I have a Blackmagic Intensity Shuttle (Thunderbolt) and a Framework Laptop 12. I’m aware that the FW12’s USB-C ports do not officially support Thunderbolt.
I’m wondering if there’s any way to get my Intensity Shuttle working. My first thought was to see if it’s possible to create a custom expansion card that could enable Thunderbolt functionality, but I suspect this isn’t feasible since the motherboard itself lacks the necessary hardware.
Has anyone in the community explored this or perhaps found a creative workaround? I have a couple of specific questions:
Has anyone managed to get any thunderbolt device working on a FW12 in any capacity?
Is there any external card or adapter that could connect my device, despite the lack of native Thunderbolt support on the motherboard?
I know this is a long shot, but I’m hoping to find a clever solution. Any information or insights would be greatly appreciated.
I think any thunderbolt through those 4 provided USB-C port will be next to impossible (just a guess) due to, as you say, the mainboard lacking the hardware
There may be other places to add an adapter or something
if the motherboard and CPU doesn’t have the hardware to over thunderbolt through the USB-C interface, it would be impossible to add the feature on. Thunderbolt is basicly SATA, PCI-E, etc over USB-C. So the motherboard needs the Hardware to have that communicated to the CPU. This means that on the motherboard there needs to be a thunderbolt IC/circuit that comminucates with the CPU over the PCI-E bus. USB also does that but with a different controller. There are a few different ways I see thunderbolt and USB intergrated, some use chips that do both, some use a MUX so switch between thunderbolt and USB. But the key operation here is that you need an PCI-E interface with the CPU for thunderbolt to work. Something that the FW12 sadly doesn’t offer, and without the PCI-E interface it’s nearily impossible to get it to work.
The only way around it is an HEAVILY bodged work around; You basicly need to desolder all the chips the framework uses for it’s USB signal and build your own custom daughterboard that connects to these points and then to the USB-C ports. I expect you run into spatial constraints when doing this, after you most likely run into driver issues, charging issues and a whole lot of hidden issues. While it’s not technically impossible, it is for all practicality impossible unless you enjoy reworking hardware like this. The level of modification would be similar to those people who take out the video chips of old consoles to use the connections with an FPGA to generate HDMI signals.
Just posting to link to the “No Thunderbolt?” thread from back in april where the official reason was given as being down to “much more expensive retimers”
I suppose 3$ (let’s say 5$ for some extra passives and pick and place time) is much more expensive than 0$ (rawdoging the usb directly to the cpu which is possible and may be what is done. I don’t know about you but I’d pay 5$ to get a usb4 port, hell I’d pay 20.
I doubt it would add only the raw component cost. Don’t manufacture my own laptops / boards, but I would assume there might be much more involved behind the scenes with such a change (supply chain, engineering, testing, etc.)… and might also have pushed launch to an even later date.
I get that some people would have loved it. Some would also love an OLED panel, two M2 slots, two RAM slots, bigger battery, converting the 13 in a 2-in-1 instead of launching the 12 in the first place.
They took note of your feedback and hopefully you’ll get USB4 / Thunderbolt in the future. For now it is what it is.
It’s definitely not just going to be the bom cost but most of the engineering hurdles they have already cleared with the intel 13th gen framework 13, it would be using the same extra parts they already have experience sourcing for that one, hell removing that probably meant more engineering than not. I they wanted to get thunderbolt certification that would of course make it a lot more expensive but I don’t think most people asked for that.
We also never got any clarification on the cost reasons, a simple “intel is lying and they are actually charging 30$ for those redrivers/they are unobtanium rn for some reason” or something statement would have would have made this a much more palatable tradeoff.
Comparatively those would be much bigger/more expensive changes.I want a plastic chassis for the 13 (once they figured out the cracking issue) XD.
I have usb4 on my 13 that’ll hopefully still last me quite a while, the 12 has enough other dealbreakers for me to not get one even if I needed a laptop rn.
Just wanted to highlight that it’s most likely not just a few bucks on top, but additional costs that in the end will be paid by customers and lead to more than $5. But yeah… I do agree that communication is not Frameworks strong suit. Their explanations are often times lacking. Here is hope they do give us future upgrades with heavy user feedback consideration.
On another tangent… the plastic chassis for the FW13 idea reminds me of the speed holes post on reddit (or was it even on this forum)? Where someone drilled holes right into the bottom lid of their FW12. Seems to work though. Around 15 °C less than without holes. Some commenters have raised concerns that this might impact the cooling of other components. As again other commenters named many products where this was a valid mod and some manufacturers even implemented the idea in newer iterations it might be well OK here too.
I too have a wishlist for the FW12 (prefer the form factor to the 13 and those Intel chips, to my surprise, are quite powerful compared to my older laptops). Posted my wishes somewhere else and hope user feedback will be picked up and baked into future upgrades.
Currently looking forward to the Desktop (batch 10) though
This is more like shipping a car with heated seats built in but not wiring up the controls for it while having a similar car that does have it fully done. Sure it is some effort and cost to wire up a button/code up some ui for it but not huge and they have already done it before so there should be few surprises.
This isn’t exactly sticking a square peg in a round hole either. They have done it on this exact platform before so the know-how is there and most of the stuff that used to be expensive comes “free” with the soc.
hmm… yeah, my analogy is not the best, but i think that adding thunderbolt would be a lot less trivial then connecting some wires or just swapping the retimers and would require more moving parts even if it is supported on the soc
also they have probably already performed extensive electrical testing and certifications (as well as ordering and manufacturing) the current mainboard, and they would conceivably have to redo a noticeable portion of that and set up scale-ish manufacturing for the new board
Well it would have to be designed in and it needs support in the pd-controller firmware but physically it really is just adding and properly wiring up the retimer. At 3$ for the retimer alone it is still quite expensive for a single component and it needs clean singal routing (then again raw-ish dogging usb 20gbit also needs pretty good signal routing so it’d probably not be a lot harder than what they are already doing).
The ship definitely sailed to add that now. It would have been relatively easy to add during the design phase but adding it later is likely not going to happen. Maybe on the next generation 12 board
