• Hey everyone, staff have documented a list of banned content and subject matter that we feel are not consistent with site values, and don't make sense to host discussion of on Famiboards. This list (and the relevant reasoning per item) is viewable here.
  • Furukawa Speaks! We discuss the announcement of the Nintendo Switch Successor and our June Direct Predictions on the new episode of the Famiboards Discussion Club! Check it out here!

StarTopic Future Nintendo Hardware & Technology Speculation & Discussion |ST| (New Staff Post, Please read)

Reminds me of this rather old photo:

2486940-0248877224-ChsSw.png


We’ll never see the leaps we used to see ever again. It’s only when you start skipping generations do you really see the differences now.

There’s definitely diminishing returns but that image is a bit of a false representation. It looks like the 6000 mesh is the base so obviously 60000 mesh won’t look any different. Like if you resize an image to be bigger. It won’t add in additional details.
 
I think diminishing returns are real... Even though each generation I say things can't get much better and there is SOMETHING that ends up amazing me... It's a lot harder to impress me each time.


However this (really old) picture doesn't exactly represent the way the industry has actually gone (even at the time)... it's sort of misleading.
1. it's only talking about polygons, which a long time ago was a majorly important part of performance, and is basically not a problem anymore.
2. that last model should have the jacket fully modelled for example as well as the bow tie and the hair should be tons of planes, that stuff was happening even in the 360 era, take a look at Assassin's Creed for example.
c92d1e60d6f4026c98b2301624448b6a.jpg

Edward had a poly count of just under 40,000 for the whole character. And there's a lot more detail in it that the bust in the picture.

Not to mention the realization of new rendering technologies... The returns are diminishing... It's just that we got REALLY good at faking it, (lighting is a good example here) now that hardware has caught up and we can do it a real way (ray tracing) the change is huge technically ... but visually sometimes you have to strain to see the differences.

I think there's still plenty of things that can be improved, things like actual hair, actual fluid simulation, better cloth sim so your character actually is wearing clothes instead of the clothes being skinned to the rig. Better real time volumetrics for fog and clouds and weather stuff. Again we're pretty good at faking this stuff ... I'm excited to see how we tackle these aspects in the future.

Also I hope I didn't trounce your post, just adding to the conversation and pointing out some things I noticed that's all.
That Mario Galaxy (?) vs Mario Odessey coin comparison was eye opening. They actually used to spend more triangles to get detail games today can get from a texture or post processing effect.
 
And we were still learning software techniques too. There is a reason that you seemed to have generational leaps within the generation. The software side of real time graphics was still being learned and developed. You can see that with the guy who has been rebuilding Super Mario 64 using modern techniques. He's pushing vastly more frames and polygons than the original game, on original hardware, just by applying 25 years of software know-how to the thing.

If we're thinking of the same video, it is a great example of software optimization and how development technique evolves. Though it's a bit misleading since if I understand it correctly, he used the memory expansion pack to implement those techniques which is not possible for the developer at the time. I wish he did a video of what's possible without the expansion pack. Still a good showcase of how RAM dependent we are now and why we leaped from PS3 having 512mb of RAM to PS4 having 8gb.
 
Last edited:
That Mario Galaxy (?) vs Mario Odessey coin comparison was eye opening. They actually used to spend more triangles to get detail games today can get from a texture or post processing effect.
and now it's getting to the point where faking it is starting to get more costly, lol
 
I'm not opposed to regional pricing, so that brings up two "problems":

The tax inclusion means that Japanese and (even moreso) European pricing will look inflated. It's 23% where I live. Nintendo can't just eat that! The exchange rate sure can't!

The worldwide audience will be paying in their local currencies, even the relatively pessimistic on price fully admit that the JPY price will have to be relatively lower.

But the US market isn't a non-rich country. It has higher incomes than many countries but the market in the US consistently displays a much higher degree of price sensitivity.

Countries dependent on the USD in place of a local currencies tend to be either essentially marketed in as part of the US for purposes of product prices, or entirely outside of Nintendo's target market. Places outside North America where you pay in USD over the odds for an imported console are not a target audience, as unfortunate as that might be for the many people in disadvantaged regions who like to play games!

