How do I split a PNG?
Is there min/max size?


What if I wanted the name to be:REDPIC4? Would it work?

when will it be out?

Now I think we have an issue of you not reading the ReadMe.
Just use any local image editing software, or online tool to split the image.

Yes REDPIC4 would work, the program only looks at the eighth character of the header to see which image half it's dealing with. The ninth through sixteenth character are used strictly for pairing images together, and providing the name.

When will it be released? I'll just follow Kerm's policy.

So how many weeks is it now?
And what about the easy converter?

Thanks I got it working.

I have 2 suggestions

1:Some color setings like: background color and text color.
2:I think the pictures names should have a max length of 16 chars instead of 8.

Thanks for your suggestions!
1. I'm adding that eventually but it's not a priority right now.
2. If it were simple to allow 16 characters I would, but as it currently stands: I find 8 characters plenty for a simple name, and there's nowhere near enough room for me to display 6 images per page and allow 16 character names. Therefore I don't think it's worth the effort to change a bunch of code and rearrange my image layout to allow for 16 char names.

My current priorities are:
1. Fix image detection (yet again)
2. Get the thumbnails to display with the correct pallet (that's gunna be a huge learning curve)

Can you add a zoom in/out function?

Got a few updates!
Thanks to PT_, image detection is working again!
Mateo is teaching me how the new thumbnail routine is going to work during our free time. Since he was busy today I started researching how to get my program to act as a subprogram for other programs.
I first tried to get the contents of Ans (which should be a string) in ICE, but it turns out ICE sets Ans to 0 when a string is stored in it. I can use a Str var instead but, I would've liked to not restrict the calling program's use of a Str var. I'll probably set it up so the calling program stores the image name (as it appears in my program) into a Str var then calls the HDPIC program.

A zoom in feature? That'll have to wait, I don't know how well the ICE routines can scale up an image. I'll also have to implement panning over an image. I suppose it's doable but, it'll be a later feature.

I worked more on getting my program to act as a subprogram and I got it to work! kinda. Unfortunately, I can't use Ans so I'm using the Str0 variable. I can't seem to pass my Str0 var from a BASIC program to my ICE program correcty. So

prgmTEST
Code:

Should immediately pull up my RGB test gradient picture, however it's not passing though my Str0 variable properly. If I check what's stored in it at the beginning of my program, it's just some random character(s).
However, if I put "HDPICV3RGBtest."->Str0 directly into my HDPIC source code, it pulls up the picture flawlessly.
I know this is possible, I tried looking at the ICESpriteCreator's BASIC source code but I couldn't see what I was doing differently. Could someone please shed light on what's going on?

EDIT:
runer112 clarified that the Str vars in ICE are not the same Str vars that you'd use in a BASIC program. I need to use a command that's something like OpenVar([720000],"r",4) to open Ans, then I can read from it.

Could it be the fact that the ICE version now doesn't have GetBASICVar()? If the new version doesn't have that and instead a substitute, I haven't updated ICE in a while...

iiirc, Get/SetBASICVar() are planned for a future version?

Do you know about this?
http://tibasicdev.wikidot.com/one-byte-tokens

The OS strings are formatted this way, as well as programs.
You need to use GetTokenString to get the ASCII values for the lowercase letters, due to them actually being two-byte tokens

<rant>
TI just has to make this sooooo difficult! Why not use standard ASCII??? Because they WANT this to be hard
</rant>

Progress update! I've got the single image thing at least partially working. If your image name is all uppercase my program seems to be handling it just fine, but give it a lowercase letter and it throws a fit. I'll have to integrate that command beck provided above!

I'm also having an issue where when Ans is something other than a string (or an invalid string), my program should just skip right over the single image detection code an just start up the main program. It seems to be doing that fine when Ans is a string, but if Ans is something other than a string, it's messing something up which causes the main program to not detect any images.

Another issue that I'll have to work out later is how my Appvars are stored. Currently, they're causing all sorts of memory issues which I posted about here. I'm not going to worry about it for my first release since that's probably going to require a code overhaul and I don't want to do that right now. I may also try and find a way to compress the images but Mateo said reworking how Appvars are stored was more important so I'm taking his word for it!

