65 replies to this topic

[megafenix]

No, No. No. No. Good god no. You cant just make crap up.

1. Marcan NEVER said the words you are putting in his mouth, he NEVER said the 476 and the 7xx series were incompatable. He said the gekko and the broadway, custom NINTENDO ONLY IMPROVED GAMING PROCESSOR BASED ON THE 750 LINE, was incompatable with the 476 series. Which is true, the gekko and the broadway are also incompatable with OTHER MEMBERS OF THE 750 LINE. They can run legacy 750 code, but other 750 processors CAN NOT RUN GEKKO OR BROADWAY PPC CODE BECAUSE THE OTHER 750 PROCESSORS DO NOT HAVE PAIRED SINGLES, and neither can or does the 476.

2. You dont understand what you are talking about. All newer processors in a series is compatable with older power processors, Of the same series. Far, FAR less processors are cross series compatable, despite them all using power isa, the differences in architecture prevents it.

3. No, the 476 would NOT be much much easier, it would be much much harder, because changing the entire architecture of the chip to be compatable with the customized nintendo specific architecture of the 750 series, with things like paired singles is faaaaaaaar more work than simply customizing the custom nintendo 750 to utilize modern technology unavailable in 1999, to support multicore.

4. Why do you keep insisting on a weaker cpu core than what is in the wii u? Seriously, you are looney toons.

 

 

go to the twiter and verify yourself, he said so

 

 

Seriously, pay attention, dont just bark, go on, just go to his twiter and look for the paired singles thing

Edited by megafenix, 03 December 2013 - 04:12 PM.

[3Dude]

go to the twiter and verify yourself, he said so

Thats right, you JUST CONFIRMED WHAT I JUST SAID. PAIRED SINGLES ARE ONLY FOR THE CPU'S IN THE GAME CUBE, AND THE WII. NOTHING ELSE HAS THEM, NOT EVEN OTHER MEMBERS OF THE 750 SERIES. WITHOUT PAIRED SINGLES YOU CANT RUN CUBE OR WII CODE.

[megafenix]

Thats right, you JUST CONFIRMED WHAT I JUST SAID. PAIRED SINGLES ARE ONLY FOR THE CPU'S IN THE GAME CUBE, AND THE WII. NOTHING ELSE HAS THEM, NOT EVEN OTHER MEMBERS OF THE 750 SERIES. WITHOUT PAIRED SINGLES YOU CANT RUN CUBE OR WII CODE.

 

 

"

1. Marcan NEVER said the words you are putting in his mouth, he NEVER said the 476 and the 7xx series were incompatable. He said the gekko and the broadway, custom NINTENDO ONLY IMPROVED GAMING PROCESSOR BASED ON THE 750 LINE, was incompatable with the 476 series. Which is true, the gekko and the broadway are also incompatable with OTHER MEMBERS OF THE 750 LINE. They can run legacy 750 code, but other 750 processors CAN NOT RUN GEKKO OR BROADWAY PPC CODE BECAUSE THE OTHER 750 PROCESSORS DO NOT HAVE PAIRED SINGLES, and neither can or does the 476.

 

"

 

nope, you just confirmed what i said, ibm modified the old 750cxe so that gekko could support paired singles and could have done the same to the 476fp fpu

 

marcan said that 476fp fpu doesnt support paired singles and that would have been an obstacle, clearly teh guy hadnt read ign report from IBM tech experts about what they did on gekko and so they could also have done to the 476fp fpu

 

 

please understand

http://www.ign.com/articles/2001/12/13/interview-ibm-details-gekko-part-i
"
Interview: IBM Details Gekko (Part I)

Two of the main designers behind GameCube's Gekko CPU give new insights in this IGNcube interview.

by IGN Staff

DECEMBER 12, 2001

IGNcube: Were you both involved with the design of Gekko?