In countries where a $450 pricetag is unachievable for so many people it cannot penetrate the market, a $400 pricetag wouldn't be a salve. Nintendo either has to engage in aggressive regional pricing or simply not launch it there until the price comes down or the country's market develops. This is something they already engage in.

When people bring up USD price expectations, they do so in the context of selling the console in the US without making a loss, ideally, making a profit.

I think $399 is entirely POSSIBLE, but I think it's also true to say $449.99 would make not making a loss easier, while also still undercutting the competition (at least the top two tiers of each's stratified strategy).
Really, the biggest hurdle Nintendo will face for the Switch 2 will be pricing, since Xbox and PlayStation has been having difficulty of pricing and their history of price cuts has comes to an end, since the PS5 didn’t get a price cut when the PS5 slim released.

Like 400$ and 450€ seems like a huge possibility, since a the PS5 in most EU regions is 550€.

Like I can see Nintendo doing way more bundles for the Switch 2, to hide the cost of the console, similar to Sony strategy of making more bundles, because the PS5 can’t receive a price cut.

Like keeping the console at the 400$-480$ is key.

And I truly believe that we’ll get a console pack in and BC will play a huge role in Nintendo marketing, similar to the PS5, since that boy was a PS4 machine in the early years.

And bringing back Nintendo select and letting it be 20$-40$ in store or digital can be huge, especially with the promise of older games supporting 1440p, HDR and fps boost.
 
Really, the biggest hurdle Nintendo will face for the Switch 2 will be pricing, since Xbox and PlayStation has been having difficulty of pricing and their history of price cuts has comes to an end, since the PS5 didn’t get a price cut when the PS5 slim released.

Like 400$ and 450€ seems like a huge possibility, since a the PS5 in most EU regions is 550€.

Like I can see Nintendo doing way more bundles for the Switch 2, to hide the cost of the console, similar to Sony strategy of making more bundles, because the PS5 can’t receive a price cut.

Like keeping the console at the 400$-480$ is key.

And I truly believe that we’ll get a console pack in and BC will play a huge role in Nintendo marketing, similar to the PS5, since that boy was a PS4 machine in the early years.

And bringing back Nintendo select and letting it be 20$-40$ in store or digital can be huge, especially with the promise of older games supporting 1440p, HDR and fps boost.
Hiding some of the cost in hardware only happened in the WIIU era, however that time had the high priced 3DS as a profitable supplement. It should be noted that Sony and Microsoft are able to sell consoles in bundles or outright at a loss because they are integrated companies, Sony did it in the ps3 era to promote their multimedia strategy, Microsoft has been in the gaming industry since 2001 to bring the hardware/service ecosystem of the pc to consoles, and only Nintendo is monetizing purely on gaming hardware & software, I really don't think they would choose to subsidize the cost. Of course I still think $399 is enough to make a profit on each switch2, considering it's really coming out in 2025 and not 2023.
 
0
My biggest concern with prices atm is that both sides in this country seem to wanna out-tariff the other side tbh

It's silly...no one is making game consoles in Ohio or Michigan...
 
Last edited:
I’m convinced Nintendo Selects will make a comeback, but digital only

Hmm, I've been against this idea for a while as Nintendo hates price drops, but selects for digital only actually makes a lot of sense for Nintendo to push digital.
Agree it will be digital only. But i am thinking They may do 'Nintendo selects voucher' program for NSO subscribers where the vouchers are specifically good for a selection of back catalog games than works out to a discount if applied. But the games will remain regular priced on eshop and retail.
 
Agree it will be digital only. But i am thinking They may do 'Nintendo selects voucher' program for NSO subscribers where the vouchers are specifically good for a selection of back catalog games than works out to a discount if applied. But the games will remain regular priced on eshop and retail.
Maybe Nintendo select will be exclusive for NSO members, that gives Switch games a 60-80% discount, with improvements on the game, which would be hitting 3 birds with one Stone.

