nesdev.org

Piracy is incredibly common. I find it odd that you'd suggest otherwise here, on an NES emulation website. It is extremely common on PC. Less common on consoles, for various reasons. In recent years, the proliferation of online leaderboards has been a huge source of statistics on just how many people pirate games. Example: http://2dboy.com/2008/11/13/90/

The piracy statement is not my rationale, it's the one I most commonly heard from publishers. The second most common thing said was that they expect poorer sales from a PC release, though the piracy problem is considered part of the problem in sales, in this rationale.

tepples wrote:The TG16 is way beyond the NES, but it's also way behind the Genesis. What the 68000 has over the 65C02 are fast multiply and divide (needed for certain physics and 3D calculations) and 16- and 32-bit addition as primitives (same). And though both have equivalents of a DMA unit, only the Genesis has hardware for compositing two background layers. And more RAM allows for more destructibility in levels; the stock TG16 is down there with the SMS and the original Game Boy in this respect, and I don't think it was easy to fit extra work RAM in a HuCard. Finally, some game designs need a third trigger button, in addition to I and II or C and B for jump and standard attack, that isn't as hard to reach as Select.

I'm not so sure that it's "way behind". PC-Engine did not support two background layers however it did support an impressive amount of sprite coverage. Some of the more impressive games manage to give the appearance of more than one layer quite well.

Now the SuperGrafx is closer to the Genesis. Sadly it didn't catch on but it did support two background layers and even more sprites on screen and per scanline. Still not as much cpu available RAM as the Genesis at 32KB, but a significant improvement over 8KB. If it had managed to catch on and take over the regular PC-Engine's rein it would have been impressive when mixed with the successful CD-ROM in Japan. It wouldn't have worked so well for North America since it launched after the TG16 launched.

Oh and there are 3 button and 6 button controllers available for the TG16, just like the Genesis later had 6 button controllers available.

MottZilla wrote:Oh and there are 3 button and 6 button controllers available for the TG16

But I don't think there were a lot of games that required one. Games still had to be designed around the controller included with the console, just as games for phones have to be designed around touch control rather than, say, a MOGA.

rainwarrior wrote:tepples, when I was professionally involved in this, the canonical answer to why most publishers preferred to release on consoles, or at least release first on consoles, was piracy on the PC.

For the record, this is probably heavily in part due to DRM being built into the console itself (say whatever you want, the CIC chip is technically hardware-based DRM). This is notorious within the NES itself: the NES didn't really see much piracy, but the Famicom was piracy heaven, to the point even the system itself got cloned.

MottZilla wrote:I'm not so sure that it's "way behind". PC-Engine did not support two background layers however it did support an impressive amount of sprite coverage. Some of the more impressive games manage to give the appearance of more than one layer quite well.

The problem is, the sprite count is comparable to that of the Mega Drive and the SNES, so that isn't an advantage for it either.

I was just gonna mention the NES' 10NES chip. That was a very effective method of anti-piracy that was difficult for pirates to circumvent. To this day, no one has deciphered how the chip actually works, and the CIC chip used in repros is allegedly a clone of Atari's Rabbit chip, itself a clone of the 10NES. That said, it can be easily bypassed by modifying your console, and some companies used voltage spikes to temporarily knock it out. It was a very effective form of DRM that was also not very invasive.

OneCrudeDude wrote:To this day, no one has deciphered how the chip actually works, and the CIC chip used in repros is allegedly a clone of Atari's Rabbit chip, itself a clone of the 10NES.

Wrong: http://hackmii.com/2010/01/the-weird-and-wonderful-cic/

Despite that, many clone CICs out there are indeed clones of Tengen's clone. There's no allegation, there's a thread about the chip's reverse engineering efforts on this very forum.

Huh, I could've sworn that a thread here mentioned that, no matter how hard they tried, they couldn't find a pattern, and by extension they couldn't replicate it. Did other consoles of the time use lockout? I believe the SMS and 7800 had software based lockout, where they had to look for a string in the code. I find it ironic Nintendo didn't use a 10NES in the GB and instead opted for a in-code lockout. Then again, they probably had to save space.