 

IBM, Mike West:

Yes we were. I'm actually a multimedia architect. My background is graphics and multimedia, and I've been working closely with Nintendo for -- probably -- it's got to be close to four years now. And Pete, he is our lead PowerPC architect who actually worked on the detailed architecture of the Gekko chip for the changes we made, over and above the PowerPC architecture. So Pete led the [design] of the actual chip architecture and I worked with Nintendo across the system.

 

IGNcube: There's a sort of general conception that Gekko is just a basic PowerPC 750 and IBM didn't really modify the core that much. Is there anything you would say to dispel that conception?

 

IBM, Mike West:

There have been some recent articles on the web that have that view, which is a little surprising. We presented a lot of details at Hot Chips and the Embedded Processor Forum, both public domains. But, Gekko is a lot more than just a PowerPC 750 and it is an integral part of the system.

 

IBM, Peter Sandon:

 Nintendo came to us and said, "Here's what we need for a graphics system. We need FPU (floating point) performance beyond what will be built into the graphics chip itself and we need to move data." So the main things we concentrated on where those two: moving data through the system as it's processed and a high level of floating point performance in order to do some of the things one might expect are done on the graphics chip, but in order to do more custom kind of lighting and geometry and so-forth that can't be done in some of the fixed-function graphics pipeline on the graphics processor, we enhanced our floating point performance.

 

IGNcube: At the Embedded Processor Forum it was said that Gekko was used for the "close-up lighting in Luigi's Mansion." What kind of assistance does Gekko provide with that and other graphics functions?

 

IBM, Peter Sandon:

Well it basically does the local lighting calculations. The lighting that's implemented in the graphics chip is for lights at infinity. So, if you want to get the effects that you get from so-called local lights that don't look like they're out at infinity, those calculations are done on Gekko.

 

IGNcube: How much flexibility does Gekko have to assist in these kinds of graphical functions? How hard does it hit the processor when you start doing something like that.

 

IBM, Mike West:

You have to understand what's done with the floating point unit. It is the conventional 64-bit PowerPC floating point unit, but the adaptations we made to it allow two simultaneous 32-bit calculations to occur. Basically each instruction is completed every cycle.

 

IBM, Peter Sandon: 

Right, and I should say that it is a significant effort to implement the paired-single floating point function beyond what is there in the standard PowerPC.

"

 

 

And Is not like IBM had to start from the scratch, just take the fpu from broadway and put it and adjust it on the 476fp

 

Still beleive they would be incompatible?

here, ask IBM

https://www-01.ibm.c.../4xx_6xx_an.pdf

"

 

Programming Model Differences of

the IBM PowerPC 400 Family and

600/700 Family Processors.

 

 

Abstract – This application note describes how the programming models differ between the IBM

400 family and the 600/700 family of PowerPC architecture processor implementations. It is

useful to system designers and programmers porting code from one family to another.

The 600/700 family is compliant with the original PowerPC processor architecture designed for

desktop applications.

 

The 400 family, designed for use in embedded environments, maintains

compatibility in the base User Instruction Set architecture. However, it defines separate Virtual

Environment and Operating Environment architectures that have been enhanced for embedded

applications. These optimizations have prompted changes in features such as memory

management, debug support, and cache access and control, and processor extendibility.

 

12 Conclusion

The PowerPC Architecture provides a high degree of code portability between different implementations

even when the implementations have significant differences in their intended applications. Porting code

between implementations is a straightforward and manageable job that is primarily limited to the kernel

and other supervisory mode code. Applications typically require almost no changes in order to be moved

to a different target processor than that which they were originally developed for.

 

Porting between processors can be greatly facilitated by the use of structured programming methods,

which then minimizes the amount of code needing modification. For example, if all access to the Time

Base is done using a single function which is used by the entire application and operating system, only

that function needs to be modified when porting to another PowerPC processor.

 

 

"

Edited by megafenix, 03 December 2013 - 04:29 PM.

