|
|---|
Marcel Verdaasdonk
Netherlands
| |
Posts 3991
16 May 2010 12:46
| Thomas Richter wrote:
|
*) Is that an unmodified version of Giana sisters or did you have to patch it up a little bit to adapt it to the different processor speed, processor capabilities etc? Maybe it's an uncritical game, but I would expect some problems from games running at a much higher processing speed than the hardware they were designed for.
|
isn't that a sign of bad programming it's in the manual not to use CPU loops as a event timer.
| |
Ajc ;)
United Kingdom
| |
Posts 688
16 May 2010 13:24
| Marcel Verdaasdonk wrote:
|
Thomas Richter wrote:
|
*) Is that an unmodified version of Giana sisters or did you have to patch it up a little bit to adapt it to the different processor speed, processor capabilities etc? Maybe it's an uncritical game, but I would expect some problems from games running at a much higher processing speed than the hardware they were designed for.
|
isn't that a sign of bad programming it's in the manual not to use CPU loops as a event timer.
|
Having worked in the games industry for several years now you'd be horrified at how often stuff like that still happens :(Andy
| |
Asaf Ayoub
United Kingdom
| |
Posts 332
16 May 2010 13:31
| In Amigame, using an adf version of the game I can hear the floppy disk drive loading, even though I dont have one. @Mr Copland Yes, some people still use software loop delays unfortunately.
| |
Marcel Verdaasdonk
Netherlands
| |
Posts 3991
16 May 2010 18:17
| people using that method should have the guts to give customers their contact info.
If you code some bad code your responceble for it the rest of your life!
| |
Flash Lab
Netherlands
| |
Posts 166
16 May 2010 18:57
| If those people are programming nuclear power plant software, you're right. If you're talking about games you are gravely exaggerating.
| |
Fahed Al Daye
Canada
| |
Posts 282
16 May 2010 19:37
| Flash Lab wrote:
|
If those people are programming nuclear power plant software, you're right. If you're talking about games you are gravely exaggerating.
|
No, no, you misunderstood me. It is not flashing LED for when the system crashes, although Natami team promised me that they will implenet it on the system....hehehe. But this flashing LED for the DVD Player and disk drive when there is an error reading or writing into it and that is unavoidable at all cost when it comes to these media devices.
| |
Gilles DRIDI
France
| |
Posts 107
16 May 2010 19:49
| Flash Lab wrote:
|
If those people are programming nuclear power plant software, you're right. If you're talking about games you are gravely exaggerating.
|
I heard thet people programming for nuclear power use interpreter.
Because the reason is stability !
Like sorting for parallel processing, after all !!!
Gilles
| |
Tony Smits
Netherlands
| |
Posts 39
16 May 2010 20:04
| @Flash Lab: But it is still bad programming.
| |
Flash Lab
Netherlands
| |
Posts 166
16 May 2010 20:14
| No argument there! Just the public stoning suggested ticked me off.
| |
Marcel Verdaasdonk
Netherlands
| |
Posts 3991
16 May 2010 21:11
| Flash Lab wrote:
|
No argument there! Just the public stoning suggested ticked me off.
|
what a great idea Flash Lab. :P
| |
Gone Gahgah
Australia
| |
Posts 237
17 May 2010 00:51
| Mr Copland ;) wrote:
|
Marcel Verdaasdonk wrote:
|
Thomas Richter wrote:
|
*) Is that an unmodified version of Giana sisters or did you have to patch it up a little bit to adapt it to the different processor speed, processor capabilities etc? Maybe it's an uncritical game, but I would expect some problems from games running at a much higher processing speed than the hardware they were designed for.
|
isn't that a sign of bad programming it's in the manual not to use CPU loops as a event timer. |
Having worked in the games industry for several years now you'd be horrified at how often stuff like that still happens :(
Andy
|
I bought an old version of Magic Carpet to run on my PC.
It runs but does so so fast I can't even tell what is happening and the game is over in a second!
And it was considered leading edge in its day!
| |
Matt Hey
USA
| |
Posts 737
17 May 2010 01:46
| gone gahgah wrote:
|
I bought an old version of Magic Carpet to run on my PC.
It runs but does so so fast I can't even tell what is happening and the game is over in a second!
|
PCTask is freely available for the Amiga now and should take care of that problem ;).
| |
Marcel Verdaasdonk
Netherlands
| |
Posts 3991
17 May 2010 01:48
| Okay now to a more serious tone.
