18:11 <d1b2> <Ben> I'm assuming the Jan 31st ship date that Crowdsupply currently gives me for my early bird is probably nonsense?

18:17 <d1b2> <Perigoso> Maybe that was the prevision before the shortage hit? But yeah that's bot going to happen afaik

22:01 <d1b2> <ewen> Last we heard the Glasgow BOM was still entirely missing a couple of key components. I'd fully expect there'd be a "good news" updatei immediately if somehow those parts had arrived in time to make even the early bird boards, so I think this is a "no news is bad news" situation. (Personally I think even the "August 2022" listed for "order now" is rather optimistic, given how specialised the missing parts are, and how few seem to be being

22:01 <d1b2> made/available. But last update on the estimates was based on ~ early 2021 estimates of availability from manufacturers which seem to have proven... overly optimistic.)

22:13 <d1b2> <trilateral> at any point will the board be redesigned?

22:14 <d1b2> <Perigoso> RevC1 i doubt will ever be "redesigned" other that some minor improvements over time

22:14 <d1b2> <Perigoso> But there are other planned "models"

22:15 <d1b2> <Perigoso> But with the shortages i think it will be a long time before we see any of that

22:15 <ebb> I'd say revC2/revC3 are some hefty changes on C1 -- although not as far as a rearchitecting

22:15 <d1b2> <trilateral> yeah but I mean to eliminate the hard to source components

22:16 <d1b2> <trilateral> this thing isn't that complicated

22:17 <d1b2> <Perigoso> Well, easier said than done

22:18 <d1b2> <Perigoso> The current parts were chosen for very specific reasons and many have no alternative

22:18 <d1b2> <trilateral> Like a voltage regulator?

22:18 <d1b2> <Perigoso> Well, actually yes

22:33 <d1b2> <trilateral> I dunno at some point it would be nice to see a plan beyond "wait it out"

22:37 <d1b2> <trilateral> Cut some functionality maybe

22:37 <d1b2> <WillC> nice thing about this project is that its open source. So you can find a new component that wont add to board dev time do it!

22:37 <d1b2> <WillC> I am so excited I want to see that happen faster also

22:38 <d1b2> <WillC> I already told them to go to china and take control of a factory and have said factory only make chips for this project. I don't understand why they haven't yet

22:46 <d1b2> <trilateral> Well yes obviously any changes will take dev time, which is why I think they're not doing that. But as a backer my patience isn't infinite.

22:49 <d1b2> <WillC> then why are you not designing it?

22:52 <d1b2> <trilateral> Dude quit simping

22:52 <d1b2> <trilateral> Got any more open source cliches to spout off?

22:52 <d1b2> <trilateral> At the end of the day, they do have an obligation to deliver

22:52 <d1b2> <WillC> As much as I love this project I don't think its as important as producing the next car to come out... Have you tried telling the auto industry just to pick different components? https://www.thedrive.com/news/43774/yes-the-chip-shortage-will-still-screw-the-auto-industry-in-2022

22:53 <d1b2> <trilateral> Dude get real

22:53 <d1b2> <trilateral> This thing is basically a fancy fx2 logic analyzer

22:53 <d1b2> <david.lenfesty> > Backers accept the risk that Products may be late, or in some cases, not deliverable at all. nope lol

22:53 <d1b2> <trilateral> Yeah that's why everyone hates crowdfunding

22:53 <d1b2> <trilateral> And thanks for continuing the cliche train

22:54 <d1b2> <trilateral> Perhaps I meant a moral obligation

22:55 <d1b2> <zyp> if you hate crowdfunding, you're free to not participate

22:58 <d1b2> <WillC> The REALEST thing is the fact there is a GLOBAL chip shortage. I understand that it might be hard to see that this effects all products that haz them micro chips

22:59 <d1b2> <WillC> THEY NEED TO GO BACK TO TUBES

22:59 <d1b2> <WillC> we wouldnt be in this mess if it was tubes all the way down

22:59 <d1b2> <trilateral> dude do you even do electronics design?

22:59 <d1b2> <WillC> nope

22:59 <d1b2> <WillC> NEVER

22:59 <d1b2> <trilateral> yes, I've been dealing with this shortage for years as well

22:59 <d1b2> <trilateral> you know what I don't do?

22:59 <d1b2> <WillC> i dont even know what this is

23:00 <d1b2> <trilateral> tell my boss there's no possibility we can ship a product for 18 months

23:00 <d1b2> <WillC> I am in this discord for the cat photos

23:00 <d1b2> <trilateral> beCuz CHiP sHorTAGe

23:00 <d1b2> <WillC> CATZ

23:02 <d1b2> <dragonmux> trilateral and WilC, please be respectful to each other and the other participants here.

23:03 <d1b2> <dragonmux> as for the problems around Glasgow's production - there's one part that 1b² has been unable to source a complete set of and which is single-supplier (TI) - the team are doing what they can to scrap together parts at not-scalped-to-hell rates

23:04 <d1b2> <dragonmux> the problem with the part is that it was chosen because it has the unusual property that it doesn't back-power circuitry the unregulated side of the regulator, which is important for the robustness of the front-end

23:04 <d1b2> <dragonmux> so it's not a simple part to replace

23:04 <d1b2> <dragonmux> as for it being a "fancy FX2 analyser" - not even close - the FX2 is only used as an easy way to transport data back from the FPGA to the host. The actual functionality is all on-FPGA

23:05 <d1b2> <zyp> I thought the unusual feature was related to overcurrent handling?

