Back
[02:09:11] <kieran491> G'day, I am attempting to read a Device Signature of an ATXMEGA32C4-AU with an AVR Dragon but i keep getting getting an error about not been able to enter programming mode when i check the output in the console i see this message "[ERROR] Got error setting up PDI mode: Device is not supported in this emulator mode. Debugger command setParameter failed., ModuleName: TCF (TCF command: Device:startSession failed.)". So i am not
[02:09:11] <kieran491> sure where to go from here, from what I've been able to understand from googling, some device weren't supported by the avr dragon or the other way around but my device appears to be supported so i am not sure where to look the wiring appears fine and it is able to read the target voltage.
[02:20:16] <Xark> kieran491: Sounds like your programmer isn't Xmega compatible (which uses PDI vs SPI programming [or something like that]).
[02:20:52] <Xark> "Note: Atmel AVR XMEGA PDI mode on AVR Dragon does NOT work for the following XMEGA devices: A3/D3 - revisions B, C and E or A1 (up to revision K)."
[02:21:13] <Xark> ^
[02:21:44] <kieran491> Xark, Thats, what i started to think but i checked Atmels Product page for the avr dragon and the ATXMEGA32C4-AU is supported from what i can see
[02:22:03] <Xark> Well, the error makes that seem incorrect.
[02:22:36] <Xark> Dragon is pretty old...perhaps get Atmel-ICE :)
[02:23:20] <kieran491> Xark, I would love to just recently got it because i didn't know where to start and saw a lot of poeple where using it\
[02:24:05] <Xark> kieran491: Well, I have a Dragon, but I never tried with Xmega. I know my AVR-ISPmk2 worked fine for PDI...
[02:26:00] <kieran491> I guess I'll either have to make a break through or bite the bullet. just a shame as it hasn't been used for any programing yet
[02:27:14] <kieran491> Its wierd thought because the AVR Dragon is also listed as a supporting tool Atmel Studio 7
[03:17:08] <kieran491> for the ATXMEGA32C4
[07:03:23] <Lambda-Aurigae> make sure you update the dragon's firmware.
[07:04:08] <kieran491> Lambda-Aurigae, I got a notice about that when i started atmel studio, As a result i updated it to the latest 1.7....
[07:06:53] <kieran491> But that hasn't fixed the problem unfortunatly
[07:07:02] <Lambda-Aurigae> then talk to Tom_itx and see if he has any of his isp/tpi/pdi programmers left to sell.
[07:10:47] <Tom_itx> mine doesn't work on the newest studio
[07:10:56] <Lambda-Aurigae> oh.
[07:10:58] <Tom_itx> and i'm not pushing them much anymore
[07:11:00] <Lambda-Aurigae> oh well.
[07:11:07] <Lambda-Aurigae> nevermind then.
[07:11:18] <Tom_itx> since they have the new programmer that does just about everything for less than the dragon
[07:11:18] <Lambda-Aurigae> I'll stop pushing them too.
[07:12:01] <Lambda-Aurigae> what's the new programmer called?
[07:13:07] <Lambda-Aurigae> http://tools.woot.com/offers/ultimate-mosquito-killer-and-pest-control-led-bulbs-4pk?ref=eml_w_to_tg_img&utm_source=Daily+Digest&utm_campaign=90c47274d5-Daily+Digest+-+20160514+-+Woot&utm_medium=email&utm_term=0_c5ca76da11-90c47274d5-283918266#tracked
[07:13:13] <Lambda-Aurigae> totally off the subject...neato though.
[07:13:26] <Tom_itx> http://www.digikey.com/product-detail/en/atmel/ATATMEL-ICE-PCBA/ATATMEL-ICE-PCBA-ND/4753383
[07:14:00] <Tom_itx> http://www.digikey.com/product-detail/en/atmel/ATATMEL-ICE/ATATMEL-ICE-ND/4753379
[07:14:09] <Tom_itx> if you want wires and a box
[07:15:04] <Tom_itx> http://www.digikey.com/product-detail/en/atmel/ATATMEL-ICE-BASIC/ATATMEL-ICE-BASIC-ND/4753381
[07:15:10] <Tom_itx> or if you just want a box
[07:15:40] <kieran491> Thanks for that Tom_itx, I've probably wired it up incorrectly or done something equally as incompetent as this is my first time messing around with xmega aswell as PDI
[07:16:21] <Tom_itx> http://tom-itx.no-ip.biz:81/~webpage/boards/USBTiny_Mkii/USBTiny_Mkii_User_manual_index.php
[07:16:28] <Tom_itx> pdi pinout is about midway down there
[07:18:07] <kieran491> Are there any common gotchas with PDI?