If software breaks because you use a faster compatible machine you should look at two things.
Timers and loops used in software.Software loops are responsibility of the programmer, period! programmable interrupt hardware are responsibility of the hardware manufacturer. See where this is a bit loop sided?
this means the hardware interface cannot change from the software perspective, no matter how fast it can go! Flash Lab see the little bump in the road i am looking at here?
| |
Flash Lab
Netherlands
| |
Posts 166
17 May 2010 06:55
| Again I understand perfectly. Just the responsible for life remark was really silly.
| |
Przemek Tkaczyk
Poland
| |
Posts 54
17 May 2010 07:27
| @Asaf That's not what I meant ;) I doubt the Team recompiled Giana Sisters :) Furthermore, shitload of the scene demos never got released as a source code.. I was just wondering if any ECS demos work on Natami at this moment...
| |
Thomas Richter
Germany
| | (MX-Board Owner)
Posts 1425
17 May 2010 07:44
| Przemek Tkaczyk wrote:
|
@Asaf
That's not what I meant ;) I doubt the Team recompiled Giana Sisters :) Furthermore, shitload of the scene demos never got released as a source code.. I was just wondering if any ECS demos work on Natami at this moment...
|
Not recompiled, of course - that doesn't make sense. But possibly patched a bit with a disk editor. Many many of the old games did not work on my 68060 based A2000 for various reasons, and I wouldn't be surprised if Giana Sisters also had issues - given its age. It is not necessarily CPU loops that count. For example, given the complicated audio state machine, software existed that - instead of waiting for Paula generating the interrupt that it is ready to take new samples - just did some other work in their game mechanics, and then came back banging the registers. It did work on an unexpanded A500 or A2000 without fast mem because the program code happened to be slow enough to have the Paula registers ready in time, but on an expanded machine sound broke down.Similar issues exist with the blitter (the blitter-busy flag is only asserted several cycles *after* the blit has been started, and becomes clear *before* the final blitter output word is written). It did work on slower machines, but the Os contains a couple of workarounds for the issue on faster machines. Maybe that issue is resolved now given that the blitter runs at full CPU speed (I guess or -hope-). Similar issues exist for IRQ handling where clearing an IRQ on a faster machine just takes so long that the CPU sees the IRQ pending while having cleared the state cycles ago such that a second bogus IRQ is generated. All tiny nasty issues programmers didn't bother about, and that caused trouble on faster hardware - not even necessarily loops for timing. Thus, I'm wondering how much work had to be spend to patch up old software. I'm not saying that anyone is doing a bad job - quite the opposite, of course! Thanks,
Thomas
| |
Team Chaos Leader
USA
| | (Moderator)
Posts 2094
17 May 2010 08:03
| Thomas Richter wrote:
|
For example, given the complicated audio state machine, software existed that - instead of waiting for Paula generating the interrupt that it is ready to take new samples - just did some other work in their game mechanics, and then came back banging the registers. It did work on an unexpanded A500 or A2000 without fast mem because the program code happened to be slow enough to have the Paula registers ready in time, but on an expanded machine sound broke down.
|
I found the same problems in various modplayer routines.
| |
Thomas Hirsch
Germany
| | (MX-Board Owner)
Posts 647
17 May 2010 08:18
| Thomas R,
the game is/was running from disk. The copy I used was from 1988/1989. Unaltered since then.Gianna Sisters is timed by vblank. This means if you run it on an NTSC machine it runs slightly faster. I tried this 20 years ago, not now. So in the case of speed this game is uncritical.
| |
Thomas Richter
Germany
| | (MX-Board Owner)
Posts 1425
17 May 2010 09:17
| Thomas Hirsch wrote:
|
Thomas R,
the game is/was running from disk. The copy I used was from 1988/1989. Unaltered since then.
Gianna Sisters is timed by vblank. This means if you run it on an NTSC machine it runs slightly faster. I tried this 20 years ago, not now. So in the case of speed this game is uncritical.
|
Wow, great then! So Natami *does* have a floppy port.Greetings,
Thomas
| |
Michael v. Parent
France
| | (Natami Team)
Posts 434
18 May 2010 20:45
| sorry I missed this info, but is natami yet capible of running any AGA progams?
| |
|
The team will post updates and news here