[3Dude]

megafenix, on 03 Dec 2013 - 7:17 PM, said:
"
1. Marcan NEVER said the words you are putting in his mouth, he NEVER said the 476 and the 7xx series were incompatable. He said the gekko and the broadway, custom NINTENDO ONLY IMPROVED GAMING PROCESSOR BASED ON THE 750 LINE, was incompatable with the 476 series. Which is true, the gekko and the broadway are also incompatable with OTHER MEMBERS OF THE 750 LINE. They can run legacy 750 code, but other 750 processors CAN NOT RUN GEKKO OR BROADWAY PPC CODE BECAUSE THE OTHER 750 PROCESSORS DO NOT HAVE PAIRED SINGLES, and neither can or does the 476.

"

nope, you just confirmed what i said, ibm modified the old 750cxe so that gekko could support paired singles and could have done the sqae to the 476fp fpu

marcan said that 476fp doesnt support paired singles and that would have been an obstacle, clearly teh guy hadnt read ign report from IBM tech experts

HE DOESNT NEED TO READ GARBAGE REPORTS FROM MORONIC GAME JOURNALISTS, HE KNOWS EVERYTHING THAT GOES ON IN THE GEKKO AND THE BROADWAY. ITS ALMOST LIKE HES THE PERSON WHO HACKED THEM OR SOEMTHING. OF COURSE HE ALREADY KNOWS THAT INFORMATION. EVERYBODY ALREADY KNOWS THAT INFORMATION. HE WAS THE FIRST ONE TO SPECIFICALLY TELL YOU IT WAS A FEATURE CUSTOM DONE TO NINTENDO FOR THE GAMECUBE.

You are holding onto a worthless pile of nothing, as if it were some golden terd of salvation dangling out of Jesus' butt cheecks. Its not, its nothing. EVERYBODY who has EVER seriously discussed nintendo hardware for the past DECADE knows about that, to the point it is old information no longer needed to even be brought up... EXCEPT FOR YOU.

Attaining backwards compatability via attempting to 'broadwaify' another, completely different architecture, that was already weaker than what they could do with the chip that already had 100% perfect backwards compatability is just. Its stupid. You can NOT just cut off the fpu customizations and slap it onto another architecture with 'a little adjusting' and get 1005 garunteed backwards compatability with a completely different architecture!!! IT Does NOT work like that! And the fact you keep bringing it up over and over and over, and posting these links you clearly dont understand, that do nothing whatsoever, over and over and over again. Never stopping, never changing, never gfazed no matter how many times you are blatantly proven wrong, you kust keep posting made up crap after made up crap.

This is enough.

Come back again after you get some self control Eye of core. I have been MORE than patient with you.

[megafenix]

"

nope, you just confirmed what i said, ibm modified the old 750cxe so that gekko could support paired singles and could have done the sqae to the 476fp fpu

marcan said that 476fp doesnt support paired singles and that would have been an obstacle, clearly teh guy hadnt read ign report from IBM tech experts

HE DOESNT NEED TO READ GARBAGE REPORTS FROM MORONIC GAME JOURNALISTS, HE KNOWS EVERYTHING THAT GOES ON IN THE GEKKO AND THE BROADWAY. ITS ALMOST LIKE HES THE PERSON WHO HACKED THEM OR SOEMTHING. OF COURSE HE ALREADY KNOWS THAT INFORMATION. EVERYBODY ALREADY KNOWS THAT INFORMATION. HE WAS THE FIRST ONE TO SPECIFICALLY TELL YOU IT WAS A FEATURE CUSTOM DONE TO NINTENDO FOR THE GAMECUBE.

You are holding onto a worthless pile of nothing, as if it were some golden terd of salvation dangling out of Jesus' butt cheecks. Its not, its nothing. EVERYBODY who has EVER seriously discussed nintendo hardware for the past DECADE knows about that, to the point it is old information no longer needed to even be brought up... EXCEPT FOR YOU.

