📒 My Thoughts
Finally it’s done. Especially the last two days were a nightmare, as usual. I’m not entirely satisfied with the result, probably because I don’t like this song as much as the last ones and it’s less melodic, so Jeff’s singing skills aren’t utilised to their full potential. But whether Santa Claus Is Comin’ to Town is good or bad, better or worse than other songs or less re-listenable is probably subjective. And the lyrics fit perfectly to the topic, I just had to use it. What I also don’t like is that this fideo is shorter than the previous ones but it’s all I could do in 1.5 weeks with few sleep. It might be shorter but I added an additional verse at the end  because there wasn’t enough space to talk about all of the toxicity-related stuff I wanted to get in. Next year, I will surely use a longer, “better” (subjectively) and more melodious song. More on that later.
Preparation I: Developer Updates
When I heard of youtube-dl through its temporary taketown, I was really excited about what it meant for devupdates/jeffsongs. I remember, like 10 years ago, my dad used and showed me an application called “Clip Grab” to download YouTube videos. But unfortunately, it stopped working after a couple of years. I used convert2mp3.net (I believe it was called) for multiple years, also for the first devupdate-based fideos. I think I found it at the top of search engine results. When that site went down, YouTube downloading went downhill. All I could find on search engines were most likely malicious websites and they were all unusable. Except for one potentially malicious desktop application that was the only working way I knew of for like 1-2 years. Until I learnt of ytdl. Which is incredible, it’s open source and supports full availability and control over available formats. A dream come true. And apparently, it’s been around for a long time, I just never found it. That gave me the idea of searching for tools on GitHub instead of web search engines. And a whole new world revealed itself to me. Working, open source and secure tools for everything. Instead of malicious websites with bad online converters or whatever. Another great tool I found and recommend is kobodl to download Kobo ebooks as drm-free epub files to read them anywhere and to actually own what you purchased instead of just a temporary right to limited access that can be revoked at any time.
To download all devupdates, I went through all of PlayOverwatch’s YouTube fideos to make sure I don’t miss any. As it turned out, there were quite a lot that were not in the official playlist and that I never used. That should expand the available vocabulary a little. There were even a 17-minute and a 5-minute devupdate with Tim from the Overwatch team and somebody else called Phil from the Overwatch team that I completely forgot about. That calls even more for a dedicated Tim fideo as there are some who call him Tim. He … is a programmer. Unfortunately, Tim speaks very fast and not as clearly which made his words very bad for a song, but maybe it’s usable for regular sentence creation.
It’s always amusing how I keep recognising parts in devupdates when I search for words and know exactly how he said a word, what else he said around it and how I used it before. And it’s interesting how much I’ve learnt about Jeff, pronunciation and Jeff’s pronunciation. Until like half a dozen of years ago, I thought it was spelt and pronounced “pronounciation”, by the way.
I executed the ytdl-devupdates.sh script and received a complete collection of all devupdates in the highest quality available for mortals. A dream come true! But of course, there is Resolve. Which doesn’t accept any codecs. It supports VP9 import but only in .mov and ffmpeg only supports VP9 in .mkv, so I had no way to get Resolve to accept the VP9 devupdates. But YouTube also has every fideo in every quality available as worse and non-free h.264. So I was about to download the h.264 versions as well. But when I tested importing one of them, Resolve of course also didn’t accept it. At first, I assumed that there was something about the way it was encoded that Resolve didn’t accept as there are many factors in fideo and audio encoding that, like how Firefox only supports h.264 and VP9 with yuv420 colors. But I couldn’t find anything unusual with the file. So I tried importing other h.264 files that I used in Resolve before. They didn’t work either. I asked the interwebs and found nothing useful, as always. But then, I checked the official codec support list. Only to find out that Resolve on Linux only supports h.264 with the Studio version. The only codec choices I had to get the devupdates into Resolve were rawvideo (which is uncompressed and therefore gigantic and would not fit on my hard drive and would take over 100 hours to convert), png (which would still be way too big and take even longer because it’s compressed) or disgustingly compressed and gigantic dnxhr. The first two options were impossible due to storage space and because I couldn’t wait 100+ hours, so I had to use dnxhr hqx 422 which made the resulting fideos look substantially different but it still looked acceptable. It took 24 hours to convert the devupdates and the ones I used took up 674 GiB. About half the size png would have been and just small enough to fit on my SSD. I could import the resulting files and, thanks to the SSD, I could actually perfectly scrub through them with no lag or stutter at all and at very high speeds. It really makes a big difference, I’m glad I’m using my SSD for fideo editing now.