[07:18:50] <liwakura> Tom_itx: need hosting?^^
[07:25:13] <Tom_itx> no
[07:25:21] <Tom_itx> it's worked fine for years
[07:31:13] <Lambda-Aurigae> he has hosting
[07:31:21] <Lambda-Aurigae> it's right there.
[07:31:25] <Lambda-Aurigae> just not on port 80.
[07:31:59] <Tom_itx> my stupid isp blocks outbound 80
[07:49:08] <kieran491> Tom_itx, Are you by chance able to tell me what this error means "Device is not supported in this emulator mode. Debugger command setParameter failed., ModuleName: TCF (TCF command: Device:startSession failed.)"?
[07:50:26] <Lambda-Aurigae> http://www.avrfreaks.net/forum/avr-dragon-does-not-see-xmega-chips
[07:50:52] <Lambda-Aurigae> it seems you have to disable jtag before pdi works on that chip.
[07:51:50] <kieran491> I can't find any mention of JTAG in the manual or am i missing something?
[07:52:03] <Lambda-Aurigae> I'm just doing google search here.
[07:52:23] <Lambda-Aurigae> http://www.avrfreaks.net/forum/xmega32a4-and-avr-dragon
[07:54:56] <Lambda-Aurigae> mostly what I'm seeing is either that particular chip is not supported by the dragon and you need jtag or you have to disable jtag.
[07:58:55] <kieran491> Theres no mention of JTAG in the data sheet i got from element14, So i don't think thats the problem. Orignally i though the problem might be the chip wasn't supported but after tripple checking, the product page for the AVR Dragon and the product page for the chip itself they both state that they're compatible as well as AS7 mentions the AVR Dragon as a supported programmer for the chip.
[08:01:11] <Lambda-Aurigae> it definitely has jtag.
[08:01:28] <kieran491> I need to get the datasheet off atmel i think
[08:01:35] <Lambda-Aurigae> http://www.atmel.com/Images/Atmel-7766-8-bit-AVR-ATmega16U4-32U4_Datasheet.pdf
[08:02:03] <kieran491> Oh wo, Completly diffrent datasheet
[08:02:07] <Lambda-Aurigae> but it looks like the jtag and pdi ports are on different pins.
[08:02:41] <Lambda-Aurigae> so, I dunno.
[08:02:49] <Lambda-Aurigae> oh
[08:02:50] <Lambda-Aurigae> wait
[08:02:53] <Lambda-Aurigae> that's the wrong chip.
[08:02:56] <Lambda-Aurigae> that's not an xmega
[08:03:36] <kieran491> Ah that would explain why :), For a second i thought i was going crazy
[08:03:38] <Lambda-Aurigae> google dropped the X for me.
[08:03:57] <Lambda-Aurigae> http://www.atmel.com/images/Atmel-8387-8-and16-bit-AVR-Microcontroller-XMEGA-A4U_Datasheet.pdf
[08:05:02] <Lambda-Aurigae> looks like no jtag on that chip.
[08:05:05] <Lambda-Aurigae> that's really screwy...
[08:05:06] <Lambda-Aurigae> but,,
[08:05:31] <Lambda-Aurigae> again, never used xmega chips myself.
[08:05:47] <kieran491> This is my first time using one myself
[08:06:02] <Lambda-Aurigae> I skipped those and went to 32bit pic chips myself.
[08:06:59] <kieran491> I bought them to experiment with and grabed the dragon at the same time
[08:08:56] <Lambda-Aurigae> I've considered them a time or two but everybody I've seen work with them has had issues.
[08:09:20] <Lambda-Aurigae> and I found the pic32 chips to be much easier to work with for me.
[08:09:30] <Lambda-Aurigae> specially as some come in a dip package.
[08:10:30] <kieran491> This was the first time I've used an external programmer, so i had no idea which one to purchase so i went for the one that seemed to have a large user base or at least i could tell from my view point
[08:12:36] <Lambda-Aurigae> you might hang around a while and ask again in a few hours...there are some around here who use the xmega chips.