  • Developers will be able to go back and improve games, with resolution and fps, most likely newer and younger workers would to it, to get the hang of the engine.
  • Selling those games at NSO for discounted price and would entice people to subscribe.
  • Previous physical and digital owners of the game, gets the upgraded free, since the developers are improving the game, for NSO.
 
I saw this comment about Monster Hunter Wilds where the Director states wanting to push hardware to the max with his game:

"To be honest, it was always clear in my mind that was the direction I wanted to take it," he told me. "Any Monster Hunter game where I'm director is always going to be focusing on the ecosystem element. As the hardware generations we're working on get more and more powerful, I want to use the specs of the hardware to the max in order to depict as convincing a living, breathing world as I possibly can."

Source:


Makes me wonder if the Switch 2 will get this game or not. What do you guys think? I thought for sure this game would come to Switch 2 but now I'm not feeling as confident if the goal when developing the game was to really push hardware. I guess it really depends on how powerful the Switch 2 really is. Still a lot of unknown in that area.
 
I saw this comment about Monster Hunter Wilds where the Director states wanting to push hardware to the max with his game:



Source:


Makes me wonder if the Switch 2 will get this game or not. What do you guys think? I thought for sure this game would come to Switch 2 but now I'm not feeling as confident if the goal when developing the game was to really push hardware. I guess it really depends on how powerful the Switch 2 really is. Still a lot of unknown in that area.
At the very least this tells me they're probably not thinking about Switch 2 as they make this game. A late port, if possible, maybe.
 
I saw this comment about Monster Hunter Wilds where the Director states wanting to push hardware to the max with his game:



Source:


Makes me wonder if the Switch 2 will get this game or not. What do you guys think? I thought for sure this game would come to Switch 2 but now I'm not feeling as confident if the goal when developing the game was to really push hardware. I guess it really depends on how powerful the Switch 2 really is. Still a lot of unknown in that area.
same as I always thought, it's not going to be hardware that prevents games from coming, but dev desire
 
Makes me wonder if the Switch 2 will get this game or not. What do you guys think? I thought for sure this game would come to Switch 2 but now I'm not feeling as confident if the goal when developing the game was to really push hardware. I guess it really depends on how powerful the Switch 2 really is. Still a lot of unknown in that area.
What hardware was he pushing? That's the key element. You can push a Sega Genesis, you can push a Series X - what's the target platform that he's trying to stretch to its limits?

Trying to guess the performance of a theoretical port of an unreleased game on unrevealed hardware is a fool's errand. I can make guesses based on the graphical presentation, but it doesn't sound like he's talking about the graphical presentation. It sounds like he's talking about the CPU-bound simulation. That's a much harder guess.
 
I saw this comment about Monster Hunter Wilds where the Director states wanting to push hardware to the max with his game:



Source:


Makes me wonder if the Switch 2 will get this game or not. What do you guys think? I thought for sure this game would come to Switch 2 but now I'm not feeling as confident if the goal when developing the game was to really push hardware. I guess it really depends on how powerful the Switch 2 really is. Still a lot of unknown in that area.
Nothing new here. If they really wanted to push hardware, I assume PS5, they either delayed the Series S version or outright not make that version. In my opinion, I've never believe when devs say they are pushing hardware because its a meaningless statement used for marketing. So in the end this says nothing about a switch 2 version
 
I saw this comment about Monster Hunter Wilds where the Director states wanting to push hardware to the max with his game:



Source:


Makes me wonder if the Switch 2 will get this game or not. What do you guys think? I thought for sure this game would come to Switch 2 but now I'm not feeling as confident if the goal when developing the game was to really push hardware. I guess it really depends on how powerful the Switch 2 really is. Still a lot of unknown in that area.
Yup, it’s to be expected, since I can see Capcom focusing more on PS5 and PC in the foreseeable future, meanwhile Switch 2 will receive last gen port.

But ,,pushing hardware to the max’’. Is something I can’t really see, since the game looks good, but not ground breaking. Like DD2 is sadly poorly optimised, I can see the same for MH, only because of RE engine, since it seems it doesn’t work the best in open world spaces, compared to RE and DMC5.
 
What hardware was he pushing? That's the key element. You can push a Sega Genesis, you can push a Series X - what's the target platform that he's trying to stretch to its limits?

