hardware synths and sample rates

Anything about hardware musical instruments.
Post Reply New Topic
RELATED
PRODUCTS

Post

Hi my friends (love this sub forum, so friendly, thanks :) )

I've started to think, with analogs such as my pulse 2, jd xi (one monophonic true analog voice), and dsi tetras, wouldn't there be a real benefit to recording them at 96K?

Believe it or not, using 96 K even on my 2011 quad i7 imac, is usable, because i barely use and vsti anymore, and 90% of uad fx are already upsampled so use the same DSP at 96k.. (ok in this case, my final medium is always 44.1, so let's say i will choose to work at 88k for this topic). To use a native effect here or there is neither here or there for cpu when all you are processing is fx on audio tracks.

The roland integra and access virus can also do 96k i believe - but surely that would have to halve their polyphony?

And then we have the rest of the range, the old modules like emu, korg, proteus, and of course my other digital synths like mininova, microkorg , nord a1.. Is there ANY point whatsoever in recording them at 96k?

The big advantage here i feel, would be the uad latency would be 1.1 round trip even through 4 uad FX.. that's insanely low to play my hardware synths especially for drums. UA won't publish the 44.1 figures, and i can't find them anywhere after hours of searching. (they even only publish the 96k ones in the manual..).. even if i change interfaces, ALL interfaces are lower latency at 88/96 than at 44/48...

second would be, i could capture a more accurate sonic footprint of the real analogs, no?

but of course, if a project is at 88 or 96, then everything is. All imported loops or samples will be upsampled, as they are all at 44.1, and what possible advantage could it have for digital synths anyway?

I guess an easier way to put this is.. what sample rare do YOU record your hardware synths at?

cheers!

Post

44/16

Post

I'm actually caring about 96k only with vstis.
Sounds to me that while you can treat analog like crap and make it sound bad it's a very different breed of bad than an aliased vsti...
Still I use vsti polis, as size and pricing hurt a lot.

Some help comes from gear like the ada...the benefit of lots of ins overweights need for 96k and embraces 44100 and 48. (aahh, multitracked drm1mkII, drool)
Plus the music I do, that keeps audiophiliac behaviors pretty far :D

Post

ok will stick to 24/44.1

that;s the minimum to submit a mastered for itunes.

Still curious to hear others, thanks guys, and please excuse silly typos in first post.

Post

44/16. My hearing is probably shot above 10-12khz anyway by now so I wouldn't hear any difference, and since I use my macbook to record, I dont want to use more disk space than I have to.

Post

48/24 is good for me. I tried 96 khz, but found even when using mostly hardware synths I'd start having issues when running plug in effects and what not. Didn't seem worth it even for the analogs. Definitely not the digital synths.
Zerocrossing Media

4th Law of Robotics: When turning evil, display a red indicator light. ~[ ●_● ]~

Post

I run VST synths at 96khz - I can hear the difference.

However, recording audio (Mic, Guitar, Analog Synth) I don't hear any meaningful difference. So for recording audio I prefer to use 44.1/48 to save resources.

Post

TheoM wrote:ok will stick to 24/44.1

that;s the minimum to submit a mastered for itunes.
Really? So they take a perfectly good file and compress it to sound crap. Wish they'd offer the 24/44.1 wave file.

Me? I wouldn't bother with 96, 24/44.1 is fine for most things. Forget 88.2 as well, I think the conversion algorithms are more complex that 'divide by 2' so I'd do 96 if I was going to bother at all.

Post

zerocrossing wrote:48/24 is good for me. I tried 96 khz, but found even when using mostly hardware synths I'd start having issues when running plug in effects and what not. Didn't seem worth it even for the analogs. Definitely not the digital synths.
in this case though the UAD will be used for 90% of fx too....

did you do any comparisons with analogs specifically as far as recording quality?

Post

Mr Arkadin wrote:Forget 88.2 as well, I think the conversion algorithms are more complex that 'divide by 2' so I'd do 96 if I was going to bother at all.
you literally have it ass backwards.

Post

Whenever I record analog sources I do that @ 24/96k for pristine preservation of the 70dB SNR and 18kHz bandwidth.
We are the KVR collective. Resistance is futile. You will be assimilated. Image
My MusicCalc is served over https!!

Post

TheoM wrote:I guess an easier way to put this is.. what sample rare do YOU record your hardware synths at?
Hardware synths or anything else at 48 kHz, because that's the converters' optimal sample rate, according to specs and my own listening tests of various converters. With some, 44.1 kHz is pretty much as good so it doesn't really matter. Listen to 44.1 kHz vs 48 kHz with most cheap interfaces like Komplete Audio 6 and you may be horrified by the very audible difference. 88.2 kHz and above is worse with everything, less so with the higher-end stuff, and there's no point using it unless I need to slow things down while keeping high frequencies. Rendering software synths is a different matter and some synths may sound very different at higher sample rates. For example String Studio, huge difference between sample rates, and 192 kHz is significantly different than even 96 kHz with it. Much less of a difference with many newer synths.
"Music is spiritual. The music business is not." - Claudio Monteverdi

Post

I got to admit there's a lot to take in this topic and a lot of different opinions. Hmmm.

I'll tell you guys about the past. In 1997 when i first opened my own studio, i was all hardware synths. Yamaha CS1X, Yamaha AN1X, Korg wavestation Rack with many rom cards, Korg o5r/w module, Novation Bass Station, Novation DrumStation, Emu orbit, Emu Planet Phatt, Roland JV 1010, Roland JV 2080 fully expanded. Overall besides the bass station, a very digital setup but also very nice, a ton of different sounds. When it came out i added a yamaha EX5R rack, a few years later i added a novation supernova in the commercial studio and eventually a virus a.. but that was the last synth i bought (and sold).

