Anandtech dashes cpus from ps3 and xbox360

Status
Not open for further replies.
Why in the world would I believe IBM? It would be in IBM's interest to destroy any of AnandTech's claims. AnandTech has nothing to gain out of this. Well besides hits.

What company would tell you that thier product isn't 10X faster then any PC today? You think Microsoft is going to tell you that Linux is great and you should use it?

Nothing that I've read has disproved AnandTech. Including that piece of information.

So Duke are you not going to listen to most of the people in this thread either. This is video games that we are playing here not typing in MS Excel or Access. I'm not really knowledgeable in this subject but I do know how to read. Most people here and at other forums are saying that this is not such a problem that it can't be beaten.

The software will be their to optimize the code so that the games can run great. And so I ask you dear sir do you think they should have went with a P4 chip instead?
 
DukenukemX said:
Shifty Geezer said:
If bandwidth were horsepower, we know bigger engine doesn't alway mean a faster car. A smaller engine with less horsepower can outperform a big, heavy motor with a huge engine. The important consideration is TOTAL SYSTEM DESIGN, not individual component performance.

Don't know about you but bigger engines always means more horsepower. Smaller engines get more power from turbos or superchargers that compress the air so the fuel would burn. You get a V10 engine from a Viper and add a turbo then no small engine will ever catch it.
You've missed the point. Total system design...

Which is faster; a vehicle with a 100 HP engine or a vehicle with a 500 HP engine? The 500, you say? What if the 100 HP engine is in an 800 kg Lotus, and the 500 HP is in a 20,000 kg 1950's tank? The small engine powers a much faster vehicle.

Which is faster; a 100 HP Lotus or a 2,000 HP drag car? Clearly the drag car you say? What, around Laguna Seca? the drag wouldn't even make the first bend, let alone get a lap time less than the Lotus!

No matter how powerful the engine you need an overall system that can use that power. A drag car hasn't the maneouverability to be faster than a small engined Lotus is most situations - only a straight line sprint. Lotus beats it on any sort of course or endurance race.

You see? GPU power or CPU power alone doesn't make for a fast console. Even though a PC GPU may have higher bandwidth (bigger engine) it has overheads and limits and other aspects that a console like XB360 can outperform it on. The eDRAM is worth maybe 35-40 GB of bandwidth (pie in the sky figure) on a conventional PC GPU bus, so that 22 GB/s main bus is still plenty if all it's got to do is provide textures and polys. Just as HP tells you nothing about the speed of a vehicle on it's own, GHz and GB/s tells you nothing about overall system performance. It only describes performance of a particular aspect of the overall system.
 
mckmas8808 said:
Most people here and at other forums are saying that this is not such a problem that it can't be beaten.
Why are anonymous people on the internet more credible than anonymous sources from Anandtech? I'm unaware of them having a history with inventing stories.

The people on forums are just speculating, they haven't actually worked with the hardware.
 
GuyMe said:
mckmas8808 said:
Most people here and at other forums are saying that this is not such a problem that it can't be beaten.
Why are anonymous people on the internet more credible than anonymous sources from Anandtech? I'm unaware of them having a history with inventing stories.

The people on forums are just speculating, they haven't actually worked with the hardware.

Because here you actually talk to the people who are spectulating. I'll put more credence to the people here than someone who's shrouding themselves behind Anand.
 
What is the problem, people?

We already expected that PC like (bad) code would run very bad, but code made for those CPUs will play very well.

Remember that is not suposed to have bad code on a console.
 
Shifty Geezer said:
DukenukemX said:
Shifty Geezer said:
If bandwidth were horsepower, we know bigger engine doesn't alway mean a faster car. A smaller engine with less horsepower can outperform a big, heavy motor with a huge engine. The important consideration is TOTAL SYSTEM DESIGN, not individual component performance.

Don't know about you but bigger engines always means more horsepower. Smaller engines get more power from turbos or superchargers that compress the air so the fuel would burn. You get a V10 engine from a Viper and add a turbo then no small engine will ever catch it.
You've missed the point. Total system design...

Which is faster; a vehicle with a 100 HP engine or a vehicle with a 500 HP engine? The 500, you say? What if the 100 HP engine is in an 800 kg Lotus, and the 500 HP is in a 20,000 kg 1950's tank? The small engine powers a much faster vehicle.

Which is faster; a 100 HP Lotus or a 2,000 HP drag car? Clearly the drag car you say? What, around Laguna Seca? the drag wouldn't even make the first bend, let alone get a lap time less than the Lotus!

No matter how powerful the engine you need an overall system that can use that power. A drag car hasn't the maneouverability to be faster than a small engined Lotus is most situations - only a straight line sprint. Lotus beats it on any sort of course or endurance race.

