Methods to de-harsh old Reason instruments?

[xalama qo]

I have the now ancient Reason 4, which is still quite a nice source of various sounds. It unfortunately suffers from that typical harshness in the high frequencies, and is almost like fingernails on a blackboard when pushing the resonance to higher levels.

Is there any way to smooth/"lubricate" the sound somewhat? I obviously don't expect magic (Reason 4 will never sound luscious like Dune 3 for example), but I've been wondering if oeksound Soothe, voxengo TEOTE, izotope RX or similar processor would work well in this case?
I know I mentioned the resonance on the filters as an example, which is quite different to acoustic room resonance or boxy acoustic guitar resonance because synth/sampler res frequencies change as they are modulated... and I don't want to confuse my question as being about resonance specifically... because "turn the res down".
I just want to butter up those harsh overtones. Any ideas are welcome, and obviously someone is going to say just EQ it!
Pfft.

[chk071]

I just want to butter up those harsh overtones. Any ideas are welcome, and obviously someone is going to say just EQ it!

Actually, I just wanted to propose that. You already said it yourself: The high frequencies. And what better way to tame those than doing it with an EQ.

What exactly do you mean though? Subtractor, Thor?

What I always noticed about Subtractor: When you sweep the filter slowly with some resonance above 50%, it actually sounds quite nice. Nonetheless, it's a pretty digital sounding synth, of course, as it hails back from times where analog modelling wasn't up to par with nowadays software tech.

What I don't like about Subtractor is that it's so limited. No unison, not a lot of waveforms, no wavetables, not a lot of modulation options. And, the modulation resolution is pretty low. Try to modulate stuff with a square wave on the LFO. It has that typical sloppy sound to it, as if it wasn't a real square wave, but, rather something more rounded. It's also a problem in Waldorf Largo, for example, but, it's much severe in Subtractor.

[xalama qo]

Hey chk071, it's pretty much the whole legacy Reason sound. ReDrum and DrRex too. I fully appreciate it's old DSP, but for lack of a better term the filters (and probably EQs too) feel like blunt instruments compared to what we're used to today. But like I said I didn't intend this to be a discussion about filters per se, it's about the overall sonic character which can sound unpleasantly cheap.
In the past I've kept Reason useful by lo-fi-ing even further. I have a Combinator preset master chain that applies a lo-fi 80s character with some Scream saturation, comp and EQ, but it's pretty heavy handed and I don't want everything to sound like that ...I mean I don't want to find Reason useful for only THAT lo-fi 80s synthwave vibe.

I'm looking more for a post treatment, a type of mastering process, that may be able to free Reason (legacy) from it's own sonic character. If at all possible
I'm thinking along the lines of increasing sample-rate to 96k, using gentle high quality saturation to soften the high end overall, etc.
More of a tonal textural massage than straight EQ surgery

[BertKoor]

I'm suspecting you might be hearing aliasing. You can prove it is aliasing by raising the pitch of a sample and listen whether you hear harmonics diving down instead.

The cure then might be to run your project at a higher sampling rate (can Reason do 192 kHz?) and convert the sampling rate with a better converter.

[xalama qo]

I was wondering about 192k also, I'd have to dig out the old FW410 for that.
It could be aliasing that I'm hearing. I know what aliasing high pitch notes sound like on older digital synths, that typical FM type sound that folds back in. I've never explored aliasing on a whole DAW before, or even paid too much attention to aliasing in sample playback.
If it is just aliasing then running at higher sample-rate should be noticeably better. Is it possible to oversample or emulate oversampling with ASIO4ALL?
Sorry for all the questions, I'd normally just check stuff out for myself but I'm away from studio computer atm. Urgh!

[Ah_Dziz]

Asio4all will only run what your actual audio device can run as far as sample rates. I'm not sure if reason runs as a plugin, but if it does than you can buy a sub host that will allow you to run the plugin oversampled and it will down sample to your regular project sample rate at the output. You can also try exporting the project at the highest available sample rate and then using an offline editor to down sample the results into a normal rate.

I'd say break out whatever interface you have so you can run the sample rate up to the max that reason supports. Even if you have to resort to using rewire, I would recommend sending all your reason tracks into a better DAW and try cleaning up the highs as much as you can. You are only going to make it so far when you're using very old DSP when it was just a matter of having nice feature sets and people spent much less time trying to avoid aliasing.

In real life though, I would just play with the stuff that still sounds good enough for you and replace other sounds with newer and better synths. I never really thought that the aliasing in reason was too bad myself. Im sure you could produce a whole album in there and nobody would notice the difference.

