Satin, different sound at different sample rates, be aware of it
-
- KVRist
- Topic Starter
- 33 posts since 24 Nov, 2021
Do you have any news on the potential release date of the fix?
-
- u-he
- 30175 posts since 8 Aug, 2002 from Berlin
I do not know. There are a couple of open questions still, but my focus is currently on something else. I might get back to Satin in December.
-
- KVRist
- Topic Starter
- 33 posts since 24 Nov, 2021
Hi Urs, Do you have any news regarding a corrected Satin release?Urs wrote: Thu Nov 24, 2022 6:02 pm I do not know. There are a couple of open questions still, but my focus is currently on something else. I might get back to Satin in December.
-
- u-he
- 30175 posts since 8 Aug, 2002 from Berlin
It's more or less fixed with at the best possible compromise. However, our codebase currently has a plethora of construction areas that make it impossible to even post a beta version. So unfortunately we'll have a few months to work on those areas before we can post any updates I'm afraid.
-
- KVRist
- 82 posts since 14 May, 2004 from ---
Question for those concerned...why do you want it to sound the same at different sampling frequencies? I mean, if it sounds the same anyway, why use different sampling frequencies? This only generates excessive data volume.
---
-
- u-he
- 30175 posts since 8 Aug, 2002 from Berlin
The issue here is that it obviously sounds wrong. The reason is that the tape model - correctly - has weird shelving filters which - at high samplerates - over-emphasize high frequency content beyond 20kHz. Tape should not have much of this content, so I've implemented an alternative way to achieve the correct frequency response that closely mimics the shelving filters at high samplerates, but without adding emphasize to content that a real tape would not have.
The problem is not necessarily immediately audible, in most cases a nudge at the Emphasis knob is enough, but in conjunction with the compander it can change the overall volume.
Unrelated to this is an issue where the visual feedback in Satin differs vastly between samplerates, which is simply a result of the resolution of the FFT algorithm used. I have improved this greatly by increasing the resolution of the FFT and by oversampling the measurement impulse. This however is impossible to achieve for the readout after using a compander or driving very hard.
The problem is not necessarily immediately audible, in most cases a nudge at the Emphasis knob is enough, but in conjunction with the compander it can change the overall volume.
Unrelated to this is an issue where the visual feedback in Satin differs vastly between samplerates, which is simply a result of the resolution of the FFT algorithm used. I have improved this greatly by increasing the resolution of the FFT and by oversampling the measurement impulse. This however is impossible to achieve for the readout after using a compander or driving very hard.
-
- KVRist
- 82 posts since 14 May, 2004 from ---
Ah understand. It has to do with authenticity of the emulation in terms of the tape machine. Ok, personally I mostly just use it for saturation and I don't really care if it sounds like the real tape machine. If it sounds good, it's good for me 
. Maybe satin is a bit oversized for my personal needs .
. Maybe satin is a bit oversized for my personal needs .---
-
- KVRist
- 274 posts since 31 May, 2017
I am struggling with following the technical stuff, but am i correct when i say that the main issue with higher sample rates is that it is just slightly too bright compared to the same settings at lower rates, and that slightly darkening via pre-emphasis mostly fixes the issue?Urs wrote: Thu Mar 23, 2023 12:30 pm The problem is not necessarily immediately audible, in most cases a nudge at the Emphasis knob is enough, but in conjunction with the compander it can change the overall volume.
I'm not using visual feedback or compounder so that doesn't affect me. But would like to be clear about what is going on until the fix is out.
-
- KVRian
- 851 posts since 12 May, 2004
Urs…if you don’t mind, where are we with this? Will there be a revised release any time soon?
On a number of Macs
-
- u-he
- 30175 posts since 8 Aug, 2002 from Berlin
After I mostly fixed the sample rate issue about a year ago, we worked on adding the missing AAX support while we were tackling a few general issues that kept us from releasing anything. During that time I shifted focus to Filterscape and Zebralette 3. AAX support for Satin was looking good about 4 or 5 months ago and the technical issues were also resolved. But by then it turned out there seem to be some remaining issues that I have to get back to and which block any effort to publish the update.
Unfortunately I won't be able to do that before Zebralette 3 is at least in public beta for a few weeks.
I wished I could task or hire other developers to work on this, but I think atm this would cost even more time and stall development even further.
Unfortunately I won't be able to do that before Zebralette 3 is at least in public beta for a few weeks.
I wished I could task or hire other developers to work on this, but I think atm this would cost even more time and stall development even further.
-
- KVRian
- 851 posts since 12 May, 2004
Thank you, Urs, for the detailed reply. A little disappointing but, I totally get the current focus on Zebralette 3….that, and the fact that there aren’t enough hours in the day.
On a number of Macs