Nintendo Switch Tech Speculation discussion

Status
Not open for further replies.
Fantastic looking game ? Yeah thx to the art, but the texture filtering seems horrible, the jaggies too. Good art with horrible picture quality is such a waste...

And by the look of it (still no aniso and aa), I think Odysey was first developped for Wii U too.

Certainly, these things are always subjective, but the art style bodes well for a timeless look that will hold up over time. I agree that the texture filtering and lack of AA is a negative, but for me, I sit about 7ft from my 42 LG 1080p tv, so I sit far enough away for some of these blemishes to be less noticeable. Jaggies drive some people nuts, but I gamed on Wii for so many years that it doesn't really bother me too much.

Mario Odyssey may have started development on Wii U, but likely changed over to Switch much sooner than Zelda BoTW. It also has a lot more time at its disposal to leverage any and all strengths that the Switch has over the Wii U. When I watch trailers for Mario Odyssey, it looks head and shoulders beyond Mario 3D World for Wii U.
 
And by the look of it (still no aniso and aa), I think Odysey was first developped for Wii U too.
Switch is likely bandwidth starved.
Wii U edram probably has more bandwidth.

When I watch trailers for Mario Odyssey, it looks head and shoulders beyond Mario 3D World for Wii U.
It does not look much better than Mario Galaxy in an emulator.

Textures and IQ in the Donk City are atrocious.
 
Could this be real? Sounds a little too much for what they've shown:

Post 1

There are no additional functions in the base, it's just an output. The main unit is actually a 1080p capacitive touch screen, surprisingly. I haven't played with the controller but from observation the key travel of LR buttons is probably very short.
I saw heatsink fins on the main unit, probably because it was an engineering sample. I haven't seen a production unit yet.
The base is very light, feels very plasticky.
The side controllers are very small. They start charging when you plug them into the sides of the main unit. The split LR buttons are positioned under your hands. Handling feels weak due to the lack of buttons.
There's a 5cm vent at the top of the main unit and you can see heatsink fins inside.
The main unit is around 1.5kg.
Each small controller has 2 separate SL SR buttons. Key travel is still very short. Doesn't feel very nice.
The silicone buttons on the controller are very average.
The brightness of the main unit isn't very good. I'm guessing it'll be just visible enough outdoors if set to max brightness, but then you can forget about battery life.
I checked again today, the main unit definitely has a 1080p screen, which is great. There are no games on it so I couldn't tell much [about display quality]. Will try to check out the motherboard in the next few days.
I had a look at the heatsink today, there's a L shaped heat pipe 0.8cm wide and about 12cm long, placed upside down. The fins on it are 5cm long and 0.8cm tall, and there's a turbine fan with 4cm diameter. Looks pretty shoddy, but should be enough since the main unit doesn't generate that much heat anyway.
The demo screen is very boring, just a bunch of fish swimming around. The factory floor is very noisy so I couldn't tell if it generates much noise. I touched it and it wasn't too warm after running for 2 hours, so it's not too bad.
Brightness really isn't very good.
As a portable console the Nintendo Switch is not bad, but don't expect it to be on par with home consoles. If you want 1080p60 you can forget about it, even PS4 Pro only barely achieves that. If The Legend of Zelda: Breath of the Wild can run at 1080p30 then it's definitely worth buying, after all it only runs at 720p on the Wii U. But if it's interlaced/interpolated then that'd be pretty embarrassing.
It's a good choice as a portable console, and the 2 small controllers allow multi-player Mario Kart too. I'll need to check the battery capacity of the controllers, not sure how good its battery life is.
Had a look at the motherboard today. The CPU looked like about 10mm2 and is made in Taiwan. There are 2 memory chips but couldn't see the model or type. The fan is pretty powerful, I can feel waves of hot air. The unit itself doesn't feel very warm. There are heat pipes so we shouldn't need to worry about frequency throttling.
On the right side of the main unit there's a 8cm2 slot that's about 0.6cm deep. Pretty sure that's for the battery, a big one at that.
There's no way Nintendo would be generous enough to include a hard drive.
Even the Nvidia GP106 is only 120cm2 [Ed: did he mean mm2 ?]. I think we can look forward to good graphics performance on the Nintendo Switch even though some chip area is taken up by the CPU.
The CPU in Nintendo Switch is what's commonly known as an APU. Its area is about the same as Nvidia GP107, but that includes the CPU too, so graphics performance won't be as good as GeForce 1050.
Reply from OP: Not an engineering sample, a verification unit assembled in Japan, should be the same as production units. You think our factory security are useless? [Its performance] beats iPhone 8 hands down, but miles behind PS4.

