Retro 7 Segment Clock (Complete)

by parallyze Jul 22, 2018
Download All Files

Please Login to Comment

the shortest print time I can get on the base frame is 4 hrs is that right? or should I change settings to print this faster? I want to finish slicing so I can start printing! thank you!

4 hrs sounds more like wrong settings or very low speeds. At least that's what my crystal ball is saying.

I'm in the middle of printing this, but looking at wiring, I'm a bit confused. Is there no +5v/GND to the first 21 neopixels? Shouldn't that not work? Or does the DIN line pull enough current to power the first set of pixels??

This is some of the best design work on the web. All parts fit amazingly well. I plan to make several for friends.Thanks for sharing your things.

Hi Sir

i have finished this project with some modification and work fine. Now i need upgrade my clock to add temperature and display every 1 minute or more. i have DTH22 sensor.
can you share or send me the arduino sketch ?
i use( 7_Segment_Clock_v2_XT) this name of sketch on my clock.
my eMail: iibb010133@gmail.com

i attached edited sketch with some clock pic.

sorry my bad English


Your clock looks great. I am building one now. The designer did an awesome job designing this; a masterpiece. I too would like to be able to add the code for temperature display. Im like you though, just a beginner with Arduino. Please if you ever get the necessary code for this, would you please share it here. thanks.


the woodwork looks great! I'm just not sure what you'd like me to do!?
Which sketch do you want? The one you attached is based on the XL version as far as I can tell. And as you already have it I'm wondering what sketch you'd like me to send?

I want display temperature every 1 minet and wait 2 second Display temperature then back to clock.
I don't know how to add code. add temperature code to XL version then send me.

nice project and very well documented .. !!

Thanks. Actually I'm just right now working on the "Tiny Edition" of this... and the documentation is the only thing holding me back right now. :D
But at least it's english and german this time :)

this is beautiful project ! thank you for sharing :) i will post a make as soon as it is ready.

Hello Sir, how does edit the source code without red led pixels(show in the photo).becouse my city and i havent the 3d printer. Sir did you send the edited code in my email address malithkumara1000@gmail.com
Thank you

Hi! Thanks for sharing! I really enjoyed making this clock, and learned a lot from the code!

I only have 1 question, yesterday I connected an LDR GL5516 because the LED's were too bright at night. Played around with the settings a bit but now the LED's are not visible in daylight. Serial monitor goes from daylight 80 to dark (blocked the LDR) 110.

" The LDR (GL5516) I've used gave me an avgLDR of about 40-0 at medium ambient room lighting and
went down to ~150-200 very quickly when blocking light from it. You may need to adjust this. "

Can you explain what i'm doing wrong? I dont have that much of a difference. Thanks in advance!


there's a few values you can play around to make avgLDR behave differently. In line #179 there's this:

readOut = map(analogRead(pinLDR), 0, 1023, 220, 0);

0 - 1023 is the "raw" value of your LDR. This is then mapped to 220 - 0. Try to vary the ranges a bit, for example:

readOut = map(analogRead(pinLDR), 0, 1023, 220, 60);

This might shift avgLDR to more usable values. Another way might be decreasing the 0-1023 range. Basic principle is:

Current brightness (160/190/240 are defaults) - avgLDR = brightness to use. So if your lowest value at daylight is already around 80 you'll only get 160 brigthness on max level (240 - 80 = 160).

Edit: Don't have one of the clocks with an LDR at hand right now, so I can't really test this for some values.

Thanks again! I played around with the values and it's working now. After uploading the code I wasn't able to change brightness with the button, dont know why. Then I reconfigured the time and after that it worked like a charm!

