Ability to set Virus to generate 48k internally when VC running if Analog output is selected. Pleeeease?

  • As the title says. I believe the virus has less audible aliasing at 48k than at 44. The analog outputs are resampled anyways to 192khz so why isn't the virus automatically generating at its highest sample rate to feed that output?
    Is it possible to alter the plugin so that you don't lose the option of setting the Virus to 48k when VC is running? Or at very least (this is probably simpler) can we first have it so that you can select 48k when using analog outs while the virus is connected via usb but VC not running?
    Any sort of workaround that would allow the virus to generate 48k internally while still being able to receive and send midi data. I doubt I'm the only one who uses the Ti as my main midi hub.
    Less aliasing & better sound is a good thing in these days of vst synths that oversample beautifully. I love the sound of synth squad's strobe when running oversampling. Some of the interface really bugs me though!

  • I'm well aware of the menu contents my man ;) The issue is that when you plug a usb into the virus you can no longer choose what clock rate you want it generating at. I found out how to get true 48k synthesis internally from it with usb connected though. I spoke with Jorg as well as ran tests at all freqs with usb vs analog out and will post the results soon. With a spec analyzer you can clearly see the frequency response difference and the difference in where the filtering happens in each.

  • it would be great to hear more about how you are setting the virus to stay at 48k with USB connections. I'm finding with OSX Lion and Logic I can't run in 48k mode well, it only runs well in 44.1. I'd love to test ways to try to keep it in 48.

  • when you plug the usb in it goes into autopilot mode. You no longer can choose the clock rate. If you have usb plugged in you have to change your audio interface to either 48k or 96k to get 48khz internally. 88.2khz runs 44k internally.

  • I work at 88.2k.. which has a little less strain on my resources than 96k. So the only acceptible options I have currently that'll get the Virus generating at 48k are to unplug the usb and run the virus midi or else work at 96k and hammer my resources that much harder.. which I'm a bit leery about since 88.2k already maxes me out occasionally. I've thought about it though. Instead I just ordered a cakewalk UM-3G midi interface to try and when it comes time to record I'll have to just unplug from USB and use it standalone. I'm interested to do a midi loopback test and record the output anyways to see if the Virus can be bested as a midi interface.

  • the UM3G is a great MIDI interface, I get on with it really well...


    Surely you can just work at 48kHz, then? Especially for recording the Virus (since you won't get more than that out of it, anyway) - I don't want to argue about the whole sample rate issue, but if your system is struggling, the argument to keep using 88.2/96 is even weaker...


    Other than doppler shifting effects, I have come across nobody who can explain an effect that would benefit from a higher than 48kHz sample rate (aside from plugins that aren't designed to work well at 48k, which is silly).

  • "Aside from plugins..."
    So you can't find anyone to explain to you the benefits aside from the main reason for it? haha. That's not a note to put in brackets as that's the entire point of it. We live in a world where projects are stacked with digital operations and that's where the difference lies. If you were recording all in analog then there would be nothing to gain there past the potential of slightly beter AD encoding from your converter.. which actually a lot of converters do see btw. I get 6db higher noise floor with mine for instance. It's not silly at all since there are limitations involved with 44 & 48k which can only be surpassed by operating at a higher rate or using oversampling.


    I won't ever work at 48k. My system is peachy at 88.2k as I've built a destroyer of a PC to handle it, but going to 96k has no benefit to me and it's preferable to stick with even multiples of your 44k final format as some processes will benefit and handle the conversion better. I can still max it out by working with unnecessarily bloated projects but I'd trim down the fat before moving anywhere from 88.2k.
    Most of the best in the industry work at higher rates in the digital environment for a reason. If there were no gains to be had past 48k then you wouldn't find some of the best plugins for instance offering 4x oversampling. It's better to work at a higher rate though than to rely on plugins sampling your audio up and down multiple times through a project. You also get lower latency which is very nice.

