WHAT IS 8-BIT? | What are 8-bit graphics, anyway? 

yet the textures can go from 16x16 to 512x512,and the rendering distances and colours used are way too many to be called 8 bit,over 256 and you can have with mods over 256 items and over 256 blocks,soo.....*ahh* *facepalm*

+JosephKing75 CBP not only that but MineCraft uses Voxel style graphics if I recall correctly.

Taylor Ringo exactly,it used to not have voxel graphics.

It still uses textures though

+Taylor Ringo It uses voxels to store information. When it renders to screen it uses polygons.

A few, is an understatement. Super mario bros used 40kb. Most games made later used 256kb.

This comment is funny after watching the video. When you say 'actual 8 bit', which one do you mean? lol

They were able to do that because they had direct or almost-direct access to the actual hardware and just wrote pointers to the specific items in the graphic array of the memory devices they were writing those games for. That would still be the case now if hardware manufacturers weren't actively fighting against this and the only access you have to the actual memory is via an API that's purposely convoluted as to obscure the actual hardware architecture.

@@rusi6219 Yeah, but doesn't change the fact that it's impressive no matter what. What a time to be alive. And this is outdated technology.

@@marcuscarana9240 I'm just pointing out that this would still be possible to do in many ways if it wasn't for the industry being anti-consumer

Most look like 32bit games on the gba

Fun fact: when 64 bit processor was created it was enough powerful that ps5 and xbox series x is a 64 bit console

@@Faoijhfr4rsgioi false. newer systems are 128 bit

@@LLotad uh no. newer systems are still 64 bit. and they're honestly probably gonna remain that way for a long, long time

@@chunkypotato idk theres no real info on what bits they use. Ps4 could be 64 or 128. I'm leaning more towards 128 because the ps2 and dreamcast were 128 bit, with the xbox being only 32 bits. either way the whole bit thing stopped mattering after the dreamcast tbh

1 byte = 8 bit

nemkutya.com/wp-content/uploads/2014/12/98253-1-hobbit-1-hobbyte.jpg xD

Here's how the big guys define what a byte is: int number_of_bits_in_a_byte = CHAR_BIT;

I know this thread is a bit old, but this isn't necessarily correct. A bit refers to the smallest amount of data that can be addressed by a cpu, and is somewhat related to processor word length. It just so happens to be 8 bits on modern systems, but can theoretically be any value. Early computers from the 50's and 60's used different sizes for their bytes.

Hydrochloric Acid Correct. I wish I remember what it was, but I was watching a video about a retro piece of hardware (I want to say it was an ancient arcade architecture) that said the CPU used 19-bit word sizes. Seriously? What sadistic person made a 19-bit word-sized CPU? But going back, you can find all kinds of odd word sizes: 9, 12, 26, 39 and even higher. The world settled in on power-of-two word sizes because they made the most sense in the context of computer science.

I think Stardew Valley looks like SNES (if you ignore all the effects, rotation and scaling that wouldn't be possible on a real SNES).

bloody_albatross scaling is possible on a snes if you're talking about what i think you're talking

I meant: Stardew Valley has actually a much higher resolution than the pixel look want's you to believe. You can see that when they didn't draw extra sprites for certain animations but just rotated and scaled the existing sprites. Then one "pixel" might be 0.5 or 1.5 the length of another pixel and other pixels are at an angle and not aligned to the supposed pixel raster.

The only 16 bit game I know is Terraria.

I think the NES era is still more nostalgic. Give it 4 more years.

@@nugget6635 Flicker is actually not a glitch, it's an intentional programming "trick" to allow for more sprites on the screen at once. I found this out after trying NES development (which is very difficult, especially for someone who struggles with coding like myself)

@@williamdrum9899 yeah. it's programmed into the game. it's not built into the hardware, like most people _assume_. the way it's programmed in games is also really difficult.

