intel quad core vs CELL

[QUOTE="ZuluEcho14"]

[QUOTE="600pc"] PROVE IT !!!! ronvalencia

Teh Cell cannot. :P

Exactly. An i7 or quad core are not graphics renders. You need a really good GPU, RAM and processor to get Crysis looking like that, not just a processor

Quad Core can play this Teh Cell cannot. :PZuluEcho14

The Cell's SPUs can handle Crysis just fine. The problem is a lack of memory--which isn't inert to the chip. The same goes for the Quad Core.

[QUOTE="ZuluEcho14"]Quad Core can play this Teh Cell cannot. :PPariah-

The Cell's SPUs can handle Crysis just fine. The problem is a lack of memory--which isn't inert to the chip. The same goes for the Quad Core.

Read my above post. Games cannot run on processor alone (yet).

Read my above post. Games cannot run on processor alone (yet).Jekken6

[QUOTE="ZuluEcho14"]Quad Core can play this Teh Cell cannot. :PPariah-

The Cell's SPUs can handle Crysis just fine. The problem is a lack of memory--which isn't inert to the chip. The same goes for the Quad Core.

http://www.gametrailers.com/player/usermovies/23663.html

Intel GMA 500 playing Quake III Arena

http://www.youtube.com/watch?v=jKxlPfUij9s

Fold@Home CPU has a different work type compared to GPU2 and CELL. GPU2 and CELL Fold@Home clients are similar.ronvalencia

I know, which is why folding at home performance shows both native and x86 TFLOP contribution, showing a score suitable for their contribution despite differences in calculations done. That said that chart I gave is a old one, I said it's from the beta client when CUDA was first introduced to Folding@Home.

you cant use folding at home because if u actually look at the number of ps3's to pc and who is breaking down the most proteins the ps3 was at the top...

here is a link to show u what i mean just read towards the bottom http://folding.stanford.edu/English/FAQ-PS3#ntoc7

stereointegrity

Instead of reading old articles how about looking at the most recent contribution figures?

you cant compare them directly. Is like trying to get 10 people in a place with a car that has a maximum speed of 200km and a fast motorbike that reach 300km. With the car you will do the route 2 times taking you total 20minutes while with the bike you will do the route 5times taking you total 37.5 minutes. Now if you only wanna take 1-2 persons sure the bike is faster. All i'm trying to say is a quad core is more efficient and most software including games is written a way that a dual core or quad core x86 cpu is faster than the cell.

Depends what you're doing. The type of processing in standard processors are designed to tackle a wide range of applications and functions while the Cell's architecture is more suited for number crunching if we're talking about the Cells being used outside of PS3.

The Cell can do some stuff that the Quad isn't designed for, like video decoding. But for general purpose processing there's no competition: the Quad murders the cell. Hell, even an early C2D would murder the Cell (im assuming the one inside the Ps3).

[QUOTE="ronvalencia"]Fold@Home CPU has a different work type compared to GPU2 and CELL. GPU2 and CELL Fold@Home clients are similar.AnnoyedDragon

I know, which is why folding at home performance shows both native and x86 TFLOP contribution, showing a score suitable for their contribution despite differences in calculations done. That said that chart I gave is a old one, I said it's from the beta client when CUDA was first introduced to Folding@Home.

you cant use folding at home because if u actually look at the number of ps3's to pc and who is breaking down the most proteins the ps3 was at the top...

here is a link to show u what i mean just read towards the bottom http://folding.stanford.edu/English/FAQ-PS3#ntoc7

stereointegrity

Instead of reading old articles how about looking at the most recent contribution figures?

FTR, that chart does not separate single-processor from multi-processor results. Also, SMP clients gets different workloads than those of the single-processor, PS3, and GPU clients (they get the heavy jobs that require lots of high-precision math--GPUs are fast but currently lack enough precision, and older CPUs are simply too slow for the beasts being assigned).HuusAsking

I linked to the section of their website that explained how the differences are handled for fairer comparison.