[astralprojection]

Reason 4. Ah, those were the days. Actually i never used 4, but 1,3 and 5. I am not sure, is Mclass suite included in 4 or 5? I think 5 right? Thats a shame. If i were you id run the project at 88k (iirc, thats the max reason had) and then just put a manual "AA filter/ low pass filter" using the MClass EQ right before the main master output. say 18k or something. but if you dont have that in 4, then ..
that should at least in theory, reduce alot of potential aliasing. however i doubt it will do anything to the synths; like subtractor. they tend to alias internally so regardless what you do at the output, the signal would still be aliased when it gets there. i think youre just stuck with that classic vintage reason sound. which honestly, i never really liked that much. just a bit of the worst of digital and the worst of analog all rolled into one. Thats not to say i didnt love using reason back then.. and still do. but im very happy the sound got upgraded along the way.

as far as post processing goes, i think your guess would be as good as mine. eq to tame the highs, really. probably shrp cuts with narrow q along the 1, 2, 4, 8, 16k 'harmonic'

[xalama qo]

Thanks for the responses and general advice guys. I like the idea of harmonic EQ cuts, will definitely try that. And thanks for the ASIO4ALL insights. I was dreading having to reinstall the M-Audio FW410 just for 192k, fussy unit! But my other cards support 96k, so if Reason can run at 88k or 96k I'm sure I'll hear if there's any improvement.

[chk071]

i think youre just stuck with that classic vintage reason sound. which honestly, i never really liked that much. just a bit of the worst of digital and the worst of analog all rolled into one.

Frankly, that's why I rather got into the VST world. I never was too fond of Reason's internal device's sound. There already was better stuff in VST 15 years ago.

[Ikaz7]

Bounce the instrument tracks to audio at highest sample rate possible and then use TDR Nova (free edition is sufficient) to tame the most prominent resonant peaks in the mids and highs using static and dynamically reacting narrow bell filters with fast attack.

Other options to try:

Toneboosters Sibilance v3 (set to freeware status recently) or the fully functional v4 demo

Vastaus by Wavegrove (says "high-end enhancer", but it's actually more of an de-harsher)

Tokyo Dawn OD DeEdger (has a fully functional - except saving plugin state - demo version that can be used to process and bounce your material)

[xalama qo]

Thanks for those suggestions, I'd forgotten about TDR De-Edger. I'll check out the demo.
This is going to be an interesting exercise, remastering old Reason projects and seeing how much I can improve them.

[tomtom1]

Thanks for those suggestions, I'd forgotten about TDR De-Edger. I'll check out the demo.
This is going to be an interesting exercise, remastering old Reason projects and seeing how much I can improve them.

i use some amount of the RV7000 reverb on every old synth (dirctly after the old device) - it makes the sound softer. ofc it doesn't sound like a up-to-date synth... but not that bad either. also there is a preset "warm and wide" in the factory bank - it is a really nice EQ-curve (just deactivate the stereo imager in that combinator) also i use reason 11, it is a really old preset - it should be there.

and what i found out the last days: the old devices sound better if they run at a higher sample rate - at least i think i can hear a clear difference between 44.1. and 96 - which sounds much warmer (especially with some Subtractor patches)

i attached two tracks. first one (flam) is Subtractor only (except the drums), second one is Subtractor (mainly), some Malstrom and PX7. of course that doesn't sound like diva. but i think (even if you probably don't like my style) - that the old synthesizers still sound good.

but maybe my ears are not that good anymore

https://soundcloud.com/treib-haus-818605447/flam
https://soundcloud.com/treib-haus-818605447/cis

[xalama qo]

Thanks for the suggestions! I'll have a look for that preset.

[tomtom1]

Thanks for the suggestions! I'll have a look for that preset.

Btw. did you test some of the things others mention above? (tdr deegder, sibilance etc.)
Did they improve the sound? I am most of the time using only Reason, and not many ppl use it (at least i have the feeling). So i am just curious how to improve the sounds of it (especially of the old devices).

[xalama qo]

I haven't been able to do any serious testing yet, it's not top priority, just on my "things to get to" list. The suggestions here have been very good though. I suspect that the most effective single thing would be to run at 96k to avoid cheap anti-aliasing filtering at lower sample-rates. I think one of the big clues there is that Europa includes Thor's wavetables, but they sound smoother/better in Europa, probably due to higher internal sample-rate within Europa itself.
When I listened back to the specific tracks I'd like to polish up most of the stuff from Reason is sampled sounds, so I'll be happy if I can just re-render those elements at higher sample-rate and just get a little more air/life in the top end. It's nothing hugely critical, I just want to how far you can push Reason to sound better before hitting a brick wall, and turning to another instrument.