[08:14:10] <kieran491> I'm quite sure it's something silly I've done and someone with a bit of experience will have a good laugh :)
[08:15:25] <Tom_itx> not alot of ppl here have chosen to use the xmegas yet
[08:15:55] <kieran491> Jump the gun did I?
[08:16:10] <Tom_itx> the dragon used to be the one to have but that one i posted is quite a bit newer
[08:16:33] <Tom_itx> the dragon was around before xmegas and support for them was added to it
[08:16:53] <kieran491> I should have come on here and asked for input before i went out and purchased it
[08:16:54] <Tom_itx> early ones wouldn't support them
[08:17:20] <Tom_itx> naw, everybody needs a good collection of programmers on the shelf collecting dust
[08:18:16] <kieran491> Starting to build a collection then :)
[08:20:04] <carabia> the atmel ice debugs arm too yes?
[08:20:05] <Tom_itx> i've got 6 different styles i can think of off the top of my head
[08:20:25] <Tom_itx> carabia, the new one is supposed to but i don't own one so can't say for sure
[08:20:53] <Tom_itx> i suggest you read about it on their site
[08:21:41] <Tom_itx> arm are jtag programmed afik
[08:22:28] <carabia> jtag and/or swd
[08:23:07] <carabia> quick search suggests that ice does swd too, while dragon doesn't. Off digikey prices, dragon's merely half the price of the ICE
[08:24:18] <Tom_itx> http://www.digikey.com/product-detail/en/atmel/ATATMEL-ICE-PCBA/ATATMEL-ICE-PCBA-ND/4753383
[08:24:20] <carabia> I have a st-link v2 clone to debug arm so yeah. Haven't got a debugger for atmegas
[08:24:23] <Tom_itx> that is the newer one
[08:24:31] <Tom_itx> offered 3 different ways
[08:24:45] <carabia> Oh really?
[08:24:53] <Tom_itx> links above
[08:24:55] <carabia> So, a plastic case costs $70? =D
[08:25:00] <Tom_itx> yup
[08:25:07] <Tom_itx> wires are extra too
[08:25:23] <carabia> ...naturally
[08:26:13] <carabia> wouldn't be so bad then. and yeah it does swd. which is like the 21st century.
[08:27:23] <Tom_itx> ttyl
[08:27:37] <carabia> troubleshooting is about 100 times easier when you can step through and read values
[08:28:10] <carabia> as i don't have a debugger for avrs atm
[08:41:40] <Lambda-Aurigae> 555 timer in 1 shot mode with a button to trigger it as the clock source for avr...one step through!
[08:41:52] <Lambda-Aurigae> and output stuff to an i/o port and display on LEDs
[08:51:14] * LeoNerd ponders options of using an INA219 and some sort of programmable 24V PSU as a single-channel SMU for LED classification
[08:54:18] <LeoNerd> 12bit ADC, basic resolution is 4mV / 100µA
[08:54:27] <LeoNerd> Sounds decent enough for simple hobby stuff
[08:56:08] <Lambda-Aurigae> you trying to measure current draw on LEDs?
[08:59:04] <LeoNerd> I want characterisation curves for my LEDs, yes
[08:59:35] <LeoNerd> So far I've been doing this really manually with two multimeters set up in V+A configuration and a power supply.. slowly twiddle the knob and write down current and voltage at successive intervals
[08:59:51] <LeoNerd> It's really slow and manual though, and prone to manual errors in writing down the numbers. I want something a little more automated
[08:59:52] <Lambda-Aurigae> more work than I want for LEDs
[09:00:02] <Lambda-Aurigae> 5V and a 330ohm resistor and I'm good for most of them.
[09:00:05] <LeoNerd> But I don't want to pay $4k for a real bench SMU
[09:00:16] <LeoNerd> I'm trying to match intensities on lots of them
[09:00:28] <LeoNerd> Admittedly this also requires a lux meter :)
[09:01:13] <Lambda-Aurigae> CDS photocell.
[09:01:35] <Lambda-Aurigae> and an ohmmeter.
