Back after 8 months. Some thoughts as a PS5 player.

1. It has levels to know where you're at. I dropped off kind of pre Rennala in an NG+ (I know only because I left a note in an Obsidian file). I get and agree why the game and the genre has the lore doctrine it has, but obscurity by design fights against just picking things up again. I feel like this is why online discussion and support is a core part of playing and understanding the game meaningfully. I believe there are folks who really do figure things out by axiom, whether its completion paths, combinations, stacks, buffs, scaling, builds, hidden areas, while believing it's not a game philosophy that works without an internet, and to be frank without the highly centralised iteration of the internet we have today. I think we should encourage players to look at guides from the get over just do it yourself alone. In much the same sense we don't tend to recommend people experience a country or a city without a bit of groundwork.
2. It's a beautiful, beautful game. Even more than I remember or appreciated. I don't just mean the art style or that ray tracing was added. It's things like level of detail, extreme use of color, particle effects, and especially, high contrast. It's all very painterly, but brush brave and stroke precise in a Sargent/Rubens/ElGreco sense, or in the sense of how great mini painters emulate insane realism in 25mm. Some main things I don't love are the washed out distance fog. I get it, but it creates a flatness than undermines (imo) one of the most beautiful, awesome, worlds ever designed in video games.
3. Menus could make it an even more beautiful game. I'm left wondering the extent FromSoftware understand what some of us are doing when we play this game and spending time in the world. Me? I'm spending a lot of time in menu and using the hud constantly. The tea drenched sepia smol boxes does little good here. I would love for the menus and hud part to get the same level of game attention as core mechanics. I feel like this is the more important criticism than saying menus are spreadsheety. Ans seriously, things are just too small. The items I'm looking at have beautiful, intricate designs, with such lore. They deserve more visual real estate and even animation (why not play the gesture or provide a show action when you select it?). I also feel like item cycling lacks slickness compared to main combat, but both matter in fights. Subjectively, I tend to get killed when trying to pick the right flask more than timing a roll. And so I think menu is underserved as part of combat.
4. Sound design is out there. I hadn't realised the extent Elden Ring imprinted its sound on me, especially combat effects and character actions. This is Nintendo / 80s Arcade levels of imprint. I would say also the audio helped me attune to combat that much faster: the game sound literally tells you if you're playing and timing well. The VAs, to be clear, are excellent imo, and I wish we knew more about them and celebrated them more as actors the way other games are doing (I'm there for the Ramon Tikaram as Godrick podcast), but the direction and writing is missing—something—somehow trapped in the era that Demon Souls was made. That said, I just did come back from months of BG3, which is the high bar here.
5. Physicality. I think the audio effects can't be underestimated to help lend weight, but when you try and swing that blasphemous/gargoyle two hander thing you have going on, you really do feel their heft, whereas when I was a moonveil spammer on my first run, it was whip fast. While being a tad critical of how the game hides its mechanics, they do matter, impressively so, in the sense of gameplay. That glove vs that glove can be night and day. And what makes this amazing is how those choices really affect your overall game approach. It's so meaningful and it feels like you're playing a different game each time. And that leads me to—
6. —Length. I'm torn. On the one hand, in the last decade,I'm playing and loving games that lean to longer play throughs (AC:O/A, HFW, BG3, Elden Ring, RDR, Rogue Trader, at a push TLOU2). On the other hand, Dungeons, Field bosses, etc, start to get repetitive and even gimmicky (oh, this rune bear also has shout, ok then). I wonder if there's an Elden Ring that cuts the game to a half or even a third such that I could play through more times. BG3, the game I'm coming off, to be clear, has the same problem, even worse, but also the same potential. I feel your build choices make the game spectacularly different, not just minimaxing on a load out. Playing as a mage over a knight over a samurai means playing a different game. This evening for example, I ended up by accident beside a crayfish in Liurnia. I had to fight, freaking out, but because of the current build a toe to fight was a reasonable thing to do. The freak out was they were so formidable in other playthroughs—because of the builds. So I'm left wondering, would a shorter game I complete more times be 'better' than a longer game I complete fewer times?

So: yeah, I'm thinking I'm back. One of the greatest games ever made.