At that particular task the Quad beat Cell, but the point was that it was just one of many different tasks that can be used for comparison purposes. People are focusing far too much on that old chart and not paying attention to the main point I was making.

you cant use folding at home because if u actually look at the number of ps3's to pc and who is breaking down the most proteins the ps3 was at the top...

here is a link to show u what i mean just read towards the bottom http://folding.stanford.edu/English/FAQ-PS3#ntoc7stereointegrity

Irrelevant..and you obviously don't understand the figures you're looking at.

If every one of those PCs were equipped with a Q6600 and folding for the same durations as the PS3 clients, the PC client figures would be beating out the PS3 figures by a landslide (based on the folding performance chart.)

Not everyone has such a processor, though. The average PC is running with a low/mid-range Core 2 Duo, at best (and I may be overestimating a bit with that.)

Core i7 pwns both :D

[QUOTE="600pc"][QUOTE="Fizzman"]

Intel bya mile.

ZuluEcho14

Teh Cell cannot. :P

[QUOTE="Pariah-"]

[QUOTE="ZuluEcho14"]Quad Core can play this Teh Cell cannot. :PJekken6

The Cell's SPUs can handle Crysis just fine. The problem is a lack of memory--which isn't inert to the chip. The same goes for the Quad Core.

Read my above post. Games cannot run on processor alone (yet).

Depends on the game.

Smooth and playable OpenGL game running on Swiftshader 2.01 Direct3D 9b JIT software renderer on Intel Core 2 Duo T7500 2.2Ghz with 4MB L2 cache. Works well on Windows Terminal Server, Windows Remote Desktop, Virtualbox, VirtualPC and VMWARE.

[QUOTE="kidcool189"][QUOTE="600pc"] PROVE IT !!!! 600pc

Note the rounding modes.

Anyway, Lawn's Xeon's complier optimization only indicated "-03" (or /03).

For specfic Intel Core 2 based optimizations I ussualy enable

1. Intel CoreTM 2 Duo Processor (/QaxT) for "Use Intel Processor Extensions"

2. Intel CoreTM 2 Duo Processor (/QxT) for "Require Intel Processor Extensions"

3. Enable Parallelization (/Qparallel)

4. Loop Unrolling set to automatic.

5. Enable "Global Optimations". 6. Enable Fast(/Qfp-speculationfast) Floating Point Speculation.

[QUOTE="600pc"][QUOTE="kidcool189"] you prove otherwiseronvalencia

Note the rounding modes.

Anyway, Lawn's Xeon's complier optimization only indicated "-03" (or /03).

For specfic Intel Core 2 based optimizations I ussualy enable

1. Intel CoreTM 2 Duo Processor (/QaxT) for "Use Intel Processor Extensions"

2. Intel CoreTM 2 Duo Processor (/QxT) for "Require Intel Processor Extensions"

3. Enable Parallelization (/Qparallel)

4. Loop Unrolling set to automatic.

5. Enable "Global Optimations". 6. Enable Fast(/Qfp-speculationfast) Floating Point Speculation.

What's thePS3 CELL's Quake III Arena's FPS again?ronvalencia

Intel by far.

[QUOTE="ronvalencia"]What's thePS3 CELL's Quake III Arena's FPS again?Pariah-

Refer to http://www.gametrailers.com/player/usermovies/23663.html

Note the Quake3's frame rate.

Running Quake3 on Swiftshader 2.01 (supports up to SSSE3) Direct3D 9b JIT software renderer+MESA OpenGL-to-Direct3D bridge (MesaWin Version 6.0) on

1. Intel Core 2 Duo T7500 (Mobile) 2.2Ghz (~35 watts) with 4MB L2 cache was smooth i.e. ~32 FPS (Demo001, Normal Settings at 640x480).

2. Intel Core 2 Duo P9500 (Mobile) 2.53Ghz (~25 watts) with 6MB L2 cache was smooth i.e. ~40 FPS (Demo001, Normal Settings at 640x480).

