The archive is available at https://twitter.jeroen-baert.be/

The post Twitter archive appeared first on Jeroen Baert's Blog.

Upsides:

• No need to run a full PHP stack and a vulnerable WordPress installation (I can still use a private instance to update the site and generate a new build)
• Pages can be minified and cached more easily by my CDN
• Much faster page load times and less load on my server

Downsides

• Losing some interactive functionality (comments, search)

I’ll see if I can retrofit some solutions for these downsides. Enjoy reading!
Update 16/02/2023: Solved some issues with older posts (>5 years ago) that weren’t converted correctly. Enjoy!

The post Moving to a static site appeared first on Jeroen Baert's Blog.

Last Update: 22/07/2023 (Links cleanup)

General resources

• SpreadMastodon (and/or JoinTheFediverse) are excellent sites to onboard family & friends. See also these comics.
• How To Get Started on Mastodon (Wired) and How to join Mastodon (Gizmodo)
• Mastodon.help
• What is Mastodon? (The Guardian) and What is Mastodon? (BBC)
• A Twitter Refugee’s Guide To Mastodon
• The Newcomer’s guide to Mastodon
• A Humane Guide To Mastodon
• joinmastodon.org <- Official website, official documentation
• Guide from @Noelle@elekk.xyz
• Nikodemus’ Guide To Mastodon
• My previous blogpost (in Dutch)
• How to use Mastodon: complete guide (video)
• Video explanation by eGPL
• General Fediverse tips
• Awesome Mastodon
• If you prefer a physical book: Mastodon for Dummies

Huh? What is a server? Why should I care?

This is where the first major difference between Mastodon and Twitter becomes clear:

• Twitter is a centralized service. One company runs, administrates and decides for the network. If you don’t agree with Twitter’s policies or monetization, there’s no other option than begrudgingly putting up with them or deleting your account.
• Mastodon is a decentralized service. There is no company that runs it. The software to run the network is free and open-source. Anyone can run, edit and adapt it to their needs.
  • All of these servers (also called instances) communicate with one another, according to a certain protocol called ActivityPub.
  • Each of these servers has a different administrator (or a team of them) running it.

Update: Since May 2023, Mastodon uses mastodon.social as the default instance using the sign-up process on mobile. This should make it easier for people to just get started.

Which server should I pick?

It doesn’t really matter *which* server you pick. The default mastodon.social server is fine if you’re starting your journey. You can also pick a server that aligns with your interests or main language, but keep in mind that it will not limit you to follow or interact with people on other servers. Once you follow someone, all that complexity goes out the window.

Think of it as e-mail: lots of people have accounts at different providers (Gmail, your ISP, Yahoo, …) but there’s a set of rules that allows all these addresses to exchange mail without problems.

There is a server list on the Mastodon website, and a curated list at Fedi.garden which groups servers per language/topic. A more advanced (and longer) server list is here. For Belgian Mastodon users, I can recommend mastodon-belgium.be.

Finding people to follow on Mastodon

Since Mastodon is decentralized it might be a bit harder to find the accounts of people you want to follow . You can’t just enter their name in the search field and find them – they will only show up in the results of your server has made contact with them before.

So check your friend’s profile on other social media or website for references. Mastodon accounts have the form of @handle@server.extension, for example @jbaert@mastodon.social. That’s Mastodon’s way of saying that I (@jbaert) have an account at (@) the mastodon.social server.

Once you have someone’s full handle, log in and paste it in the search bar of your Mastodon web client or app. Once the account shows up under the “accounts” tab, open it and follow it. Voila!

Finding people from Twitter

Some tools have also been developed that try to find people on Mastodon based on your Twitter follow list. These tools require you to login with your Twitter account to fetch your info.