Argument for the PC master race: http://www.polygon.com/2014/9/8/6120417 ... controller

Genesis, Game Boy, and Game Boy Advance require games to contain a valid header with the console maker's name as text (Genesis) or image (GB, GBA). This "Trademark Security System" (TMSS) was ruled ineffective in the USA in Sega v. Accolade. Atari 7800 games are flat-out digitally signed, something other companies' consoles wouldn't widely adopt until Xbox. Patent issues?

rainwarrior: You have a point about accessibility. For people who can't work a standard controller, it's easier to build a custom controller for a PC game. But then this breaks when a game uses XInput, which works only with Xbox 360 controllers (and with Xbox One controllers since a recent Windows update). All Windows Store games that use a gamepad use XInput, and all Windows RT games are Windows Store games. Therefore, at least for the purpose of controller accessibility, we can lump Windows RT with Xbox 360.

tepples wrote:

MottZilla wrote:Oh and there are 3 button and 6 button controllers available for the TG16

But I don't think there were a lot of games that required one. Games still had to be designed around the controller included with the console, just as games for phones have to be designed around touch control rather than, say, a MOGA.

Not sure on the amount that required said gamepads, but the 3 button pad is just the same as the Genesis (the 'select' button is moved over next to the action buttons). Quite a few games were setup to use select in it probably would on the Genesis (quick select/option button).

Some later games used the extra buttons of the 6 button pad, if it was detected. But it was mostly for quick item switching and such (a minor convenience).



As for the SGX, I'd put it above the Genesis (though not by a lot). It has twice the vram (128k), it has a combined 512 sprite pixels per scanline, combined 128 SAT entries, having two BG layers make the 16 of the 32 subpalettes just for BG layers even more usable/capable. The window registers (a priority mixing chip) allows for allows for complex graphic tricks and transparency shapes. The SGX has the same res modes as the PCE, but making the higher 7.16mhz and 10.74mhz dot clock pixels modes incredibly useful because the heavy sprite scanline pixel bandwidth to back it up. Both res modes are higher than the Genesis. The audio is still the same, but the SGX has a new revision of the audio chip which allows switching on/off channels without 'blips' - allowing you use each of the six channels 32 byte buffer as a PCM sample play back buffer (a 1khz timer routine can do 32khz stream streaming). I had demos up and running in no time doing this - I wonder if dev's would have eventually done this as well. The original audio chip can do this trick too, but you have to use/waste a spare channel to invert the blip (caused by sudden volume change), so it's more practical on the SGX. The Genesis still as a few nice tricks to its own in comparison, but overall all I'd take practical features (vram, sprite bandwidth,etc) over specific case special effects (shadow/highlight, column scrolling, etc).

At that first video you posted, I see the game doing some screen warbling effect. Could the NES do something similar? I know Kickle Cublicle does something similar.

Yeah, and more extreme too (see: road deformation in racing games). It's basically just linescrolling.

OneCrudeDude wrote:At that first video you posted, I see the game doing some screen warbling effect. Could the NES do something similar? I know Kickle Cublicle does something similar.

What part of Kickle Cubicle?

The sine wave animation? That's just to show that the transparency parts/transitions/edges aren't limited to just straight edges or lines. The hblur effect is to show that it can enabled for just the transparency part (making the light areas fuzzy; aka poorman's glow) - though not for the sprite (only BG areas). The two videos approach the transparency effects in different ways (and have their own advantages/disadvantages). But yeah, both are H-int effects (common called 'linescroll' effects). They can be done on the NES with an external interrupt+timer (or line reader like the mmc3 does IIRC). So they can be done on the nes, and full screen like that, but you'll be cutting into your already limited cpu resource. You're looking at 224-240 interrupts per frame, so multiply that with the amount of cycles the routine takes. Plus you have setup/prep time that you need to add to that (usually a table). On the NES, I'd probably do it every two lines and apply the savings for something else.

There's a third video, but it's nothing special that other systems can't already do (other than it's more practical to do since there's a lot of sprite bandwidth per line than they other systems). Both Genesis and PCE can do it, and I believe a couple of games do this trick on both systems. And if I have my fact straight, that trick can also be done on the NES. But the limited sprite bandwidth makes it near impractical do in application on the nes. And, IIRC, you can't do this on the SMS because the BG selects the priority of the sprite instead of the sprite table entry itself (someone correct me if I'm wrong here).

On that note, do any official NES games do that trick (sprite priority->BG trick)? I mean for transparency/silhouette (and not for popping a piece of the BG over the sprite in dynamic areas).

tomaitheous wrote:

OneCrudeDude wrote:At that first video you posted, I see the game doing some screen warbling effect. Could the NES do something similar? I know Kickle Cublicle does something similar.

What part of Kickle Cubicle?

https://www.youtube.com/watch?v=VIOo3DZRPWk at 14:23

According to the same game, there's also a scene where the sea is parted (vertically) to reveal a castle underwater. I never knew the NES could split backgrounds vertically and scroll them, unless they simply redrew the BG tiles and used sprites to make it look as if it's a transition.

Another good example of sine wave line-scrolling effect is the title screen of the Japanese version of Fantasy Zone on the Famicom.