DFPSR»Blog

Made an official website for the project

Recently read about how important it is for open source projects to let their users find them in order to get more contributors, and I do need help with making the project grow with all the big plans I have. Then I made a search for my old engine and only found one post mentioning it, which was about how such a good product could be completely unheard of for ten years. My skills in promoting my projects can probably improve, so I've spent the last week building a new flashy website with a short top domain, so that visitors have more than just the github page to bookmark. Read about search engine optimization but failed to get it indexed by any search engine in the end. Designed some banners that people can use when linking to the website, which will hopefully keep it from being completely invisible when people explicitly search for my full name.
https://dawoodoz.com/dfpsr.html


Decided to trademark the site's name using the ™ symbol without registering it, because it's been with me for decades any someone else using it would feel like identity theft. Buying the dot com domain should be enough to scare away any criminals trying to steal the name. If they buy a strange local domain, I can just laugh at them for getting the domain people can't remember.

Because my old site was on classic Google sites, which will soon reach the end of life, I also moved my old GPU renderer from 2010 into the new site. Many years ago, I assumed that my site could stay on Google Sites because they wouldn't go bankrupt any time soon, but instead they became crazed with power and closed it anyway. Google is really adamant about bending users to their will and did not even offer to keep the existing themes from Google sites. This was the third nail in the coffin for using Google after they ruined my YouTube account that I used to showcase my engine and locked down Google code where my code was hosted. It's back to plain HTML then.


Tried many times to get CSS to work for more than just the index page, but eventually had to give up and just paste the style into each page's head. A system for automatically generating HTML from my own text format would probably be more efficient, because I also have a top menu repeated on each page, which must be defined in the HTML body. When designing a website, I'm torn between using free tools that are tied to a service (which will be gone after only a decade), and having to write HTML by hand (can at least be copied to the next host). Should probably create my own tools for generating websites with correct character encoding and tag endings automatically, because I'll need it for the software renderer's HTML documentation. HTML is way too verbose even with CSS and Markdown is not explicit enough with all the dialects.
Simon Anciaux,
The following might sound negative but my intent is to provide some useful feedback. It's only my opinion, I don't pretend it's the truth. I'm not a designer or web developer and my own website probably has some of the problem I mention.

I think your site doesn't look good, and isn't setup for a good reading experience. I'll only talk about the dfpsr page, but it applies to all the site.

When I first arrive after following the link, the main visible thing is the screenshot. The image (and text) is left aligned. If the browser is maximized, it doesn't feel comfortable as I'm centered in front of my monitor and the right half of the screen is just black and unused. The screenshot itself isn't interesting to me. I understand it's suppose to show what the engine is capable of, but it's dark with no atmosphere, the scene isn't interesting, the lighting isn't interesting, the composition doesn't exist and the format isn't interesting (the image is almost a square, you could use a widescreen format). The integration of the engine (or is it just a renderer, that's not clear) name isn't good and should be more stylized (~logo).

In my opinion you should center the content on the screen, use only a portion of the available width (have a max width a 1024 for example) as it's easier to read text if it doesn't cross the whole screen.

The menu doesn't look good, the separators don't look good, the green on black colors look like all the sites that were created when The Matrix came out. It seems you're going for an oldish style, but in my opinion it doesn't look good and makes me think that the content of the site is old and was there for the last 20 years without being touched. I think you should go for a lighter (well just not black everywhere) theme, and a bit more diversity in color for text and UI.

The text itself isn't that useful. It's not clear to me what the renderer provides feature wise, for whom it is created, for whom the site is created (user, or contributor). The first paragraph is a bit weird to me, it seems like I started to read in the middle of a larger paragraph.

The following line doesn't make sens to me:
Want a custom rendering pipeline or image format? Just copy and paste until you have found a unique style.


The following seems suspicious. It seems to me that you're saying that if you throw stuff at a GPU without any care it will work worst than a well thought out solution on the CPU, which isn't a fair comparison to me.
2D and isometric graphics that a GPU might struggle to render in 60 frames per second due to a poorly utilized 3D graphics pipeline and call overhead can be drawn in 300 frames per second on an affordable desktop CPU without messing up your code.


Overall you seem opinionated about GPU vs CPU and Open Source vs Proprietary, which is OK but can come out wrong and discourage potential user.

The banners you provided will most likely clash with any other website and I think almost no website uses that kind of things nowadays. Press kits contain higher resolution images that publications can edit to fit there layout and different logos that fit with different backgrounds.

As for tools to create a site I like the ideas in this post from Fabien Sanglard.
Edited by Dawoodoz on
Thanks for the comments. I will add more content and clarify those points when I figure something out. I was thinking about adding some decorations to the square images, but hate it when sites mask images using circles (feels like something's covering my sight).

