No, this is a family shaving game for family shaving people.
Show us yer programming projects!
- nyarla
- Baaaaaa!

- Posts: 2820
- Joined: Tue Jan 14, 2003 1:53 pm 0
- Location: i said kiss me you're beautiful these are truly the last days
- Contact:
Re: Show us yer programming projects!
Naughty version would be horribly painful to watch 
No, this is a family shaving game for family shaving people.
No, this is a family shaving game for family shaving people.
- super charlie
- Baaaaa!

- Posts: 1809
- Joined: Wed Mar 31, 2004 7:53 pm 0
- Location: Ipswich, suffolk
- Contact:
Re: Show us yer programming projects!
You don't fancy teaming up to make Magnetic Shaving Space Phallus then?nyarla wrote:Naughty version would be horribly painful to watch
No, this is a family shaving game for family shaving people.
Cheers
Charlie
Visit my website here to have a look at my game.
- nyarla
- Baaaaaa!

- Posts: 2820
- Joined: Tue Jan 14, 2003 1:53 pm 0
- Location: i said kiss me you're beautiful these are truly the last days
- Contact:
Re: Show us yer programming projects!
haha 
razors and cocks. a heady combination.
razors and cocks. a heady combination.
- piku
- Sheepie Whore

- Posts: 20960
- Joined: Wed Apr 24, 2002 10:44 am 0
- Website: Http://Www.ncot.uk
- Location: Wakefield
- Contact:
Re: Show us yer programming projects!
OK, so razors are a bit dangerous... So how about Magnetic Hot Wax Derby.
return (0);
Re: Show us yer programming projects!
I didn't know wax was magnetic.
Re: Show us yer programming projects!
Could be Hot Wax with Iron Filings in it?GoatFoam wrote:I didn't know wax was magnetic.
"Oh dear lord, I slightly soiled myself. Oxfam won’t be pleased."
Re: Show us yer programming projects!
Added RPI filters (from Snakey's Kega emulator) to ZXSpin:

But I have to say, Snakey's been a bit naughty with these - couldn't get certain plugins to work at all. Access violations up to the eyeballs, but only sometimes. Well, most of the time. But occasionally, I'd make a small change and it would work only to fail on the next run through....
Didn't help that the link to the scalers' source provided on Eidolon's Inn is a broken link, so I had to download the source to older scalers. After matching the source to delphi's disassembly of the compiled scaler, I found some downright nasty behaviour.
You start the filter by passing the source and destination surface pointers. In the case of 2xSAI, Snakey (Assuming it was him that wrote it, of course) is doing this:
Which means that the line "pcmpeqw mm1, [eax+colorE]", with ColorE being a constant 0, is causing the access violation. This is due to the earlier "sub eax,ebx" - eax points to the source pixel, and ebx contains the width of the source image. Sheesh. It's stepping up a line from the very first pixel in the surface where no line exists. Instant A/V 
It does it at the bottom of the image too - reads from pixels situated completely off the allocated surface memory.
I've fixed it by adding borders to the image to the top and bottom, and rendering the source bitmap at a vertical offset. The worst part is that there's no documentation about this at all - and it's very difficult to infer from the source of the filter. Took bloody ages to find that! Also undocumented is the scaling factor that the filter returns in the initialisation:
Can you see the intentional error?
That's right - if a scaler declares to the calling app that it scales by 3x, it sets both RPI_OUT_SCL1 and RPI_OUT_SCL2 as well as RPI_OUT_SCL3 (which is a combination of 1 and 2!). It's such a small detail that I completely missed it and meant that my code, which tested for each scaling flag in turn with a bitwise AND, reported that 3x scalers were all requiring 1x destination images. Grr.
You could maybe forgive that last one (in asm, it would be trivial to shift right, mask off the top 5 bytes and read the low byte of the register as the scale), but reading from outside the surface is not something I'd expect from a coder of Snakey's calibre. He may not have written the scalers, but they seem to be the "official" scalers from Kega, so it's possible he did. If he maintains images deliberately larger than the internal screen then it's understandable that such a mistake may be missed - A little documentation mentioning this would have been nice though!
Mind you, it was worth it - they're only marginally slower than ZXSpin's own versions of the filters.

