|
55906
|
Terminal Invaders by @TheDespicableJam | Stardance
|
TheDespicableJam
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Terminal Invaders by @TheDespicableJam | St Title: Terminal Invaders by @TheDespicableJam | Stardance...
|
1793865251.4340415
|
1794729251.4340415
|
|
55916
|
automatic sd card to usb copier by @Krzysztof | St automatic sd card to usb copier by @Krzysztof | Stardance...
|
Krzysztof
|
0
|
3.08
|
0
|
[]
|
[{"post_id": 58413, "ts": "202 [{"post_id": 58413, "ts": "2026-08-30T20:27:31Z", "hours": 1.9764, "text": "I wrote whole cpp code it was a pain in the butt lol"}, {"post_id": 58297, "ts": "2026-08-30T17:53:14Z", "hours": 1.1011, "text": "I hopped into kicad designed a schematic then putted everything on a pcb last step was routing the pcb (I flippin hate this step) :)"}]...
|
Title: automatic sd card to usb copier by @Krzyszt Title: automatic sd card to usb copier by @Krzysztof | Stardance
Devlogs:
I hopped into kicad designed a schematic then putted everything on a pcb last step was routing the pcb (I flippin hate this step) :)
I wrote whole cpp code it was a pain in the butt lol...
|
1793865253.1758494
|
1794729253.1758494
|
|
34518
|
Pomodoro Timer by @Gaby1000 | Stardance
|
Gaby1000
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Pomodoro Timer by @Gaby1000 | Stardance
|
1793865254.0406764
|
1794729254.0406764
|
|
47366
|
Gotcha v2 by @sterbolt | Stardance
|
sterbolt
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Gotcha v2 by @sterbolt | Stardance
|
1793865257.9164102
|
1794729257.9164102
|
|
55941
|
Untitled project by @kaboom | Stardance
|
kaboom
|
0
|
3.12
|
0
|
[]
|
[{"post_id": 59124, "ts": "202 [{"post_id": 59124, "ts": "2026-08-31T19:08:39Z", "hours": 3.1222, "text": "sfse"}]...
|
Title: Untitled project by @kaboom | Stardance
De Title: Untitled project by @kaboom | Stardance
Devlogs:
sfse...
|
1793865260.830171
|
1794729260.830171
|
|
7417
|
Untitled by @tecknet | Stardance
|
tecknet
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled by @tecknet | Stardance
|
1793865262.5494583
|
1794729262.5494583
|
|
36044
|
BSMacOS by @jackpolanczuk | Stardance
|
jackpolanczuk
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: BSMacOS by @jackpolanczuk | Stardance
|
1793865265.454562
|
1794729265.454562
|
|
19280
|
My awesome macropad! by @otiliamuntean48 | Stardan My awesome macropad! by @otiliamuntean48 | Stardance...
|
otiliamuntean48
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @otiliamuntean48 | Title: My awesome macropad! by @otiliamuntean48 | Stardance...
|
1793865268.3899925
|
1794729268.3899925
|
|
194
|
Untitled project by @anchith | Stardance
|
anchith
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @anchith | Stardance
|
1793865270.103069
|
1794729270.103069
|
|
39146
|
Asciistic.sh by @renanfartes | Stardance
|
renanfartes
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Asciistic.sh by @renanfartes | Stardance
|
1793865271.8304
|
1794729271.8304
|
|
14308
|
MacroShot by @miniaturity | Stardance
|
miniaturity
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: MacroShot by @miniaturity | Stardance
|
1793865274.714073
|
1794729274.714073
|
|
17701
|
My very own WebOS by @Nicholas-Reyne | Stardance
|
Nicholas-Reyne
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @Nicholas-Reyne | Star Title: My very own WebOS by @Nicholas-Reyne | Stardance...
|
1793865276.483922
|
1794729276.483922
|
|
25936
|
π¬πππ-ππππππ by @Uncool | Stardance
|
Uncool
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: π¬πππ-ππππππ by @Uncool | Stardance
|
1793865281.181236
|
1794729281.181236
|
|
52451
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865282.91452
|
1794729282.91452
|
|
52456
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865285.5331037
|
1794729285.5331037
|
|
52461
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865288.6807911
|
1794729288.6807911
|
|
52462
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865290.402529
|
1794729290.402529
|
|
52464
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865295.7162898
|
1794729295.7162898
|
|
52492
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865296.5455477
|
1794729296.5455477
|
|
52493
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865299.088918
|
1794729299.088918
|
|
22778
|
ARC ( Auto recon craft) by @kennator | Stardance
|
kennator
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: ARC ( Auto recon craft) by @kennator | Star Title: ARC ( Auto recon craft) by @kennator | Stardance...
|
1793865302.3360891
|
1794729302.3360891
|
|
55995
|
My awesome macropad! by @Bobfish6598 | Stardance
|
Bobfish6598
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Bobfish6598 | Star Title: My awesome macropad! by @Bobfish6598 | Stardance...
