norly has quit [Remote host closed the connection]
pabs3 has quit [Ping timeout: 265 seconds]
ccnnjj has joined #neo900
pabs3 has joined #neo900
Humpelstilzchen has quit [Ping timeout: 272 seconds]
Humpelstilzchen has joined #neo900
trx has quit [Ping timeout: 250 seconds]
trx has joined #neo900
pabs3 has quit [Ping timeout: 255 seconds]
arossdotme has quit [Ping timeout: 245 seconds]
pabs3 has joined #neo900
arossdotme has joined #neo900
infobot has quit [Read error: Connection reset by peer]
infobot has joined #neo900
nicksydney has joined #neo900
pabspabspabs has joined #neo900
pabs3 has quit [Ping timeout: 264 seconds]
<ZetaR>
I have been studying the IR paper. Have you considered thermal runaway in your protection mechanism? LED current increases exponentially with temperature (with constant voltage), so a lot of high-power LED designs use constant-current sources (where the increase is logarithmic instead).
nicksydney has quit [Ping timeout: 244 seconds]
nicksydney has joined #neo900
pabspabspabs is now known as pabs3
nicksydney has quit [Ping timeout: 244 seconds]
nicksydney has joined #neo900
nicksydney has quit [Read error: Connection reset by peer]
nicksydney has joined #neo900
<DocScrutinizer05>
the design *is* constant-current
<DocScrutinizer05>
you can tell from the circuit diagram as well as the simulations
nicksydney_ has joined #neo900
nicksydney_ has quit [Read error: Connection reset by peer]
<DocScrutinizer05>
operating an LED from constant-voltage is as silly as it gets
nicksydney has quit [Ping timeout: 272 seconds]
<DocScrutinizer05>
even the dirt cheap keyring lamps work only because of the CR2032 or whatever batt is used have a noticeable impedance ESR working to limit the current through LED
cybiko123 has quit [Quit: Leaving.]
cybiko123 has joined #neo900
cybiko123 has quit [Changing host]
cybiko123 has joined #neo900
<DocScrutinizer05>
>> Example current-limiting circuit: continuous current through the IR LED D1 is limited by R1 and R3. Short pulses in the 100 μs range can draw extra current from C1. The peak current is then limited by the constant current transistor circuit.<<
<DocScrutinizer05>
rather you should ask where we found a GPIO capable of sourcing 3V3. answer: we didn't
<ZetaR>
DocScrutinizer05: I meant that the I-V curve is exponential w.r.t. constant current. It is different with a resistor, but still super-linear. On their own R1 and R3 only limit it to 178mA, so whatever limiting is left depends on the temperature of the LED (ignoring the drop across T1). Also, the voltage I see in the circuit is 4.2, not 3.3?
<ZetaR>
oops, I meant exponential w.r.t. constant voltage.
<ZetaR>
I can do some calculations to see when it would become a problem, but that will have to wait until tomorrow.
<DocScrutinizer05>
thanks, but as you can see in IR whitepaper, we already did extensive calczulations and even simulations
<DocScrutinizer05>
and there is no problem
<DocScrutinizer05>
again: >>The peak current is then limited by the constant current transistor circuit<<
<ZetaR>
Oops, I missed that sentance. Sorry.
<ZetaR>
(thought it said "resistor" for some reason")
<DocScrutinizer05>
in the red Pr.It LED current curve you can see how current gets limited pretty hard to sth around ~0.4A before capacitor depletes and the resistor based average current limit kicks in and reduces I gradually to ~100mA
<DocScrutinizer05>
the parameter sweep for transistor beta = 10, 100, 1000 shows it's quite stable a circuit for anything beta in a realistic range of 50 to 500+
<ZetaR>
Right. It is a complicated circuit and I interpreted the transistor as being just a switch when I first read it.
<ZetaR>
Probably shouldn't read these this late in the day.
<DocScrutinizer05>
simplicity is an art
<DocScrutinizer05>
;-)
<ZetaR>
And complexity is engineering? =)
<DocScrutinizer05>
this circuit accomplishes quite a few functions with as little as 3 transistors and one Zener
<ZetaR>
Yeah... simple circuits often seem more complicated than "complex" ones.
<DocScrutinizer05>
:nod:
<ZetaR>
Though it makes studying things like op-amps *very* interesting. Such a simple set of rules to produce very complex behavior.
<DocScrutinizer05>
hehe, see the IR RX
<DocScrutinizer05>
Werner didn't dare publishing my Qucs simulations for that one ;-)
<DocScrutinizer05>
I admit the circuit is a tad more critical by dimensioning of components
<ZetaR>
Yeah, it was a bit of "this is a this thing, oh wait no it is something else, hmm".
<DocScrutinizer05>
yeah, it's an amp and a filter and a schmitt trigger / comparator and a monoflop
<ZetaR>
I caught the amp and filter when I first saw it. I started the head scratching at schmitt trigger. It is a very interesting circuit.