But I have to say, Snakey's been a bit naughty with these - couldn't get certain plugins to work at all. Access violations up to the eyeballs, but only sometimes. Well, most of the time. But occasionally, I'd make a small change and it would work only to fail on the next run through....
Didn't help that the link to the scalers' source provided on Eidolon's Inn is a broken link, so I had to download the source to older scalers. After matching the source to delphi's disassembly of the compiled scaler, I found some downright nasty behaviour.
You start the filter by passing the source and destination surface pointers. In the case of 2xSAI, Snakey (Assuming it was him that wrote it, of course) is doing this:
Code: Select all
_2xSaILine:
; eax now points to colorE
sub eax, ebx
; Main Loop
xLoop: push ecx
movq mm0, [eax+ebx+colorA] ;mm0 and mm1 contain colorA
movq mm2, [eax+ebx+colorB] ;mm2 and mm3 contain colorB
movq mm1, mm0
movq mm3, mm2
pcmpeqw mm0, [eax+(ebx*2)+colorD]
pcmpeqw mm1, [eax+colorE]
pcmpeqw mm2, [eax+(ebx*2)+colorL]
pcmpeqw mm3, [eax+(ebx*2)+colorC]
It does it at the bottom of the image too - reads from pixels situated completely off the allocated surface memory.
I've fixed it by adding borders to the image to the top and bottom, and rendering the source bitmap at a vertical offset. The worst part is that there's no documentation about this at all - and it's very difficult to infer from the source of the filter. Took bloody ages to find that! Also undocumented is the scaling factor that the filter returns in the initialisation:
Code: Select all
#define RPI_OUT_SCL1 0x000010000
#define RPI_OUT_SCL2 0x000020000
#define RPI_OUT_SCL3 0x000030000
#define RPI_OUT_SCL4 0x000040000
That's right - if a scaler declares to the calling app that it scales by 3x, it sets both RPI_OUT_SCL1 and RPI_OUT_SCL2 as well as RPI_OUT_SCL3 (which is a combination of 1 and 2!). It's such a small detail that I completely missed it and meant that my code, which tested for each scaling flag in turn with a bitwise AND, reported that 3x scalers were all requiring 1x destination images. Grr.
You could maybe forgive that last one (in asm, it would be trivial to shift right, mask off the top 5 bytes and read the low byte of the register as the scale), but reading from outside the surface is not something I'd expect from a coder of Snakey's calibre. He may not have written the scalers, but they seem to be the "official" scalers from Kega, so it's possible he did. If he maintains images deliberately larger than the internal screen then it's understandable that such a mistake may be missed - A little documentation mentioning this would have been nice though!
Mind you, it was worth it - they're only marginally slower than ZXSpin's own versions of the filters.
"Oh dear lord, I slightly soiled myself. Oxfam won’t be pleased."
- piku
- Sheepie Whore

- Posts: 20960
- Joined: Wed Apr 24, 2002 10:44 am 0
- Website: Http://Www.ncot.uk
- Location: Wakefield
- Contact:
Re: Show us yer programming projects!
Haha you've managed to emulate the shitness of my old TV 
Can you now add a feature where randomly the 'TV' will detune itself. And simulate bad tapes too.
Can you now add a feature where randomly the 'TV' will detune itself. And simulate bad tapes too.
return (0);
Re: Show us yer programming projects!
Hmm. I already emulate bad tapes - turn off tape acceleration and emulate the hiss and wobblepiku wrote:Haha you've managed to emulate the shitness of my old TV
Can you now add a feature where randomly the 'TV' will detune itself. And simulate bad tapes too.
I feel really bad for the VBA-M guys, who've implemented these scalers and got around the problems I found by just passing image info that's larger than their actual surface... The scalers will in some cases just go wandering around on other memory allocations which will almost certainly show up as a bug at some point! It'll be a bitch for them to find... Should I tell someone? :-)
"Oh dear lord, I slightly soiled myself. Oxfam won’t be pleased."
- The Caffeine Kid
- Baaaaaaa!

