Back
[00:03:24] <sirpatrick> Does it take up more memory if I split functionallity into multiple libraries vs not using helper libraries or functions? Like I have an SPI library, a UART library, etc.
[00:04:20] <sirpatrick> I know for bigger projects it is best to split it up
[00:23:37] <twnqx> shouldn't make a difference at all
[00:23:57] <twnqx> unless you have identical function parts
[03:41:34] <OndraSter> one for programmers:
http://anongallery.org/img/8156/push-rejected-rebase-or-merge-git-push-force.jpg
[03:44:53] <w|zzy> lol
[09:58:38] <carabia> here's a tough one, any ideas out of which junk one might find a to220 3v3 reg?
[10:07:07] <megal0maniac_afk> carabia: That pretty much entirely depends on what "junk" you have available to you
[10:08:21] <carabia> Looking around, ... But you see the problem is i'd prefer it to be to220 or similar, not smd
[10:08:31] <carabia> So i'm thinking it's a no-go
[10:09:12] <megal0maniac_afk> Not unless the "junk" is pretty old. 99% of new stuff will be smd where possible
[10:09:23] <carabia> yep..
[10:09:41] <amee2k> yeah... for low power they use SMT, and stuff that needs more juice has a switcher
[10:10:16] <carabia> well an smps could work but most likely smd as well
[10:10:17] <amee2k> i've even seen switchers in like <5W gear like SOHO-class shit ethernet switches
[10:10:25] <carabia> yeh
[10:10:40] <carabia> well they're dirt cheap
[10:10:59] <megal0maniac_afk> And won't be 3v3
[10:11:33] <carabia> I guess i'll have to desolder one out of my protoboard if i want to do anything tonight :x
[10:11:54] <megal0maniac_afk> Use a zener diode, and regular diodes to drop or offset the voltage :)
[10:12:01] <carabia> nope
[10:12:26] <carabia> maybe i'll have time to grab the packet from the store on friday so then i can just replace the 1117 in my proto board
[10:12:37] <carabia> fucking work. getting in the way of... hobbies
[10:14:57] <megal0maniac_afk> Gosh darnit
[10:36:27] <blathijs> Someone here who can comment on how gcc finds the crt*.o files? I'm trying to work with -mmcu=attiny13a and I get ld: cannot find crttn13a.o: No such file or directory. Using -mmcu=attiny13 works as expected. This seems to be related to the fact that the file is in /usr/lib/avr/lib/avr25/crttn13a.o, while the non-a version is both in /usr/lib/avr/lib/crttn13.o and /usr/lib/avr/lib/avr25/crttn13.o
[10:37:07] <blathijs> From reading avr-libc changelogs, the tn13 was copied back out of avr25 "to work with older gcc version", but I'm using 4.8 which I'd expect to be new enough?
[10:39:58] <blathijs> Hmm, seems
https://savannah.nongnu.org/bugs/?35407 and
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=695514 are relevent
[10:44:27] <megal0maniac_afk> blathijs: The binaries should work on either device. Not that's the point, but..
[10:45:12] <blathijs> megal0maniac_afk: Yeah, I've been just using attiny13 so far. But I need some of the 13a specific registers now (and I could of course just copy their definitions from iotn13a.h, but that's not very pretty)
[10:45:51] <blathijs> It seems create a symlink from /usr/lib/avr/lib/avr25/tiny-stack to /usr/lib/ar/lib/avr25 fixes things. Apparently the tn13a has a 1-byte SP and thus uses tiny-stack
[10:46:05] <blathijs> Not sure why it works for tn13 though, which also has a 1 byte SP I think
[10:46:11] <megal0maniac_afk> Should do
[10:47:25] <bsdfox> the registers are different on the 13a? I thought it was just power saving improvements
[10:47:30] <bsdfox> and cost inflation
[10:47:36] <blathijs> Ah, according to that savannah bug, gcc defaults to avr2 when it doesn't find any applicable library in other places
[10:47:49] <blathijs> bsdfox: There are some power-save related registers :-)
[10:48:02] <megal0maniac_afk> Yeah, what blathijs said
[10:48:32] <blathijs> bsdfox: BODCR for disabling BOD during sleep and PRR to disable the timer or ADC
[10:48:39] <blathijs> possibly others too
[10:49:04] <megal0maniac_afk> mega328 / mega328p have similar differences
[10:49:34] <bsdfox> ahh neat. BOD used to be only a fuse setting and I guess those two peripherals didn't have the clocks gated?