• Fedifinder
• Twitodon
• Mastodon Flock
• Debirdify (Update 22/03/23: Debirdify has been suspended by Twitter)
• Movetodon (Suspended because of Twitter API changes)

Finding and following more interesting people

There is no algorithmic recommendation engine driving Mastodon. It’s up to you to fill your timeline with interesting people. It requires a bit of work, but it’s definitely worth it! These sites are promising projects to boost discovery on Mastodon:

• Followgraph: Tool that explorers your follower’s followers. Very good way to quickly find interesting people.
• Fedified: A list of notable accounts on Mastodon
• FediDirectory: Hand-curated list of cool accounts to follow.
• FediPages: Search engine who crawls popular Mastodon servers and allows you to search.
• Fediverse.info: Tool to find people by interest.
• Find Academics: Tool to find academic Mastodon users by research field
• Trunk: Wiki that maintains list of interesting people per category.
• A list of news/magazine bots
• Google Sheets of Notable Accounts, Journalists and fun bot accounts to follow

You can also usually browse the directory (account list) of a Mastodon server. For example, for the Mastodon-Belgium server, it’s at https://mastodon-belgium.be/directory.

I can also recommend the FediAct extension, that allows you to easily follow and perform actions when you browse servers that aren’t yours. Another handy tip: You can choose to only follow posts by a user in a certain language! This is handy for big international accounts who alternate between multiple languages.

Fedi.Tips has some good tips on finding friends and getting more followers.

What Mastodon clients should I use?

Unlike Twitter, Mastodon supports and encourages third party clients. The ActivityPub protocol it runs on is documented and open-source. The Mastodon website maintains a list of compatible apps.

The official client is generally considered good (enough), but power users might look for more functionality in other clients. You can use multiple clients, there is no limit! For a more extended list of clients, check here.

Android

• Official client (<- Recently got a revamp, see this blogpost)
• Tusky (<- I’m currently using this one)
• Megalodon for Mastodon
• Subway Tooter

iOS

• Official client
• Ivory (this one seems very popular among my iOS friends)
• Metatext
• Toot! for Mastodon
• IceCubes
• Manny
• Mammoth (from the devs of Twitter app Aviary)
• Wooly (from the devs of Twitter app Fenix)
• Mona (from the devs of Twitter app Spring)

Web

• Elk (<- I use this one)
• Mastodeck
• Phanpy

Can I search Mastodon for …?

Because of its decentralized nature and because it searching can be used as a tool for harassment (searching for a term and bullying everyone who uses it), there is no default full text search on Mastodon. You can, however, search for users and hashtags.

Some servers might enable full text search by installing an additional component, ElasticSearch. There is also an external project to also search the content of posts: Tootfinder. You can opt-in to offer your content for full-text search.

Changing Mastodon servers

There may be many reasons why you would want to change servers. Maybe you don’t agree with the moderation policy anymore, you think the server is not performing well (although that might very well be temporarily), or you just feel like switching. Mastodon doesn’t hold you or your account hostage! There is a built-in way to move servers.

The correct way to switch is explained by developer Eugen in this post.

1. Sign up on NEW server
2. On NEW server: Go to Account -> Moving FROM another account
3. Enter old account’s handle
4. On OLD server: Go to Account -> Moving TO another account
5. Enter new account’s handle and submit

The following guides also explain it thoroughly: This guide from Patricia Aas, this guide from fedi.tips, guide from DjNavarro

Your followers should automatically transfer as well, though this can take a while. A lot of these guides recommend manually exporting them on your old instance and importing them on the new one to speed up the process. Your posts will remain on the old server, where your account will become a referrer to your new spot!

I forgot my password, and Mastodon doesn’t recognize my username … huh?

This is a common pitfall: only the server you registered the account at can help you with this problem. There is no single knows-it-all Mastodon server. It’s a decentralized system, remember?

A lot of people get confused because at first glance, for example mastodon.social and mastodon.lol look very similar visually (of course, they are both running Mastodon with the default UI!). They might have the same look, but these are separate servers, and making an account on one of them doesn’t automatically give you an account on the other one. So make sure you check the URL before trying to log in! It’s easy to get confused!

Why do I have multiple timelines?

Mastodon offers you three timelines by default:

• Home: Posts from people you follow, and the things they boost (=retweet)
• Local: Posts from everyone on the same server as you
• Federated: Posts from the whole Mastodon universe that your server just picked up

There is no algorithmic recommendation engine. An external attempt can be found at FediFeed.

Does Mastodon have lists?

