"I did all of this myself. I drew every single one of those damn pictures myself. I came up with which parameters were needed and which weren’t myself. I designed the architecture of the sub-parameters and wrote the descriptions for all my ideas and specs myself.
You people have latched onto this AI thing so much. I used it for translation, and I am exhausted from writing this everywhere. I do not speak English, so I used it to translate. On Reddit, everyone jumped on me at once like I was the enemy. I even had to delete the post because there was a strong feeling that you don’t care about the concept or the development of the device, but just want to trash everything and pick on the AI part.
That’s all from me. I am not accusing you specifically of anything, it’s just very hurtful that everything I spent a long time thinking through was dismissed as being made by AI.
And yes, here is the list of parameters from all nine pages and their impact on the processor. If, after this, someone still claims that the Polyend can’t handle this, then any store calculator is more powerful than the Polyend—which is obviously not true, right?"
1. Morph ( on Track Length / Play Mode page ) It is just a “lookup table” (LUT). All values are already in the memory. The CPU doesn’t calculate anything new, it just “looks at the paper” and picks a number. It’s super light, maybe even lighter than the current Fill randomizer.
2. Glide ( on Note / Micro-tune page ) Bipolar knob: left for Always Glide, right for Legato (like TB-303). It’s just a simple pitch slide. Any digital player can do this easily. The CPU just slowly changes the playback speed of the sample. No magic here.
3. Note Length ( on Note / Micro-tune page) This is just a software timer for the Gate. I added 0.25 precision for short “stabs” and micro-rhythms. Zero CPU load.
4. Advanced Filters ( on Filter Cutoff / Resonance page ) The idea is to use the filters that are ALREADY inside the Play+ synth engines. We just “re-route” them to the audio tracks. You swap one filter for another – the load is the same, but the sound is much better.
5. One Shot Mode ( on Sample Start / Sample End page ) basic thing. Just play the sample as it is. Lowest load for the system.
6. Warp Mode ( Virtual strech on Sample Start / Sample End page ) IMPORTANT: This is NOT time-stretching like in a computer! This is “virtual slicing”. The sample just starts from a specific point (Offset) to stay in sync with BPM. This is how Digitakt works. It’s very easy for the hardware.
7. Grid Mode ( Virtual slicing on Sample Start / Sample End page ) Same as Warp. We just tell the CPU: “play the sample from slice #15”. Same load as a normal One Shot.
8. AHD Envelope ( Sample Attack / Sample End page ) Added a Hold knob via the screen button. Plus “FREE” mode for smooth transitions without clicks. Simple volume logic, very light.
9. Arp Mode (Arpeggiator on Sample Start / Sample End page ) Just a note generator. For the audio engine, it looks like you are just pressing pads very fast. The voice limits are the same, so no CPU problems.
10. Drive ( on Distortion / Bit depth page ) Simple “waveshaping”. These algorithms are already in the synths, we just move them to the audio tracks.
11. Pre/Post Filter Routing ( on Distortion / Bit depth page ) This is just a “switch”. What goes first – filter or distortion? The CPU just changes the order of blocks in the chain. It’s “free” for the resources.
12. Chorus and Phaser ( on Reverb / Delay Send page ) I agree this is some load. But I put them in Send Effects. They work only once for the whole mix, not for every track. The load is fixed and predictable.
13. Move (on Move / Micro-move page ) Step geometry. The CPU just moves notes in the grid. It’s just simple math.
14. Fill ( Your own library without a computer ) This is a killer feature. Make a pattern, tweak the sounds, and save it as your own preset inside the box. Then just load it in any new project via Fill. This is just SD card memory work, zero CPU load during playback.
15. Smart Freeze (RENDER) – THE MOST IMPORTANT If someone is worried that the CPU can’t handle heavy FX — here is the solution. We just “freeze” the track. Play+ renders it to audio with all FX and plays it back as a simple file. Then the CPU load for this track drops to ZERO. This removes all hardware limits.
16. Master Compressor ( like in Digitakt 2 ) Yes, this takes some resources. Но (but) it’s only one on the Master bus. The ARM CPU in Play+ is strong enough for one good compressor with sidechain. We need this for a professional sound “out of the box”.
17. Distortion for internal Synths Added a separate drive block inside the synth engines. So synths can “growl” before they hit the master bus.
18. Key Tracking Link parameters to notes. Formula: Value = Note \* Coeff. Even a calculator can do this.
19. Perform Mode Automation Record your hand movements into the grid. For the CPU, there is no difference if you turn the knob or if the automation does it. It’s just a data stream.
Wild stuff man. It shows you put some real though and effort into this.
I was really hoping that you had actual evidence that this stuff could work on the Play+, but it seems more of an AI hallucination than something realistic. I mean this with kindness, but you should really grab a Play+ before trying to influence it’s functioning and spend some time looking into the facts of DSP and reverse engineering if you’re going to use AI for the specs, as it’s been known to give inaccurate information. I was pretty stoked about this idea, but it doesn’t seem like it’s possible to do this on firmware alone
maybe an actual dev from the polyend team could jump in and give us a little more scope about the realistic application of your ideas and then adjust the feature request to match that