|
1793865304.0819232
|
1794729304.0819232
|
|
55999
|
bob by @Bobfish6598 | Stardance
|
Bobfish6598
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: bob by @Bobfish6598 | Stardance
|
1793865309.935981
|
1794729309.935981
|
|
2716
|
Untitled project by @ttt600161 | Stardance
|
ttt600161
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @ttt600161 | Stardance
|
1793865318.1463816
|
1794729318.1463816
|
|
29085
|
KARTX by @Eben-Siyabalapitiya | Stardance
|
Eben-Siyabalapitiya
|
0
|
8.78
|
0
|
[]
|
[{"post_id": 19630, "ts": "202 [{"post_id": 19630, "ts": "2026-07-01T00:27:14Z", "hours": 8.7792, "text": "DEVLOG 1 β Setting Up Vision Model Training & Project Structure βββββββββββββββββββββββ TRAINING Found a grocery dataset on Roboflow with 80k+ images across 17 categories. Set up training on Google Colab using a free T4 GPU with YOLOv8n, 50 epochs, image size 640, batch 16, output saving straight to Google Drive. Left it running overnight. Woke up to find it stopped at epoch 16. Colab hit the free GPU usage limit and killed the session mid run. Checked the weights folder in Drive and it looked empty at first, thought all progress was gone. Turns out Colab had been auto incrementing folder names each restart so the actual checkpoints were in kart2/weights/ the whole time. Both last.pt and best.pt were there. Opened results.csv and confirmed 15 full epochs completed with mAP50 around 0.87, precision and recall both near 0.99. Nothing lost. Training is paused at epoch 15/50 waiting for GPU quota to reset. Resume plan is loading last.pt with resume=True so it picks up exactly where it left off. βββββββββββββββββββββββ BUILDING THE APP While waiting on GPU access built out the full local pipeline. detector.py -runs model.track() each frame using ByteTrack andreturns detections with track IDs cart.py - handles a virtual tripwire line, items crossing down get added to the cart, crossing back up removes them, grace counter stops brief tracking losses from wiping items instantly prices.py - reads prices.json at startup and calculates subtotal plus Ontario HST 13% display.py - draws the tripwire line and bounding boxes onto the camera feed each frame main.py - ties everything together, Tkinter dark themed panel showing the live cart and total, OpenCV camera window showing the annotated feed, both running together ββββββββββββββββββββββ PROBLEMS Several source files were empty on disk, had to repaste everything. Stale pycache was loading a broken cached version of detector.py. Ultralytics and opencv were not installed locally. prices.json was empty and throwing a JSON error on startup. Wrong camera index was hitting the built in webcam instead of the external one. Local best.pt was a placeholder with nothing in it, temporarily swapped in yolov8n.pt to test the pipeline then pulled the real checkpoint down from Drive. βββββββββββββββββββββββ WHERE THINGS ARE AT Pipeline is fully working end to end. Camera feed shows the tripwire line and bounding boxes, items crossing get added to the cart panel with their price, subtotal and HST update live, total shows in green. Detection is decent for 15 epochs, right general categories but not perfectly precise yet which is expected. Once GPU access comes back finishing the remaining 35 epochs then swapping in the final best.pt."}]...
|
Title: KARTX by @Eben-Siyabalapitiya | Stardance
Title: KARTX by @Eben-Siyabalapitiya | Stardance
Devlogs:
DEVLOG 1 β Setting Up Vision Model Training & Project Structure βββββββββββββββββββββββ TRAINING Found a grocery dataset on Roboflow with 80k+ images across 17 categories. Set up training on Google Colab using a free T4 GPU with YOLOv8n, 50 epochs, image size 640, batch 16, output saving straight to Google Drive. Left it running overnight. Woke up to find it stopped at epoch 16. Colab hit the free GPU usage limit and killed the session mid run. Checked the weights folder in Drive and it looked empty at first, thought all progress was gone. Turns out Colab had been auto incrementing folder names each restart so the actual checkpoints were in kart2/weights/ the whole time. Both last.pt and best.pt were there. Opened results.csv and confirmed 15 full epochs completed with mAP50 around 0.87, precision and recall both near 0.99. Nothing lost. Training is paused at epoch 15/50 waiting for GPU quota to reset. Resume plan is loading last.pt with resume=True so it picks up exactly where it left off. βββββββββββββββββββββββ BUILDING THE APP While waiting on GPU access built out the full local pipeline. detector.py -runs model.track() each frame using ByteTrack andreturns detections with track IDs cart.py - handles a virtual tripwire line, items crossing down get added to the cart, crossing back up removes them, grace counter stops brief tracking losses from wiping items instantly prices.py - reads prices.json at startup and calculates subtotal plus Ontario HST 13% display.py - draws the tripwire line and bounding boxes onto the camera feed each frame main.py - ties everything together, Tkinter dark themed panel showing the live cart and total, OpenCV camera window showing the annotated feed, both running together ββββββββββββββββββββββ PROBLEMS Several source files were empty on disk, had to repaste everything. Stale pycache was loading a broken cached version of detector.py. Ultralytics and opencv were not installed locally. prices.json was empty and throwing a JSON error on startup. Wrong camera index was hitting the built in webcam instead of the external one. Local best.pt was a placeholder with nothing in it, temporarily swapped in yolov8n.pt to test the pipeline then pulled the real checkpoint down from Drive. βββββββββββββββββββββββ WHERE THINGS ARE AT Pipeline is fully working end to end. Camera feed shows the tripwire line and bounding boxes, items crossing get added to the cart panel with their price, subtotal and HST update live, total shows in green. Detection is decent for 15 epochs, right general categories but not perfectly precise yet which is expected. Once GPU access comes back finishing the remaining 35 epochs then swapping in the final best.pt....
|
1793865320.919575
|
1794729320.919575
|
|
1977
|
Purrfect Day by @juhi | Stardance
|
juhi
|
0
|
0.0
|
0
|
[]
|
[{"post_id": 343, "ts": "2026- [{"post_id": 343, "ts": "2026-06-01T17:24:51Z", "hours": 0.0, "text": "π"}]...
|
Title: Purrfect Day by @juhi | Stardance
Devlogs: Title: Purrfect Day by @juhi | Stardance
Devlogs:
π...