<DocScrutinizer05>
it is basically a bistable circuit toggling at edges of input signal, while the offset (to compensate for ambient light) is made form a lowpass filter
<ZetaR>
The complexities of feedback in op-amps reminds me of simple recurrance relations in mathematics producing chaos. Here is an example you might find interesting: https://en.wikipedia.org/wiki/Logistic_map
<DocScrutinizer05>
from*
<ZetaR>
Interesting. Makes sense.
<ZetaR>
Have you done a prototype of it?
<DocScrutinizer05>
partially
<DocScrutinizer05>
complete prototype only makes sense with the original components
<ZetaR>
So the Neo900 will be a bit different, then?
<DocScrutinizer05>
no
<ZetaR>
Oh, not sure what you meant by "original" then.
<DocScrutinizer05>
I had no access to e.g. the type of opamp and photodiode
<ZetaR>
Ah, I see.
<DocScrutinizer05>
the dimesnions of the resistors are a tad touchy in this one
<DocScrutinizer05>
and change quite a bit with the type of photodiode
<DocScrutinizer05>
I guess the type of opamp is rather irrelevant, but anyway
<ZetaR>
Is that going to be okay with normal tolerances?
<DocScrutinizer05>
yes
<ZetaR>
Okay, that's good.
<DocScrutinizer05>
tolerances are still way beyond 10%
<DocScrutinizer05>
after all the circuit is adaptive
<DocScrutinizer05>
it adjusts itself to vasty changing light intensities
<ZetaR>
I wonder if this circuit would see widespread use then.
<DocScrutinizer05>
the dimensions are so it should work down to 1Hz and in a range from 0 to 30cm witha 50mW LED
<DocScrutinizer05>
3cm to 50cm with a 200mW LED
<DocScrutinizer05>
and even a 5mW LED shall still work in 0cm
<DocScrutinizer05>
which is pretty much as good as it gets with the limitations in space etc we need to consider
<ZetaR>
Zero distance is a bit short don't you think? =)
<DocScrutinizer05>
no
<DocScrutinizer05>
it's the recommended operation mode
<DocScrutinizer05>
touch the Neo900 with your IR-remote or RS232<->IR adapter
<ZetaR>
Ah, okay, so just something like <1cm.
<DocScrutinizer05>
yes
<ZetaR>
Can't you get better distance by just tightening the beam width? I guess angle sensitivity might be more of an issue then, though.
rjeffries has joined #neo900
<DocScrutinizer05>
for the RX we hardly can do that
<ZetaR>
Oh, oops, confusing TX/RX for some reason.
<DocScrutinizer05>
for TX we don't want too tight beam, this sucks for TV-remote and similar stuff - targeting like in an egoshooter needed
<ZetaR>
My TV remotes will even work facing away from the TV.
<ZetaR>
Though, my room isn't that big, and the walls are white.
<ZetaR>
(IDK if the color changes the IR reflectance)
<DocScrutinizer05>
that's what we expect to work with Neo900 as well. With N900 forget about it
<DocScrutinizer05>
color changes IR albedo, though in a way you'd not expect all times
<ZetaR>
I noticed the paper said that. I didn't realize that remote control LEDs were that bright.
<ZetaR>
Yeah, I guess the changes would be more intuitive if we could see IR well. =)
<ZetaR>
You can actually see in these frequency ranges, but only if it is very dark and your eyes are adjusted.
<ZetaR>
Even then, the light looks pretty dim.
<DocScrutinizer05>
yes
<DocScrutinizer05>
but I suspect not everybody can (genetic differences?). Some of my friends can't see the IR cameras (their IR LED illumination) in cinema, while I always see when they're active
<DocScrutinizer05>
it only works for the shortwave IR though
<ZetaR>
Interesting.
<ZetaR>
It would be a difference in the rods. Genes are on the X chromosome.
<ZetaR>
So there might be a difference between sexes as well.
<DocScrutinizer05>
IOW the 870nm. No chance to see 950nm
<DocScrutinizer05>
OOOHH that's interesting
<ZetaR>
Fun fact: you can actually see into the X-ray spectrum too. People from a more innocent time noticed that if you stared into a high-intensity X-ray beam with your eyes closed you could see a faint blue glow. Then dying of radiation poisoning was discovered by Marie Curie.
<DocScrutinizer05>
haha, that's radioluninescence
<DocScrutinizer05>
what's glowing blue is the water in your eyeballs
<ZetaR>
Yep. =)
<DocScrutinizer05>
astronauts report dim flashes they see with closed eyes, whenever a high energy quantum passes their eyeballs
<MonkeyofDoom>
other fun fact: you can see into UV (400-300nm) if you remove the lenses from your eyes
<ZetaR>
(genes) There is actually a slightly mutated version of one of the genes for the red cone. Since it is on the X-chromosome, some women can get both kinds and have two slightly different forms of red reception.
<ZetaR>
Its quite amazing what surgeries they did before modern medicine.
<DocScrutinizer05>
yeah, old Egyptians already drilled holes in skull to do brain surgery. And they poor victims actually sometimes survived that
* DocScrutinizer05
feels nausea
<ZetaR>
Here is quote I recall from the Hypocratic oath: "And I shall not cut those laboring under the stone, but leave this to those skilled in that craft" (paraphrased from memory) Referring to kidney stones
<DocScrutinizer05>
weird
<ZetaR>
Yep, so the ancient Greeks had professional surgeons for kidney stones.
