1,371,886 events, 863,422 push events, 1,160,487 commit messages, 66,156,292 characters
converts outputs from male and female to man or boy and woman or girl
i might fucking hate my life but i hate C# more. -> migrate off of ET onto r2api -> add contenthandler.cs, which contains shit for buffs/bodies/masters/effects/skills -- basically everything r2api fucking fails at ->add display rules back -> unfuck tar river -> add item display rules back -> add temp-cube for custodian
Added Jada Life Collection and Cyberpunk Samurais
I have added my Jada Life Collection and I have noticed I did a mistake with the Cyberpunk Samurais. I added the wrong policy! I am so sorry, I feel so stupid :( And I dont find a way to update a already merged project.
Here is my twitter: https://twitter.com/Nft4R And my site: https://nftforged.com/
Once again so sorry for my mistake!
"10:40am. Let me chill just a bit and then I will start. I seem to be wasting too much time in the morning, but that is fine.
10:50am. Ok, focus me. Time to do my first trace. I won't get better just sitting here.
10:55am. The image I've picked is a screenshot from one of the early eps with her looking down angrily. As much as I want to draw only her thigs and waistline, I'll want to draw the background and the face for practice as well.
11:15am. Wow, it took me this long just to do the left hair tail. It is not as easy as the vids make it look. Instead of using the pen tool, I find myself using the curvature tool and then adjusting the paths using the achor point tool.
12:40pm. I am just about done with the hair. It is easy to get into this. The curves themselves aren't hard to place, but I am in awe of the animators who can do all this in under an hour. At the rate I am going I'll be stuck working on this image for days. But that is fine. Think about how much time I've wasted in 2019 and 2020 learning random shit.
That I might be able to get 1,000x speedups using AI chips is optimistic speculation on my part. The realistic range is somewhere between 10-10,000x with most of the probability mass between 10-1000x.
Would something like a 50x speedup be enough? Who knows. If I trained the agent would it be reliable enough? Would I be able to get past the UI automation obstacles?
Sigh, who can deal with all of that.
I picked poker as the most sensible goal, and I've aimed my mind at accomplishing it, but I was wrong. Automating poker in real life is actually quite hard.
To be sure of success, I need the neocortical algorithm and the AI chip to implement it on.
Forget probabilistic programming. Forget such long shots. I've been way too early in my pursuit of the Singularity. It is going to happen when the pieces fall into place and not a day earlier.
I think ultimately, I've been an artist all along. I thought that picking money as one of my explicit goals would get me the best of all worlds, but in the end that just became a part of my ideology rather than something I'd want to pursue directly. Anybody with such a goal would have gotten a job long ago.
12:50pm. But the path is viable. I'll learn to draw and compose and then I can create games all by myself with no problem. That should get me some income.
My programming skills themselves are extremely strong right now and I won't have trouble implementing whatever I want. That is one thing I've attained for my past 6.5 years of effort.
Let me have breakfast here. After that I'll have the chores waiting for me. My dad brought me a computer to install Linux + Hotspot on and that will take me a while. If I had my own income I could reject such tedious tasks, but now I am not in a position to do so."
Forward go mod tidy
stdout/stderr
This is a command that can fail (in my case I think for stupid reasons in a hell of my own construction, but nonetheless). Right now we just get
$ go run github.com/Khan/webapp/dev/cmd/gqlgen
tidy failed: go mod tidy failed: exit status 1
exit status 3
which is not the most informative. Now, instead, we'll forward its output to our own stdout/stderr rather than devnull.
HOW DID A '~' APPEAR IN MY NVIM DIRECTORY I THOUGHT IT WAS A SYMLINK I FUCKED UP ALL OF MY LUA CONFIG AND WHO KNOWS WHAT THE FUCK ELSE I DELETED WITH THE RM COMMAND I AM HONESTLY SO PISSED. IT'S 12AM AND I HAVE TO REDO MY LUA CONFIG AND CONSTANTLY WORRY ABOUT RANDOM SHIT THAT I DELETED WTF JUST HAPPENED. I HATE IT SO MUCH WHEN RANDOM SHIT HAPPENS IN MY DIRECTORIES WHEN I SERIOUSLY DIDN'T CALL FOR IT
holy shit so many changes holy motherfucking shit anyways push it
Massive Keymap Reformat! + Layout/Splash updates
I have the name: LudvikKoutny This user started a thread on: https://devtalk.blender.org/t/2-8-keymap-file-formatting-mess/5279 about how blender just chews up keymaps, dropping items etc. And, oh boy, was that my experience! The past handful of updates have all been to try and resolve issues with one or more commands not functioning in my template vs blender default and industry standard, but the rub has been that after each update (editing the keymap inside of blender) I export it from blender, which corrupts it, and then I end up with different issues the next time.