[10:53:33] <blathijs> bsdfox: I think so, yes
[10:57:06] <blathijs> In any case, it seems that the problem is that avr-libc 1.8.0 does not offer tiny-stack versions, but avr-gcc (>= 4.7.1 I think) does automatically select tiny-stack / -msp8 for devices that have an 8-bit SP but does not fallback to avr25, only to avr2
[10:57:14] <blathijs> So, just symlinking seems ok for now
[10:58:38] <zmo> hi!
[10:58:58] <zmo> does anyone know the difference between a -AU and -AUR package for an AVR MCU?
[10:59:07] <zmo> (same question with -MU and -MUR)
[10:59:11] <bsdfox> blathijs, you should submit a bug report
[10:59:41] <bsdfox> zmo, pretty sure the R is RoHS
[10:59:51] <bsdfox> so it's lead-free
[11:00:43] <zmo> bsdfox - actually I think it's more about the Reel packaging
[11:01:03] <zmo> the -AU is RoHS
[11:01:40] <bsdfox> oh so reel instead of tray?
[11:01:54] <blathijs> bsdfox: Yeah, there is one. It seems it's already fixed in trunk as well:
https://savannah.nongnu.org/bugs/?35407#comment2
[11:50:50] <megal0maniac_afk> zmo: Yeah, it's just tray vs reel. Look in the datasheet
[12:05:32] <zmo> megal0maniac_afk - did not find it in the datasheet
[12:05:35] <zmo> but thanks :-)
[12:07:51] <megal0maniac_afk> zmo: Should be under "ordering information", if it's there
[12:08:55] <zmo> ok
[12:40:06] <N1njAway> Suppliers will also often do the kitting out as well.
[12:40:32] <N1njAway> So someone like DigiKey can transfer tape in to tray, or tray in to tape at their end.
[15:11:01] <megal0maniac_afk> Reading datasheet for some old UV EPROM
[15:11:14] <N1njAway> THOSE ARE CLASSIFIED!
[15:11:22] * N1njAway burns megal0maniac_afk's datasheet quickly.
[15:11:26] <megal0maniac_afk> "Low power requirement. Active: 788mW
[15:11:38] <megal0maniac_afk> Standby: 184mW"
[15:12:31] <megal0maniac_afk> "Fast access time: 300nS max"
[15:12:44] <megal0maniac_afk> The wording hasn't changed, just the numbers :)
[15:13:59] <amee2k> now the marketing department can push the decimals around more before the customers start sending shit back as defective :>
[15:14:12] <megal0maniac_afk> I should buy some. Think I could erase it by putting it outside? The ozone layer isn't what it used to be :D
[15:14:30] <amee2k> maybe try peeing on it lol
[15:14:55] <megal0maniac_afk> That's a terrible idea
[15:15:10] <amee2k> i'd expect the datasheet to specify what kind of exposure is required to erase it
[15:15:26] <amee2k> (and what kind of exposure is acceptable without data loss)
[15:15:53] <megal0maniac_afk> Good point
[15:16:20] <N1njAway> Generally you need pretty intense exposure to 350nm or shorter UV. My old EPROM eraser had a germicidal lamp in it and you'd generally cook them for 20-30 minutes
[15:16:39] <N1njAway> Most times you'd put stickers over the windows to prevent long-term issues.
[15:16:51] <N1njAway> But yeah, odd that it's not on the datasheet.
[15:17:08] <megal0maniac_afk> I'm checking now. Didn't actually look
[15:17:22] <megal0maniac_afk> And then you burnt it so whatever :/
[15:20:01] <megal0maniac_afk> Spec says 253.7nm @ 12000uW/cm^2 for 15-20 minutes
[15:20:16] <megal0maniac_afk> And anything below 400nm will do the trick. Albeit slower
[15:20:33] <megal0maniac_afk> They talk about angstroms :)
[15:21:26] <N1njAway> Angstroms are just tenths of nanometers, convenient since they approximate the size of an atom.
[15:21:31] <megal0maniac_afk> I feel the need to buy one. For posterity's sake
[15:21:45] <N1njAway> megal0maniac_afk: An EEPROM?
[15:21:53] <megal0maniac_afk> UV EPROM
[15:21:58] <megal0maniac_afk> I have lots of EEPROM
[15:21:58] <N1njAway> Er EPROM
[15:22:12] <N1njAway> I have a bunch of EPROM's - how many do you want and what size?
