Standard AI response. Similar to " production-ready", "according to industry standards" or "common practices" to justify and action or indicating it is done, without even compiling or running code, let alone understand the output. An AI can't hear, and even worse, relate this. Ask it to create a diode ladder filter, and it will boost it created a "physically correct analog representation" while output ting clean and pure signals...
For context, I'm working on a proper SPICE component-level Diode Ladder.
I tried this for laughs with Gemini 3 Pro. It spit out the same ZDF implementation that is on countless GitHub repos, originating from the 2nd Pirkle FX book (2019).
Since there is a Ursa Major project on github, made by an owner, who reimplemented this also based on observation, made into a plugin, I wonder how much was regurgitated by the AI agent.
Indeed, same questions few days ago when somebody shared a "generated" NES emulator. We have to make this answered when sharing otherwise we can't compare.
At some point the llm ingested a few open source NES emulators and many articles on their architecture. So i question the llm creativity involved with these types examples. Probably also for dsps.
Right, the amount of hallucinated response data I see at work using any of these leading models is pretty staggering. So anytime I see one of these “AI created a 100% faithful ___” type posts that does not have detailed testing information, I laugh. Without that, this is v0 and only about 5% of the effort.
> i question the llm creativity involved with these types examples.
Indeed but to be fair I'm not sure anybody claimed much "creativity" only that it worked... but that itself is still problematic. What does it mean to claim it even manage to implement an alternative if we don't have an easy way to verify?
I’m not claiming a 100% faithful physical recreation in the strict scientific sense.
If you look at my other comment in this thread, my project is about designing proprioceptive touch sensors (robot skin) using a soft-body simulator largely built with the help of an AI. At this stage, absolute physical accuracy isn’t really the point. By design, the system already includes a neural model in the loop (via EIT), so the notion of "accuracy" is ultimately evaluated through that learned representation rather than against raw physical equations alone.
What I need instead is a model that is faithful to my constraints: very cheap, easily accessible materials, with properties that are usually considered undesirable for sensing: instability, high hysteresis, low gauge factor. My bet is that these constraints can be compensated for by a more circular system design, where the geometry of the sensor is optimized to work with them.
Bridging the gap to reality is intentionally simple: 3D-print whatever geometry the simulator converges to, run the same strain/stress tests on the physical samples, and use that data to fine-tune the sensor model.
Since everything is ultimately interpreted through a neural network, some physical imprecision upstream may actually be acceptable, or even beneficial, if it makes the eventual transfer and fine-tuning on real-world data easier.
Well I'm glad you find new ways to progress on whatever you find interesting.
This honestly though does not help me to estimate if what you claim to be is what it is. I'm not necessarily the audience for either project but my point remains :
- when somebody claims to recreate something, regardless of why and how, it helps to understand how close they actually got.
It's not negative criticism by the way. I'm not implying you did not faithfully enough recreate the DSP (or the other person the NES). I'm only saying that for outlookers, people like me who could be potentially interested, who do NOT have a good understanding of the process nor the initial object recreated, it is impossible to evaluate.
I had the hardware for both units and use them extensively so 100% familiar with how they sound.
And I'm not doing it based off of my ears. I know the algorithm, have the exact coefficients, and there was no guesswork except for the potentiometer curves and parts of the room algorithm that I'm still working out, which is a completely separate component of the reverb.
But when I put it up for sale, I'll make sure to go into detail about all that so people who buy it know what they're getting.
Why would I? When you buy a car part, they don't print on the box they used AutoCAD in order to build it. When you rent a movie they don't talk about using DaVinci Resolve to edit it, right? People use AI now to build software. I don't think that's going to change any time soon.
Maybe the OP has the hardware and can compare the sound both subjectively and objectively? Does it have to be 100% exact copy to be called the same? (Individual electronic components are never the same btw)
The OP didn't clarify. But if there's a claim of 100% faithful recreation, I'd expect something to back it up, like time- and frequency-domain comparisons of input and output with different test signals. Or at least something. But there isn't anything.
The video claims: "It utilizes the actual DSP characteristics of the original to bring that specific sound back to life." The author admits they have never programmed DSP. So how are they verifying this claim?
Well it's a new project so give it some time. I feel confident that I'm not lying so I can make that claim.
Also its target market is not a technical crowd but people who make music. I'm optimizing more for what they want to see (which are sound demos) rather than what a programmer would want to see.
Standard AI response. Similar to " production-ready", "according to industry standards" or "common practices" to justify and action or indicating it is done, without even compiling or running code, let alone understand the output. An AI can't hear, and even worse, relate this. Ask it to create a diode ladder filter, and it will boost it created a "physically correct analog representation" while output ting clean and pure signals...
For context, I'm working on a proper SPICE component-level Diode Ladder.
