/vr/ - Retro Games

>>1420841
>considering the SNES' slow CPU

I've never noticed any slowdown in Pocky & Rocky, the fuck are you even talking about?

Man sega autists have brainwashed people into thinking that the SNES was this archaic ass slow cpu, while the genesis was this technological marvel that is unrivaled in speed by anything, when in reality the difference between the two is marginal at best

>>1420878
>final fight snes port
>slow as fuck when there's more than 3 sprites on the screen

>>1420896
And yet that has nothing to do with the the SNES's better performing CPU.

>>1420878

Doesn't even have blast processing, yo.

>>1420894
SNES was basically the RPG console, Nintendo had to cut corners because they already allocated all the console budget to the graphic design. The SNES was originally supposed to be backward compatible with the NES, but they scrapped that too.

>>1420896
>badly programmed game
>it's the system's fault

>>1420901
Correct, because the SNES had a worse performing CPU. The SNES port performs much worse.

>>1420902
No, it has a crazy varied library with games in multiple genres, this is also one of those genesis fanboys things they say but is utterly shit

>>1420904
>What is Ghouls n' Ghosts
>What is Turtles in Time

Shit man, the SNES was fucking slowdown city.

>>1420908
I've never experienced slowdown with either of these

>>1420906
I didn't say the SNES was bad, granted, it did have some good games. But you can't deny that it was horribly crippled by slowdowns and limitations imposed by its crappy CPU.

>>1420910
Why don't you actually do some research before spewing shit

>>1420894

lol this is completely untrue

>>1420910
It wasn't, it has a ton of amazing games that I love, a couple of moments of slowdown in hectic games isn't going to ruin the whole console for me, what am I, a segautist?

Snes' slowdown is the only angle genesis fans have to work against it, and man do they ride that dick hard

>>1420898
Blast processing is DMA. It has DMA.

>>1420905
Which would be a valid complaint if you also bitched about the even worse performing CPU on the genesis. But you don't.
If anything you could bitch about the variable access mode if improperly used.

>>1420894
>when in reality the difference between the two is marginal at best

2/10. Either that or you're just retarded.

Phalanx

>>1420918
>the even worse performing CPU on the genesis.

Lol, this is what Nintendrones actually believe.

>>>/v/

>>1420921
snes can do more colors, genesis does sprites faster. that's about it.

>>1420923
>Lol, this is what Nintendrones actually believe.
Reality?
I love the genesis, I play it more than the SNES, but calling a faster CPU slower is just nuts.

>>1420918
>Blast processing is DMA.

[citation needed]

Why are you even talking out of your ass when on this slow board, you have all the time in the world to do research on this subject you clearly know nothing about.

>>1420925

in practice this meant that genesis games had more detail, could have more things on screen at once, and could have smoother more fluid animation

20 years later and you fags are still falling for that "blast processing" bullshit.

>>1420928
The SNES had a slower CPU. CPU means Central Processing Unit. The SNES' 65816 was a result of Nintendo cutting corners.

Super Mario World
Super Metroid
Yoshi's Island
F-Zero

>>1420936

it was a trade off for better sound and more colors, kinda like how the n64 has no dedicated spu and most games have shit sound to compensate for graphics

>>1420934
Blast processing was just the Genesis' 7mhz 68000 processor, which was faster than the consoles of its time. Games on the Genesis didn't suffer SNES-style slowdowns. They are two different consoles with strengths and weaknesses.

>>1420938
Those were exclusives that took advantage of the SNES' primary weakness (slow CPU), ports were often worse on the SNES because the CPU was complete shit and game developers wanted fast sprites moving everywhere. There's a reason the SNES only has two decent shmups.

Also, Sunset Riders showed how fucking bad the SNES was at handling multiple sprites.

>>1420951

sunset riders is better on the snes though

>>1420951
OP asked for fast games and I provided them. Take your shitty Genesis pissing match elsewhere.

>>1420959

>>1420951
>There's a reason the SNES only has two decent shmups.
lol, and which ones would those be?

