Jump to content


Photo

Is the wii u THAT underpowered?


  • Please log in to reply
160 replies to this topic

#61 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 29 May 2013 - 07:48 PM

I think there is some truth to this- games will start running more smoothly as developers become more familiar with the system.  Nintendoland is really smooth and crisp.  Madden '13 is a little choppy at times.  If they were going to make a madden '14, I'd imagine it would be smoother.

I'd hope so, Tiburon were given less than 6 months to port Madden.  Same with all of the other early ports.  The large publishers didn't have final devkits until midyear, which gave them little time to get crackin on the ports, and even less time to actually understand what the system has and how it works.  I place most of the blame for that on Nintendo, but a little honesty from publishers about it would have been nice.



#62 Rockodoodle

Rockodoodle

    Chain Chomp

  • Members
  • 677 posts

Posted 29 May 2013 - 07:50 PM

I think there is some truth to this- games will start running more smoothly as developers become more familiar with the system.  Nintendoland is really smooth and crisp.  Madden '13 is a little choppy at times.  If they were going to make a madden '14, I'd imagine it would be smoother.

 

 

No the Wii U is not underpowered, but Nintendo alone can't prove it isn't,even if the graphics of Retro's new game blows minds, people will still say the Wii U is weak simply because the game was produced by a first party Nintendo developer, Nintendo needs a really good looking third party game to break the perception that the Wii U is underpowered



#63 GAMER1984

GAMER1984

    Lakitu

  • Members
  • 2,036 posts
  • NNID:gamer1984
  • Fandom:
    Nintendo

Posted 29 May 2013 - 07:52 PM

It already has that, in Need For Speed. People will never be satisfied because all they have is their confirmation bias and they aren't going to let it go no matter how good a game looks or who makes it. Similarly, many of those people who are Sonic fans were super hyped when Sonic was rumored for Xbox One, but now Sonic is irrelevant and overdone because it is Nintendo exclusive. People will find any way they can to believe stupidity so long as it confirms what they thought all along with some of this stuff. "Obscure, incorrect information stating that the Wii U CPU has fewer DMIPS than PS3? I'll take it!"


No I think its a racer so its not a very demanding game. But let's say that the Crysis 3 port was released on Wii U and it showed higher resolution, better frame rate and overall better visuals then I believe people saying the Wii U is underpowered would be less than it is now.

#64 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 29 May 2013 - 08:04 PM

No I think its a racer so its not a very demanding game. But let's say that the Crysis 3 port was released on Wii U and it showed higher resolution, better frame rate and overall better visuals then I believe people saying the Wii U is underpowered would be less than it is now.

It might not be a very demanding game, but the 360 and the PS3 would not be able to run the game as it is configured for Wii U.  Really all they took advantage of was more RAM, the rest was just butter.  The frames are smooth.



#65 Nintyfan86

Nintyfan86

    Bob-omb

  • Members
  • 262 posts

Posted 29 May 2013 - 08:06 PM

Guys,

 

I honestly believe the following:

 

  1. Nintendo's exit with the Wii hampered Wii U excitement.
  2. Confusion over the controller compounded with #1 has a lot to do with less enthusiasm. 
  3. Nintendo honestly believed the gamepad would be the next Wii Controller with consumer response. 
  4. Item #3 hampered focus from graphic fidelity to gamepad integration and focus. Hence the lack of tools (possibly one, non-intentional reason, but this could explain why things were so poor at launch).
  5. Finally, items 3&4 led third parties to believe functioning software, with gamepad integration, would provide a decent ROI. Given the low install base and probable attachment ratio, the investment did not need to be too high to begin with.  

