I love threads like this as a engineer they are fun and will lead to stupid, good, out of the box ideas. Only 10% will be useful but it’s a nice problem solving game to play. Has very little todo with why most of us ordered the h2c i hope and even less for me at least in defending it. Pondering how h2c works and how this tech can evolve is just pure fun for some of us… I am sure bambu software and hardware teams are playing the same game…
2 Likes
I, on the other hand, see something a bit worrying in how Bambu Lab plays this game. They don’t seem to care much about the products that are currently on people’s desks, and that’s a bit annoying. Imagine buying a car and every workshop suddenly refusing to service it after five months just because a new model came out — you wouldn’t be thrilled about that, right?
What I mean by them neglecting existing machines is easy to see if you look at the tungsten nozzles. Firmware updates land for the P2S and H2S before the H2D, even though the H2D is the only one that actually verifies which nozzle is installed. That’s the kind of thing that makes it feel like they’re playing a game where they keep scoring own goals.
And the really entertaining part in all this is the community reaction. Not even a year ago, the forums were on fire about how “closed” Bambu Lab is — people were talking about burning their printers, smashing them, blowing them up, you name it. Now they release the H2C, which will pretty much shut the door on third-party nozzles, and a lot of the very same people are jumping on it without blinking. One can’t help but gently ask: what happened to the whole open-source philosophy everyone was ready to die on a hill for? 
I also see a small niche that the H2C could have filled, but not in a way that makes the price feel justified — and definitely not with how Bambu Lab is treating their existing customers. There are limits to how you can play the game before people decide they’ve had enough
That said, I also enjoy pushing my machines beyond their limits, but I wouldn’t buy one just to torture it. When I get a new machine it’s because I need it to solve a real problem — the stress testing is just a side hobby. But hey, everyone has their own way of having fun with hardware 
3 Likes
Im praying that they now start banging away on all the software and firmware. Theres a rumor that more dji engineers are moving to bambu. To the point that dji is now investing in elegoo, out of spite. Maybe they are about to become the apple that theyve always wanted to be and cover hardware, software and support with an equal vigor. Hopes n dreams. I still see a p2x and new round of bedslingers coming. Bedslingers will get a small refresh. They also still have a whole world of resin to play with. Guesses.
There’s a few undesirable things going on, but I’m not really smelling it boils down to mistreating existing device users, at least, not quite.
It’s pretty obvious they spent disproportionately more resources on new product launches, but as long as existing printers do get fixes and whatever new capacities feasible that should be fine.
Is bambu delivering fixes to older devices on time? Kind of, mostly. Major problems will get fixed fast but after that, smaller ‘bearable’ ones will get assigned to thinly staffed teams, Back burner. But again, this applies to newer printers. My H2D and H2S felt identically supported, they are miles better than A1 but from a purely post launch level, it’s kind of the same story.
Is H2D for example, missing out/getting less on newer stuff compared to H2S/H2C? Again, some caveats, but I think generally, no.
On Tungsten nozzles, I get why you feel it’s a de-prioritization, but it seems more like the camera recognization needs software development time of which they have allocated not enough resources, Back burner. It’s not because H2S/P2S are newer, it’s only because they didn’t need that camera algorithm update. Possibly relevant info: The Tungsten nozzles are also still not available in China at all, at least for non pro users. Distribution/Manufacturing also seems to be stretched thin with lesser stuff backlogged.
The other edge H2S has over H2D is revamped main AC board heat sink, but that’s hardware and swappable.
I will give them some s*** on timing BBS new feature releases only with new hardware launch, it’s not actually consequential, but it can very much stirs negative user feelings. These features aren’t even locked to the newest stuff smh.
Like the just released V2.4.0, nice feature for H2D/H2C.
.
I think it’s more boring underperformance of corporate R&D managements, and not really something nefarious. Yes, older devices need more resources, but they are not forgotten either. Besides incremental hardware changes, You don’t actually get it better with less problems on newer devices. It’s not like you will see a better device in H2S/H2C over H2D in those contexts.
1 Like
I get what you’re trying to say, but from where I sit it does add up to existing users and existing hardware being treated as expendable whenever there’s a shiny new launch.
A few concrete reasons why.
1. Regressions that break working workflows for months
This isn’t just “minor bearable stuff” being put on a back burner.
- Multi-material TPU from external spools was flat-out blocked by a Studio/firmware combo with the “Not support printing 2 or more TPU filaments” error. That hit people running farms on external spools, i.e. exactly the kind of power users who already invested heavily in the ecosystem.
- After a firmware update on the X1C, TPU from external spool started throwing 07FF-8012 “unknown error” while the same g-code ran fine from SD. That’s not a marginal corner case; it’s a core feature that suddenly stopped working on an already-sold machine.
