New official VERGE build (finally) released!
Displaying 1-20 of 31 total.
12 next
Please enter a numerical value for the importance of this sticky.
Enter 0 to unsticky.
vecna

Yes, this build is WAY late, but here it is. You can view the full changelog, and here's a new v3vergec.txt reference. Hopefully you'll see the online doc section updated to reflect the new hotness soon. Short version is there's several key bugfixes, including big improvements with line number error reporting. New stuff including FlipBlit(), a set of functions to get/set a variable value by its name, some new functions to interact with the windows clipboard, and a number of new variables, including curmap.tileset which provides an image handle to the current map VSP, that you CAN draw to (the image will be 16 pixels wide and 16*numtiles tall).

Planned binary-tree based optimizations did not make it into this build in the interests of just getting it done so you'll believe me when I say we're not dead. I did however add a new verge.cfg option that controls whether or not vcc_verbose.txt is generated during compilation, which defaults to off, and on large projects, this will reduce compile time by a noticable amount. Tree optimizations later on will improve compile time further as well as the lookup time for CallFunction and the new variable Get/Set functions.

We also added a small splash screen that appears when V3 is compiling to let people know that v3 didn't crash on startup or anything when compiling large projects that aren't in releasemode. I just whipped this one up in photoshop in literally thirty seconds. I will happily accept submissions for sexier screens. The general guidelines are that it should be a 320x200 GIF that will be displayed centered in the screen for larger resolutions. If anyone is interested in making a better compile screen, feel free to just post it in the comments for this post.

You may not believe me, and I won't blame you, but I do mean to try to get back on a every-week-or-two release schedule for the engine. So feel free to submit additional feature requests and bug reports here as well. I read the forums as much as I can but stuff does slip through. Grue also tells me you all can expect a new Sully build in a few days, so if he doesn't come through, I encourage you to throw rotten vegetables at him, such as brocoli or cabedge or perhaps carots.

Posted on 2004-08-02 03:23:25 (last edited on 2004-08-02 03:24:16)

Toen

FIRST POST

Posted on 2004-08-02 03:38:06

gannon

I was wondering if you could get at the ints in structs with GetInt (you can't could you put that in please)
I found it crashes if it can't find the varible with SetString (I didn't test the others)
thank you for the get and set int and string arrays that will be of great help

Posted on 2004-08-02 04:21:59

Ness

So I put the new exe in my dev directory, start it up in compile mode, and it compiles IN 2 SECONDS. As opposed to around 40 for the old version.

All I can say is .....word.....

Posted on 2004-08-02 04:32:33

mcgrue

Quote:Originally posted by Ness

So I put the new exe in my dev directory, start it up in compile mode, and it compiles IN 2 SECONDS. As opposed to around 40 for the old version.

All I can say is .....word.....


VCC Verbose: the silent killer!

Yeah, Sully compiles in mere seconds now too. I had no idea it was a massive log-writing excersize that was slowing everything down.

The best thing about this build IMO tho is the improved error reporting. Yaaay!

Posted on 2004-08-02 04:40:52

vecna

vcc_verbose wouldn't be that slow if it just was outputting variable and function declarations, but it actually outputs for each and every library function call as well (the idea being that if I had a VC compiler crash, that would help me narrow down exactly where it was crashing). And of course, because its opening, writing to, and closing the log file for each statement.

Posted on 2004-08-02 04:45:29

mcgrue

Yeah. Likewise, if you run the Sullycode with verbose logging on, you can be waiting for... quite a while.

Posted on 2004-08-02 05:27:04

RageCage

My god... I'm in love... the power... the POWER!!!

Posted on 2004-08-02 05:51:25

Zip

Wohoh! SetRandSeed is fixed! TIme to go and play moar...

Zip

Posted on 2004-08-02 10:27:47

Buckermann

Great work Vecna!
And you forgot to mention the new sexy icon for the verge binary :)

Posted on 2004-08-02 15:44:20 (last edited on 2004-08-02 15:49:42)

Troupe

Quote:Originally posted by vecna

I encourage you to throw rotten vegetables at him, such as brocoli or cabedge or perhaps carots.


omgcabedge *wip*

Posted on 2004-08-02 20:49:04

rpgking

Wow, this looks really good. I'm glad to see things like button remapping in there. The variable get/set functions look really useful too, and I'll have several possible uses for that VSP pointer.

Posted on 2004-08-02 21:20:10 (last edited on 2004-08-02 21:31:49)

Ness

Heh, it even crashes faster when you do an array out of bounds

Oh and error reporting on the line that actually caused the error is nice to

Posted on 2004-08-03 00:42:36

Zip

I've quickly doced some of the stuff I've played with today, but there's still plenty to be done. Having the renderstring running is TOO nice. :D

Troupe - I was gonna make a similar comment, but I thought after having half the stuff I asked for implimented, it would be rather unfair. The song however, rox. :)

Zip

Posted on 2004-08-03 01:55:16

Falthorn

New Sully to show off the sexification?
eh? EH?

And where the fuck are those diagonal directions, vec?!
EH?!

Posted on 2004-08-03 04:06:32

vecna

oh diagonals yes yes soon just please stop hitting me ;_;

Posted on 2004-08-03 04:16:08

zaril

Could we get an Ascii(string chr)?

Posted on 2004-08-03 11:02:59

resident

Hurrah! Entity.visible, Entity.obstruct and Entity.obstructable. Back to work with my zombie themed game then.

Posted on 2004-08-03 12:32:26

Technetium

Alright, here are my questions about a lot of the changes...

Rendermap(): Can this render areas larger than what would normally fit on the screen? For example, if I made a new image that was 5000 by 5000 pixels, and used rendermap at (0,0), would it render as much of the map as it could fit? Also, does it render from 0,0 on the map, or does it begin dependant on the xwin and ywin variables?

CopyImagetoClipboard: I'm confused on the use of this function. Does it let programs outside of Verge access this clipboard? If not, can't we just get the same effect by creating a new image and duplicating the old one over it?

curmap.rstring: is this read-only, or can we write to it as well?

layer.lucent: I assume this is an array. Is it layer[].lucent or layer.lucent[]?

Posted on 2004-08-03 14:14:33

vecna

Hurray for zombies!

Re Technetium's post:

RenderMap() - yes, it will fill as much of the image as possible, with the upper-left corner of the image being the coordinates specified at xwin, ywin. So you can render more or less of the map than is otherwise on the screen. You could very well blit the entire map to an image.

Clipboard functions: Yes, these functions are referring to the windows clipboard. They may have limited utility, but I can think of a few things they're useful for. Personally I use it side by side with photoshop to do certain effects, and I also wanted to make a program in VERGE to assist me in tile-ripping from like emu screenshots and stuff.

renderstring is read/write.

It's layer.lucent[]. I meant to add more layer variables, as well as some zones, but didn't have time. Probably next build.


Posted on 2004-08-03 17:12:55


Displaying 1-20 of 31 total.
12 next
 
Newest messages

Ben McGraw's lovingly crafted this website from scratch for years.
It's a lot prettier this go around because of Jon Wofford.
Verge-rpg.com is a member of the lunarnet irc network, and would like to take this opportunity to remind you that regardless how babies taste, it is wrong to eat them.