[09:01:58] <kieran491> I saw on youtube a guy just used a solor cell and measured the voltage off that to determine the output of a "bulb"
[09:02:00] <LeoNerd> Yeah but then you gotta know the transfer function of them and they're usually quite weird
[09:03:00] <LeoNerd> Ideally I want a two-channel four-quadrant thing so I can do transistors and allsorts too
[09:03:19] <LeoNerd> Buuuut.. INA219 isn't bidirectional so I couldn't use it in negative voltage area
[09:03:31] <Lambda-Aurigae> kieran491, that's how many lux meters work...just a little voltmeter and a solar cell to drive it.
[09:03:35] <LeoNerd> Hrm. Oh but I coooould perhaps use it with a split +12/-12 setup and keep the "common" in the middle
[09:03:52] <Lambda-Aurigae> LeoNerd, differential power!!! lots of fun.
[09:03:58] <kieran491> Lambda-Aurigae, Well there ya go the more you know :)
[09:04:11] <Lambda-Aurigae> the more you tear apart the more you know.
[09:04:27] <LeoNerd> I had been hoping for a chip like the INA219, but can do negative bus voltage too. That way I can just have two channels that can do +/-ve voltage and current, and a common ground
[09:05:35] <Lambda-Aurigae> kids these days don't tear things apart to learn about them.
[09:05:46] <Lambda-Aurigae> they just sit around and simulate the world or play video games.
[09:07:18] <LeoNerd> Eh... I think that's genuinely changing more lately
[09:07:36] <LeoNerd> Much as we deride it, I think things like arduino are responsible for that, in a good way
[09:07:59] <Lambda-Aurigae> maybe where you live...
[09:08:13] <Lambda-Aurigae> around here they even simulate farming!
[09:08:19] <Lambda-Aurigae> nobody wants to get their fingers dirty.
[09:12:32] <LeoNerd> Hrm.. What would you call a voltage sensor that can cope with negative voltages?
[09:12:43] <LeoNerd> Like, if I power it with +5V and gnd, but it can measure say, + or -20V
[09:12:56] <Lambda-Aurigae> ummm..
[09:12:58] <LeoNerd> "bidirectional" refers more to a current sensor I feel
[09:13:03] <Lambda-Aurigae> differential
[09:13:06] <Lambda-Aurigae> no
[09:13:08] <Lambda-Aurigae> that's not it.
[09:13:17] <LeoNerd> Bilateral?
[09:13:46] <Lambda-Aurigae> dual rail
[09:13:48] <Lambda-Aurigae> ?
[09:14:09] <LeoNerd> Hmm.. but it's not dual rail. I don't want to -supply- it with negative voltage.
[09:14:12] <LeoNerd> Well.. I suppose I could
[09:14:15] <Lambda-Aurigae> negative voltage supply
[09:14:18] <Lambda-Aurigae> and dual rail
[09:14:43] <Lambda-Aurigae> https://www.maximintegrated.com/en/app-notes/index.mvp/id/4050
[09:15:05] <Lambda-Aurigae> negative rail current sense amplifier there.
[09:16:59] <Lambda-Aurigae> https://e2e.ti.com/support/amplifiers/current-shunt-monitors/f/931/t/310480
[09:17:24] <Lambda-Aurigae> how to do it with the ina138 with a -30V rail
[09:18:09] <Lambda-Aurigae> gotta scroll down though.
[09:18:35] <Lambda-Aurigae> http://cds.linear.com/docs/en/application-note/an105fa.pdf
[09:18:47] <Lambda-Aurigae> feed it with an opamp...page 2 left side.
[09:19:52] <LeoNerd> Hrm.. all of these are purely analog components though. The beauty of the INA219 is it has an onboard 12bit ADC
[09:20:12] <LeoNerd> So you just talk I²C to it. Otherwise, I'd be looking into a separate ADC chip with associated cost/size implications
[09:20:28] <LeoNerd> (because I don't know any MCUs with better-than-10bit ADC)
[09:21:12] <Jartza> evening
[09:21:38] <kieran491> Does there need to be a load on the PDI clock?
[09:54:04] <kieran491> Huh, so when i connect my scope onto the clock line everything works fine, did a quick google search it appears adding a 22pf cap wmight fix the problem
[09:54:08] <kieran491> http://www.avrfreaks.net/forum/programming-atxmega-dragon-pdi-mode
[10:45:46] <carabia> Don't know about you LeoNerd but I'm getting pretty stable 24MHz operation off the salaea clone
[10:46:14] <LeoNerd> Maybe yours is a different clone
[10:46:25] <LeoNerd> What software?
