I purchased both the 7900xtx and a 4090. Kept the 7900xtx. Absolutely adore this card. Being a Linux nerd made it a no brainer, but the performance vs price ratio wasn’t there.
Fuck that thing. THREE slots? I had to abandon one project and look for alternatives because my GFX card already block one slot from being used, but blocking two is crazy.
Speaking as someone else who had a 7900XTX, this was sort of common. It’s just that a lot of game devs aren’t designing their games and apps with anything AMD (maybe Intel too) GPU related. The market is mostly Nvidia so that’s what their games are looking for when it tries to search up compatibility.
Most of the time, it’s just that they don’t officially support it but it still runs without any issue. Since their game is looking for “Nvidia” but sees “AMD”, it’s assuming you aren’t using a compatible GPU. Even more so because Alyx came out in 2020(?) and the 7900XTX came out in late 2022 so the game really doesn’t know that this GPU exists.
The game/app runs without any issue. The game just doesn’t have this specific card to check for a quick compatibility and throws this error as a “just in case”.
There were some rare times where the game/app legitimately did not run. The Oculus app was one of these. VR worked perfect fine through SteamVR but never with the Oculus app since that app was designed for Nvidia GPUs.
I've got it on ultra settings and have never noticed a frame rate drop. The game is around 4 years old and has a number of bugs, Valve kind of just abandoned it despite it being one of the flagship VR experiences. The software not recognizing a CPU that came out 2 years after launch and perhaps a good six months since it's been updated isn't a huge surprise.
I'm using a sandwich style case with the PCIE port at the top of the case and the 3 scews for the slots are at the bottom. It basically pushing the card up and backwards into the PCIE extension cable's slot. No ass sag, but I can tell there's a limit to how much it would be able to handle like that. I've got a 7900xt (not xtx) so it works for what I need right now.
I'll get it shared tomorrow. It's designed after a bottle jack but is completely threaded to allow for adjustment and doesn't have any of the handle components, obviously.
I may be misinterpreting what you're describing, but it kind of sounds like you've reinvented the machinist jack.
Not throwing shade, it makes sense and there's a reason that machinists the world over use them to tackle the same sort of problem of how to support sagging parts- it's simple and it works. And coming up with the same solution shows that you have a good understanding of the issue and how to tackle it.
Just pointing it out because I love the overlap between different fields and hobbies, and maybe if you didn't already know that machinist jacks were a thing you'll find inspiration for a version 2.0 if you ever feel the need to make one.
I'd also like to see your version when you get around to sharing it.
And yes, I designed this in a vacuum, but it is actually a common machinists jack, it seems. Minus a few neat features I saw around the web (like a lock screw component) it's a 1:1 match for functionality.
Like you, I have a love for projects and tools that are shared across different fields. Mostly due to my ADHD, I have a slew of hobbies that focus around mechanical engineering, chemistry and biology. It's extremely cost effective to commingle different hobbies, actually. My 3D printing and CAD work is universal, my collection of chemistry glass is used for rando chemistry things as well as mycology. All of my fittings, connectors and electronic components are also universal... etc. etc.
I feel like it's a pretty common experience for a lot of us maker types- we come up with a solution to a problem, maybe we tried searching for one and came up empty, or maybe we just thought it would be fun to make from the get-go, then sometime later we stumble upon someone selling pretty much the exact thing we made and think "huh, so thats what these things are called" maybe with a touch of disappointment that your idea wasn't as original as you thought, or maybe just intrigued because you just added another term to your vocabulary and you have a better idea what to search for next time.
Yours looks good though, It's probably unnecessary but I'd be tempted to add like a U-shaped cradle piece to the top to help keep it in place if your computer ever gets jostled for any reason, and maybe a jam nut at the bottom to make sure it stays where you set it. Again probably totally unnecessary, but I always figure that if it's worth doing it's worth overdoing.
What sort of printer and filament/resin are you using? I'd worry a bit about it warping or sagging over time from the heat inside a computer. I know some materials can handle the heat better or worse than others, but I haven't dipped my toes into 3d printing myself yet, so I may be overestimating how much of an issue it is.
And can your printer make decent screw threads, or do you have to clean them up afterwards with a tap and die or something? I'm not really up on the current state of 3d printers, but one of the first 3d printed objects I remember ever handling was back in high school 15 or so years ago. One of my teachers went to a conference where they were showing off new gadgets for computer and shop classes, and he brought back a couple 3d printed crescent wrenches for us to fondle. I remember the screws being really crunchy and they almost but didn't quite work, so in the back of my mind I've always thought of functional screw threads as something 3d printers can't quite do, so it's wild to me if we've gone from barely able to make even a coarse thread with huge tolerances work to being able to make pretty fine threads with pretty tight tolerances.
Of course back then, they hadn't even really settled on calling it "3d printing," I remember that teacher calling it a "rapid prototyping machine" when he was telling us about it, and described it as being "like a 3d printer"
In addition, I can't tell people enough that most "specialty' tools are most likely common tools from other fields, but maybe include a fancy adapter or something. And yes, I don't want to know how many times we have solved problems that were already solved. (A benefit is that we understand the true problem to its core and integrate those solutions much more efficiently into future projects.)
My "jack" is printed in PETG as it has a higher glass transition temperature (80°-85°C) than PLA. I have had PLA prints on my shelf start to sag due to age but most of that is likely negated by additives in new generations of PLA+. (or PLA Pro, or whatever buzzword is currently used.) PETG should be just fine for this. ABS would be another option, too.
I currently have a Prusa Mini that is my go to, but a large format Prusa XL with 5 separate hotends is squarely in my sights at the moment. (I have had several other printers that eventually got disassembled for parts.)
The threads on my print are super tight and no post processing was needed. It just screwed together after it was done. I did go to the trouble of designing a shoe looking adapter thing for more surface area against the card, but the weight of the card is enough to keep it from moving around. The thread angle is such that the downward pressure keeps enough friction on the threads so it's not going anywhere through any jostling. (Rough threads are a feature, in this case.)
But yeah, printers these days are much more accurate and less prone to errors. Just 10 years ago, every other print I did was a failure. Now? Failures are quite rare and that is mostly because of the tech. Now that layers are consistent enough these days, most threads I print are with 0.2mm layers which is good enough if I keep the tolerances wide enough in CAD. (My go-to is 0.25mm tolerance for fitted parts or less depending on the application.)
Yeah! I remember those wrenches quite well. Print-in-place models are somewhat mainstream now. I randomly put one on Printables and it is far more popular than I ever expected. The success rate seems quite high given the number of random printers that are out there: https://www.printables.com/model/489431-print-in-place-hinge . I only bring this up as it's a testament to how far the tech has come, and only recently.
I literally just got my first resin printer and have yet to have a failure. Getting consistent prints at 0.03mm has been a breeze and quality is limited by the model, not the printer. (Resin printer fanatics are likely to argue with me on that point.) Getting ABS plastic strength with hyper-detail is amazing! However, it's messy and smelly, so that sucks. (I am going to setup some real precision testing this evening as I have had some projects on hold for a couple of years due to FDM limitations.)
Not to compare, but I remember I had put a few olden (20 years old at least) lego bricks from my childhood together and made a support castle underneath an XFX brand 6700 XT with 3 fans.