I've recently completed my third year of computer engineering at UofT, and I wanted to share my thoughts on the courses I've taken. I hope this will be beneficial to incoming first, second, and third-year engineering students. My area of depth is 6 (Software) and my areas of breadth are 4 (Control) and 5 (Computer hardware/networks).
Difficulty: How hard it was to comprehend course material and/or complete practice problems accurately. Keep in mind that difficulty is subjective but I tried to eliminate personal bias and be as objective as possible.
Workload: The amount of work to complete outside of lectures (labs, assignments, homework, studying for tests/exams).
All engineers take the exact same courses in their first year first semester and similar ones in second semester. I took these courses in 2021 during covid so some information might be outdated.
First Year Semester 1:
APS100 Orientation to Engineering:
Difficulty: 1/10 Workload: 3/10 Course Average: N/A
There is no technical material in this course. All coursework consists of Ethics analysis, creating a schedule for time management, creating your resume, and discussion board participation. During tutorials, the TA typically asks behavioral questions to the class and covers slides pertaining to engineering principles, the learning environment, test readiness, and related topics.
APS110 Engineering Chemistry and Materials Science
Difficulty: 6/10 Workload: 5/10 Course Average: B
This course blends elements of chemistry, physics, and mathematics, requiring lots of formulas and memorization. The assignments and online quizzes were moderately difficult. Not sure if he’s still teaching but Scott Ramsay was a really good professor. Really enjoyed watching his video demonstrations.
APS111 Engineering Strategies and Practices I
Difficulty: 3/10 Workload: 6/10 (Variable) Course Average: B
Students are organized into groups of 5-6 and are assigned the task of developing a solution for a client, following the engineering design process. Success in this course heavily relies on your team. Throughout the semester, there are several milestones where you need to submit large reports with your team, making it crucial to have capable teammates. Many teams resorted to last-minute cramming the night before each milestone (not a good idea). Additionally, there are two peer reviews, so make sure you contribute to avoid being perceived as an unproductive member. In most cases, you don’t need to worry about this though. You’d likely all be good friends by that point (and give each other a perfect review).
CIV100 Mechanics
Difficulty: 9/10 Workload: 9/10 Course Average: C+
The consensus among most students in my year was that this was the toughest first year course. It's incredibly calculation heavy, with questions that may appear straightforward in theory but require multiple steps to solve. Even a minor error can screw up your entire calculation process. The assignments are extremely time-consuming, and the quizzes pose significant challenges. I suggest you work in a group for the assignments. Furthermore, both the final exam and midterm were exceptionally difficult.
MAT186 Calculus I
Difficulty: 4/10 Workload: 4/10 Course Average: B
If you’ve taken IB or AP this course will be a breeze. Even without that background, it's still relatively straightforward. Very little on integrals, mostly derivatives, rates of change, and graphs. PCEs and WebWorks were trivial, and the midterm and exam were easy.
MAT188 Linear Algebra
Difficulty: 7/10 Workload: 7/10 Course Average: B-
The majority of the course revolves around relatively straightforward operations on matrices and vectors. However, it does introduce abstract concepts like higher-order dimensions, which may require some deeper thinking to fully grasp. Webwork assignments were slightly more challenging compared to MAT186's, and completing assignments took a while. Once again, you should work with a partner or as a group. Pay attention in the MATLAB practicals because you’ll be using MATLAB pretty often in future courses.
First Year Semester 2:
APS105 Computer Fundamentals
Difficulty: 6/10 Workload: 5/10 Course Average: B-
If you’ve taken programming in high school, you will likely find this course very easy (2/10 difficulty). The course covers simple programming syntax, along with basic sorting algorithms, without delving into more advanced concepts beyond linked lists and binary trees. I think that this was one of the more enjoyable courses of first year.
APS112 Engineering Strategies and Practices II
Difficulty: 5/10 Workload: 7/10 (Variable) Course Average: B+
This course is essentially a continuation of APS111, but with the added component of being assigned an actual client with a real-world problem statement. Each group works with a unique client, and at the end of the term, you present your final report directly to them, along with your supervisors. As with APS111, the quality of your team plays a crucial role in your success. Another thing to note is that unlike APS111, there were multiple individual assignments to complete on top of your team reports.
ECE110 Electrical Fundamentals
Difficulty: 8/10 Workload: 6/10 Course Average: C+
Definitely one of the harder first year courses. This course marks your introduction to circuits. Topics covered include capacitors, Ampere's law, Ohm's law, Kirchhoff's laws, AC/DC circuits, magnetism, Thevenin and Norton equivalents, and more. The material is extensive and requires thorough understanding, as many future courses will build upon its concepts. Make sure that you understand the content well, given its foundational importance in subsequent courses.
ECE191 Introduction to Electrical and Computer Engineering
Difficulty: 0/10 Workload: 1/10 Course Average: N/A
Seminar course. You just need to show up to the lectures (one per week) and submit a 100-ish word report about what you learned.
MAT187 Calculus II
Difficulty: 6/10 Workload: 5/10 Course Average: C+
The workload in this course is comparable to Calc I, but the content is more challenging (obviously). It delves deeper into integrals and introduces concepts like ordinary differential equations and polar coordinates. But again, you’ll manage just fine if you did decently in Calc 1, and especially if you’ve taken AP math or IB HL math.
MIE100 Dynamics
Difficulty: 7/10 Workload 4/10 Course Average: C+
I personally found this to be the most challenging course in first year, but it seems that most people disagree with me. It shares similarities with CIV100, but you’re instead dealing with dynamic systems (moving) rather than static ones (not moving). Some concepts may prove difficult to grasp initially. However, if you focus on memorizing the relevant formulas and understanding when to apply them, you should be okay. Expect an emphasis on physics in this course.
Second year was the worst (for ECE) and a very big step up from first year. If you struggled in first year, you better lock in for second year.
Second Year Semester 1:
ECE201 ECE Seminar Course
Difficulty: 0/10 Workload: 0/10 Course Average: N/A
Another seminar course, except you just need to attend 7/10 lectures to pass. Most students just scan their Tcard at the entrance then leave right after. The lectures are pretty useless except the Magellan one where they show you how to choose courses for 3rd and 4th year.
ECE212 Circuit Analysis
Difficulty: 7/10 Workload: 8/10 Course Average: B+
This course provides a deeper exploration of circuits, extending beyond circuit solving to include frequency and sinusoidal analysis. Questions tend to be long so careless errors will cost you. Personally, I found the labs challenging, particularly because they required working with oscilloscopes at a higher level than I was used to, and I could not for the life of me figure out how to use that thing.
ECE241 Digital Systems
Difficulty: 7/10 Workload: 8/10 Course Average: B-
This hardware course focuses on the foundation of hardware. A lot of the content is at the fundamental level so it’s not too hard to comprehend. However, the main challenge lies in learning Verilog, a hardware description language. I hate Verilog. All the labs are done using Verilog, and frustratingly, the course provides no instruction on the language, nor are there many helpful resources available online. This lack of guidance made what should have been straightforward labs much more difficult. Thankfully if you are in CE, you’ll never need to take another course with Verilog again. The midterm was easy but the final was borderline impossible.
ECE244 Programming Fundamentals
Difficulty: 7/10 Workload: 7/10 Course Average: B-
This course is the sequel to APS105, delving into more advanced concepts like pointers, constructors, polymorphism, and inheritance. Many of these topics may feel familiar if you have prior programming experience, particularly from high school. Consequently, if you're already proficient in programming, as many computer engineering students are, you should be able to navigate through this course well. The labs were generally manageable, although a couple of them were time-consuming.
MAT290 Advanced Engineering Mathematics
Difficulty: 7/10 Workload 8/10 Course Average: B
In this course, you’ll learn a lot of miscellaneous mathematical concepts that will be relevant in future courses. The goal is to establish a strong foundation to facilitate understanding in third and fourth-year courses. There is also a quiz every week. They were terrible from a time management standpoint, but as long as you spent 3-4 hours per week completing the assigned textbook questions, you should be alright since one (out of two) of the questions on the quiz is guaranteed to be an assigned textbook question (at least, that was the case when I took the course).
MAT291 Calculus III
Difficulty: 9/10 Workload 9/10 Course Average: B-
The entire workload comes from studying for those damn weekly quizzes, much like MAT290. Personally, I found it to be one of the most difficult courses I've ever taken. Think of calculus 2, but now everything is in three dimensions, encompassing concepts like surface integrals, flux integrals, and more. Additionally, the course introduces abstract topics that I still struggle to grasp. Balancing the demands of this course alongside everything else was an incredibly stressful experience.
Second Year Semester 2:
ECE216 Signals and Systems
Difficulty: 8/10 Workload:7/10 Course Average: C+