Glad to hear it's working now for you :)

Yeah, brightness can be tricky at times. The way i read/store values to eeprom is a bit messy. So in case you changed one of the brightness levels(min/med/max) by hand and that was the value stored to eeprom at that time it wouldn't be able to set it correctly when using the buttons afterwards. Ending setup resets brightness, so that's one possible fix. Another one would be me finally cleaning up that sketch a bit... ;D

sir,many Thank for share your knowledge with us.sir i need the make a clock without any 3d prented parts.becouse my province is havent any 3d printers.i hope a making the clock 3 led per segment on the pvc board.
Sir,how to change the skatch without any not lightning leds? (Ex: All leds count is 88)
Same as this picture

Printed the parts a while ago but finally got around to soldering and assembling the clock. Amazing!! The fitment is so great.

Am going to rework the back box I think though, so it can house 2 proper buttons (I got 10mm ones) on top and have a place to put an LDR in between them

Hi! Always good to hear all parts did fit as expected :)

Regarding the LDR: Try to think of the place where the clock will end up before fixing the LDR in a certain direction/angle. One of the clocks I gave away had the LDR fixed to the left side on the back. Later it was put right next to a TV and the LDR had to move to the other side to stop the clock from reacting to the TVs screen :)

Really nice looking clock, most certainly going to print one of the 3 versions you created!
What filament did you use for the diffusers? Is it plain white or a semi transparent version?

I was using either transparent or natural material (PLA or PETG, whatever was loaded on one of my printers :) )

White PLA/PETG will block way too much light, can't recommend using this for the diffusers.

Ah thank you, will need to purchase some of that then if white doesn't work out with fewer bottom layers. Also, I assume the Arduino Pro Mini is the version with 5V/16MHz? The LED strips require 5V signal at least.

by the way, I'm currently working on a tiny edition of this one. It won't be using diffusers, I'm going for something like on the grid clock/lazy grid clock (2 sheets of paper or inkjet film). Don't know when I'll finish this one as it isn't really a priority right now. But it'll be max. 200mm width for all 4 digits and the ":" between using 1 led per segment.


Did you ever continue on this 1 led design? A lot of people around me loved this one, so gifting them a smaller version would be cool :)
Could always design something myself of course, especially after seeing how you planned to route them in your video

I had to postpone work on the 1 led-version because of other projects. But as it turned out that one is driving me nuts... I wanted it to be printable as easy as possible and wanted to totally avoid screws. Dozens of iterations later I'm still not happy with how it turned out, i.e. still don't have a good idea of how to finish this one. Can't remember any of the 7 segment things where I wasted so much plastic.... :D

The easiest and fastest to build would probably the SE-edition. If the build volume allows to you can align the two frames so they'll print as one piece. For the 1 led-version I wanted to avoid printing 30 diffusers, so this is currently more a mixture of 7 segment clock / lazy grid clock (pics attached).

Maybe I'll do some work on this one this week, let's see... ^^

Looks great! Was expecting the front to have cutouts, did not expect the full white front. Looks great though, different from the rest.
Might also give it a go to design something usable, perhaps not as refined as your designs..

A few screws in the back won't hurt in my opinion

Just a quick update: I haven't forgot about the tiny one. Actually I'm printing some parts right now.... as I totally misaligned some screw holes and yesterdays print just went to the trash bin... ^^

Check it out, used 10mm buttons and a LDR in the middle.

Looks good, I should think about adding at least a hole for the ldr. But you may not like the way the case turned out for the tiny one I'm working on currently. This probably is nowhere near about enough space for the 10mm buttons?

Made this one 6mm thicker as well to fit the counter nuts on the inside. Not ideal, your solution makes for much thinner cases!
Bought these in multiple colors, are easier to press:

Ah, I see. I once bought a bag full of those small 6x6 push buttons I'm using, so I must get rid of them. ;)

The case is always a shot in the dark somewhat. For standalone things like clocks I prefer using a Arduino Pro Mini + rtc (google for "ds3231 for pi", they are quite small. Especially if you desolder/cut off the pins). But some people just use a single esp8266 and sync time using ntp, others prefer using a more simple Arduino Nano (no ftdi adapter required) and a somewhat bigger ds3231 rtc (like the ones I'm using in the pictures of this thing).

