MSR *WILL* be finished, come hell or high water... hell more likely.

[TheOldMan]

Let's take this to a new topic... I'd really like to know....

zeograd.com has the last released version, but it hasn't been updated since 2005. Tomatheous gave me a few new routines for it, and I added a couple of things as well, but such modifications are kept in-house. HuC desperately needs an official update, but its maintainer has abandoned it.

What would you change/fix in Huc?
I have an include flag patch in place, along with some minor bug-fixes (ie, not skipping spaces correctly).
I'd like to split the library stuff up, so you could include only what you need, and -maybe- re-write the asembler to output linkable modules. Anyone interested in helping/testing?

[Arkhan]

They said they abandoned updating it because not enough people were using it.

But, I think not enough people are using it BECAUSE it needs an update...

yeah.

[nodtveidt]

I will test the f*ck outta any changes made to HuC. We abused the hell out of it with MSR, exposing its myriad of limitations and bugs. Just the other day I came across yet another flaw, involving variables set global at the source level (as opposed to global in globals.h, which seems to be at an even higher scope). Slow array access is the biggest thing in dire need of fixing, and from there, proc shuffling for minimum space waste... or, as you guys call it, "backpacking".

If no one uses your stuff, there's no need to fix it. But if you don't fix it, no one will use it. It's a catch-22.

[Arkhan]

Slow array access is the biggest thing in dire need of fixing,

yeah , and IIRC, >> and << aren't any faster than multiplies/divides...

pointers didn't work right at all really, and there aren't any structs... we have a list of complaints over here already.   What other ones do you have Rover, lol.

at least the Aetherbyte HuC Tweak extravaganza makes it a little more useable.  I'm using it for Atlantean and will be using it for the other crap too.

[nodtveidt]

Stuff I would want to see:

-Fix array access (as above)
-"Auto-backpacking" (as above too)
-Remove the utterly useless gfx plotting routines (already did this here)
-Remove the broken sound code that just wastes space in hucard building
-Optimize scroll()
-Better mapper functions that support more than 256 unique tiles
-Getting rid of scan_map_table() and making the mapper support height/width greater than 256
-Add ACD and SGX capabilities to hulib (Tom did this already)
-Modify compiler/assembler to take advantage of the extra 24KB of SGX work RAM
-More intelligent verbose code output (backslashes screw it up at present)
-struct support
-C++ style comments (would help with verbose output, I think)
-goto keyword (being forced to do this in assembly is lame)
-Expansion of the overlay table (50 is fine until you get serious)
-The addition of a 24 bit variable type for addressing (there's one internally but it's not exposed)
-Much more intelligent use of zero page
-ADPCM streaming (Tom coded one, so I have it in my hulib)

I probably have more but this is about all I can think of for now. I've tried to fix things myself but it's just a nightmare... the code is a mess and things aren't done optimally.

[Arkhan]

-Much more intelligent use of zero page

I loled.

I wonder who would actually use a new / fixed HuC aside from You, and Aetherbyte.

Everyone else seems to just do asm.

[nodtveidt]

A few people have expressed interest, but without something worthwhile of recommendation... well, you know.

ChrisBee wants to start up his own group, but he said he doesn't really code. I think he wants to learn though. mrhaboobi is obviously interested too. I think HuC's limitations really squashed any early attempts at greatness... people like Trilinear just packed up their toys and went home once they found out how broken it was.

And we mustn't forget about hcf. He's a pretty serious coder too.

[TheOldMan]

I took the HuC discussion to a new topic, so everyone knows. (What Needs Fixed In HuC)

I wonder who would actually use a new / fixed HuC aside from You, and Aetherbyte.

I think HuC's limitations really squashed any early attempts at greatness

Genearal opinion seems to be "If it wasn't such a pos, maybe a lot more people would use it" :0

[Arkhan]

Genearal opinion seems to be "If it wasn't such a pos, maybe a lot more people would use it" :0

Yeah, but the creator (and his closer companions) are of the opinion that there is no point in fixing it if noone is using it... so either we fix it (as usual? ), or enough people start using HuC and make 900 bug reports.

I believe one of them also said "The only way to fix it really is to just rewrite the whole thing".  Not a great sign there.

[nodtveidt]

Just because one says it doesn't mean it's correct. Yeah, it's broken, but yeah, it can be fixed without reinventing the wheel.

[Arkhan]

A few people have expressed interest, but without something worthwhile of recommendation... well, you know.

ChrisBee wants to start up his own group, but he said he doesn't really code. I think he wants to learn though. mrhaboobi is obviously interested too. I think HuC's limitations really squashed any early attempts at greatness... people like Trilinear just packed up their toys and went home once they found out how broken it was.

And we mustn't forget about hcf. He's a pretty serious coder too.

My question was who would care about a new C tool?  The question becomes "is it worth it to organize a widescale, done-up-nicely" released version with documents/explanations/etc..... or a "yeah 3 of us care so heres the .zip of the current fix" hack job.

Doesn't hcf do asm?

I know Tom is of the "go asm or go home" walk.

Touko seems to be fine in asm, but maybe he would like a new HuC?  Touko? What say you!

If some of the original Zeograd.com demo creators showed up expressing interest, man that would be awesome.  Where did they all go?

Just because one says it doesn't mean it's correct. Yeah, it's broken, but yeah, it can be fixed without reinventing the wheel.

Oh I wasn't saying it was correct, I was just giving reasons why HuC is dead in the water right now.  I asked questions, made comments, got responses, and drew my own conclusions about the status of the thing.

[nodtveidt]

Well, it almost makes me want to just reinvent hulib for cc65. heh.

[Arkhan]

Well, it almost makes me want to just reinvent hulib for cc65. heh.

Who knows, it might be worth the effort.  lol

[mrhaboobi]

I think you would find that newbies to any platform ( like myself ) would rather attack it in C then ASM in the first instance, have a decent C plaftorm would make things easier, yaroze was C and then converted to C++, Psyq ( the proper pfx tools ) did allow C and ASM, and sure ASM makes alot of sense, but personally for me it would be easier to start in C and then transition if i need to.

[touko]

oh, a new release of huc will be fine ..
For now i'am using huc only for banks mapping (because his ASM generated code is too dirty) , and i have 85/90% of assembly in my current project ..
And of course with custom libs .

And i agree with rover for goto in C, this is very  usefull for speed, and code size reduction ..