Swiftshader 2.01 is not optimized for SSE4.1/SSE4.2.

http://www.transgaming.com/business/swiftshader/

"SwiftShader is the world's fastest pure software 3D renderer with DirectX® 9.0 features"

[QUOTE="ZuluEcho14"]

[QUOTE="600pc"] PROVE IT !!!! TheGreatOutdoor

Teh Cell cannot. :P

One word RAM

That isn't even a fair comparison. A 2008 Quad Core compared to 2006 or earlier CELL? Come on, at least make it a legit comparison.Zoso-8

To be honest I have to question the point of Cell in gaming, at least in consoles and PC.

The GPU does what it is good at and the CPU does the same, why do we need a CPU adapted to be a middle ground? It just makes it worse for both areas, it makes a weak GPU and a poor CPU. If you use it for either you rob performance from the other area, Cells ability to do graphics on the PS3 is limited by the typical CPU tasks Cell is also expected to juggle.

The argument of course is to produce a CPU that can do both general purpose work as well as heavy number crunching like a GPU. Frankly I think GPGPU has made it obsolete, it gains similar advantages to Cell but with existing technologies. Instead of designing a whole new architecture to do this why not just distribute work load to the components best at that task?

Sometimes it's not about creating the ideal technology to make something work, it is about using the most practical method.

AnnoyedDragon

A very good post.

[QUOTE="Zoso-8"]That isn't even a fair comparison. A 2008 Quad Core compared to 2006 or earlier CELL? Come on, at least make it a legit comparison.HuusAsking

Remember

1. PowerXCELL doesn't have the "economic of scale" benefit when compared to AMD's FireStream i.e. it's RV770 is same as Radeon 4800 RV770 equivalent.

2. PowerXCELLdoesn't equal PS3 CELL i.e. not the same core.

[QUOTE="ronvalencia"]

[QUOTE="600pc"] 3.2 GHz Cell with 8 SPEs delivering a performance equal to 100 GFLOPS on an average double precision Linpack 4096x4096 matrix.(wikipedia)HuusAsking

Note the rounding modes.

Anyway, Lawn's Xeon's complier optimization only indicated "-03" (or /03).

For specfic Intel Core 2 based optimizations I ussualy enable

1. Intel CoreTM 2 Duo Processor (/QaxT) for "Use Intel Processor Extensions"

2. Intel CoreTM 2 Duo Processor (/QxT) for "Require Intel Processor Extensions"

3. Enable Parallelization (/Qparallel)

4. Loop Unrolling set to automatic.

5. Enable "Global Optimations". 6. Enable Fast(/Qfp-speculationfast) Floating Point Speculation.

The people are saying core 2 quad because its a good CPU but do any of us truly have enough knowledge of both of them to actually compare?:question:

Ultramarine777

Read my last post because I just explained why the ps3's number crunching ability is applicable to games if developers take the time. We are talking abotu games right? Yeah as a pc the cell could have issues. But the cell isn't in pc's as of yet. The chip in the PS3 was a join design between IBM and Sony to be utilized in the ps3.

That article goes on to explain that if the cell were ever to be used in PC's it could ultimately but it's not being targetted for that as of yet. As far as for what the ps3 is trying to achieve it's more than capable.

All this talk about general processing power and out of order execution is nonsense and just lazy know it all people who don't want to take the time or don't know how to take advantage of the cell or the development tools are not there which i have a hard time believing at this point. When the ps3 was orignally released yes that was an issue, but you are going to see more and more developers come around on the cell processor.

"The downsides of an in-order PPE are minimized as much as possible by making the core only 2-issue, meaning that at best, it could execute two operations in parallel. So, execution potential lost to in-order inefficiencies are minimized in a sense that at least there aren't a lot of transistors wasted on making the PPE an extremely wide chip. A good compiler should be able to make sure that both issue ports are populated as frequently as possible, despite the fact that the microprocessor is in-order. The PPE is also capable of working on two threads at a time, also designed to mask the inefficiencies of an in-order core for general purpose code.

Architecturally, if anything will keep Cell out of being used in a PC environment, it's the PPE. A new Cell with a stronger PPE or an array of PPEs could change that, however."

Walker34

As a stream processor array, there's nothing special about CELL's SPEs. SPE is not the only stream processor array product in the interactive entertainment market. On the PC, these stream processor array products are known as GPUs e.g. AMD/ATI Radeon HD or NVIDIA Geforce 8/9/GT2x0. Remember "ATI" is short for "Array Technologies Incorporated".