Preparation II: The Song
its now 3 am, i should continue writing tomorrow but i alsop dont want to because i want this fideo to be over and forgotten as always and the last time i went to bed instead of writing the description because i was too exhausted was last december and that resulted in me never actually writing that description which is very unfortunate because many interesting things are now probably forgotten | good yawning, i woke up at 13 and couldnt get up until 15
The two topics I still had in mind from last year were toxicity and saying thanks to the developers as something positive. Since this was not the best year, I planned to do a positive thanks song this year to lift everyone’s spirits. But aside from that general idea, I didn’t really know what to put in the lyrics. And I had no idea what song to use. I already used the three good and melodious ones with lyrics and music videos that I could base the cover on. And titles that could be modified to fit the cover. I listened to all Christmas songs I know multiple times and wrote down ideas for them. The only things I could think of were “you know tracer and …”, “play, they told me …” and “do they know their game at all / something about people who dont play ow and stuff like "thank god its those pitiful people who dont know the game instead of you" and "where noone is ever toxic, all theyve got is fun"”. One song that I really like and that sounds melodious is Christmas Time, but I didn’t have an idea for the lyrics and while it does have a music fideo, it’s just the singer singing while getting buried in snow. But I thought the lyrics could fit the positive topic, so I decided to use it and started writing lyrics. But it never clicked and all I could come up with was We waited all through the day For the evening, far away When we could play together In Overwatch“ and “And now the time has come To say thanks to everyone On the Overwatch team for this game we love to play”. Sounds unimaginative, boring and not the positive, interesting and uplifting song I wanted to make.
But then, I remembered Santa Claus Is Coming to Town. Jeff Kaplan is coming to town would be the obvious and wisible change. And then, I could let Jeff tell people not to do toxic things. And I immediately came up with the chorus of “I see you when you’re bad I know when you are mad I know you blame your team for that But get good for fuck’s sake” and i absolutely loved that idea. The lyrics made click, I had ideas and I wanted to continue. So I abandoned Christmas Time, for which I also had no idea how I could change the title, and switched to town. The lyrics were perfect for the toxicity topic. Which is great but also unfortunate because I wanted to make something positive. And, while the lyrics and title were perfect, the music isn’t as great as other songs. Apparently, there are hundreds of versions. The one I like the most is the one I used by Frank Sinatra. It’s good but not as good or long as other Christmas songs. That already made me know that this year’s song wouldn’t be as good as the previous ones and it would be a worse Christmas fideo for the first time instead of a better one every year. But I also didn’t have much time left and a longer and subjectively better song would probably take too long. So I continued, hoping that the lyrics would be interesting enough to make it acceptable. But one thing that I did improve from the last years is that this time, the lyrics consist of 100 % rhymes and 100 % crimes. I didn’t bother with rhymes in the last years to get more time for the actual song. But if this year’s song has to be carried by its lyrics, it has to have rhymes. I’m glad I did that and I like the result. Fresh D would be proud.
Santa Claus Is Coming to Town is (was) not on my Christmas playlists. Actually, I don’t even know if I knew it before ugotaphonenumber’s masterpiece. I certainly didn’t know any of the versions on Spotify and YouTube I checked out for my cover.
The original lyrics are quite something. They tell children of a powerful man who watches them in their sleep, sees everything they do and who punishes them for being sad. They have to be “good” or otherwise he will get them. Why are so many children’s things so weird? And I do not at all support the idea of doing something positive because they are forced to do so to avoid punishment instead of because they also agree that it’s good and they want to do good for the sake of goodness. Literally, not figuratively. That reminds me of a debate between John Cleese, Michael Pallin and two very clergy men. And one of the latter said that religion is good because Mother Theresa did what she did because that’s what Jesus said, unlike all the others who only do good because of the idea that everyone should be happy. And if Jesus either didn’t exist or even just was widiculed, she would stop doing good things. That they not only say that religion is needed to force the bad and sinful humanity to not kill each other by telling them of an almighty man in the sky with total surveillance who will punish them but also that people that don’t need to be forced to do good and do it because they want do do good is actually worse than being forced is even more outrageous. And I don’t know anything of Mother Theresa besides her name but I believe I’ve heard that she wasn’t as good even with Jesus forcing her.