- There are several threads where H2D + external spool logic goes completely off the rails (printer obsessively trying to route to external, AMS ignored, etc.), again tied to recent firmware/Studio changes.
These are regressions introduced by updates whose main purpose is to support new models / new logic. When that breaks established, documented use-cases on older machines and stays that way for a good while, it’s hard to spin as “everyone is equally supported”.
2. Tungsten hotends: marketed now, supported… later
On the tungsten hotends, the official story is literally that the product is here but the proper firmware support will arrive in waves:
- Resellers are listing a firmware schedule where the P2S gets correct parameters + smart recognition earlier, whereas H2D and H2S are pushed back to a later window.
- In the meantime, the official instruction is basically:
Select “Hardened Steel Nozzle” and, on H2D, hit “Skip” when the printer auto-detects the tungsten hotend.
So you pay “pro” money today, and then run your fancy tungsten hotend in a downgraded compatibility mode for months while waiting for the privilege of having correct parameters and smart recognition.
If this was only about “camera recognition dev time”, you’d expect the rollout to be synchronized, or older devices to get it first (bigger installed base). Instead, the schedule explicitly prioritizes a newer platform (P2S) ahead of H2D/H2S. That’s product-tier prioritization, not some neutral queue.
3. New hardware launches as feature gates
Your own example actually proves the point: features that clearly are not hardware-locked get timed to ship with new hardware:
- Studio 2.4.0’s multi-extruder improvements are marketed around the H2C launch, even though the release notes explicitly say they “unlock the full potential of multi-extruder printers such as H2D and H2C”.
So the capability is shared, but the communication and timing are structured to make the new box look special first, and only afterwards it trickles down as “oh, and by the way, this also benefits your existing H2D”.
You call that “not actually consequential”, but psychologically and practically it is: fixes and QoL for existing owners get batched up and tied to marketing cycles instead of being pushed as soon as they’re ready.
4. The authorization / cloud-lock saga hits existing users hardest
The whole “Authorization Control” / Bambu Connect requirement is another example where resources go into tightening control over already-sold machines instead of squashing day-2 bugs:
- The X-series firmware update that introduced mandatory authorization for key operations effectively forces users into installing Bambu Connect and routing things through their cloud, which sparked a fairly massive backlash.
- The ToS / firmware wording leaves open the possibility that refusing future updates can result in printers being blocked from starting new jobs. That is a huge stick hanging over existing owners, and it arrived long after those machines were paid for.
That whole exercise brings exactly zero functional benefit to someone who just wants their already-bought H2D/X1C to stop throwing new errors with TPU or external spools. It’s resource spent on controlling the installed base, not on improving their day-to-day experience.
5. “Kind of, mostly” on fixes is not good enough at this price point
You’re saying:
“Major problems will get fixed fast but after that, smaller ‘bearable’ ones will get assigned to thinly staffed teams, Back burner.”
The issue is that the things being treated as “bearable” are, for a lot of us, core use-cases:
- External spool + TPU
- Multi-TPU workflows on farms
- Full, documented support for the premium hotend we just bought
- Not having to tiptoe around firmware in fear that an update silently breaks a key workflow again
When the same pattern keeps repeating—new hardware hype, regressions on existing machines, vague “soon” on fixes or firmware support windows measured in quarters—that’s not just “they spend more on launches”. That is what mistreating existing users looks like in practice.
So yeah, they do ship fixes eventually, and sometimes those fixes are decent. But the consistent theme is:
- Launch new hardware / features.
- Break or sideline workflows on existing machines.
- Put regressions and long-promised support (like tungsten) on a slow track while the next launch is already being teased.
Calling that “mostly fine” feels very generous.
As was my last paragraph, I didn’t say the whole thing is mostly fine, at all, things did need to change. I just didn’t arrive at quite the same interpretation conclusion about Bambu’s intent as you, but you are welcome to have a different conclusions on that, these are just opinions of ours. The TPU stuff you added is welcomed info, but that’s not stuff I’m trying to hide.
If you want a term that better fit a different mood, it’s called sucks. Same thing tho.
I didn’t call it satisfactory, again. For the price, I agree, or otherwise.
A better way to describe what I’m trying to say is, my conclusion is more of them doing a bad jobs in X aspects for Y reasons, with the overall picture that more resources is needed in Z. Rather than, boiling it down to an attitude that in a way anthropomorphizes the whole company because it’s usually more complicated than that. It’s also most likely that certain decision makings within the company definitely have such nudging intents, but if you look at the whole H2 series so far, it hasn’t been a big factor, imo at least.
The intent here is clear, as you said, but the real impact, besides pissing some users off, is minimal, as in, after all, everyone still gets it, its not pushing new users towards exclusively newer things.