[15:22:19] <N1njAway> You're in the US, right?
[15:22:38] <megal0maniac_afk> HAHA!
[15:22:39] <megal0maniac_afk> No
[15:22:46] <megal0maniac_afk> South Africa
[15:22:47] <OndraSter__> you should have told me
[15:22:49] <OndraSter__> I have plenty :D
[15:22:52] <OndraSter__> with many firmwares
[15:22:57] <megal0maniac> lol
[15:22:59] <OndraSter__> Z80, 8031s...
[15:22:59] <N1njAway> Oh that's right.
[15:23:00] <OndraSter__> 8051s
[15:23:02] <megal0maniac> Too late
[15:23:14] <OndraSter__> anywhere from 16Kbit to 1Mbit
[15:23:20] <megal0maniac> N1njAway: I wish I was in the US
[15:23:28] <OndraSter__> do you?
[15:23:32] <megal0maniac> Stuff wouldn't be so bloody difficult to get
[15:23:39] <megal0maniac> OndraSter__: Sometimes
[15:23:41] <OndraSter__> ok
[15:23:48] <N1njAway> megal0maniac: Clearly you should move here so that I could ship you EPROMs for less postage.
[15:23:52] <OndraSter__> :D
[15:24:05] <megal0maniac> I can't fault your reasoning. I'll pack my bags
[15:24:30] <megal0maniac> There are a bunch on ebay, upwards of $2
[15:24:39] <N1njAway> "... caught smuggling a briefcase full of EPROMs through customs...."
[15:24:40] <OndraSter__> free shipping?
[15:24:50] <megal0maniac> China, so yes
[15:24:53] <OndraSter__> oh ok
[15:25:10] <N1njAway> "Come with free egg-roll. Store your data long time!!"
[15:25:51] <megal0maniac> Haha!
[15:27:06] <megal0maniac> I was playing with the idea of getting some parallel RAM
[15:27:22] <megal0maniac> And making something completely analogue with it
[15:27:42] <megal0maniac> Well, analogue control, anyway. Gates and stuff
[15:28:17] <OndraSter__> huh?
[15:28:20] <megal0maniac> Nevermind. I'm talking about homebrew digital which is the antithesis of analogue
[15:28:26] <megal0maniac> I'm confusing myself
[15:29:17] <megal0maniac> N1njAway: Thanks for the offer, though :)
[15:29:29] <N1njAway> Sure
[15:29:55] <N1njAway> I have a lot of stuff around here that I don't need, but I don't want to throw out - would rather see said items go to someone who may want them.
[15:32:24] <megal0maniac> Maybe we can chat about it one day. Postage isn't _that_ much more expensive for international :)
[15:32:44] <megal0maniac> Surprising cheap from CZ actually
[15:33:16] <megal0maniac> ~6 to 15 USD from US, depending on the state it seems
[15:33:22] <N1njAway> Yeah but then I have to spend the extra 90 seconds making out the stupid green customs form.
[15:33:54] <megal0maniac> "Gift, $1"
[15:34:04] <N1njAway> Yep
[15:34:15] <megal0maniac> That's how China rolls
[15:34:32] <N1njAway> Had to ship a crapton of International for our Kickstarter project the other year, which meant many hours at the post office since they have to type all that crap in by hand.
[15:34:55] <N1njAway> But did get to know the postal worker really well, and he's a cool dude. So now I have a go-to guy at the local office, which is surprisingly useful.
[15:37:32] <megal0maniac> I never really thought about that. That must have sucked
[15:37:47] <megal0maniac> But contacts are always good
[15:38:15] <N1njAway> Yeah, pretty much did. The USPS website and software is completely f'ing next to useless, especially for international.
[15:39:07] <N1njAway> The Click 'N Ship manager software you can download very faithfully reproduces the overall experience of going to the US Post Office in person with regard to lack of speed, frustration, and inconvenience.
[15:41:41] * Gerritjan is away (auto-away after 2hrs of inactivity) - (10:26 pm)
[15:44:13] <megal0maniac> If our post office was anything like the website, then people would be falling asleep or dying mid-transaction. Usually before anything useful had been done. Luckily, the physical post office is pretty good. Website is next to useless. Only nice thing is the tracking, which works sometimes
[16:48:36] <Fornaxian> megal0maniac_afk, I have 6800 series microcontrollers with uvEPROMs on them.