I tried this for laughs with Gemini 3 Pro. It spit out the same ZDF implementation that is on countless GitHub repos, originating from the 2nd Pirkle FX book (2019).
Ha! Textbook... Literally.
Since there is a Ursa Major project on github, made by an owner, who reimplemented this also based on observation, made into a plugin, I wonder how much was regurgitated by the AI agent.
Indeed, same questions few days ago when somebody shared a "generated" NES emulator. We have to make this answered when sharing otherwise we can't compare.
At some point the llm ingested a few open source NES emulators and many articles on their architecture. So i question the llm creativity involved with these types examples. Probably also for dsps.
Right, the amount of hallucinated response data I see at work using any of these leading models is pretty staggering. So anytime I see one of these “AI created a 100% faithful ___” type posts that does not have detailed testing information, I laugh. Without that, this is v0 and only about 5% of the effort.
> i question the llm creativity involved with these types examples.
Indeed but to be fair I'm not sure anybody claimed much "creativity" only that it worked... but that itself is still problematic. What does it mean to claim it even manage to implement an alternative if we don't have an easy way to verify?
I’m not claiming a 100% faithful physical recreation in the strict scientific sense.
If you look at my other comment in this thread, my project is about designing proprioceptive touch sensors (robot skin) using a soft-body simulator largely built with the help of an AI. At this stage, absolute physical accuracy isn’t really the point. By design, the system already includes a neural model in the loop (via EIT), so the notion of "accuracy" is ultimately evaluated through that learned representation rather than against raw physical equations alone.
What I need instead is a model that is faithful to my constraints: very cheap, easily accessible materials, with properties that are usually considered undesirable for sensing: instability, high hysteresis, low gauge factor. My bet is that these constraints can be compensated for by a more circular system design, where the geometry of the sensor is optimized to work with them.
Bridging the gap to reality is intentionally simple: 3D-print whatever geometry the simulator converges to, run the same strain/stress tests on the physical samples, and use that data to fine-tune the sensor model.
Since everything is ultimately interpreted through a neural network, some physical imprecision upstream may actually be acceptable, or even beneficial, if it makes the eventual transfer and fine-tuning on real-world data easier.
Well I'm glad you find new ways to progress on whatever you find interesting.
This honestly though does not help me to estimate if what you claim to be is what it is. I'm not necessarily the audience for either project but my point remains :
- when somebody claims to recreate something, regardless of why and how, it helps to understand how close they actually got.
It's not negative criticism by the way. I'm not implying you did not faithfully enough recreate the DSP (or the other person the NES). I'm only saying that for outlookers, people like me who could be potentially interested, who do NOT have a good understanding of the process nor the initial object recreated, it is impossible to evaluate.
3 replies →
I had the hardware for both units and use them extensively so 100% familiar with how they sound.
And I'm not doing it based off of my ears. I know the algorithm, have the exact coefficients, and there was no guesswork except for the potentiometer curves and parts of the room algorithm that I'm still working out, which is a completely separate component of the reverb.
But when I put it up for sale, I'll make sure to go into detail about all that so people who buy it know what they're getting.
Can you sell it, or would you have to do some renaming in order to get around trademark/etc ?
Consider reaching out to Audiority - I know they have some virtual recreations of Space Station hardware.
https://www.audiority.com/shop/space-station-um282
Luckily the trademark is public domain!
2 replies →
Are you also going to go into detail about the use of AI to generate the code?
Why would I? When you buy a car part, they don't print on the box they used AutoCAD in order to build it. When you rent a movie they don't talk about using DaVinci Resolve to edit it, right? People use AI now to build software. I don't think that's going to change any time soon.
4 replies →
Sell it?
Wat?
Perhaps a subjective evaluation based on how it sounds.
It’s bold to call it 100% faithful without some rigorous test harness though, isn’t it?
[dead]
[flagged]
He included "100% faithful" in the prompt!
“You are an elite DSP programmer who never makes mistakes..”
Maybe the OP has the hardware and can compare the sound both subjectively and objectively? Does it have to be 100% exact copy to be called the same? (Individual electronic components are never the same btw)
The OP didn't clarify. But if there's a claim of 100% faithful recreation, I'd expect something to back it up, like time- and frequency-domain comparisons of input and output with different test signals. Or at least something. But there isn't anything.
The video claims: "It utilizes the actual DSP characteristics of the original to bring that specific sound back to life." The author admits they have never programmed DSP. So how are they verifying this claim?
Well it's a new project so give it some time. I feel confident that I'm not lying so I can make that claim.
Also its target market is not a technical crowd but people who make music. I'm optimizing more for what they want to see (which are sound demos) rather than what a programmer would want to see.
That might make it 100% faithful for OPs use cases, but not necessarily anyone else's.