But is that how it is? I don’t have one, but reading the sentence just below it,
“- H2S and P2S : No smart recognition function, Simply select the Hardened Steel Nozzle parameter to use the hotend.”
the slicer simply treats it like a hardened nozzle, doesn’t seems to be anything more to it. It really reads to me that the only thing missing is H2D camera not being able to read these QR tampos, so no auto-read these like regular hardened, but otherwise you are not missing out on features or functionality or performance over.
Again, it’s because imo the H2S/P2S merely needed some registry.exe level tweaks. I’m not saying it’s fair or right, you can call it corporate incompetence, I just don’t think it’s based on mistreatment intents here.
1 Like
bambu forum dropping reply marks as usual.
anyway, two more small things I might add,
If you print a lot of advanced TPUs, as you described, that must have sucked, so I can understand how that impact your perspective seeing everything else. But I barely do TPUs, so, please understand that it’s a bit inevitable that I won’t see them as negatively.
Unrelated, I also didn’t care about the authorization saga at all, never did.
H2D will throw a warning and you have to press Continue, because the AR codes are neither HF nor standard nozzles. So yes, it absolutely does impact the workflow.
It’s really as simple as implementing the exact same handling in H2D as in H2S and adding that AR code to the firmware. It’s not as if they’re reinventing the entire vision system from scratch.
So there is essentially no reason to push H2D support back, other than to keep P2S and H2S owners happy and, as a nice little bonus, piss on the H2D users.
And then there’s the rotary tool that was promised for H2D – will that quietly appear with the next flagship, where H2C has to line up and play the waiting game and H2D never sees it at all? Looking at how they’ve handled things so far, it’s hard to understand how anyone can accept this behaviour. Your new flashy printer is “old” in about five months, and then you get to enjoy the same treatment current customers are experiencing today.
2 Likes
Added this 38 sec time into my comparison table, together with 247’s 4 second longer 42 second timing - for comparison with U1, INDX etc.
2 Likes
Based on your wording, sounds like, if you don’t sit around for ‘skip’ the print just won’t start. If that’s the case I’d consider these nozzles borderline unusable, lol.
Pretty much, yes – the true “pro” experience:
Pay extra for the fancy nozzle, then sit there like a human OK button so the printer will agree to use the thing you just installed.
If I have to babysit a warning popup every time I hit print, we’ve gone a bit beyond “quirky UX” and landed firmly in “borderline unusable”, yeah.
Then probably that wave has already arrived for the H2D.
When I scan the Tungsten Carbide nozzles from the printer menu, it reads them without any problem and nicely reports the correct nozzles.
Also Bambu Studio doesn’t complain.
1 Like
That’s nice – mine will never detect them. It’s currently dead, waiting for spare parts after both the hotend cooling fan and the TH board failed.
I still don’t really see the benefit of tungsten nozzles beyond the “cool factor”. My HF nozzles either last a very long time or only a few hours, and if tungsten – at that price point – ends up surviving no longer than some of my HF nozzles have, it’s simply not economically defensible.
It is good to hear that your printer actually detects them though. I run beta firmware and mine didn’t detect them at all before it died – and yes, I tested it even though I don’t actually own a tungsten nozzle.
I bought them in the sale for 2.5 times the price of a hardened steel one.
I am getting more and more into CF filaments, so if it hold out 2.5 times longer it at least saved me 2 additional nozzle swaps 
PCD seemed like the real prize but bambu will never make one in-house, hopefully others will soon.
The ROI calculus for Tungsten Carbide seems a bit meh, as these two materials are not too dramatically different in areas that might significantly improve print quality - friction/thermal conductivity/heat expansion, which PCD Diamond does, and apparently promises better layer adhesion, surface quality over hardened steel.
That said, it’s still better to use different nozzle for each filament types so it can get pretty absurd for hobbyists. But if I’m selling quality focused prints even PCD’s pricing will be a no brainer.
True, but be aware that even tungsten can and will clog with CF filaments, so let’s hope they last long enough for your investment to pay off.
I rarely print with non-abrasive filaments, and at the moment I can’t see a return on investment that makes tungsten worthwhile. I’d still like to try them though, that might happen someday.
Idk about you but I’ve never actually killed a non-brass nozzle thru wear, only ever thru incidents. So the “amortization” mindset of it lasting longer than any other is a little weird imo. It’s gonna crash into the plate and get bent, not wear out.
1 Like
If it’s like the H2D, and I suspect that it is, looks like there are sensors right at the buffers and they’re being read. Either way, if it is switching between the right and left nozzles, and it cuts and swaps induction hotends before printing with the left, it can pre-stage a new color on the right nozzle or at least at the buffer. while the left is doing it’s thing.
1 Like
How did you get that feature to show up?! Im refering to the filament assignment for sparse and solid infill. Im on V2.4.0 and have an H2C and would love to change the infill material.