>>1420942
Compared to the Playstation, which was more well-rounded in its design, the N64 actually had a lot of cut corners. The board was simplistic as fuck and replied on the CPU to handle most things. The entire architecture was a clusterfuck that 3rd party developers complained about and is the reason we don't have decent N64 emulation to this day.

But, at least the N64 was cheaper. Every kid could afford it.

This is a good tl;dr: http://en.wikipedia.org/wiki/Nintendo_64_programming_characteristics

>>1420945
No it wasn't. It was DMA.
And no, the processor had a faster clockspeed but lower IPC. Nothing on the 6800K takes less than 10 cycles per instruction and can even be as high as 20 to 30.
The 65c816 Typically has low cycles in the 4 cycles per instruction. It's impossible to say how it'll perform in any one specific case without looking at that case, but general mips the 68K is 80% of the 65c816. In general the SNES was faster but also had limited versatility in sprites.

>>1420970
The N64 also had actual 3D. Which you know is pretty fucking good for 'cutting corners.'
Every 3D game looks better for it.

>>1420970
This. Nintendo's marketing strategy was always about making the cheaper, budget console that parents are more likely to buy, and they'd succeed through quantity over quality and developing games. The SNES, N64, and even the Wii were all underpowered on release.

You'd have to be a retard if you haven't realized this yet.

>>1420975
The N64 was 1996 though, wasn't it.

>Every 3D game looks better for it.

Lol no, we all know how inferior Megaman 64 was.

>>1420976

then why were the games 3x as expensive

>>1420973
>mips the 68K is 80% of the 65c816.

[citation needed]

The inferiority of the SNES's CPU (which was a result of including a better GPU) was the primary reason Nintendo needed to add a math coprocessor to it. And even then, it didn't get rid of the famous SNES slowdowns. It had prettier graphics, but that's about as far as the 65c816 went.

>>1420976
>that parents are more likely to buy,
But parents weren't more likely to buy it. PSX sold nearly 3x as much. Unless you meant more likely to buy than not buy anything at all, of course if they were intending on buying a console they'd be choosing from whatever is on the market anyway.

Wild Guns

>>1420983
Cartridges and lack of Sony magic to afford a disc drive for each console. And even then, you had Blockbuster for N64 games.

>>1420982
True, it wasn't the case for the N64. It ended up being more expensive and harder to develop for, compared to the PS1.

>>1420989

my parents bought me a n64 but i got a ps1 in 1999 for $100

>>1420893
>(The game would slow down considerably every time there was a flying enemy onscreen.)

>slow down
>slowdown

>>1420980
>Lol no, we all know how inferior Megaman 64 was.
And yet the 3D space was handled far better. The lines are straight and not shakey as shit as typical of the PSX which has no Z buffer.

>>1420984
>
The inferiority of the SNES's CPU (which was a result of including a better GPU) was the primary reason Nintendo needed to add a math coprocessor to it. And even then, it didn't get rid of the famous SNES slowdowns. It had prettier graphics, but that's about as far as the 65c816 went.

[citation needed]

>>1420997
My parents only bought me an SNES because my family is poor.

>>1420973
The level of cluelessness in this thread is unreal. I can tell you've never tried to program for the m68k or the 65816. Let's start with a crucial difference for those of us actually writing assembly: register count. Also, instructions that don't hit memory more than once typically take 4 clocks; where you got 10 I don't even know, it's bizarre.

>>1421002
I have never experienced what you're talking about. You're talking about the Tengus, right? Or those bubble crabs in stage 2? Never had any slowdown involving them.

>>1421029
>The level of cluelessness in this thread is unreal.

This is what happens when Nintendrones try to convince you that their budget console is better in every single way than any other game system at the time.

>Now they're gonna tell us the SNES was more powerful than the CD-i

The SNES had some good games and those games had some good graphics. I don't think anyone contests this. The fact is, the SNES had a slow CPU that developers had to work around with. That was the SNES' achilles heel, so to speak.

>>1421040
That's still arguing theoretical performance, and it doesn't even say how much better which chip is at what.

>>1421037
Indeed. There's also no denying that the SNES had a lot of excellent special-purpose hardware, which compensated for the weak CPU.

>>1421040
Yeah, code density is a huge problem for the m68k, probably the biggest problem with it.

