Let’s see how long DHH & co can keep harvesting low hanging fruit of Linux laptop problems. I’d expect they’ll plateau soon but I would love to be surprised.
I’m torn between my instinct to classify anything from DHH as mostly hype, my faith in Linux kernel developers, and my cynicism toward Linux kernel developers.
At the very minimum, omarchy and omakub already provide out of the box seamless fixes for all the common issues that are “fixed” but need tedious involved configuration nonetheless.
Honestly, given my experience from distro hopping, I am certain that collecting solutions across distributions and implementing them in one can go very far. It's almost as if distributions contributors too rarely try out other distributions to then steal what the other distribution does better.
Small enthusiast distributions with a bit of a hype can gain good features in by just pulling in knowledgable users missing things from their previous distro - and they can move a lot faster than the Debians or Fedoras of the world can, no committee decisions to be made first.
Windows has automatic Modern Standby to Hibernate support (by default after 5% battery drain in standby). If it is not entering Hibernate, there may be an application keeping the system awake.
This works great for me on my Lenovo, I don't even have to close the lid, it just works. I think I configured it to be a little more aggressive about going to sleep when on battery than the default, but the settings work.
this sounds like a dumb workaround for fundamental issues that should ideally be addressed by beating app, driver, and hardware developers into submission in a way that would probably fall afoul of antitrust law
I’ve been on both the same side and the opposing side of debates with him, both in person and over internal discussion threads. His public persona and private behavior match. I viewed it positively, though per the topic of the thread, not everyone did.
That’s very different than simply observing someone in public. Which is what my post was referring too to and so it remains accurate.
FWIW I like carmack from what I have seen publically (and Romero, who I have interacted with) but I wouldn’t pretend to know who either of them really are from my observation of them.
We're testing those specific modules now. Initial testing shows that it should work the same as on 7040 Series, but we will complete validation before updating our Knowledge Base article on memory compatibility.
Repaste is a likely solution, since that generation was before we switched to Honeywell phase change thermal interface material. Traditional thermal paste will slowly pump out over time. We do have Honeywell material in our Marketplace. You can also reach out to our support team for help.
What's the bottleneck in getting a Macbook like touchpad experience in a modern non-Apple laptop? Is it software? Some specific company that won't sell parts to anyone else? Patents?
I have tried several laptops, and nothing has even comes close in the last ten or so years.
I am hoping you might have some unique insight into this!
PS: Framework Laptop 16 looks great, will order one later this year and then get a GPU with more vram whenever available in future.
If you haven't seen it, this "Linux Touchpad like Macbook" project is related, the last/best effort I've seen in this direction. Here's a random update from a few years ago:
I'm very happy to see Framework lead the USB power delivery pack by supporting 240W, 48V/5A charging on the Framework 16. As the first company to ship laptops using this spec, what problems or quirks, if any, have you seen from 48V charging in the field?
Since we designed both the power adapter and the product that is using it, it just works. There aren't a lot of other devices that support 240W USB-C for us to test against.
How thick must be the cable that delivers 48V @ 5A? I suppose not much thicker than normal, but still? Does this require a specialty cable to achieve that?
Increasing the voltage doesn't require a thicker cable, and 5A power starts with 100W USB-C PD. USB-C cables that support 5A aren't that thick, but they do need a special chip so that the devices know it's safe to send 5A.
I'm very impressed you managed to get nvidia to give you access to the 5070! I have one queetion though, is the 5070 limited to 100W because of the docking connector, or for cooling reasons?
In theory I'm the perfect audience for the Framework 16! The only thing holding me back is the lack of a 4K display. It's so good for dense text on the screen (e.g., code with lots of split buffers), I can't go back. Still waiting patiently for this to become an option.
It would also be a huge benefit to use a replaced mainboard as a homelab base WITH ECC support in the future.
Same goes for the Framework Desktop, which features Strix Halo without ECC support, whereas ECC IS possible with Ryzen AI MAX+ 395+ PRO (e.g. HP Z2 Mini G1a).
Can't speak for Framework, but AMD themselves doesn't necessarily support this configuration so it would be on Framework to develop/test/certify a compatibility list... they'd likely only support first party modules, if anything. At least for non-pro AMD CPUs.
Likely mostly down to resources/time as to the lack of official support.