<DocScrutinizer05>
who knows, maybe already used acoustic shockwave to smash the stone? :-)
<ZetaR>
Wishful thinking. =) Though you would try almost anything with kidney stones, even some random guy cutting them out. The pain is horrible from what I hear.
<ZetaR>
They actually were able to squeeze the endocellular fluid out like from a tube of toothpaste, and measure it and the membrane seperately.
Pali has joined #neo900
sparetire_ has quit [Quit: sparetire_]
freemangordon has quit [Ping timeout: 245 seconds]
freemangordon has joined #neo900
freemangordon has quit [Ping timeout: 245 seconds]
freemangordon has joined #neo900
qwazix has quit [Ping timeout: 258 seconds]
qwazix has joined #neo900
modem has joined #neo900
mrcaaattt has joined #neo900
<wpwrak>
ZetaR: pity that the FSF SIM doesn't seem to have SWP. i'm a bit concerned about our ability to actually test that stuff. it seems to be pretty hard to find SIMs that a) aren't tied to some contract, and b) explicitly state to have SWP support.
<wpwrak>
so i guess we'll just have to try some critters and hope one of them responds ...
<JamesJRH>
Hi! I've just found out about the Neo900 and think that it's great!!!
<JamesJRH>
I want to get one, however, first thing that strikes me as an issue, being a developer of an Android application that makes use of multitouch, is that only 2 pointers are planned.
<JamesJRH>
If you can surpass 2, preferably have at least 10 (for 10 digits of a human), then I'll be wanting to port my musical instrument over to the Neo900.
<JamesJRH>
I don't see a need for more than 10, and 10 seems pretty standard nowadays, but less than 5 would be a pretty big problem, for me and others. I don't mind what the underlying technology is, be it resistive or capacitive, so long as it can support enough independent pointers.
<JamesJRH>
Other than that, the specs are looking pretty damn good!
<JamesJRH>
Battery hotswap sounds awesome!!!
<JamesJRH>
Is the CPU at least dual core?
<Wizzup>
no, I do not think it is
<JamesJRH>
Hmm.
<Wizzup>
also, I don't know how hard it would be to do 10 finger mt
<Wizzup>
(or if it is possible)
<Wizzup>
where did you read the part about two pointers?
<JamesJRH>
Well, single-core is a little annoying, I'd prefer a heterogenous dual-core using for example big.LITTLE (https://en.wikipedia.org/wiki/ARM_big.LITTLE ), but this isn't as bad as only having 2 pointers.
<bencoh>
it wont be more than dualtouch
<Wizzup>
JamesJRH: that won't happen (@big little)
<Wizzup>
The CPU is chosen because it is similar to the n900 one
<JamesJRH>
2 pointers is a deal-breaker for it being my main phone; it'd have to be secondary. That doesn't rule-out me getting 1 though, but hmm, it means I'm not freed from another device.
<bencoh>
looks fun ... dunno how I'd like it as a piano player though :)
<JamesJRH>
If Neo900 could get at least 5, preferably 10 pointers, I'll do my best to port it or rewrite it for the Neo900.
<bencoh>
it's a resistive screen, so it should be pretty accurate, btw
<JamesJRH>
Cool, yes, I saw that on the FAQ.
<bencoh>
but unfortunately it wont be appropriate for what you have in mind
<JamesJRH>
It does work with just 2, or 1, pointer, but it's not the full deal.
<bencoh>
yeah
<JamesJRH>
With my ‘single-touch chording’ feature, though, one can play major and minor triads with only 2 pointers. Playing chords with 4 notes, will be limited to only 1 type of chord, though I forget the name of it.
<Wizzup>
JamesJRH: you could use the hw keyboard?
<Wizzup>
:)
<JamesJRH>
True.
<bencoh>
uh, right :-)
<JamesJRH>
I have mapped my laptop keyboard to the Wicki-Hayden layout before I took-on Hexiano.
<JamesJRH>
I used VMPK, with JACK, and a synthesiser, iirc..
<JamesJRH>
Okay, to be fair, most basic stuff will be fine. The HTC Desire was playable, but had very poor accuracy, especially for 2 fingers. By comparison, the improved accuracy should make it far better than the Desire, but not as ‘full’ as most phones or tablets.
<JamesJRH>
It will be very good until you hit the limit – until you want to play a chord that requires more than 2 pointers.
<JamesJRH>
I often play a tune that requires 3 pointers at least once, but I guess that I could live with 2 for a while, provided that the successor to the Neo900 has more pointers.
<JamesJRH>
I /want/ to use the Neo900.