Now, this is really item 6, and this deserves it's own paragraph. Publishers have watched year over year decreases in game sales. They anticipate the PS4 and Xbone to reverse this trend. Hence forth, they are spending as little as possible to bring ports to a new market (Wii U), while holding back until the 'big boys' launch. At this point, the Wii U's install base is bigger, and they know that one of two things will happen:

 

  1. The new consoles will face shortages, which will increase Wii U sales.
  2. The Wii U install base will grow.

Why post this in a thread about the Wii U's technical prowess? I think everyone can see why, but if not, allow me to conclude:

 

Ports from the Xbone are coming. The architectures are too similar, and the costs of developing for the PS4/Xbone will be too great to not have the game on every platform possible (these new games will not be like the PC games of console ports on very high, but rather, new, more ambitious titles).  The game market will not explode to the levels we saw in 2007-2009. The tablet did not exist, and the overall gaming options that you can get, for free, on PC were not as sophisticated (for casuals). Plus, the market is always changing, as people age and demographics shift (my 7 year old wanted an Ipad Mini over a Wii U:(). 

 

The point? This is a business. Money is the final arbiter. If I have this figured out, you can bet Nintendo does too.



#66 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 29 May 2013 - 08:26 PM

Guys,

 

I honestly believe the following:

 

  1. Nintendo's exit with the Wii hampered Wii U excitement.
  2. Confusion over the controller compounded with #1 has a lot to do with less enthusiasm. 
  3. Nintendo honestly believed the gamepad would be the next Wii Controller with consumer response. 
  4. Item #3 hampered focus from graphic fidelity to gamepad integration and focus. Hence the lack of tools (possibly one, non-intentional reason, but this could explain why things were so poor at launch).
  5. Finally, items 3&4 led third parties to believe functioning software, with gamepad integration, would provide a decent ROI. Given the low install base and probable attachment ratio, the investment did not need to be too high to begin with.  

Now, this is really item 6, and this deserves it's own paragraph. Publishers have watched year over year decreases in game sales. They anticipate the PS4 and Xbone to reverse this trend. Hence forth, they are spending as little as possible to bring ports to a new market (Wii U), while holding back until the 'big boys' launch. At this point, the Wii U's install base is bigger, and they know that one of two things will happen:

 

  1. The new consoles will face shortages, which will increase Wii U sales.
  2. The Wii U install base will grow.

Why post this in a thread about the Wii U's technical prowess? I think everyone can see why, but if not, allow me to conclude:

 

Ports from the Xbone are coming. The architectures are too similar, and the costs of developing for the PS4/Xbone will be too great to not have the game on every platform possible (these new games will not be like the PC games of console ports on very high, but rather, new, more ambitious titles).  The game market will not explode to the levels we saw in 2007-2009. The tablet did not exist, and the overall gaming options that you can get, for free, on PC were not as sophisticated (for casuals). Plus, the market is always changing, as people age and demographics shift (my 7 year old wanted an Ipad Mini over a Wii U:(). 

 

The point? This is a business. Money is the final arbiter. If I have this figured out, you can bet Nintendo does too.

They didn't think that the Wii U controller would cause the same amount of excitement as the Wii did, but they were under heavy pressure by investors to release new hardware, and fast, because sales of the Wii were declining rapidly.

 

Now, they had a general idea of where they would like to go with it hardware wise, but they weren't finished revising the hardware until mere months before launch.  As a matter of fact, it was the week of e3 2012.  This left them with little time for optimizing and revising the OS, stress testing and getting a solid feel for the hardware and how it performed in real world operations (theoreticals don't really help with the actual kit), or get decent tools out to developers who needed them to work on games with very little time to push to market.



#67 Nintyfan86

Nintyfan86

    Bob-omb

  • Members
  • 262 posts

Posted 29 May 2013 - 10:07 PM

They didn't think that the Wii U controller would cause the same amount of excitement as the Wii did, but they were under heavy pressure by investors to release new hardware, and fast, because sales of the Wii were declining rapidly.
 
Now, they had a general idea of where they would like to go with it hardware wise, but they weren't finished revising the hardware until mere months before launch.  As a matter of fact, it was the week of e3 2012.  This left them with little time for optimizing and revising the OS, stress testing and getting a solid feel for the hardware and how it performed in real world operations (theoreticals don't really help with the actual kit), or get decent tools out to developers who needed them to work on games with very little time to push to market.


I don't know, none of it adds up. Having a content filled 2010, followed by killing the Wii in 2011 with very little support, and expecting a half baked successor to fill the gap? I would think the board of directors would not approve.