|
1793865322.7745385
|
1794729322.7745385
|
|
27854
|
My very own WebOS by @tengenrusei | Stardance
|
tengenrusei
|
0
|
0.0
|
0
|
[]
|
[{"post_id": 17907, "ts": "202 [{"post_id": 17907, "ts": "2026-06-28T20:51:41Z", "hours": 0.0, "text": "I honestly do not understand the tutorial but I was able to organize my files and place my os name and time but it has some errors :("}]...
|
Title: My very own WebOS by @tengenrusei | Stardan Title: My very own WebOS by @tengenrusei | Stardance
Devlogs:
I honestly do not understand the tutorial but I was able to organize my files and place my os name and time but it has some errors :(...
|
1793865326.3026793
|
1794729326.3026793
|
|
13511
|
OSO Keyboard by @vamshikrishnaramasamy | Stardance
|
vamshikrishnaramasamy
|
0
|
44.0
|
0
|
[]
|
[{"post_id": 13374, "ts": "202 [{"post_id": 13374, "ts": "2026-06-21T22:09:24Z", "hours": 25.0, "text": "HOLY I KNEW nOTHING aBoUt PCB desingngnnnnn. i made so many mistakes both on the pcb and on the cad. i had to go and fix those, it toookk ssooooooooooooo long πππππππππππππππππππππππππ"}, {"post_id": 10866, "ts": "2026-06-17T20:27:45Z", "hours": 1.0, "text": "i think it more or less done, weβll seeβ¦"}, {"post_id": 7640, "ts": "2026-06-12T19:22:04Z", "hours": 5.0, "text": "PCB is SO annoying. I keep running into problemsss"}, {"post_id": 6669, "ts": "2026-06-11T05:37:33Z", "hours": 6.0, "text": "Getting PCB ready to be printed!!!"}, {"post_id": 5998, "ts": "2026-06-10T01:57:39Z", "hours": 4.0, "text": "OOF. Huge session but extremely productive. I got a lot of work done on the circuitry side! I revamped it to use an ESP32 as the brains so that the entire thing would be more programmable and customizable!"}, {"post_id": 5401, "ts": "2026-06-09T04:44:35Z", "hours": 3.0, "text": "Just added my, ehm, special feature! more on that soonβ¦"}, {"post_id": 5268, "ts": "2026-06-08T23:49:42Z", "hours": 0.0, "text": "working on circuitry for first time (GPT helped)!"}, {"post_id": 5219, "ts": "2026-06-08T22:35:30Z", "hours": 0.0, "text": "Working an open source moddable kb!"}]...
|
Title: OSO Keyboard by @vamshikrishnaramasamy | St Title: OSO Keyboard by @vamshikrishnaramasamy | Stardance
Devlogs:
Working an open source moddable kb!
working on circuitry for first time (GPT helped)!
Just added my, ehm, special feature! more on that soonβ¦
OOF. Huge session but extremely productive. I got a lot of work done on the circuitry side! I revamped it to use an ESP32 as the brains so that the entire thing would be more programmable and customizable!
Getting PCB ready to be printed!!!
PCB is SO annoying. I keep running into problemsss
i think it more or less done, weβll seeβ¦
HOLY I KNEW nOTHING aBoUt PCB desingngnnnnn. i made so many mistakes both on the pcb and on the cad. i had to go and fix those, it toookk ssooooooooooooo long πππππππππππππππππππππππππ...
|
1793865327.157883
|
1794729327.157883
|
|
27878
|
Maestro by @yaasshh | Stardance
|
yaasshh
|
0
|
10.38
|
0
|
[]
|
[{"post_id": 20015, "ts": "202 [{"post_id": 20015, "ts": "2026-07-01T14:43:58Z", "hours": 10.3825, "text": "I have been working on a custom macro pad, which is based on the HackPad. I had an idea of making it usable with external apps through APIs and other integrations. I completed the PCB quite a while ago. I have put about 14 hours of total work into this, although I have only logged close to 10 hours. The other four hours were unlogged because I was working on a different laptop, but that is fine since it is only a few hours. The device itself is a macro pad with: Six MK-style switches One rotary encoder An OLED screen The PCB is quite impressive; I added some silkscreens to make it look aesthetically pleasing, which I hope will come across well in the finalized product. I designed the case in three parts: (a) The base (b) The plate (c) The top part that encloses everything The top part leaves space so that the screen can be viewed, making for a really nice, aesthetic-looking piece of hardware. I have decided to call it Maestro because it fits the theme I have going for it. It is going to be something I use in my daily life to shorten my work time and streamline my workflows. I am pretty excited for the funding to go out so I can start working on this."}]...
|
Title: Maestro by @yaasshh | Stardance
Devlogs:
I Title: Maestro by @yaasshh | Stardance
Devlogs:
I have been working on a custom macro pad, which is based on the HackPad. I had an idea of making it usable with external apps through APIs and other integrations. I completed the PCB quite a while ago. I have put about 14 hours of total work into this, although I have only logged close to 10 hours. The other four hours were unlogged because I was working on a different laptop, but that is fine since it is only a few hours. The device itself is a macro pad with: Six MK-style switches One rotary encoder An OLED screen The PCB is quite impressive; I added some silkscreens to make it look aesthetically pleasing, which I hope will come across well in the finalized product. I designed the case in three parts: (a) The base (b) The plate (c) The top part that encloses everything The top part leaves space so that the screen can be viewed, making for a really nice, aesthetic-looking piece of hardware. I have decided to call it Maestro because it fits the theme I have going for it. It is going to be something I use in my daily life to shorten my work time and streamline my workflows. I am pretty excited for the funding to go out so I can start working on this....