> Something that we hear over and over again across our entire product lineup is that people want pointing sticks. You might know it as trackpoint from other brands. The little nubs that you can use as a mouse. Obviously, if you're a ThinkPad user, former ThinkPad user, that might be something that you're very familiar and comfortable with. And so, it is something that actually on all of our products several times over the last 5 years, we have tried to prototype and make work. The big challenge on this actually is just that there's very, very little space here. That the Zstack here is incredibly thin.
> And for a keyboard, it works because the keys are compliant. If there's force that's put on the lid, like let's say you're got your laptop in a backpack with a book or something, it's just being pressed on like that, the screen is going to touch the keys and the keys are going to give way because they're just on these uh on these scissor mechanisms and the screen will be okay and you may get a little smudge you have to wipe off. You've got like finger grease on there.
> A pointing stick though is not compliant. Not compliant in that way. So, you've got this like sharp point basically sticking out from the keyboard. And if there's pressure placed on the lid, that's going to go right onto that point on the tracking stick and end up damaging screen or have a high likelihood of damaging the screen.
> And so, we've just kept over and over, we've kept trying this and seeing if we could get a low enough profile pointing stick solution to make that work, not risk the screen at all. And so far, that doesn't exist. That is something that we keep going into the supply base to try to find.
> Hopefully we that is something we find in the future because of course with this input module system on framework laptop 16, it would end up being relatively straightforward for us to just make an input module a keyboard that you can swap in that's got that pointing stick unlike uh you know even our other laptops where you'd have to have an entirely new input cover to get that kind of functionality.
Most laptops have rubber bumpers around the screen to create a small airgap when the screen is closed, and even if you stack several laptops on top of each other, the bottom one's screen won't flex enough to touch the keyboard. Maybe with the more rigid lid, they can enable this?
Given the longevity goals with their modular designs, I'm guessing they're unlikely to make dramatic changes to their overall design in less than 5-7 years time from initial release. Such a screen change would likely require new hinge design as well as a thicker display casing, not to mention the risk of someone putting the trackpoint keyboard in a model that doesn't have the thicker display section.
I am thinking that something with a nub on a 2-axis slider as opposed to rocker switches could be an option, but that would potentially have drift issues. Not to mention the Framework keyboards themselves are probably mostly a COTS solution, where something like I'm thinking would require custom R&D an likely be limited release. If Framework, Dell and Lenovo could work together, they could probably come up with a good solution... though Lenovo likes the Fn button in the corner, where most others prefer Ctrl then Fn.
Yeah, nowadays the trackpoint is just a bad pointing device. As laptops get slimmer, trackpoints get less precise and less usable. And they hurt our wrists.
We just have to let go. A haptic trackpad is miles better now.
I've used it on a Dell laptop too, but it was far inferior to the Trackpoint. I think IBM/Lenovo had a patent on the specific technology but it might be expired now.
I don't think more modern ThinkPads are much better... they have a shorter keyboard depth than the earlier models and the feel overall is significantly reduced... I understand as most people don't generally want a "thick" notebook.
Awesome job on following through with the upgradability! I love the nvidia support!
Are you coming out with another coolermaster case for the 16 mainboard?
I want to make a custom dock with fans to force more cooling over the radiators. Could it be possible to "unlock" the 100W TDP of the 5070 in firmware or are there other hardware limitations like the interconnect?
Was adding the USB C power input on the GPU necessary to get full power? I see in the specs on github that VADP_GPU can take 100W into the mainboard and VSYS_GPU can supply 240W to the GPU. Are there any tradeoffs powering the system from the back ports vs the GPU?
Was the previous version of the AMD GPU not sending the display signal directly to the panel via the edp mux but instead via the igpu? If not is that something you can update in firmware? Can you publish how this was done so someone can make an oculink expansion board with displayport input?
Thanks to everyone at Framework for making such awesome hackable products!
I would love more keyboard options! If I could have a laptop with a keyboard more like a standard tenkeyless, I would be over the moon. In the meantime, I'll have to see about hacking the standard keyboard somehow, although there's no fixing those arrow keys.
Wouldn't it be possible to have a side add-on similar to the numeric keypad but the width of the spacer or LED matrix to accomplish this? Maybe a revamped keypad that has home, end, page up, page down, and then a bunch of assignable macro keys would work well.