I think everyone hated this course. It was so poorly taught and the midterms felt haphazard and unpredictable. I relied on the power of BS to pass (worked surprisingly). Other than working with many graphs and MATLAB, I don’t remember much about this course because it was so uninteresting.
ECE221 Electric and Magnetic Fields
Difficulty: 8/10 Workload: 5/10 Course Average: C+
This course serves as an introduction to fields, while also expanding on the principles covered in ECE110. Many concepts build upon intuitive ideas, such as the repulsion of identical charges and attraction of opposite charges, which can make seemingly complex topics less intimidating. There are also many procedures that you need to know to solve specific types of questions. The labs are generally straightforward. You just need to follow the steps on the lab worksheet and answer a few simple questions from the TA.
ECE231 Intro to Electronics
Difficulty: 9/10 Workload: 7/10 Course Average: C+
This course is basically circuit analysis but significantly harder. It combines everything you've learned about circuits thus far, introducing new topics on top of that foundation. There are so many new concepts to learn and the final pretty much tests you on all of them. I’ve encountered multiple scenarios where I saw two concepts that I thought were completely separate from each other being integrated into a single question on an exam. Also very calculation heavy. Despite my experience in this course, I must admit Khoman Fang was a great professor.
ECE243 Computer Organization
Difficulty: 7/10 Workload: 6/10 Course Average: B
This course is a direct continuation of ECE241, and surprisingly, I found it a lot more enjoyable than its prerequisite. The focus is on Assembly language programming, which I found to be way more understandable compared to Verilog. Assembly operates at a fundamental level, so you’re technically working with the most basic building blocks in programming. The labs often involved seemingly simple tasks like counting to 10 on a HEX display, but the nature of Assembly meant that achieving this could require hundreds of lines of code. I appreciated these labs as they provided a refreshing change from the usual programming tasks. Final and midterm exams were relatively easy.
ECE297 Software communication and design
Difficulty: 8/10 Workload: 9/10 (Variable) Course Average: B+
Despite the demanding workload, this course was one of my favorites. The main project involved collaborating in teams of three to develop a geographical information system (similar to Google Maps). Unlike other courses where teams are assigned, here you have the freedom to choose your own teammates. Starting off, your team is given only the OpenStreetMap database, and the APIs, granting you complete control over your project's design, functionality, and optimization. Performance enhancement and pathfinding algorithms were key components of the project. It's crucial for all team members to contribute effectively for success in this course; And trust me, this is not a one or two person project. To succeed in this course, all three members must be doing their share.
Third Year:
They say that third year is the easiest year for ECE, but that depends on what courses you choose. For me, it was only slightly easier than second year but still significantly harder than first year.
APS360 Fundamentals of Deep Learning
Difficulty: 8/10 Workload : 8/10 Course Average: B+
This course felt fundamentally very different compared to the other programming related courses. If you’re unfamiliar with deep learning, the content can be overwhelming but the inner workings behind AI are both fascinating and challenging to grasp. Fortunately, much of the coding in the course revolves around preparing data, setting up AI architectures, and training/testing models, rather than building an AI from scratch. While the labs are manageable, the final project can be very time consuming, particularly as training a single model can require hours or even days. Regarding this aspect, make sure you start early. Both the midterm and final exams were quite difficult.
ECE302 Probability and Applications
Difficulty: 7/10 Workload: 3/10 Course Average: B-
The first half of this course is very straightforward. Just an introduction to probability. The second half is where the difficulty quickly ramps up. Make sure you don’t underestimate the second half of the course. I made the mistake of assuming the material was easy after performing well on both midterms (midterms were very easy), leading me to neglect the final third of the course so I flunked the final. FYI, our marks were determined solely by the 2 midterms and the final.
ECE344 Operating Systems
Difficulty: 8/10 Workload: 10/10 Course Average: B
This course had the most difficult and demanding labs I’ve ever seen in my life. They are due every two weeks and some of them took most students more than 15 hours to complete. I’ve seen people staying up all night in the GB computer rooms to finish these labs. One positive aspect of the course if Jon Eyolfson is your professor, is that all lectures are both recorded and live-streamed, allowing attendance in person, online synchronously, or asynchronously. But if you have Ding Yuan, good luck. I heard he is not only bad at teaching, but also makes you program an operating system from scratch. The course content is quite interesting and the Midterm and final exams were fair.
ECE345 Algorithms and Data Structures
Difficulty: 8/10 Workload: 7/10 Course Average: B+
If you're one of those programming geniuses who have been doing Leetcode since like 3 years old, you’d have likely already learned all course content before you’ve even started the course. You could probably write midterm and final exams without attending any lectures. However, if you're not as experienced, attending every lecture, taking detailed notes, and staying focused is crucial. Personally speaking, some algorithms were incredibly difficult for me to grasp, while others were simple. Assignments typically include both coding and long-answer components, which can be time-consuming, but you can work with a partner.
TEP444 Positive Psychology for Engineers
Difficulty: 1/10 Workload: 3/10 Course Average: A-
This course overall was very enjoyable and a nice refreshing experience from the other technical courses. You go on excursions to various locations with your group and write reports. It's an excellent opportunity to expand your social circle and make new friends. If you aim for an A- grade (80-85), you can breeze through with minimal effort. However, achieving a higher grade (A or A+) would require significantly more work (like 5 times more, no joke), which is not worth the extra effort in my opinion.
CSC343 Introduction to Databases
Difficulty: 7/10 Workload: 10/10 Course Average: Unreleased
The content of this course is actually pretty interesting and is taught well. But the assignments were excessively time consuming. There are only three in total, but each one demanded such a significant investment of time to the point where I contemplated dropping the course altogether. You do work with a partner (going solo isn’t even an option if you want to complete the assignments) but even so, be prepared to invest 15-30 hours per person, per assignment (they also mark hard). Despite the workload, I do recommend taking this course because knowing SQL can substantially benefit you in the future. Just start the assignments early. On a positive note, the midterm and final were relatively straightforward.
ECE311 Introduction to Control Systems
Difficulty: 10/10 Workload: 5/10 Course Average: C+
This course didn't come with a heavy workload, consisting only of three labs and two midterms. However, I found the content to be very challenging. It felt like they condensed and combined every concept from every engineering math and physics course into one. Topics ranged from circuits and dynamics to Laplace transforms, matrices, Bode plots, and more. It didn’t help that Prof Scardovi didn’t record any lectures, has messy handwriting, and uses a damn chalkboard instead of modern technology like other profs do. The midterms and final were somewhat lenient compared to the lectures, but this course still ended up being my lowest mark.
ECE361 Computer networks I
Difficulty: 8/10 Workload: 9/10 Course Average: B
This course requires a lot of memorization compared to other courses. You’ll have to understand calculation processes but also memorize a lot of general knowledge about networks. A problem with this course is that some concepts that show up in tests are not adequately explained in lectures. By far the best way to study is to do past exams. Professor Valaee often recycles questions, sometimes even directly copying past midterms and finals (like he did in my cohort). Unlike other courses, there is no big assignment/lab for this course, but it makes up for it with quantity. There’s a quiz every week, a wireshark lab every 2 weeks, a programming lab every 2 weeks, and a midterm. The programming labs are already difficult by themselves, but this cumulative workload felt very overwhelming.
ECE472 Engineering Economics and Entrepreneurship
Difficulty: 7/10 Workload: 5/10 Course Average: B+
This is straight up an economics course, no engineering at all. You’ll learn about investments, equity, financing, and related areas. There are 4 quizzes throughout the semester, with quiz 3 (quiz 3 specifically) being by far the hardest (average in the 50s). Since you're able to bring your textbook to the quizzes and exam, my biggest tip is to write past quiz and exam solutions in your textbook, as there's a limited pool of questions, and new questions often resemble some kind of variation of past ones.
ECE421 Introduction to Machine Learning
Difficulty: 9/10 Workload: 7/10 Course Average: unreleased
While the course content itself may be challenging, surprisingly, the midterm and final exams were pretty easy. There's a slight discorrelation between the lecture material and what appears on the exams, but relying on past exams for preparation should be enough. The course leans heavily on mathematical concepts, much more so than APS360. The assignments were difficult and tedious but I recommend you work with others to complete it.
With that, I've covered my thoughts and tips for all the engineering courses I've completed. Transitioning into your next year in engineering, particularly if you’re transitioning from high school, can feel daunting. However, remember that you're not alone in this experience – many of your peers are facing similar challenges. If you need any more advice or have questions, feel free to reach out to me via DMs!