Already thinking about a way to mount the buttons facing up. That way I could at least make the height somewhat adjustable by scaling the case along Z only. But the one thing I'd be more interested in is giving the software sketch a massive cleanup. :)

So funny, I am actually printing the SE as we speak. Got the frames and covers done and am now printing a modified electronics cage with real buttons

You're probably right and I shouldn't waste too much time in thinking about snap fits. The front is meant to be customizable, one of the ideas which makes constructing everything "smooth and clean" somewhat complicated. In one of the pictures you can see the alternate front with cutouts. I wanted this to be chooseable for people building the clock so anyone can print that optional part and put it in front of everything. :)

Too late, no way back already :D
Looks funny to use on a desk for example though! For a 'Nanoleaf' replica I did recently I used a sheet of cheap glossy photo paper as a diffuser and that worked really well, also looks nice when turned off

:D :D

Paper can work very well sometimes, have been using it quite often. But as soon as sun light enters the room it somehow doesn't look that nice anymore. :D

Inkjet Film works really well but isn't exactly cheap at ~1 EUR for a single sheet A4. And most of the time it takes 2 layers to avoid seeing the led. Have a look at the S/VM grid thing, somewhere there's a picture showing paper, inkjet film and canvas next to each other :)

Ah, here it is: https://cdn.thingiverse.com/assets/77/2c/bd/d4/17/007_diff.jpg

Top right is inkjet film. It diffuses very nicely without the "structure" you can see when using paper (bottom right). Also the colors look more vivid, but there's usually good tuning potential using color correction.

White material is almost useless as diffusive material. Even with 2-3 thin layers at 0.1mm you'll end up with less light coming through than through 2 sheets of regular copy paper.
Yes, the 5V version of the pro mini is used. Be aware there's some clones out there using the ATmega168 instead of the ATmega328P. They only have 1kb SRAM (instead of 2kb) so you might need to remove some parts of the code (serial outputs) to make it "fit".

I saw you mention that in previous comments, will pay attention to it.
Thanks a lot for the quick responses! I will order what I don't have and start printing

Have fun and remember to print these using the recommended extrusion width/line width (0.6mm) to get fast print times :)

Bei mir ist in der Datei Base_Frame_v1.stl ein Fehler. Egal was ich in Simpify3D einstelle.
Kann man im Screenshot gut sehen dass unten links und oben rechts Löcher in der Wand sind.

Ich kann mir nicht vorstellen dass diese Fehler nur bei mir zu sehen sind.

Määhääh... Antwort editiert, wurde als "flagged for moderation" markiert und anscheinend kann ich gerade gar nicht mehr im Thread antworten. Gleich noch einein dritten Versuch hinterher... xD


Oh, Raten. Klasse... ^^

  • Es geht nicht um die Uhr hier, oder? Ist das der Frame vom SE-Remix?

Der Rahmen der Remixes ist nämlich aus geschnittenen Stücken dieser Basisversion zusammengesetzt. Deshalb ist es wichtig zu wissen, von welchem File hier genau die Rede ist...

  • Wenn Simplify3D so einen Kram auswirft, ist vermutlich eine aktuelle Version (neuer als v4.0.0) am Start... kommt das hin?

Hab das Thema auch bei S3D im Bug Report-Forum. Versionen nach v4.0.0 verhalten sich teilweise sehr komisch bei Multipart-STLs. Betrifft nicht nur das spezielle File oder meine STLs, auch andere (mal z.B. den Multipart-Benchy hier von Thingiverse in v4 und v4.1.x vergleichen. Oder einfach 20mm Boxen exakt aneinander ausrichten und die Unterschiede bestaunen). Ich hab diverse Factory Files, die ich mit v4.1.x nicht mehr sauber gesliced bekomme. Sollte ich da eine Rückmeldung bekommen, schreib ich's hier dazu. Den letzten Bug allerdings habe ich zwar im Januar bestätigt bekommen, dann aber bis Ende März zur Veröffentlichung des Bugfixes in 4.1.2 nie wieder was von denen gehört...