Regarding CPU versus GPU, it's not a comparison between optimized versus not optimized engines. It's actually the GPU hardware itself that's limiting due to PCI express memory latency, initialization overhead and low clock speeds. If the target platform only has an old OpenGL driver, the overhead gets even worse. If the game uses low resolution 2D with serial dependencies in filters, the CPU will actually beat the GPU hardware-to-hardware by making high frequency and low memory overhead more important. There's however an old myth that GPUs are simply faster at everything. A serial dependency is not just a problem to remove, because reusing results can turn O(N⁴) into O(N²) using algorithms that would be slower on a general graphics API.

I selected this design because most websites after 2008 make me remember the Soviet era (yes, I'm that old). Generic groceries had the same blue white generic style as Facebook and LinkedIn. All looking the same without any variation.

Everever I browse, all sites look the same to me, images crammed into ugly circles, overly large texts, menus trying to look like a mobile application... If everyone just follows the herd, it will become even more horrible for each year. I guess all styles will eventually look old to those remembering the previous cycle, just like old people see Lady Gaga as unoriginal.

Show a young person a flat square with big text and the association is mobile devices. Show me the same and I will think of Windows 3.1, Norton Commander and groceries from 1979 (much older looking).
Edited by Dawoodoz on
I updated the arguments with a long list of facts to sound less opinionated. Including all arguments for and against software rendering that I could think of. To make readers understand the use cases, I added some of my personal experience where GPU rendering did not stand the test of time or just felt wrong for the situation.
Edited by Dawoodoz on
Made a search for the latest web trends and this blog pretty much sums upp everything that I hate about today's annoying websites.
https://www.tisindia.com/blog/mus...n-elements-for-modern-web-design/

1. Full width video background. Is wasting data, discriminating poor countries and enabling sound at the wrong time supposed to be mobile friendly?

3. Inifnite scrolling. Will either drive the visitor insane and leave or create a dangerous social media addiction.

5. Adaptive design. I have already decided which minimum text size I want on my phone and hate when sites try to mess around with the settings for me. Disabling zoom for those who explicitly pinches the page because they cannot see a small image is just evil. Any text bigger than 18 pixels is screaming at the reader.

7. Web fonts. So now the news story about a war looks like a mobile game for children. (applauding)

11. Card layouts. People hated it in Windows 8, but lets try to force it on everyone on the web while ruining access for blind visitors.

I guess this is what one gets if only measuring sales rather than the visitor's mental health after the visit.
Hello,

Just wanted to mention that finding proper color combinations that won't make your visitors' eyes bleed is a full-time job on its own; a lot of people have figured it out already, and you can find color combinations collections like this one if you search for a bit:
https://visme.co/blog/website-color-schemes/

Next, I'll talk a little bit about my experience. Before rolling my own HTML static site generator, I used Hugo for a while. It's too bloated for my need (all I need is some occasional blog posts), but it allowed me to figure out what I really need. What is really useful is their collection of ready-to-use themes (again, to be easy on the eyes of your visitors), so you can use one of them, tweak it to your needs and stay focused on your own project instead of losing time on your website. You can even just look at an example site using a theme and try to mimic it (don't forget to properly credit it :) ) if you want to completely avoid using Hugo.
https://themes.gohugo.io/
Number 33 "Striking citrus colors" is around the look I'm going for. Just have to make sure that the details don't distract in the background if using transparency.
Edited by Dawoodoz on
Google's mobile friendliness test is apparently completely broken. Tried their advice about adaptive sites just to see what it would look like. The text became way too big (4x maximum acceptable) on my phone and Google still complained that the text is "too small". It looked fine on my phone before they said anything, because it scales up the smallest text automatically to user preferences.

Why can't they just have an HTML tag saying "It's plain text, stupid! Let the user decide the font size."? A browser considering plain text to not be readable on a mobile device is a broken browser. We cannot keep changing the HTML standard for each new fad being pushed by companies.
Abner Coimbre, Edited by Abner Coimbre on Reason: Initial posts.
I think making an official website is great and it's inspiring to see you take constructive criticism in stride and improving it.

I would also love to see visually punchy and long-form content over time. I follow your short posts because I'm already invested in the network, but I think new readers would need something like Riccardo's The Art of Syzygy series to be drawn in.

The Handmade Network Twitter has a strong following, and they might tweet about your longer articles if you put in the work and then contact staff (not that you're not putting in good work already.)
I might put up a tutorial on SIMD vectorization based on my experience from computer vision development. First a generic guide that can be searched for, and then an introduction to the hardware abstraction layer. My coworkers used to complain about not finding any good ARM NEON guides.
Simon Anciaux,
I would be interested in a tutorial on SIMD (preferably SSE for me).
Mārtiņš Možeiko,
Here's a good high-level one: http://const.me/articles/simd/simd.pdf
Simon Anciaux,
Thanks.
mmozeiko
Here's a good high-level one: http://const.me/articles/simd/simd.pdf


I could probably add many standard industry tricks on top of that.