  • You can reduce the buffer to the corresponding number of samples for any given time period - latency has nothing to do with it.


    The fact is bigger numbers sell better (and oversampling is an entirely different issue). Why do you think they put 8/12MP sensors in mobile phones, despite the fact that there is no way you can get enough light in their tiny lenses to get that kind of spatial resolution? BUT if you are given the choice between two products you cannot differentiate, you're going to pick the one that says it's got more megapixels or more bits resolution or higher sample rate or whatever.


    It's also interesting that you get a higher noise floor at a lower sample rate - I wonder what ADCs/DACs your system is using and how well they are managing the sample clocks around the system at those sample rates. If you use a half decent Delta-Sigma CODEC then it should be oversampling well into the MHz band spreading the noise power all the way across it meaning post-decimation you should get very little noise...


    The noise performance of any of my sound cards is not affected by sample rate - you should get that checked (update drivers or something?)...

  • AtonyB, I have the feeling you may be missing nms' point. :P
    A lot of VST plugins offer an oversampling option, because it's an easy way to reduce the imperfections of digital processing. It's only a mathematical issue, we're not even talking about ADC/DAC here.
    Although I don't work at 96khz, I'm not suprised to hear that some plugins sound better at this rate. It's a way to force all the plugins to do oversampling, even the poorly designated ones.
    And if you only work with greatly designated plugins who do oversampling, it's still a way to avoid having each of them downsampling its output when the next one in the chain is going to upsample it again...

  • Ace has it exactly.
    AntonyB - Do you really think that so many of the top guys use higher rates because they're simply caught up in sales hype though? That's silly as we're talking about the best in the business. I don't buy into going above 96k though personally which is why I recently picked up a Mytek 96 ADC rather than spent more for a 192k converter.


    Here's what I'm talking about regarding slight converter improvements in 44 vs 88k. I get the same differences whether going MR816 DA into Mytek AD, or MR816x DA/AD. Taken from Right Mark Audio Analyzer via DA/AD loopback test:


    Noise level: -97.5db @44k, -100.6db @88k (3.1db difference)
    Dynamic Range: 101.4 @44k, 103.6db @88k (2.4db difference)


    To me none of that would be any deciding factor though. The deciding factor and most audible difference comes from the stacked effect of all your plugins and digital synthesis. There's no easier way to get a better sound from them as a whole than going up to 88.2k. The leading software developers and high end gear manufacturers most always recommend production at 88/96k. But anyways this thread isn't meant to be a debate about the topic, rather a small feature request.

  • nms: What I do to combat this lo-fi reality is I use direct outs on the Virus TI and input them into my interface at a higher resolution. This solves two problems at once:


    1. The ability to get the raw undithered hi-res Virus TI's 192 kHz outputs.
    2. The ability to resample the higher outputs into a lower resolution without aliasing your audio.


    In other words, until this baby can pump out 192 kHz through USB, this is the best way to "hear" the Virus TI.

  • It's a way to force all the plugins to do oversampling, even the poorly designated ones.


    You can't force a plugin to do anything it's no programmed to do - if you have, say, a delay effect, all increasing the sample rate will do is make it take up more memory in sample buffers (OR have a reduced maximum delay) - It might make a phaser/flanger/chorus effect's life easier by reducing the amount of interpolation it has to do, but since you are only doubling the sample rate you haven't helped by much



    Zitat

    AntonyB - Do you really think that so many of the top guys use higher rates because they're simply caught up in sales hype though?


    Yes. Not all sound engineers are DSP experts. Also, bigger numbers sell better.




    If your plugin doesn't sound great at 44.1kHz then you are doing something wrong. The Virus is the best digital synthesizer i've ever heard whether at 44.1 or 48, so Access manage just fine with their internal upsampling and subsequent downsampling.




    All that aside - this is nonsense, the Virus will only generate sample rates of either 44.1 or 48. If your project works at 88.2 then you will be better served (with your desire to pursue inperceptible differences) by upsampling 44.1 by a factor of 2 rather than 48 up by a factor of 1.8375 - clearly the integer scalar will give you better results.