I wanted to share my detailed experience with the Kia EV9, which I purchased on January 3rd. I had eagerly anticipated this vehicle for years, and it topped my list due to my admiration for the Kia Telluride, even though I never owned one. I appreciate the direction Kia has taken over the past decade, and I was excited to support a company other than Tesla. I previously owned a Ford Mustang Mach-E, which I traded for the EV9, primarily for its larger size and superior features compared to the EV6, which I still own. Although the EV6's interior felt cheaper than the Mustang, it excelled in nearly every other way.
For some context, I purchased the very first EV9 in Vermont, making me the only owner in the state at the beginning of the year.
Here’s a detailed look at my service history with the EV9:

1. Climate Control Screen Issue:
   • Problem: The climate control screen intermittently failed to display anything. Disconnecting the 12-volt battery did not resolve the issue.
   • Solution: The dealership replaced the entire screen unit, a process that revealed the cost of this part to be about $4,000, though it was covered under warranty. This issue seems to be relatively common based on other forum discussions oai_citation:1,KIA EV9 Issues Kia EV Forum oai_citation:2,Error Message: Check Tire Pressure Monitoring System Kia EV Forum.
2. Back-Right of Driver's Seat Movement:
   • Problem: I felt movement in the back-right of the driver's seat during left turns, which was difficult to isolate.
   • Solution: The first dealership couldn't find the issue, but the second one did. It took about two months to receive and install the seat-track assembly, which resolved the problem.
3. TPMS System Warning Light:
   • Problem: A warning light indicating that the TPMS system needed to be checked kept appearing. Initially, the TPMS sensors were communicating correctly, but the issue persisted.
   • Service Visits:
     • First Visit: The dealership verified that the TPMS sensors were properly communicating and ruled them out as the issue.
     • Second Visit: They removed the negative and positive terminals from the battery, connected the leads to each other (thereby removing the battery from the loop), and created a complete 12-volt circuit to discharge any stored power in the car’s capacitors. They left it this way for 20 minutes, which caused the error to go away temporarily.
     • Follow-Up: I avoided a third visit by speaking with the technicians over the phone. They submitted a follow-up case to Kia TechLine, which recommended replacing the Body Dynamics Module (BDM). However, Kia placed a freeze on shipments of this part due to a known defect. This means it could be weeks or months before a replacement is available as the part needs to be potentially redesigned and shipped from Korea to the United States. In the meantime, I found that removing the black wire from the 12-volt battery for a few seconds can reset the issue temporarily oai_citation:3,Kia EV9 Forum Kia EV Forum oai_citation:4,Error Message: Check Tire Pressure Monitoring System Kia EV Forum.