Trying to guess the performance of a theoretical port of an unreleased game on unrevealed hardware is a fool's errand. I can make guesses based on the graphical presentation, but it doesn't sound like he's talking about the graphical presentation. It sounds like he's talking about the CPU-bound simulation. That's a much harder guess.
current gen hardware, is that not oblious? Capcom want to push Xbox Series X/S and PS5 to the limit
 
current gen hardware, is that not oblious? Capcom want to push Xbox Series X/S and PS5 to the limit
there's a million different little limitations in hardware. DD2 pushed the CPU hard (possibly to unreasonable limits), and other games push the GPU hard. Wilds being an open world game, it's a question which limit they go for
 
I will say that the very negative reaction to Dragon's Dogma 2's performance will probably help the Switch 2 by causing devs to avoid taxing the CPU in the future, lol.

Probably not good for game design variety though.
 
I will say that the very negative reaction to Dragon's Dogma 2's performance will probably help the Switch 2 by causing devs to avoid taxing the CPU in the future, lol.

Probably not good for game design variety though.
Despite its negativity, it’s one of Capcom best selling game of the year.

It’ll only happen when the profitability is at stake.
 
0
And we were still learning software techniques too. There is a reason that you seemed to have generational leaps within the generation. The software side of real time graphics was still being learned and developed. You can see that with the guy who has been rebuilding Super Mario 64 using modern techniques. He's pushing vastly more frames and polygons than the original game, on original hardware, just by applying 25 years of software know-how to the thing.
first: really impressive project.
But mostly: almost nothing is modern knowledge, but optimization through understanding of the architecture in hindsight while it stayed static as well as better documentation (even if most of the stuff was known in some areas and fields back then, just reaching that knowledge level was hard back then).
It also helps that hes not developing the principles of 3d gaming while doing the optimization, but has a clear picture whats the output, its hard to optimize if your project is shifting all the time, give them 1-2 more years JUST optimization time and im sure they would have improved a lot of it.

What im saying: most of the improvements where not out of the question on paper back then. (at least from the videos I've seen from him)
This does neither diminish the accomplishments of the SM64 team or of this project.
 
current gen hardware, is that not oblious? Capcom want to push Xbox Series X/S and PS5 to the limit
No, it's not obvious when the same team saying "push hardware to the limit" is also saying "PC is our primary platform." If their target hardware config is PC, they might be grinding the PS5 to dust, or they might be going much lighter. It entirely depends on how many PC players they're trying to reach.

Fami: "Is Switch 2 a target platform for this game?"
Devs: "We pushed our target platform hard on the environment"
Fami: "I guess Switch 2 isn't a target platform."

That is a silly conclusion. It doesn't tell you if it's a CPU limited game or a GPU limited game. It doesn't tell you RAM requirements, or if it's 60fps on PS5. This is one sentence in a PR piece, editing from what is almost certainly a panel interview. It tells you nothing.
 
No, it's not obvious when the same team saying "push hardware to the limit" is also saying "PC is our primary platform." If their target hardware config is PC, they might be grinding the PS5 to dust, or they might be going much lighter. It entirely depends on how many PC players they're trying to reach.

Fami: "Is Switch 2 a target platform for this game?"
Devs: "We pushed our target platform hard on the environment"
Fami: "I guess Switch 2 isn't a target platform."

That is a silly conclusion. It doesn't tell you if it's a CPU limited game or a GPU limited game. It doesn't tell you RAM requirements, or if it's 60fps on PS5. This is one sentence in a PR piece, editing from what is almost certainly a panel interview. It tells you nothing.
Pushing hardware, always make it seem like it’s not going to be best optimised, since Starfield and DD2 faced performance issues at all consoles and PC and they were paraded to be NG, which is true, but at a lost.

But I feel like there’s more pessimism with third party support, mostly because Midori was revealed to be a liar, and he was the one who brought the idea of Switch 2 getting phenomenal support, heck he even mentioned KH 4 getting a Switch 2 port, but that now seems unlikely.