Natok the Watok Yes plus

Still goes after the nostalgia factor that the actual 8-bit games have so it makes sense what they're referring to

What people called "x-bits" in the late 80s-90s was always very vague, and more of a buzzword than anything. At first they meant the processor word size, but then you had processors that had 16 bits words but only had an 8 bit external bus size, so their speed was limited. Then you also had the number of bits per pixel used to display graphics. So got things like the TurboGrafx-16 were the CPU was 8-bit, it had supposedly a 16-bit graphics processor, but could only make use of 9-bits of color. In the end what actually matters is the quality of the games.

customsongmaker yes he did at :47...

X-bit refers to the number of binary values the processor can work with. 4 Bit processors work with 4-bit binary (0000 or 0101). 8 bit processors work with 8-bit binary (00000000 or 00101011). 4 bit, like said in the video, can only store values from 0-15. 8 bit can store values 0-255. Today, we are using 64 bit processors (mostly, there are some 32 bit processors being used). They are capable of storing values from 0-9,223,372,036,854,775,807 (9.2 quintillion). processors with higher bits support more RAM and can run more efficiently. (I also believe that some programs are smaller when 64 bit than their 32 bit counterparts.) 32bit processors only support 4 GigaBytes of RAM. 64 bit Processors Support up to 16.8 Exabytes or 16.8 million Terabytes.

It means nothing. There are very many things to consider (ALU width, registers width, data bus width, plus multiplexing on the top), so it is not really a tech term, more of marketing stuff. 6502 can be a good example of very 8-bit architecture, though, as it has 8-bit ALU, 8-bit registers including the stack, 8-bit data bus, sees the address space as set of 256-byte pages, and is not capable for atomic 16-bit calculations. For most of other CPUs it is not that clear. For one, Z80 considered 8-bit, but has 4-bit ALU, but also capable for atomic 16-bit calculations.

I thought basically it meant that with 8 bits the programmers binary that makes up the programming of the game has 8 spaces for the 0's and 1's to make up information.....that may be oversimplifying. ..

But with 10% of the animation budget.

Tangy ™ hand drawn animation i think

But traditional animation is 2D animation. Are you saying that tradigital animation is not the same as traditional animation? Because they're similar, but one is done digitally and the other is done on paper and cels.

SeepMan pixel art. the same

@@TotallyGoodatGames paper and cels are traditional

@@supahcomix Yes.

Thank God I Thought I was the only one!!!! 😫😫😫

Yes, he's a guest in this video

Uh... Google. search "how to avoid clones"

Sounds like it...

Mark Brown!!!

Actually, there's a 8x16 sprite mode, but it's not very useful unless the game scrolls vertically. It also limits you to 32 sprites, so in essence, there's still 64 sprites, but they are automatically glued together for you by the hardware.

+JohnRegularYum This guy (@britishgaming) actually makes videos like this on his own channel (called Game Maker's Toolkit) as well - I would recommend checking it out if you haven't already. :)

That's exactly what I was just thinking! I wonder who he is...

I just commented “are you game makers toolkit?” And then saw this comment. Thanks for clarifying. Subbing solely based on who you are. I’m sure this channel is full of quality video if you’re here.

@@tanner_titus I haven't worked at Pocket Gamer / AppSpy for about 7 years so don't expect anything from me here :P

These are a caveman's version of video games, It brought down the video game industry as low as it could go,it's proof that jesus died in vain and legally changed his middle name to FUCKING

It's a shitty version of the Virtual Boy... Yes, I said that!

stufaman as if the virtual boy wasn't shitty enough

***** It did have a head strap though.

Bits cannot go down into fractions...

You missed the entire point of the video. There's no "8bit" style, only a replication of what popular 8bit consoles at the time could produce. Shovel knight isn't any "bit", and if you consider CPUs then it would be 32/64bit with an aesthetic similar to consoles like the NES. The "bitness" of the machine does not always determine graphical power. The Atari ST was 16bit but only could produce 16 colours at once, and the PC Engine has an 8bit CPU, but has similar graphics to the SNES/Megadrive. Early IBM PCs were 16/32bit but depending on the graphics card, only could produce 4/16 colours at once.