And the lyrics of Do They Know It’s Christmas are very … remarkable as well. Those poor, uneducated, pitiful and ignorant people in this weird, faraway place called Africa don’t even know the day in which some religion celebrates the day of capitalism, throwaway society and pretense? There are people who don’t have snow when the northern areas of the world do? All they have is life instead of amassing useless products to compensate for not living the life one has? And of course, there are no rivers or vegetation in Africa. But thank god that he only punished some faraway people with these things and not you.
Step I: Sentencing
I always used Notepad++ to find Jeff’s words. But that’s a Windows application. So I used VSCodium instead. The search for all open files in the left sidebar unfortunately doesn’t show the beginning of the line the result was found in if the result is far-right. But I learned to research issues. So I did and I learnt about the button on top of the search sidebar that looks like a “new file” icon. It opens a search as a tab that shows complete lines, file names and optionally the line before and after. This was perfect. It was only unfortunate that I couldn’t have that search open and then double click on a result to open that transcript while still having the search open. It just switches to the transcript’s tab, away from the search tab. But I didn’t have to go to the actual transcripts, apart from having to correct some mistakes, and if I have to know what is around the result, I can just activate displaying the lines before and after the result on the search tab.
Otherwise, there’s nothing to talk about. Only that getting good words for songs is harder because they have to be spoken clearly and slowly to have long vowels that can be stretched out and they have to be as monotonous as possible for the tuning to sound good. If they aren’t monotonous, they sound like “Junkrat” in Balance. And Jeff still hasn’t said Christmas. And the only time I made him say Christmas was apparently in Balance as well, which was made in Fideo deluxe, meaning that I had no way to reuse it. Or even open it to look for the word pieces I used because I don’t have Fideo deluxe installed. But it didn’t sound good anyway, at least the mAS part. So I made a new Christmas, this time out of three parts only. cORISmess. “Orisa” still doesn’t disappoint. The result sounds really good. Unlike some other stitched-together words like blAIM. But I didn’t have enough time to spend hours to maybe find better word snippets.
Step II: Tuning
I hoped that FL Studio would work properly on Linux. I had it installed with Wine, but who knows if it actually worked? And I immediately found an issue: I couldn’t merge clips in Newtone with shift+alt+click. I tried it in a Windows 10 VM but that didn’t work either. Then I remembered that I had shortcut issues in FamiTracker as well that were caused by a shortcut already being defined by the operating system. So I looked for a shift+alt shortcut but didn’t find one. Research on the interwebs didn’t lead to anything, of course. After hours, I found a different way to merge clips: If I permanenty activate cut mode by pressing C instead of just temporarily toggling it with shift, I could hold ctrl which activated the merge cursor and worked. At least, I think it was ctrl. Doing it this way is not listed on the official hotkey list or anywhere else. But it worked, so I’m not complaining. And I could continue and tune my first Jeff word on Linux.
Actually, there was another thing I researched. The worst thing about FL Studio or Newtone has always been that there is only one undo step. Pressing ctrl+z a second time just re-does it. And this time, I found out that there are unlimited undo steps if you press ctrl+alt+z instead. My first reaction was: … why? My second reaction was: … well, at least I know now, that should make it so much more comfortable and remove the constant possibility of everything getting permanently messed up.
I couldn’t have pauses between clips in Newtone. Whenever I cut out silence and then moved the part to the left or right of the gap, the other side would immediately be stretched out to snap to filling the gahp. It wasn’t like this on Windows, it only snapped together when it was close. So I had to cut tiny parts out and set their volume to 0 to be able to have breaks.
Otherwise, tuning Jeff’s voice wasn’t as bad as it usuall was. Besides the breaks, I didn’t have any issues or bugs, unlike usual, and the possibility for unlimited undos was very comforting, even though I only needed it a couple of times, unlike before. After a bit, I stuck to the strict pattern of cutting syllables -> setting note heights -> setting lengths -> setting volumes. That let me always focus on one thing at a time and made me more efficient and resulted in better results. I also prepared the notes beforehand in LMMS so I can just look at them on my other monitor instead of also having to listen to the original song and analysing the notes at the same time as cutting, pitching and lengthening.