I’m currently at the same boat of ,, Will third party truly support the system?’’ Maybe it’s my pessimism talking, but, it truly seem extremely quite, since I would expect hearing developers mention Switch 2 ports, maybe it’s an NDA thing with Nintendo.
 
And we were still learning software techniques too. There is a reason that you seemed to have generational leaps within the generation. The software side of real time graphics was still being learned and developed. You can see that with the guy who has been rebuilding Super Mario 64 using modern techniques. He's pushing vastly more frames and polygons than the original game, on original hardware, just by applying 25 years of software know-how to the thing.
That's pan.... something Koak. I am sorry for butchering his name. But there were things he did such as changing the optimisation flags. Reduce code lines, optimize the way how mario turns. I love it. His famous line is mario traveling 4 parallel universes away.

When has a new video I always tuned in.


Edit: it was Kaze emanuar
 
Is this the one?

Talk about a really complete and impressive analise.

Read it and think it will be in other node than 4NM is impossible.

But the CPU performance worried me. Isn't just 40% of the current generation too low? Like, I think even the Deck is better than that, right?
 
0
The potential pandemic sales spike was weaker for Sony and MS due to hardware shortages at the time.
Not necessarily true. SONY had a monster fy 2023 (2022 to 2023) which led them for forecast 25 million PS5s for 2023 and 2024.

The decline was in fy 2024 when demand tapered and all 3 mfg saw declines in hardware and led to the revised forecasts. Except Nintendno which forecasted a decline but beat it by 500k
 
0
That's pan.... something Koak. I am sorry for butchering his name. But there were things he did such as changing the optimisation flags. Reduce code lines, optimize the way how mario turns. I love it. His famous line is mario traveling 4 parallel universes away.

When has a new video I always tuned in.
Kaze Emanuar is who you're thinking of, although Pannenkoek2012 is notable for making so many observations about the engine.

And yeah, what Kaze has shown off on N64 is truly incredible. Just utterly ridiculous. 60FPS with more geometric detail than the base Mario 64.
 
I basically won't be playing any big third party games on the switch because I'll be playing them on pc, I might be in a position to buy one or two big third party games on the switch2 with the idea of testing out the visuals as well as the performance, but I'll still be playing mostly Nintendo first party games as well as tons of indie games.Focusing on whether or not big third parties are on the switch2 isn't a particularly worthwhile concern for Nintendo (I said they don't care, not that they don't want to)
 
Kaze Emanuar is who you're thinking of, although Pannenkoek2012 is notable for making so many observations about the engine.

And yeah, what Kaze has shown off on N64 is truly incredible. Just utterly ridiculous. 60FPS with more geometric detail than the base Mario 64.
Yeah, I feel a shame of forgetting his name. But it was just genius the amount of changes he made.
 
Last edited:
What im saying: most of the improvements where not out of the question on paper back then. (at least from the videos I've seen from him)
At least some of them are - he's applying compiler optimizations that weren't yet available. I don't know what C compiler the original devs were using, but he's rewriting loops so that GCC can unroll them, an optimization that I'm not sure was even detectable by compilers of that era.

I believe he's replacing the RSP firmware. When SGI developed the RSP for the N64, they based it on work they were doing for CAD systems and offline 3D animation. The default firmware for the RSP was high precision to support those operations. SGI didn't realize that such a firmware was overkill, and a performance suck, for real time game consoles. Because they'd never built one.

Nintendo, using the firmware, didn't realize that it was too high precision either. They were building their first 3D engine. SGI was the company for 3D technology at the time, by a mile, and Nintendo the most advanced game studio, and the only company shipping their second 3D accelerator (see: Star Fox). Yet they missed this basic thing.

Sure, math is the same everywhere. So anyone who understands math could have achieved many of these finer grained optimizations. And yes, the SM64 engine is a mess dictated by the release date. But that doesn't change my point of developers learning to use hardware, which feeds into new hardware designs.
 
But I feel like there’s more pessimism with third party support, mostly because Midori was revealed to be a liar, and he was the one who brought the idea of Switch 2 getting phenomenal support, heck he even mentioned KH 4 getting a Switch 2 port, but that now seems unlikely.
Out of curiosity I looked up MysticDistance related info and it turns out he said the rumor that p5r would be a switch exclusive, so I don't think he has any credibility on the atlus game logging in to switch2 blurb.
 
