CAD files for the external shell (surface topology) of Steam Deck
gitlab.steamos.cloud
gitlab.steamos.cloud
The fact that they wrote this into the readme, even though strictly speaking it's unrelated to the outside topology, makes me wanna work for valve.
They're seriously one of only a few tech companies I really admire, and that's despite the fact that I don't have much inclination towards gaming.
Thus, they can afford to set themselves apart from the likes of Apple and tell their customers to have fun.
Pessimistically, I imagine that if their hardware business boomed exponentially, their support costs would rise and the hardware designers would be pressured to reduce PEBKAC risks...
Unlike most companies we discuss here, Valve is not publicly held, and doesn’t seem particularly interested in being so. That, combined with their dominance of the PC gaming market, means their culture is more free to let them target things orthogonal to making money, when the whim strikes them.
The public aim to make as much money as possible or at least increase valuation. The private however can be content with making enough of it. Which can really change the long term plans.
Also being public company and not increase enough valuation always runs the risk of being acquired and someone else doing it.
Speaking of which, there’s a lot of negative sentiment around these parts for Apple and Google taking 30% of their stores (“rent seeking”), but not so much around Valve for doing the same. Possibly because HN skews more towards app developers who pay fees to the former and less towards game developers.
That alone makes a huge difference and is nothing that either Apple or Google offer.
As an existing Steam user though, this is great for me! It means one less third party DRM platform I need installed.
And regarding DRM: People often forget that the use of DRM with Steam is in the hands of the Devs/Publisher. There are DRM-free games on steam. Now, how convenient it is to back up those games is another thing, but it’s often possible.
The PC remains an open platform, even MacOS to a degree (albeit apparently less so as of late...) and Valve have aggressively campaigned for it to stay that way. Competition there is possible.
I remember the debates from ~10 years ago around Microsoft's moves towards a more AppStore-like experience on Windows, where Valve really led the charge against it. The move towards SteamOS and Steam Machines seemed to be part of their reaction at the time. And it seems to pay off big-time now, not just for them but also for users, Linux users especially of course. :)
That said, negative sentiment about Steam has been a heavy subject in years past. There's a reason Epic came in swinging with super low fees.
The other part is that you have many actually useful alternative avenues to selling PC games. Yes, Steam is far and away #1, but lots of devs go direct to consumer themselves, and some use alternate stores like Epic, GOG, or Itch.
Having a mature linux support helps with that.
In the case of SteamDeck the lack of Microsoft tax also helps the price.
Its tool for doing that is consumer sentiment.
This all plays into Valve's business strategy. They're fighting to justify ye ole 30% cut.
That is the tool for basically doing anything productive as a business... it isn't as if they've done some about face at the emergence of genuine competition - they've always been unusually well liked (especially considering the industry).
And fortunately for Valve even 3 years after release Epic Games Store don't even have half of Steam features. Also Epic is so consumer-hostile (or as they call it pro-developer) that they won't ever implement player-reviews into their system.
So it's not that hard for Valve for keep their mindshare. Far larger danger for them is Microsoft with it's Xbox game subscription model and unlimited amount of money.
Meanwhile the large developers, the ones who are making $50M+, those developers have enough pull to make the user install a custom client like Origin. That’s why Valve cuts them a better deal.
Disclosure: I work at small indie PC gamedev studio.
I just point out how Valve changed rules for big companies back in 2018. So they have different revenue share compared to both Apple / Google and consoles.
However, there are many times more small indie titles than there are big successes, and it's the big titles that draw users to your platform and bring in the money.
If Valve flipped their model, those indie titles would be even more of a net drain on their profitability, while there would be more incentive for the big titles to release on their own platforms.
> Furthermore, please be aware that opening the outside enclosure of your Steam hardware will void a warranty you may otherwise enjoy (https://support.steampowered.com/kb_article.php?ref=4577-TUJ...).
I'd say that significantly diminishes the good will statements like "you have every right to open up your Steam Deck and do what you want with it." buys them though.
Allowing you to make the choice between hacking or not seems like perfectly good will to me. You can't have your cake and eat it too, as they say.
I don't expect a warranty from that, but I very much do expect a warranty if I crack it open and don't break something while messing around.
Thankfully the actual warranty doesn't say it's voided by opening.
Some of the exclusions on there are worrying, though. Improper cleaning? Refurbs have no warranty? "Commercial use" is much too vague.