[10:47:15] <carabia> using salaea's logic 1.1.15
[10:47:38] <carabia> LeoNerd: it was a year or two ago you linked me this personally off ebay
[10:47:46] <carabia> So I believe the unit is the same
[10:49:58] <carabia> could be the usb-cable too, i guess. I have a super short cable i made myself
[11:12:50] <theBear> with ebay the same seller/description is no reason to expect a matching unit
[11:27:23] <hetii> hi :)
[11:30:57] <hetii> hmm
[11:31:11] <Lambda-Aurigae> you gotta yell louder.
[11:32:05] <Thrashbarg> https://www.youtube.com/watch?v=-xfA4_j04zA
[11:32:08] <Thrashbarg> ^ this
[11:32:34] <hetii> I just consider what I try do to :) So I play a bit with my espstk project. I`m able to reprogramm by it avrs and add support for spi flash and serprog for flashrom software
[11:32:54] <hetii> now I plan add also support for eeproms that use i2c
[11:33:24] <hetii> but I use there 74hc244 buffer before my esp
[11:33:41] <hetii> and this is a bit tricy cause i2c use bidirectional communication
[11:33:51] <Lambda-Aurigae> so?
[11:34:06] <hetii> so suppose will need wire two buffer and act as one
[11:34:09] <Lambda-Aurigae> spi is bidirectional too.
[11:34:57] <hetii> yes but use different pins for output/input and i2c use the same pin as input and output
[11:35:20] <Lambda-Aurigae> is the problem voltage levels?
[11:35:23] <hetii> no
[11:37:16] <hetii> its more how to toggle my buffer to be able read/write data on single line
[11:41:03] <hetii> not sure if its possible If will use two line as SCL and two AS SDA
[15:48:50] <phinxy> PROGMEM tells the compiler to put data in the non-volatile program memory. Does this mean the only difference between storing data with progmem Vs. defining a variable is that the variable is pushed to the RAM?
[15:49:07] <phinxy> ..push to the RAM when its booted
[15:50:10] <Lambda-Aurigae> no.
[15:50:39] <Lambda-Aurigae> when you set a variable such as int a=15; then a is stored in flash and the variable created at runtime and assigned the value.
[15:51:05] <Lambda-Aurigae> in progmem, the data is stored in flash, period, and accessing it is not just like accessing a variable normally.
[15:51:33] <Lambda-Aurigae> http://www.nongnu.org/avr-libc/user-manual/pgmspace.html
[15:53:02] <Lambda-Aurigae> you have to actually use the pgm_read_byte() and similar macros to read the data.
[15:53:17] <Lambda-Aurigae> http://www.nongnu.org/avr-libc/user-manual/group__avr__pgmspace.html#ga349c2c134daa01aeea62256f1fd7489c
[15:53:32] <phinxy> sounds like what i said
[15:53:44] <phinxy> what dont you agree with?
[15:54:08] <Lambda-Aurigae> ok..rereading...yeah,,,mostly.
[15:54:15] <Lambda-Aurigae> more or less,,yeah.
[15:54:54] <Lambda-Aurigae> I misread...I thought you meant progmem data was pushed to ram automagically.
[15:55:30] <phinxy> when is 'a' pushed to RAM? when it needs access to it the first time or directly at boot? can you define this behaviour?
[15:55:44] <Lambda-Aurigae> ummm..
[15:55:53] <Lambda-Aurigae> I think when it is defined..
[15:56:05] <Lambda-Aurigae> by scope...global, local..
[15:56:21] <phinxy> ah
[15:57:02] <Lambda-Aurigae> global, at "boot time"
[15:57:05] <phinxy> a function for example could grab the default variable data from flash when it starts running, each time
[15:57:22] <Lambda-Aurigae> local variables in functions would be declared when the function is called and destroyed when the function closes.
[15:57:29] <Lambda-Aurigae> unless it's a static variable.
[15:57:45] <phinxy> are you allowed to define a static var inside a function?
[15:57:50] <Lambda-Aurigae> sure.
[15:58:18] <phinxy> hmm, does it tells the function not to destroy the data in there from the last time it ran the function if you get what im saying..
[15:58:23] <Lambda-Aurigae> http://stackoverflow.com/questions/5033627/static-variable-inside-of-a-function-in-c
[15:58:45] <Lambda-Aurigae> basically...declaring it static in a function tells it to create it the first time then maintain the data between calls of the function.