After I tuned the first verse, I recorded it with Audacity, as usual, because I’m not paying 300 Bucks for a buggy application I only use hwhonce a year. But when I put it in Resolve and let it play along with the instrumental … it sounded horrible. I already formed excuses to myself in my head about how this year’s song is not only shorter and less melodious but also sounds horrible, how I went from improving a lot every year and making great fideos even I really liked to something horrible that hurts to listen to. But then I tested something. I used Resolve’s feature to decrease the pitch by one semitone. That made the overall sound very weird, but it actually sounded right now and not disgusting. I put the same Jeff verse below a later instrumental verse and it sounded bad again. I removed the pitch change and it sounded right. There was actually a key change for the verses after the middle part. I did notice that the last two notes of the original verse sound exactly like G# A instead of A A#, but Frank’s singing voice takes a lot of liberties both in pitch and timing, so it’s not easy to even know what note he’s supposed to be singing. I assumed his pitch was just a little off at the beginning, because other notes in A# major seemed to fit. But I apparently based the notes I put in LMMS on a later A# part. When I realised that and changed the key of the first two verses and first pre-chorus in FL STudio to A major, everything sounded right and, while not as good as the previous songs, acceptable. It’s interesting how horrible it sounded together, even though the instrumental part doesn’t even play the vocal’s notes. Music and how pitches fit or don’t fit together is fascinating.
One new thing I did before tuning this time is putting the entire finished un-tuned lyrics into Newtone and setting them all to one note to see if the words I used are usable for tuning. So I could hear if any words are not good enough and I can prevent a Junkrat word beforehand. There were a couple of words that sounded horrible, one in particular took some time to find a usable replacement for. But most were good enough because I was so strict with using only slow, clear and monotone words. And thanks to this extra step of testing and replacing badly tunable words first, every word sounded acceptable when I went to the actual tuning.
One thing that definitely didn’t sound good after tuning were Tim and Phil’s parts. Those were the best instances of those words or assembled word snippets I could get. Tim especially talks very quickly and mumbly. Relatable. I really wasn’t sure if I should even use them. But I already made them and I thought they were something different and interesting. I had to at least make something new this time. And maybe they’re bad in a way that is wisible but not so bad that it’s too bad to be listenable. I hope.
Step III: Synchronising Jeff’s Face
This is the second time I used optical flow frame interpolation for Jeff’s visuals, making it perfectly smooth 60 fps at any stretched out speed. It really makes a great difference. Only a couple of times, there were duplicated frames in the original, making the interpolated frames between them completely static. But I could remove the frozen part which resulted in a perfect motion again. Even the professional fideo people at the Overwatch team can’t make fideos without duplicated and dropped frames. Then again, they also can’t set the intro to a reasonable volume or Jeff’s voice to the same volume in each devupdate.
The speed of a clip can be changed by pressing ctrl+r on an active clip and then dragging it on the new top part. This would be perfect, as I could just cut the syllables and then drag the beginning and end to where they have to be according to the tuned voice. Unfortunately, dragging the beginning or end like this does not at all properly keep the beginning and end frames. When I drag a clip out to twice the length, it’s not at half the speed, but faster than that with new frames at the end. So I have to constantly drag the speed changing part and then drag the regular part to get rid of the additional frames at the end, or to get them back when I want to increase the speed. The fact that speed changing clips doesn’t properly keep their original length means that I can’t just cut out all syllables/letters and then drag them to span the needed length. I have to change the speed of the entire word, constantly adjust the length and then cut the first part off if it’s right, continuing with the next part. This is the only way to make the parts of different speeds have perfect transitions because the beginning frame doesn’t change when changing the speed but the end frame of a clip does. This is annoying, but it’s doable and it works.
Somehow, I messed up the audio tracks while I moved stuff in Jeff fideo tracks. Part of the original cut and untuned voice clips were gone and other parts overrode the song. Jeff’s tuned singing was still all there and in place, but even the instrumental parts that were still there were at the wrong time. So I had to import the instrumental again, synchronise it to Jeff’s singing this time and cut out and add the fifth verse at the end again. Which took some time because I couldn’t get it to fit for a while this time. By the way, at first, I used a copy of the first verse as fifth verse. But when I realised that the verses 1 and 2 are in a different key, I had to use verse 3 instead. The transitions ended up being good and unnoticeable, I believe.
