Loss of synchronization in SC2-120 mode

[IU4ALH]

1 )- When receiving in SC2-120 mode and decoding has reached about half the screen, then synchronization is lost. Sync is restored after receiving a few lines, but with obvious image distortion. See photo attached.
Is it possible to fix this?
This problem happens with all the images I received for testing and only in the aforementioned SC2-120 mode.

2 )- Would it also be possible to add new modes that are very popular among us radio amateurs, such as ROBOT 24/36/72 and Scottie DX?

Greetings by IU4ALH

[IU4ALH]

Hello.
I wanted to let you know that the vast majority of radio amateurs who transmit in SSTV use their abbreviation in the bottom right corner of the image they transmit.
For this reason, I've moved the text indicating both the mode in use and the numerical line count from its original position in the bottom right to a more convenient location in the bottom left (I'm attaching an image I received).
Do you think it would be appropriate to make this very small change to the firmware you published on the site?
Furthermore, in relation to our previous discussion, I confirm the real need for ROBOT modes, so I'm eager to see your updated firmware.
Congratulations to you and greetings with my sincere admiration.
IU4ALH Antonino

[dawsonjon]

Hi, Thanks that's very neat! If you wanted to submit a pull-request, or share the change I would be happy to include it. There are more updates on the way including transmit capability and SD card storage. When I get a little time I will see if I can add the robot modes too. Thanks very much 👍

[IU4ALH]

Hi. The modification is very simple; just adjust the parameters of the two lines that print the black background rectangle and the text with the modes and the row count. In practice, these two lines: line 185 and line 225 ..... 185 display->fillRect(320-(21*6)-2, 240-10, 10, 21*6+2, COLOUR_BLACK); //BLACK RECTANGULAR BACKGROUND at the bottom right with this :- 185 display->fillRect(128-(21*6)-2, 240-10, 10, 21*6-3, COLOUR_BLACK); //BLACK RECTANGULAR BACKGROUND at the bottom left and line 225 display->drawString(320-(21*6), 240-8, font_8x5, buffer, COLOUR_WHITE, COLOUR_BLACK); // POSITION WRITTEN MODE BOTTOM RIGHT with this :- 225 display->drawString(130-(21*6), 240-7, font_8x5, buffer, COLOUR_WHITE, COLOUR_BLACK); // POSITION WRITTEN MODE BOTTOM LEFT I also removed the white writing at the bottom left of the initial splash.h image, because it was aesthetically out of place when decoding started and the modes writing appeared. By the way, I wanted to ask you which PC application you used to create the splash.h file from an image? Because I tried to replace it with another image and succeeded with the "lcd-image-converter.exe" program, but when inserted into the sketch, the image appears with very strange and different colors, making it unusable. I used the R5B6G5 16-bit color conversion method. Best regards to you and I await your kind reply. IU4ALH Antonino

…

Il 11/07/2025 22:29 CEST dawsonjon ***@***.***> ha scritto: dawsonjon left a comment (dawsonjon/PicoSSTV#7) #7 (comment) Hi, Thanks that's very neat! If you wanted to submit a pull-request, or share the change I would be happy to include it. There are more updates on the way including transmit capability and SD card storage. When I get a little time I will see if I can add the robot modes too. Thanks very much 👍 — Reply to this email directly, view it on GitHub #7 (comment), or unsubscribe https://github.com/notifications/unsubscribe-auth/AYCJW6IGRN6ZJURXZTAEKZL3IANEHAVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTANRTG4ZDQMBQGM. You are receiving this because you authored the thread.Message ID: ***@***.***>

[dawsonjon]

Sorry for delayed reply. Thanks, that's very useful. I will make the change when i get a moment. I used a simple python script to generate the image.h file, you can find it here: https://github.com/dawsonjon/PicoSSTV/blob/main/image2fbuf.py. Cheers Jon

