Backwards Compatibility: poll

[TheChefO]

My approach is to consider the problem unsolvable and will wait until there's enough information to go on, which in this case there will never be.

Such an approach isn't suitable for any kind of prediction/design. One has to take the data which is available, lay it all on the table, weigh the pros and cons of different options.

In the months since talking about this, some of the knowledgeable members have introduced some interesting alternate hardware.

My take on such an approach is this: If the hardware is fast enough to emulate a foreign architecture in realtime, then it would seem a suitable upgrade to dump any legacy architecture as the new one has proven to thoroughly outclass the alternate.

Similar to how there are emulators out for many older consoles which can run on modern PCs and some even run on a smart phone.

Obviously, ARM doesn't have a 6800 or z80 in it, yet the ARM used in most android devices is so fast, it can emulate the hardware without breaking a sweat.

When someone can introduce an architecture which can emulate Xenos/Cell without issue (and within the die budget) then it will have proven to be a suitable replacement and at that point I don't care if it's x86, MIPs, Alpha, Cray, or ARM.

As long as it fits the budget and isn't taking away die space from the GPU.

All I can talk about are generalisms. If you won't accept them without these impossible datapoints then that's that. Especially when you've decided BC is just a matter of picking a CPU with a compatible ISA.

Not just a compatible ISA, it would also have to have VMX128 (or better), the cache locking feature (or suitable replacement), at least 6 HW threads, and the ability to chew through each thread at least as fast as Xenon so that games aren't stuttering along.

Cell is a different animal, as your Cell emulation thread found, the SPE is pretty much impossible to emulate on a mm vs mm basis. So, Sony can:

1) Completely ignore the desires of Devs that have complained about Cell and literally just ramp it up (6-16SPE)

2) Modify SPE's with larger local stores, OOOE, or more registers.

3) Shoot themselves in the foot and drop SPE's and deal with the repercussions of destroying the concept of a platform that Sony had built up since 1994.


I personally think Sony should keep the SPE's on board and find a use for them. Even if Devs NEVER have to directly access the SPE's in the design, Sony should still include them and find a way to easily utilize them.

- Sony could institute a mandatory hardware AA method which utilizes the minimum 6 SPEs.
- Sound obviously fits in the SPE mold.
- A Physics Middleware kit which runs on the SPE's would also be a great venture for Sony to handle for their devs.

With Sony instituting the minimum 6 SPE's needed for compatibility, they can use the rest of the die budget for Better/More PPE's. Bottom line, Cell is an interesting and potentially dominant performer. To dump it would destroy BC, would negate all the code bases and technique lessons which devs have developed over the past 6 years, would be a waste of all the Cell R&D investment up to this point, and a waste of the IP.

All of this, for what alternative?

If it's something which is powerful enough to emulate Cell (not likely), then great. If not, seems a bit backwards innit?

Why doesn't the GPU get a mention?

The XENON is indeed a tricky proposition with all of the oddball functionality which was thrown in, MEMEXPORT, the tessellator, not to mention the EDRAM/logic. However, AMD/ATI is handling the new GPU, so I'm assuming they know their baby better than anyone and will not have an issue replication the functionality with GCN's more flexible and programmable nature.

GCN/Tesla should have Zero issue emulating the machine code of RSX. Sony may just have to pony up for the licensing fee to Nvidia if the go with GCN.

Now you've shifted the goalposts a little. If you agree that Cell is better than Xenon, and PS4 with Cell2+GPU will be better than XB3 with Xenon2+same GPU, then there is a cost for scaling XB360; it'll lose performance.

No, it actually isn't losing performance. Xbox never had Cell.

I'm not convinced MS would be willing to license the tech from Sony and vice versa. Nor am I convinced it is CLEARLY the better CPU. Harder to dev for, but higher theoretical throughput. I've explained my vision for how to handle Cell in ps4 above.

It could be the better CPU, but many of the performance advantages which Cell has can be lessened via VMX type execution units.

It's up to Sony/MS to take advantage of the CPU architectures they have and improve/modify them where they feel necessary. But I'm not convinced either architecture is so inefficient that it would be better to start over from scratch rather than modify/extend what is already there.

 

[TheChefO]

I have never cared about BC until this generation. If the literal hundreds of digital downloads I have purchased can't move over then I'm not either. I don't make a habit of keeping previous gen systems if I have moved on to the next generation. I have about 25+ games from Games on Demand and 2 or 300 XBLA titles so I will either be sticking with the current platform for quite some time or until such time that enough software/services have interested me enough to want to jump to the new generation of hardware. Now if the new generation has a new box and then some portable hdd/apple tv sized dongle or whatever that gives me full BC then I'll make the move.