Attaining backwards compatability via attempting to 'broadwaify' another, completely different architecture, that was already weaker than what they could do with the chip that already had 100% perfect backwards compatability is just. Its stupid. And the fact you keep bringing it up over and over and over, and posting these links you clearly dont understand, that do nothing whatsoever, over and over and over again. Never stopping, never changing, never gfazed no matter how many times you are blatantly proven wrong, you kust keep posting made up crap after made up crap.

This is enough.

Come back again after you get some self control Eye of core. I have been MORE than patient with you.

 

 

nope, you just confirmed what i said, ibm modified the old 750cxe so that gekko could support paired singles and could have done the same to the 476fp fpu

 

marcan said that 476fp fpu doesnt support paired singles and that would have been an obstacle, clearly teh guy hadnt read ign report from IBM tech experts about what they did on gekko and so they could also have done to the 476fp fpu

 

 

please understand, the answer is not just in the ign interview, but also in the users manual you will see later

http://www.ign.com/articles/2001/12/13/interview-ibm-details-gekko-part-i
"
Interview: IBM Details Gekko (Part I)

Two of the main designers behind GameCube's Gekko CPU give new insights in this IGNcube interview.

by IGN Staff

DECEMBER 12, 2001

IGNcube: Were you both involved with the design of Gekko?

 

IBM, Mike West:

Yes we were. I'm actually a multimedia architect. My background is graphics and multimedia, and I've been working closely with Nintendo for -- probably -- it's got to be close to four years now. And Pete, he is our lead PowerPC architect who actually worked on the detailed architecture of the Gekko chip for the changes we made, over and above the PowerPC architecture. So Pete led the [design] of the actual chip architecture and I worked with Nintendo across the system.

 

IGNcube: There's a sort of general conception that Gekko is just a basic PowerPC 750 and IBM didn't really modify the core that much. Is there anything you would say to dispel that conception?

 

IBM, Mike West:

There have been some recent articles on the web that have that view, which is a little surprising. We presented a lot of details at Hot Chips and the Embedded Processor Forum, both public domains. But, Gekko is a lot more than just a PowerPC 750 and it is an integral part of the system.

 

IBM, Peter Sandon:

 Nintendo came to us and said, "Here's what we need for a graphics system. We need FPU (floating point) performance beyond what will be built into the graphics chip itself and we need to move data." So the main things we concentrated on where those two: moving data through the system as it's processed and a high level of floating point performance in order to do some of the things one might expect are done on the graphics chip, but in order to do more custom kind of lighting and geometry and so-forth that can't be done in some of the fixed-function graphics pipeline on the graphics processor, we enhanced our floating point performance.

 

IGNcube: At the Embedded Processor Forum it was said that Gekko was used for the "close-up lighting in Luigi's Mansion." What kind of assistance does Gekko provide with that and other graphics functions?

 

IBM, Peter Sandon:

Well it basically does the local lighting calculations. The lighting that's implemented in the graphics chip is for lights at infinity. So, if you want to get the effects that you get from so-called local lights that don't look like they're out at infinity, those calculations are done on Gekko.

 

IGNcube: How much flexibility does Gekko have to assist in these kinds of graphical functions? How hard does it hit the processor when you start doing something like that.

 

IBM, Mike West:

You have to understand what's done with the floating point unit. It is the conventional 64-bit PowerPC floating point unit, but the adaptations we made to it allow two simultaneous 32-bit calculations to occur. Basically each instruction is completed every cycle.

 

IBM, Peter Sandon: 

Right, and I should say that it is a significant effort to implement the paired-single floating point function beyond what is there in the standard PowerPC.

"

 

 

And Is not like IBM had to start from the scratch, just take the fpu from broadway and put it and adjust it on the 476fp

 

Still beleive they would be incompatible?

here, ask IBM

https://www-01.ibm.c.../4xx_6xx_an.pdf

"

 

Programming Model Differences of

the IBM PowerPC 400 Family and

600/700 Family Processors.