So yes, opening it up and improperly cleaning it (for example, you take a q-tip with rubbing alcohol to clean off some gunk and accidentally dislodge a capacitor) could void your warranty.
----
[0] https://www.ftc.gov/news-events/press-releases/2018/04/ftc-s...
My complaint is that the warranty page only talks about whether you caused the damage for the clause about opening it up or making modifications. The clause about warranty loss for cleaning "in any manner other than as specified in the Hardware manual" is unconditional.
But when the put a screw back too tight, which then a year later cracks the plastic support its screwed into, falls out, and shorts some terminals on the voltage regulator, and the whole thing goes up in smoke...
I’m sure that if people opening their devices, breaking them, and expecting warranty to cover it is such a big problem, then there is enough financial incentive to find an actual solution.
> you have every right to open up your Steam Deck and do what you want with it.
We already have this right for things we own, on exactly the same terms, which is it voids warranty.
https://www.ifixit.com/News/11748/warranty-stickers-are-ille...
"Any damage you do will not be covered by your warranty – but more importantly, you might break your Steam Deck, or even get hurt!"
My point is that you're probably better off being repulsed at someone else because in the scheme of things Valve really isn't that bad, IMHO.
I have no doubt that some people spend money they can’t afford to even in Dota 2/CS GO, but that feels more of a choice than other games.
You must be kidding. I mean at that point you'd hope some preservation instinct kicks in and you realize you are trading real dollars for a few bits on a server, and you go buy a chess set and play it down at the park.
It is amazing to me that people will just walk in with their pay check and leave drunk with pennies. Some serious neurochemistry going on there.
It also does not account for the presence of microtransactions influencing the design of the rest of the game towards selling more microtransactions which affects you even if you never buy anything. Granted, this could be considered as "payment" for a free game but once microtransactions/lootboxes/NFTs/whatever have or will become acceptable there, they spread to paid games too.
...which is a pretty fair business in my opinion.
And beyond that, the rest of the game is not being designed independently of microtransactions and that affects you even if you never buy any.
They certainly weren't on a blockchain though, or give the user exclusive ownership over them.
That being said... it was a lot of fun until I got frustrated with it! It's on the PC game pass.
But, yes: if you change the inner mould line there's the risk of some components not getting the airflow they need. It might not even be obvious, as not everything has a temperature sensor.
Then again the eventual 'Water-cooled Steam Deck' modders would know about this already(Water-cooled Xbox/PS have very high memory chip failures due to improper cooling).
But TBH, A compute hardware which is designed to be opened from a mainstream company is something I never thought would see again. I really hope general consumers dig into this and demand it from others.
The complement of video games is video game consoles.
I'm not surprised, use of NX or SolidWorks is expected for such products design.
> That's according to the internal info side the .stp files
2D drawing also produced with NX (according PDF properties)
ADDITION: Just tried to open this STEP-file with gCAD3D[0] CAD app — no luck.[1]
[2] https://github.com/FreeCAD/FreeCAD
[3] https://github.com/realthunder/FreeCAD_assembly3/releases
That it produces decent Step files which FreeCAD can open is even better. :)
If I can also dock it to my screen, keyboard and mouse then I can see myself doing some light development on it too.
If I end up getting one, I'd probably install Arch over the Arch-based distro they use. I don't mind the OSTree approach (I guess it makes sense for consumer tech), but I'd like to get the most out of that thing; ProtonGE and MangoHUD included.
Is that really relevant? The Steam Deck is not trying to be Desktop Linux - it boots directly to Steam and most users won't ever care what runs underneath. Just because you can run other Linux programs does not mean that problems with Linux (even the self-inflicted kind) will apply to the Deck.
I'd like to think so. Linus' conclusion was that Linux is inappropriate for any PC gaming, so I'd really like to hear his line of reasoning that has given him newfound hope in the platform if his own experiences were so rough.
> it boots directly to Steam and most users won't ever care what runs underneath.
Correct. It still uses DXVK though, and it's arguably even worse than desktop Linux for gaming since it's harder to tweak the files that might be necessary for smooth gameplay. We'll see how things turn out, but certain titles like Fallout 4 and the Battlefield games are borderline-unplayable on Linux without hacking a few INI files.
Valve can now easily claim that their CAD files were used even if they were not.
(I haven’t looked to see it the supplied files include enough of the interior to be able to produce parts that properly fit all the required internals…)
CC-BY-NC allows you reproduce even modify for non-commercial use, so each user may even 3D print it for yourself or create a gift for friends.