<bencoh>
or you could add some hwkbd mapping as suggested ... I kinda like the idea of being able to use those two
<bencoh>
as a hwkbd doesnt provide the same feedback
<JamesJRH>
If I port it, I'll definitely support mapping the hardware keyboard for the Jammer layout when in landscape mode and for the Sonome layout when in portrait mode (due to the way that the hexagons are arranged to match the hardware keyboard's roughly hexagonal arrangement).
mvaenskae has joined #neo900
<JamesJRH>
Janko would also map in landscape, but I don't plan to port/rewrite that because it's not really an isomorphic layout and so will very much get in the way of other features that I'm planning that /assume/ the isomorphism.
<JamesJRH>
So anyway, moving on from hardware, what software can be available? Will there be a way to emulate Android applications from F-Droid or otherwise?
<JamesJRH>
Over 90% of the smartphone applications that I currently rely on are from F-Droid.
Oxyd76 has quit [Ping timeout: 265 seconds]
<Wizzup>
JamesJRH: you could probably run android on it.
<Wizzup>
or some port of android.
<Wizzup>
The neo900 project does not focus on the sw side, there is the FPTF (Fremantle Porting Task Force)
<JamesJRH>
I don't want to /run/ Android, if I have the choice, actually.
<JamesJRH>
Okay.
<Wizzup>
likely they will ship with a debian image initially
<Wizzup>
or something like it.
<JamesJRH>
Ideally, I'd like to run NixOS on it.
<Wizzup>
That is, hopefully many projects will support the neo900, but there is no 'pledged support' by the neo900 team itself (please correct me if I am wrong)
<Wizzup>
JamesJRH: isn't NixOS mostly a package manager?
<Wizzup>
I mean, Nix is
<JamesJRH>
That's Nix.
<JamesJRH>
Yeah.
<Wizzup>
So why can't you run it in a chroot?
<JamesJRH>
NixOS is a distro.
<Wizzup>
Or, what does nixOS as a distro offer you
<Wizzup>
that you must have on a phone ;)
<JamesJRH>
Nix.
<Wizzup>
well, it's more than a phone... sure
<JamesJRH>
:-)
<Wizzup>
so, a chroot works fine for that purpose :)
<JamesJRH>
I must have Nix.
<Wizzup>
feel free to bring the distro to the neo900
<Wizzup>
;)
<JamesJRH>
Okay, but which way.
<JamesJRH>
Sure.
<bencoh>
I dont see any reason why you wouldnt be able to "port" it ;)
<bencoh>
actually if there is an arm port it should work almost out of the box
<JamesJRH>
I'd prefer Android chroot's onto NixOS, rather than the other way round.
<bencoh>
provided appropriate kernel
<Wizzup>
JamesJRH: definitely, but I didn't suggest using android as host
<JamesJRH>
I'm working on that.
<Wizzup>
as for android emulation, I know Jolla has it
<Wizzup>
I don't know about other projets - perhaps if you run chrome..
<Wizzup>
Doesn't chrome ship with android emulation these days?
<bencoh>
jolla runs on top of an android kernel
<Wizzup>
I guess it's not avail. for arm like that
<bencoh>
(heavily modified)
<JamesJRH>
Cool, Jolla, I'll look it up.
<Wizzup>
bencoh: all actually android stuff on Jolla is done by their android emu/vm
<bencoh>
Wizzup: you still need the underlying graphic drivers in kernel ... and that's one of the reasons their kernel is a modified android one
<Wizzup>
I think it was never released
<JamesJRH>
Chrome Android emulation? Everyone suspected that Google would want to merge their 2 software platforms, but I didn't know it has now happened. :-)
<Wizzup>
bencoh: I thought jolla had libhybris for that
<Wizzup>
And I don't think you need android graphics drivers to emulate single android apps.
<Wizzup>
You just need compatible graphics drivers.
<Wizzup>
s/drivers/interface/
<Wizzup>
[see the video]
jonwil has joined #neo900
<JamesJRH>
Very interesting.
<JamesJRH>
So just because I have an operating system that can seamlessly run Android applications doesn't mean to say that those have to be the main runtime.
<JamesJRH>
So can I write applications in something /other/ than Java/Dalvic?
<JamesJRH>
(And of course still run those that /are/ written for Android.)
<bencoh>
for ? neo900 ?
<JamesJRH>
This is why I keep saying that I plan to “port/rewrite” my application – I want to rewrite it in Haskell, if possible.
<JamesJRH>
Yeah.
<bencoh>
neo900's primary goal is to run maemo (well, some kind of maemo port)
<JamesJRH>
Okay, so can I write in Haskell for Maemo?
<bencoh>
but neo900 isnt responsible for this port (or for any other port) and they'll be community-driven, so nothing stops you from porting your own distro
<bencoh>
as for haskell&maemo, it has no package for it yet, but building one shouldnt be too hard, unless it has too recent deps
<JamesJRH>
Well I would like to run NixOS, but to port Maemo's application runtime to NixOS such that I can run Maemo application on NixOS. Or the other way of looking at it is that I want to replace Maemo's package manager with Nix.
<bencoh>
well, it's a dpkg-based distro, so ...
<bencoh>
feel free to try
<JamesJRH>
And then also use Haskell for Maemo, and run Dalvik too.
<Wizzup>
p.s.