However, they were not expecting the Wii to be a hit, and they had the HD console on stand by in case it flopped. Yet it was a huge success.

With what your saying, the only logical business strategy is to bank on the Wii brand name to stimulate early adoption. It seems as though it is the logical choice given the parameters.

But...on the technical end, it was not as simple as releasing a machine that played current ports better. Instead they had to optimize for games 2 years out and beyond.

So, I am left trying to reverse engineer their business strategy. The machine,s power is not up for debate until we see actual next gen titles on it IMO.

#68 Rockodoodle

Rockodoodle

    Chain Chomp

  • Members
  • 677 posts

Posted 30 May 2013 - 05:45 AM

Maybe there's not much money in the gamepad or it requires too many resources, but it would be cool if more than one could use it. I would think that a family would want more than one.  I realize they don't want to cannibalize their 3ds sales, but it would be cool if the gamepad also could be a portable system to a degree at least- maybe play eShop games on the fly....I would think that with hardware components perpetually dropping in price, some additional functionality would not be out of the question- at minimum there should be some Wii U/3ds integration....

 

I would think that when Apple TV comes out, people will be able to use their iPads as a controller for what they see on TV.....I think to a degree Nintendo made this controller in anticipation of that

 

 

They didn't think that the Wii U controller would cause the same amount of excitement as the Wii did, but they were under heavy pressure by investors to release new hardware, and fast, because sales of the Wii were declining rapidly.

 

Now, they had a general idea of where they would like to go with it hardware wise, but they weren't finished revising the hardware until mere months before launch.  As a matter of fact, it was the week of e3 2012.  This left them with little time for optimizing and revising the OS, stress testing and getting a solid feel for the hardware and how it performed in real world operations (theoreticals don't really help with the actual kit), or get decent tools out to developers who needed them to work on games with very little time to push to market.



#69 MorbidGod

MorbidGod

    Hammer Bro.

  • Members
  • 1,717 posts

Posted 30 May 2013 - 06:05 AM

Guys,

I honestly believe the following:

  • Nintendo's exit with the Wii hampered Wii U excitement.
  • Confusion over the controller compounded with #1 has a lot to do with less enthusiasm.
  • Nintendo honestly believed the gamepad would be the next Wii Controller with consumer response.
  • Item #3 hampered focus from graphic fidelity to gamepad integration and focus. Hence the lack of tools (possibly one, non-intentional reason, but this could explain why things were so poor at launch).
  • Finally, items 3&4 led third parties to believe functioning software, with gamepad integration, would provide a decent ROI. Given the low install base and probable attachment ratio, the investment did not need to be too high to begin with.
Now, this is really item 6, and this deserves it's own paragraph. Publishers have watched year over year decreases in game sales. They anticipate the PS4 and Xbone to reverse this trend. Hence forth, they are spending as little as possible to bring ports to a new market (Wii U), while holding back until the 'big boys' launch. At this point, the Wii U's install base is bigger, and they know that one of two things will happen:
  • The new consoles will face shortages, which will increase Wii U sales.
  • The Wii U install base will grow.
Why post this in a thread about the Wii U's technical prowess? I think everyone can see why, but if not, allow me to conclude:

Ports from the Xbone are coming. The architectures are too similar, and the costs of developing for the PS4/Xbone will be too great to not have the game on every platform possible (these new games will not be like the PC games of console ports on very high, but rather, new, more ambitious titles). The game market will not explode to the levels we saw in 2007-2009. The tablet did not exist, and the overall gaming options that you can get, for free, on PC were not as sophisticated (for casuals). Plus, the market is always changing, as people age and demographics shift (my 7 year old wanted an Ipad Mini over a Wii U:().

The point? This is a business. Money is the final arbiter. If I have this figured out, you can bet Nintendo does too.

Third party publishers like Activision expect the same thing to happen with the other consoles.

I don't know, none of it adds up. Having a content filled 2010, followed by killing the Wii in 2011 with very little support, and expecting a half baked successor to fill the gap? I would think the board of directors would not approve.

However, they were not expecting the Wii to be a hit, and they had the HD console on stand by in case it flopped. Yet it was a huge success.

