Work continues on the Life On Mars cover video. I actually had a "complete" version of all the music, and was relatively happy with it. But, in one of many learning points during this project, I gave a listen to the original (which I should have been doing more often) and realized my intro was just too fast of a tempo. I had tried various ways to approach the order in which I laid down tracks. I think drums first is still probably the best way, although in this tune they don't come in for the first minute. After learning that I really couldn't keep track of the drums while trying to play with the original, I tried using a midi version of the song with which I could include a click track and that seems the best approach.
There are multiple ways to approach this sort of incremental song construction. Due to problems with latency, it was important for me to keep the recording setup between the drums and my DAW as simple and uncluttered as possible. Latency is a killer with drums and percussion if you're trying to record in time. I managed to get a version down that was close enough for my satisfaction. I did end up making some minor timing corrections, mainly for the occasional rushed kick drum on the "and" of 4.
Probably the single biggest challenge of this project is that I'm recording to video. Making corrections to timing mistakes and the occasional clam note is easy in a DAW, but trying to get that synced up to video becomes exponentially harder. There are ways around this with creative video editing, but it was important to me that the foundation of the video was going to be good performances on all the parts as the baseline. Although I'm mainly doing this for my own entertainment (and internet points, because, internet points), I have felt this might serve as a sort of audition tape that I could use as needed for bands, further academic endeavors, etc.
My approach has been to keep attempting complete takes until I get one that's close to "perfect" (my version, not Rick Wakeman's). I then stop recording and know that I will use the last takes. In some cases this might be the 10th+ take overall.
Once I looked at the rough draft of the video performances and had worked on the mix (I was on version 14 or so), I realized I just wasn't going to be satisfied with my version if it didn't have the same vibe as the original. I never desired or expected to make a perfect note for note rendition that would sound exactly like the original, but I felt like my version was too rushed and too produced sounding. It was lacking the dynamics and character of the original, and that's kind of the whole point.
I resigned myself to have another go at the intro, as that was the biggest problem and I managed to have a "decent" go at the piano part yesterday. Although it was useable, I was suffering my version of a hangover after drinking four beers the previous day, and I wasn't giving myself the best opportunity for a good take. I also realized that my approach to the green screen needed some work. I've learned a small amount about videography, and one thing I've gathered is that your green screen needs to be as flat and "invisible" as possible and it needs to be lit appropriately.
Looking at my various takes, there was a variety of lighting and screen real estate occupied by my green screen backdrops. I think for my project, the best approach will be one where the green screen completely occupies the background. There may be future projects where I have the green screen as a sort of object in the background that doesn't necessarily fill up the screen.
All that being said, I need to re-shoot the entire thing so that I can have a consistent lighting and size for my green screen. I am learning to never see an individual performance as too precious. This is a growing process as I learn to accept not every take will be perfect, but become willing to keep trying until I get as close as I'm capable. I know when I've done my best, and that's usually when I'm just practicing or playing without the "red light". The good thing is that I've learned to get over red light syndrome because you can always make another attempt. Part of the comfort level comes from really knowing the part well and having confidence that you can play it right.
Tying in to that was finally settling on an arrangement from which I would build the song and giving myself time to practice it. I think I've decided on just playing the intro piano part independently without a rigid timekeeper and then locking in with the drums at the first chorus when they initially come in. I may tweak this as I discovered that the intro was fairly close to 113.5 beats per minute, but then it increases to 128 at the first chorus. It sort of slows down again at the end of the guitar interlude, but returns to that during the rest of the song. My dissatisfaction with the piano performance largely stemmed from timing and feel issues related to trying to play to a metronome/backing track when I had been accustomed to playing it more freely.
Dealing with this tempo change has been instructive and it gives me some experience for the future as this issue will come up again. I'm definitely convinced that you need to ensure you practice the piece exactly the way you intend to record it. This seems obvious, but in this case I had been playing the piano part for so long that I didn't think this would be much different. And it wasn't really, but it was different enough that my playing become too mechanical and lost the vibe.
As mentioned, I'm also planning on tweaking the camera setup. My plan now is to use my GoPro and iPhone cameras so I have at least two angles of each performance. This just adds another layer of complexity, but I always enjoy performance videos with multiple angles (if done well) more than just the simple straight ahead shots. The GoPro is ideally suited for "neck cam" on the guitar and bass, and I'll also be able to get some alternate overhead shots as well as a shot of kick drum performance (this will be used sparingly, but I like this angle on drum vids).
The ostensibly biggest obstacle in this project (famous last words) is getting the piano and drum performances down solid. These are the two that I tend to make the most mistakes (albeit few/small) compared to the vocals, guitar and bass. I find repeating takes of those performances is less arduous and time consuming than piano and drums. My plan is to approach it in this order - drums, piano, bass, vocals and guitar. My main reason to redo the bass & guitar is because I want the additional camera angles and I need to fix the way I use the green screen. I also need to help Aeyong install a window blind for our little half moon accent window which is still letting too much natural light in which can illuminate behind the screen and that's a problem.
Another impetus to repeating the entire process is that I've purchased my next equipment upgrade, which is the Slate Virtual Microphone System. This is comprised of a large diaphragm condenser microphone, preamp and a software based modeling system that models microphone, preamp, and compressor (as desired). This system was announced a few years ago, but it's just now hitting the market. I've heard all positive reviews so far, but I don't know that it's hit enough of a market saturation to be certain.
Based on the microphone alone and considering its price point ($1K), it's a good upgrade from Blue Baby Bottle (which is still a great mic) and should fill a niche for several years. Philosophically this system is the microphone equivalent of the AxeFx. As the technology has progressed over the years, modeling has narrowed the gap with analog gear, and most listeners don't know the difference. I've always embraced technology and I while I love analog gear, I also love the digital stuff for the flexibility, variety and power it provides. The three microphones it models (at onset, more will be added later) would be way beyond the wallets of all but the most successful (rich) producers. It models the Telefunken 251, Neumann U47 and Sony C800G microphones. Bought on the market (especially for classic versions), this three mic locker would likely run upwards of $50K. Of the reviews I've seen so far, people are having a difficult time telling the difference between the original and modeled versions. I'm not sure how much I buy that, but for the price and what's a decent sounding (extremely flat) microphone it's more than worth it. It will be interesting to try this new mic out on the vocals for this project.
There are more aspects to the video project to come, to include filming our dogs in various costumes against the green screen and the subsequent film editing and use of effects. These are all essentially new skill sets I will develop as part of this project. I'm always going to focus on music first, but an occasional video project is fun and a way to keep it interesting for the short modern attention spans (including mine).