<Wizzup>
you can already try all this on a n900
<Wizzup>
pretty much
<JamesJRH>
Except for Haskell on Maemo (which I don't know about), it seems that these things have all been done, just not simultaneously.
<Wizzup>
uh... you just get a compiler on arm and compile for arm?
<Wizzup>
linking against specific libs should also not be too hard
<JamesJRH>
I.e. Maemo–Dalvik integration and using Nix instead of dpkg.
<Wizzup>
it seems like a lot of work that would achieve ... not so much (at least wrt nix)
<Wizzup>
but, feel free to try it of course
<JamesJRH>
Okay, but what about APIs?
<Wizzup>
Depends on the API.
<Wizzup>
That is, depends on what you want to do.
<bencoh>
and the amount of time you're willing to spend on it
norly has joined #neo900
<wpwrak>
Wizzup: SWP (Single Wire Protocol) is a protocol used to talk to the Secure Element in some SIM cards. it's something that can be used with NFC
<JamesJRH>
I'm willing to spend a lot, because I currently find that I'm very limited by my Android phone, and I think that others will benefit too, if I break the barriers.
<bencoh>
building a fully-functionnal (including phone features) nix-based distro might represent some work ... not to mention keeping compat with current maemo apps
<bencoh>
JamesJRH: that's the spirit <3 :)
<JamesJRH>
:-)
<JamesJRH>
Okay. Does Maemo have a name for its runtime environment?
<JamesJRH>
Like does it have an equivalent to Dalvik?
<Wizzup>
wpwrak: right, so you were talking about the NFC version of the gnupg cards?
<bencoh>
if you want to start playing with the software part, grab a n900 (it's not that expensive these days) and happy hacking
<Wizzup>
I didn't realise that
<bencoh>
and stay around here for the neo900-specific bits
<bencoh>
JamesJRH: also, #maemo
<JamesJRH>
Okay, cool...
<Wizzup>
JamesJRH: fwiw you can start with a chroot on a sd card and running maemo, and then have certain apps from the chroot run in maemo
<wpwrak>
Wizzup: no, just the SIM card there. the description says it communicates over the "traditional" interface (like a modem would), not SWP.
<wpwrak>
Wizzup: i was just hoping that it may be a well-documented card with SWP capabilities
<Wizzup>
wpwrak: hm, I think there if a NFC version of the card
<Wizzup>
but perhaps I am not understanding you well
<Wizzup>
ah, I think I get it now..
<bencoh>
pabs3: dunno about neo900; n900 has issues with some combinations
che11 has joined #neo900
che11 has quit [Ping timeout: 240 seconds]
<arossdotme>
with the neo900 worldwide mob chip, that has no 4g right? wheres the Europe and America,etc mob chips do have 4g
<arossdotme>
thats the difference right?
jonwil has quit [Quit: ChatZilla 0.9.91.1 [SeaMonkey 2.33.1/20150321194901]]
ZetaR has quit [Ping timeout: 246 seconds]
ZetaR has joined #neo900
mvaenskae has quit [Ping timeout: 276 seconds]
mvaenskae has joined #neo900
mvaenskae has quit [Ping timeout: 265 seconds]
mvaenskae has joined #neo900
paulk-collins has joined #neo900
<ZetaR>
wpwrak: Couldn't the OMAP/MCU act as a go-between for secure crypto/auth cards and the NFC? Then SWP support in a smartcard/SIMcard could be optional.
<ccnnjj>
Where is the free-mantle/FPTF work being co-ordinated? (I may have missed it, but IÂ didn't see a lot of information on talk.maemo after some initial info dumps).
<ZetaR>
ccnnjj: I think most of the focus is on hardware right now, but I would wait for an answer from someone more knowledgeable.
lobito1 is now known as lobito
<wpwrak>
ZetaR: since the MCU sits between NFC and SIM, this is exactly what it does :) whether it just passes data from one side to the other without looking at it, or whether it does something more interesting, that choice is left to you (as a user / developer)
<ZetaR>
wpwrak: Exactly. So testing doesn't just need to be limited to SWP either.
xes has quit [Quit: leaving]
xes has joined #neo900
<wpwrak>
well, the regular SIM connection is a separate issue. but we'd also like to know whether SWP works
<wpwrak>
(also in production testing, not only in the lab)
dal2 has joined #neo900
dal2 has quit [Ping timeout: 255 seconds]
dal has joined #neo900
panais has joined #neo900
dal has quit [Ping timeout: 255 seconds]
dal has joined #neo900
dal has quit [Ping timeout: 255 seconds]
dal has joined #neo900
<ZetaR>
wpwrak: I had a look around, the stuff I found was systems with complicated corporate setups like Mastercard's mobile PayPass, or whole-system NFC testing rigs (that were not clear about SWP specifically), or no-name SWP test SIMs from China.
panais has quit [Ping timeout: 258 seconds]
modem has quit [Ping timeout: 256 seconds]
<JamesJRH>
Hi again.
<wpwrak>
ZetaR: no-name from china sounds appealing in this case :)
<wpwrak>
got any links ?