Cura, Slic3r, ReplG und S3D bis einschl. v4.0.0 zeigen mit dem Baseframe der SE jedenfalls keine Auffälligkeiten.

Sorry, also es geht um die Version "Sources for Retro 7 Segment Clock SE (Small Edition) - http://www.thingiverse.com/thing:3095715 "
Ich benutze Simplify3D in der neuesten Version 4.1.2 und habe es davor mit Version 4.1.1 getestet und auch da das Problem mit den fehlenden Dreiecken.

Retro 7 Segment Clock SE (Small Edition)

Ja, beides Versionen neuer 4.0.0, die (nicht nur bei mir) einige Probleme bereiten. S3D behandelt die STLs (auch OBJs) seit v4.1.x anders, aber ob das genau so gewollt ist konnte und hat bis jetzt noch keiner vom Support beantwortet.

Anbei ein Factory File. Da sind die 18 Teile der Front einzeln drin, das bekommt S3D 4.1.x hin. Aber speichert man diese 18 STLs nun in eine einzelne Datei und lädt diese, kriegt S3D v4.1.x es schon nicht mehr hin. v4.0.0 und andere Slicer haben mit der exakt gleichen STL aber 0 Probleme.
Auch die Reparaturoptionen in S3D verhalten sich da anders. Während z.B. v4.0.0 das zusammengesetzt STL als zusammenhängende Fläche korrekt erkennt, versucht v4.1.x das Ding zu reparieren - und danach ist's noch schlimmer...

Ich kann aber nicht empfehlen, das so zu drucken. Denn die Rückseite/das Elektronikgehäuse wird ähnliche Probleme zeigen.

Aktuell kann ich selbst viele Sachen auch nur mit der 4.0.0 erfolgreich slicen. Ich habe testweise auch die STLs aus anderen Programmen exportiert, S3D ab v4.1.x zickt als einziges momentan rum. Egal ob Fusion, Inventor, Solidworks als Quelle etc.

Falls die Updates auf dem Rechner durchgeführt wurden, ist vermutlich noch die 4.0.0 installiert - neue Versionen installieren sich immer parallel in einen neuen Ordner mit Versionsnummer. Wenn die 4.0.0 noch da ist, bitte mal mit der testen. Da treten keine Probleme in der Art auf.

Leider kann man die alten Versionen bei S3D anscheinend nicht herunterladen. Wer also mal alte Sachen runter geworfen hat und/oder erst bei 4.1.1 gekauft hat, kommt da wohl nicht dran... :-/

Vielen Dank für den Tipp mit der älteren Version. Gleich mal schauen ob ich die noch finde.
Ansonsten Danke für Deine Mühe und die angehängten Dateien.

Nachtrag: Ich hab bei allen meinen Things die "problematischen STLs" neu hochgeladen, nachdem ich alle koplanaren Teile entfernt habe. Jetzt sollte kein Slicer damit ein Problem haben (auch netfabb etc. zeigen alles "grün"). Hab den Tipp von S3D bekommen, mal schauen wie die zukünftigen Versionen sich verhalten...

Gern. Ich hab gerade alle STLs meiner Things hier getestet. In S3D bis v4.0.0, Cura und Slic3r tauchen 0 Probleme auf. Ärgert mich, ich hab jahrelang neben S3D gar nix anderes mehr installiert gehabt und jetzt krieg ich damit nix mit mehr als 2 Komponenten hin... schräg... ich hoffe mal auf ein gutes Update in naher Zukunft. :)

Comments deleted.

I build one.
The led strip is very durable its hard to break, i did cut the 1st led shot and had to spice the strip,even this isn't too bad. i also shorted one and had to replaced one led at end of chain.

In the ardrino sketch it says 5v 1am at the start does this limit the led strip to a 1 amp draw? My 6seg clock seems to pull 0.8 amp or so.

clock can be a little hard to read close up in the light any body using a smoked glass filter on front?

