OK,
Here we go.
Here is a roundup of most of the Tape emulations I have access to.

Cakewalk FX Tape Sim - Slate Digital Virtual Tape Machines - Waves Kramer Master Tape - Toneboosters Reelbus - Tesla Pro and Ozone 4 thrown in for good measure.

Updated video exploring aliasing using 11.7 khz sine wave


Remember I am not really sure about everything going on with what I am seeing it is just an experiment to help uncover what might be going on behind the scenes of these fascinating plugins.


Enjoy!

Here's the cassette emulator TapeHell2. What's up with that enormous spike at 8KHz?


I just like the idea of a cassette simulator so much that I'm still going to try this one out, despite the aliasing.

Wow, the Kramer really tosses bogus frequencies all over the place, doesn't it? No wonder so many people dislike it. 

I noticed some aliasing when you showed Ozone 4, which initially surprised me because I hadn't seen that here. On closer inspection I realized that you were pushing it all the way up and I'd only gone to about 60%. I arrived at that level as I tried to match the harmonic levels in my baseline plugin, which was ReelBus. So it would appear that even Ozone can alias if you try hard enough.

I've been just sitting here listening to some trad jazz while A/Bing all these tape sims, trying to step away from measurements and allow a more subjective evaluation. I'm enjoying the sound of ReelBus best, but TesslaPro is right up there (and I have to keep reminding myself that it's a frickin' freebie).

BTW, the cassette simulator was a bust. 

bitflipper

........ I'm enjoying the sound of ReelBus best, but TesslaPro is right up there (and I have to keep reminding myself that it's a frickin' freebie).

Thanks bitflipper, Rain, Simeon and others for a most delightful and educational thread!  


Bitflipper:  I was a little surprised to see you say you are enjoying the sound of ReelBus best as I remember that in another thread [http://forum.cakewalk.com/tm.aspx?m=2616606] you said: Last night I popped Reelbus into several projects just to see where it might fit. One of the old projects I brought up already had a tape sim on the master, the ancient FX:TapeSim that's been bundled with SONAR since Day One. I was therefore able to A/B it against Reelbus and hear how superior the new TB product was - except that it wasn't. The old Cakewalk plugin sounded better.  I realize that was several weeks ago, and that you have been experimenting with it some more.  Does this current post mean that you are finding ReelBus to be promising or useful?  I happen to like it for the price and for its sound, but I have aging ears which means not always hearing all that is or is not there, and I value what you and others here have to say about these plugins.  

Thanks again to all of you for a great thread!!!!

Rain: There was a very long thread where somebody had discovered that *surprise!* limiters can cause aliasing, followed by page after page of pictures showing one limiter after another and wild speculation about how this might explain the popularity of limiter A over limiter B. What nobody seemed to notice was that the aliased components were at -93db or lower. It was a silly and pointless exercise. 

The screenshots in this thread, however, do not illustrate inaudible or even subtle artifacts. When you introduce spurious inharmonic frequencies a mere 30db below the main signal, well, you can hear that, and it's not going to sound good. Perhaps more significant, it certainly won't sound like tape, which is the whole point of this class of plugin.


cecilius: it's been a few weeks since I bought and first started experimenting with ReelBus, and at the moment I have to admit having mixed feelings about it. When it fits, it works well. But I'm still struggling with where and when a tape sim fits! 

Truth is, I've never been into tape sims and never understood the fascination with them. I worked with tape for a long time (a lot longer than I've worked with digital) and I really don't miss the limitations and quirks of tape machines. I actually like the sound of digital audio, which can be as clean or dirty as you want it to be, and it only demands that you follow a few simple rules to stay out of trouble.

But I'd been feeling like I must be missing out. It seemed as though everybody else was using tape sims. There were long debates on kvr about which one's best. High-end vendors were getting hundreds of dollars for them. So when Jereon's ReelBus came along to rave reviews - and for cheap - I figured I'd see what the fuss was about.

Here's what my experience has been. The tape sim was either too subtle to bother with, or it sucked the microdynamics out of anything percussive. It raised average RMS but in an unpredictable nonlinear manner, while lowering the crest factor and muffling the "sparkle" in cymbals and acoustic guitars. It did help glue the mix together, but I felt sticky like a kid after a crafts project.

Bottom line: I think ReelBus does what it's meant to do, and is a bargain compared with comparable products. I'm just not sure I need it.

Hi Bit
Have to say I'm with you on this - The only Tape Sims I have are Magnetic (which came in a bundle) and Reelbus and I only purchased Reelbus a few weeks ago.
I have only found a few things I like them on - just used Reelbus on a drum bus and it did sound pretty good. I've hardly used Magnetic - once or twice when I was going for an effect as opposed to using everywhere to try and improve a mix, if you see what I mean.
Nigel

Regarding spectra, and aliasing vs. IMD. If you read my initial post to test for aliasing, I suggested to use a sweep tone. That makes it obvious to see aliasing in a spectrum. As the tone rises in pitch, the aliased components will lower in pitch. The IMD as a result of the (modeled) bias frequency won't. (I think... someone check me on that.)

...I suggested to use a sweep tone. That makes it obvious to see aliasing in a spectrum.

Really hard to take a screenshot of it, though.


Emulating bias oscillator intermodulation distortion seems a little extreme. It got me thinking about other aspects that could be simulated in the name of realism:
- require waiting 30 seconds between takes to rewind and position the "tape"
- have the virtual azimuth gradually go out of alignment after X hours of operation
- simulate the tape stretching every time you rewind
- gradually lose high frequencies as the oxide sheds, then supply a virtual Q-tip to clean the virtual capstan and rollers
- have adjacent tracks bleed into one another if you haven't opened the project in a while
- occasionally overwrite or erase tracks
- and the ultimate in tape-sim realism: random breaks (not sure how you'd simulate the tape piling up on the floor, though)