Yes, Mastodon has lists to collect people you follow, but they are private-only at the moment. You cannot share them. You can find them in the right sidebar of the Mastodon web client, and in your Mastodon client. Like Mike Masnick explains in this excellent article, Mastodon List Manager is a great external tool to manage your lists. I use it often!

Enabling the Mastodon advanced web interface makes it look more like Tweetdeck, allowing you to have multiple columns of lists side by side. Also see Mastodeck.

Does Mastodon have DM’s?

Yes, but they work very differently from the DM’s in Twitter. It’s implemented as a private conversation that looks like a post. More info here. They are not end-to-end encrypted (your server host can read them) and should not be used for sensitive information. Use Signal for private communication.

How can I get verified?

There is no verification system in Mastodon similar to Twitter’s dreaded “blue ticks”. This means that identity has to be proven externally. People you see with a checkmark in their Mastodon name simply placed it there as an emoji – it has no meaning, other than being a joke. You could just as well place a monkey or poop emoji there :)

If you want to prove your identity on Mastodon, you can do it by listing websites in your profile that are under your control. There is a way to formally verify these links by including a piece of HTML code on the page you’re linking to. The Mastodon docs have more information. This guide from fedi-tips is also very good. Also check out this excellent Mastodon Link Debugger tool.

For example, in my profile, I prove that I am the owner/user that has the rights over jeroen-baert.be and my Github profile. This should give users high confidence that I am who I claim to be.

Mastodon is not Twitter (and it doesn’t have to be)

Although they are both microblogging platforms, there are subtle and less subtle differences in the ways both platforms work and operate. You might have to take a while to adapt to a new way of doing things.

• No recommendation engine or promotion for viral tweets: giving someone a favorite(=likes) just communicates to the author that you appreciated the post. favorite away!
  • If you still want to keep track of the dopamine hits of people liking your content, check out MastoMetrics
• Every instance is run by volunteers. There might be outages, slowness or hiccups. Be patient.
• Reverse chronological timeline only: no ads, just the posts of people you follow, with the most recent ones at the top.
• Use boosts (=retweets) liberally: Boosts are important for the discoverability of Mastodon users. If you like something and want to spread it to your followers, boost the post!
• No quoting, by design. Explained here and here. (Update: they are coming.)
• Use Content Warnings
• Use image descriptions to help people with disabilities

Mastodon is just the start of the Fediverse

The fediverse is bigger than Mastodon! There are lots of great decentralized services that are worth checking out. Thanks to Sean Tilley for collecting these:

• Facebook replacement: Friendica
• Instagram replacement: Pixelfed
• YouTube replacement: PeerTube
• Spotify replacement: Funkwhale
• MeetUp replacement: Mobilizon
• Reddit replacement: Lemmy
• Podcasting replacement: Castopod
• GoodReads replacement: BookWyrm

Crossposting between Twitter and Mastodon / Posting to both

There are tools that can post tweets to Mastodon and vice versa, saving you some time from typing or copy-pasting your content twice, if you choose to engage on both platforms with the same stuff.

Buffer is a tool that allows you to post to multiple networks: Twitter, Mastodon, LInkedIn.

There is an open-source Mastodon-Twitter Crossposter, developed by the awesome Renato Lond. It has shut down, so if you want to use it, you’ll have to host your own instance. Another crossposter tool that you can self-host is this Mastodon-Bot by Yogthos.

Not all content is interchangeable: Retweets or Quote Tweets lose their meaning on Mastodon and intrinsically link to features on Twitter (accounts, t.co links, …). My advice is to not blindly forward or copy retweets or quote tweets. If you have to link to Twitter content, I can recommend Nitter, a tool that offers Twitter content without all the tracking.

Export your Mastodon data

Mastodon servers don’t hold your data hostage. Go to preferences, choose import and export, then data export and request a zip file of all your data.

Export your Twitter data

Just in case Twitter ever goes down, it’s nice to have a local backup of all the tweets and content you wrote over the years. You can request a big zip file with all that information on the Twitter website.

• Click More in the main navigation menu to the left of your timeline.
• Select Settings and privacy.
• Select Account under Settings.
• Click Your Twitter data under Data and permissions.