…

On Sat, 12 Jul 2025 at 13:08, Antonino ***@***.***> wrote: *IU4ALH* left a comment (dawsonjon/PicoSSTV#7) <#7 (comment)> Hi. The modification is very simple; just adjust the parameters of the two lines that print the black background rectangle and the text with the modes and the row count. In practice, these two lines: line 185 and line 225 ..... 185 display->fillRect(320-(21*6)-2, 240-10, 10, 21*6+2, COLOUR_BLACK); //BLACK RECTANGULAR BACKGROUND at the bottom right with this :- 185 display->fillRect(128-(21*6)-2, 240-10, 10, 21*6-3, COLOUR_BLACK); //BLACK RECTANGULAR BACKGROUND at the bottom left and line 225 display->drawString(320-(21*6), 240-8, font_8x5, buffer, COLOUR_WHITE, COLOUR_BLACK); // POSITION WRITTEN MODE BOTTOM RIGHT with this :- 225 display->drawString(130-(21*6), 240-7, font_8x5, buffer, COLOUR_WHITE, COLOUR_BLACK); // POSITION WRITTEN MODE BOTTOM LEFT I also removed the white writing at the bottom left of the initial splash.h image, because it was aesthetically out of place when decoding started and the modes writing appeared. By the way, I wanted to ask you which PC application you used to create the splash.h file from an image? Because I tried to replace it with another image and succeeded with the "lcd-image-converter.exe" program, but when inserted into the sketch, the image appears with very strange and different colors, making it unusable. I used the R5B6G5 16-bit color conversion method. Best regards to you and I await your kind reply. IU4ALH Antonino > Il 11/07/2025 22:29 CEST dawsonjon ***@***.***> ha scritto: > > > dawsonjon left a comment (dawsonjon/PicoSSTV#7) #7 (comment) > > Hi, Thanks that's very neat! If you wanted to submit a pull-request, or share the change I would be happy to include it. There are more updates on the way including transmit capability and SD card storage. When I get a little time I will see if I can add the robot modes too. Thanks very much 👍 > > — > Reply to this email directly, view it on GitHub #7 (comment), or unsubscribe https://github.com/notifications/unsubscribe-auth/AYCJW6IGRN6ZJURXZTAEKZL3IANEHAVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTANRTG4ZDQMBQGM. > You are receiving this because you authored the thread.Message ID: ***@***.***> > — Reply to this email directly, view it on GitHub <#7 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAFPFX3Q4NVLDG4TOEWOPR33ID3CBAVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTANRVGE2DCMRRGM> . You are receiving this because you commented.Message ID: ***@***.***>

[IU4ALH]

OK... thanks. Anyway, I also planned a position in the top right of the modes and line count text, so I put three lines in the sketch (one for each desired position), and each user can choose the desired position simply by placing the classic // in front of the two lines they don't need. Kind regards. IU4ALH -- Inviato da Libero Mail

[IU4ALH]

Ciao John I tried using your procedure to create the splash.h file with the Python routine, but I can't because I get the error you can see in the attached photo. Do you have any advice for me? Thanks in advance. IU4ALH -- Inviato da Libero Mail

[dawsonjon]

Hi, I can't see the photo? I haven't tested the script widely, but it should be easy to fix. Can you let me know the error message?
Cheers
Jon

[IU4ALH]

Hi. I've provided the script below, including the input and output image file names (please check for any errors), and the error message below. The splash.png file is yours (320x240), and I'm using it for testing. ===== # _ ___ _ _____ _ _ # / |/ _ \/ | |_ _| |__ (_)_ __ __ _ ___ # | | | | | | | | | '_ \| | '_ \ / _` / __| # | | |_| | | | | | | | | | | | | (_| \__ \ # |_|\___/|_| |_| |_| |_|_|_| |_|\__, |___/ # |___/ # # Copyright (c) Jonathan P Dawson 2024 # filename: image2c.py # description: # # Converts an image file into a C array # License: MIT # import imageio import sys input_file = sys.argv[1] name = input_file.split("splash.png")[0] output_file = name+"prova.h" im = imageio.imread(input_file) h, w, c = im.shape pixels = [] for y in range(h): for x in range(w): r, g, b = im[y][x][:3] color = ((r & 0xF8) << 8) | ((g & 0xFC) << 3) | b >> 3 color = (color >> 8 | color << 8) pixels.append("0x%04x"%color) pixels = ",\n".join(pixels) contents = "static const uint16_t __in_flash() %s[] = {%s};"%(name, pixels) with open(output_file, "w") as outf: outf.write(contents); ================================= PS C:\Users\nino\Desktop\imagepy> & 'c:\Users\nino\AppData\Local\Programs\Python\Python313\python.exe' 'c:\Users\nino\.vscode\extensions\ms-python.debugpy-2025.10.0-win32-x64\bundled\libs\debugpy\launcher' '56308' '--' 'C:\Users\nino\Desktop\imagepy\image2fbuf.py' Traceback (most recent call last): File "C:\Users\nino\Desktop\imagepy\image2fbuf.py", line 19, in <module> input_file = sys.argv[1] ~~~~~~~~^^^ IndexError: list index out of range PS C:\Users\nino\Desktop\imagepy> ===================================== IU4ALH

