having a cool beer with my studio partner (who works on Cubase since 20 years) at the moment, we are completely thrilled (not: drunk, yet!) by an almost magical phenomenon that probably everyone of us knows, but noone dared to talk about yet:
The recording of the space bar sound!
I admit this sounds like as it would better be put into the media lounge, but after hours of thinking, we came to the conclusion that Nuendo/Cubase are time machines, allowing us to travel forward in time. And the invention of a time machine clearly needs a prominent place to be posted.
Place a mic in your monitoring room (i.e. the room where you have your PC keyboard)
Record something nice (well, not mandatory)
Stop recording by hitting the space bar.
Listen to the previously recorded track, can you hear the space bar being hit at the end ?
Probably yes. And here is where the magic starts: Hitting the space bar ends the recording. In the moment you hit it, no more audio is recorded. But you actually CAN hear it. It has always been like this, since 1997 at least (where we found the first space bar sounds in our archives).
This is because Steinberg software let’s you travel in time! Unfortunately, it’s only about 20-100 ms, so it is not yet able to predict the results from the lottery.
Joke aside, at least all bedroom producers must be annoyed by that fact, not? Always cutting the last few ms off the audio?
Because some plugins introduce latency, delay compensation “shifts” some audio earlier in time .
I.e. these plugins get their audio earlier than other plugins, so when the latency is added (when audio has passed through the plugin), it is back in sync with everything else.
However, moving back in time is impossible.
So what a DAW does is moving everything later, except for that one track/plugin.
Which means that when you start playback, the actual playback starts later.
Hit the spacebar:
-The DAW “waits” to play (pauzes), but delayed tracks start to play in the background. Project is visually not running.
-Once the delay has passed, the project starts to play.
===>So when your project is @ 1 minute, the project has actually played 1 minute and some miliseconds.
Now you can figure out why you hear the spacebar at the end of the project.
I like that “feature” a lot. Because I recieve bedroom-recorded stuff for pimping the sh*t out of it on a regular base - which often includes truckloads of editing, even when the Band only wants “a mix” - in that case I usually not need to listen through all the takes, I only listen to the “clack” at the end… The way the “producer” hit the space bar you can usually judge about the take itself… I mean… if the “clack” is coming to early and to hard - it was not a good take. If the clack is sounding satisfied and with great vibe - this is usually a great take.
Unfortunately they usually do not use Blumlein setups for recording the space bar…
I agree with Chewy. It is only logical. Based on the construction of the physical spacebar key that noise may be generated by the key before the actual underlying switch makes contact. Also, it is also expected that some minimal delay is introduced by the keyboard encoder and the serial (in case of PS/2) or USB (for more modern) interface before the event is captured and made available to the application (i.e. Nuendo).
I would say that, too. Was my first thought when reading the post - but that means that the recording/the file will end exactly in the middle of the “clack”. It will be present - even without post-roll-recording. So Matthias needs to do some test-recordings, recording all kinds of spacebars (ok, one is enough) and with all kinds of buffer size (latency) - if there is a gap behind the clack - then we have post-roll-recording. If the gab differs depending on the buffer sice (maybe with loaded plugins as well - for example latency hogs like UAD Precission Multiband or something like that) - then we have an explanation.
Well, in fact I do not mind when there is a post-roll recording. Why not. A somewhat more nasty thing was in an older Nuendo Version that (in a multitrack situation) the recorded files were not from identically length. Often it was possible to enlarge the recorded events a little on the end. Not only a couple of samples, more something like 500 ms or so. That caused files from different size, which I personally disliked. Recording 20 Tracks, hitting stop - having Files which are differing about at least 10 kb.