Download that zip file and store it somewhere safe. It’s a good idea to do this for other services (like Google) too. If you want to, you can convert all this data to something more manageable later. This is also a nice tool to convert your Twitter archive.

The post A guide to Mastodon (for Twitter users) appeared first on Jeroen Baert's Blog.

I’ve merged this guide with the one here, since a lot of the topics were similar. Go read over there!

The post Mastodon Q&A appeared first on Jeroen Baert's Blog.

Hieronder een thread die origineel op Twitter verscheen over Mastodon, een gedecentraliseerd microblogging platform dat na de overname van Twitter door Elon Musk onder de aandacht kwam. Ik bewaar hem hier voor mensen die rustig willen nalezen. Ook voegde ik een verzameling goeie algemene resources over Mastodon toe.

Bedankt aan iedereen die vragen instuurde, Mastodon probeerde en de thread deelde.

Algemene Mastodon resources

Goede algemene Mastodon-info in het Nederlands

• joinmastodon.org
• Hoe begin je met Mastodon?
• Wat is Mastodon? bij Het Laatste Nieuws
• Wat is Mastodon? bij Pocket Lint
• Wat is Mastodon? (commentaar van mij bij VRTNWS, ook bij Humo) (NL)
• Wat is de Fediverse? (Nu.nl)

Goede algemene Mastodon-info in het Engels

• What is Mastodon? (Youtube)
• Video explanation by Savjee
• Gids van Gizmodo: How to join Mastodon
• Gids van @Noelle@elekk.xyz
• 10 snelle Mastodon tips
• What is Mastodon? (The Guardian)
• What is Mastodon? (BBC)
• Mastodon.help
• Goeie blogpost over hoe Mastodon verschilt van Twitter, en waarom
• Algemene tips over de Fediverse
• Reeks artikels van DiggingTheDigital, technischer
• Officiële documentatie van Mastodon, technischer

Mijn Twitterthread (NL)

Dit is een poging om Mastodon uit te leggen voor mensen die van Twitter komen. Het is een ander systeem, en ik begrijp de verwarring die ontstaat!

Twitter is een gecentraliseerd systeem. Er is 1 bedrijf dat de Twitter-server uitbaat. Dat bedrijf (en zijn nieuwe opperbaas @elonmusk) beslist wat er kan en niet kan op het platform, tegen welke prijs, wie wel en wie niet, …

Als zij morgen beslissen dat het gedaan is met Twitter is er geen Twitter meer. Dat zit denk ik niet meteen in het vooruitzicht, maar met die gekke @elonmusk als kapitein gaat het schip hier wel stevig schudden. En klotsen er mogelijk ongure types op het dek. (Ik kan de stukken van Mike Masnick over Musk en free speech en moderatie zeer hard aanraden.)

Mastodon is een gedecentraliseerd systeem. Het systeem wordt geschreven door een community vrijwilligers. De volledige broncode van de software is beschikbaar. Er is niet één Mastodon server. Er zijn er 100den. Elk van die servers heeft z’n eigen spelregels.

Meteen werpt dat vragen op.

Wat als mijn vrienden op een andere server zitten? Hoe ga ik hun berichten lezen?
Geen probleem, je kan iemand volgen die op een andere server zit. De berichten worden doorgestuurd, volledig transparant. Ook als andere mensen ze quoten.

Daarom zijn Mastodon gebruikersnamen opgesteld als naam@server. Bvb “@jbaert@mastodon.social“. Daaruit kan je afleiden dat ik (@jbaert) op een specifieke server (mastodon.social) zit.

Volledige handle (@jbaert@mastodon.social) ingeven in je client en je volgt.

In godsnaam, hoe voeg ik iemand toe van op een andere server?

Vul de volledige handle (bvb @jbaert@mastodon.social) in in de zoekbalk van jouw server, ga naar het tabblad accounts en klik daar op toevoegen.