Post 2
Can't take any photos. I've tried to describe all I could. Haven't checked the battery capacity of the side controllers yet, will take a look soon.
The controllers feel okay when attached to the main unit. It sucks that there's no + button, but I guess it's a compromise to allow two player games.
Our factory doesn't have the traditional controller so I can't say anything about that.
The main unit structure is pretty simple, should be easy to repair.
The side controllers are very complex, make sure you take good care of them. If you break the gold pins on the bottom then you can't charge them.

Post 3
The battery in the small controllers is about 5x2x0.5cm. You guys can guess the capacity. I don't want to go into the clean room, that'd look too suspicious.
The advantage of Nintendo Switch is no frequency throttling. There's no lag whatsoever after running for 2 hours straight. Not like mobile phones that can last only 1 minute on full performance.
I almost took a controller button home today, but then I thought it's pretty useless so I didn't do it. If I get caught the punishment will be severe [Ed: lit. "they'll cut off my hands"].
The most expensive component in the NS, besides CPU and motherboard, would be the side controllers. The quality is very high. Feels good to use when connected together.
I'm confident in the small controller's battery life after seeing the battery dimensions.
Quality control is very strict, you can buy with confidence.
I think the base is pretty useless besides charging. It's mostly a portable console.
The base is easily scratched because it uses the shittiest plastic.
If you want to upgrade your portable console then buy this. The display is high definition and not blocky.
Nintendo Switch is definitely the best portable console.
The chip looks like it uses the same packaging as the Nvidia GPUs. Looks very big and powerful. It's made in Taiwan, there's no TSMC logo but we all know who made it. The surface of the chip is smooth like GP106. Have a look at your video card and you'll know what the NS CPU looks like.

Post 4
Diagonal size of the screen is about 17. [Ed: they didn't say what unit.]
Our production line made 2,100 units today and we're still increasing output. The whole factory floor should be able to make 20,000 units per day.
I saw the blue and orange versions of the controller today.
Nintendo Switch has active cooling with heat pipes and fans, you can't compare it to mobile devices. It won't throttle frequency even after prolonged use. Its performance beats any mobile device hands down, including iPad Pro.
There's a very important thing I still need to find out, will let you guys know in a few days.
What's the maximum distance of the controllers? This is an important question if you want to play 2-player Mario Kart with split screen.
I haven't seen the OS. When they do testing I'm also working, when I'm off shift I'm too tired to go check.
About the 1080p screen, I'm not 100% sure about my visual estimation. The testing computer says output is 1080p. I guessed it's 1080p based on the fact that the motherboard in the base is too simple. [Ed: is this in response to a question about whether it supports 4K? Some context would be nice here.]
The base doesn't enhance performance in any way. There's only the power connector, a USB 3.0 and a HDMI port on the inside, and 2 USB 2.0 ports on the outside.
Nintendo Switch is a traditional portable console. It's major selling point isn't innovation. The only thing extra is a video-out.
One characteristic of the NS is that everyone has a different view on how it should be used. For me, it's a great portable console.
Sorry for not being able to check the details of various things. I can only say that I haven't found any motion sensing capability yet, and our factory doesn't have any testing for it either. Maybe I just haven't found it yet! I'm too tired today, will reply tomorrow.
The motherboard in the base is way too simple. Just think of it as a charger.
The main unit has 2 small holes on the sides, and a heat vent at the top. Dust getting inside could be a problem.