 

 

Abstract – This application note describes how the programming models differ between the IBM

400 family and the 600/700 family of PowerPC architecture processor implementations. It is

useful to system designers and programmers porting code from one family to another.

The 600/700 family is compliant with the original PowerPC processor architecture designed for

desktop applications.

 

The 400 family, designed for use in embedded environments, maintains

compatibility in the base User Instruction Set architecture. However, it defines separate Virtual

Environment and Operating Environment architectures that have been enhanced for embedded

applications. These optimizations have prompted changes in features such as memory

management, debug support, and cache access and control, and processor extendibility.

 

12 Conclusion

The PowerPC Architecture provides a high degree of code portability between different implementations

even when the implementations have significant differences in their intended applications. Porting code

between implementations is a straightforward and manageable job that is primarily limited to the kernel

and other supervisory mode code. Applications typically require almost no changes in order to be moved

to a different target processor than that which they were originally developed for.

 

Porting between processors can be greatly facilitated by the use of structured programming methods,

which then minimizes the amount of code needing modification. For example, if all access to the Time

Base is done using a single function which is used by the entire application and operating system, only

that function needs to be modified when porting to another PowerPC processor.

 

 

"

 

your attempts are useless, you may have marcan, but I have IBM, and if marcan says something that goes against IBMs words then something is fishy

Edited by megafenix, 03 December 2013 - 04:47 PM.

[3Dude]

Oh my god. Seriously you have no idea what you are reading.

They are talking about the power isa instruction level code base. That does not matter. AT all. It does not mean what you think it does, in the slightest, at all. That will do nothing, whatsoever, for physical hardware architecture differentiations JUST LIKE IT STATES IN THE PART YOU FREAKING UNDERLINED.

To put it more precisely, if your architecture is so unique that it has dozens of calls initializations, operations, units and specializations NOT PRESENT in other architectures you wish to port to, you CANT simply straightforward copy paste code. Simple instructions that only use parts of the machine common in both processors, will run, but anything unique to one processor and not the other is a DEAD END.

Its just describing power instruction architecture man. There is no secret here, its not some super proof proof, its Performance Optimization With Enhanced Risc Instruction Set Architecture. OF COURSE the instruction set architecture is going to be the same for all those processors! That does NOT mean what you are trying to state it means.

Its the same damn thing as x64/86 is to intel processors. all those changes you would need to make are handled by the damn compilers anyways unless you want to go into assembler, but there is no way in hell sharing the same basic instruction core is going to magically make code designed for parts of hardware that dont exist in other processors run. And heavily customizing a completely different architecture to match a COMPLETELY different architecture is a HUGE investment of resources, which makes absolutely no sense when you are using a less spowerful processor, why? WHY? WHY would anyone want to spend so much money and go so far out of their way to make a weaker processor that would NOT have 100% backwards compatability no matter how hard they tried, when they could simply make very basic, very easy modifications enabled by modern technology to improve performance and enable multicore coherency with the original line heavily customized for use in videogames? If this was such a simple matter why is EVERY SINGLE OTHER SOLUTION TO BACKWARDS COMPATABILITY TO STICK THE ORIGINAL CPU OF THE OLD PRODUCT INTO THE NEW PRODUCT OR SOFTWARE EMULATE? ITs because your shenanigans are not a real thing! Thats not something that is practical to do.

This is done. Over. You have done nothing, but post the same unending garbage back to back to back. There is no longer any point to this, you have never brought anything to this discussion, again, I have been MORE than patient with you. You are done for now.

In fact, this thread is done for now. At least until the rest of that documentation surfaces, it seems its original purpose has ran its course anyways. We got all the information out of the actual official overview page of the espresso documentation.

its only useful now for bringing back old trolls to try and use new tricks.