You're not alone in this opinion.

Digital distribution and online interactivity are more important now than ever, and by-design.

For Sony or MS to not cater to the market THEY CREATED would be very poor form.

IMO.

 

[Shifty Geezer]

No, it actually isn't losing performance. Xbox never had Cell.

Your thread was saying how different BC approaches cost 25, 12.5 and 0% of 'power' from within a console performance budget. That needs to be measured against the alternatives. If MS going with Xenon2 means 20% less 'power' than if they went with Cell, that's a 20% cost of 'power' for scaling their hardware and keeping BC.

Also, of fundamental importance, is whether 'raw power' is much use. A choice of another processor architecture might mean less power but more functionality, flexibility, cost effectiveness (retail price), etc.

It could be the better CPU, but many of the performance advantages which Cell has can be lessened via VMX type execution units.

It could also be that whatever Cell's performance advantages are in some scenarios are mitigated by the majority of workloads and Sony should just ditch Cell and go with a stronger single thread performance for the best next-gen experience. That's something your poll isn't really measuring - impact on the next-gen experience in supporting the previous gen experience. Yes, I know, you feel there's a whole platform that needs to carry over and the company that doesn't run their old games will lose to the rival that does. That discussion has been had; no need to repeat it.

It's up to Sony/MS to take advantage of the CPU architectures they have and improve/modify them where they feel necessary.

Why? Why should they improve on existing architectures instead of be free to try anything else? What if...Cray's barrel-processor has appeal, and implementing it would be much cheaper and easier if no regard is paid to running existing Cell or Xenon code? Why shouldn't Sony take advantage of their new experience with ARM and PVR on Vita, or MS build on their mobile and PC hardware support?

But I'm not convinced either architecture is so inefficient that it would be better to start over from scratch rather than modify/extend what is already there.

No architecture that finds it's way into a successful console is going to be that bad that it's not competitive. But that doesn't make it the instant choice for the sequel. The only reason to consider updating the existing systems without regard for other options is to support BC. Forget about BC and you have a whole world of options open up, around which you can engineer some BC if needed. Perhaps within those evaluations the idea of scaling up your console makes sense, such as Nintendo and Wii where they intended to compete outside of the power race. More often than not the idea of scaling up your old hardware is poor though hence whey PS2 didn't just scale PS1 and PS3 didn't just scale PS2. There's always the option of dual SKU's with added hardware for those who still want BC. That comes at 0% power cost in your poll too.

 

[TheChefO]

Your thread was saying how different BC approaches cost 25, 12.5 and 0% of 'power' from within a console performance budget. That needs to be measured against the alternatives.

Indeed. But Cell isn't an alternative for MS. At least not according to what I know of the relationship of Sony with Cell, and MS with Sony.

And like I said, I'm not convinced the benefits of Cell couldn't be replicated with VMX extensions and/or dedicating SPE die budget toward GPGPU.

Also, of fundamental importance, is whether 'raw power' is much use. A choice of another processor architecture might mean less power but more functionality, flexibility, cost effectiveness (retail price), etc.

Indeed.

Functionality.
Like the ability to keep a customers DLC, or Disc games, or Developers code libraries.
Functional.

price/performance is important. I'm not discounting it. Additional functionality I've brought up as an option via an additional ARM processor. I think it would make a lot of sense from both Sony and MS for cross platform functionality. But such a decision does not mean that the entire CPU would need to be ARM based. Just as Wii wasn't ARM based, but still had an ARM chip.

It could also be that whatever Cell's performance advantages are in some scenarios are mitigated by the majority of workloads and Sony should just ditch Cell and go with a stronger single thread performance for the best next-gen experience. That's something your poll isn't really measuring - impact on the next-gen experience in supporting the previous gen experience. Yes, I know, you feel there's a whole platform that needs to carry over and the company that doesn't run their old games will lose to the rival that does. That discussion has been had; no need to repeat it.

It's not so much that Cell and Xenos is perfect as is and would never need to change to reflect a changing development environment. But that they can be modified to reflect the growing needs and desires of the development community while at the same time, still maintaining compatibility with existing code bases.

Why? Why should they improve on existing architectures instead of be free to try anything else? What if...Cray's barrel-processor has appeal, and implementing it would be much cheaper and easier if no regard is paid to running existing Cell or Xenon code? Why shouldn't Sony take advantage of their new experience with ARM and PVR on Vita, or MS build on their mobile and PC hardware support?