This is insane:-)
unique use of addressable led strips thought these were standard 5050 stirips that didn't make sense
3 leds per segment 1 lost top to bottom so 22 per led 1 hidden led wont light?
3 unused linking left and right digit
42 lit 4 unlit?
5 in between


Almost. It's 47 LEDs per Module (2 digits):

3 per Segment (14) = 42
2 lost between top/bottom digit #0/#1 = 44
3 between digit #0/#1 = 47

I use those numbers inside the sketch to "address" the positions of each digit/module.

Those leds have gotten pretty cheap so I prefer "wasting" a few of them instead soldering 3 wires at each cut. Still thinking about a v3 of this clock where only 1 strip is used... xD

If you like the way it's build you might find the lazy grid clock interesting (1 strip, x/y 7x11 matrix) or the spiral/vase mode grid I uploaded recently.. :D

2 meters should do all 6 digits?
would there be any advantage to use ws2811 for 12 v leds?

2 meters = 120 leds. (3x 47 per module) + (2x 5) in between - 151 leds. So no, 2 meters won't be enough for the whole thing with 6 digits (XT). ;)
The sketch is using FastLED, so ws2811 would be supported. But it depends on what you want to achieve to notice any differences between ws2812b / ws2811. Most important thing is to watch strip dimensions (not thicker than ~1mm between leds, not wider than 10mm). APA102/SK6812 should work also but require to scale the parts along z to accomodate 12mm wide strips. If using 12v strips and feeding them directly like in the instructions make sure you're feeding the arduino using vin/raw, not +5V.

Is there a reference to what the strip numbers mean ? Will probably go with ws2811 just thought it might be possible to make it brighter with a higher voltage strip.

Higher voltage doesn't mean anything about resulting brightness. WS2811 is just a driver IC, while WS2812/B come in a single package, usually a 5050 SMD LED. APA102/SK6812 are chipsets/leds from other brands, WS in WS28xx stands for "World Semi".

But I doubt any of them will be noticably brighter. But if so they'll draw more voltage and/or current and you have to make sure you're using the right wire gauges. I wrote a lot about possible power requirements in the instructions already.

Unless you're trying to light the moon ws2812b can be pretty bright. Unless someone decides to print the diffusers in white PLA or anything, that is...

Using WS2812Bs and the default sketch with its power limit this thing is readable if lit directly by a 250W studio light. The reason I added ldr support was because even the lowest brightness is enough to slightly light a 30m² room at night... i wonder why people always want something to be brighter, without knowing how bright it already is... if you're looking for something to be readable in direct sunlight this is the wrong thing for you either way. You'd need a few Watts per LED and this design won't survive a power draw of 20 Watts per digit...

El cheapo 60x40 studio light vs. 7 segment clock with default sketch and ldr enabled at 1m distance... it's a cheap light but at that distance you wouldn't want to sit in front of it... ;)

7 seg vs light

Just completed mine tonight. Thanks for the excellent project!

I'm always done with this fantastic project. Really very good design!
But after running the test sketch with good result the other sketches wont compile because an memory error:

De schets gebruikt 14886 bytes (103%) programma-opslagruimte. Maximum is 14336 bytes.
Globale variabelen gebruiken 1143 bytes (111%) van het dynamisch geheugen. Resteren -119 bytes voor lokale variabelen. Maximum is 1024 bytes.

It's dutch but I think you can work it out. Can you please help me? I'm not good in coding Arduino's at all..

Thank you in advance!


Hi Eddy,

I remember someone having the exact same problem with another thing. Could it be you're using a Pro Mini clone with an ATmega168? The Pro Mini should have an 328P but the 168P ones come with only 1 KB of SRAM.


Try commenting out all "Serial.println" stuff you see. Should only be 2 or 3 lines and this should save enough memory to upload to 1kb.

Edit: Try commenting out the following lines:

105 - 107 (serial.begin and 2 lines with serial.print in void setup())
261 (entering setup mode in setupClock())
307 (setup done in SetupClock())

That should get you down to about 860 bytes memory usage.