  • It might make a phaser/flanger/chorus effect's life easier by reducing the amount of interpolation it has to do, but since you are only doubling the sample rate you haven't helped by much

    Agreed. x2 isn't a big oversampling factor.


    If your plugin doesn't sound great at 44.1kHz then you are doing something wrong. The Virus is the best digital synthesizer i've ever heard whether at 44.1 or 48, so Access manage just fine with their internal upsampling and subsequent downsampling.

    I've no doubt about that :) my Virus sounds great at both sampling rates.



    If your project works at 88.2 then you will be better served (with your desire to pursue inperceptible differences) by upsampling 44.1 by a factor of 2 rather than 48 up by a factor of 1.8375 - clearly the integer scalar will give you better results.

    Seems logical to me, but I wouldn't talk about imperceptible differences, though. They can clearly be heard in the file of this topic : http://www.virus.info/forum/in…page=Thread&threadID=3031
    What remains to be tested is if this difference will still be audible after a SRC to 88.2k.

  • No offense Antonyb, but you are way over your head here with this discussion.

    Zitat von AntonyB

    You can't force a plugin to do anything it's no programmed to do. all increasing the sample rate will do is make it take up more memory in sample buffers

    Oversampling is when a source is upsampled by a given number of multiples of itself. 88.2 yields the equivalent benefits of oversampling every plugin in your project x2 except without the downsampling back to 44 that would happen every time. Any plugins that oversample will have at least x4 oversampling. Your theory that it's all a big hoax by software developers and that the best engineers in our industry are being fooled by it all when really there's no difference is quite a stretch and just plain wrong. That's coming from someone who's tested it extensively and directly discussed it with a few of the best developers in the industry. Are you not familliar with aliasing or brick wall antialias filters at all?


    If your plugin doesn't sound great at 44.1kHz then you are doing something wrong. The Virus is the best digital synthesizer i've ever heard whether at 44.1 or 48, so Access manage just fine with their internal upsampling and subsequent downsampling.All that aside - this is nonsense, the Virus will only generate sample rates of either 44.1 or 48. If your project works at 88.2 then you will be better served (with your desire to pursue inperceptible differences) by upsampling 44.1 by a factor of 2 rather than 48 up by a factor of 1.8375 - clearly the integer scalar will give you better results.

    Is that right... so what do you have to say to this: Virus-Saw-44k-vs-48k-source-downsampled.wav
    That's the sound of a 147hz saw from the Virus.. it's what I like to call an idiot proof test since it exposes the differences in a clearly audible way. The first half of the loop is generated at 44k, second half is 48k downsampled to 44. I think you'll find those "imperceptible differences" audible even on crappy laptop speakers.


    Zitat von Ace17

    Agreed. x2 isn't a big oversampling factor.

    Actually it is by far the most relevant and audible oversampling factor. From there it's majorly diminishing returns.
    As for SRC, with a top quality SRC like Izotope's 64bit SRC there's nothing to worry about from the SRC upsampling from either 44/48k to 88/96k.


    nms: What I do to combat this lo-fi reality is I use direct outs on the Virus TI and input them into my interface at a higher resolution. This solves two problems at once:
    1. The ability to get the raw undithered hi-res Virus TI's 192 kHz outputs.
    2. The ability to resample the higher outputs into a lower resolution without aliasing your audio.
    In other words, until this baby can pump out 192 kHz through USB, this is the best way to "hear" the Virus TI.