With what your saying, the only logical business strategy is to bank on the Wii brand name to stimulate early adoption. It seems as though it is the logical choice given the parameters.

But...on the technical end, it was not as simple as releasing a machine that played current ports better. Instead they had to optimize for games 2 years out and beyond.

So, I am left trying to reverse engineer their business strategy. The machine,s power is not up for debate until we see actual next gen titles on it IMO.


No, they expected the Wii to be a success. They launched the DS as a test for the causal market. When that sold, the HD console you speak of ceased development and they finished the Wii concept.

That's why the DS was called the third pillar, and they said the Game Boy brand wasn't dead. You seen any new Game Boys lately? (still hoping for a return someday lol)

They didn't kill the Wii intentionally, it just ended up doing that. Nintendo made a lot of mistakes, but the graphics didn't suffer because of gamepad development. The only thing that hindered graphics is it was released in 2012. If it was released in 2013, it would be on the same level as the others and we wouldn't be having this conversation.

However, Nintendo released in 2012. And you know what? The difference between these three consoles, aren't that huge if a gap like Wii vs PS3 & 360.
Whovian12 -- Nintendo Network ID.

#70 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 30 May 2013 - 06:51 AM

Third party publishers like Activision expect the same thing to happen with the other consoles.


No, they expected the Wii to be a success. They launched the DS as a test for the causal market. When that sold, the HD console you speak of ceased development and they finished the Wii concept.

That's why the DS was called the third pillar, and they said the Game Boy brand wasn't dead. You seen any new Game Boys lately? (still hoping for a return someday lol)

They didn't kill the Wii intentionally, it just ended up doing that. Nintendo made a lot of mistakes, but the graphics didn't suffer because of gamepad development. The only thing that hindered graphics is it was released in 2012. If it was released in 2013, it would be on the same level as the others and we wouldn't be having this conversation.

However, Nintendo released in 2012. And you know what? The difference between these three consoles, aren't that huge if a gap like Wii vs PS3 & 360.

Pretty much what I was saying.  Well put.



Maybe there's not much money in the gamepad or it requires too many resources, but it would be cool if more than one could use it. I would think that a family would want more than one.  I realize they don't want to cannibalize their 3ds sales, but it would be cool if the gamepad also could be a portable system to a degree at least- maybe play eShop games on the fly....I would think that with hardware components perpetually dropping in price, some additional functionality would not be out of the question- at minimum there should be some Wii U/3ds integration....

 

I would think that when Apple TV comes out, people will be able to use their iPads as a controller for what they see on TV.....I think to a degree Nintendo made this controller in anticipation of that

They've already said that Wii U supports more than one gamepad, but it will come at a performance cost, because the GPU has to render three separate images at the same time, which is probably the reason they went with an SD screen for the gamepad, much less burden on the framebuffer and the GPU in general.



#71 Nintyfan86

Nintyfan86

    Bob-omb

  • Members
  • 262 posts

Posted 30 May 2013 - 01:18 PM

Third party publishers like Activision expect the same thing to happen with the other consoles.

No, they expected the Wii to be a success. They launched the DS as a test for the causal market. When that sold, the HD console you speak of ceased development and they finished the Wii concept.
That's why the DS was called the third pillar, and they said the Game Boy brand wasn't dead. You seen any new Game Boys lately? (still hoping for a return someday lol)
They didn't kill the Wii intentionally, it just ended up doing that. Nintendo made a lot of mistakes, but the graphics didn't suffer because of gamepad development. The only thing that hindered graphics is it was released in 2012. If it was released in 2013, it would be on the same level as the others and we wouldn't be having this conversation.
However, Nintendo released in 2012. And you know what? The difference between these three consoles, aren't that huge if a gap like Wii vs PS3 & 360.

I forgot about the third pillar, lol. I figured then that it was a fancy way of saying transitioning into a different demographic before expanding the product across the entire demographic. Including GBA BC made me think this way.

I did not know they expected immediate success with the Wii. I presumed they expected to expand into the blue ocean while offering the established market something the competition was not. Hence all of those Wii's gathering dust next to the HD Twins awaiting exclusives.