Besides the memory, one of the difference between AMD FireStream and Radeon HD 4980 is that FireStream doesn't have display connectors. An add-on card for the PC must add something of the value to the end user or it will end up as PhysX PPU or 3DLabs WildCat VPU (200 GFLOPS from it's programmable stream processors during NVIDIA Geforce FX5900/6800 days).

Remember a stream processor array without a proper middleware support is a dead paper weight.

[QUOTE="mismajor99"]

The Cell in it's current PS3 configuration is subpar to Intel's and AMD's current lineup, especially for gaming. The Cell in the PS3 is like having an M16 with blanks, It's handicapped by a gross lacking of vram and a proper GPU, so the whole argument is mute to begin with. Simple math is all that has been proven that the cell does well, but it's cores' aren't Porche's, they're more like Hondas.

Walker34

That's a load of crap.

Sorry, "AMD GAME!**" system platform kills PS3.

**AMD's marketing label for it's gaming level AMD CPU, AMD core logic and AMD GPU ecosystem.

Already did.. Key word smart developer. Like i was saying not every developer knows everything man, and just the fact you say go ask a developer doesn't mean they knwo what they are talking about. No one knows everything and a lot of developers don't know what they are doign when it comes to the cell. Fact. Or they just don't want to take the time. The fact is the ps3 is perfectly capable for gaming and even moreso if developers take the time.

"In-order microprocessors suffer because as soon as you introduce a cache into the equation, you no longer have control over memory latencies. Most of the time, a well-designed cache is going to give you low latency access to the data that you need. But look at the type of applications that Cell is targeted at (at least initially) - 3D rendering, games, physics, media encoding etc. - all applications that aren't dependent on massive caches. Look at any one of Intel's numerous cache increased CPUs and note that 3D rendering, gaming and encoding performance usually don't benefit much beyond a certain amount of cache. For example, the Pentium 4 660 (3.60GHz - 2MB L2) offered a 13% increase in Business Winstone 2004 over the Pentium 4 560 (3.60GHz - 1MB L2), but less than a 2% average performance increase in 3D games. In 3dsmax, there was absolutely no performance gain due to the extra cache. A similar lack of performance improvement can be seen in our media encoding tests. The usage model of the Playstation 3 isn't going to be running Microsoft Office; it's going to be a lot of these "media rich" types of applications like 3D gaming and media encoding. For these types of applications, a large cache isn't totally necessary - low latency memory access is necessary, and lots of memory bandwidth is important, but you can get both of those things without a cache. How? Cell shows you how.

Each SPE features 256KB of local memory, more specifically, not cache. The local memory doesn't work on its own. If you want to put something in it, you need to send the SPE a store instruction. Cache works automatically; it uses hard-wired algorithms to make good guesses at what it should store. The SPE's local memory is the size of a cache, but works just like a main memory. The other important thing is that the local memory is SRAM based, not DRAM based, so you get cache-like access times (6 cycles for the SPE) instead of main memory access times (e.g. 100s of cycles).

What's the big deal then? With the absence of cache, but the presence of a very low latency memory, each SPE effectively has controllable, predictable memory latencies. This means that a smart developer, or smart compiler, could schedule instructions for each SPE extremely granularly. The compiler would know exactly when data would be ready from the local memory, and thus, could schedule instructions and work around memory latencies just as well as an out-of-order microprocessor, but without the additional hardware complexity. If the SPE needs data that's stored in the main memory attached to the Cell, the latencies are just as predictable, since once again, there's no cache to worry about mucking things up.

Making the SPEs in-order cores made a lot of sense for their tasks. However, the PPE being in-order is more for space/complexity constraints than anything else. While the SPEs handle more specified tasks, the PPE's role in Cell is to handle all of the general purpose tasks that are not best executed on the array of SPEs. The problem with this approach is that in order to function as a relatively solid performing general purpose processor, it needs a cache - and we've already explained how cache can hurt in-order cores. If there's a weak element of the Cell architecture it's the PPE, but then again, Cell isn't targeted at general purpose computing, despite what some may like to spin it as."