|
1793865328.9021776
|
1794729328.9021776
|
|
44093
|
WORK BOARD by @juhi | Stardance
|
juhi
|
0
|
17.0
|
0
|
[]
|
[{"post_id": 41142, "ts": "202 [{"post_id": 41142, "ts": "2026-08-05T05:02:26Z", "hours": 17.0, "text": "A simple productivity dashboard that brings together everything you need while studying or working"}]...
|
Title: WORK BOARD by @juhi | Stardance
Devlogs:
A Title: WORK BOARD by @juhi | Stardance
Devlogs:
A simple productivity dashboard that brings together everything you need while studying or working...
|
1793865330.6590016
|
1794729330.6590016
|
|
52509
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865332.4049087
|
1794729332.4049087
|
|
45914
|
AURA Core by @vishruthathreyas | Stardance
|
vishruthathreyas
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: AURA Core by @vishruthathreyas | Stardance
|
1793865334.153093
|
1794729334.153093
|
|
52544
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865337.05176
|
1794729337.05176
|
|
44097
|
My very own WebOS by @Tanojimo | Stardance
|
Tanojimo
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @Tanojimo | Stardance
|
1793865339.9122567
|
1794729339.9122567
|
|
52549
|
NULL
|
NULL
|
0
|
0.0
|
0
|
[]
|
[]
|
|
1793865341.6900816
|
1794729341.6900816
|
|
50772
|
Something something by @CassianWray | Stardance
|
CassianWray
|
0
|
9.58
|
0
|
[]
|
[{"post_id": 51179, "ts": "202 [{"post_id": 51179, "ts": "2026-08-20T18:00:06Z", "hours": 1.95, "text": "I finally completed the case. I am so happy with this even though its not that good looking. Designing the case probably took me the longest out of all. Not pretty at all and I know it looks basic af but I created it and I am happy that I did."}, {"post_id": 50527, "ts": "2026-08-19T18:29:39Z", "hours": 3.45, "text": "Finally Completed the PCB design properly. I started with the case design and I am completely lost and donβt know what to do. Like I cannot even use a designing software properly."}, {"post_id": 49780, "ts": "2026-08-18T18:23:56Z", "hours": 4.1836, "text": "I worked on the designing of the keyboard. My plan is a 12 key keyboard with the essential keys required for gaming so I can use it for gaming myself. Anyway I have been at it almost the entire day but most of the time has been spent trying to figure out how to operate the kicad, watching youtube tutorial and reading documents so only 4 hr was logged. Anyway I have completed the pcb design but I donβt know how its getting some warnings. This is my first time making a hardware project so I donβt even know what is happening. Anyone who knows hardware please give me some feedback."}]...
|
Title: Something something by @CassianWray | Stard Title: Something something by @CassianWray | Stardance
Devlogs:
I worked on the designing of the keyboard. My plan is a 12 key keyboard with the essential keys required for gaming so I can use it for gaming myself. Anyway I have been at it almost the entire day but most of the time has been spent trying to figure out how to operate the kicad, watching youtube tutorial and reading documents so only 4 hr was logged. Anyway I have completed the pcb design but I donβt know how its getting some warnings. This is my first time making a hardware project so I donβt even know what is happening. Anyone who knows hardware please give me some feedback.
Finally Completed the PCB design properly. I started with the case design and I am completely lost and donβt know what to do. Like I cannot even use a designing software properly.
I finally completed the case. I am so happy with this even though its not that good looking. Designing the case probably took me the longest out of all. Not pretty at all and I know it looks basic af but I created it and I am happy that I did....
|
1793865343.7813513
|
1794729343.7813513
|
|
42594
|
CMHP (Clean Minimalistic Home Page) by @Tuguldur | CMHP (Clean Minimalistic Home Page) by @Tuguldur | Stardance...
|
Tuguldur
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: CMHP (Clean Minimalistic Home Page) by @Tug Title: CMHP (Clean Minimalistic Home Page) by @Tuguldur | Stardance...
|
1793865345.5174341
|
1794729345.5174341
|
|
56011
|
asd by @RespectableDot | Stardance
|
RespectableDot
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: asd by @RespectableDot | Stardance
|
1793865346.359065
|
1794729346.359065
|
|
56018
|
bob by @Bobfish6598 | Stardance
|
Bobfish6598
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: bob by @Bobfish6598 | Stardance
|
1793865350.7103517
|
1794729350.7103517
|
|
56019
|
bob by @Bobfish6598 | Stardance
|
Bobfish6598
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: bob by @Bobfish6598 | Stardance
|
1793865352.5179374
|
1794729352.5179374
|
|
39943
|
COSMOCRYPT VT-80 by @NITHIN | Stardance
|
NITHIN
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: COSMOCRYPT VT-80 by @NITHIN | Stardance
|
1793865355.4333534
|
1794729355.4333534
|
|
56027
|
Clock by @EhtishamCodes | Stardance
|
EhtishamCodes
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Clock by @EhtishamCodes | Stardance
|
1793865357.1489527
|
1794729357.1489527
|
|
56030
|
My own site! by @imraghav012 | Stardance
|
imraghav012
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My own site! by @imraghav012 | Stardance
|
1793865360.131312
|
1794729360.131312
|
|
35563
|
PLEASE DELETE by @Lokpriyanth | Stardance
|
Lokpriyanth
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: PLEASE DELETE by @Lokpriyanth | Stardance
|
1793865363.0304294
|
1794729363.0304294
|
|
56036
|
Anims-simulation by @Freaked | Stardance
|
Freaked
|
0
|
2.71
|
0
|
[]
|
[{"post_id": 59105, "ts": "202 [{"post_id": 59105, "ts": "2026-08-31T18:51:10Z", "hours": 2.7089, "text": "#Devlog 1 : I am working of a cool 3D black hole simulator , it was really time taking but i finally got how my simulator will look and i am thinking of first making a net (space fabric) to show accuracy of black hole and its massive gravity i think i should use some cool animation to make it more attractive. what you guys think what will be best."}]...