<JamesJRH>
I've been talking on #maemo a bit, then #nixos a bit. The NixOS on ARM situation seems to have progressed massively, so my plan of porting NixOS to the Neo900, packaging Maemo's Hildon Gtk runtime to NixOS, writing applications in Haskell and Hildon (including a Hexiano rewrite), and packaging Dalvik to support ‘legacy’ (for me) F-Droid applications, all seems feasible!
<JamesJRH>
Albeit, a big challenge.
<JamesJRH>
The good thing, though, is that each of these things is an independent task that has its own benefits, before the others have been completed.
<JamesJRH>
For example, porting NixOS to an ARM platform, for me, will double-up with my existing efforts of wanting to run NixOS on the BeagleBone Black, and other ARM open hardware.
<JamesJRH>
So that part of the challenge I'm already facing.
<JamesJRH>
The same sort of argument applies to writing Haskell Gtk applications for desktop, I'm going to do that anyway, so the Haskell+Hildon effort should be somewhat divided.
<JamesJRH>
After I get NixOS working on the Neo900, the task of packaging Hildon and other necessary components will then enable the use of Maemo applications (which I'd also need to package, but hey, packaging is relatively easy having packaged many things now by this point).
<JamesJRH>
And finally, packaging Dalvik and F-Droid applications enables those applications.
<JamesJRH>
So after a point, this can be a work-in-progress but already be getting a lot of benefit out of it, and benefitting others..
<JamesJRH>
s/.././
<JamesJRH>
So, now that what I'm going to do about the software is cleared, I'd like to financially support the project.
<JamesJRH>
Though I'm still quite miffed that you can't get more than 2 pointers.
<JamesJRH>
Anyway, how much is the phone likely to be?
<JamesJRH>
I'm expecting to pay at least 500£. My current and previous phone were both about 500£.
<JamesJRH>
But this is open hardware, which, to me, is more important than pretty much anything else about a phone.
<JamesJRH>
about hardware in general.*
<JamesJRH>
And I want to support open hardware, as well as it being an opportunity to get a fully-fledged GNU+Linux system in my pocket.
<JamesJRH>
Rather than shoddy Android.
<JamesJRH>
That doesn't even have decent shell untilities by default.
<JamesJRH>
utilities*
<JamesJRH>
Let alone GHCi and Haskell support.
<JamesJRH>
I'd rather preörder the full price of the device now, rather than give a cautiously small donation now and the full price later.
<JamesJRH>
I don't mind the wait, just when you're ready, but you have the money upfront.
<JamesJRH>
Hmm, maybe I should put this in an email.
<Wizzup>
JamesJRH: if you order at the shop
<Wizzup>
you pay the first part
<Wizzup>
and you can add more money
<Wizzup>
which will be subtracted from the final price
<JamesJRH>
Okay, cool!
<Wizzup>
See the shop howto in the recent news letter
<JamesJRH>
I only signed-up to the newsletter earlier today, I think around the time that I joined this channel.
<ZetaR>
JamesJRH: If you wish to have >2 touch on the Neo900, then there is always the hackerbus (DIY expansion port in the battery case), or USB OTG (USB host and client. Most phones only get client). Either of those could be used to add on some kind of chording device, or basically whatever you can get the hardware together for.
<ZetaR>
I think its funny that Ebay thinks that outlet wallplates and pipe elbows are related to "NFC SWP".
<JamesJRH>
How much will 990€ roughly be in pounds, after tax, exchange fees, shipping, etc.? Am I looking at about the same in pounds? I.e. about 1000£?
<JamesJRH>
That's doable.
<JamesJRH>
But I'm really not happy about the lack of pointers.
<ilon>
JamesJRH: do you actually use more than two touch points at the same time on other devices?
<JamesJRH>
Yes, see earlier re. Hexiano.
<ilon>
I think I've never done that myself, despite using clunky android gear for about 3 years :/
<ilon>
oh, keyboard
<ilon>
then I get why
<ilon>
JamesJRH: regarding only 2point touch, its related to the fact that the old n900 screen will be kept, and just enhanced with a better controller + software to "hack" it to get 2 point touch
<JamesJRH>
Wizzup, ZetaR: I'm not going to pay ~1000£ for something that I'm going to be thinking from day 1 “I really can't wait for version 2”, but be out of pocket for version 2 if it did come along, but on the other hand, I really don't want to /not/ support you. I regret that I didn't support the Neo FreeRunner, though it was a similar sort of dilemma of wanting a version 2.
<wpwrak>
ZetaR: excellent, thanks a lot !
<ilon>
JamesJRH: which is usually not the case for resesistive screens
<ilon>
JamesJRH: I think that if you assume that there wont be another phone like the neo900 in the next 10 years, i think you'll get quite close.
<JamesJRH>
So is it at all possible that I can ‘downpay’ for version 2, such that you still get upfront funding for R&D, but that I'm not wasting money on an intermediate phone which is too much for what it's worth to me?