Mixa No, the developer wanted to make a game that looks like a NES game

CodeNameZ That was just marketing speech, I don't really care what the devs have said when the end product looks more like Super Metroid or Castlevania 4 rather than Ducktales or Super Mario 3. It only adapted the gameplay elements from nes era, but graphically it's definitely snes style.

I'd say it is very PC Engine style, not SNES. Similar palette, similar capabilities, including sound.

PC Engine games were generally more colourful - more close to the Megadrive in capabilities.

yes, the square wave on the nes was more like 3-bit. not sure about the triangle or noise channels. the samples channel was 1-bit (delta modulated) but with a fairly high sample rate (which i believe was variable, and had to share clock cycles with cpu instructions)

Actually, pulse (not square, because of the duty control) wave is 1-bit by the nature, wouldn't be pulse/square otherwise. But pulse channel DACs are 4-bit, providing 16 levels of volume. The triangle is 4-bit, a counter sends 0-15-0 sequence to the DAC. Noise is 1-bit by the nature of digital noise generators (LFSR), but also has 4-bit DAC to control the volume. DPCM uses 1-bit encoding for the data, but sample channel DAC is actually 7-bit, although in the DPCM mode it only uses 6 bits. So one could count a lot of bits in the NES sound hardware, but 8 is nowhere to be found.

Haha, yeah, thats what i meant

Chiptune is any fucking waveform you want. But I guess you have only heard of the NES.

umm well we were talking about the term "8 bit music" and using NES as an example. yeah obviously chiptunes encompasses any music produced by a chip (or at least music that sounds like it is)

Seeing the pixels rotate makes me very upset. Games like Mega Man 9 and 10, Shovel Knight, Freedom Planet and Sonic Mania have incredible Sprite work for modern systems however.

@Vikrinox 3u,bxbw9wc

I believe that true 8-bit sprites have separate, rotated versions when needed.

I can't stand when people talk about 8-bit but only refer to the NES. SID was better sound processor, so there! Though I do feel your pain, it appears that to the young 'uns 8 bit means nothing more than "composed solely to square waves"

***** Hate to go all Four Yorkshireman on you, but back when I were a lad, t' Commodore 64 only 'ad THREE VOICES and only had t' choice of square wave, sine wave, sawtooth, noise, and "pulse" whatever that were... Mind, we did 'ave adjustable Attack, Decay, Sustain and Release on t' different voices and we could reprogam t' different channels to t' different waveforms anywhile, it were reet good in fact!

chip tunes

Señor Inteligente There are 2 square channels, 1 triangle, and 1 noise. (dpcm could be a 5th channel, but if you want TRUE chip tune, forgo the dpcm channel) The square channels can have 3 different wave forms and the volume can be adjusted. The triangle had 1 form, and no possibility to for volume adjust. Noise has 2 forms, and volume adjust is possible.

Yeah, because only the NES is true 8-bit. Fuck the rest. Fuck the Master System. Fuck the C64. Fuck the Game Boy and especially the PC Engine.

flappy bird looks more like a gba game

Yes Indeed!! One of my favorite NES games.

@@202-Uptown-Legend same

Search for "Mark Ferrari". He has a whole bunch of great 8-bit art pieces.

it's also not 8-bit, i don't know why he says this

sorry I'm late to the party, but that really is 8-bit. It actually can run on the Atari 800XL.

Yeah. A more accurate look would be to have other "costumes" for the sprite that are pointed up or down.

Not to mention is the pixels of the sprites not alighning with the backround pixels. They will always align, and without a single exception. None.

Miljörör för språkpartiet lol du menar väl språkrör för miljöpartiet lol

bd

32 bit >>

Super mario world, sonic rhe hedgehog...

All of those just fall under 8 bit these days. The kids wouldn't be able to tell the difference anyway.

Yeah, I suppose so. My wife always calls anything vaguely retro "8 bit", even 80s retro soundtrack music.