You can upgrade to any one of the new parts. In the Framework Marketplace, you can set compatibility filters to see the new parts (some of which are on waitlist because we'll be shipping them later this year): https://frame.work/marketplace?compatibility%5B%5D=laptop_16
- Bazzite takes community contributions; whereas, SteamOS is packaged and distributed by Valve, and
- Bazzite is based on Fedora, so the work to support Fedora should bubble over to Bazzite.
I'm curious though, is there a big difference in functionality for SteamOS vs Bazzite. Are there things that work in Bazzite that wouldn't in pure SteamOS?
We manufacture our systems and many of our modules in Taiwan, and have had less tariff impact than other electronics companies as a result. Currently, laptops are also exempt from some tariffs.
Not specifically about the new laptop but will you guys ever do a 15 inch non-gaming laptop that looks similar to a macbook?
I really want to buy one of your products but 13 inch is a bit too small for me while the Laptop 16 is a bit too bulky.
I understand there's two versions of requirements for the NVidia 50 series - the higher end 5070Ti and up, and the lower end 5070 and down. What's the chance of releasing a 5070Ti/5080 version?
I wish I could buy it, but I'm in the unfortunate situation of being in Norway, instead of almost any other European country. Will you ever ship here? ;-(
Is it possible to put the trackpad "above" (as in closer to the display) the keyboard? This is to me the obvious correct place to put it, but I'm in the clear minority.
I literally just want a touchpad with buttons. These new 'clickpads' are the bane of my existence. They are so much slower, and certain workflows are impossible. I must use an external mouse now with modern laptops.
Why can no laptop manufacturer even make this an option?
Aside: what if frame.work site had a place for popular vote for features? (With proper registration, etc.) E.g. Digital Ocean has this, and it seems helpful, they follow up on some of the most upvoted feature requests.
It's sort of free market research.
Kickstarter style upfront payment may be more useful. You want touchpad with buttons; it would cost 300k to develop if enough people paid, the company would build it. If it is a niche, let open ecosystem take care of it.
Can only speak for myself, but for me the issue with traditional clickpads comes down to their mechanical diving board nature. Even the best ones are not nice to use due to the unavoidable variance in pressure and click feel across the pad that is exacerbated as the size of the pad increases and the mechanism wears over time.
The type that doesn’t move at all and simulates a click with haptics on the other hand I find just fine. MacBooks do this of course but there’s also a few x86 laptops equipped with pads like that.
So in my opinion, mechanical clickpads should disappear entirely and laptops should offer two options: a static haptic clickpad and traditional trackpad with buttons.
Honestly, I'm here half wondering why we need the click at all. One finger drag for move, quick one finger tap for left click, tap and half for click and drag, two finger tap, two finger drag for scroll covers all the common interactions.
Which isn't to say I don't use the click functionality at all. I will subconciously use it in some scenarios, but not in others, but if it were missing I would adapt very quickly, since I use the gesture alternatives so often, that I would automatically fall back to them.
I suppose I need the click for some obscure interactions like right click drag, but honestly except in games I've almost never seen that used. My surface laptop as currently configured literally wouldn't even allow some other rare ones like hold button and scroll (I'd need to turn on right side scroll-wheel for that) and I've never even noticed the absence of that ability until I tried it just now.
I've missed them every time I've been in the unfortunate position of dealing with someone else's macOS system. It's all a matter of what you're accustomed to.
Hardware-wise, no, I've had plenty of PC trackpads that are better than Apple trackpads. But MacOS tends to have better built-in support for advanced gestures, which seem to be impossible on Windows and must be manually configured on Linux (but gives you enormous power once you do).
Apple's palm rejection is also top tier, though other systems have been getting better. My current Dell seems fine so far, but at my last company the Dell I had was almost unusable due to my cursor just teleporting around my document randomly if my hands got too close to the trackpad (which is where they have to be to type).
Not sure if it's a hardware (Dell) or software (Ubuntu) improvement, but thank god.
You could try putting a trackpad from a macbook into the framework. AFAIK the palm rejection is all in the firmware. The apple trackpad is USB. If you look at the code for Asahi Linux it could tell you more.
... it's a software problem afaik. The trackpad may be slightly better quality, but it's the drivers and the OS integration that make even some games playable without a mouse on Mac OS.
Don't think any one x86 laptop manufacturer can fix it.