>>1421047
>SNES had a lot of excellent special-purpose hardware, which compensated for the weak CPU

For most SNES-type games, this is true. Other games, on the other hand, were not designed with a slow CPU in mind, so the Genesis was a better system for ports. The SNES was a better system for home-grown games.

>>1421049
He didn't even take instruction sets or cycles into account either. It's as if he actually believes a processor's performance is its bandwidth.

>>1421049
You can only argue theoretical performance you two bit dipshit.
The only way to not is to compare specific programs and the programs won't be the same. It will always be a theoretical comparison.
But in practical terms for MIPS the 65c816 is faster, theoretically.

>>1421040
Ok, what people may not be understanding from this is that from this comparison, the m68k is clearly the faster processor, but the 65816 isn't a complete piece of shit. The bandwidth advantage stated is mostly advantageous for code density, not speed. Meanwhile, register count makes a huge practical difference for performance.

>>1421058
Except it is. Don't need to read the thread, I've seen the MIPS rated and nothing in this thread disproved that at all.

>>1421054
>the 65c816 is faster, theoretically

No it is not. Did you even read the thread? For all practical cases, the SNES CPU is much slower.

>>1421061
... have you even written a line of assembly code in your life? There's a reason we don't benchmark computers in terms of MIPS.

>>1421061
Read >>1421059

The SNES CPU would be better if the nature of programming actually meant that bandwidth is the #1 important aspect in determining performance.

>hurr snes can do the same 2 operations over and over again faster

So can an oscillator.

The SNES CPU was and is crippled, it shows in its games and it proves that its bandwidth advantage doesn't translate well into the real world. The 68K was rarely bandwidth limited in the first place, while the SNES is constantly being shown as limited. It's the reason why Nintendo released so many games with add-on chips.

Again, nobody's saying the SNES was a bad system, it had good graphics but the CPU was slow. Slow CPUs can be worked around with, by slapping effects on, but as a matter of fact, most games are WRITTEN to be mostly executed on a CPU. That is what made the SNES suffer. The AVERAGE GAME wasn't designed with the SNES in mind, the SNES couldn't handle high sprite games with lots of shit moving everywhere, which was due to the arcade craze in the 90s. Arcade machines generally had fast CPUs and that was it. The SNES lacks that, and (rightly) relies on trickery to achieve good looking games.

I could keep repeating the same shit over and over, but I don't think Nintendrones like you can do anything but parrot the same technical GameFAQs post written by a well-intended and knowledgeable 12 year old (who lacks historical experience) that you managed to come across.

>>1421072
You show me the specs where the 68K isn't doing 12 cycle instructions at least 50% of the time. Then maybe you'll have a point. But since that's not reality, you can fuck off nintendrone.

What we need is to have the strongest assembly programmers from each side implement all the benchmarks from benchmarksgame.alioth.debian.org. Might need to make the challenges easier so such old systems can actually complete them.

Overall winner should be the system with the most individual wins, not total or average time, because games are written taking into account the strengths and weaknesses of each system.

>>1421081
That's really the only real way to compare them.

>>1421081
If you want to get even more practical, just take into account the slowdowns of various arcade games that got ported to SNES, compared to the Genesis.

That should be all.

>>1421008
N64's RDP z-buffer was notoriously slow and hard to program for. It was occasionally disabled and a software depth buffer was used on the CPU instead.

>>1421084

there weren't any good home ports of arcade games until the saturn

genesis ports usually looked better graphically but had ear raping sound

some snes ports had added content but the genesis port did not

>>1421104
>there weren't any good home ports of arcade games until the saturn
R-Type for the TG-16 says hi.

Neo Geo games got better ports on the Genesis because they both had fast CPUs to handle the crunching and large sprite sizes/count.

Pocky and Rocky was basically an anomaly for the SNES, considering I don't see any game mentioned that fits the bill.

>>1421104
That's exactly why it's a good practical test. To see how much better the Genesis was at handling a high-end task than the SNES. Sure, the Genesis versions of Street Fighter or Samurai Shodown weren't comparable to the originals, but they were better than the on the slowdown-ridden SNES.