Despite these issues, I absolutely love the EV9. The climate control switches, infotainment system, phone integration, seats, and seven-seater layout are all fantastic. It's frustrating to see a warning light on the instrument cluster of a car I've desired for so long, but I'm hopeful for a permanent fix soon.
For context, my experience with the EV6 was very similar. It had a bunch of issues during the first few months of ownership, but since those were resolved, it has had no problems.
I find these issues completely unacceptable in a brand-new vehicle. However, at this point, Kia has put in around $6,000 to $10,000 of work into this one car. They have really worked through things quickly. I suspect most EV9s out there have at least one of these problems, if not all three. So getting it replaced and going with a lemon law situation isn't really in my best interest, since I would just purchase another one of these vehicles anyway, and then that one might have these problems. The dealership, and particularly the service center, said that I was a real early adopter, to which I responded that I was just dumb for buying two EVs, well, three if you include the Mustang, very, very early in their release cycle—realistically, only a month after they'd been released. So the next time I buy a car, I'm probably going to wait. Although, I will probably buy another EV9 when I turn in the four-year lease on this one. By then, I suspect they may even have a refreshed, facelifted version.
Long story short, I love this car. I wish it didn't have these initial problems. But alas, it is literally the perfect car for me, and I want it to be perfect. And it will be. Plus, Kia Customer Service plans to compensate me generously for my troubles, as they did for my EV6.

9 months ago I did anabolic steroids for 4 months first cycle ever but it gave me rosecea I work outside now when I come in after a day of work my skin is red on my face and hot as fuck and doesn’t stop for a bit wtf do I do. Always had pale skin it goes back to my normal skin tone if I haven’t been out in a weather element for a while but when I do and get back into regular temperature my face gets red and hot and it’s driving me fucking bad shit idk wtf to do

This is going to be a long post, so please bear with me. For starters, this is my second time applying to PA school--I was at a disadvantage last cycle, as I applied late July, almost early August. I was working towards my masters degree during that time, and had a bunch of in-progress courses that I need to retake. I wasn't the best student my sophomore year of college, and Organic Chemistry I was the death of me, had to retake it and passed on my third try. I have shadowed a surgical PA for a few months, but because of the pandemic, the hospital was not allowing any students to shadow anymore. I like the idea of PA's bridging the gaps in healthcare, as with everyone. Although lateral mobility and good work-life balance are enticing factors in being a PA, I don't see that as being true. It really is specialty-dependent, an ideally, I think is best to stick with one speciality and perfect your craft. I don't mind the level of autonomy that comes with being a PA, I feel that everybody can be a leader within their own scope of practice.
However, amidst this journey of applying to PA school, during college, and even high school, medical school and being a physician has been nagging in the back of my mind. I come from two loving parents who never pressured me to pursue a career in medicine, but something that will make me satisfied and have a good life. Being a physician means having the breadth of knowledge, thoroughly understanding the mechanism of actions in medication and diseases. Its better to know and comprehend the whole story and the why, instead of just knowing surface level information. I'd like to think being a research assistant is helping me in solidifying my decision to be a physician--I understand that its two completely different things, but I enjoy showing medical students and other graduate students how to dissect a mouse, I love teaching them the anatomy of the mouse, and what each part is special in, and how it pertains to the research that I do. I love being the leader, and I appreciate when my colleagues come to me for questions and ask for my perspective in different things. I enjoy being a leader, yet I enjoy working in a team. For reference, here are my stats, they are not the greatest, but I'd like to think my experiences showcases that:
-graduated college with a 2.943 gpa, had to retake most of my sophomore year's worth of classes, failed organic chemistry I and passed on my third try. with all retakes and masters, raised to 3.02. despite failing most of soph year, i made deans list numerous times.
-got my masters in biomedical sciences, got a 3.4 gpa
-last 60 credits, 3.66, last 45, 3.61
-1955 hours as a night shift float pct, certified in phlebotomy, ekg, and as a pct.
-1392 hours in biochem research, vitamin D regulation focused. abstract sent out, will be presenting in symposium in fall
-new job as mental health specialist on back up call center, about 60 hours since i just started position
-990 hours in social and affective neuroscience research, presented thesis to psych department
-volunteering: sunday school teacher for 2nd graders, habitat for humanity, health screener in grad school, volunteer in hospital in high risk pregnancy and geriatric wings, summer camp counselor in church, food bank.
-teaching experience: learning assistant for psych stats course, helped teach 37 juniors and seniors as a junior, about 42 hours bc during semester.
-hobbies: food festivals, legos, parrots, painting, cooking and baking.
I have not taken my MCAT, and I told myself that if I do not get far in PA school applications, I will start studying for the test come the fall. The masters program that I am apart of has a linkage with the medical school, so I will try to see what my options are when the time comes. All in all, both professions are excellent, but I don't want to spend the rest of my life wondering what could have been. Some honest advice and feedback would be helpful.

Open-source GitHub Repo Paper Describing the Process
Aside: If you want to take the course I did online, the full course is available for free on YouTube.
When I was a graduate student at Carnegie Mellon University, I took this course called Intro to Deep Learning. Don't let the name of this course fool you; it was absolutely one of the hardest and most interesting classes I've taken in my entire life. In that class, I fully learned what "AI" actually means. I learned how to create state-of-the-art AI algorithms – including training them from scratch using AWS EC2 clusters.
But, I loved it. At this time, I was also a trader. I had aspirations of creating AI-Powered bots that would execute trades for me.
And I had heard of "reinforcement learning" before.. I took an online course at the University of Alberta and received a certificate. But I hadn't worked with "Deep Reinforcement Learning" – combining our most powerful AI algorithm (deep learning) with reinforcement learning
So, when my Intro to Deep Learning class had a final project in which I could create whatever I wanted, I decided to make a Deep Reinforcement Learning Trading Bot.

Background: What is Deep Reinforcement Learning

Deep Reinforcement Learning (DRL) involves a series of structured steps that enable a computer program, or agent, to learn optimal actions within a given environment through a process of trial and error. Here’s a concise breakdown:

1. Initialize: Start with an agent that has no knowledge of the environment, which could be anything from a game interface to financial markets.
2. Observe: The agent observes the current state of the environment, such as stock prices or a game screen.
3. Decide: Using its current policy, which initially might be random, the agent selects an action to perform.
4. Act and Transition: The agent performs the action, causing the environment to change and generate a new state, along with a reward (positive or negative).
5. Receive Reward: Rewards inform the agent about the effectiveness of its action in achieving its goals.
6. Learn: The agent updates its policy using the experience (initial state, action, reward, new state), typically employing algorithms like Q-learning or policy gradients to refine decision-making towards actions that yield higher returns.
7. Iterate: This cycle repeats, with the agent continually refining its policy to maximize cumulative rewards.