I assume some gestures are simply not possible. Like click-to-drag and scroll simultaneously. Not every app handles gutter-hover-to-scroll in a usable way. On a mouse or a pad with buttons, you can keep the left click held down and scroll with the wheel or gesture. Uni-pads make this impossible.
"Without issues" is a stretch. You need to use two hands or be skilled with one. Its trivial on a mouse or a pad with a discrete buttons.
But ok, what about just dragging a long distance where you would normally lift the mouse or finger? Is there some hidden gesture for this? Maybe once your initial drag finger hits the edge you need to use two more to do a move gesture? But I've seen that trigger scroll and/or pinch-to-zoom.
This reminded me of a feature in windows where if you were dragging something and reached the end of the touchpad, the cursor would continue on the same trajectory as long as you kept your finger at the edge of the touchpad. Then you could overshoot a little so you could bring your fingers back to the middle to regain maneuverability. I haven't missed it since I switched to linux but now that I'm thinking about it that was a very nice touch (no pun intended).
> what about just dragging a long distance where you would normally lift the mouse or finger?
This is why you set Trackpad speed to "fastest", and take advantage of the aggressive trackpad acceleration. When you move your finger quickly you'll easily reach the far side of the screen before your finger reaches the edge of the pad, and slow finger movements will still be precise
I'm probably missing some context, but on my Mac I'm using three fingers drag and I can lift fingers and (quickly) reposition them without breaking the drag.
I have the external apple trackpad (the most recent usb-c version at that) and this click-to-drag and then attempting to scroll does not work on Linux. Seems like this might have been a particular attention to detail on part of macOS devs.
As far as I know touchpad implementations just report finger locations and its up to software to interpret what a combination of these gestures means.
How does that work? You've got to tap the touchpad to trigger the initial click, don't you? For some reason, I really HATE tapping a touchpad (let that be an Apple or otherwise), it breaks my flow, I suppose? (like, you have to pause at the cursor's location, lift, tap twice to initiate a dragging event, then finally move on) whereas on the ThinkPad I daily drive I do all the cursor movement/scrolling with my right hand and the selection/clicking with my left thumb on the physical key that sits on the top of the touchpad sensitive area. That makes click&drag workflows super efficient, I find.
You click and drag with one finger and you are free to scroll with two other fingers during the drag. It is a multitouch gesture. (I don't use "tap to click" since I always found it cumbersome)
Just for me to understand, you navigate to that thing you want to drag, then press harder (without the double-tap+move in short sequence, do there's that), and that registers a drag event?
Could you do the "press harder" part with, e.g. a thumb in an other region of the touchpad instead of the finger that did the navigation?
Yes, you can. As long as one finger is "pressing hard-ish" a second finger can command the drag position, but if the finger that is pressing (you do not need to press very hard to trigger a click) is not the one that is also moving, then you will have issues when also scrolling with two other fingers, because at that point you have 4 fingers touching the trackpad, and by default you get anoter gesture registered (probably a zoom out to see all the windows in the workspace, called "expose"). If the fingers touching the trackpad are "only" three, you can drag and scroll, with the window that receives the scroll being the one under the pointer/item being dragged.
Thanks for clarifying, it seems analogous to having a physical mouse button, then (except that the haptic feedback is simulated, which strangely isn't off-putting to most, I've personally always felt the sensation uncanny)
> Why can no laptop manufacturer even make this an option?
Because it’s a variation of both the case and the internals that brings a higher failure rate, more dust ingress, more moving parts, and, most importantly, would rarely be chosen.
> They are so much slower,
They are objectively faster because you can click anywhere rather than moving a finger to a button or keeping one finger always on the button.
> They are objectively faster because you can click anywhere rather than moving a finger to a button or keeping one finger always on the button.
I have no problem with the current trackpad (and prefer it), but when I used a trackpad with dedicated buttons, i'd use my index finger to track and my thumb to click, so I wouldn't have to move my fingers around at all.
Regardless, why do we feel the need to argue with people's personal preferences? You don't have to agree with someone on this. It's fine. People can prefer other things.
This is my issue! All these years later, I am still not used to it, and I accidentally trigger multi-tap nonsense that I didn't intend because I am trying to click with my thumb.