I discovered that blender had just deleted 30-40 items from the Sculpt, Font, Image Paint and a few more from 3D View, it had also added empty keymaps for Gizmo and Header, which weren't present when I compared against the built-in keymap.
Sooooooooooooooooooo, I manually reformatted and indexed the BLUi keymap. It is functional, but the focus was on trying to increase readability and editability for humans. It will be a while before I trust blender to export a keymap properly again, so it made sense to create something that I could easily edit manually in a text editor.
Splash screen is now a little more generic, to permit version #, credits etc
Layout has changed slightly, I wasn't happy sculpting in the little viewport, so I renamed the existing workspace to 'Layout', and added a second workspace called 'Model / Sculpt' that doesn't have the File Browser, Outliner or Properties showing, and doesn't hide the Tool Settings by default.
"1:50pm. Let me chill just a bit more and then I'll do the chores. I remember that PC being particularly difficult last time and taking me forever to deal with because I had to mess with config files.
2:35pm. Done with the first part of chores. Now let me get to work on the other part.
2:40pm. First step is to burn the Ubuntu Server to an USB stick. Now let me bring in the laptop so I can see the Grase Hotspot install instructions. Ok, now let me disconnect my own computer so I can plug in the other. All this will probably take me the remainder of the day. Here is hoping that I do not have to mess with config files too much.
3pm. I burned the USB in wrong mode and it is trying to mount is as a CD-ROM. These stupid timewasting hassles is why I've learned to hate Linux. I need to repeat the process. It has been years since I last did this. I hope that I get money at some point in my life so I can turn this nonsense down.
3:50pm. Ok, I got Linux and Grase 1.8 installed on that PC, but now comes the hard part. I remember years ago fiddling with the network card settings. The stupid thing has 3 lan cards and one of them has to be named eth0
instead of pxp1
. I need to go over my journal entries in turn.
3:55pm.
///
10:05am. https://askubuntu.com/questions/217635/how-to-rename-an-ethernet-interface
I am going to try editing '/etc/udev/rules.d/75-persistent-net-generator.rules' to do what it says here.
I hate Linux so damn much. It is always takes a humiliating song and dance to get anything to work properly on it. I should be starting my day by chilling and then moving to programming, but I am wasting my time with this instead.
10:15am. For some reason it does not seem like the udev rules are doing anything.
10:30am. https://groups.google.com/a/grasehotspot.org/forum/#!topic/grase-hotspot/mz-i-_03YlE
I hadn't known that sudo update-grub
recreates the grub.cfg
file. I've been using that command sporadically.
GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0"
I keep seeing this as being GRUB_CMDLINE_LINUX="net.ifnames=1 biosdevname=0"
elsewhere and it is confusing. Let me try the first one.
11:05am. Ok, I understand now. I finally got the hotspot under control.
GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0"
Modifying this line it /etc/default/grub
and then running update-grub
to recreate the grub config file, and then reseting causes the /etc/udev/rules.d/70-persistent-net.rules
file to get generated with the appropriate interface cards. Then I can just change the names directly, no need to write all of that shit (that I did like a dozen times) out.
Changing eth1
to eth0
for the first card and vice versa for the second gets the thing to work. Well, hopefully it will stand the test when my father gets back.
///
I found this from 1.5 years ago (3 journal files back.)
Let me write some of this down on a piece of paper. It really is a good thing I wrote this down 1.5 years ago otherwise it would have been hell to figure it all out again.
4:10pm. Ok, I have the tabs open on the laptop. Let me switch again.
4:30pm. Ok, I got the rules
file to generate and can rename the interfaces. There are 3 so I am not sure what to rename. Also I've lost access to the internet now that the rules
file has been generated. I tried plugging the LAN cable into the other two cards, but I can't get anything to work.
I've decided to just wait for dad to get back and have a look. Setting up Internet connections is literally his job, so he should have some insight on this.
4:35pm. Now that I've gotten to this point, I think the last step of making the hotspot work should not be too hard. Let me get back to drawing.