I also used the occasion of having to re-synchronise the voice and music to moving the entire fideo to make the song start after one second of silence. Unfortunately, I realised later that doing that snapped all audio clips to frames. In Resolve, audio clips don’t snap to frames like video clips, they can be moved with a seemingly infinite amount of accuracy. Which is great for synchronising. But this messed up the synchronisation a little. I have to keep this in mind. I have to avoid this in the future by moving audio and video separately if the audio clip is not snapped to the fideo’s framerate.
I also used the incredible and iconic Junkrat hand movement again to fill the gaps between verses after there was no space for it last year. I absolutely love it. That’s the best thing that came out of the Jeff songs.
Preparation III: Ingame Scenes
There was no original music fideo I could base mine on which would make it an interesting reference for those who know the original. I didn’t have any idea what to do and it was almost Christmas. Eventually, I came up with the story of a team that doesn’t work together, is toxic and bad, but they blame it on Hanzo who is just as bad as them but picked a hero the others consider a throw pick and therefore has the sole responsibility for everything going wrong in the match. To illustrate that, they would all turn around to look at him, attack him, use disapproving emotes and Doomfist would throw because it’s already lost by switching to Mei and blocking off the exit as the round starts. The last scene was supposed to be most people switching to a composition with more synergy and then leaving spawn together as a team with a renewed spirit while the camera pans to show the christmas tree in spawn. More on why some of that wasn’t in the fideo later. I wasn’t convinced by this idea but I couldn’t come up with anything better and it was already late on the last day. I even considered not doing ingame scenes at all because I didn’t want to make people do these boring and unimaginative things and then maybe even have them watch a bad video with them in it. It also didn’t help that my script required a full team of 6, so I had to get a lot of people. But I didn’t have anything else to fill the non-singing parts, so I eventually booted Windows and started to ask for volunteers in the Overwatch global chat once again. After I didn’t do it in March because I could do all Link Remastered scenes with Workshop bots.
Step IV: Ahcting!
Getting together enough people took more than an hour, I believe. Unfortunately, I didn’t start asking in global chat hours or even a day in advance, telling people when we would record, as I did last year. When I had 5 other people, we got to the Mei wall scene. Unfortunately, I apparently had already let the round start, so the spawn doors were open. So I set the match time to 0, ending the game to get a new one with closed spawn doors. Unfortunately, four people left while the match was ending. So I had to look for people again. Incredibly, DEADL0CKEd did not only wait for like more than an hour, they also asked in global chat and got one person. I did not at all expect this degree of patience and helpfulness for a random person on the interwebs with 1, 2 suckscribers who wants to make a bad fideo. I can’t say thank you enough!
In a stroke of luck, inkygames came online. I asked them and they were eager to be an unimportant secondary character again and also brought along someone else. Shortly after, RubberCalyps came back and we were 6 again. This time, we did all scenes. The only scene that was recorded during the game instead of later in the replay was the chat part. As it was bound to happen, GenEnorF … I mean, Shadowplay deactivated itself exactly in that moment and refused to turn on again. I expected this and immediately gave up on it, telling the others to wait while I get OBS ready. But DEADL0CKEd told me I can make them owner of the custom game and take time to restart my PC. So I did that and it actually fixed it, Shadowplay worked perfectly after that. Another great thing DEADL0CKEd did for me. I would have recorded everything with OBS which would probably have been stuttery, as it has always been when I had to use it instead. I must have been blessed with a heavenly helper sent from above on Christmas Eve Eve. Up they come and save me, bloody do-gooder.
Apart from that, the ingame scenes went well this time. The Tracer 1v1 took longer than expected but the way it happened was perfect. I envisioned it as Tracer blinking in and immediately killing me. But she took longer because the enemy team is bad as well, missed a pulse bomb and even took the health pack that was next to Hanzo the entire time when he wanted healing but didn’t take. It turned out so much better than planned, another Christmas miracle. Thanks to Sun!