Basically, if it didn't involve SuperFX, Mode 7 trickery or other Nintendo tricks, it simply didn't look as good as the average game.

>>1421104
Midnight Resistance is better (including sound) on Genesis.

>>1421104
Well no shit, this was the mid-90s, you literally had to have an arcade to get an arcade version. We had to settle with Genesis and SNES ports until we either could afford an AES, or until Pentium III got cheap enough to buy and run MAME on.

So basically, your best bet of getting arcade games and ports at home was buying a Genesis until around 2000 or so. Nintendrones don't want to admit this, because the only games they know are first-party Nintendo games that use specific, never-seen-before features that relied on coprocessors and proprietary techniques, while other games were just written for CPUs - SNES's weak point.

Genesis and Super Nintendo each had different types of games. The SNES wasn't better in every way.

Surprise, surprise.

I'm actually glad the SNES had a shit CPU, Nintendo invented new genres and new kinds of gameplay over the early 90s. It would've been boring if we just had only CPU-bound games during that era.

I mean shit, Pilotwings didn't have the best graphics, but with all of that trickery mashed together, it had a unique style I've never seen before. Imagine opening up Pandora's Box and making it so CPUs never got exponentially better. We'd be coming up with more and more DIFFERENT looking games with various add-on chips, and pumping in millions of dollars to (painfully) develop games this way. Sort of like Cthulhu, trying to imagine shit you're not going to understand.

>>1421137

snes had way more quality games, it wasn't competitive there

the best made genesis games were made in the last 1-2 years of its life

>>1421149
It's about the same, except SNES had way more shovelware bullshit titles and horrendous ugly, horrible 3rd party games. And not to mention the influx of generic JRPGs.

I mean, if the only games you've played from the 16-bit era are Final Fantasy VI and LttP, I can entirely understand why you believe the SNES had "more" games.

>>1421157

nice ad-hoheim bro

>>1421160
Nice lack of examples faggot.

>>1421164

who could forget those great first party genesis games like alex kidd, spellcaster and revenge of shinobi LOL

>>1421165
>Mario Is Missing
>Winter Gold (lol shitty desperate Super FX game that Nintendo actually developed topkek and tried to forget)
>Super Play Action Football
>Yoshi's Island
>All of the shit SNES games that are utterly raped with slowdown mentioned in this thread

Congratulations, you are legally retarded. You can't even succeed with your own moronic bullshit.

>>1421173

mega drive shmumps look like vomit at least super aleste & axelay look good and don't have sound effects that sound like dry farts

>>1421178
>thunder force

Wow holy shit you're an idiot. I can't believe you actually cherry picked a screenshot from a THUNDER FORCE game, of all the choices you could've picked. Hell, you were too stupid to even use Toejam and Earl, but whatever. Anyway, here's a YouTube video for you, kid.

https://www.youtube.com/watch?v=fL81XjT_xxQ

Have fun with your 16-bit JRPGs. Maybe one day you can leave the cage of Nintendium you've built for yourself.

>>1421178
>the game you like is shit, it doesn't have the Nintendo Seal of Quality on it

>>1421185

even compile new genesis was a failure didn't even bother remaking this game for it or the saturn and opted for snes instead

>>1421196

https://www.youtube.com/watch?v=hEpTUXH-Vn0

No 2player though.

Could the SNES handle RECCA?

>slowdown minimal
>SNES

Pretty sure if you want games that don't involve watching slow moving sprites shuffle around a screen while awful midi trumpets blare, you should buy a Genesis. Or an NES

I dislike the Nintendoes-what-segdon't bickering in the thread; I'll try to answer the OPs post of "awesome action games on the SNES."

The Natsume games on the SNES are probably their best work, OP. Pocky and Rocky 1 and 2, The Ninja Warriors Again (play the JPN version!), Wild Guns, Mighty Morphin' Power Rangers (ok maybe this one is just for me) - all display a competency for action games that many SNES developers couldn't match (imo).

Games like Majyuuou/King of Demons, Hagane (good luck finding that one in the wild) are also very "genesis-esque" with their emphasis on speed and fast action.