The SNES was the best 16bit console in terms of hardware by far. 1: Sound was clearly and unambiguously better with SNES because of more stereo channels. 2: SNES color pallet and simultaneous color limits were vastly larger than Genesis as well as sprite limits were much higher. 3: Way more buttons on the game controller and well designed for ease of access as well. The Genesis's CPU was considerably faster than the SNES, but this is a moot point. The SNESs video memory and sound peripherals were so efficient that the SNES did not require such a fast processor to match the performance of the Genesis, and even surpass it. Sega tried to market their faster Genesis processor in the early 90s calling it " blast processing ". It was a misleading sales gimmick.

Yes, great, but we talking about retro style games.

It obviously didn’t help that the Genesis had “16-BIT” blasted all over it.

I think the snes could, if you really wanted it to, run a very simplified version of minecraft. XD After all, it ran doom, and it could do 3d graphics and basic texture mapping with the right add-ons. Would look very little like the real thing though. It does sound like an amusing project to implement, come to think of it. XD

there was a virsion of minecraft called minecraft 4k, where the entire game was stored in just 4 kilobytes. the snes or even the nes could definitely run that

@Katarzyna Tomczak ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-gKpLUAhLjOc.html

kernelPANIC Because it is actually him. *mumbles about reading the discription*

The music however was made to closely emulate the NES's limitations and highly resembles Mega Man 1-6's sound front.

Hey, Retro City Rampage was ported to DOS, so who knows lol. With the appropriate tweaks (no parallax, etc.) it definitely could.

Julian Rey and yet no games on the NES looked like Shovel Knight. anyway, I take back what I said 2 years ago about hurting. I fear that people will hurt me

I think "8-bit" style games exist as nothing more than an aesthetic choice and to play on the nostalgia of old farts like me who remember their Commodore 64s from the first time round. What they mean to these young'uns I have no idea, but I'm guessing it means simple, easy and throwaway, the complete opposite of actual 8 bit games which took their cue from the arcades. Arcade games were designed to kill you quickly and frustratingly so you'd keep chucking in 10p pieces to beat your high score. Though back in the day arcade games were referred to as "coin ops" (short for coin operated) rather than arcade games as you'd find them in quite a few other places apart than arcades in the olden days (at least in the UK)

GeoNeilUK I've played my fair share on arcade machines... on emulators. Got fed up of pressing the "insert a coin" button to be honest. I know it's not an arcade game or even 2D, but have you read up on how much effort was put into Crash Bandicoot? They had to keep tweaking the levels and whatnot, let it build, often overnight, then see if it actually worked well on the PS1, mostly due to hardware limitations. Crazy stuff! And nowadays there are plenty of shoddily-programmed "8-bit" games which I can't even get running smoothly on mid/high-range hardware. Also, you seem to know your stuff (or at least sound like you do). Do you know of anywhere (preferably Scotland) with a real pinball machine? I should probably look in some arcade-type places but it's never really crossed my mind at the right time.

The King of Fighters is shifted to 3D models because the 2D Pixel Art was too expensive. The rendering of a pixel art game is kinda tricky because of the resolution and the way it works on different aspect ratios. It's not that it's easier or cheaper. It's an esthetic choice, the sad part is that the audience/the players see it as a limitation or as a retro-inspired visuals not as a choice like hand-drawn or cellshading...

GeoNeilUK I've never seen a single modern "8-bit" game that I've thought looked like it was on a retro console.

As an indie game developer, that's absolutely accurate. Good art is difficult and time consuming, and if it's too high-quality there are more code complications to take into account than when you can control each pixel almost individually. Some people try to emulate older games' art styles as a way to hide this facet or as an attempt to look better, as it worked for those games, whereas other developers try to adapt to larger color palettes and more precise rendering.

Then they really werent your friends in the first place.