This iterative learning approach allows DRL agents to evolve from novice to expert, mastering complex decision-making tasks by optimizing actions based on direct interaction with their environment.

How I applied it to the stock market

My team implemented a series of algorithms that modeled financial markets as a deep reinforcement learning problem. While I won't be super technical in this post, you can read exactly what we did here. Some of the interesting experiments we tried included using convolutional neural networks to generate graphs, and use the images as features for the model.
However, despite the complexity of the models we built, none of the models were able to develop a trading strategy on SPY that outperformed Buy and Hold.
I'll admit the code is very ugly (we were scramming to find something we could write in our paper and didn't focus on code quality). But if people here are interested in AI beyond Large Language Models, I think this would be an interesting read.
Open-source GitHub Repo Paper Describing the Process
Happy to get questions on what I learned throughout the experience!

The Sonic franchise had a large gallery of villains, both the original saga and the Archie Comics and IDW comics, there are many to choose from to adapt them in the movies or series of the Sonic cinematographic universe.
Let's talk about what villains I could or would like to see in the future of SCU
One of my favorite things about the Archie and Idw comics is their excellent variety of villains outside of Eggman.
in Archie: we have warlords, wizards, dark gods, alternate universe doppelgangers, aliens, a corporate mastermind, and then some.
IDW: we have two traumatized cyborgs and a group of criminals
Scourge :I love because, well, could anyone resist taking such an inherently good character, and turning him evil? I love the dynamic he and Sonic shared, being able to get into each other's heads, but ultimately staying themselves. It is true that it would be super repetitive to have a hedgehog that is an evil counterpart to Sonic, let's go (let's have Shadow) of course it can make Scourge see that unlike Sonic, he was not afraid, even so he used his powers to obtain power and control in his planet it would be perfect to see more hedgehogs than just being limited to sonic, amy, shaodw and silver
dr Finitevus :I always loved him; he wasn't overwhelmingly powerful, although he had unique abilities and intelligence. It was his knowledge of the world that allowed him to pit the world against the heroes or the heroes against each other. When he didn't need to be somewhere, he just wasn't there. He fought proxy wars by recruiting other characters to, knowingly or unknowingly, carry out his orders. It was a refreshing change of pace from Sonic's usual bad boy.
It will be an excellent way to get more out of the echidna lore, what a bad taste the knuckles spinoff left in our mouths, this tribute already deserves to adapt its entire arc in the comics to the SCU since it would be great to see more echidnas and give it a try. the treatment you deserve
Enerjak: a character full of energy and creativity with a cool and creative design, I always liked this character and he generated some pretty good plots and conflicts. It would be another option along with finitrevus to be a personal threat to Knuckles and the others on how Knuckles should confront one of his species and prevent him from following the steps of going astray.
Mammoth Mogul:I like his powers, his suit and his elegant way of speaking. But I don't think he has much personality. and he is conceited, he keeps escaping from justice and causes problems for Sonic, a mafia lord would be perfect in a detective plot either with the chaotix or with sonic it is also related to the chaos emerald of course they could change this that He looks for them because they are a legend
metal sonic :Its design needs no explanation.When he is in control, he can be as smart as Eggman and as terrifying as one of Sonic's god-level villains.
His rivalry with Sonic has a lot of depth. Thematically they are natural rivals of the franchise, nature versus machine, but also in terms of characters, one is a free soul and the other is in a prison of multiple things.
Metal has very personal problems derived from his rivalry with Sonic. His identity is built around the idea that he is the best version of Sonic, he is the real Sonic, but every loss shows him otherwise and all he can do is face it and try to say that Sonic is the copy and Metal is the real one. . But no matter how strong he is, how fast he is, even if those traits are more than what Sonic is capable of, he loses. That's the tragedy of him, he's in a cycle of not accepting reality and he's punished for it and so far we've gotten great ways to show that character. We got healthy parts of his character, scary parts, interesting parts and sad parts.And the good thing is that, unlike other rivals, he is a villain, not an antihero or a redeemed character, a villain who can be explored without being the good guy.He's also had some amazing boss fights over the years.This would be a perfect character to take on Eggman's legacy and use it for your own benefit.
Dr. Starline: Amazing villain that you love to hate, he played his role to perfection and left lasting effects long after his death in the form of the equally good Surge and Kit. What I like most is that he could have easily been like everyone the other sidekicks and try to one-up Eggman, but instead he still idolized him and was ultimately used to cement how good Eggman is as a villain.
He would be another perfect candidate to take Eggman's place and he is a thousand times better than Agent Stone or any generic Eggman assistant who only exists to provide a moment of comedy that is not funny.
The metarex and the dark oak (sonic x)
I think the writers did a great job creating a new arc for sonic x
It's a fun sci-fi take on Sonic that gives us a galaxy-spanning adventure. You still have individual episodes that are reminiscent of your weekly episodes of Star Trek or whatever that have their own independent stories on individual planets, but also the overall story of the season is still moving forward. I feel like this is the best way to do episodic television.
I love the idea that we have a planet full of animals, but in Sonic's universe there are also plants.
The plot twist of Metarex being formerly plant people (Seedrians, if I remember correctly) turns them from this vague threat into a completely personal threat, as we've had plenty of time up to that point to see Cosmo become part of the Team Sonic. .
It's always good to have that underlying message that plants and animals can get along despite their differences. In today's world, many people have yet to discover the real-life equivalent.
Plus, you have the entire Tails and Cosmo arc and the culmination at the end. Don't lie, you always cry too!
prefect official as the final boss of the sonic franchise
Merlina :Merlina doesn't understand that death is a natural part of life, which is why she is afraid of it. This is a real fear called thanatophobia, but it is a fairly extreme case. Anyway, after defeating her, Sonic consoles her by telling her to make life worth living with the time she has, thus resolving her internal conflict and also ending her character arc. This type of deep writing is something that Sonic Adventure to Black Knight excel at.
Well, that was all, write in the comments which villain from the Sonic franchise you would like to see adapted into SCU in the future. Leave your opinions in the comments. I would love to read them.