I had some fun tracing Milim, but I've decided that this is too easy. Sure my technique needs work, but it is just a matter of time until I draw everything.
Learning to draw this is not really a challenge.
Let me do what that painter guy suggested and try drawing my hand. If I can be successful at drawing with an intermediate step, instead of just tracing things out on the screen then I can be sure that I exceeded my old self in drawing skill. I'll put my hand on the table in front of me and try drawing them.
The hardest part for me would be the outline.
4:45pm. I drew a very rough outline using straight lines. But how do I improve on this? If I was tracing straight over the reference, getting this right would be a lot easier.
5pm. There is an idea I want to try. The way I drew the outline was using very broad lines, but maybe I should have done the opposite. I should have put down a large number of paths that I could potentially manipulate later. This is similarly to how in sketching one would put down a large number of lines and strengthen the one that matches. Once I've gotten it to match I could always smooth it out later.
And instead of painting my own hand, I found an image of a real left hand online. I'll use that as a reference.
I am not skilled enough to try and draw anything from still life yet. Ok, first of all, let me just try tracing over it.
5:05pm. Right now I am using short straight lines to draw the outline instead of long curves. The result feels quite accurate.
5:20pm. Done. Still this is just tracing. Imagine that I did not trace right over the image, but had it somewhere on the side.
Could I get the proportions right in that case?
5:30pm. I drew a small part of the thumb, and it is off. Like by about a milimeter, but it would be noticeable in the final product. I actually could not see it until I traslated the object over the image. It is really difficult for me to draw the outline even though I have things side by side. As I keep drawing I get the sense that something is off, but I do not know how to correct for it.
If only I could get the very broad outlines correct first.
Let me focus on that, just like when I drew my own hand.
Hmmm, putting small lines is a different matter than putting many large ones.
Maybe the question I should be asking myself is how to sketch in a vector program?
5:40pm. This is why I was so bad at art as a kid.
When I tried drawing from still life, I'd start by drawing a line. Being what it is, no matter how much I focused would be off by a little bit. I'd use the endpoint as an achor and iteratively draw more lines based on that. The end result would be greatly off in proportions as a result of the errors compounding.
5:45pm. I do not really know how to correct for errors as I draw based on a reference.
5:50pm. https://www.youtube.com/watch?v=6hRQYy7s7f0 Tiger Illustration Process - Sketching on Photoshop and Coloring on Adobe Illustrator CC - Speed Art
Let me watch this while I think about it.
6:05pm. Let me watch some sketching tutorials specifically in Illustrator. This is the area where I am the weakest. I do not want to completely rely on tracing existing stuff.
https://youtu.be/_Ocaa2k9m5c?t=177
The way he does this is interesting, but isn't there a way of doing this without having to rely on a tablet?
6:30pm. I am thinking about it. The reason why I suck is because I try to do everything in a connected fashion right off the bat.
I mean, I think I knew about sketching as a kid, but I did not see the point of scratching a bunch of lines on the paper first. It still came out wrong for me either way.
6:35pm. Right now, suppose I tried doing a hand, but instead of doing it in a connected fashion, I try isolating all the pieces out first, kind of like Lego blocks for the later product.
Right now even if I get the proportions a little wrong, everything ends up being a mess.
But I can't really just scrible things in the Illustrator either since I am using a mouse. I need to first discover a novel way of sketching. After that I should focus on aligning and scaling things properly.
7:10pm. No nevermind this. It is too hard. I have no idea how to deal with compounding of minor errors. Forget beating my old self. Being a good artist matters not even a little.
7:45pm. Let me put down a few words regarding the hotspot issue and then I will close. Dad took a look at it and managed to connect to the Grase app using the laptop via LAN cable, but I now have the issue that the Internet is not working since I made the rules file. Sigh, I did not have this issue last time.
It just stopped working since the interface rename. I'll have to take a look at my notes more closely. I'll leave that for tomorrow. Right now, let me play Bloodstained.
It is not as good as Ender Lilies or Hollow Knight.
Ender Lilies just set too high of a standard for me. Hollow Knight while good feels too much like a flash game, Ender Lilies had the perfect atmosphere. Bloodstained feels like MMORPG in places.
I am not a kid anymore, so I am not into doing fetch quests or kill quests. I hate how Vigil foists those on me when I just want to kill monsters."