So,launching in 2012 makes sense with regards to the global economic recovery, yet 2011 fit better with Nintendo's product cycle.

What I have problems understanding, from an R&D standpoint, is how the architecture could not be finalized by E3 2011, with dev tools and software being developed from then on? I believe everything in it existed at this point, may be wrong. Clock speed would be all that remained.

I appreciate routerbad and yourself for being so generous with your time in regards to discussing these concepts.

#72 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 30 May 2013 - 01:24 PM

I forgot about the third pillar, lol. I figured then that it was a fancy way of saying transitioning into a different demographic before expanding the product across the entire demographic. Including GBA BC made me think this way.

I did not know they expected immediate success with the Wii. I presumed they expected to expand into the blue ocean while offering the established market something the competition was not. Hence all of those Wii's gathering dust next to the HD Twins awaiting exclusives.

So,launching in 2012 makes sense with regards to the global economic recovery, yet 2011 fit better with Nintendo's product cycle.

What I have problems understanding, from an R&D standpoint, is how the architecture could not be finalized by E3 2011, with dev tools and software being developed from then on? I believe everything in it existed at this point, may be wrong. Clock speed would be all that remained.

I appreciate routerbad and yourself for being so generous with your time in regards to discussing these concepts.

No problem at at all.  

 

The real reason why they didn't have final hardware much earlier will probably never be known.  Technically, none of the components existed until Nintendo and IBM developed them.  They are based on previous architectures, like Power7, PPC, and AMD Unified Shaders, but the final product is anything but.  

 

The target hardware for the wii u early on was rumored to be something like a high end HD4XXX series GPU, with a pure Power7 CPU.  Because of all of the difficult R&D that needed to be done to develop a wireless video standard with broadcom that worked with zero lag and no tearing, artifacts, etc, the GPU had to be designed at the same time to work flawlessly with that standard.  My guess is that it took them longer to develop the tech behind the gamepad than anything else, and that the GPU was finalized once they were settled on that.

 

Also, as 3Dude pointed out in another thread, they wanted eyefinity, which only debuted with very high end 5XXX series GPU's, and made more broadly available for HD6XXX.


Edited by routerbad, 30 May 2013 - 01:26 PM.


#73 MorbidGod

MorbidGod

    Hammer Bro.

  • Members
  • 1,717 posts

Posted 30 May 2013 - 07:07 PM

I forgot about the third pillar, lol. I figured then that it was a fancy way of saying transitioning into a different demographic before expanding the product across the entire demographic. Including GBA BC made me think this way.

I did not know they expected immediate success with the Wii. I presumed they expected to expand into the blue ocean while offering the established market something the competition was not. Hence all of those Wii's gathering dust next to the HD Twins awaiting exclusives.

So,launching in 2012 makes sense with regards to the global economic recovery, yet 2011 fit better with Nintendo's product cycle.

What I have problems understanding, from an R&D standpoint, is how the architecture could not be finalized by E3 2011, with dev tools and software being developed from then on? I believe everything in it existed at this point, may be wrong. Clock speed would be all that remained.

I appreciate routerbad and yourself for being so generous with your time in regards to discussing these concepts.


No problem at all ^_^.

I agree with what routerbad said completely. Just want to add, that, it's good all the tech wasn't done by E3 2011 in a sense. If it was, that would mean the Wii U in video card generations would be two generations behind instead of just one. And that could end up making a difference in graphics that people think exists today.

Although developers I am sure developers hated that.
Whovian12 -- Nintendo Network ID.

#74 Nintyfan86

Nintyfan86

    Bob-omb

  • Members
  • 262 posts

Posted 30 May 2013 - 08:57 PM

No problem at all ^_^.

I agree with what routerbad said completely. Just want to add, that, it's good all the tech wasn't done by E3 2011 in a sense. If it was, that would mean the Wii U in video card generations would be two generations behind instead of just one. And that could end up making a difference in graphics that people think exists today.

Although developers I am sure developers hated that.

 

 

No problem at at all.  