Konami-released games are also up-to-par on the system, with Cybernator improving on the Genesis' Target Earth in every conceivable way, and Sunset Riders and TMNT being very respectable and fun ports (though these latter two do contain their share of slowdown).

>>1422754
>awful midi trumpets

That's one of my favorite things about SNES music. ;_;

I think that the SNES was a great single-player system.

>>1422763
Thanks for this post, I'll check them out. I'm interested in Ninja Warriors Again (JP) the most. I'll also add NBA Jam to that list, because it's both multiplayer and fast. (Turbo shoes, yo)

>>1422754
Or an AES. (You'll also need to shell out another $200 for the 161-in-1 and a converter)

>>1422763
>I dislike the Nintendoes-what-segdon't bickering in the thread;

It's because the SNES fanboys got pushed out of /v/ and onto /vr/. And believe me, I really do mean SNES fanboys.

>>1422754
Indeed, the action NES games still hold well today. I should buy a flash cart and an NES clone with RGB/SCART output.

>>1422735
I don't think the SNES could handle HALF of what was happening on screen. So much shit at a fast pace.

>>1422949
I think it is bull, /v/ has a heavy Nintendo bias.

The fuck is Pocky Pocky? That's Yokai for the PC Engine.

>>1422964
SNES could handle Recca if they put the NES CPU on the cartridge.

In 16-bit? Then they'd have to slap the Genesis CPU onto the cartridge.

The fuck is Pocky and Rocky? That's KiKi KaiKai for the PC Engine.

>tfw american games

>>1422974
Weeb faggot.

>>1422974

>>1422979
I've never heard of Pocky and Rocky, I had a TurboGrafx and Sega consoles growing up not Nintendo.

>tfw Hudson is dead

>>1422982
>tfw vidya is dead

Merging into Konami was for the best. The developers could pull an SNK if they wanted to, though. As in, leaving the company and making a new one.

>>1421029
He also fails to account for the fact that the SNES doesn't actually run at its cited speed of 3.58 megahertz -- it drops to 3/4 of that whenever it does anything with RAM or ROM. Some cartridges had faster ROM that didn't have this problem (though RAM always did), but it seems like most used the slower kind. I don't really know for sure though.

The SNES's CPU is actually underrated but it was still a pain in the ass in many ways, especially when combined with the system's other quirks and bottlenecks. And SNES games definitely slowed down more and had less onscreen, even though both systems slowed down plenty (just look at Thunder Force IV).

>>1423058
>it drops to 3/4 of that whenever it does anything with RAM or ROM. Some cartridges had faster ROM

Does emulation account for this, or are we always playing at 100% speed?

>>1423028

I dunno, man. After playing Heavenly Guardian (same developers) it's probably best that P&R stays dead and buried.

>>1422964
The SNES's CPU is twice as fast as the one in the NES, and there is no magical slowdown sauce that would prevent it from running that game. Really, NES games slow down with a lot less going on, you just don't notice it because the rest of the game is so primitive and your expectations are lower. Recca is exceptional, and the same exceptional things could be done on SNES -- but they usually weren't because programmers aren't superhuman and they don't know how every detail of the game is going to end up working as they're writing it. The Genesis gives you some nice headroom in that regard, which was great as a practical matter, but technically the SNES has enough CPU resource to run most Genesis games if you made them as optimized as Recca. It's just a lot more painful to do it.

I'm pretty sure TG-16/PCE games actually have the least slowdown of all; not sure why people are always talking up the Genesis in that regard.

>>1423081
Genesis had more Neo Geo arcade ports and games that pushed the hardware compared to the PC Engine. You can't be completely blind when taking slowdowns into consideration. Yes, Thunder Force IV had slowdowns on the stages, but it also spammed a buttload of effects. It was a tradeoff.

>>1423075
If you pushed the hardware of the Genesis, then the SNES definitely would not be able to run it. Practically speaking, nobody's going to spend billions of dollars porting a game to 1/2 the hardware. It's pretty moot.

>>1423058
It was mentioned in the thread.

>>1423061
Yes and no.
Emulation depends on the emulator and how it's emulating.