Post 5
Here's some standard specs: CPU 1750 MHz, GPU 912 MHz, EMC 1600 MHz. (I remembered wrong, the GPU is actually 921 MHz. It sure looks like it [Ed: likely in reply to someone saying it's similar to some other chip], but this one is 16nm so it's probably a new product.)
Battery capacity is 4310 mAh. (The battery is not replaceable.)
Today I saw a cube module near the bottom of the controller, could be a gyroscope. I'll ask around when I get a chance.
GPU frequency is half of a desktop Pascal.
The controllers are very light, around 20–30 g. Together with the main unit they're less than 1,000 g.
Correction, the CPU is 1785 MHz.
Battery is built in.
One unit has been under test for 11,750 minutes in our factory, still working well.
It can be charged using a standard USB Type-C cable.
There's no such thing as frequency throttling when not plugged in. The performance isn't very high to begin with, if it throttles then the games will lag like hell.
There is indeed an infrared-looking light at the bottom of the right controller.
I've confirmed today that there will be a 4G version.
We have an updated motherboard and the old motherboard. Probably different firmware, I can't really tell. Battery life is about the same, probably just a different version.
Today we had units shipped to Japan and Australia. I was confused.
Today I saw a colleague making some accessory with a network port. The main unit is now thicker, with 2 layers of back cover. The plan is to make 2,000 first. (This is the 4G version, we're making 2,000 as a trial, I didn't see the unit myself, heard it from some other guy in my dorm. I'll take another look the day after tomorrow when I go back to work. There's a new circuit board behind the back cover, I reckon it's an accessory. I'll check it out on Monday when I go back to the factory.)
[to be continued]

Post 6
Here's a spec for the standard version, the standard main unit without controllers is 302 g.
Each controller is 50 g.
Today I saw the enhancer. There are still many questions but I'll tell you what I know.
The processor in the enhancer is even bigger than the main unit CPU, at 200mm2.
Dimensions looked like 12x18, on par with GP106.
There are 2 extra memory chips. (Pretty sure it adds 4 GB memory, for a total of 8 GB.)
It connects to the back of the main unit motherboard via some sort of PCI bridge.
There are 2 Wi-Fi [antennae], 1 HDMI, 1 Mini-DP, 1 network port, 2 unknown circular ports, 3 network activity indicator lights.
The enhancer is much more complex than the main unit. It also has 6 or 7 unknown storage chips.
The enhanced main unit also looks different than the standard unit. It's got extra bits on the motherboard, including the bridging connector.
What I don't know is whether this is a development unit or if it'll be officially announced, and of course I have no idea if it'll go on sale at all.
The enhanced version doesn't have a base yet.
The enhanced version can connect to a TV/display without the base. It also has a power input.
I don't know if the processor on the enhanced version includes CPU.
If it doesn't include a CPU, then it can at most upscale existing games to 4K, it won't bring any dramatic improvements. The weak CPU might be the bottleneck.
If the main processor on the enhanced version only has GPU, then it'll be more powerful than PS4 Pro.
At the moment our factory floor only plans to make 2,000 enhanced units. (I also think it's probably a development unit. Let's wait for Mr. Ren to give us more info.)
The enhanced version is very powerful, but also weighs more and feels worse in the hands. It's purely to cater for 4K TVs. I'm also anxious to know if it'll go on sale.
The performance will be off the charts with the main processor plus enhancer processor. I've never seen such a huge 16nm chip on a mobile device, especially because the main processor is already 100mm2.
The main processor is dwarfed by the enhancer processor.
There's no additional battery on the enhancer. You probably have to plug it in when playing.
The power adapter is also built in.
I've already said, this might be a development unit. I want to tell the whole story, please go back and have a look at my previous posts. (The base only has a USB-C port, it won't handle a huge amount of data transfer.)
The main unit is unexpectedly light, at 300 g. I'm liking the standard version more and more.
I weighed it myself with an electronic scale.
High definition is finally a reality.
Each smaller controller is 50 g. When combined with the main unit the total is 400 g. Is that acceptable to you guys?
Yeah I think the enhancer is a development unit too, but the performance is really something.
There's no way to boost the performance of the standard version via the base. It only has a USB-C connector, at most USB 3.0, that's obviously not enough bandwidth.
Alright I've said a lot. Flame on.
!translated

Original thread: https://www.reddit.com/r/NintendoSwi...aybe/?sort=top
 
Last edited by a moderator:
Lol, those posts are excellent comedy material!! Serious next-level trolling. I remember myself pre Xbox One though; I believed it was going to dwarf PS4, had display lanes, extra ultra special EDRAM, Data move engines. If you are a fan of a brand, you want it to succeed, believe me I have been there :(
 
Fantastic looking game ? Yeah thx to the art, but the texture filtering seems horrible, the jaggies too. Good art with horrible picture quality is such a waste...

And by the look of it (still no aniso and aa), I think Odysey was first developped for Wii U too.