|
Title: Anims-simulation by @Freaked | Stardance
D Title: Anims-simulation by @Freaked | Stardance
Devlogs:
#Devlog 1 : I am working of a cool 3D black hole simulator , it was really time taking but i finally got how my simulator will look and i am thinking of first making a net (space fabric) to show accuracy of black hole and its massive gravity i think i should use some cool animation to make it more attractive. what you guys think what will be best....
|
1793865364.7556484
|
1794729364.7556484
|
|
39060
|
PomodBros by @belbel | Stardance
|
belbel
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: PomodBros by @belbel | Stardance
|
1793865366.4671743
|
1794729366.4671743
|
|
4707
|
Ai by @AnantPrakash | Stardance
|
AnantPrakash
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Ai by @AnantPrakash | Stardance
|
1793865368.5492213
|
1794729368.5492213
|
|
36726
|
Demo#1 by @ritam | Stardance
|
ritam
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Demo#1 by @ritam | Stardance
|
1793865370.28479
|
1794729370.28479
|
|
44103
|
A website π€©π by @nourzain160 | Stardance
|
nourzain160
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: A website π€©π by @nourzain160 | Stardance
|
1793865371.9929492
|
1794729371.9929492
|
|
38690
|
Pipasso | An Auto Drawing Bot by @Dragon | Stardan Pipasso | An Auto Drawing Bot by @Dragon | Stardance...
|
Dragon
|
0
|
40.9
|
0
|
[]
|
[{"post_id": 48266, "ts": "202 [{"post_id": 48266, "ts": "2026-08-16T15:55:07Z", "hours": 1.8519, "text": "Devlog 23 The new paths looked good in code, but moving a real mouse along pixel steps can still look rough. I spent time checking that the automation engine could follow those close-together points safely. What I did Updated the mouse engine to expand longer point-to-point moves into safe drag steps. Checked pause, stop, and mouse-button release behaviour with continuous paths. Updated the draw-style controls and help text so Continuous is clearly the normal choice. Kept the live preview and placement ghost matched to the same processed mask. Current result Continuous paths are ready for real cursor automation. The output stays faithful to the source mask instead of drawing through empty areas."}, {"post_id": 48178, "ts": "2026-08-16T13:59:46Z", "hours": 2.0469, "text": "Devlog 22 So I was thinking: if every dark pixel is part of a little map, the cursor should be able to walk through that map without taking shortcuts through the white background. That became the new continuous drawing style. What I did Added a Continuous draw pattern and made it the default. Grouped ink pixels into connected regions. Used a depth-first walk to make long, pixel-adjacent pen paths. Let paths backtrack over already-drawn pixels instead of lifting the pen or crossing blank space. Kept larger connected shapes first so the drawing starts with the main details. Current result The bot can draw connected regions as unbroken, human-like strokes. There are no skipped scan lines in Continuous mode. The original fill patterns are still available for shading-style drawings."}, {"post_id": 46942, "ts": "2026-08-14T14:43:47Z", "hours": 1.6844, "text": "Devlog 21 I watched the bot draw and finally noticed why it looked a bit like a printer instead of a person: it was doing separate scan lines with spaces between them. It was technically drawing the right shapes, but the movement did not feel right. What I did Looked at how the fill paths were being created. Kept the older horizontal, vertical, and random scanline styles as optional faster modes. Planned a new default that would stay on real image pixels instead of jumping between rows. Current result I understood that the problem was the path algorithm, not the mouse speed. The next task was to make a proper continuous tracer."}, {"post_id": 46421, "ts": "2026-08-13T16:45:27Z", "hours": 1.9192, "text": "Devlog 20 The final part was testing with different pictures and fixing the little problems that only show up when you actually use the program. Then I wrote the README so future-me knows what all of this does. What I did Tested line-art, threshold, and colour-group paths with sample images. Checked placement, cancellation, stop behaviour, and cursor release. Added installation, usage, safety, and limitation notes to the README. Listed dependencies in requirements.txt . Current result A working AutoDraw desktop project with documented setup and safety guidance. Plenty of future ideas, but a complete usable version now."}, {"post_id": 45167, "ts": "2026-08-11T14:14:03Z", "hours": 1.5628, "text": "Devlog 19 I realised most people would double-click the app, not run commands. A console window popping up behind it made it feel unfinished. What I did Added main.pyw for no-console Windows launching. Guarded console output for environments without stdout/stderr. Checked the project can also be packaged with PyInstaller. Current result The project is closer to something a normal user can open directly."}, {"post_id": 44683, "ts": "2026-08-10T16:40:31Z", "hours": 1.7878, "text": "Devlog 18 By now it worked, but the controls felt scattered. I spent this session making it feel more like a small Windows utility than a school experiment. What I did Organised controls into Settings, Thresholding, and Action sections. Added labels for interval, click delay, luminance, and draw pattern. Improved the image and threshold previews. Current result The main workflow is easier to understand at a glance."}, {"post_id": 42588, "ts": "2026-08-07T12:32:51Z", "hours": 2.4092, "text": "Devlog 17 The bot canβt magically swap the pen colour in every app, so I made the program honest about that limitation instead of pretending it was solved. What I did Added automatic pauses between colour groups. Made an always-on-top colour notice with an RGB swatch. Added continue and stop controls for each layer. Current result Multi-colour mode guides the user through manual pen changes."}, {"post_id": 41977, "ts": "2026-08-06T14:48:48Z", "hours": 1.7783, "text": "Devlog 16 Then I wondered if it could do more than black lines. I experimented with grouping similar colours, even though it makes the workflow more complicated. What I did Added k-means colour clustering. Ordered groups from dark to light. Generated fill paths for each colour group. Current result The processing engine can split an image into separate colour layers."