…

Il 21/07/2025 16:33 CEST dawsonjon ***@***.***> ha scritto: dawsonjon left a comment (dawsonjon/PicoSSTV#7) #7 (comment) Hi, I can't see the photo? I haven't tested the script widely, but it should be easy to fix. Can you let me know the error message? Cheers Jon — Reply to this email directly, view it on GitHub #7 (comment), or unsubscribe https://github.com/notifications/unsubscribe-auth/AYCJW6O2YIKZEWIC7RLNBP33JT23HAVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTAOJXGAZTQNJXHE. You are receiving this because you authored the thread.Message ID: ***@***.***>

[dawsonjon]

The error message is occuring at line 19, it's expecting the filename to be passed as a command line argument i.e.
py image2buf.py splash.png.

It looks like you have hard coded the file name on line 20. If you want to take this approach I would suggest changing lines 19 to 21 with something like:

input_file="splash.png"
output_file="splash.h"

It's worth noting that I haven't ever tried this script on a windows platform, so it's possible that we might need to make further changes to get it working.

[IU4ALH]

Unfortunately, there's nothing I can do. I've tried many things, including what you suggested, but nothing works. Thanks anyway. IU4ALH

…

Il 21/07/2025 20:39 CEST dawsonjon ***@***.***> ha scritto: dawsonjon left a comment (dawsonjon/PicoSSTV#7) #7 (comment) The error message is occuring at line 19, it's expecting the filename to be passed as a command line argument i.e. py image2buf.py splash.png. It looks like you have hard coded the file name on line 20. If you want to take this approach I would suggest changing lines 19 to 21 with something like: input_file="splash.png" output_file="splash.h" It's worth noting that I haven't ever tried this script on a windows platform, so it's possible that we might need to make further changes to get it working. — Reply to this email directly, view it on GitHub #7 (comment), or unsubscribe https://github.com/notifications/unsubscribe-auth/AYCJW6O7JPZIUDL5FKIDSUL3JUXWVAVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTAOJXHE3DGMBXGA. You are receiving this because you authored the thread.Message ID: ***@***.***>

[IU4ALH]

In this way, which I'm writing below, the script doesn't give me errors, but it creates a splash.h file with all the lines 0x0000, which is a black screen. Do you have any ideas? IU4ALH ============================= # _ ___ _ _____ _ _ # / |/ _ \/ | |_ _| |__ (_)_ __ __ _ ___ # | | | | | | | | | '_ \| | '_ \ / _` / __| # | | |_| | | | | | | | | | | | | (_| \__ \ # |_|\___/|_| |_| |_| |_|_|_| |_|\__, |___/ # |___/ # # Copyright (c) Jonathan P Dawson 2024 # filename: image2c.py # description: # # Converts an image file into a C array # License: MIT # import imageio.v2 import sys # input_file = sys.argv[1] input_file ="splash.png" output_file ="splash.h" im = imageio.v2.imread(input_file) h, w, c = im.shape pixels = [] for y in range(h): for x in range(w): r, g, b = im[y][x][:3] color = ((r & 0xF8) << 8) | ((g & 0xFC) << 3) | b >> 3 color = (color >> 8 | color << 8) pixels.append("0x%04x"%color) pixels = ",\n".join(pixels) contents = "static const uint16_t __in_flash() %s[] = {%s};"% (h, pixels) with open(output_file, "w") as outf: outf.write(contents); ===========================