    Despite previous theories, the cleanest signal you can possibly get from the virus is with your project rate set to 48khz and running via usb. The audio that goes out the virus DA is taken from that same source. The "192khz DA converter" on the way out is essentially just for marketing. It's not producing a better sound than what comes out the usb. The converter chip may be capable of processing that high quality of a source, but the source has already been created at 48khz. If you hear a difference between the analog and digital outputs it's because you're fooling yourself or listening to the degradation of the DA to AD conversion. Maybe you like the coloration from your interface, or maybe you are listening to the specs in your head about the 192khz DA.. but you won't get any clearer signal than the digital source piped through the usb direclty and the Access team will say the same. If you are running your project at 44khz and the usb is plugged in, the source that feeds the analog output will only be 44k unfortunately. The one way the analog outputs can lead the pack is if you don't have USB plugged in and run it via midi, allowing you to choose 48k synthesis internally which then goes to the analog outs and gets streamed into your interface at whatever rate your project is running at. You're still sending it through a full DA/AD cycle though unlike if you'd ran it usb. This is why I've decided to just switch up to 48khz in my project when it's time to track something and use the usb, then switch back to 88k and resume use. It takes me about 15 seconds to open that take in soundforge and resample to 88k using the Izoptope SRC. That way I get the cleanest source possibile tracked to audio. I do like to process with my outboard hardware.. but doing it this way allows me to route it out for outboard processing and try whatever I like, while meanwhile I've already captured the take I wanted from the virus perfectly.

  • Interesting - it seems that the antialiasing filter on the virus doesnt perform so well as there is a distinct drop off on 44.1 from 16kHz even (bizzare or possibly deliberate, who knows) and the first lobe is quite prominant - much more so than with 48kHz (from my own measurements).


    I'll have to admit, I've always run the virus in 48k as that is the default sample rate for my ASIO driver, so I haven't noticed that difference, albeit a specific issue that can be remedied by an EQ and would be buried in any mix.


    Aside from the quirks of the particular piece of equipment being used - there's no reason why you can't run at 44.1kHz as there should be no audible difference (ie identical up to 18kHz since, sampling aside, getting the analogue equipment to be reliable up there is hairy enough and nobody cares at any higher than that even if they can hear it). I will concede that running at 48kHz does give you a little more room for the aliasing filter to go from unity (ideally at 18kHz, which is the case for the codec I use when run at 44.1kHz) to -100 or so but it is not crucial.


    Above 48kHz, however, besides internal upsampling for plugins, gives you nothing extra in the frequency ranges that anyone can hear (and are probably best removed for improving dynamic range, although they will be so low amplitude that you needn't worry anyway).


    Oversampling is only of benefit if the plugin makes use of it, such as for sample interpolation (but you don't need to upsample for that, you can generate the inerpolated sample(s) on the fly if you wish). Speaking of leading industry experts - I will point out that I have heard such people being extremely pleased with themselves for 'modelling capacitors (or condensers as they were saying)' in their emulation of hardware and using prewarp - things which are the bread and butter of DSP...


    Incidentally, I would point out that I am actually doing a PhD in DSP so I do actually know what I'm talking about, rather than simply regurgitating misapprehended concepts to order.

  • excellent, if you're doing a PhD on this then you'll find this all the more useful.


    That sample I posted is particularly relevant because it's not a high frequency sound.. it's a 147hz sound with a lot of harmonics yet listen to how much it's affected when you isolate and compare it in a way that makes the difference obvious.


    I used a 2 pole filter fully open for that. If you look in a high res spectrum analyzer you can see the differences extend down pretty far and not just 1 or 2k. You should know better than to assume an eq can easily reverse that. You can't put back what's been filtered out, only artifically boost what remains. If you're talking about some background pads it's probably not going to be an issue. If you're talking about synth leads and elements that are in the forefront then there is most definitely capability to produce better, sharper, clearer, more detailed sounds. Do you like the sound of the analog 4 pole filter but hate how dark it makes a lot of stuff sound? Try it at 48khz.


    You have to understand that while 48k is an improvement over 44, it still only raises the available frequency ceiling 2.5khz higher than the usual 20k limit and is certainly not the be all end all sample rate where digital processing and synthesis are concerned.