Edit: …in an automated fashion. For example, is there an app where I can put an object on a turn-table, point my phone at it, and give it a spin? (Clearly that would work only for certain shapes.)
For something like this, use a contour gauge at certain offsets from a set point will give you a set of curves you could model through.
You can, if course, just measure and draw it out yourself. But the rounded sides are likely going to be guesses.
So, major kudos on this release. Could probably just sketch in blender and then just standard difference tricks to put your case on one of these.
Use a caliper measuring device and a flatbed scanner.
With the flatbed scanner, make a high res (600 dpi, or maybe even 1200 dpi) 2D scan of the front of the phone. Keep the screen of the phone on while scanning, and have it display some image that contrasts well against the edges of the screen, so you can clearly see where the edges are. (Ie if the phone is black or other dark color, use a an all white image. If the phone is a bright color, use a dark image.)
Scan the back of the phone as well, so you can see where the camera lenses on the back of the phone are.
Import the 2D images into your 3D modeling software of choice and create a flat polygon covering the front view. Extrude into a 3D shape.
Scale according to physical width, height and depth of the phone that you measure with the caliper.
Make room for the lenses using the scan of the back of the phone.
Buttons on the side and charging port on the bottom you can decide on how to fit from using caliper measurements, and adding some additional room around those as that is often desired.
Then model a shell that fits around it.
It will probably take a couple of test prints and refinements to get it right even so, but I think said approach will work well in terms of benefit to effort.
That’d work pretty well for phones (like the PinePhone) that are basically “rectangular prism slabs with uniformly rounded edges and corners”. It’d be much harder to model this Samsung S4 I have here, with its curved ends and it’s compound curved back…
Take a photo of the object from all angles, I usually end up with 40 photos. Don't move or spin the object, the static background details help align the images and you want the lighting to be consistent. Smartphone cameras actually work best for small objects because their tiny aperature keeps everything in focus.
I've used the turntable laser scanners (NextEngine specifically) but it is far from automated, requires hand alignment of point clouds. Photogrammetry approach was much lower effort.
But really this gives you a mesh, which is far from being a CAD drawing. Taking measurements and drawing from scratch is usually how I've seen it done when accuracy matters.
You touch the object in as many points as you need and translate the touch coordinates from the angles of your tool's arm to XYZ. Or some systems allow you to do it optically.
I've worked with these, they are super cool. Just the laser part of the hand-held scanner was detailed enough to show the lettering on a laptop keyboard because of the differing depth of the molded-in characters.
Here's a random video of someone doing that:
It’s also great material to learn if you are interested in mechanical engineering.
I’m not saying that every company should do it, or that they are better than others. For them it’s really not something that makes money or is a core product, so they probably don’t care so much about competitors etc. Also even if you mess with it and they need to replace it while on warranty, it’s still probably ok. They want you to buy games, not the console.
So they are not some messiah of eco-socialism that will save the world from greedy corporations. They just want to make some more money :) but that’s ok, and I appreciate they did it.
This goes for the gameplay and physics too - yeah, VR makes more things possibly but there really isn't anything revolutionary in in HL:A compared to other VR games.
Meanwhile player-initiated exploration has been replaced with an in-game currency that you need to collect to upgrade your weapons which makes it a chore rather the excitement of coming across a lambda cache. Speaking of weapons, there are far fewer than in previous HL games and the ones that are there are hardly anything new. Even the gravity gloves are not much more than a VR-adaption of the gravity gun - going as far as bringing back the supercharged version from EP1 for the final part of the game which is really fun but nothing revolutionary or creative.
Instead of level-specific puzzles, the game is filled with same-ish hacking minigames that while cool at first are overused and as a result are just an annoyance towards the end.
The choice of making the player character voiced in a medium that is all about immersion when previous games already had established the trope of silent player characters was also puzzling - it doesn't help that the Alyx's lines are pretty meh in this game and often directly contradict what the player might feel/want.
Graphics are impressive, but they are also cheating here by having many dark corners that you can't inspect because your flashlight only magically turns on where the game wants it to.
HL:A is a decently polished experience, a good game and certainly better than no Half-Life game, but hardly anything worthy of being called Half-Life 3, which is why it isn't called that. I won't go as far as calling it a soulless sequel that you might expect from a big publisher trying to milk a franchise, but I am also not calling it the opposite of that.
Finally, HL:A is also a VR exclusive requiring an expensive PC, additionall HW and physical space to play it in, which means for most people it might as well not exist.