…

Il 21/07/2025 20:39 CEST dawsonjon ***@***.***> ha scritto: dawsonjon left a comment (dawsonjon/PicoSSTV#7) #7 (comment) The error message is occuring at line 19, it's expecting the filename to be passed as a command line argument i.e. py image2buf.py splash.png. It looks like you have hard coded the file name on line 20. If you want to take this approach I would suggest changing lines 19 to 21 with something like: input_file="splash.png" output_file="splash.h" It's worth noting that I haven't ever tried this script on a windows platform, so it's possible that we might need to make further changes to get it working. — Reply to this email directly, view it on GitHub #7 (comment), or unsubscribe https://github.com/notifications/unsubscribe-auth/AYCJW6O7JPZIUDL5FKIDSUL3JUXWVAVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTAOJXHE3DGMBXGA. You are receiving this because you authored the thread.Message ID: ***@***.***>

[dawsonjon]

I made a couple of tweaks to your script to get it working with version 3 of imageio. With this script I get an identical splash.h to the one in the repo. There is quite a lot of black in the picture, so the array is mostly zeros, but if you scroll down far enough there are some colored pixels too.

import imageio.v2
import sys

name="splash"
input_file ="splash.png"
output_file ="splash.h"

im = imageio.v2.imread(input_file)
h, w, c = im.shape

pixels = []
for y in range(h):
  for x in range(w):
    r, g, b = im[y][x][:3]
    color = ((r & 0xF8) << 8) | ((g & 0xFC) << 3) | b >> 3
    color = ((color & 0xff00) >> 8 | (color & 0xff) << 8)
    pixels.append("0x%04x"%color)
pixels = ",\n".join(pixels)
contents = "static const uint16_t __in_flash() %s[] = {%s};"% (name, pixels)



with open(output_file, "w") as outf:
  outf.write(contents)

[IU4ALH]

Yes..!!! Now Is functioning !! Thanks Very much.! I remember you for the ROBOT modes... -- Inviato da Libero Mail

[IU4ALH]

Hi John, I apologize for the delay in responding, but I've had some problems. Your image conversion sketch now works perfectly (though I found an online converter that works well, too). I also received many beautiful images, and I'd like to share my findings: - S1, S2, and PD modes, the decoding is almost perfect; - SC2-120 works well, but about halfway through the image decoding it loses synchronization (as I already pointed out to you at the time). - Martin modes 1 and 2, although the image is excellent, a column appears on the left of the image that suggests a shift value adjustment. For M2, this column is about double that found in M1. In this regard, I'm sending you a photo comparing the two modes, with some information on it so you can visualize what I'm saying. Anyway, this is a secondary point, in my opinion, because the priority would be the addition of the ROBOT modes. Best regards, IU4ALH Antonino

…

Il 24/07/2025 20:07 CEST dawsonjon ***@***.***> ha scritto: dawsonjon left a comment (dawsonjon/PicoSSTV#7) #7 (comment) I made a couple of tweaks to your script to get it working with version 3 of imageio. With this script I get an identical splash.h to the one in the repo. There is quite a lot of black in the picture, so the array is mostly zeros, but if you scroll down far enough there are some colored pixels too. import imageio.v2 import sys name="splash" input_file ="splash.png" output_file ="splash.h" im = imageio.v2.imread(input_file) h, w, c = im.shape pixels = [] for y in range(h): for x in range(w): r, g, b = im[y][x][:3] color = ((r & 0xF8) << 8) | ((g & 0xFC) << 3) | b >> 3 color = ((color & 0xff00) >> 8 | (color & 0xff) << 8) pixels.append("0x%04x"%color) pixels = ",\n".join(pixels) contents = "static const uint16_t __in_flash() %s[] = {%s};"% (name, pixels) with open(output_file, "w") as outf: outf.write(contents) — Reply to this email directly, view it on GitHub #7 (comment), or unsubscribe https://github.com/notifications/unsubscribe-auth/AYCJW6KYCFT4WXRILPRNVGT3KEOF5AVCNFSM6AAAAAB75SLKHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMJUGM3TSMBSGA. You are receiving this because you authored the thread.Message ID: ***@***.***>

[dawsonjon]

Glad you got it working! Can you share the online converter you used? Did you spot the recent pull request that adds scottie DX and robot 36? Should be fairly straightforward to add the other robot modes.