Open-source GitHub Repo Paper Describing the Process
Aside: If you want to take the course I did online, the full course is available for free on YouTube.
When I was a graduate student at Carnegie Mellon University, I took this course called Intro to Deep Learning. Don't let the name of this course fool you; it was absolutely one of the hardest and most interesting classes I've taken in my entire life. In that class, I fully learned what "AI" actually means. I learned how to create state-of-the-art AI algorithms – including training them from scratch using AWS EC2 clusters.
But, I loved it. At this time, I was also a trader. I had aspirations of creating AI-Powered bots that would execute trades for me.
And I had heard of "reinforcement learning" before.. I took an online course at the University of Alberta and received a certificate. But I hadn't worked with "Deep Reinforcement Learning" – combining our most powerful AI algorithm (deep learning) with reinforcement learning
So, when my Intro to Deep Learning class had a final project in which I could create whatever I wanted, I decided to make a Deep Reinforcement Learning Trading Bot.

Background: What is Deep Reinforcement Learning

Deep Reinforcement Learning (DRL) involves a series of structured steps that enable a computer program, or agent, to learn optimal actions within a given environment through a process of trial and error. Here’s a concise breakdown:

1. Initialize: Start with an agent that has no knowledge of the environment, which could be anything from a game interface to financial markets.
2. Observe: The agent observes the current state of the environment, such as stock prices or a game screen.
3. Decide: Using its current policy, which initially might be random, the agent selects an action to perform.
4. Act and Transition: The agent performs the action, causing the environment to change and generate a new state, along with a reward (positive or negative).
5. Receive Reward: Rewards inform the agent about the effectiveness of its action in achieving its goals.
6. Learn: The agent updates its policy using the experience (initial state, action, reward, new state), typically employing algorithms like Q-learning or policy gradients to refine decision-making towards actions that yield higher returns.
7. Iterate: This cycle repeats, with the agent continually refining its policy to maximize cumulative rewards.

This iterative learning approach allows DRL agents to evolve from novice to expert, mastering complex decision-making tasks by optimizing actions based on direct interaction with their environment.

How I applied it to the stock market

My team implemented a series of algorithms that modeled financial markets as a deep reinforcement learning problem. While I won't be super technical in this post, you can read exactly what we did here. Some of the interesting experiments we tried included using convolutional neural networks to generate graphs, and use the images as features for the model.
However, despite the complexity of the models we built, none of the models were able to develop a trading strategy on SPY that outperformed Buy and Hold.
I'll admit the code is very ugly (we were scramming to find something we could write in our paper and didn't focus on code quality). But if people here are interested in AI beyond Large Language Models, I think this would be an interesting read.
Open-source GitHub Repo Paper Describing the Process
Happy to get questions on what I learned throughout the experience!

I've seen the technical graph (provided on occasion by Random) that shows the transition stages of solids-to-liquids-to-gases over time/temperature/pressure, and became curious: can the overall drying time be reduced with lowehigher temperature cycling and greater pressure coefficients?
From a practical point of view, if "Person A" buys a commercial FD while "Person B" buys a small HR FD, will the freeze/drying time remain generally the same regardless of the chamber size? (let's assume the exact same amount of food in each unit and similar atmospheric conditions for a 1:1 test)
I think I understand the temperature/pressure chart enough to be dangerous, but it seems like current freeze drying capabilities do not vary enough to make a significant difference in production time whether you're a small shop or mass producer. 8 months ago Salvarius mentioned to a reader that lower and lower mTorr readings would not reduce cycle time. But would lower mTorr -AND- lower temperature/higher heating cycles affect production time at all?
Apologies if I missed an older message on this. And I understand the latest firmware also does more to measures mTorr levels and temps in order to cycle most efficiently. My question applies to the science of freeze drying, in general. Thanks!

I work at an education access program, and we are analyzing our inquiry and application data. I would like to try to see if there are any patterns to the timeline in which families interact with us. I have the date of inquiry, the date they attended an information session and the date they submitted an application. I am hoping I can map this out in a graph/chart that would show the general timeline (stretch of time) where we are having interaction points (actions taken by families), with our organization. In my head, I'm picturing something like my crude drawing posted here, similar to a gantt chart, only with ~100 lines to signify each applying family. Is this possible in excel?
https://preview.redd.it/qz8pzfts6t0d1.png?width=638&format=png&auto=webp&s=923be86bee256f99b36ada380df9879765131f30

I’ve been grappling with something for a while now, and I feel like I need to get it off my chest. I’ve always been the type of person who strives for excellence in everything I do, especially when it comes to academics. However, lately, I’ve been facing a string of failures that have left me feeling completely mediocre.
It seems like no matter how hard I try, I just can’t seem to excel in my studies. I’ve always been a diligent student, putting in hours of hard work and dedication, but despite my efforts, my grades continue to fall short of my expectations. It’s incredibly disheartening to watch my peers succeed while I struggle to keep up. I would excel in about two subjects in high school and pass the rest. Everyone around me called me intelligent, praised my successes and wouldn't mention my failure. Now it feels impossible to not put those in the spotlight.
What’s worse is that this constant cycle of failure has started to take a toll on my self-esteem. I can’t help but feel like I’m not good enough or that I’m just not cut out for academic success. It’s a vicious cycle of self-doubt and disappointment that’s been weighing me down.
I know I’m not alone in feeling this way, but it’s still tough to talk about. Everyone around me seems to have it all together, while I’m over here barely keeping my head above water. It’s hard not to compare myself to others and wonder what I’m doing wrong.