Now these were all going into a mackie 32/8 and being recorded mostly one by one as audio tracks into Logic 3 on a pentium 2 300mhz, it was 256 mb ram from memory.. everything sounded fine. All recorded at 44.1 khz. Sometimes we had to go to a dat tape and sometimes we went into a tc finaliser then a dat tape. The only plugins i had were a couple waves direct x plugins, sonic timeworks reverb (i later added compressor x when it came out), and logic's stock plugins. Effects processors included a lexicon multi FX called the MPX1, a lexicon pcm 90 reverb, dbx blue compressor, focusrite red compressor, a behringer composer pro compressor and de esser, focusrite green pre, and of course the pres on the desk. We only used desk EQ! I know right! Talk about a big omission. Had a lovely patchbay set up to patch the effects on whatever channel i wanted.
Mackie HR824's, first ever revision.
I was really happy with mixes overall.And my hearing was much better then so if there was any issues recording stuff at 44.1 i would have heard it (plus i had BOTH my ears). This little home studio racked to the hilt, was purely born to compose music, and was bought due to all the cash i collected on my 21st birthday (and all the cash i had saved since a kid), my grandfather gave all his grand kids rather generous 21st birthday money. In hindsight, if i had invested that 30K of his into apple shares on the exact date he gave it to me, i'd be worth over 12 million bucks now LOL! Oh well.:(

When I partnered with a friend to open a commercial studio in late 1999, we added a bunch more focusrite pres, sold the mackie desk, and went with a mac for the first time (G3 500 ), and a pro tools mix cube system (25K in OZ back then, ouch, including interfaces). All our synths were directly patched in to pro tools, we had this studio desk with racks built in and it was all pretty awesome to tell you the truth. The age of good waves and mcdsp plugins was born, logic had drastically increased the quality of it's own plugins, and we finally had a good EQ plugin (we used the mcdsp filterbank, i still love that plugin to this day). Started dabbling in vocoding (orange vocoder, was a revelation), and all sorts of fun things. Had a band room and vocal booth.. Never added any hardware after that day. We ended up getting the virus TDM plugin too (called indigo) and everything was going nicely. Basically it was the pro tools early millennia version of what the UA does with the apollo now (or avid do with HDX).. with the added benefit that the latency was just as low, but everything was patched in real time through logic's environment, (logic had pro tools 100% dae compatibility then), and therefore, we never had to leave the DAW.. all synths were permanently patched in to pro tools TDM interfaces (we had an adat bridge with tangos feeding it for the synths as well as an 1622).

Now long story i know, but again, everything was recorded at 44.1khz, and 16 bit for a long time too - if my memory serves me correctly, logic didn't even do 24 bit then and i can't even remember if the pro tools interfaces were more than 20 bit anyway.

But yeah, the bass station was the only analog thing we had, and even that used dco's, but it was still analog and sounded great. We never ever recorded it higher than 44.1khz.

But now Shy has confused me saying record hardware synths at 48. Surely since itunes is 44.1 and that's where i am selling my music, that 44.1 makes most sense.. and what does he mean that 88.2 sounds worse? Remember the only soft synths i am using are sample based anwyay (halion/nexus/sampletank/kontakt and to be perfectly honest, i want to get rid of everythign and just keep a good orchestral , ethnic and guitar library.. that's it, that's all i want to use for VI's, so if the VIs are sampled at 44.1, no point going higher anyway, but some kontakt libraries are at 96). If i could get a really good orchestra hardware synth, i'd stop VI's 100%.

edit, interesting:
http://www.soundonsound.com/sos/jun05/a ... 0605_6.htm

so supports my theory although an old article. Also instead of dithering to 16 bit now, itunes is at 24 so there's no need for that. So final output is 24/41

I will try out of curiosity.. recording one project at 96 the downsampling to final output of 44, also will record at 88, and will just record it at 44.. and see what difference i can hear especially in the recorded analog synths.. will use izotope re sampling at highest quality for best results when down sampling.

Post

TheoM wrote:But now Shy has confused me saying record hardware synths at 48. Surely since itunes is 44.1 and that's where i am selling my music, that 44.1 makes most sense..
Ah, I didn't mention that I just downsample the final audio to 44.1 kHz. Why not just record at 44.1 kHz? Because the converter may sound a bit worse, depending on the input, and its low-pass filter may effect the frequency response too much. But with some converters there's just no audible difference, so it just depends. And what about the software sample rate converter's effect on the sound? None, with any decent one.
and what does he mean that 88.2 sounds worse? Remember the only soft synths i am using are sample based anwyay
88.2 kHz is at least a little worse mainly in transient response and low frequency response, with most converters, and it has much more noise than 44.1 / 48 kHz with any converter, but that may not matter at all, especially with synths which have a way higher noise floor than the converter's anyway. In any case, whatever sample rate you use, it's not likely to make an important enough difference to destroy your mix if it's not the optimal one.

With sample-based software instruments that don't have filters or some other effects applied, which have 44.1 or 48 kHz samples, it would of course be best to use the same sample rate as the samples'. With some that make extensive use of filters for example, they may sound better at higher sample rates, because the sample player plugin's algorithms are of the type that could benefit from oversampling, but don't have enough or any oversampling applied, so the higher sample rate can make it sound "smoother" and nicer, but of course that depends on the specific plugin and instrument, so just gotta listen and decide.
"Music is spiritual. The music business is not." - Claudio Monteverdi

Post Reply

Return to “Hardware (Instruments and Effects)”