Last edited:
0
Is this one of the first “leaks” at the likely GPU clocks?

3.5 Tflops 👀
This is a fami->internet->fami situation. They're pulling those clock numbers from the Nvidia hack, but there is wide doubt that they mean what the poster assumes they mean. Even if they do mean that, there is a wide gap between tested clocks in 2022 and final clocks in 2025.
 
This is a machine translation, but it is an interview with a developer about Monster Hunter.


Ryozo Tsujimoto ("Mr. Tsujimoto"):

We released "Monster Hunter: World" ("World") in 2018 and "Monster Hunter World: Iceborn" ("Iceborn") in 2019, but the development of "Monster Hunter Wilds" ("Wilds") started before Iceborn was released. We started the development of "Monster Hunter Wilds" (hereafter referred to as "Wilds") before Iceborn was released.

This includes the development of the concept and the verification of its feasibility. During the development of World, we decided that Wilds would be done on the next model with the next specifications, and based on this, we decided to aim for the "pinnacle of what we could do.

Citation
 
I think Twilight Princess is a mostly pretty game that shines in interiors or 'manmade' environments like castle town and dungeons. But I never quite vibed with how the overworld looked, aside from the smaller areas like Lake Hylia and Snowpeak.

Oh you're definitely not the only one that thinks this. TP is one of the biggest mixed bags in terms of graphics for a Nintendo game. Some parts look good to great like Zora's Domain and the Twilight Palace. But other parts look very, let's say..."lacking" like Ordon Village and Hyrule Field. For the latter, especially, we were still using hardware at the time that wasn't powerful enough to render large open environments of that size, so you'd see soupy low res ground textures, and limited geometry and such. Though that being said, TP probably did as decent a job as it could have on that front given the limitations it had.
 
0
At least some of them are - he's applying compiler optimizations that weren't yet available. I don't know what C compiler the original devs were using, but he's rewriting loops so that GCC can unroll them, an optimization that I'm not sure was even detectable by compilers of that era.

I believe he's replacing the RSP firmware. When SGI developed the RSP for the N64, they based it on work they were doing for CAD systems and offline 3D animation. The default firmware for the RSP was high precision to support those operations. SGI didn't realize that such a firmware was overkill, and a performance suck, for real time game consoles. Because they'd never built one.

Nintendo, using the firmware, didn't realize that it was too high precision either. They were building their first 3D engine. SGI was the company for 3D technology at the time, by a mile, and Nintendo the most advanced game studio, and the only company shipping their second 3D accelerator (see: Star Fox). Yet they missed this basic thing.

Sure, math is the same everywhere. So anyone who understands math could have achieved many of these finer grained optimizations. And yes, the SM64 engine is a mess dictated by the release date. But that doesn't change my point of developers learning to use hardware, which feeds into new hardware designs.
I think that he also got rid of a lot of redundant and useless variables. Merged a few functions. Cut down some instructions. And a community help him find algorithm to for further optimization. Outside of front end and back end, I am not much of a coder, but this shows me some hardcore real world algorithms
 
0
There’s definitely diminishing returns but that image is a bit of a false representation. It looks like the 6000 mesh is the base so obviously 60000 mesh won’t look any different. Like if you resize an image to be bigger. It won’t add in additional details.
You can tell that the 60K is the base if you look closely. It's a lot smoother and more rounded.

The problem is that you have to look closely.
 