I guess imaginary ones for the sake of jokes don't exist

It does actually. As a result, an 8-bit cpu has no issues dealing with 16-bit values but the programmer has to treat them as two 8-bit chunks instead and add the carry to the higher 8-bit part during additions for example. Thankfully though, pretty much every CPU detects carry/overflow so it's easy to do.

+za909returns so thats how the timer can go past 256!

Sorry for the long comment, Bedsitdweller, although it's actually not as long as it looks if you skim over the enclosed sample of code marked off with ">" signs. //I thought that 8 bit meant the processor could process 1 byte at a time and a 16 bit processor could do 2.// No, that's clock speed. Even the oldest game consoles process way more than 1 or 2 bytes at any given time, or else the screen would only light up 1 pixel at a time! (Even if "at a time" meant tiny fractions of a second, it would still cause a strobing effect if that were the case.) Bit architecture (8-bit, 16-bit, etc.) is the number of bits in a single character of source code. Despite popular belief, you could theoretically have today's graphics rendered from 8-bit character code in ASCII, and that's because a color on a pallet or an angle on a polygon or sprite (and so on and so forth) need NOT be limited to a single character on a sheet of source code! (However, doing so would tank your console in sales, precisely because most of the general public holds the incorrect belief that bit architecture is a measurement of graphical capability.) Have you ever seen a sheet of source code? Something that looks like this? > RossiterUp=c(11,2,3*(20/18),4*(20/18)) > RossiterUp [1] 11.000000 2.000000 3.333333 4.444444 > RossiterUpPerMinute=c(0.55,0.1,0.17,0.22) > RossiterUpPerMinute [1] 0.55 0.10 0.17 0.22 > RossiterDown=c(13,6,7,6) > RossiterDown [1] 13 6 7 6 > RossiterDownPerMinute=c(0.65,0.3,0.35,0.3) > RossiterDownPerMinute [1] 0.65 0.30 0.35 0.30 > CreeksideUp=c(9,10,4,14) > CreeksideUp [1] 9 10 4 14 > CreeksideUpPerMinute=c(0.45,0.5,0.2,0.7) > CreeksideUpPerMinute [1] 0.45 0.50 0.20 0.70 > CreeksideDown=c(4,3,10,9) > CreeksideDown [1] 4 3 10 9 > CreeksideDownPerMinute=c(0.2,0.15,0.5,0.45) > CreeksideDownPerMinute [1] 0.20 0.15 0.50 0.45 > CherryTreeUp=c(11,15,5,4) > CherryTreeUp [1] 11 15 5 4 > CherryTreeUpPerMinute=c(0.55,0.75,0.25,0.2) > CherryTreeUpPerMinute [1] 0.55 0.75 0.25 0.20 > CherryTreeDown=c(7,6,13,8) > CherryTreeDown [1] 7 6 13 8 > CherryTreeDownPerMinute=c(0.35,0.3,0.65,0.4) > CherryTreeDownPerMinute [1] 0.35 0.30 0.65 0.40 > > mean(RossiterUp) [1] 5.194444 > mean(RossiterUpPerMinute) [1] 0.26 > mean(RossiterDown) [1] 8 > mean(RossiterDownPerMinute) [1] 0.4 > mean(CreeksideUp) [1] 9.25 > mean(CreeksideUpPerMinute) [1] 0.4625 > mean(CreeksideDown) [1] 6.5 > mean(CreeksideDownPerMinute) [1] 0.325 > mean(CherryTreeUp) [1] 8.75 > mean(CherryTreeUpPerMinute) [1] 0.4375 > mean(CherryTreeDown) [1] 8.5 > mean(CherryTreeDownPerMinute) [1] 0.425 For the sake of intellectual honesty, this section of code is not *actually* from a video game or from the underlying engine of one. It is in fact from a session I spent in R (a specialized programming language for Biology-related statistics) for part of my Master's Thesis. Nevertheless, you get the idea. Let's consider the line "RossiterDownPerMinute=c(0.65,0.3,0.35,0.3)". In 8-bit code (which is the minimum for ASCII to operate, by the way), the capital R at the beginning of that line would be 01010010. The lowercase r at the end of the word "Rossiter" would be 01110010. The opening parenthetical in the Create Set function ("c(...)") would be 00101000. The closing parenthetical would be 00101001. In 16-bit code, 8 0's would be added to left of each: 0000000001010010, 0000000001110010, 0000000000101000, and 0000000000101001, respectively. The real advantage of higher bit architectures is not better or more complex graphics, but rather, the ability to incorporate higher-order Unicode characters in the underlying source code. A 64-bit character architecture can theoretically accommodate the first 18.4 *quintillion* characters, and the actual Unicode standard hasn't even been developed to anywhere near that number of characters! Again, though, most of the general public holds the incorrect belief that bit architecture is a measurement of graphical capability.