I'm now playing 2016's Ratchet&Clank reboot on the PS4 Pro. I think what they claim to be using Temporal Injection to do both upscale and AA, and it's running in HDR in my TV.

It uses stylized art and it's freaking gorgeous at that. Ratchet&Clank is hard proof of that theory claiming "Nintendo doesn't need more powerful hardware for their stylized games" being pure bullshit, IMO.





Lol, those posts are excellent comedy material!! Serious next-level trolling.

This guy's posts from November (shortly after the Switch started production according to some sources) gave away the precise numbers for tablet weight, JoyCon weight and each device's battery capacity.


I remember myself pre Xbox One though; I believed it was going to dwarf PS4, had display lanes, extra ultra special EDRAM, Data move engines. If you are a fan of a brand, you want it to succeed, believe me I have been there :(
Proof so far point to him being a factory worker, not a fan.
 
Last edited by a moderator:
Nintendo's games benefit from more powerful hardware just like they all do, the difference is that Nintendo's games already look very nice even though they aren't bleeding edge. Going back to Wii, the Mario Galaxy games look very nice to this day, they actually do look fantastic using the Dolphin Emulator to render them in 1080p. Same goes with Zelda Skyward Sword. Compare this to Resident Evil 4, a fantastic looking game at its time, doesn't hold up nearly as well over the years, and doesn't benefit nearly as much from rendering in 1080p on Dolphin.
 
One concern I have is with post effects. Mobile devices typically can't do 60 fps full-screen post effects (I've learnt from experience :(). I think that's because they mostly work as TBDR and post effects don't fit the tiles? Will this be a problem for Switch? Because post effects make a huge difference to general art and are an essential part of any game these days, and if Switch is hampered in the same way other mobile devices are then there probably will be few 60 fps games.
 


The thread with the full translation is here:
https://www.reddit.com/r/translator/comments/5shm34/chinese_english_foxconn_leak_about_the_nintendo/



Some parts I'd like to highlight from the guy's posts:

Today I saw the enhancer. There are still many questions but I'll tell you what I know.
(...)
It connects to the back of the main unit motherboard via some sort of PCI bridge.
(...)
The enhanced main unit also looks different than the standard unit. It's got extra bits on the motherboard, including the bridging connector.
(...)
At the moment our factory floor only plans to make 2,000 enhanced units.
(...)
What I don't know is whether this is a development unit or if it'll be officially announced, and of course I have no idea if it'll go on sale at all.

He's calling it "the enhancer" that connects to the main unit's motherboard via "some sort of PCI bridge". But then he says the "enhanced main unit" is a bit different, mostly from the aforementioned port on the back.
We also know that the Switch has no hidden ports on the back.
He's also claiming they're only making 2000 "enhanced units", meaning they're producing 2000 "Enhanced Main Unit + Enhancer" combos. It looks like whatever this is, it didn't reach production phase, but 2000 is certainly enough for a wide enough number of devkits.
That said, we're either looking at:
1 - Full handheld "Switch Pro" that is more powerful but runs Switch vanilla games
2 - "Switch Enhance-able" that can connect to the "Enhancer". This means a future Switch SKU that brings upgradeability to the table.
3 - Regular Switch being able to use USB-C's alternate mode to carry PCI-Express or maybe some proprietary comm like NVLink (or a fraction of what nvidia calls NVLink given the limited number of pins).

Regardless, the guy isn't confident this will ever release for the market. I don't know if it makes sense to produce 2000 units of something that may not be coming to market.. internal prototypes would be made in much lower numbers IMO, but this guy may know more than me in these matters.
It also looks like he mistook this enhanced version with the "4G version". He may have seen a different motherboard and assumed it would be for a 4G version, but ended up being something completely different.


The enhanced version doesn't have a base yet.
The enhanced version can connect to a TV/display without the base. It also has a power input.
(...)
The enhanced version is very powerful, but also weighs more and feels worse in the hands.
(...)
There's no additional battery on the enhancer. You probably have to plug it in when playing.
The power adapter is also built in.

There's no battery and when the enhancer is connected, you have to plug it in for the combo to work.
I think this means the "enhancer" is actually a dock in its final form, though it just isn't being built as such for these devkits. Maybe to avoid rumors of a "Super Dock" coming down the road, which would make some sense given the secrecy around the Switch so far.