Step V: What Remains of Edithing
But then, everything went downhill. There were three replays because I accidentally pressed the button to set the timer to end the match after the Mei wall scene and because of when I restarted my PC. The middle replay with the scenes outside spawn was perfectly fine but the other two couldn’t be opened, just giving the error message “Unable to view replay of an undecided match”, even though it says loss on all of them and they all ended as wins for the defenders because I set the time to run out. I know from previous fideos that matches have to have a proper conclusion for the replays to be accessible. Setting the timer to 5 and letting the match end as a win for defence or the TDM team with more points has always worked. Until this time. I was devastated. I didn’t want to waste more of the voluntary actor’s time and some of them went offline. And getting new people would take hours again. But I also couldn’t do that because that would be unfair to the previous actors. They spent time and effort to help me and then I don’t even use their scenes? And just using the outside scenes from the working replays and then the other scenes with completely different people would also be inconsistent in the fideo. But I also still had the working replay from the first try with partly other people. And very fortunately, while the Mei wall scene wasn’t acted, Mei still placed an ice wall there and Sigma even walked towards it a little, making it look a little like the planned running against it and then attacking it and Mei, as we did it absolutely perfectly in the second try. So I recorded that and the other beginning spawn scenes in that replay and then the outside scenes from the other working replay. I hoped that the scenes from outside spawn would be enough to fill the middle part.
I let ffmpeg convert the recordings to dnxhr, but that took like 2 hours, so, after a while of working on other stuff like the thumbnail and fixing synchronicity errors, I felt so exhausted that I couldn’t keep going. That was when I tweeted the title and thumbnail of Jaiden’s video about overworking oneself because the title “The Closest Feeling to Death that isn't Death” came into my mind at that time, being a perfect description for how I felt. I write notes about noteworthy things that happen while I work on a fideo to make sure I don’t forget to mention them when I write the description novel. At that time, I wrote this: “my head hurts my brain hurts as mr gumby said i cant think i feel dizzy and cottony inside and as if my soul has already half-left my body and is losing control over the latter and its perception i just want to lie down and sleep forever but then ill be in the other world full of nightmares”. So I went to sleep at like 5 am. When my alarm went off … I don’t even remember what I did. I always have my phone on my sofa at night so I have to get up to switch off the alarm, otherwise I wouldn’t be able to get up. I don’t know what happened, but I do know that I had a dream in which I somehow did something that enabled me to load the broken replays and I was really excited. Only to be greeted by another error message after the map already loaded. I woke up immediately and wanted to try what I just did, but then I realised that I forgot what it was that I did in the dream, that it was probably nonsense and that I felt my half-dead and incredibly tired body again, so I closed my eyes again and fell asleep again.
I actually got up at 11. Which was way too late, considering the fact that I still had to edit the ingame scenes, finish the thumbnail, “animate” the outro with all ingame actors, make subtitles, prepare posts and of course write this description. In addition to the fideo rendering, uploading and being processed by YouTube, which takes hours. I continued editing and I was able to fill the middle part, although the healing scene is way too long, in my opinion. I hope it didn’t bother others too much. But the final scene was still missing. So I had the “idea” of just panning out over festive King’s Row with everyone grouped up on the high ground. That would at least show the actors from the second session that was partly lost more. And give purpose to me telling them to wait up there when they weren’t (physically) in the scene. That reminds me, the plan was to have Moira spam “Need Healing!”. But I didn’t see it in chat. At first, I thought she just forgot to press the button. But I heard a sound. And then, the others told me that only healers and the healee see it in chat. I’ve been playing since the open beta and I didn’t know that. Maybe because I main healer.
Since this song reminds me of ugotaphonenumber’s cover, I wanted to get an easter egg of that fideo in. My first plan was to actually use the “to toOOOOOOOOOOOOwn” part once instead of Jeff singing it, like how I used “screaming” once in Balance, if I recall correctly. But then I didn’t do that and just put Soldier’s face in the corner, cutting in and out with the Jeff cuts, making it very inconspicuous. Did you notice it? Maybe you did now that Phoenixgamer commented that he found it. I also put one other easter egg in, the Ness pixel art when Jeff sings “kindliNess”. Rest in peace, along with my mum’s tortoises.
I finished the fideo at 17:00. Rendering it as uncompressed rawfideo (to HDD because my SSD was full from devupdates, it would have taken 15 minutes) took only 25 minutes thanks to the devupdates being on my SSD. That’s incredible. Unfortunately, converting that giant 235.5 GiB rawfideo to VP9 (with the best settings) takes 50 hours. And I need 3 VP9 versions, main, web1 and web2. So I let Resolve render a dnxhr lb version as well which I used as final render before my recent codec and ffmpeg adventure, so I could upload it to YouTube on the 24th and not 50 hours later. That took only 5 minutes. Absolutely incredible. All thanks to the SATA SSD.
