riklaunim

One is the cable the other is the connector. Just note that there are also wider 8i variants for PCIe x8 which GPD does no have. There are adapter boards on Aliexpress, Ebay and the like. You will also need an ATX power supply so it won't be a mobile eGPU setup. Also depending on GPU vendor there can be additional tricks needed to make it working.
 
To have a PC that is ultraportable in a pocket or small bag. It can be docked and then you have a quite good workstation. Also sometimes for field use with various external tools.
 
And IMHO even 13-14" would be somewhat too small to use as a primary screen for extended periods so it might as well be super small and docked when needed.

Python dominates universities because it makes programming very approachable. That is attracative for research. Automating some data analysis, doing matrix or array stuff, repeating the same thing an a few dozen or hundreds of dataset etc. is made simple and a great deal of your day-to-day science does not require that kind of stuff to be programmed in a low level language like C(++). Even if it is code that will need performance down the line, something like Python is great to jot down a first iteration to flesh out the idea. Once everything is clear you can then put effort in a better defined optimisation. All the core stuff I use is written in C++ and uses/can use platform-specific optimised math libraries like MKL or AOCL. Python serves as the glue around it for stringing things together. Even when things are done "in Python", they use libraries that usually are bindings to C(++) or Rust code as well.
You can write junk code in every language. That has nothing to do with Python or C++ and there are many reasons why code can be suboptimal. Maybe you don't have the mathematical or programmatical knowledge to optimise it further, maybe you don't understand the target domain well enough to write proper code for that use case, or maybe you write something very general and thus have to make sacrifices in the base product in order to accomodate flexibility.
Why does Haskell equal basic quality? Someone will make a bad performing game on par with today's complaints with your supposedly superior Haskell engine, I'm willing to take poison on that. Simply because they don't fully understand your engine, don't have the in-depth Haskell expertise or do things inefficiently for other reasons. Then we can all happily bash Haskell and rinse, lather, repeat.
 
Another part is status quo. There are an insane amount of resources for game development in C(++) making it unattractive to leave that ecosystem. Someone will have to come with the "Unreal Engine 5 Haskell Edition", but that will take amounts of time and effort(and money most likey) that few have or are willing to put in given the risk.

It's just a glue while you code CUDA kernels separately, the same with other aspects of GPU computing.

To use more cores you would need to parallelize more and more work and that may not be available to a game engine or an app. Drawing geometry, shading etc can be parallelized and DX12/Vulkan/Metal is quite good and scheduling draw calls via multiple cores. Game logic like current world state and combat is based on conditional calculations and you can't split it as one equation needs the result of the previous one - that's why often games have one "primary" core with higher load and few/multiple with lower load for everything else that can be parallelized. The game may trade consistency for parallelization speed but then gamers would be angry their spell did no damage or it did 10 seconds after it should etc. For single-player games, it could even pass but for anything multiplayer and barely competitive it's pretty much a no-no.

If you check WoW technical support forums you will notice that a lot of players have such stutter problem in Dragonflight. Some from the launch even.
 
If it's the game then you can't do much about it. If it's not the game - there are few things to check/test:
 
* What is the capacity on the SSD and how much free space does it have left?
* Sometimes Nvidia drivers conflict with other drivers causing excess CPU load or just pure stutter/high latency in the system. This usually can be fixed by removing Nvidia drivers with DDU and installing them fresh.
* Also check if any Windows service or app isn't using a lot of CPU while WoW is having problems
* Disable any overlays (Discord, MSI Afterburner and alike), disable every addon and test again. If you have multi-monitor setup disconnect all monitors aside of the main and test again.
 

Publish your code on for example github and then ask the questions - here, on reddit and so on. Don't ask to ask 😉

Publish your code on for example github and then ask the questions - here, on reddit and so on. Don't ask to ask 😉

Publish your code on for example github and then ask the questions - here, on reddit and so on. Don't ask to ask 😉

Within your limit there should be Ryzen 7 laptops (4700U/4800U/5700U and maybe even 5800U/5825U). That combined with 16GB of RAM would be really solid for more than just learning to code. What you should look at before buying is the laptop screen and noise levels - look for reviews of given model to see if it has some impactful weakness.
 
Not sure what's your goal are but with tkinter you rather won't get any good apps and simple desktop apps are rather something that should be a web app. For hobby/fun use it likely will be fine, but like commercially not really.

There is limited emphasis on game development in Python. There is PyGame and few other libraries but most traffic (guides, tutorials, resources and tools) are for Unity and Unreal (C#).
 
And don't expect to master anything faster than in few years at minimum. You can watch a guide for some simple showcase game, try to create something similar and hit a wall of actual complexity of game dev. And if you want to get a job later on - for Python it will be mostly web dev so you should know that. Plus if it's boring for you then I doubt you get  a job as for junior positions there is a lot of applicants. Bad code, lazy/lack of skill is first to get rejected.

There is limited emphasis on game development in Python. There is PyGame and few other libraries but most traffic (guides, tutorials, resources and tools) are for Unity and Unreal (C#).
 
And don't expect to master anything faster than in few years at minimum. You can watch a guide for some simple showcase game, try to create something similar and hit a wall of actual complexity of game dev. And if you want to get a job later on - for Python it will be mostly web dev so you should know that. Plus if it's boring for you then I doubt you get  a job as for junior positions there is a lot of applicants. Bad code, lazy/lack of skill is first to get rejected.