}, {"post_id": 41194, "ts": "2026-08-05T08:08:30Z", "hours": 1.5864, "text": "Devlog 15 I noticed some images looked better depending on how they were filled. That became a good excuse to add different drawing patterns. What I did Added horizontal scanline filling. Reused the same logic for vertical strokes. Added a random mode that shuffles and sometimes reverses strokes. Current result The user can choose how the drawing progresses without changing the final ink mask."}, {"post_id": 41172, "ts": "2026-08-05T06:32:42Z", "hours": 1.8778, "text": "Devlog 14 Drawing individual pixels was obviously not going to be practical. I changed the mask into longer strokes, one run of pixels at a time. What I did Found continuous runs of ink pixels in each scanline. Converted each run into one two-point stroke. Used zig-zag ordering on alternate rows to reduce travel. Added a fill-spacing setting. Current result Threshold drawings are much faster and smoother than pixel-by-pixel drawing. [!NOTE] The image shows how it used to draw and now how it draws, the drawing is still kind of wobbly, will keep adjusting till its flawless"}, {"post_id": 40841, "ts": "2026-08-04T16:59:46Z", "hours": 2.3481, "text": "Devlog 13 Edge detection looked cool, but it sometimes missed things or made weird extra details. I tried a simpler idea: draw only pixels within a brightness range. What I did Added min/max luminance settings. Created a black-and-white ink mask from the selected range. Made a live threshold preview in the GUI. Current result Users can choose which brightness levels become ink. The result is easier to tune for simple images."}, {"post_id": 39993, "ts": "2026-08-03T10:48:42Z", "hours": 1.81, "text": "Devlog 12 I wanted a way to know whether the drawing would land in the right place before the mouse started moving. So I made a little ghost version follow the cursor. What I did Captured a screenshot behind the placement overlay. Displayed a transparent processed-image preview under the cursor. Added an outlined bounding box to make the size easier to notice. Current result Placement is visual instead of a blind click."}, {"post_id": 38252, "ts": "2026-07-31T15:34:48Z", "hours": 1.7439, "text": "Devlog 11 At first I was typing coordinates, which was annoying and easy to get wrong. A click-to-place workflow felt much more natural. What I did Added a full-screen crosshair positioning window. Captured the next left-click as the drawingβs top-left origin. Added Escape cancellation before drawing begins. Current result The user can point at a canvas and place the drawing directly."}, {"post_id": 37767, "ts": "2026-07-30T18:01:59Z", "hours": 1.6058, "text": "Devlog 10 I really did not want a bug to leave the mouse holding down forever. This session was mostly about making failure less stressful. What I did Added stop, pause, and resume state handling. Ensured the mouse button is released on completion, stop, or error. Enabled PyAutoGUIβs corner fail-safe. Started wiring an Esc-key emergency stop. Current result The bot has multiple ways to stop safely. Pausing mid-stroke no longer risks a stuck drag."}, {"post_id": 37157, "ts": "2026-07-29T17:33:23Z", "hours": 3.7783, "text": "Devlog 08 and 09 Devlog 08 This was the slightly scary part: allowing the project to control the real cursor. I tested on a safe blank drawing area, with my hand ready to stop it. What I did Added PyAutoGUI mouse movement and mouse-down/mouse-up actions. Translated local canvas points into screen coordinates. Added adjustable movement, click, and gap delays. Current result A simple path can be drawn using real mouse input. Timing is configurable instead of hard-coded. Devlog 09 I clicked draw and the whole window froze while the mouse worked. It wasnβt technically broken, but it felt broken. That sent me down the threading rabbit hole. What I did Moved drawing work to a background thread. Added start, progress, and finished callbacks. Kept GUI actions on the interface side of the program. Current result The interface stays responsive while automation is running."}, {"post_id": 36845, "ts": "2026-07-29T05:06:14Z", "hours": 1.7694, "text": "Devlog 07 My first path order was basically random, and the cursor would bounce around like it had too much coffee. I made a simple nearest-neighbour ordering system to calm it down. What I did Measured distance between path ends. Ordered contours by the closest next path. Reversed paths when approaching their end was shorter. Current result The cursor should spend less time travelling between separate strokes. Line drawings feel more intentional."}, {"post_id": 35858, "ts": "2026-07-27T15:50:26Z", "hours": 1.9178, "text": "Devlog 06 Edges are still just pixels, though. I had to figure out how to make them into actual strokes that a mouse could follow without constantly clicking every single point. What I did Found contours from the edge image. Converted contours into lists of (x, y) points. Simplified overly detailed paths and ignored tiny unusable contours. Added PathGroup so related strokes can travel together. Patched some parts of image rendering issue mentioned in the 5th devlog , so it at least works. Current result The processor now produces drawable paths, not only a black-and-white image."}, {"post_id": 35536, "ts": "2026-07-27T01:16:32Z", "hours": 2.2169, "text": "Devlog 04 and 05 Devlog 04 I added an image preview, then noticed a problem: the original picture and the drawing area could have different shapes. A wide image was getting squished, which looked terrible. What I did Added a preview area to the interface. Resized images while keeping their aspect ratio. Letterboxed them on a white canvas so the processing area stays consistent. Current result Images fit the drawing canvas without stretching. The preview is much closer to what the bot will use. Devlog 05 So I was thinking that drawing every pixel would take forever. I tried edge detection instead, because an outline is often enough to recognise a picture. What I did Added OpenCV and converted images to grayscale. Used blur plus Canny edge detection to find visible outlines. Saved the generated edge image for previewing and debugging. Had a lot of problems with the back end when converting PNG / SVG Current result The app can turn an image into line-art data. Some rendering issues when using a PNG or a SVG I can see why some pictures work better than others. (1st Image is the gray scale version of , the 2nd image)"}, {"post_id": 34685, "ts": "2026-07-25T16:29:03Z", "hours": 1.9825, "text": "Devlog 03 The next question was, βCan it even open the kind of images I use?