WillStrop2008 What the hell? You just throw Randoms things together. You are talking about character encoding. That is no real relation to the Bit System you are using, there are different standards with different bit usages, but that's it. How you explained it, we could easily use 8bit architecure for everything. But we also can allocate more RAM, save bigger numbers or even calculate more precisely. 3D Games wouldn't be possible how they are today if we would still use 8bit. Also alles new consoles are 64bit not 128bit.

MrDavibu Character encoding has *everything* to do with Bit Architecture. Bit Architecture, by definition, is the number of bits in each character of source code. Just on the off chance you thought "character encoding" specifically meant the visible characters you see on the screen (I've run into other people who thought that), No. That's not the nature of source code. Unicode is an extension of ASCII, and unlike raw ASCII, it extends into characters that require more than 8 bits to encode. While other (far less common) standards do exist, *all* of them require a consistent number of bits per character. (So no, I never said "everything" was possible with 8 bits per character. Not if you wish to encode foreign alphabets like Greek and Hebrew.) Critical question: Which aspect of game play mechanics or graphics is limited in principle to only a single character of source code? If the answer is "nothing," than you can see how the number of bits in a single character is irrelevant to those higher level aspects of game programming. Storing larger numbers can be accomplished by storing the same number in multiple addresses, as long as those addresses are consecutive. That technique can affect more precise calculation as well (via floating point values), but again, the consecutive addresses thing doesn't work for character encoding. Well, the DreamCast and GameCube were 128-bit, but later console generations may well have gone back down to 64-bit architecture for the very reason I explained above. Thanks for the heads up if they did. Anyway, keep in mind that we're talking about the length of a "word" in Binary, which, despite being called a "word," corresponds to a single letter of Source Code. Multiple "words" (letters) can be used consecutively to encode more complicated information.

actually, thats boolox

@@livelongandprosper70 shut up

Yes except sometimes it doesn't. When talking about an 8-bit processor yeah, when talking about an 8-bit console... maybe, and when talking about "8-bits game/graphics" all bets are out, probably means the programmer did the graphics himself and used that as an excuse lol Not saying that there aren't cool looking pixelated/low-res graphics out there.

@@trinidad17 Then 8bit-style design... but not 8bit.

Heh, nice to see I wasn't the only one to notice. And it looks you're a fellow Saint Seiya fan too!

Thank you for actually writing it down. I never heard of the game and heard him say "Lat Ma Luna". When I tried to google the name I found absolutely nothing at all which was frustrating as hell.

I don't laugh at any graphics based on their crudeness, just on poor artistic choices. Games have never been about faking the real world to me, so as long as it fits together in its own world and enables quality gameplay it's fine.

The majority of people want graphics and gameplay experience.

Or a loud minority. Loud minorities can be confused for the majority.

Not likely. Graphics technology evolves much much slower these days. Not that much happens in ten years anymore.

So Action 52 was a perfect game that wasn't buggy or glitchy?