Hoe kies ik dan een goeie server?
Het maakt dus niet zo veel uit, want je kan mensen over de servergrenzen heen volgen! Kies een server waar de beschrijving en regels je van aanstaan, en bouw je netwerk. Een overzicht vind je op: joinmastodon.org/servers

Enkele Belgen hebben een Belgische server gestart: mastodon-belgium.be

Wat als de server waar ik op zit ermee ophoudt / ik niet meer akkoord ben met de manier waarop die gerund wordt?
Dat is de kracht van Mastodon: je kan je data volledig meenemen en op een andere server verder doen. Er kan niet één iemand beslissen om de stekker uit het netwerk te trekken, en jij blijft de baas over je data.

De opdeling in servers maakt het ook makkelijk om je eigen ervaring op het netwerk te vormen. Als je niks wil horen van mensen die broccoli lekker vinden blokkeer je de BroccoliLovers4Life Mastodon-server.

Wie betaalt dat dan?
De meeste servers worden gedraaid door vrijwilligers. Omdat het verdienmodel niet rond advertenties of privacygevoelige data verhandelen draait kunnen zij voor hun harde werk donaties gebruiken.

Deze mensen doen hard hun best om een omgeving te cureren waar iedereen op die server zich goed bij voelt, en zij betalen voor de serverkosten. Enkele centen go a long way!

Nog een tip voor mensen die liefst op beide platformen blijven
Update: Deze crossposter houdt ermee op. Uitleg en alternatieven hier.

Met deze excellente tool crossposter.masto.donte.com.br kan je tweets (Twitter) en toots (Mastodon) uitwisselen. Instellen wat er met RT’s en quote tweets moet gebeuren, en je priceless wisdom verschijnt op beiden.

Momenteel is het druk
Dankzij de interesse en toestroom hebben vele grotere Mastodon servers (http://mastodon.social/, http://mastodon.lol/, etc …) het even moeilijk en hebben ze vertraging / accepteren ze geen nieuwe gebruikers.

Dit is normaal! Al deze servers worden gerund door niet of -onderbetaalde vrijwilligers die achter de schermen bezig zijn om capaciteit bij te zetten en oplossingen te zoeken. Geef ze tijd en eventueel geld. Een netwerk waar je niet betaalt met je data is niet gratis :)

“Mastodon” is dus niet kapot of down, maar enkele van de populaire servers hebben het moeilijk. Dit is het gevolg van een gedecentraliseerd systeem. Na enkele uren herstelt de synchronisatie zich. Dat is best knap te noemen!

The post Eerste hulp bij Mastodon appeared first on Jeroen Baert's Blog.

I found out the hard way that an Iomega / Lenovo Storcenter ix2-ng NAS doesn’t have any onboard flash memory to store the operating system. So after yanking out and repurposing both the original disks that came with the system I was left with a device that wouldn’t boot or recognize new disks.

I’ve pieced together a way to reset the system using information from MyHackBlog and some experimentation, which I’ll put down here, for my own amusement and to save anyone else some time.

Requirements

• USB Drive (> 4 GB)
• SATA Hard drive (any size > 2 GB should do)
• A linux system where you can mount this hard drive
• Lenovo Storcenter ix2 + power adapter
• Network cable and network (for configuring the device). I suppose this is optional, but there aren’t any other ways to check whether or not the reset has worked.

Preparation

1. Download the zip file containing all tools from MyHackBlog (AllFilesZipped <- direct link). As far as I can see, these tools were obtained through Iomega / Lenovo support at one point, but they are not available from Lenovo support at the moment. I’ve ran all files through ClamAV, and they come up clean.
2. Unzip everything to a location you can easily remember (i.e. Desktop)

Prepare USB stick

1. Format the USB drive to FAT32 (instructions for Windows, instructions for Linux)
2. Run h4c-4.1.102.29716.exe (I checked this with Virustotal: it’s legit, two false positives because it’s a packed file)
3. Select the correct drive to write to and let the tool finish

Prepare Hard Drive

I found out the hard way that you cannot just chuck in an empty disk and let the device format it. The tool on the USB expects zImage and initrd (two essential startup-related tools) to be in the first few sectors of the disk.

The zip file contains compatible versions of zImage and initrd in the \emctools\ix2-ng_images folder.