>still arguing about the merits of SNES vs. Genesis architecture

The console wars were 20 years ago... just let it go guys. Just let it go.

>>1424165
No anon. They continue to this day because somehow people actually bought Xbones.

>build your product identity and mascots arount BEING FAST

>3D consoles roll around
>you have trapped yourself into having to make things like Sonic 3D Blast and Sonic Adventure

>>1424217
>Sonic Adventure

Bro I don't even like sanic hegehog and I thought SA was good.

Star Fox 2 had multiplayer IIRC.

>>1420893
There's slowdown in the first level topkek. When you make your way just before the first boss, the game runs at like 20% speed (which makes it a lot easier to just dash through the entire thing)

>>1422754
>implying genesis music wasn't trash

>>1425059
>doesn't like chiptunes

>>>/v/

>>1425078
>thinks genesis had better music than SNES

>>1425091
>can't listen to chiptunes unless it uses samples

>>1424797
That was because a lot of the dirtywork was done by a specialty chip. On it's own, the SNES can't handle multiple things on screen to save it's own ass and the NES was just as bad. Which is why it had so few Shmups and Run & Gun games

>>1425748
>implying stupid shit like a big dumb idiot.
The SNES had plenty of shmups and run 'n' guns.

Umm, didn't the SNES allow cart chips to have their own CPUs?

>>1425776
Nope.

>>1426065
Yes they did. SA1 and SuperFX are the two most notable examples, and one obscure Japanese Shogi title has an ARM3 in it.

>>1426584
>and one obscure Japanese Shogi title has an ARM3 in it.

Whoa, this sounds badass. Link?

>>1426584
Wrong. The SA1 and the SuperFX are not CPUs.

>slowdown-minimal games for the SNES

Fat chance, should have went with blast processing.

>>1426661
The SA1 is a 65816 just like the CPU in the SNES, just clocked several times higher. The SuperFX isn't really a CPU though; it's a DSP.

>>1426619
This thing.

http://en.wikipedia.org/wiki/List_of_Super_NES_enhancement_chips#ST018

Nobody mentioned smash tv?

Why didn't they make any Pocky & Rocky games that was like the first one?

You could tell they obviously outsourced Pocky & Rocky 2, the game had more slowdowns and the gameplay was nothing alike.

Wait, people are honestly blaming all SNES slowdown on bad programming? It is fact that the Genesis handled action better than the SNES. I'm not inciting a flame war since I'm more of an SNES guy but try comparing SNES shmups to Genesis sometime. Gradius III has some of the most hilarious slow down of all time.

>>1429752
It's always a mix of things. For a hyper-optimized game, the SNES is well past the point of diminishing returns in terms of what CPU resource can do for you in a sprite-based game. Given realistic constraints on devs of the time though, the Genesis's arguably minor edge with its CPU became a much bigger deal.

Game engines do all sorts of stuff, dev teams were on tight schedules, and flexibility in the game dev process can be key to making a good game that isn't too limited or repetitive. The Genesis CPU was nicer to work with, so devs had a lot of freedom to do whatever and their games could still run well, but on the SNES it could take a lot more planning and optimizing to make a CPU-intensive game. This wasn't always possible or desirable, so SNES devs tended to set more modest targets in terms of what the CPU would have to handle, and made less action-packed games designed to handle fewer routines at once.

Also, a little bit of slowdown was considered tolerable for that time, and devs took that into account -- Gunstar Heroes has some. However, on the SNES the standard was lower (just like the graphical standard was lower on Genesis) so they didn't sweat it too much rather than cutting down on the game even more.

>>1429752
I'd say most SNES games were programmed rather well, because they were done by major companies and using assembly when necessary.

We could've had the SNES's shit processor perform better if the companies spent TWICE the amount of development time for EACH GAME optimizing everything like Recca.

And even then, Genesis was much easier to code for in the first place, because the hardware gave a lot of leeway.

>>1429752
More or less my thoughts. Good games will try and fit into the system's limitations, much like good Genesis music.

>>1427767
>>1427767
>>1427767
http://www.youtube.com/watch?v=4AapB7dW3HA

>>1430059
>all dat blood

Must've been before ESRB.