23:05 <d1b2> <dragonmux> not backpowering on OCP

23:05 <d1b2> <dragonmux> yes

23:05 <d1b2> <dragonmux> compromising on that part compromises on product safety

23:06 <d1b2> <trilateral> ...or you implement it without a lego piecee

23:06 <d1b2> <dragonmux> it's a vreg.. you can't leave out a vreg

23:06 <d1b2> <dragonmux> the entire front-end won't work if it's missing

23:07 <d1b2> <trilateral> you know what "IC" stands for, right?

23:07 <d1b2> <trilateral> just de-I it

23:07 <d1b2> <dragonmux> it really is far more complex than that.. IC stands for Integrated Circuit which just refers to anything silicon more complex than a diode or a transistor

23:07 <d1b2> <trilateral> yeah, and pretty much anything that's been integrated can be factored into a more discrete circuit

23:07 <d1b2> <dragonmux> not VRegs

23:08 <d1b2> <trilateral> baloney

23:08 <d1b2> <dragonmux> the trickery done in silicon for voltage regulation doesn't work in discrete circuitry

23:08 <d1b2> <dragonmux> and yes, we design electronics and are involved in the open ASIC community.. we might know a thing or two about that

23:09 <d1b2> <trilateral> not enough to ship a product, apparently

23:09 <d1b2> <zyp> uh, a linear regulator is just a pass transistor with feedback regulation, building a discrete linear regulator isn't too hard

23:09 <d1b2> <zyp> although absolutely not worth it

23:10 <d1b2> <dragonmux> zyp: with the feature set in question it would be hard as no multi-gate FETs, and it's not an open source part so we don't know how exactly they're doing their OCP backpower magic

23:10 <d1b2> <dragonmux> if it were a simple linear, then sure.. doable but horribly inefficient

23:11 <d1b2> <zyp> doesn't matter how they're doing it, what matters is the end result

23:11 <d1b2> <bob_twinkles> don't a lot of IC VRegs rely on extremely closely matched transistor pairs, which is impossible to do with discrete components

23:11 <d1b2> <zyp> but I'm just playing the devil's advocate here, not actually arguing for doing it

23:12 <d1b2> <dragonmux> trilateral: please don't take this wrong but we're not on the team handling Glasgow.. please don't mouth off at us like that or assume things are simpler than their reality. We understand your frustration - really we do - we are only offering honest explanations based on what we've been informed is the situation

23:13 <d1b2> <dragonmux> bob: correct.. and multi-gate FETs which also don't exist as discretes..

23:13 <d1b2> <trilateral> that's fine but I think you're water carrying then

23:14 <d1b2> <trilateral> it's absolutely possible to implement reverse current protection without this particular IC

23:14 <d1b2> <trilateral> I understand that it costs time and effort for them to update design

23:14 <d1b2> <trilateral> but at some point it's the right thing to do

23:15 <d1b2> <dragonmux> also, consider this: you're asking the team to redesign a complex 4-layer PCB to be larger and with hundreds of extra components to discretely implement something done with a 3x2mm IC at the moment - this is not economical and it would blow the form-factor right out the window along with the case design

23:15 <d1b2> <dragonmux> just because you feel.. slighted?

23:16 <d1b2> <trilateral> I mean, if they'd actually shipped anything, the chinese would have already cloned it

23:16 <d1b2> <trilateral> maybe they should not have designed with zero clearance

23:17 <d1b2> <zyp> why do «the chinese» need to wait for the product to ship when the hardware design is already available?

23:17 <d1b2> <dragonmux> we understand you haven't understood how board design works or just how many discretes are involved in what you're asking, so we'll just leave it at that point

23:18 <d1b2> <trilateral> Because they don't waste time without a proven market

23:20 <d1b2> <Attie> @trilateral a lot of work goes in to selecting a part like that, verifying its behaviour, and including it in a design - a lot of thought and care has been put into designing a robust tool. it's a significant oversimplification to suggest the team "just replace it" or "implement it as discretes".

23:21 <d1b2> <trilateral> Yeah, the whole point of backing is I'm not interested enough for that

23:22 <d1b2> <trilateral> Otherwise I'd just make my own

23:22 <d1b2> <dragonmux> then please respect what a board design professional is telling you

23:22 <d1b2> <trilateral> Dude you're not the only professional get off your high horse

23:23 <d1b2> <trilateral> But I've definitely been schooled on oshw/crowdsupply, don't worry.

23:26 <d1b2> <WillC> I would love to see it! Ill even buy one from you how long do you think it will take?

23:31 <d1b2> <Perigoso> Funny i was thinking you should be getting of your high horse

23:31 <d1b2> <Perigoso> And don't just throw "dude" around...

23:31 <d1b2> <trilateral> Reading comprehension isn't your strong suit huh?

23:31 <d1b2> <Attie> can we stop lashing out at each other please

23:32 <d1b2> <WillC> I really wanted to buy one

23:33 <d1b2> <WillC> thats just mean I struggle with English

23:52 <d1b2> <bob_twinkles> huh, if the only part that is still missing is the TLV733 (1.2v DQN footprint), can the board be modified to use the SOT23 version?

23:52 <d1b2> <bob_twinkles> TI claims they have like 200k of those

23:52 <d1b2> <bob_twinkles> TLV73312PDBVR

23:52 <d1b2> <bob_twinkles> it looks like there's enough space but I'm not a power wizard

23:53 <d1b2> <dragonmux> unsure, though we know that one of the team's goals was to not modify the board if possible just for the campaign - but that's over to @esden to answer as changes like that are on him

23:55 <d1b2> <bob_twinkles> yeah, I know the board design is intended to be frozen. But as long as they haven't been built maybe it's possible

23:55 <d1b2> <bob_twinkles> March isn't really that far away though, assuming they actually deliver

23:56 <d1b2> <bob_twinkles> My brain is tickling me to say that I've seen discussions about using thew SOT23 version and it was rejected for some reason but I can't remember the details