 

The real reason why they didn't have final hardware much earlier will probably never be known.  Technically, none of the components existed until Nintendo and IBM developed them.  They are based on previous architectures, like Power7, PPC, and AMD Unified Shaders, but the final product is anything but.  

 

The target hardware for the wii u early on was rumored to be something like a high end HD4XXX series GPU, with a pure Power7 CPU.  Because of all of the difficult R&D that needed to be done to develop a wireless video standard with broadcom that worked with zero lag and no tearing, artifacts, etc, the GPU had to be designed at the same time to work flawlessly with that standard.  My guess is that it took them longer to develop the tech behind the gamepad than anything else, and that the GPU was finalized once they were settled on that.

 

Also, as 3Dude pointed out in another thread, they wanted eyefinity, which only debuted with very high end 5XXX series GPU's, and made more broadly available for HD6XXX.

This may help others reading, and I know it will help my general understanding:

 

My impression of console development is that it gets finalized at some point. The foundation, so to speak, is there. In this case, Nintendo, AMD, and IBM know they are going with a GPU, most likely custom from the 6 series (2010), and the tri-core Power 7/750 base chip. Leaving the gamepad out for a moment, I would presume the only thing left for the box itself would be the clock speeds, amount of cache, and so forth. Or, put another way, that initial dev kit, with the 4850 inside, it should have been the target for the OS. 

 

As they got closer with the gamepad and finalized specs, I would have thought that the OS would have been patched along the way. 

 

What I am having trouble understanding is how the OS was in such a state at launch, and how the dev tools were to the point where launch games were using 2 cores. 

 

Note: I am also under the impression that the MCM is simply a customized e6760 and IBM  solution to fill in the gaps. Without a real target game from Nintendo to show off technical capabilities, it is hard for someone not really tech savvy (like myself) to rationalize the issues they have had outside of the gamepad development (which is an amazing achievement along with the low power consumption). 



#75 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 30 May 2013 - 09:11 PM

This may help others reading, and I know it will help my general understanding:

 

My impression of console development is that it gets finalized at some point. The foundation, so to speak, is there. In this case, Nintendo, AMD, and IBM know they are going with a GPU, most likely custom from the 6 series (2010), and the tri-core Power 7/750 base chip. Leaving the gamepad out for a moment, I would presume the only thing left for the box itself would be the clock speeds, amount of cache, and so forth. Or, put another way, that initial dev kit, with the 4850 inside, it should have been the target for the OS. 

 

As they got closer with the gamepad and finalized specs, I would have thought that the OS would have been patched along the way. 

 

What I am having trouble understanding is how the OS was in such a state at launch, and how the dev tools were to the point where launch games were using 2 cores. 

 

Note: I am also under the impression that the MCM is simply a customized e6760 and IBM  solution to fill in the gaps. Without a real target game from Nintendo to show off technical capabilities, it is hard for someone not really tech savvy (like myself) to rationalize the issues they have had outside of the gamepad development (which is an amazing achievement along with the low power consumption). 

The original prototypes and the software environment was built around weaker hardware than ended up in the system, that is true.

 

Because it took so long to get the hardware finalized, the OS itself was not optimized for it by the time they were preparing for launch.  This is a big reason behind the utter lack of marketing, Nintendo are sandbagging the Wii U intentionally (they know early adopters will buy into it anyway, they just thought there would be more) because they themselves felt the software environment was not ready.

 

The real problem is that the GPU is so very divergent from the original target hardware, that all of the tools would have had to be rewritten both to take advantage of the GPU at a basic level, and to properly use the gamepad.  For the gamepad they needed to essentially upgrade the target hardware in order to get the performance they were looking for with the gamepad, and AMD tech like eyefinity fits perfectly.

 

It isn't just a customized 6760, its the other way around, in fact.  Its a custom chip that happens to utilize the AMD unified shader cores from the 6760 that they licensed.  It's a brand new animal that no one was prepared for.

 

I assume they still haven't nailed down the best clock speeds, and the cache on die was based on their budget for the chip.  They really thought of everything, and companies like Shin'en that really understand GPU tech through and through are able to be authoritative on the fact that there is plenty of power there if you know how to use the system properly, and everything is designed for extreme efficiency.  No clock cycles wasted, as it were.

