Reply To: BEST IN SHOW

#22579
AXP
Participant

Hi Tele,

I’m kinda used to the time difference, because the company I work for has an HQ in the U.S.
After reading about you having good and bad news I thought you were going to say the interface worked fine, but the sound sucks 🙂
Glad to hear that you like it. In fact it’s based on the generally acclaimed tube modeling approach, but the real trick is that it’s very computationally efficient. Plus I’ve got a very fast oversampling engine, which I’ve spent a lot of time on perfecting it out. It takes about 230 CPU cycles to process a single mono sample without oversampling and less than 3500 samples with 16x oversampling, which includes 16 executions of the triode processing loop as well as 2 passess of the antialiasing filter, making it sub-linear in performance.

I would definitely like to see the screenshots. There are contact details in the plugin’s ‘about’ page (just click the logo). I am still not able to reproduce it. Reaper’s “clear cache and rescan” completes perfectly, on 5 different PC I’ve got. It works even through the remote desktop session! I’ve also got a trial version of Cantabile 2.0 and it works just fine as well. I think I’ve got Ableton Live bundled with my audio interface, so I’m going to test it too. Moreover, I’ve tested it with some audio editors – Audition 2.0 and Ocenaudio 2.0 with both working without a glitch. It even works through a wrapper in the Foobar player. There must be something very specific about your configuration that escapes me.
May I ask you to run ‘msinfo32’ on your PC and then save and mail me its output? (alternatively it’s in Start->All Programs->Accessories->System Tools->System Information)

About the clicks when changing the parameters. Actualy it was even worse in the beginning, but I’ve implemented a quick workaround in 1.0.1 that somewhat alleviated the issue. Well, it still sounds better than swapping real-life tubes would 🙂 The reason for clicks is that different tube models will have different bias points and circuit capacitors need some time to settle on a new one. If you look at the output signal in the editor, you’ll see that such transients look exacly like they would in the real life.

Pushed by your report though I’ve just spend an evening working on the proper solution and now I can say that it’s resolved completely. I’ll release the new version shortly.

I smiled while reading that the Gain knob is causing pops too. In fact ‘leds’ have nothing to do with the actual value. The knob has 100 internal steps from 0.0 to 1.0. The value is truncated for display, but the full precision is used for controlling the DSP part. What you’ve actually encountered I think is that the gain is very sensitive to mouse movement and each pixel of vertical movement causes a significant dB change. You can check if it’s smooth enough by using the automation input.

I would be grateful if you could spread this piece of software around, but let’s iron out those bugs first 🙂
Meanwhile I’m working on some more stuff, because obviously this one is going to be just a building block for more serious things which you might be also interested in.

So yeah, send me your screenshots and system info log, and keep looking for updates.

Thanks again,
/AXP