Stweam Me, Centuwion
I streamed part of the creation of this fideo. Which was actually great. It felt like I was with someone even though I wasn’t (and without the stress that usually comes with social interaction as an introvert) and it kept me awake, efficient and didn’t allow me to overly question and doubt what I was doing. Unfortunately, the connection was so bad at times that the stream was partly unwatchable and even completely ended once. Connection losses happened every couple of seconds. According to bandwidth tests, my upload was as low as 0.3 kbit/s. Or rather as high as that, because it constantly went down to nothing at all. And probably because of this, my ping went up and down between 12 and 500+ ms. That explains the up to 50 % stream frames dropped by the connection on most streams since I started streaming a while ago. I just never noticed it because I haven’t played a fideo game since almost a year. Well, except for checking the shops in ACNH every day when I’m on the toilet and brushing my teeth at the same time for maximum efficiency. Streaming was really great this time and I wasn’t nervous at all, probably because I just streamed fideo and audio editing which is just mechanical work that has one way to do it and a specific correct outcome. Drawing or 8-bitting on the other hand has infinite possible ways and outcomes, almost all of them being bad. So I’m nervous because others could see me fail and realise that I actually can’t do anything and if I do something that’s not bad, it was just luck.
And What Brings Me to That Conclusion?
It’s now 2 am on the 26th. And I haven’t eaten since breakfast to get this description done. As I said, I just want to get this over with and forget all the work and misery that went into this, as always. Along with the result. So, see you after the next overworking session after I’ve done nothing for a couple of hours (instead of having fun because having fun means being lazy and wasting lifetime) to fill my bars up to the point where I can just function again.
The result turned out acceptable, not terrible as I feared when I used the wrong key. It’s still disappointing. I really liked the last two Jeff songs, and me really liking something I did is very rare. Last year, I made a great song, even though I was double sick in the end of November and beginning of December. My expectations rose each year, every song had to be better. And now it’s not only not better, but worse. Like when my mum told me early in the year that I’m stagnating and wasting my potential by making what I make instead of becoming a highly paid Microsoft Fuhrer. But in my current state of completely overworking myself, there really is no difference in me doing what I’m doing and having a highly paying horrible job, at least for my well-being. I have to change this. But I’ve also had the idea of a New Year’s 8-bit cover for a long time and was really looking forward to it. I want to cover the PMD2 credits theme and show an image of every fideo I’ve made this year. But the PMD2 ending theme is long and I’m completely exhausted from this fideo and also from … the entire year, actually. There are also other songs that I would rather cover right now. I never do what I want to do, always what I somehow tell myself I have to do. I rather just sit here, not doing anything and just feeling bad instead of playing the game I want to play, watching something I want to watch, covering the song I want to cover and so on. Instead, I play games because I feel like I have to, like Overwatch because it’s been my main game for a long time and because I know two people there who want to play with me and where I’m very afraid of disappointing my teammates or like ACNH where I just check the shops, recipes and so on, efficiently while I’m on the toilet and brushing my teeth, to get good things for when I one day have time to actually play the game. Now that I think about it, I don’t remember the last time I’ve actually done what I wanted. Well, I did do and enjoy the Pokémon Center and Drenched Bluff covers. But even when I do what I want to do, I don’t allow myself to have fun. And thinking that something has to turn out better when I have fun also puts more pressure on myself and results in the fun disappearing and the result being worse. I don’t think I’ll be able to do the New Year’s cover in time. I could try, of course, but I probably shouldn’t. I won’t have fun and it will turn out bad when I have to rush it, especially in my current state. And being bad is what I fear. I should take some time off and learn how to do what I actually want to do. I remember Akai tweeting about learning to do exactly that as well. It’s fascinating how much we humans can struggle to live when it seems so easy in theory. I should start right now. But I don’t want to disappoint myself by not doing the New Year’s cover. But I’m also afraid I’m not good enough for it yet, like how the more difficult covers of Dialga’s Fight to the Finish and Cynthia didn’t turn out that great. What do I do? Do I rush it for the 31st? That is probably out of the question or the song will be bad and I will be dead. Do I take my time and finish it a little later? Maybe, but I’m still extremely exhausted, my sleep schedule is messed up and other stuff. And I’m afraid I’m not good enough yet, aside from my current exhaustion. Or do I start doing what I want to do right now by doing a different cover I’d rather do and postpone it to next year?