1. Connect the drive to a working Linux system (where you have access to the devices under /dev). I used a Raspberry Pi with a sata attachment, but any Linux-based system should do.
2. Copy zImage and initrd to that system.
3. Copy them, byte by byte, to the hard disk. Double check which disk you’re copying to! It might (and probably won’t) be at /dev/sda for you! Writing to the wrong disk might render your system unbootable.

sudo dd if=/where/you/saved/zImage of=/dev/sda seek=2048
sudo dd if=/where/you/saved/initrd of=/dev/sda seek=8192

Reset Storcenter ix2

1. Mount the prepared hard disk into the first bay of the Storcenter ix2
2. Plug in the prepared USB stick into the Storcenter ix2
3. Hold the reset button on the back of the device down (using a paperclip or pencil) and hold it for a long time (at least a minute)
4. While doing that, insert the power adapter
5. The Storcenter will start resetting the device. (You can probably hear the hard disk head moving). If your USB stick has an activity light, it will flash.
6. Wait until the Storcenter shuts itself down
7. Remove the USB stick
8. Plug in the ethernet cable
9. Turn on the device
10. Wait a few minutes, and figure out what IP the new device was assigned (using your router or the included ipscan24 tool in the ZIP file)
11. In the web interface, go to software updates and select the ix2-ng-4.1.114.33421.tgz file (this update is known to be good, you can update further later, if required)
12. Wait for the update to install and the device to reboot (remember, the CPU on this thing is pretty slow, can take a while)

And this should get your device back in a functioning state. Now remember, the management software is installed on that first disk. Don’t remove that first disk without adding a second one, or you’re back to square one.

The post Restore an Iomega / Lenovo Storcenter ix2 without original disks appeared first on Jeroen Baert's Blog.

COVID-19 en simulaties

Alle ogen zijn vandaag gericht op wetenschappelijk onderzoek naar een mogelijk COVID-19-vaccin. Een van de pistes die daarin onderzocht wordt is hoe bepaalde proteïnen (kleine fascinerende moleculaire machines) interageren met elkaar interageren, om zo mogelijke methodes te ontdekken waarop een vaccin/remedie zou kunnen werken.

Het probleem is dat er echt een gigantisch aantal combinaties en interacties van proteïnen en hun structuur (want structuur beïnvloedt werking) mogelijk zijn. En een gigantische zoekruimte, dat is waar computers en rekenkracht uiteraard in the picture komen. Niet elk onderzoeksinstituut beschikt over een peperdure supercomputer (of de tijd om daar berekeningen op uit te voeren is zeer kostbaar).

Dit is waar Folding@Home in the picture komt: door de rekenkracht van vele computers te combineren kan er een ongelofelijk resultaat behaald worden: Momenteel haalt het project meer dan een één exaflop!

Sinds de COVID-19 crisis is Folding@Home enorm onder de belangstelling gekomen. Hoe ze onderzoek ondersteunen en wat hun plannen zijn voor de toekomst vind je op hun blog.

Mee folden met je computer

Op de site van Folding@Home zijn clients beschikbaar voor alle gangbare besturingssystemen. De client zorgt ervoor dat er telkens een klein beetje werk (zie het als een klein stukje van de puzzel) wordt opgehaald, je computer de berekeningen uitvoert, en het resultaat terugstuurt.

Je kan de client zo instellen dat hij enkel werkt als jij je computer niet gebruikt (als je computer idle is), alsook voor welke projecten je je wil inzetten. In dezer dagen staat dat standaard op “Covid-19”, maar er zijn ook projecten rond kanker, alzheimer, etc… beschikbaar.

Voor Windows is er een extra applicatie beschikbaar die je in staat stelt om een 3D-visualisatie van het proces te bekijken.

Standaard worden zowel je CPU (processor) als GPU (grafische kaart) ingezet om proteïnen-folding te simuleren. Die laatste is daar een pak sneller in, maar niet iedereen heeft een compatibele grafische kaart.

Je kan de client instellen op light, medium of full. Hou de temperaturen van je systeem in de gaten. Een onvoldoende geventileerde laptop of desktop kan erg warm worden bij het folden – dan is het misschien de moeite om een versnelling lager te schakelen.