We didn't exist for trillions of years, and then one day we suddenly appear because two people decided to breed. Now, we're automatically forced into the rat race, competing in everything from school and college to relationships and jobs. It's all just a completely pointless rat race. We're here to compete and strive to reach the top, only to lose it all in the end. When you die, your experiences and your money wont come with you. You'll return to the pre-birth state of non-existence. You go from nothing, to something, and then back to nothing.
It's such a pointless cycle. If life were at least entertaining, I wouldn't mind its pointlessness so much. But unfortunately, life is not only pointless but also boring. Birds can fly, tigers can run at incredible speeds, and monkeys can swing around like Naruto characters. What do we humans have? Nothing we do is interesting or cool. We spend our time in concrete boxes because we can't tolerate certain weather conditions. We can't do anything impressive like fly, run fast, or even fictional things like teleportation or invisibility.
Instead, we spend most of our lives working and doing boring, menial tasks. For 99% of people over the age of 18, most of our waking hours are consumed by school or work. The sheer dullness of life and our physical limitations make the pointlessness of life irredeemable.
Personally, I wish I had never been born. If I had to exist on this earth, I would have preferred to be a plant or a cool animal. But now that I'm here without my consent, I have no choice but to exist and be a slave until I die, trapped in a boring human body incapable of doing anything interesting. I honestly wish we had never evolved from monkeys because having so much intelligence and awareness of the world around us has been our downfall. Ants don't become depressed when they have to work together to make a living. Neither do most animals. But since we are the most intelligent and aware, that really just makes our lives shittier because we know how depressing and pointless it is.
But now that I'm here, there's nothing I can do but wait until I die. I have no choice but to be a slave to money until I die. Having to pay taxes and bills just to exist. I'm trapped in a life where I'm judged and belittled for not excelling in this pointless rat race. A rat race that doesn't matter, because we are all going to die in the end. And I didn't ask to play this stupid game. I was forced into it by two fools.
Human life is probably the most boring, mundane, monotonous game there is in existence. I am jealous of plants. They are alive, but their existence is very brief, simple, and there is no suffering. Even when they die there is no suffering. It simply isn't fair at all, but hey at least one day this garbage will be over.

Hello Everyone,
I have really found this /r helpful, and I really appreciate everyone here. I'm an unusual applicant, and I'm really unsure how I should proceed. I am 33, and when I was in Undergrad I experienced some terrible, publicly documented abuse of a novel type (at least in the academic literature). This caused me to have a 2.3 UGPA, I also had to work full time, and am a URM, 1st gen college grad, had to work to support my immigrant wife who couldn't work, (we're still married lmao), etc.
Since then, I have 2 master's degrees, one of which has a 4.0 GPA. As a part of that second 4.0 masters degree, I conducted the largest qualitative/quantitative research on the population that I'm a part of (survivors of this type of abuse), and produced a 100 page thesis on the topic, with some pending submissions to academic journals. A part of it was categorized under legal theory, which is so cool! I also created the first publicly listed college course on that type of abuse as a professor at the university I'm at. I teach on the topic here in the USA, and I do Collaborative Online International Learning on the same topic in Japan & Costa Rica. I have won a few faculty awards for excellent teaching, helped found an academic certificate and now manage it as a faculty member, and have spoken at two huge academic conferences on my research. I'm also a part of the two largest (to my knowledge), advocacy groups for this type of abuse, one of which I founded and manage as well. I teach like 9 different courses at my uni.
I'm looking to get a JD because a big part of my research involves survivors taking legal action as a recourse for this type of abuse. I'd like to be a professor of Law one day, and I know that's hard as hell, but I know I want to teach, because I do in my area, and I love it. I'm also genuinely interested in being an attorney. My LSAT practice is 174-178 consistently (AUG and on). I'd like to go to any top 14 school.
I was wondering if anyone had first hand experience with those admissions counselor networks from last cycle? What was your experience like? did it help? anyone or anything to avoid?
I can get amazing letters of REC from deans, chairs, academic committee members and others that have seen my work. They can all attest that my 2.3 from 15 years ago isn't representative of me. I think I have a really cool story, going from survivor, to advocate, to scholar on my topic. Even defining it with social science evidence and some publications/conferences etc. But I honestly don't know if top schools care. It's been a journey and truly hard as hell to do, but I'm happy I've done it, and this is the next step for me any my family.
Also, am I just screwed due to my low GPA?

We didn't exist for trillions of years, and then one day we suddenly appear because two people decided to breed. Now, we're automatically forced into the rat race, competing in everything from school and college to relationships and jobs. It's all just a completely pointless rat race. We're here to compete and strive to reach the top, only to lose it all in the end. When you die, your experiences and your money wont come with you. You'll return to the pre-birth state of non-existence. You go from nothing, to something, and then back to nothing.
It's such a pointless cycle. If life were at least entertaining, I wouldn't mind its pointlessness so much. But unfortunately, life is not only pointless but also boring. Birds can fly, tigers can run at incredible speeds, and monkeys can swing around like Naruto characters. What do we humans have? Nothing we do is interesting or cool. We spend our time in concrete boxes because we can't tolerate certain weather conditions. We can't do anything impressive like fly, run fast, or even fictional things like teleportation or invisibility.
Instead, we spend most of our lives working and doing boring, menial tasks. For 99% of people over the age of 18, most of our waking hours are consumed by school or work. The sheer dullness of life and our physical limitations make the pointlessness of life irredeemable.
Personally, I wish I had never been born. If I had to exist on this earth, I would have preferred to be a plant or a cool animal. But now that I'm here without my consent, I have no choice but to exist and be a slave until I die, trapped in a boring human body incapable of doing anything interesting. I honestly wish we had never evolved from monkeys because having so much intelligence and awareness of the world around us has been our downfall. Ants don't become depressed when they have to work together to make a living. Neither do most animals. But since we are the most intelligent and aware, that really just makes our lives shittier because we know how depressing and pointless it is.
But now that I'm here, there's nothing I can do but wait until I die. I have no choice but to be a slave to money until I die. Having to pay taxes and bills just to exist. I'm trapped in a life where I'm judged and belittled for not excelling in this pointless rat race. A rat race that doesn't matter, because we are all going to die in the end. And I didn't ask to play this stupid game. I was forced into it by two fools.
Human life is probably the most boring, mundane, monotonous game there is in existence. I am jealous of plants. They are alive, but their existence is very brief, simple, and there is no suffering. Even when they die there is no suffering. It simply isn't fair at all, but hey at least one day this garbage will be over.