[15:59:17] <Lambda-Aurigae> without the static in the variable declaration it would start at whatever it was declared.
[16:46:08] <liwakura> on the avr, its allocated outside the stack then, a de facto global variable
[16:48:48] <Lambda-Aurigae> heap is for variables.
[16:49:31] <Lambda-Aurigae> I think.
[16:49:37] <Lambda-Aurigae> I could be wrong...kinda cold and tired.
[16:50:02] <Lambda-Aurigae> maybe heap is for malloc..
[16:50:08] <Lambda-Aurigae> it's too late in the day to think.
[17:00:25] <Jartza> static data goes outside heap, yes.
[17:00:41] <Jartza> something called "heap" is for malloc and free
[17:00:53] <Jartza> but usually that's just stupid thing in tiny mcu
[17:01:15] <Jartza> as you malloc and free the memory, you leave gaps and eventually it's bad
[17:04:08] <Jartza> start of memory is .data, then .bss, then heap and stack starts at RAMEND and "grows" downwards towards the heap
[17:06:11] <Jartza> when using PROGMEM, the data goes to flash and basically you just get pointer to flash address, data itself must be read by "other means" and it's not automatically copied to ram
[17:06:55] <Jartza> but if you use "c-strings", they end up in .data usually and eat up RAM
[17:07:05] <Jartza> .data is populated at start of your code
[17:07:43] <Casper> string, the ram eater
[17:08:05] <Jartza> http://www.atmel.com/webdoc/AVRLibcReferenceManual/mem_sections.html
[17:08:26] <Jartza> Casper: yes, strings in AVR code should always go to progmem
[17:10:36] <Jartza> many people think "const" does the trick, but it doesn't
[17:10:48] <Jartza> const should never be used to tell where the data is located
[17:11:52] <Jartza> and on avr it doesn't even work
[17:11:55] <cehteh> thats the problem with C
[17:12:11] <cehteh> const can express more than one semantic, but nothing exactly
[17:12:32] <Jartza> const char foobar[] = "Some pointless string";
[17:12:37] <Jartza> that'll end up in RAM anyway
[17:12:46] <Casper> yup
[17:13:10] <Casper> all it does is: this will never change, throw an error if something try to write to that memory address
[17:13:17] <Casper> (and even then, it may fail)
[17:13:41] <Jartza> cehteh: yeah, depends where the const is used :)
[17:14:08] <cehteh> no thats a C deficiency
[17:15:20] <Jartza> the pitfall many people run into happens with "const int* ptr;"
[17:15:24] <Lambda-Aurigae> it's a feature!
[17:16:18] <Jartza> cehteh: well, it's also good to note that const works differently depending if it's C or C++
[17:16:21] <Jartza> :)
[17:16:27] <cehteh> imo const is 90% useless .. in many cases the (any modern) compiler can deduces constness by itself
[17:16:59] <cehteh> one can use it to qualify the intended usage of a variable
[17:17:11] <cehteh> but there are so many glitches where it does not work out
[17:17:11] <LeoNerd> const can be useful to declare in your headers so other programmers can see it
[17:17:34] <LeoNerd> "my function does not intend to modify your data"
[17:17:41] <cehteh> yes
[17:18:34] <cehteh> but how about const returns? how do you initialize const stuff? not everything can be statically initialized even if later it should never ever be changed
[17:18:44] <cehteh> C lacks a lot in that department
[17:19:32] <cehteh> you'll end up with terrible hacks like casting constness away for some special cases
[17:19:45] <cehteh> which breaks other assumptions the compiler may do
[17:21:14] <Jartza> I can't count how many times I've seen programmers trying to "protect" their pointers with "const int *ptr;"
[17:21:26] <Lambda-Aurigae> I just don't use const
[17:21:48] <Lambda-Aurigae> I make no assumptions in programming and make sure I do what I need to do with my code.
[17:22:32] <cehteh> often thats the most sane way
[17:22:55] <cehteh> back in time it was meant to help compiler optimizations, but thats not the case anymore
[17:23:15] <Lambda-Aurigae> back when you had one or two platforms to compile for and they were much the same...
[17:23:31] <Lambda-Aurigae> rather than a dozen different microcontrollers that handle memory different one and all.