 

The great thing is that many of us are already satisfied with how the system basically operates, and that says something, because Nintendo isn't, and they will continue to improve all aspects.


Edited by routerbad, 30 May 2013 - 09:15 PM.


#76 Nintyfan86

Nintyfan86

    Bob-omb

  • Members
  • 262 posts

Posted 30 May 2013 - 09:58 PM

The original prototypes and the software environment was built around weaker hardware than ended up in the system, that is true.

 

Because it took so long to get the hardware finalized, the OS itself was not optimized for it by the time they were preparing for launch.  This is a big reason behind the utter lack of marketing, Nintendo are sandbagging the Wii U intentionally (they know early adopters will buy into it anyway, they just thought there would be more) because they themselves felt the software environment was not ready.

 

The real problem is that the GPU is so very divergent from the original target hardware, that all of the tools would have had to be rewritten both to take advantage of the GPU at a basic level, and to properly use the gamepad.  For the gamepad they needed to essentially upgrade the target hardware in order to get the performance they were looking for with the gamepad, and AMD tech like eyefinity fits perfectly.

 

It isn't just a customized 6760, its the other way around, in fact.  Its a custom chip that happens to utilize the AMD unified shader cores from the 6760 that they licensed.  It's a brand new animal that no one was prepared for.

 

I assume they still haven't nailed down the best clock speeds, and the cache on die was based on their budget for the chip.  They really thought of everything, and companies like Shin'en that really understand GPU tech through and through are able to be authoritative on the fact that there is plenty of power there if you know how to use the system properly, and everything is designed for extreme efficiency.  No clock cycles wasted, as it were.

 

The great thing is that many of us are already satisfied with how the system basically operates, and that says something, because Nintendo isn't, and they will continue to improve all aspects.

I see, so it is not as simple as releasing specs for software, and making the hardware engineers work around it (this was also playing into my thoughts, like how all different parts will work with Windows regardless of architecture changes within the x86-64 spectrum to a point). Rapping your head around a closed system is very difficult. It is like capital budgeting, but locking yourself in without any alterations, for years. I never considered it from this angle. 

 

The different levels of memory cache are interesting as well. What is even more pertinent is that Microsoft is using a similar strategy, yet Sony is going with the GDDR5. I have read, but have not fully understood, the concerns on Sony's approach. I also have trouble understanding how 8 Jaguar cores and a 7970m are going to merge into an APU. 

 

One aspect with Sony's system that seems to be apparent is that the power will be exploitable from day one, presuming a game exists that can take advantage of it. However, their approach seems unbalanced to me. I do not know how that cpu/gpu combo will be able to use half of that ram without running into bottlenecks of it's own. Not to mention the obvious reserves for the OS (which I suspect some of those cores/gpu compute unites to be reserved for, and why MS is lower spec in comparison for undoubted OS efficiency vs Sony). 



#77 MorbidGod

MorbidGod

    Hammer Bro.

  • Members
  • 1,717 posts

Posted 31 May 2013 - 05:06 AM

This may help others reading, and I know it will help my general understanding:

My impression of console development is that it gets finalized at some point. The foundation, so to speak, is there. In this case, Nintendo, AMD, and IBM know they are going with a GPU, most likely custom from the 6 series (2010), and the tri-core Power 7/750 base chip. Leaving the gamepad out for a moment, I would presume the only thing left for the box itself would be the clock speeds, amount of cache, and so forth. Or, put another way, that initial dev kit, with the 4850 inside, it should have been the target for the OS.

As they got closer with the gamepad and finalized specs, I would have thought that the OS would have been patched along the way.

What I am having trouble understanding is how the OS was in such a state at launch, and how the dev tools were to the point where launch games were using 2 cores.

Note: I am also under the impression that the MCM is simply a customized e6760 and IBM solution to fill in the gaps. Without a real target game from Nintendo to show off technical capabilities, it is hard for someone not really tech savvy (like myself) to rationalize the issues they have had outside of the gamepad development (which is an amazing achievement along with the low power consumption).