Gamification: Join het Nerdland-team

Als motivatie / bragging rights wordt er online een scorebord bijgehouden van wie het meeste “werk” (gemeten in Work Units) verzet. Verschillende gebruikers kunnen ook samenwerken in een team.

Met Nerdland hebben we zo’n team, en ons teamnummer is 239672. Dat kan je ingeven bij het instellen van je Folding@Home Client.

Gedetailleerde statistieken van Team Nerdland en zijn deelnemers vind je hier op de site van Extreme Overclocking, die dagelijks de Folding@Home statistieken uitlezen en in mooie grafiekjes gieten.

Op moment van schrijven zijn we ‘s werelds 288ste beste team – een enorme prestatie! Bedankt aan iedereen die meerekent :)

Je kan voor de lol ook een “officieel” certificaat ophalen op de stats pagina van Folding@Home.

The post Folding@Home met Nerdland appeared first on Jeroen Baert's Blog.

I’ve found a bit of time to work on my CUDA polygon mesh voxelizer, a command line tool to turn a polygon mesh into a voxel grid representation. The tool seems to be used by some people, always nice to see!

• Support for outputting to an .obj point cloud, which is a nice additional format to have (I only had .binvox previously, which is not that widespread)
• Fixed a nasty bug that occured at large voxel grids (2048x2048x2048 and higher)
• CUDA 10.2 support
• VS 2019 Solution

As always, releases and code on Github.

UPDATE 2020 / 06:

Now also with actual models instead of point clouds.

The post Cuda Voxelizer v0.4: OBJ Support! appeared first on Jeroen Baert's Blog.

We all looked forward to the release of cowboy-em-up Red Dead Redemption 2 on PC. The game has a lot of graphics settings options, and has proven to be quite a beast to get running well. The built-in presets are system-dependent, which makes comparing settings and performance between systems a bit harder.

I made a little tool to parse the game’s settings file (system.xml) into a nice text block to share settings with in online discussions. (The tool is based on my previous GTA 5 Settings parser – the XML structure of the engine settings file is very similar)

• The tool is live and can be found here: https://rdr2.forceflow.be/
• The Github repository is here: https://github.com/Forceflow/rdr2_settings_parser

Example output of the parser:

NVIDIA GeForce RTX 2070 SUPER, Vulkan
1920 x 1080, 144hz, Windowed: No, Vsync: Yes, Triple: On

Textures: Ultra
Anisotropic: 16x
Lighting: High
Global Illumination: High
Shadows: High
Far Shadows: High
SSAO: High
Reflection: High
Mirror: High
Water: Custom
Volumetrics: High
Particles: High
Tessellation: High
FXAA: Off, TAA: High, MSAA: 0

Near Volumetrics Quality: High
Far Volumetric Quality: High
Volumetric Lighting Quality: High
Unlocked Volumetric Raymarch Resolution: Off
Particle Lighting Quality: High
Soft Shadows: High
Grass Shadows: High
Long Shadows: On
Full resolution SSAO: On
Water Refraction Quality: High
Water Reflection Quality: High
Water Physics Quality: 50%
Resolution scale: Mode1o1
TAA Sharpen Intensity: 100%
Motion Blur: On
Reflection MSAA: Off
Geometry Level of Detail: 100%
Grass Level of Detail: 66%
Tree Quality: High
Parallax Occlusion Mapping Quality: High
Decal Quality: High
Fur Quality: High
 
Generated with Forceflow's RDR2 settings parser

The post Red Dead Redemption 2 Settings Parser appeared first on Jeroen Baert's Blog.

For the Nerdland Science Podcast (with ao Lieven Scheire), we posed a Traveling Salesman Problem for the song “Ambiance, Ambiance” by Sam Gooris, this connecting popular culture with an NP-hard CompSci problem! That’s why they pay us the big bucks.

More info and Python code on the Github repository:
https://github.com/Forceflow/Ambiance_TSP/blob/master/README.md

The post Solving a Traveling Salesman Problem in Python for fun appeared first on Jeroen Baert's Blog.