β I used Pillow because it made loading and converting images much less scary. What I did Added a file picker for image files. Loaded images with Pillow and converted them to RGB. Put the image-loading work into image_processor.py instead of mixing it into the GUI. Current result Users can choose an image from their computer. The program has a separate place for image-related code."}, {"post_id": 34055, "ts": "2026-07-24T14:45:34Z", "hours": 1.7711, "text": "Devlog 02 I started with the least exciting but most reassuring thing: making an actual window appear. Seeing a blank app on screen made the project feel real. What I did Set up main.py to start the program. Created a basic Tkinter app shell. Added a title, a simple layout, and a place for image controls. Current result The app launches without needing a terminal-focused workflow. There is a foundation to build the interface onto."}, {"post_id": 33523, "ts": "2026-07-23T15:29:11Z", "hours": 1.4511, "text": "Devlog 01 β The idea So I was thinking: what if I could give a program a picture and have it draw it with my mouse? It sounded simple at first, but then I realised the computer needs a list of tiny movements, not just an image. What I did Defined the goal: load an image, turn it into strokes, and reproduce them on screen. Chose Python because I already knew enough to experiment quickly. Picked Tkinter for the desktop window and wrote down the main parts the project would need. Current result A clear plan instead of just βmake a drawing bot.β The first rough project folder and entry-point idea."}]...
|
Title: Pipasso | An Auto Drawing Bot by @Dragon | Title: Pipasso | An Auto Drawing Bot by @Dragon | Stardance
Devlogs:
Devlog 01 β The idea So I was thinking: what if I could give a program a picture and have it draw it with my mouse? It sounded simple at first, but then I realised the computer needs a list of tiny movements, not just an image. What I did Defined the goal: load an image, turn it into strokes, and reproduce them on screen. Chose Python because I already knew enough to experiment quickly. Picked Tkinter for the desktop window and wrote down the main parts the project would need. Current result A clear plan instead of just βmake a drawing bot.β The first rough project folder and entry-point idea.
Devlog 02 I started with the least exciting but most reassuring thing: making an actual window appear. Seeing a blank app on screen made the project feel real. What I did Set up main.py to start the program. Created a basic Tkinter app shell. Added a title, a simple layout, and a place for image controls. Current result The app launches without needing a terminal-focused workflow. There is a foundation to build the interface onto.
Devlog 03 The next question was, βCan it even open the kind of images I use?β I used Pillow because it made loading and converting images much less scary. What I did Added a file picker for image files. Loaded images with Pillow and converted them to RGB. Put the image-loading work into image_processor.py instead of mixing it into the GUI. Current result Users can choose an image from their computer. The program has a separate place for image-related code.
Devlog 04 and 05 Devlog 04 I added an image preview, then noticed a problem: the original picture and the drawing area could have different shapes. A wide image was getting squished, which looked terrible. What I did Added a preview area to the interface. Resized images while keeping their aspect ratio. Letterboxed them on a white canvas so the processing area stays consistent. Current result Images fit the drawing canvas without stretching. The preview is much closer to what the bot will use. Devlog 05 So I was thinking that drawing every pixel would take forever. I tried edge detection instead, because an outline is often enough to recognise a picture. What I did Added OpenCV and converted images to grayscale. Used blur plus Canny edge detection to find visible outlines. Saved the generated edge image for previewing and debugging. Had a lot of problems with the back end when converting PNG / SVG Current result The app can turn an image into line-art data. Some rendering issues when using a PNG or a SVG I can see why some pictures work better than others. (1st Image is the gray scale version of , the 2nd image)
Devlog 06 Edges are still just pixels, though. I had to figure out how to make them into actual strokes that a mouse could follow without constantly clicking every single point. What I did Found contours from the edge image. Converted contours into lists of (x, y) points. Simplified overly detailed paths and ignored tiny unusable contours. Added PathGroup so related strokes can travel together. Patched some parts of image rendering issue mentioned in the 5th devlog , so it at least works. Current result The processor now produces drawable paths, not only a black-and-white image.
Devlog 07 My first path order was basically random, and the cursor would bounce around like it had too much coffee. I made a simple nearest-neighbour ordering system to calm it down. What I did Measured distance between path ends. Ordered contours by the closest next path. Reversed paths when approaching their end was shorter. Current result The cursor should spend less time travelling between separate strokes. Line drawings feel more intentional.
Devlog 08 and 09 Devlog 08 This was the slightly scary part: allowing the project to control the real cursor. I tested on a safe blank drawing area, with my hand ready to stop it. What I did Added PyAutoGUI mouse movement and mouse-down/mouse-up actions. Translated local canvas points into screen coordinates. Added adjustable movement, click, and gap delays. Current result A simple path can be drawn using real mouse input. Timing is configurable instead of hard-coded. Devlog 09 I clicked draw and the whole window froze while the mouse worked. It wasnβt technically broken, but it felt broken. That sent me down the threading rabbit hole. What I did Moved drawing work to a background thread. Added start, progress, and finished callbacks. Kept GUI actions on the interface side of the program. Current result The interface stays responsive while automation is running.