    Working at 88/96 isn't about capturing frequencies above 20k.. not in the least. It's about having enough frequency headroom to push the aliasing and filters WAY above the range our music exists in so it's not messing with it.
    Do you want to listen to the music? Or do you want to listen to the music through a bunch of stacked filters you have no control over and digital garbage that's been folded back into the audible range?


    It's only been a year since I switched to 88.2k from working exclusively at 44. In the past year and a half I have researched, tested, studied, and debated the topic to death. I also learned a lot discussing it with Andy Simper (owner of Cytomic & developer of 2 of the best plugins released - The Glue compressor & FXPansion's Dcam synth squad) as well as Alexey Lukin (of Izotope fame and developer of the Izotope 64bit SRC).. and have also bugged Marc & Jorg from Access a lot about the different aspects relating to the Virus. The things I've said on the topic are a result of what I have learned first hand. Chalk it up to my perfectionist nature and crusade towards producing the best sounding music I am able to. Rest assured, I don't just buy into theory or marketing hype here.. every piece of my rig from hardware to software gets rubber glove tested from top to bottom.

  • nms, for me it is my converters. I think that the Virus outputs sound way cleaner then the USB outputs. That is why I choose to record my outputs into my interface instead of just recording through USB.


    On that same subject... I am using a Presonus Firebox right now, but I am considering an Apogee Duet or a RME Fireface 400. Since you seem to be on the up and up with converter... Do you think the difference I will hear will be slightly better or extremely better if I upgrade?

  • It doesn't sound as if you read much of what I said.


    I said that increasing the sample rate can relax the demand on the interpolation filter - perhaps to make it cheaper, but I would have thought the cost of extra bandwidth, etc. would offset that. BUT on an 'off-the-shelf' Texas Instruments (/Burr brown) CODEC running at 44.1kHz the interpolation filter is unity gain until around just shy of 20kHz.


    48kHz goes even further (unity until just shy of 21.5kHz). This means that no higher sample rate can represent frequencies up to that point any better, and those above are unimportant as no-one can hear them assuming the audio equipment even lets them through.


    I did notice that there was some frequency shifting going on between the sample rates where the peaks didn't seem to align that well, but this might just be due to different spectral leakage due to the differing sample rates - you get an artefact where the pitch wanders off when you change sample rate then comes back in so its hard to tell if it lines up once it stabilises by ear (and I'm not going into the lab to find out) - this is made more difficult by the fact that the oscillators (deliberately or not) arent consistent cycle to cycle anyway and the differing sample rates may alter the performance of that also. Non-integer small shifts in sample rate can be the biggest pain in the arse when trying to get cross sample rate consistency (I haven't checked, so I don't know if the virus is fixed or floating point so I couldn't begin to guess why, either, to suggest how you would rectify it).


    Either way, I'm not surprised, if you go into the minutest detail, that you see a difference, because they are different, the sample times are different, the banwidth is different and, yes, tinkering with the virus at 44.1 you do notice a little less sharpness at the top end which really surprises me..


    I don't know what codec they use in the Virus, but thats irrelevant to the audio you get via USB as that won't go anywhere near a codec so the interpolation filter must be implemented as an FIR or otherwise on the DSP which may explain the effect you see...


    Again, all I've heard arguments for is why 48 is better than 44.1 which is perhaps a deptracated problem, mainly because 48 is pretty standard for everything besides CD now and any codec I'ved used is fine on both for human ears, even on unrealistic tests, but historically may have given you a bit more margin for error. I guess it's also associated with something to do with the virus, and not having seen a line of code for the os on the thing, is all guesswork.


    I'm confused, anyway. When it comes to recording the virus, can you not just change the sample rate temporarily while you do it to 48k or 96k, whatever floats your boat, and then change back when you are done? If you are using it TI mode then it has to be consistent with sample rates, so I'm not sure what they can do for you without implementing some sort of SRC in the VC plugin beyond simply doubling up the samples, which I guess is happening somehwere along the line already for 44.1k...