PerfectTube12 donkey kong country, super mario, TES Arena, etc... i dont know action 52

Da.shArk87 It's an NES game that contained 52 games on one cartridge, and it was horrific; games filled with glitches, broken mechanics, terrible graphics, etc. Broken games, licensed or otherwise, did exist on the NES, but because they were broken, they were long forgotten. The games you mentioned were good games that people still talk about to this day and are getting re-released on Virtual Console and other platforms.

ohh maybe it was an indie studio that sold Action 52, many indie studios made games at that time.. I had a smiliar cardridge like you on my original NES with like 70 games only of olympic sports, it was quite fun.. i miss those games actually :)

Da.shArk87 there were TONS of game breaking glitches. this fantasy is a lie spread throughout the internet that is easily disproven. most roms you download have been patched. in fact, new glitches are found all the time as people go back and replay old games. you are just wrong.

When it comes to micro-processors it commonly refers to the word size, some 8-bit processors had 16 bit data buses, while still being able to operate on 8 bit words at a time. Said that is not as clear cut as some did have narrower ALUs, and some had instructions for working with 16-bit data.

Keep fighting that good fight, my man!

We don't need to know about your menstrual flow... And technically, Ps4 uses a 64 bit multi core processor. We have been in the 64 bit era since N64. N64 wins. MENSTRUAL FLOW.

NAAAAAAAAY

try making a game on android for android, oh yeah YOU CAN'T

I think the idea is you can make a game without needing a full time professional 3d animator.

RozumariSama I would say 16-bit and above is fine though, and plenty of those games look amazing, with a lot of talent and effort.

Alex Ander Most pixel art games are imitating 16bit/early 32bit generation, usually running at 256 or more colours. I don't see these pixel art games emulating the colour scheme of the ZX spectrum or the limited palette of the NES for instance :)

+RozumariSama Maybe they AREN'T all trying to mimic an 8-bit system's limitations... I think some want to imitate the Genesis or SNES type of bits (I forget if it's called 16 or 32 bit tbh) But I can see where your frustration comes from instead of them marking all pixel games as "retro", perhaps they should just say "pixel art style." Cause there's a difference between "retro" and "pixel art."

I would buy them.

so would I

What about the PC Engine? That was technically an 8bit system.

***** The PC engine had an 8 bit CPU but it also had 2 16 bit graphics processors, so it was an 8 bit system with 16 bit graphics

While the Atari ST was 16bit yet only could display 16 colours. The later MSX computers were 8bit and yet could display up to 256 colours at once. 16bit != better graphics automatically.

Couldn't 32-bit processors have used PAE to access more RAM, though? I heard that the main reason Intel was forced to go 64-bit on x86 was because of AMD.

@@jeremyandrews3292 Yeah addressing is usually not the same as the word size. 8-bit processors did commonly have 12 or 16-bit addressing, and some 16-bit processors had 20 or 24bit addressing available. x86 real mode has 20-bit addressing, and there are many other different modes in x86 itself, some with pagination including PAE that you mentioned. With PAE you have a virtual 32-bit address space for each process, but the OS has 64-bits page table entries but less than than can be used for addressing, as it stores flags and other housekeeping there, it's fairly complex and hierarchical. But long story short on 32-bit x86 you can have up to 64GB of RAM using PAE, which is one of the x86 native memory management modes. Aside from addressing, there is the whole data bus thing which can have a width on its own, the data being where the processor reads/writes the data it addressed using what we mentioned above, that also varies a lot depending on what architecture we're talking about, and also doesn't match the internal word size in some cases. So, unless you specify what you're talking about "bits" doesn't actually mean that much. But when we say 8-bit processor the common thing is to assume we're talking about word size, or at least the word size the processor exposes to the developer as some had 4-bit internal words... it can be complicated to say the least heh, but the common thing we we're talking about the processor is to assume it to be the word size.

Yeah, but unlike 8-bit, actual rape is still happening.

@@a.wadderphiltyr1559 I don't get what oil has to do with rape, but I'm sure someone out there enjoys your comment.

Then again, it's really interesting how old games work :)

And now 8bits dominating the NFT spaces. Even the old games are coming back like Faketown which is 8bits too. the old styles are coming back and its really good.