch of small units of varying inductance, resistance, and capacitance. There's more to it than that, but it's not exactly reinventing the wheel
S
Solarflares
Guest
No. I know exactly what you’re talking about.
I can only bounce back and point out that another way is to intelligently predict the effect a virtual circuit would have on a complex waveform, and redraw it.
Chistopher
malapterurus electricus tonewood instigator
That is what a lot of companies are doing now. I know for a fact the QuadCortex does that. It doesn't exactly work with non-linear systems though. You couldn't model a Fuzz Factory through your method, you could with mine however.
S
Solarflares
Guest
Define non-linear systems.
You are not being clear.
The end result is an altered waveform, whatever method.
80's_Metal
Underglazed Hair Metalologist
One could argue this.
Chistopher
malapterurus electricus tonewood instigator
Nonlinear means you can't apply a constant transformation to a signal. More generally, changes in input don't produce proportional, and easily predictable changes in output.
In a nonlinear system, the input signal at t=0 might effect the output signal at t=1. Or the amplitude of input frequency A might effect the amplitude of output frequency B
Chistopher
malapterurus electricus tonewood instigator
Also FPGA is real-time. Not even the best algorithm can match that
i love ac/dc and think the brothers both get great tone, but theres nothing ground breaking about an sg into a cranked marshall played with attitude, even if it kicks ass. now the shorts... that might be a first
S
Solarflares
Guest
I can see that FPGA has relevance in a hybrid system, if we consider it to be an analogue substitute.
Which leaves me thinking that the next obvious move would be quantum technology with an FPGA buffer at each end.
marcello252
Well-known member
the more I read the more I love my small tube old Princeton
Chistopher
malapterurus electricus tonewood instigator
Yeah, I could also see them working in parallel. FPGA's weakness is that it models the hardware but doesn't "know" the sound, and Quad Cortex has the opposite problem.
S
Solarflares
Guest
Now you’re getting there.
I think what you will ultimately see is a Quantum matrix core with multiple FPGA insert points, with the ‘encompassing’ FPGA able to access all areas of itself simultaneously. We can then use AI, and teach the matrix why a certain thing will happen, so it has more ‘prediction’, and increasing sympathy toward user preferences.
Last edited by a moderator:
uOpt
Something Cool
Don't underestimate how much compute power modern CPUs have.
i have a nice stable of tube amps, doubt ill need anything else in the foreseeable future
S
Solarflares
Guest
All I want Jeremy, is for my Mullard 161’s, and Seymour Duncan Convertible 100 to keep going until my final breath.
And they can bury me with my two ‘87 guitars.
Chistopher
malapterurus electricus tonewood instigator
I'm being literal. Even if the CPU had 0 latency, my design would in theory be faster.
Cynical
Active member
How would this make any sense at all? All you can do with FPGAs is look up Boolean logic tables; guitar signals aren't discretes, they're not gates and muxes, so this doesn't even begin to work unless you think you have an FPGA with a LUT large enough to do an entire analog-to-digital conversion, an entire ampsim's worth of programming logic, and a digital-to-analog conversion.
Does VHDL's real type even have enough bits to be useful for studio audio?
Similar threads