First question: OS, why was it in such a bad shape? Honestly, Nintendo sucks at OS creation. The Wii was vwry basic, and the Wii U was an upgrade to it's design. I think they should use Linux, but they haven't listened yet.

As for the developer tools, Nintendo had no excuse. I don't know why they didn't give proper help to the developers. They learned from that mistake, but it was too late.

Hopefully that helps.
Whovian12 -- Nintendo Network ID.

#78 tboss

tboss

    Pokey

  • Members
  • 1,242 posts

Posted 31 May 2013 - 05:27 AM

ninty said parts of it was developed by different and completly unrelated teams. i think ninty didnt start the OS till close to the last second when  they new what the hardware was, but still waited to long.

 

or didnt want to spend R&D working on it till the wiiU was finished.


Edited by tboss, 31 May 2013 - 05:27 AM.


#79 routerbad

routerbad

    Lakitu

  • Section Mods
  • 2,013 posts
  • NNID:routerbad
  • Fandom:
    Zelda, Mario, Halo, Star Trek

Posted 31 May 2013 - 06:11 AM

I see, so it is not as simple as releasing specs for software, and making the hardware engineers work around it (this was also playing into my thoughts, like how all different parts will work with Windows regardless of architecture changes within the x86-64 spectrum to a point). Rapping your head around a closed system is very difficult. It is like capital budgeting, but locking yourself in without any alterations, for years. I never considered it from this angle. 

 

The different levels of memory cache are interesting as well. What is even more pertinent is that Microsoft is using a similar strategy, yet Sony is going with the GDDR5. I have read, but have not fully understood, the concerns on Sony's approach. I also have trouble understanding how 8 Jaguar cores and a 7970m are going to merge into an APU. 

 

One aspect with Sony's system that seems to be apparent is that the power will be exploitable from day one, presuming a game exists that can take advantage of it. However, their approach seems unbalanced to me. I do not know how that cpu/gpu combo will be able to use half of that ram without running into bottlenecks of it's own. Not to mention the obvious reserves for the OS (which I suspect some of those cores/gpu compute unites to be reserved for, and why MS is lower spec in comparison for undoubted OS efficiency vs Sony). 

The GPU will not be a 7970, for starters, at the very most we're looking at a 7770.  Sony is big on putting out numbers that tend to not be true in real life situations, so I wouldn't even count on that at this point.  They claimed RSX had 1.8TFLOPS as well.

 

You are right in assuming that they'll run into hardware bottlenecks long before memory bottlenecks, where those hardware bottlenecks are is anyone's guess at this point.

 

The reason MS went with a lower spec, and even Nintendo went with a lower spec is that much of that additional power will be wasted.  PS4 having more power will not show on screen due to the diminishing returns we're seeing in the GPU space, and why GPU makers just add more shaders linearly along with more RAM for their more powerful cards.  Those cards are MUCH more power hungry, run MUCH hotter, yet produce maybe a few more frames than a midrange or budget card.  The benefit of too much power in a closed system like this is limited.  You have to find a balance that works.



First question: OS, why was it in such a bad shape? Honestly, Nintendo sucks at OS creation. The Wii was vwry basic, and the Wii U was an upgrade to it's design. I think they should use Linux, but they haven't listened yet.

As for the developer tools, Nintendo had no excuse. I don't know why they didn't give proper help to the developers. They learned from that mistake, but it was too late.

Hopefully that helps.

Exactly, Microsoft is an OS company, and that shows through brilliantly on the XBONE.

 

Nintendo is a game company, OS's and online infrastructures aren't in their natural wheelhouse, though I'm sure they're learning very quickly to make it their wheelhouse. 



#80 BrosBeforeGardenTools

BrosBeforeGardenTools

    Hot dog vendor who spilled condiments on himself

  • Members
  • 1,864 posts

Posted 31 May 2013 - 06:16 AM

I'm going to go ahead and bring up the argument I used against a friend. We are not at the point of marginal returns. Console makers make their consoles less powerful so as not to lose so much money and then want you to believe that. Until we are at a molecular level in games, there will always be room for improvement.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

Anti-Spam Bots!