[16:49:06] <N1njAway> Someone buy me a new chemical filter for our air purifier.
[16:49:16] <Fornaxian> use an old sock.
[16:50:59] <N1njAway> Filled with coal
[16:59:10] <Tom_itx> wait for christmas and you may get one free
[16:59:28] <Tom_itx> ( sock full of coal ) that is
[17:06:46] <N1njAway> I just need to order another one so we don't have to keep venting, gets cold in the winter. Problem is they're $1250 ea and special-order from the UK :(
[17:11:39] <Fornaxian> chemical filter
[17:11:40] <Fornaxian> bah
[17:12:00] <Fornaxian> it's just a little skin cancer,,,,and don't mind the coughing and itching,,,that'll go away in a year or so...
[17:12:51] <N1njAway> It's fine for the oven, but since the laser-cutter is also plumbed in, that's the one that really matters.
[19:02:11] <DanFrederiksen> is there a long term clock in an atmega? if you want to count seconds, minutes, hours? without counting microseconds
[19:03:15] <N1njAway> No RTC built in, just set up a timer and increment a variable.
[19:03:39] <bsdfox> DanFrederiksen, some of them allow a 32kHz crystal to be used with one of the timers
[19:06:14] <DanFrederiksen> but that requires an interrupt to increment the value? which interferes with the already strained realtime operation. I have to watch for a mains zero crossing anyway so I guess I can use that occasion to also count time
[19:07:39] <bsdfox> use a faster processor if you're running out of cycles..
[19:07:43] <N1njAway> Yep.
[19:08:30] <N1njAway> Or just configure the interrupt to fire once per second, if seconds are the finest granularity you need.
[19:08:53] <N1njAway> Then you have REALLY low overhead.
[19:13:43] <DanFrederiksen> bsdfox, it's not that it isn't fast enough. my concern is that with real time operation the interrupt will eventually delay something vital however rare. or can it prioritize?
[19:16:30] <DanFrederiksen> I'm doing PWM with current limiting in a high voltage system <400V so things change quickly
[19:17:38] <DanFrederiksen> but it's on mains so watching the zero crossing anyway will be a good opportunity to count time
[19:18:13] <DanFrederiksen> I don't need exact time, just a sense of long term time to watch a voltage grow
[19:18:21] <DanFrederiksen> charging battery
[19:27:13] <Valen> use mains zero as your time base
[19:27:17] <Valen> they use it for clocks
[19:30:43] <Fornaxian> do like the commodore vic-20 did and do an RTC in software that counts mains pulses..
[19:31:07] <Fornaxian> it was real annoying taking a vic-20 meant for the USA down to Australia....it lost time horribly.
[19:31:21] <N1njAway> Fornaxian: Does it run at different speeds on 50Hz vs 60Hz then? :D
[19:34:04] <Fornaxian> N1njAway, I think they used a different prescaler for the two...
[20:44:41] <Valen> is it ok to leave usb powered up when the rest of the AVR is powered down?
[20:44:48] <Valen> I'm thinking the answer is no
[20:45:06] <Valen> but I haven't seen it really mentioned as a problem
[20:45:22] <Tom_itx> i'm not sure why not
[20:45:37] <Tom_itx> it's somewhat separate from the rest isn't it?
[20:45:52] <Tom_itx> i mean, you can power the rest of the chip 2 or 3 different ways
[20:46:04] <Valen> i'm not too sure, I just see the "max voltage on any pin"
[20:46:18] <Tom_itx> what chip?
[20:46:21] <Tom_itx> U4?
[20:46:22] <Valen> 32u4
[20:46:24] <Valen> yeah
[20:46:42] <Valen> things like body diodes and the like worry me
[20:46:45] <Tom_itx> what do you mean by powered down?
[20:46:59] <Tom_itx> are you using external power for part of it?
[20:47:03] <Tom_itx> 3.3v?
[20:47:48] <Valen> i mean power is off
[20:47:53] <Valen> no power on any vcc
[20:47:55] <Valen> vcc = gnd
[20:48:00] <Valen> uvcc = 5v
[20:48:15] <Tom_itx> vcc=gnd doesn't sound good
[20:48:21] <Valen> somebody plugs their device into usb without plugging it into power first
[20:48:29] <Tom_itx> i think it would bleed power from the USB
[20:48:50] <Valen> that is my concern, but it must be a common issue, i just don't see it addressed anywhere
[20:49:02] <Tom_itx> i'd run it by dean
[20:50:01] <Valen> though i spose its happy running 5v on usb when the chip itself is at 3.3v
[20:50:12] <Valen> so they must have seperated it
[20:50:16] <Tom_itx> they do
[20:50:44] <Tom_itx> have you looked at the various wiring senarios?