You are right about everything! Works like a charm now! So happy with the result I will post my make tomorrow!

Al the best,


That's great to hear, have fun tinkering around with it :)

Someone mentioned an alternative to commenting stuff out here: https://www.thingiverse.com/thing:3241802/comments

Lazy Grid Clock

Was ein suuuuuper cooles Projekt!
Absolut Klasse!

Ich plane grad die Beschaffung der Teile die mir noch dafür fehlen.
Dabei kam in mir die Frage auf wo im Gehäuse der Fotowiderstand untergebracht wird?

Danke für das Teilen dieser Arbeit mit uns.
Ich weiß den Aufwand sehr zu schätzen.

Danke :)

Für den Fotowiderstand habe ich keine Halterung vorgesehen. Ich habe ihn einfach vor dem Löten seitlich durch eins der beiden Löcher für die Kabel zu den Modulen durchgeführt. Anbei ein Bild - inklusive ganz, ganz viel Staub ^^

Alternativ hab ich ihn auch schon unten beim USB-Kabel gehabt, bei einer Sache die etwas höher an der Wand hing. Da hat der LDR direkt dahinter nichts mehr mitbekommen... :)

Ah ok sowas in der Art hab ich mir schon gedacht. Ist ja aber auch legitim da man es eh nicht sieht :)

Very nice project and build. I'm having a blast too. One question though: Did you add a resistor to the LDR from the 5v or is it just a strait feed?

glad you like it. I added the LDR like can be seen in the arduino playground example. One side goes to +5V, the other
side goes to A1 and through a 10k resistor to ground, see attached picture.

red: +5v
orange: A1
black: ground

It's alive!!! Here is an update to your code. 1: changed behavior to ignore brightness button if autobrightness is set on (1). 2. Extended the palette to add red, green, white, and orange (though that looks yellow).

Thanks again for the design and code.

Comments deleted.

What about taking a look at "Thing Files"?

Built this clock.. full credit to you, great design and easy build. https://www.youtube.com/watch?v=32HIUN9NOVs

my problem that the Test.ino will work but the Main Program will not working.The i2c Scanner find the RTC.Have you a Idea what the problem?

which sketch version were you using? Does it output something on the serial console when starting up?
Or do the LEDs work and you're having problems with the rtc? Sorry, but it's hard to tell where the problem might be.
One thing I can think off: Is it possible you did wire the buttons wrong? They do have 4 pins and if you're using the wrong side to connect the wires, the sketch will act like if the buttons were pressed all the time.
This would make the clock enter setup and wait for the buttons to be released (which will never happen, basically they are shorted)... but you should see this on the serial console ("entering setup...").

How do you remove the power limitation?

Oh, that's a really complicated one, glad you're asking!

Look for the line which says "Limit power usage" within the sketch... -.-

Also (I don't know anything about writing code or else I would try our myself) could you make a countdown mode? I think that would be a cool feature. Maybe even switch between hours/minutes/seconds and days/hours/minutes

Sure I could. But why would I? If you're looking for lots of features there's quite a lot of clocks including webservers, bluetooth, wlan and so on...

Additionally the two buttons are already used for 5 functions. Adding this to the current clock would only complicate usage.

I'm just thinking of improvements. I like tinkering you know? And you could always add a third button.

Sure I could add a 3rd or 4th button. I could also add a rotary knob or some capacitive sensing. Same goes here... why would I?

I like tinkering around, too. That's why I've built that thing. And because I like supporting beginners who are interested in this I decided to not only give them the STL files. No, I decided to give them some detailed instructions, a fully working sketch and even commented most of that sketch so people could adjust it to their likings. Something to tinker around with.

The sketch explains pretty much how the leds get grouped to segments, how those segments are drawn and how to display numbers. It shows how to use the DS3231 module. So everything one needs to start adding own stuff is already there. Additiionally to all the great FastLED/neoPixel/WS2812-Guides, DS3231 samples/instructions out there. You don't want to learn how to add this on your own, you don't want to tinker with the sketch. That's fine, it's your choice. As it is your choice to remove a power limitiation without even knowing why and what it is good for (without other code changes you won't notice anything at all).