<ilon>
JamesJRH: I've personally been cursing over both iphones and android phones since i abandoned my last n900 :/
paulk-collins has quit [Ping timeout: 276 seconds]
<JamesJRH>
Yes, but I'm relatively poor. 1000£ is currently over half of my savings.
<ilon>
JamesJRH: I dont know if I were unclear.. The following is just my personal opinion as a donator / supporter of the project: I dont think there ever will be a version 2.
<ilon>
JamesJRH: I totally understand you, I really had to convince myself to cash out for it as well
<Wizzup>
JamesJRH: if it interests you at all I wrote some sw to forward input devices over the network
mvaenskae has quit [Ping timeout: 264 seconds]
<Wizzup>
I guess it's not of much use to you
<ilon>
JamesJRH: but in the end, since the radio on my last n900 started to behave shady (dropping calls etc) and my wife forced me to switch to something else, I've been constantly unsatisfied with my phones, I've had iphone 4, Samsung Galaxy Grand Duos, Leagoo lead 2, and currently a LG G3 858HK
<JamesJRH>
But either way, I don't like being in this situation where I want to support the project for fear of there not being a version 2, but really it's the version 2 that I want, so why can't I preörder that? The company can operate in credit, and not debt, during R&D.
<ilon>
So if this phone last for atleast 5 years (which seems to be the case when they calculate on components failure rates), it would easily be worth it.
sparetire_ has joined #neo900
<JamesJRH>
Wizzup: That is interesting, but doesn't help me with this issue.
<JamesJRH>
I can't go 5 years with only 2 pointers, without also carrying another device… which I guess is okay for now.
<JamesJRH>
Hmm, I mean even just 3 or 4 pointers would be enough to get by.
<JamesJRH>
I'm pretty novice even at my own application, but I can play a few tunes and some of them require 3 touches even with my ‘single-touch chording’ feature enabled which allows 4ths and 5ths to be played with just 1 touch, so a major or minor triad only requires 2, and an augmented chord only requires 3.
<Wizzup>
why does a keyboard not suffice?
<Wizzup>
I didn't look at the app, btw.
<ccnnjj>
It's a musical keyboard app, e.g. a piano/synthesizer. I can see the usecase, and is an unfortunate situation.
illwieckz has quit [Remote host closed the connection]
<ccnnjj>
A physical (typing) keyboard would be less natural than the touchscreen. Although I'd expect it'd feel cramped on the screen as it were.
dal has quit [Ping timeout: 255 seconds]
<ccnnjj>
It's a good excuse to learn the Theremin. There's an app in extras for that.
<JamesJRH>
Um, I mean: s/an augmented chord only requires 3/a dominant seventh only requires 3/
<JamesJRH>
An augmented triad cannot make use of single-touch chording, due to no adjacent keys, as neither can a diminished seventh, a 4-note chord which also doesn't contain any 4ths or 5ths.
modem has joined #neo900
<JamesJRH>
So 4, really, is the minimum for reasonably basic music.
<Wizzup>
would it not be possible to replace the digitizer, technically?
<JamesJRH>
But I often play something that has a dominant seventh in it, so personally, less than 3 is really not good.
<JamesJRH>
I don't know.
<JamesJRH>
Are you asking me?
<Wizzup>
I really wonder if you can even fit 8 fingers on the n900 screen :)
<Wizzup>
JamesJRH: well, I think it should be possible, but likely not easy
<Wizzup>
but it's possible -- and then connect to the hackerbus
<wpwrak>
or just connect a nice big usb-midi keyboard to usb :)
<Wizzup>
yep :)
<Wizzup>
which you can try now on the n900 even :)
<JamesJRH>
Yeah, 8 isn't something I'd push for, but it's not hard to fit 6 on most phones.
<JamesJRH>
Such as 3 on each hand.
<Wizzup>
does it need to be touch?
<Wizzup>
touchscreen*
<JamesJRH>
Um, actually, with the thumbs, 8 pointers is not hard to fit, but still, it's not a critical thing. I certainly haven't yet got the skills to play with 8 pointers.
dal has joined #neo900
<JamesJRH>
Though, seeing as I use all 10 digits to correctly touchtype, I'm sure that over the years I'll eventually be able to play more advanced stuff.
<JamesJRH>
wpwrak: I'm not aware of proper Wicki-Hayden keyboards that are not concertinas/accordians, and if you get MIDI 1 of them they probably cost more than a Neo900. :-P
<JamesJRH>
1s*
<Wizzup>
midi usb keyboards of cute size are like 40 euro
<JamesJRH>
Hey, if that 10.1in is only 39.99$...
<Wizzup>
I am not saying it will easily work, but it should be possible
<Wizzup>
Please note that likely maemo may or may not support the multitouch APIs
<Wizzup>
I think mpx didn't even exist at the time
<Wizzup>
let alone XI2
<Wizzup>
but you can read raw events likely
<JamesJRH>
Really, how much would it cost to put a better screen on the Neo900?