- Posts: 5691
- Joined: Mon Mar 01, 2004 9:24 pm 0
- XBL Gamertag: TheCaffeineKid
- PSN: caffeinekid
- Yahoo Messenger: caffeinekid
- Location: Wakefield
- Contact:
Re: Show us yer programming projects!
Maybe it shouldn't... but that image looks a lot "nicer" than a perfect crisp picture. 
It takes me back... and feels more real somehow.
It takes me back... and feels more real somehow.
- piku
- Sheepie Whore

- Posts: 20960
- Joined: Wed Apr 24, 2002 10:44 am 0
- Website: Http://Www.ncot.uk
- Location: Wakefield
- Contact:
Re: Show us yer programming projects!
My game engine now has sprites, some fancy Catmull-Rom spline paths and a nice bit of XML to glue it all together.
And a video, it seems important to make videos of our progress now, rather than screenshots.
http://www.vimeo.com/8015369
The jerky behaviour is partly due to screen capture software, and also due to unoptimised debug-compiled code. And 200 sprites.

And a video, it seems important to make videos of our progress now, rather than screenshots.
http://www.vimeo.com/8015369
The jerky behaviour is partly due to screen capture software, and also due to unoptimised debug-compiled code. And 200 sprites.
return (0);
- sack
- Baaaaaaa!

- Posts: 7077
- Joined: Fri May 10, 2002 4:13 pm 0
- XBL Gamertag: evilgopher
- PSN: sixFiveOhTwo
- Location: Rijswijk, The Netherlands
- Contact:
Re: Show us yer programming projects!
nice... but you are aware I can do *infinite* numbers of sprites like that on an 8MHz Amiga? 
- Orangy Tang
- Baaaaa!

- Posts: 1794
- Joined: Sun Apr 25, 2004 5:25 pm 0
- Location: My secret lab
- Contact:
Re: Show us yer programming projects!
I'm getting stuck into writing a little roguelike:

You can only really walk around and hit things at the moment, getting the line of sight right has taken up most of my time.
Along the way I'm realising that roguealike's simple graphics mask some rather complicated goings on under the hood, especially when all the different elements have to interact in sane ways. Oh, and Unangaband's dungeon generation is complex and facinating.
You can only really walk around and hit things at the moment, getting the line of sight right has taken up most of my time.
Along the way I'm realising that roguealike's simple graphics mask some rather complicated goings on under the hood, especially when all the different elements have to interact in sane ways. Oh, and Unangaband's dungeon generation is complex and facinating.
[ Triangular Pixels.net - play Snowman Village, Rescue Squad, NiGHTS 4k and more ] [ Development Journal ]
- piku
- Sheepie Whore

- Posts: 20960
- Joined: Wed Apr 24, 2002 10:44 am 0
- Website: Http://Www.ncot.uk
- Location: Wakefield
- Contact:
Re: Show us yer programming projects!
Are they all real sprites that you can interact with?sack wrote:nice... but you are aware I can do *infinite* numbers of sprites like that on an 8MHz Amiga?
(plus how's the infinite sprites thing work? I did once know, but I've forgotten)
return (0);
Re: Show us yer programming projects!
Well, first you need infinite memory and infinite time....
A tree based infinite memory architecture is best, as it makes it possible to access your infinite memory with a finite bus, although access time is on the order of log(infinity).
It helps if you also have an infinite resolution display, or the sprites just make a mess.
A tree based infinite memory architecture is best, as it makes it possible to access your infinite memory with a finite bus, although access time is on the order of log(infinity).
It helps if you also have an infinite resolution display, or the sprites just make a mess.