Walker34

On CUDA processors, you have multi-thousand registers(e.g. 32768 32bit registers for Geforce 8600GT i.e.8192 32bit registers (32K) per SP) being stored next to ALUs and ithas both software and hardware managed cache.

Refer to http://www.gametrailers.com/player/usermovies/23663.htmlronvalencia

[QUOTE="600pc"][QUOTE="Walker34"]

I've been reading a bunch. Ill see what i can find. I don't get into all the technical crap too much either although i have CS degree.

But my understanding is 1 core and 7 spu's is > quad core pc in straight number crunching scenarios though. It's actually more powerful because of its architecture and the way things are allocated. The difference is that will not translate into the real world right nwo because operating systems and branching is not set up to handle the cell as of yet. The software hasn't caught up. It may never because it's a war and intel and microsoft have their own plans on what they think is better right now and what they can sell.

It's a little scary because Microsoft is an american company. It's like the japanese and american car battles of old pretty much. The japanese are coming up with all these v-tek and v stuff and the americans are just plopping multiple cores into a straight line. The japanese probably think we are silly and their math geniuses are probably laughing at us because there is no need to have multiple cores and it's a waste of material. Silly Americans.

The americans will use the general purpose argument and they probably have one, as having multiple cores does have some advantages for raw power, but I tend to think having 1 core and 7 spu's compounds itself to the point, where that advantage is not what it seems like it would be.

Problem number 3, I am not using a Cell. None of us are in our PC's, if the cell is the might god chip then were are the innovatinos leading from it? Were is IBM? If the cell really was more powerful why are we not using it? The fact is that it is not.

thegoldenpoo

[QUOTE="thegoldenpoo"]

Problem number 3, I am not using a Cell. None of us are in our PC's, if the cell is the might god chip then were are the innovatinos leading from it? Were is IBM? If the cell really was more powerful why are we not using it? The fact is that it is not.

killzowned24

That is still meaningless. Put enough Cell processors together and hey presto you have the worlds fastest/ floating point super computer. Put C2Qs or I7s together and you achieve the same result. Put enough P4s together and you could feasibly do the same.

Also bare in mind, the reason the Cell is used is because of its cost effectiveness for achieving high numbers in floating point operations, not because it is some artifact with hidden power!! :lol:

[QUOTE="killzowned24"][QUOTE="thegoldenpoo"]

Problem number 3, I am not using a Cell. None of us are in our PC's, if the cell is the might god chip then were are the innovatinos leading from it? Were is IBM? If the cell really was more powerful why are we not using it? The fact is that it is not.

DAZZER7

That is still meaningless. Put enough Cell processors together and hey presto you have the worlds fastest/ floating point super computer. Put C2Qs or I7s together and you achieve the same result. Put enough P4s together and you could feasibly do the same.

Also bare in mind, the reason the Cell is used is because of its cost effectiveness for achieving high numbers in floating point operations, not because it is some artifact with hidden power!! :lol:

[QUOTE="600pc"][QUOTE="kidcool189"] you prove otherwise-GeordiLaForge-

[QUOTE="DAZZER7"]

[QUOTE="killzowned24"] Well its tech with a modified CELL does help run the fastest supercomputer on earth:)killzowned24

That is still meaningless. Put enough Cell processors together and hey presto you have the worlds fastest/ floating point super computer. Put C2Qs or I7s together and you achieve the same result. Put enough P4s together and you could feasibly do the same.

Also bare in mind, the reason the Cell is used is because of its cost effectiveness for achieving high numbers in floating point operations, not because it is some artifact with hidden power!! :lol:

[QUOTE="killzowned24"][QUOTE="DAZZER7"]

That is still meaningless. Put enough Cell processors together and hey presto you have the worlds fastest/ floating point super computer. Put C2Qs or I7s together and you achieve the same result. Put enough P4s together and you could feasibly do the same.

Also bare in mind, the reason the Cell is used is because of its cost effectiveness for achieving high numbers in floating point operations, not because it is some artifact with hidden power!! :lol:

Dynafrom