[20:51:00] <Valen> Voltage on any Pin except RESET and VBUS
[20:51:00] <Valen> with respect to Ground(8) .............................-0.5V to VCC+0.5V
[20:51:52] <Valen> vcc +0.5v
[20:53:05] <Valen> Voltage on VBUS with respect to Ground..........-0.5V to +6.0V
[20:53:29] <Valen> no mention of vcc there
[20:53:35] <Valen> so sounds good right?
[20:55:17] <Tom_itx> look around P254
[20:55:23] <Tom_itx> 21.3 in my pdf
[20:55:25] <Tom_itx> but it's old
[20:56:08] <Valen> yeah they show it with the 3.3v (or so) power
[20:56:13] <Valen> just not with no power
[20:56:18] <Tom_itx> buss powered vs self powered
[20:56:43] <Tom_itx> 21.5 3.4 to 5.5v io
[20:58:35] <Tom_itx> i _think_ it would be ok like that
[20:59:06] <Tom_itx> but i got nothing to back that up
[21:02:05] <Valen> I'm going to say it is for the next few hours because plugging and unplugging the USB will suck ass during programming this thing lol
[21:02:23] <Tom_itx> heh
[21:02:49] <Tom_itx> you're programming via isp instead of bootloader?
[21:03:26] <Valen> nah bootloader, but mains powered device i don't want to leave running for ages
[21:12:52] <twnqx> so make it bus powered :3
[21:13:10] <Valen> too many watts
[21:13:12] <twnqx> two power domains, one only for the avr, one for the rest, avr can be dual powered
[22:20:49] <jerkey> help why does an atmega2560 read only 262106 bytes? shouldn't it be 262144 bytes?
[22:28:20] <Tom_itx> does it have a bootloader?
[22:28:30] <Tom_itx> look at the memory map of it for sure
[22:28:56] <tzanger> 262144-262106 is 38 bytes. too small for a bootloader, but perhaps that's interrupt vectors that can't be used for code.
[22:29:06] <Tom_itx> is ram part of that map?
[22:30:01] <Xark> tzanger: But the vectors would still be in the flash image, I would expect...
[22:46:29] <Casper> maybe it do not have 256kB exactly?
[22:49:23] <Xark> Casper: Perhaps not. However missing 38 bytes is pretty odd. :)
[22:49:29] <N1njAway> ATMEGA2560 has 57 interrupts for the vector table, so 57 words or 114 bytes
[22:50:44] <Xark> N1njAway: OK, but the vectors are "data" that need to be in the flash anyways. Also, IIRC on AVR, the "vectors" are actually asm code (i.e., typically a jump).
[22:51:40] <N1njAway> Yes, but they will be part of the flash image.
[22:51:50] <N1njAway> How are you reading the file out?
[22:52:10] <Casper> waitwaitwait
[22:52:11] <N1njAway> er flash
[22:52:28] <Casper> signature and fuses maybe is the rest of the bytes?
[22:53:37] <Xark> Casper: That makes some sense...can you read fuses with pgm_read_bytes? :)
[22:54:15] <Casper> probably not
[22:54:26] <N1njAway> Casper: He's coming up short reading the flash, no over.
[22:54:37] <N1njAway> Xark: What is resulting in a short read?
[22:54:41] <Casper> N1njAway: yes, that's what I mean
[22:55:08] <Casper> the fuses and signature is most probably part of the flash
[22:55:30] <N1njAway> It's not in the flash region, no.
[22:56:46] <Casper> then explain where it is?
[22:58:17] <Casper> I doubt it's in the eeprom
[22:58:25] <Casper> and it's surelly not in a dedicated part
[22:58:25] <N1njAway> The fuses are their own thing, as is the eeprom
[22:58:38] <Xark> N1njAway: The issue is from jerkey, not me (I just found it curious).
[22:59:50] <Casper> N1njAway: explain why sometime when you erase the flash it kill the signature and reset some fuses on older chip?
[23:01:29] <N1njAway> Chip erase will clear the lock bits and nuke the program flash, and will optionally clear EEPROM if EESAVE is not set.