You see? GPU power or CPU power alone doesn't make for a fast console. Even though a PC GPU may have higher bandwidth (bigger engine) it has overheads and limits and other aspects that a console like XB360 can outperform it on. The eDRAM is worth maybe 35-40 GB of bandwidth (pie in the sky figure) on a conventional PC GPU bus, so that 22 GB/s main bus is still plenty if all it's got to do is provide textures and polys. Just as HP tells you nothing about the speed of a vehicle on it's own, GHz and GB/s tells you nothing about overall system performance. It only describes performance of a particular aspect of the overall system.


nice post, and it was better than that entire Anand article 8)
 
GuyMe said:
Chris Hecker (who works at EA now) aired similar complaints about the NextGen console's CPUs at the GDC in March. I haven't read the entire thread, so I apologise if this has already been discussed in this or any other thread.

http://www.gamespot.com/news/2005/03/18/news_6120449.html (starts at the last quarter of the page)

...

Here is the terrifying realization about the next generation of consoles. I'm about to break about a zillion NDAs, but I didn't sign any NDAs so that's totally cool!

I'm actually a pretty good programmer and mathematician but my real talent is getting people to tell me stuff that they're not supposed to tell me. There we go. Gameplay code will get slower and harder to write on the next generation of consoles. Why is this? Here's our technical slide. Modern CPUs, like the Intel Pentium 4, blah, blah, blah, Pentium [indiscernible] or laptop, whatever is in your desktop, and all the modern power PCs, use what's called 'out of order' execution. Basically, out of order execution is there to make really crappy code run fast.

So, they basically--when out of order execution came out on the P6, the Pentium 6 [indiscernible] the Pentium 5, the original Pentium and the one after that. The Pentium Pro I think they called it, it basically annoyed a whole bunch of low level ASCII coders, because now all of a sudden, like, the crappiest-ass C code, that like, Joe junior programmer could write, is running as fast as their Assembly, and there's nothing they can do about it. Because the CPU behind their back, is like, reordering that guy's crappy ass C code, to run really well and utilize all the parts of the processor. While this annoyed a whole bunch of people in Scandinavia, it actually…

[laughter]

And this is a great change in the bad old days of 'in order execution,' where you had to be an Assembly language wizard to actually get your CPU to do anything. You were always stalling in the cache, you needed to like--it was crazy. It was a lot of fun to write that code. It wasn't exactly the most productive way of doing experimental programming.

The Xenon and the cell are both in order chips. What does this mean? The reason they did this, is it's cheaper for them to do this. They can drop a lot of core--you know--one out of order core is about the size of three to four in order cores. So, they can make a lot of in order cores and drop them on a chip, and keep the power down, and sell it for cheap--what does this do to our code?

Well, it makes--it's totally fine for grinding like, symmetric algorithms out of floating point numbers, but for lots of 'if' statements in directions, it totally sucks. How do we quantify 'totally sucks?' "Rumors" which happen to be from people who are actually working on these chips, is that straight line gameplay code runs at 1/3 to 1/10 the speed at the same clock rate on an in order core as an out of order core.

This means that your new fancy 2 plus gigahertz CPU, and its Xenon, is going to run code as slow or slower than the 733 megahertz CPU in the Xbox 1. The PS3 will be even worse.

This sucks!

Well now they have to have good code. :LOL:
 
pc999 said:
What is the problem, people?

We already expected that PC like (bad) code would run very bad, but code made for those CPUs will play very well.

Remember that is not suposed to have bad code on a console.

Sorry this just annoyed me.

PC Code is not BAD........ or at least no worse than your average console code.

And I hate PC's!
 
mckmas8808 said:
The software will be their to optimize the code so that the games can run great. And so I ask you dear sir do you think they should have went with a P4 chip instead?

My feeling is that the answer is yes if they wanted to make life easier on developers and have better quality PC->console ports right from the start.

Instead, it sounds like there will be a huge ramp up just to reach PC levels of performance let alone the theoretical maximums we've all been hearing about.
 
seismologist said:
mckmas8808 said:
The software will be their to optimize the code so that the games can run great. And so I ask you dear sir do you think they should have went with a P4 chip instead?

My feeling is that the answer is yes if they wanted to make life easier on developers and have better quality PC->console ports right from the start.

Instead, it sounds like there will be a huge ramp up just to reach PC levels of performance let alone the theoretical maximums we've all been hearing about.

Developers will learn to live with it. I don't see how you can honestly say "NO" to Multi-Core PCs when AMD and Intel are following the same route with their own CPUs. Face it, these type of CPUs are the way of the future and developers who don't meet this head on and just complain will fail.

EDIT: Just because this type of technology is in its infancy with PC's and Consoles doesn't mean that you should put it off till it more mature either. I like that Console makers (Sony, Microsoft) are future-proofing their CPU's now because about 3 years down the line when PCs have Multi-Core code optomized for it and you see amazing Physics and AI and Console only have P4's....then people will complain why Console makers didn't go the "OTHER" route.