My only point of reference is my MacBook Pro trackpad but I have no problem using my finger to move the cursor with my thumb resting on the trackpad and clicking. Not exactly the same because you lose some of the tactility of discrete buttons but it seems somewhat possible at least.
Yeah, most of the non-apple implementations of this don’t work with using a second finger to click in my experience. They expect you to click with the pointing finger.
Someone should scoop up the niche market of anguished ThinkPad devotees, with a TrackPoint and a good, non-chiclet keyboard. Maybe Framework, leveraging its modular system. Maybe a Framework-compatible third-party.
I just watched Framework's other video talking about what they're working on[1], and they mentioned they're trying to get the trackpoint working, but they're having issues with it damaging the screen due to the low clearance.
I think you're supposed to touch the touchpad in particular ways with your thumb, while using your index finger on the TrackPoint stick.
I wonder what the set intersection is, between people who want TrackPoint sticks, and people who don't want TrackPoint buttons.
ThinkPad industrial design the last several years seems focused on looking thin and sleek -- like an Apple product, only in matte black, with a red accent in the middle of the keyboard -- but some of the human factors changes aren't intuitive to me.
Current models of the X1 have the option of having buttons or not. Which seems fine, but hopefully it isn't a sign that they're trying to drop the option of buttons.
Some Dell business models have them as well... I used to be a fan, but at this point I prefer the Mac touchpad experience. The closest I've felt are Razor and a few higher end Chromebooks (that I won' t buy). I'm hoping other mfgs get a lot closer to the Apple touchpad experience as patents start to expire in the next few years.
There's a few that are close, but still not close enough. Also, Mac slightly changed their default settings (regarding the physical click behavior), I never recall what it is but only that I change it back when starting out on a new machine.
With the trackpads that have built-in clicks in the pad itself, I've always found it really difficult to drag-and-drop stuff if it has to be pulled longer than a few pixels. Just moving and pressing against a surface seems to not be a super accurate movement in general.
Although it gets buggier with every release, macOS has a three-finger-drag operation, and there's a grace period when you lift your fingers if you need to adjust your position over the trackpad. It also lets you just fling one finger.
I don't think I have, long time ago I had to use a laptop myself, just remember that being difficult last time I had to do it together with someone else. Probably depends on the software/hardware itself also, how well something like that would work.
Have you? The precision vs a touchpad with buttons isn't even close. It may well be a driver issue in Linux, all I know for sure is that it's an issue that does not exist with touchpads and that I have already spent far too much of my life fiddling with settings trying to get it to behave.
I never had problems with precision on HP high end models, Windows XP up to 2008 and several Ubuntu versions since then to 2022, then Debian.
I own an nc8430 and a ZBook 15 first generation. I use the lower row of three mouse buttons as left, middle and right click. Those touchpads don't move and don't bend. I disabled tap to click as buttons are much better and never move accidentally the pointer by design. Palm detection works very well, basically no issues. I use two finger scroll and pan. Several gestures work but I don't really like them. I disabled everything. I rather use keyboard shortcuts. I defined some of my own especially to navigate among virtual desktops.
There is still one ZBook Fury model with buttons, every other model lost them.
I have, yes! I do it a lot when dragging and dropping on Linux. Start the drag with one finger, and then bring a new finger in to move it. It's just like using a button.
I've not had to configure anything to make this work for a number of years now in Plasma. Though I've been running Linux on Macbooks for a long time, so maybe it's about specific hardware support.
While I prefer an external mouse, I can manage ECAD and some 3D modelling if I have buttons. It's great in a pinch. I'm getting nauseous even imagining it with a clickpad.
My current plan is to retrofit buttons to my clickpad. Earlier this week I ordered a few different styles of touchpad buttons from AliExpress to test. I'll build a custom little USB HID device for it once I've picked my favourite one.
I don't think I can rely on laptop manufacturers to buck the clickpad trend any time soon, so I'll do it myself.
> These new 'clickpads' are the bane of my existence.
But only because they are all worse than Apple's version. What you really want isn't a touchpad with buttons, is a "clickpad" that doesn't suck. And as far as I know only Apple makes them.
Not a question about the Framework Laptop 16 specifically, but why are the upgrade kits[0] for the Laptop 13 still marked as "register interest" on the Framework Marketplace? The Ryzen AI 300 motherboards, RAM, and Wi-fi cards are all available separately already.