[17:23:35] <Jartza> I still use const but mainly because old habits die hard
[17:24:41] <cehteh> also __restrict which came like 20years too late :D
[17:25:56] <Jartza> I can't count how many times I've told programmer that "int *const ptr;" is really what they want
[17:31:53] <Jartza> but I agree, __restrict was late :)
[17:37:33] <Jartza> luckily we have assembly
[17:37:43] <Jartza> no need to guess what happens ;)
[18:20:19] <LeoNerd> Hrm. It seems that few USB devices I test via my so-called "USB3 hub" actually appear to work, yet are fine direct
[18:20:26] <LeoNerd> I begin to wonder if the hub is in fact not
[18:27:38] <Lambda-Aurigae> LeoNerd, hardware usb devices or v-usb based?
[18:31:21] <LeoNerd> Well, I know the V-usb ones will be crap. But the pololu programmer doesn't either. :/
[18:31:21] <LeoNerd> And that's a proper PIC
[18:31:41] <Lambda-Aurigae> interesting.
[18:32:05] <eszett> Hi! I have a question, can i run a crystal specificed for, say 20 pF, with 18 or 22 pF caps too?
[18:32:24] <LeoNerd> I doubt that will make a huge difference
[18:32:27] <Lambda-Aurigae> eszett, depends on the implementation, but, yeah, in my experience they aren't real critical.
[18:33:05] <eszett> Lambda: ahh very good, thx!
[18:33:38] <Lambda-Aurigae> eszett, I've run chips with crystals on solderless breadboard with no caps...the stray capacitance on the breadboard is often enough to get the circuit started oscillating.
[18:33:46] <eszett> so practically there is a window from 0-10 pF i can go up or down with my caps, against the specs?
[18:33:59] <eszett> oh..
[18:34:51] <eszett> Lambda: well, in reality i have a pcb ..
[18:35:15] <Lambda-Aurigae> the point is, with AVR chips, it's not real critical.
[18:36:33] <eszett> well then, that answer is sufficient for me :-)
[18:37:00] <Lambda-Aurigae> for a commercial product, stick with what it's supposed to have.
[18:37:23] <eszett> alright
[19:12:10] <eszett> was bedeutet "gurten" (Digikey)?
[19:13:00] <cehteh> eszett: just use one 18p and one 22p :D
[19:13:07] <eszett> haha
[19:13:20] <eszett> and they both compensate to 20 pico farad?
[19:14:02] <cehteh> gurten .. most likely 'gegurtet' from reel
[19:17:16] <eszett> i dont need that, just a few, say 40 crystals. and they want a fee for "gurten" i hope i dont have to pay that when i take 40 crystals from tape
[19:32:39] <eszett> someone wants to make a Groupbuy order from Digikey to save shipping?
[19:33:41] <Lambda-Aurigae> would still have to pay to have it shipped all over.
[19:33:54] <cehteh> mhm
[19:34:04] <cehteh> where in germany are you?
[19:34:12] <eszett> probably in makes only sense for people in same country
[19:34:21] <eszett> cehteh: frankfurt am main
[19:34:41] <eszett> cehteh: you are joining in?
[19:34:52] <cehteh> moment checking something
[19:34:56] <cehteh> i am in karlsruhe
[19:35:30] <eszett> i think "trust" is no problem neither from you to me or the other way around
[19:35:48] <cehteh> does digikey ship to mere mortals?
[19:35:53] <Lambda-Aurigae> sure
[19:35:57] <Lambda-Aurigae> I've ordered from them many times.
[19:35:58] <eszett> they ship free above 65€ order
[19:36:26] <eszett> which means: they say implicitely "little tinkerers, go away!" :-)
[19:36:50] <eszett> but we may hit the 65€ limit
[19:36:53] <Lambda-Aurigae> I'm just a "little tinkerer"
[19:37:06] <Lambda-Aurigae> I've had no problem hitting their free shipping limit when I order.
[19:37:51] <eszett> well little is a kind of definition, my budget is about 20-30€ for a few smd parts
[19:38:19] <cehteh> are you in urge?
[19:38:37] <eszett> not at all, im planning for the next weeks or month..
[19:39:49] <cehteh> ah then i can ask around some friends, i dont have much in mind now, i'd like to test some iso1540 chips, but thats just evaluation and they are expensive, so 2-3 of them at most
[23:00:21] <kieran491> G'day