The processor in the enhancer is even bigger than the main unit CPU, at 200mm2.
Dimensions looked like 12x18, on par with GP106.
There are 2 extra memory chips. (Pretty sure it adds 4 GB memory, for a total of 8 GB.)
(...)
The enhancer is much more complex than the main unit. It also has 6 or 7 unknown storage chips.

It sure looks like a discrete GPU, along with more eMMC storage.
The increased storage makes me think the idea of this "enhancer" is to bring processing power similar to the PS4Bone/Pro for (mostly multiplatform) titles that will be exclusive to the enhancer. Enhancer games will only install in the enhancer dock, and not gobble up the main unit's internal 32GB.
So instead of launching a home console that plays the big multiplatform games, Nintendo would launch an add-on for the Switch that allows running these multiplatform titles. Big question is if the CPU in the SoC could ever handle such a thing.

Maybe releasing this "super dock" will depend on market research. For example, if they feel the home console market is too saturated by the time the super dock is ready for production, they'll cancel it. Though the 2000 unit production would make me think otherwise, really.




As for the clocks, it's 921MHz GPU and 1785MHz CPU. Completely different from any of the numbers given by Eurogamer. We'll have to wait and see about that.
An early teardown of the Switch will also tell us if there's something odd connecting to that USB-C port, and/or if it's a different SoC than the old TX1.
 
maybe its just a switch 2 dev unit ? The enhancer is really just the soc the switch 2 will use and its just cheaper to make it into a small device that plugs into a switch ?
 
One concern I have is with post effects. Mobile devices typically can't do 60 fps full-screen post effects (I've learnt from experience :(). I think that's because they mostly work as TBDR and post effects don't fit the tiles? Will this be a problem for Switch? Because post effects make a huge difference to general art and are an essential part of any game these days, and if Switch is hampered in the same way other mobile devices are then there probably will be few 60 fps games.

Seeing as how TX1 thumps the majority of mobile processors in mobile devices, Switch isn't going to run into the same degree of troubles. Mario Kart 8 Deluxe runs 1080p 60fps on Switch, so at the very least you have a baseline of what Switch can do at this resolution and framerate.
 
Absolutely, if they were going with 16nm Finfet, it seems unlikely that they would need to reduce the clocks to this extent. I still wouldn't be shocked if we end up seeing the 28nm HPC+ process used. Its the least expensive option, and they refined the 28nm process over the years with consistent improvements to energy efficiency. The chip would be a little bigger than the 20nm TX1, but I am not convinced that's a big deal.

Zelda BoTW didn't make the move to Switch until spring of last year, so it is certainly a game optimized from the beginning for Wii U. Mario Odysey and perhaps whatever Retro has been working on will be the titles to really show off what a game designed from the ground up for Switch can accomplish. I suppose you could make an argument for Xenoblade 2 as well. Regardless, Zelda BoTW is a fantastic looking game with visuals that should hold up well to the test of time. Very similar to how well Zelda Wind Waker holds up.

Odyssey looks like a game that maybe started on Wii U, but since it's not actually coming to Wii U they could do more with the graphics. It looks better than 3d world. Xenoblade looks like the Wii U game with tales esque character design, the OG Wii game still looks the best haha.

And like someone else said, the biggest technical issue with all these games is the lack of AF, and the lack of AA on the 720p games. Hopefully Mario is 1080p at launch. Really grinds my gears when devs use poor texture filtering and AA for the sake of a few polygons.
 
Switch is likely bandwidth starved.
Wii U edram probably has more bandwidth.

It does not look much better than Mario Galaxy in an emulator.

Textures and IQ in the Donk City are atrocious.

Wii U's eDRAM might have more by the numbers bandwidth, but even if it does, due to memory compression in the nvidia tech the Switch will have more practical bandwidth.
Actually most consoles in fact could use more bandwidth, it's the biggest reason for the lack of AF, which is thought of as an extra by developers. If consoles had an abundance of bandwidth we'd see much better IQ. The Ps3 and 360 were bandwidth starved, particularly the 360.
 
Seeing as how TX1 thumps the majority of mobile processors in mobile devices, Switch isn't going to run into the same degree of troubles. Mario Kart 8 Deluxe runs 1080p 60fps on Switch, so at the very least you have a baseline of what Switch can do at this resolution and framerate.
I'm talking specifically about post effects which don't fit the architecture. I don't think that's an issue with Tegra because it's the same as the PC architecture. Also the resolution is much lower I guess. A decent phone SOC tends to be coupled with an insane screen res having to process 4+ more times as many pixels as Switch.
 
