mcpersso wrote:In autumn 2015 I bought a Eminent 310U. I was a happy man. It was the right Jarre sound that I had hunted in various plugins.
But I missed that one could quickly change in what had been recorded without having to record the whole effort again. I had just gotten reason 9, and there was a plugin called Combo 310 Unique Organ. It was a very nice interface and fairly sound, but stringer part failed bravely.
I thought there had to be better. And then I started (if I had known then how much work lay ahead, I would never have done it).
My first requirement was that the sound quality should be top notch. So I decided that it should sample at 24-bit / 96 kHz. In addition, having the noise removed. With RX3 and x-hum succeeded after many tests.
I have never understood the issue you've had with the string ensemble on Combo 310U, it sounds quite fine to me: in no way is it a failure, and no-one else has had an issue with it. If I recall correctly you kept claiming the samples were flat and had no movement, which is bonkers, given it's what my 310U sounds like - it's a DI recording (ignoring the entirely separate and somewhat experimental stereo recording of the speaker output). I disagree entirely with your assessment; I find the Combo 310U DI samples warm, rich, lively and naturally free of distortion or noise.
As I pointed out the last time you posted a complaint here (and probably on RT too), samples of organs will never
quite sound the same because of the lack of oscillator sync in samples. Within that natural and expected limitation, Combo 310U sounds as close to a 310U as I've heard, with higher fidelity samples than any previous attempt I am aware of, such as those I licensed for the Retrospective ReFill (and even those still pretty cool actually, the ones from Hollow Sun and Per Kristian), save perhaps GForce's VSM, which I've not heard, though now I'm curious to compare. However, that's somewhat a different thing anyway, a general strings ensemble mixer, rather than 310U emulation.
But consider this, the sampled E310's I mentioned above sound a hell of a lot better than the 310U Strings for Reaktor, which while oscillator-based, and thus does have the oscillator sync, which is good, on the other hand is it's far too bright, and worse, aliases really badly, making it entirely unusable above about C4 (4') or C5 (8').
It's hard to judge listening to the SoundCloud demos as you've swamped it in the phaser to hide the actual string ensemble, but to me it sounds ragged. Take the phaser off and it sounds loud but thin.
You're talking about wanting the sound quality to be "top notch" and the usual "24/96 sampling", but frankly, we all do that, and really, it's the 24-bit and using a DI box to prevent ground loops that's important at the recording stage to provide the best SNR at the destination when recording these types of things.
But that's by the by: it's the next bit you write that has me baffled: you state you had to remove noise
with RX3 and x-hum. I know this is going to sound like sour grapes but since you were repeatedly and publicly so critical of me for really no reason other than you wanted to do your own version, it would be very remiss of me not to point out that if there was that much noise from your E310* that it needed RX3 and X-Hum to clean up the results, the sample quality at the recording destination couldn't have been that good to start with from the source, and I know from testing RX in the past the noise removal will have an impact on the overall tone. I don't know RX3+, I think it was RX2 I tested back when I was working on a project with Hollow Sun on the DK Synergy; I remember we hated the RX2 result so much I abandoned it and where relevant I truncated the sample decays to just before the compander brought the noise level back up to unusable levels. It actually worked quite well. The moral there is while I can't predict or claim to know how much impact it would have, you certainly can't say your samples are an exact recording of your organ: by your own admission you've changed the tone to some extent by using multiple de-noisers.
Nonetheless, based solely on my experience of the 310U I own, there is very little noise from the line outputs, nothing, in my view, that would need post-recording surgery at sample level**, and that little bit which is present I'd argue precisely one wouldn't
want to remove as that's a crucial element of the make-up of the sound: turn up the volume and listen to that opening bass note of Equinoxe again. Vintage gear is not noise-free.
If your samples had that much noise into a modern 24-bit DAC (or rather, ADC) above the low but noticeable level of Jarre's original 310U recorded*** on a dodgy old 8-track, or whatever it was, that they needed to be heavily cleaned up with two different hum removal algorithms, then either your 310 line outs are faulty, or, and I suspect given their nature this might be more likely given that line outs in good order have a surprisingly high SNR****: did you sample via the headphone output? Those have a spectacularly poor SNR that one would have to run through lots of filters to in order clean up.
* I appreciate this might be seen as pedantry, as they are the same thing in a different box, but you are marketing a product for sale so you need to avoid being disingenuous or misleading: I do note your website photo clearly shows a 310 Theatre, not a 310 Unique, yet your product text and this thread state consistently you sampled a 310U. Is the photo not, then, the instrument you sampled?
** as opposed to standard HP/low-high shelf/notch filtering/EQ that could be applied from a mixer or insert effect at song mixing level
*** and much of that was in mono. When using the phaser, the 310U stereo out could only be connected to it in mono; it was the phaser output then that was made stereo via delay.
**** even after 45ish years. I doubt mine has ever seen an engineer, and dog knows I tried to find and pay one to give it a health check.