But my choice is to not put time into something only a single person has been asking for and I absolutely don't need for myself. Especially if this would mean hardware changes and redoing most of the instructions. But this topic has already been covered in the comments.

Alright. Sorry I asked. Just a suggestion.

Thanks for this great design! One thing, could you please put the source code on github, so others can work on and maybe even improve it

Hi neb1,

thanks! But I wonder what the source code on github would be good for!? There's nothing holding people back from downloading the INO and customize it to their likings. And I don't really see the need for "collaborative work" on a simple clock.

Speaking about source code: I've made some updates and will upload v3 of the INO sketch later. It has some features added and has an example of how to color the leds to your likings.

It's free, why not? :D

People could fix bugs, add features like different color themes/animations, add support to receive radio controlled time signals like DCF77 or as I had planned via GPS (should work even indoors).

With custom options using DEFINEs in the source code like e.g. Marlin does, everyone could customize it to their needs.

Maybe we can even switch to an ESP8266 (uses Arduino IDE too!) with Wifi connectivity and NTP time synchronization.

Just a note: If you're planning to use DCF77 you might have a look at the limitations when using Arduino/FastLED and interrupts! That can be really tricky at times: https://github.com/FastLED/FastLED/wiki/Interrupt-problems

I don't get it... it absolutely doesn't matter if someone downloaded the source code from this thing or github...?! I know about ESP8266s and so on - there's already dozens of projects out there for almost every combination.

If you wanted to add GPS or DCF77 functionality, fine. What prevents you from doing so after downloading the sketch from thingiverse? And if you're done with it and happy AND do think others might find this useful - what prevents you from doing a remix?

If somebody thinks he needs to improve/convert this for a ESP8266, great! Download the sketch, change it -> do a Remix. Done. Where's the need for github?

What do you mean with "custom options using DEFINEs"? Custom options can already be set using variables (brightness, color change interval, 12/24h, displaymode and so on)... declaring values using DEFINE/CONST/whatever doesn't tell if a user can change it without problems or not - that's relevant to the compiler/pre-processor. A few months ago I've built a plotter cutter and used marlin for this. I've downloaded it from github, did my modifications and uploaded it to the arduino. I see no difference if I had to download marlin from thingiverse instead. Just because it is on github doesn't mean anyone will work on that for free and add anything people can think about.

There's just absolutely no benefits I can see, maybe you can shed some light on this. The only thing I'd see happening:
Somebody changes something on github and people keep asking here why it doesn't work for them. Or why GPS/ESPxxx-version doesn't work for them. This thing is here to provide people with parts, instructions and a working sketch for the clock in a single place. Tell my why I should care about other peoples modifications to this maybe not working for some?!

"It's free, why not?" - Look at the license... you always can do a remix of this and put the source on github on your own. If your time is absolutely free and you want to maintain that, go ahead...

Yes I think that added functionality would benefit others and therefore should be shared. That's the spirit of open-source after all. It's the same thing in software as you would have it here with remixes.

Using pre-processor commands would tailor the resulting firmware for your specific device, when there are multiple individual features or hardware combinations. One example would be your extension parts to display seconds.

If it weren't for dozens of people that previously contributed to open-sourced Marlin, you wouldn't be able to use it for your plotter cutter. See, there is your benefit. :)

When my hardware arrives I'll have a look at the code and maybe start a repo. Thanks too, for your heads-up with the interrupts in the other reply. :)

  • lots and lots of stuff about general usage of tv/github -
  • aaand lots and lots more of stuff and rants about general usage of tv/github -

But we eventually managed this :)

Git vereinfach die Sourcecode-Verwaltung. Ja, vielleicht ist es etwas perfektionistisch gedacht, für ein Projekt dieser Größenordnung, aber whatever. Du MUSST es ja nicht machen, es war ja nur ein kleiner Vorschlag im Eingangspost. Kein Grund hier eine Riesen-Diskussion über Grundsätze etc zu starten. Aus mir nicht nachvollziehbaren Gründen werde ich das Gefühl nicht los, dass du das ein wenig zu persönlich nimmst...