{{ form.as_p }}
 
This renders all of form fields and those fields are listed in HotelForm.
 

That example uses a simple file input. Copy/paste image from clipboard is not supported by it while drag and drop on an file input would depend on the browsers. Usually if you need drag and drop and clipboard paste you have to do it in JavaScript and then also handle how the data is sent to the backend (usually via AJAX requests to a custom API endpoint).

For comparison charts you could look at relative radar charts, assuming the data is good to compare in such format, like some of my attempts:

 
The faster test run is 100% and the slower one is a % of that.  AMD started to use radars as well. If there is enough data points, the lines have clear separation it can give a quick view on how 2-3 things compare to each other.

There is limited demand in desktop apps in Python. Most jobs or projects are either web or some service/API. For actual games there are some bindings in Python to game engines/libraries (RenPy, PyGame...) but they are niche, majority goes into Unity and Unreal Engine via C#. Note that making a game takes way more than just coding 😉

Yoda would have a quote for that but the short answer - you should not.

If you will like JS and SPA as a frontend then you may look at node and master that ecosystem on the backend and frontend. Also note that nowadays software stacks are the driving force. You should be looking at picking software stacks, frameworks to masters while language would be a side effect of that (check job offers to see what and how is listed in them, especially in your city, vicinity).
 
A tool that makes work easier and better. Django in Python is a go-to framework for web development - it has everything you need to make a dynamic web application - templating, database support, ORM, testing, even generic admin panel. For one it makes your job easier but also standardizes code for a project and partially for any django app out there. Multiple devs can work on same project so consistency and quality is important. And there are other Python frameworks aiming at different use cases as well - like API servers, microservices and so on - so they are chosen based on what fits given project best and what the teams wants to work with.

So you don't have funding, you don't have staff, you aren't leveraging any existing distro and yet you want to make it highly non-standard... Sorry but it won't work. Era of "after hours" codding is long gone. Not to mention security aspect of a distro made by no-name entity.

This will depend on what type of programming you do. If you run like Docker + Kubernetes with multiple containers eve 16GB can be low. If you work with large datasets, databases that need RAM the same.
 
For average tasks, learning etc there are no such high requirements, just note that you can't add RAM later. Also check software compatibility - not everything can be available on macOS or on ARM version. Python is available.

So you don't have funding, you don't have staff, you aren't leveraging any existing distro and yet you want to make it highly non-standard... Sorry but it won't work. Era of "after hours" codding is long gone. Not to mention security aspect of a distro made by no-name entity.

Nowadays you don't really "learn" programming languages but software platforms and best practices. If you want to use PHP then you have to learn PHP top frameworks/software stacks. Same for Python. Frontend with JS/CSS/HTML will push you through jQuery to also SPA frameworks like ember.js, vue or angular - if you want to be a fullstack developer.
 
It's completely useless. He isn't going on Computer Science course and on top of that as self learner aiming at doing very specific high level stuff he will likely never grasp any meaning or importance of it. To become a good full stack / web developer he will have to learn a lot, some related to what you are implying but it will be learned from a totally different angle.

Zen2 APUs are somewhat available in EU, got 4650G locally here in Poland, even come in with part of that tray AMD sells them. Aliexpress has them too at pretty much same price which is bit odd. AFAIK some got one of those from Amazon too. 4750G is nearly double the price of 4650G for 2 extra cores and 1 Vega "core" - so IMHO not so worth it.
 
And my World of Warcraft 4650G benchmarks: https://rk.edu.pl/en/benchmarking-world-warcraft-amd-apu-integrated-graphics/
 

Zen2 APUs are somewhat available in EU, got 4650G locally here in Poland, even come in with part of that tray AMD sells them. Aliexpress has them too at pretty much same price which is bit odd. AFAIK some got one of those from Amazon too. 4750G is nearly double the price of 4650G for 2 extra cores and 1 Vega "core" - so IMHO not so worth it.
 
And my World of Warcraft 4650G benchmarks: https://rk.edu.pl/en/benchmarking-world-warcraft-amd-apu-integrated-graphics/
 

JS can be divided into two parts:
* widgets and scripts used on websites
* single page applications built entirely in JS via Ember.js/Angular/Vue/Flavor of the month framework
 
The first one is generic web design and it's that complicated the second one is quite specific specialization and quite often specialization into one software stack. For freelance webdesign it's rather to narrow.
 
There is to much juniors and wannabies that don't learn and way to little mid/senior developers. If you learn you will have more jobs than time allows.
 
If you are willing to learn then it would be cool to find a job in a company (even small one) that have good reputation, senior developers and one that mentors/tutors new developers - this will teach you a lot, really a lot and after that you will be able to move your career as you see fit.

That's bit to many cores. IMHO 16-cores is max so 1920X, 1950X, 2920x, 2950X. Those will be slower single core than Ryzen 5900X, 5950X or even 3900X, 3950X but also quite often extremely discounted due to that (and available here and there as leftovers). On the bonus side you get lots of PCIe lanes (lots of NVMe storage if needed, multiple GPUs even in the future if needed) and quad channel RAM.
 
And check those Ryzens too. If you want to look for bargains things like Ryzen 3900 (no-X) can be cheaper as a slightly cut down version of 3900X. The 5000-series would be the most performant but it's up to you if you want to push it that hard vs price. I'm still using 1920X for my dev box.