<JamesJRH>
Hmm.
modem has quit [Ping timeout: 245 seconds]
<Wizzup>
'better' is relative
<Wizzup>
and it's likely not easy, but doable
<ilon>
The screen factor is kind of unusual
<ilon>
also, at the scale of <500 devices, it would probably be rather expensive
<JamesJRH>
Wizzup: Notice that I didn't say capacitive, though. The screen technology is irrelevant as long as it performs, and I accept that resistive is a lot better than when I first tried it many years ago.
<JamesJRH>
Are there resistive screens with more pointers?
* Wizzup
afk
<Wizzup>
zzz
dal has quit [Ping timeout: 255 seconds]
<JamesJRH>
ilon: Well, maybe, but I'd like to see more than 500 people use it. And I'd also like the other Neo900 users to be able to get more out of my application, if I go to the effort of making it native for Maemo.
<JamesJRH>
Even just 4–6 pointers, rather than the 10 pointers of most devices of the last 3 or 4 years.
<JamesJRH>
My old LG Optimus 3D had 10.
<JamesJRH>
That's circa 2011.
paulk-collins has joined #neo900
mvaenskae has joined #neo900
<JamesJRH>
It /is/ very interesting though, a /resistive/ multitouch screen could, with enough pointers, be far better than even the highest-end capacitive screens. A while back I tried to implement a velocity feature that uses the pressure sensitivity to change the velocity (strength; loudness) of the note.
<JamesJRH>
It was disappointing, and I abandoned the idea.
<JamesJRH>
The pressure sensitivity was just too unreliable, too inaccurate.
<JamesJRH>
Apparently it's actually not pressure at all, but the area of contact that your digit makes with the screen.
<JamesJRH>
Sometimes hitting hard would barely make a sound, other times playing softly would be interrupted by a rouge and intrusive loud high-velocity note.
<JamesJRH>
s/rouge/rogue/
<JamesJRH>
Potentially, modern resistive technology, resistive being based-on /pressure/, could be the way forward.
<JamesJRH>
Just, why can't it do more pointers?
<JamesJRH>
Well, it can, it surely can – if it can do 2 then there's no fundamental reason why it can't do more than 2, surely.
<JamesJRH>
But how much more would such panels cost, and are there drop-in replacements for the planned panel? I.e. are there better models in the same product line of panels, or that are otherwise compatible?
<JamesJRH>
Hmm, maybe my reasoning about version 2 is bad. Version 1 isn't just to bridge the gap financially and the intermediate cost isn't totally ‘wasted’ per se, especially not at this tiny scale. It's actually a vital source of feedback and community testing for version 2, and that which looks wasted was actually ‘spent’ on testing.
panais has joined #neo900
<JamesJRH>
At least at this tiny production scale; for mass production it's unacceptable because they can easily afford to test better but rather they don't mind us replacing our devices every couple of years.
<JamesJRH>
So I guess that I'm doing a service to the future community by adopting version 1, not just for financial reasons but also for testing reasons.
<JamesJRH>
That helps me justify the cost, I guess.
<JamesJRH>
Although, if the first thing I'm going to say is that more pointers are needed, this isn't really in need of testing, I can tell you that right now, so this argument doesn't apply to everything.
<JamesJRH>
But for other things, I guess this is something that helps me justify it.
<JamesJRH>
And of course for other things still, there'll be obvious advantages over Android, making it look like it's actually Android that's the one in development. :-P
<JamesJRH>
There are problems in Android that I don't think should have made it out of the version 1 beta, yet they're still there now.
<JamesJRH>
Actually what triggered me to check back at the Neo FreeRunner, which then made me to realise that Neo900 existed, is that I had an idea for a new simple application to make booting live CDs really easy, making use of the g_mass_storage kernel module, and then decided that Android is not going to make that easy, and certainly not without requiring root access, which not everyone has.
<JamesJRH>
It would involve selecting an ISO file from the filesystem and then running a command such as this:
<ZetaR>
JamesJRH: The resistive touchscreen tech I have seen would not be able to support >2 touch AFAIK, just because of how the screen works. You might be able to tell there were 3 touches, but the location would be ambiguous. But I'm not really sure why you would want to do that on the touchscreen, it seems like it would be more apropos to use a device with buttons since you would need a really big screen to do chords comfortably.
<ZetaR>
You could probably do this on the N900 and Neo900 with the hardware keyboard.
<ZetaR>
Also, I'm not sure if you mistook me about additions to the Neo900. I was not referring to a V2 (or were you commenting in reply to someone else?), but rather to how the Neo900 will have access to basic system busses via a connector in the battery compartment.
burmas has joined #neo900
<ZetaR>
This would allow you to basically add almost any kind of device that you could create a circuit for.
<ZetaR>
You could even create a modified backplate with keys on it, if you wanted to.
modem has joined #neo900
Pali has quit [Remote host closed the connection]
dal2 has joined #neo900
burmas has quit [Quit: Konversation terminated!]
afics has quit [Ping timeout: 245 seconds]
afics has joined #neo900
pigeons has quit [Ping timeout: 245 seconds]
pigeons has joined #neo900
pigeons is now known as Guest18168
che11 has joined #neo900
che11 has quit [Ping timeout: 244 seconds]
paulk-collins has quit [Remote host closed the connection]