Developers are human, and the ability to learn may not be as high as a child but I think theres room in developers heads to learn how to correctly create a game using these type of CPUs.
 
To ERP

Sorry, I dont really mean that PC code is bad once that there is already few bad code because of out of order,but in consoles there is a diference bettwen good and bad code, and a PC like code will be bad in these CPUs.

But we already expect that PC like code would became bad code for consoles, they "just" need to rework the code for consoles.
 
seismologist said:
mckmas8808 said:
The software will be their to optimize the code so that the games can run great. And so I ask you dear sir do you think they should have went with a P4 chip instead?

My feeling is that the answer is yes if they wanted to make life easier on developers and have better quality PC->console ports right from the start.

Instead, it sounds like there will be a huge ramp up just to reach PC levels of performance let alone the theoretical maximums we've all been hearing about.

Consoles won't struggle to reach PC levels of performance simply because they are not doing the same thing.

Using CPU benchmarks is going to give a very bad indication of overall performance, consoles have thin API layers and aren't battling the OS for resources.

PC processors are just a bad thing to compare against.
 
mckmas8808 said:
He hasn't even worked on the hardware. Read Shifty's last post you could learn something. People like you (not to offend you or anything) and alot of websites treat Anand like he is some kind of god.
I was saying that both speculative forum posts and Anand articles should be taken with a grain of salt, not that the Anand article was accurate. The thing about them not making stories up was meant as a separate point though I can see how it was a bit convoluted.

Also, just because a reworking of code could make games run better than they are on dev-kits doesn't mean the hardware itself is efficient.
 
ERP said:
Consoles won't struggle to reach PC levels of performance simply because they are not doing the same thing.

Using CPU benchmarks is going to give a very bad indication of overall performance, consoles have thin API layers and aren't battling the OS for resources.

PC processors are just a bad thing to compare against.

I know the performance is there but it seems like it's difficult to get at (even with a thin OS layer).
I meant the developers will have a huge ramp up to reach PC levels performance. based on what we've heard about first gen games being mostly single threaded only running on a single CPU core. It almost reminds me of the early days of the PS2 when nobody was using the vector units.
 
Early compilers tend to suck, developers are lazy and not all of them necessarily have a good grasp yet on the way the processor works (or processors in general for that matter).

For instance, if you have a microbenchmark consisting of non unrolled loop with a small body then you are going to get atrocious performance (compared to processors with register renaming). Do most developers know about that? Almost certainly. Would any developers if they want to whine about the processor try to use results from such a microbenchmark to back up their point? Who knows ...

I dont put stock in the words of anonymous cowards.
 
mckmas8808 said:
The software will be their to optimize the code so that the games can run great. And so I ask you dear sir do you think they should have went with a P4 chip instead?

it's about cost. If they could've gone with a P4 or A64 they would have, but they needed to own the IP.

it cheaper for them to produce 3 weak multi-core processors, than to pay for one top of the line single core, not only that but they don't have to pay royalites to AMD or Intel.

The trade-off? life just got way harder for game developers.

They made that mistake on XBOX one, going with Intel's chip, which is why they can't afford to keep selling them and are forced to kill the console. They aren't going to make that mistake again.

Same goes with sony, they're trading costs for developer workload, it much much cheaper for Sony to use CELL in the PS3, after investng nearly 1.2 billion dollars into it, than it would be developing a powerful single threaded core that would really make Dev's lives easier.

And near the end of the consoles life, both these CPU's will be produced for dirt cheap, which would never happen with an AMD or P4 core. It's all about cost.

BTW Here's the original article reposted:
http://www.ansonwilson.com/anandreview.htm
 
mckmas8808 said:
So Duke are you not going to listen to most of the people in this thread either. This is video games that we are playing here not typing in MS Excel or Access. I'm not really knowledgeable in this subject but I do know how to read. Most people here and at other forums are saying that this is not such a problem that it can't be beaten.

The software will be their to optimize the code so that the games can run great.

Are you a developper ? No offense but i don't think you have a clue
about programming (like the majority of the posters here ... )
Game engines and commercial softwares (you cited Access ...) use both extensively integer instructions and control flows ...
General purpose code = integer instructions with control flows
I mean more than 80% of the game engine (executed by the cpu) is composed of these instructions.
That's why a fast general purpose processor is needed to execute the non graphics section of game engines.
 
Complaining that out-of-order code runs poorly on an in-order closed box system is like complaining to an eskimo that his igloo lacks powerpoints. Perfectly accurate, yet utterly stupid.
 
PARANOiA said:
Complaining that out-of-order code runs poorly on an in-order closed box system is like complaining to an eskimo that his igloo lacks powerpoints. Perfectly accurate, yet utterly stupid.

You've seemed to have hit the nail on the head....
 
Status
Not open for further replies.
Back
Top