Devlog 10 I really did not want a bug to leave the mouse holding down forever. This session was mostly about making failure less stressful. What I did Added stop, pause, and resume state handling. Ensured the mouse button is released on completion, stop, or error. Enabled PyAutoGUIβs corner fail-safe. Started wiring an Esc-key emergency stop. Current result The bot has multiple ways to stop safely. Pausing mid-stroke no longer risks a stuck drag.
Devlog 11 At first I was typing coordinates, which was annoying and easy to get wrong. A click-to-place workflow felt much more natural. What I did Added a full-screen crosshair positioning window. Captured the next left-click as the drawingβs top-left origin. Added Escape cancellation before drawing begins. Current result The user can point at a canvas and place the drawing directly.
Devlog 12 I wanted a way to know whether the drawing would land in the right place before the mouse started moving. So I made a little ghost version follow the cursor. What I did Captured a screenshot behind the placement overlay. Displayed a transparent processed-image preview under the cursor. Added an outlined bounding box to make the size easier to notice. Current result Placement is visual instead of a blind click.
Devlog 13 Edge detection looked cool, but it sometimes missed things or made weird extra details. I tried a simpler idea: draw only pixels within a brightness range. What I did Added min/max luminance settings. Created a black-and-white ink mask from the selected range. Made a live threshold preview in the GUI. Current result Users can choose which brightness levels become ink. The result is easier to tune for simple images.
Devlog 14 Drawing individual pixels was obviously not going to be practical. I changed the mask into longer strokes, one run of pixels at a time. What I did Found continuous runs of ink pixels in each scanline. Converted each run into one two-point stroke. Used zig-zag ordering on alternate rows to reduce travel. Added a fill-spacing setting. Current result Threshold drawings are much faster and smoother than pixel-by-pixel drawing. [!NOTE] The image shows how it used to draw and now how it draws, the drawing is still kind of wobbly, will keep adjusting till its flawless
Devlog 15 I noticed some images looked better depending on how they were filled. That became a good excuse to add different drawing patterns. What I did Added horizontal scanline filling. Reused the same logic for vertical strokes. Added a random mode that shuffles and sometimes reverses strokes. Current result The user can choose how the drawing progresses without changing the final ink mask.
Devlog 16 Then I wondered if it could do more than black lines. I experimented with grouping similar colours, even though it makes the workflow more complicated. What I did Added k-means colour clustering. Ordered groups from dark to light. Generated fill paths for each colour group. Current result The processing engine can split an image into separate colour layers.
Devlog 17 The bot canβt magically swap the pen colour in every app, so I made the program honest about that limitation instead of pretending it was solved. What I did Added automatic pauses between colour groups. Made an always-on-top colour notice with an RGB swatch. Added continue and stop controls for each layer. Current result Multi-colour mode guides the user through manual pen changes.
Devlog 18 By now it worked, but the controls felt scattered. I spent this session making it feel more like a small Windows utility than a school experiment. What I did Organised controls into Settings, Thresholding, and Action sections. Added labels for interval, click delay, luminance, and draw pattern. Improved the image and threshold previews. Current result The main workflow is easier to understand at a glance.
Devlog 19 I realised most people would double-click the app, not run commands. A console window popping up behind it made it feel unfinished. What I did Added main.pyw for no-console Windows launching. Guarded console output for environments without stdout/stderr. Checked the project can also be packaged with PyInstaller. Current result The project is closer to something a normal user can open directly.
Devlog 20 The final part was testing with different pictures and fixing the little problems that only show up when you actually use the program. Then I wrote the README so future-me knows what all of this does. What I did Tested line-art, threshold, and colour-group paths with sample images. Checked placement, cancellation, stop behaviour, and cursor release. Added installation, usage, safety, and limitation notes to the README. Listed dependencies in requirements.txt . Current result A working AutoDraw desktop project with documented setup and safety guidance. Plenty of future ideas, but a complete usable version now.
Devlog 21 I watched the bot draw and finally noticed why it looked a bit like a printer instead of a person: it was doing separate scan lines with spaces between them. It was technically drawing the right shapes, but the movement did not feel right. What I did Looked at how the fill paths were being created. Kept the older horizontal, vertical, and random scanline styles as optional faster modes. Planned a new default that would stay on real image pixels instead of jumping between rows. Current result I understood that the problem was the path algorithm, not the mouse speed. The next task was to make a proper continuous tracer.
Devlog 22 So I was thinking: if every dark pixel is part of a little map, the cursor should be able to walk through that map without taking shortcuts through the white background. That became the new continuous drawing style. What I did Added a Continuous draw pattern and made it the default. Grouped ink pixels into connected regions. Used a depth-first walk to make long, pixel-adjacent pen paths. Let paths backtrack over already-drawn pixels instead of lifting the pen or crossing blank space. Kept larger connected shapes first so the drawing starts with the main details. Current result The bot can draw connected regions as unbroken, human-like strokes. There are no skipped scan lines in Continuous mode. The original fill patterns are still available for shading-style drawings.
Devlog 23 The new paths looked good in code, but moving a real mouse along pixel steps can still look rough. I spent time checking that the automation engine could follow those close-together points safely. What I did Updated the mouse engine to expand longer point-to-point moves into safe drag steps. Checked pause, stop, and mouse-button release behaviour with continuous paths. Updated the draw-style controls and help text so Continuous is clearly the normal choice. Kept the live preview and placement ghost matched to the same processed mask. Current result Continuous paths are ready for real cursor automation. The output stays faithful to the source mask instead of drawing through empty areas....
|
1793865374.8985126
|
1794729374.8985126
|