[23:03:11] <N1njAway> FYI, if I read one of my ATMEGA2560's here I get exactly 262,144 bytes of flash, as expected.
[23:03:11] * Xark notes the ATmega2560 PDF "claims" flash address space is from 0x0000 to 0x1FFFF (16-bit words).
[23:03:49] <N1njAway> Yes, 256KB total, or 128K words
[23:04:17] <Xark> Yep. But no mention of any "special" areas (but it does mention that the space is divided for boot-loader/application).
[23:05:07] <N1njAway> The space is flat - usage of the bootloader is put somewhere in to the flash memory based on what the BOOTSZ fuses are.
[23:05:29] <N1njAway> There are no special areas in the flash, it is a set of 262,144 bytes. That's it.
[23:05:39] <N1njAway> EEPROM is a seperate area, as are the fuse settings.
[23:05:54] <Casper> but Xark, I suspect it's a bug in the programmer
[23:06:06] <Xark> Casper: I tend to agree (or "quirk" at best).
[23:06:19] <N1njAway> Casper: It has to be, since Atmel's tools read out exactly what they should.
[23:07:20] <Casper> and N1njAway... some avr are known to kill their signature and some reserved fuse bits from time to time with a chip erase, keep eeprom command
[23:07:37] <Casper> hence why I suspect the signature and fuses are in flash
[23:07:53] <N1njAway> Using what programmer?
[23:07:59] <Casper> any and all
[23:08:23] <N1njAway> I have reprogrammed literally many THOUSANDS of AVR devices in production here over the past 12 years and have never had that happen.
[23:08:41] <Casper> it's an hardware issue it seems that is triggered by unknown situation, possibly due to slightly unstable supply or something
[23:08:55] <Casper> atmel know about that issue, but couln't reproduce it
[23:08:58] <Xark> Losing the signature pretty much kills the chip I would think (or at least makes it tricky to program).
[23:09:09] <Casper> yes, tricky to program
[23:09:13] <N1njAway> What do they mean by "lose the signature" ?
[23:09:16] <Casper> need to force the programmer to ignore the signature
[23:09:36] <Casper> N1njAway: the signature go 0xFF 0xFF 0xFF
[23:09:44] <N1njAway> Does the part still work?
[23:09:57] <Casper> which cause the programmer to think that no device is connected
[23:10:00] <Casper> yes
[23:10:05] <Xark> Casper: Also tricky to diagnose I bet, as "bad signature" is pretty much generic problem 101 for programmers. :)
[23:10:10] <Casper> if you tell it to ignore the signature and detection
[23:10:27] <Casper> Xark: yeah
[23:10:32] <N1njAway> That's very odd. I have never had it happen to me personally.
[23:10:44] <N1njAway> I've thrashed chips hundreds of times when doing firmware dev, too.
[23:10:49] <Casper> some has been able to rewrite the signature somehow, but I couln't figure out how they did
[23:10:58] <Casper> it happened to me once
[23:11:43] <Casper> I think it was on an at90s8535
[23:16:38] <N1njAway> Very strange.
[23:17:32] <Casper> yup
[23:19:11] <Casper> man weather can be weird
[23:19:20] <Casper> yesterday: -7C
[23:19:27] <Casper> tomorrow: +7C
[23:19:52] <Casper> it always amaze me to see that many degree difference between so little days
[23:19:59] <Casper> and sometime in 1 day...
[23:20:35] <N1njAway> Casper: Where are you at again? :D
[23:20:45] <Casper> somewhere on the notrh
[23:21:48] <N1njAway> Singapore?
[23:21:59] <N1njAway> Oh, thats the server :)
[23:22:41] <Casper> yeah that's the server
[23:22:49] <Casper> but isn't singapore in the south?
[23:23:09] <N1njAway> Cleveland, OH here. It's 0C here now
[23:23:31] <Casper> 1C here
[23:23:51] <Casper> and I hope that by tomorrow my windshield washer fluid will not be a solid anymore
[23:25:09] <N1njAway> And yeah, Singapore is just south of the equator
[23:25:12] <Casper> need to change my tires this week end and replace that fluid... before it get frozen for the winter...
[23:26:41] <Casper> I can tell you that summer tires in the cold isn't the best thing at all
[23:26:58] <Casper> I can understand why in the south there is so many accident as soon as there is a bit of snow