I've also updated the todo list. When all of the 'Higher Priority' things are done, I'll finally publish a V1 release!

I have a suggestion, why not open an image from BOSshell? (since you're already using Ans to run the program)
this page for details on the structure of BOSshell's 'packet' file (basically just stores data for the subprogram to read)

It's been a while since my last update but I got sick so I've had some time to work on it again. I remembered this feature request from over a year ago:

So I decided to focus on the converter itself and I managed to add a feature that allows for you to select multiple images at once then wait for the converter to cycle through all of them!
I had one small issue where my program was running too quick so convPNG would only convert one of the images but I've added a WaitForExit delay and it seems to be working just fine. I've uploaded it to github (first commit in 5 months, yay) so you can try it for yourself. (You'll need to build the ICE source code since it's not compatible with the old beta release)

EDIT: Due to my wacky scheduled I'm thinking about redesigning my UI. Instead of converting each of the 6 image's palettes to one consistent palette (which has proved quite difficult) I'd be changing it so only one thumbnail image is rendered at one time. I'd be reserving 2 colors (most likely white and black) in every image palette so I can keep the selection UI looking consistent. This will make it much easier on me to push out a full release and maintain, but that's two out of the 256 palette entries that are taken up by colors that may not be used in the image. Which would you guys like me to do? Keep my current UI or switch to an easier one?

Old : Concept : Conecpt 2

I like this one.


I'm glad to see this bring worked on.

I like the idea of the second one, but I would probably want a small thumbnail to show what it looks like (scaled down obviously). I like the color and look of the third, but the image size is just weird.

I have tried to use the .8xv files I made using HD Picture Converter CE on my calculator, but it doesn't read the 3 files (CREMES.R,CREMES.P,CREMES.L) in the program HDPICV. After that I of course tried to let it read the example image (dogger), with the appvars ExampleR, ExampleP and ExampleL. The program doesn't read them either. The program does read the appvars ExampleR and ExampleL with the landscape with a moon. I also tried making my own image using the .ini file and windows_convpng, wich also creates just 2 images, and the program does read those but then in a corrupted way. What am I doing wrong?

There's been a sudden influx of people (3) attempting to use this program and not succeeding due to its complexity and my poor instructions. Unfortunately, this is going to be my only post for the next three or four days so I wont be able to help.

The only help I can provide is: don't use the dev versions, they most likely don't work (you can try storing any string into Ans and running the program again, sometimes (for reasons unknown) that helps). Use beta 2 found here. Be sure that your image is 320x240 to start with. Do only one image at a time. Unfortunately beta 2 only gives you a standard definition image but at least I know it works.

I apologize for hyping up the 'full release' of HP Pic then just fading away after a few days, I'm still working on it along with the much more reliable C# converter, but IRL stuff has been holding me back.

Good luck everyone, when I get back I'll focus on answering your specific issues.

I prefer the second UI image you show. I think that would be easier for you.

Hello, after much working with spaghetti code over the past couple days I have finally put together an easy-to-use(-but-hard-to-read) wrapper for TLM's beta 2 image converter specifically.

Thank you, TheLastMillenial for your guide and apologies for sending you that semi-sour PM the other night.

Here is the GitHub Gist link... https://gist.github.com/Luxray5474/42fcb189c3aa5545fd4830aa9a7c4f77

You must download both of the files (convpng.js, convpng.template) into the folder where the windows_convimg.exe is located.

Features:

* Automatically fills in the convpng.ini file just from arguments you type in the terminal!
* Automatically resizes and splits the image with ImageMagick, so no more wasteful clicking and moving your mouse inside Photoshop or GIMP!
* Can convert to PNG if your input is a JPG.

I tried to make it easier to read while maintaining conciseness but due to the nature of JavaScript...

Requirements, instructions, and more are stated on the top of the script itself as a comment. I might include more options like auto cleanup of excess files and moving the .8xv's into a folder, but I am currently too tired of working with this code Anyone who can optimize this code or port it to a less messier language - please, be my guest!

Thank you again, TheLastMillenial for your excellent software!