Sure, be free to explore whatever is available... But if the alternate architecture which offers an advantage which is impossible to incorporate into/onto the existing Xenos/Cell design, then surely the alternate design which is clearly superior would have no problem emulating Cell/Xenon.

Yes?

In that case, I say, Great! Bring on the magic sauce and let the revolution begin!

Design fundamentals of alternate hardware such as the Barrel processor technique for executing massive multithreading could be implemented in a Power ISA. Or if doing so would take too long, and the existing Cray design would be cost effective and performant enough to emulate Xenon/Cell, then why not?

If not, then it clearly isn't that much faster than the existing design, and therefore, why change?

No architecture that finds it's way into a successful console is going to be that bad that it's not competitive.

Glad we see eye to eye here.

The only reason to consider updating the existing systems without regard for other options is to support BC.

I don't think I've given this impression that they shouldn't evaluate the market alternatives, but if I did, I can see why you'd think I was being short sighted. My view is that the level of engineering in these "modern" console CPU's is significantly higher than it's ever been. There is no clearly superior alternative out there as there was in the days of yesteryear where engineering was consistently finding new alternate methodologies and novel new ideas.

In many cases where minds have been free to wonder and come up with drastically radical and different architecture, the end result hasn't always been as intended.

Some have come out of nowhere and captured a performance edge.

Rarely though has a new architecture come out which significantly outperformed the competition on a per mm basis while at the same process to where the alternate hardware was so fast that it could emulate the alternate design.

Even Cell which most would agree in its current form would outperform a Xenon on the same process, is also significantly bigger. 176mm vs 233mm

Now if MS invested that extra die space in an extra Core and additional .33mb cache to match the die size of Cell, perhaps the tables turn.

Perhaps invest the extra 57mm2 into significantly beefier VMX units, more cache, OOOE, exectuion units ...

The point is Cell is larger, so it should out perform Xenon.

Performance per mm at a given process is a real issue and both Sony and MS should evaluate where Power architecture sits in the landscape of available alternatives.

But, at the same time, they should be keeping an eye on how these alternatives would perform running existing code and/or what about the alternate hardware could be integrated into their existing designs while at the same time, keeping in mind that they will have the option of offloading a substantial sum of computing to the GPGPU anyway.

Forget about BC and you have a whole world of options open up, around which you can engineer some BC if needed. Perhaps within those evaluations the idea of scaling up your console makes sense, such as Nintendo and Wii where they intended to compete outside of the power race. More often than not the idea of scaling up your old hardware is poor though hence whey PS2 didn't just scale PS1 and PS3 didn't just scale PS2. There's always the option of dual SKU's with added hardware for those who still want BC. That comes at 0% power cost in your poll too.

The Wii example is a weak argument and nowhere near what I'm suggesting, and rather insulting to the concept I'm proposing.

A better example would have been IF Microsoft had found some suitable business arrangement with intel, xb360 could have "scaled" from the existing xbox celeron architecture to a tri-core Pentium M (252mm2 with 6mb cache).

And would currently be "scaling" to a 172mm quad core IvyBridge i7 (minus the GPU of course would bring that die size down).

Hopefully that makes the concept I'm proposing a bit more clear in that it is nothing like what Nintendo did with Wii.
______________________________

Dual SKUs including redundant hardware is wasteful, and negates the "platform" strategy, and message.

Now if the old cpu architecture was so outclassed with modern alternatives, I'd agree. But we aren't talking about a 6800 in the existing consoles with this new fangled "polygon" stuff ... "no idea what that old 6800 cpu would be used for, but with this vastly new 3d world stuff, we need a MUCH different cpu".

ps4/xb720 is an extension of the same workflow, data types, and concepts as what exists today.

They just need more. More polys, more pixels, more physics, more objects, etc.

The above doesn't necessitate tossing the existing arch.
Scaling it to meet future demands is plausible, and reasonable.

 

[TheChefO]

A bit less participation than I'd have liked to see, but the poll trend is certainly telling...

 

[fearsomepirate]

I still play last-gen games a reasonable amount because last gen was when 3D graphics started to not be horrible, i.e., decent resolution, decent frame rates, graphics looked like reasonable approximations of what the artists wanted, etc.

I would like it if the PS4 could play PS3 games primarily because I can see myself still picking up the odd current-gen game for the next several years. But in all likelihood I will be done with console gaming next-gen. I'm guessing they're going to make games phone home in order to cut off the used games market, or they're all going to go with that "online pass" stuff, at which point the advantages of consoles over PC will be gone to me.