I'm talking specifically about post effects which don't fit the architecture. I don't think that's an issue with Tegra because it's the same as the PC architecture. Also the resolution is much lower I guess. A decent phone SOC tends to be coupled with an insane screen res having to process 4+ more times as many pixels as Switch.

But you already knew the Tegra X1 was a Maxwell Gen 2 architecture, so is there another point you were alluding to regarding post processing effects on Switch? It certainly wont be an architecture problem, seeing as how Maxwell Gen 2 is a very modern architecture with a rich feature set.

http://www.nintendolife.com/news/20...witchs_chipset_is_very_close_to_tegra_x1_spec

ARM confirms Switch uses four A57 CPU cores. So anyone holding out hope for A72/3 cores can put that to rest.
 
It's because I don't specifically know what the issue is with mobile SoCs. It's nothing to do with modern-ness as even the latest mobile SoCs can struggle. If it's an issue of TBDR on mobile SoCs versus normal rendering on X1, then it's not a problem. If it's a bandwidth issue though, or something else, it could well be an issue on Switch.

Putting this the other way, if Switch has no problem with post effects then it'll have a graphical advantage over other mobile devices.
In answer to my own question , I've found this which shows 60 fps with moblur and bloom.
 
http://www.nintendolife.com/news/20...witchs_chipset_is_very_close_to_tegra_x1_spec
ARM confirms Switch uses four A57 CPU cores. So anyone holding out hope for A72/3 cores can put that to rest.
It was a Facebook repost of the androidpit's article that admittedly copy/pasted eurogamer's specs and clocks. It was posted by some community manager (not a tech guy) and he/she probably just assumed it was public knowledge. Poor guy probably didn't ever imagine the can of worms he opened.
That post has since been deleted and ARM has officially retracted any reference to the CPU architecture in the Switch.






I was watching Nintendo's "hardware overview video" and I noticed there's one more thing the Foxconn worker revealed that came through:

There is indeed an infrared-looking light at the bottom of the right controller.

So the full list confirmed specs that came exclusively from the Foxconn worker back in November:

1 - 4310mAh battery in the main unit
2 - 525mAh battery in each JoyCon
3 - 300g tablet weight measured in a digital scale
4 - Each JoyCon weights 50g
5 - Shoulder buttons called SL and SR on each JoyCon
6 - Exact number and type of I/Os present in the dock -> 2*USB 2.0 + 1* USB 3.0 + 1*HDMI
7- IR LED in the right JoyCon
8 - "Orange and Blue" Joycons -> Units are actually bright red and blue (the orange -> bright red difference is probably just semantics or a translation thing)


There's just no way this guy isn't legit.
I also find it very hard to believe they were testing the CPU at 1785MHz in what seems to be production hardware, just to clock it at almost half the frequency with the shipping firmware.
 
I also find it very hard to believe they were testing the CPU at 1785MHz in what seems to be production hardware, just to clock it at almost half the frequency with the shipping firmware.

Depends on what they were testing for. I don't find it hard to believe.

Just because the silicon can do that doesn't mean it's right to clock that high for the final product (battery life, reliability, heat, noise, charging issues e.g. battery temps).

There might yet be a mode of operation (e.g. GC/Wii CPU emulation targeting a single core) that uses different clock arrangements.
 
Depends on what they were testing for. I don't find it hard to believe.

Just because the silicon can do that doesn't mean it's right to clock that high for the final product (battery life, reliability, heat, noise, charging issues e.g. battery temps).

Are you familiar with SoC testing methodologies within the assembly lines?
Honest question, because the difference is so great that if we are to look at ARM power curves on 20nm Cortex A57 (1800MHz consumes 3.5x more power than 1000MHz on quad-core operation), it just doesn't seem to make sense to emulate "environmental conditions" unless Nintendo was trying to emulate people playing inside an active volcano.


There might yet be a mode of operation (e.g. GC/Wii CPU emulation targeting a single core) that uses different clock arrangements.
And they were testing only single core operation? The guy checked the testing station over a period of several days and they kept testing this specific operation for emulation?
Plus, if they were testing for Wii emulation, why require the GPU to run at 920MHz?
 
Status
Not open for further replies.
Back
Top