Ah, gut. Auf deutsch geht das vielleicht etwas besser, manchmal steckt's ja in den Details. :D
Das Problem ist einfach: Ich will das Ding so einfach wie möglich halten und vor allem: Sicher sein, dass das was hier gedownloadet wird, auch funktioniert wie auf den Bildern zu sehen.

Wenn jemand einen Remix macht, dann wird das automatisch oben verlinkt und jeder kann die Remixes anschauen. Aber wenn ich es bei Github hochlade, muss ja immer noch einer den Maintainer machen, Änderungen prüfen, einpflegen und dafür sorgen, dass das auch zu der Anleitung hier passt. Und ehrlich: Ich wüsste nicht, warum ich das machen sollte. Ich hab die Uhr als Projekt aus Langeweile gebaut, das Ding hab ich Freitag Abend mit ´ner leeren virtuellen Seite Papier begonnen. Deswegen habe ich aber den Sketch dran gehängt: Damit eine Basis da ist, auf der wer auch immer aufbauen kann.
Das kann ja dann gern bei Github über ´nen Remix landen, wenn jemand die Basis für gut genug befindet und Zeit reinstecken will. Aber bloß weil jemand das machen mag, bin ich ja nicht verpflichtet, den Github zu maintanen und meine Anleitung hier anzupassen, oder? Alternativ lasse ich die Version hier auf einem festen Stand und starte parallel bei Github.
Ich verstehe den Einwand mit der Verwaltung, ja. Aber hier sehe ich einfach nicht, dass überhaupt mal aktive Entwickler kommen, die mehr als simple Anpassungen machen. Persönlich mag ich ja Open Source, ich finde es auch ätzend wie wenig Leute überhaupt was zurück liefern. Aber hier halte ich es einfach für "mit Kanonen auf Spatzen schießen", solange nicht abzusehen ist, dass da wirklich von mehr als 1-3 Personen Interesse besteht. Der Großteil liest doch nicht mal, die bauen das nach Bildern nach und kommen höchstens zum meckern wenn sie was falsch gemacht haben. ;))

Und nein, nix persönlich. Aber deine Antwort hat halt (für mich) jetzt das erste Mal erklärt, dass Du da halt schon recht optimistisch in die Zukunft blickst mit vielen Entwicklern. Und ich wollte halt Argumente, was es -jetzt- für einen Unterschied macht, wo der offene Quelltext zum Download liegt. :)

Ich verstehe, worauf du hinaus willst bzw. dass dir das Komplettpaket wichtig ist.

Warten wir einfach mal ab, wieviel Interessenten es so an dem Projekt geben wird. Dann kann man ja immer noch entsprechend loslegen mit entweder Remixes oder Git.

Schade, dass ich noch auf den LED-Strip warten muss. Arduinos, ESP8266, STM32 etc hätt ich ja immer genug herumliegen :D Hoffentlich bleibt mir die Motivation bis dahin erhalten, einige Ideen für Erweiterungen hätt ich ja schon im Kopf :D

Ein kurzer Einblick, ich komm ja eigentlich aus der Elektronik-Bastler Ecke. Das Hobby 3D-Druck hab ich mir anfangs nur angeeignet, um mir schöne oder zumindest nützliche Gehäuse für meine Projekte drucken kann. Wenn ich also so eine schöne 7 Segment Anzeige sehe, kannst du dir ja sicher vorstellen wie meine Bastler Augen aufleuchten voller Ideen xD

  • not relevant any long due to pm -
  • not relevant any long due to pm -

Thanks it is great when people list the components needed for a project with the description "Hats Off Well Done"

WOW - thanks for sharing such a complete, well documented project. :)