At least some of them are - he's applying compiler optimizations that weren't yet available. I don't know what C compiler the original devs were using, but he's rewriting loops so that GCC can unroll them, an optimization that I'm not sure was even detectable by compilers of that era.
i give you the compiler optimization improvements. There was real science with formal models made on universities to make mathematically better optimizations.
I believe he's replacing the RSP firmware. When SGI developed the RSP for the N64, they based it on work they were doing for CAD systems and offline 3D animation. The default firmware for the RSP was high precision to support those operations. SGI didn't realize that such a firmware was overkill, and a performance suck, for real time game consoles. Because they'd never built one.
is he replacing it? missed that. fair there to, but its not a "new tech".Again, for me new tech is new general scientific discoveries and techniques how you do something, replacing a firmware for a specific hardware is not general progression. In this case: them not realizing that its overkill is exactly a "not enough time to optimize", and not new developments, since there is no new science/formulas/knowledge, just more experience with existing tech.
Nintendo, using the firmware, didn't realize that it was too high precision either. They were building their first 3D engine. SGI was the company for 3D technology at the time, by a mile, and Nintendo the most advanced game studio, and the only company shipping their second 3D accelerator (see: Star Fox). Yet they missed this basic thing.

Sure, math is the same everywhere. So anyone who understands math could have achieved many of these finer grained optimizations. And yes, the SM64 engine is a mess dictated by the release date. But that doesn't change my point of developers learning to use hardware, which feeds into new hardware designs.
"Yet they missed this basic thing." that's the thing. its not new techniques, its not new hardware or formulas, its not using the available tech and tools to its fullest because the lack of knowledge, and not endless time to optimize instead of trying to ship a industry moving product.
But that doesn't change my point of developers learning to use hardware, which feeds into new hardware designs.
Oh im not disputing that. And im not trying to be contrarian, i just wanted to point out the progress in that project is less knowledge of 25 years and more "better documentation and development tools (faster pcs, higher resolution screens, great analysis tools) no binding time and cost limit to ship a retail product.

If the team back then had all the time of the world, and the only goal was: optimize it so that it runs more efficient, im 100% confident they could have achived a similar goal. But no company is gona pay for that. I had my fair share of "i know i could make it 5% better, but it would take 15 days instead of 2 to work on that feature". Or going back to a feature half a year later, with better understanding what the long term vision ist, and knowing where a lot of optimizations could have been made, independent of the technology.

In essence: solo tinkering and commercial development are 2 worlds, and i to often see the "if one person can do that, why could/can't the team do it".
Im just trying to set the development reality somewhat into perspective =)

there are other projects (mostly remakes), where it IS clear that technological progress and new developments opened up some possibilities.
 
~ 10 year

from

WiiUVC_TheLegendOfZeldaALinkToThePast_09.bmp


to

wind-waker.jpg



and from this

Wii_TheLegendOfZeldaTwilightPrincess_04.bmp


to

1800.png
Oh, I remember I saw a picture of
"when you think 20 years ago was this:"
Shows a link to the past.
"But it really is this:"
shows picture of Wind Waker.

I hate that technology is reminder of how fast life moves. Not to mention, I have more white hair popping up. Just a couple of months ago, it was two or three. Now, it is about five or six.
 
And we were still learning software techniques too. There is a reason that you seemed to have generational leaps within the generation. The software side of real time graphics was still being learned and developed.

Yeah, one of the reasons Super Mario Galaxy looks significantly better than Sunshine is cause EPD Tokyo seemed to have become accustomed to the hardware enough that they seemed to have an easier time implementing things like bump mapping. They improved in other areas as well, of course, but just the additional bump mapping alone helped out substantially. Sunshine had extremely limited bump mapping, and while it was still a good looking game, most of it ended up looking very flat and drab. But with SMG, they implemented BM practically everywhere and helped everything feel much more "solid" and realistic by comparison.

edit: since we're on the subject, did SMG or SMG2 have any normal mapping?
 
Last edited:
0
You can tell that the 60K is the base if you look closely. It's a lot smoother and more rounded.

The problem is that you have to look closely.

Most modeling tools have a subdivide tool to automatically smoothen the model at the expense of more polygon. It is inefficient though and will not add additional details but fast and very effective. That’s what it look like they did to the 6k model. It could also be that that 60k really is the base model but they didn’t leverage the additional triangles. Like me making a completely green 4k image saying there’s no improvement in detail compared to a green 1 pixel image
 
Last edited:
Please read this new, consolidated staff post before posting.

Furthermore, according to this follow-up post, all off-topic chat will be moderated.
Last edited by a moderator:


Back
Top Bottom