2005/2006 PC versus PS3/Xbox 360

whooleo

Newcomer
PC specs:

GPU: GeForce 7900 GS (core @ 550MHz, memory @ 722MHz, 20 pixel shaders 7 vertex shaders, 16 ROPS)
CPU: Intel Pentium 4 540J @ 3.2GHz with Hyper Threading
Motherboard: ASUS P5WD2 Premium socket LGA 775
RAM: 2x1GB DDR2 in dual channel mode @ 800MHz
OS: Windows Vista Business Class x86
HDD: 2xWestern Digital Caviar 7200RPM ATA 80GB (not in RAID)

how does this PC compare to a PS3 or Xbox 360 in terms of horsepower?
 
OS and driver overhead in combination with lack of proper porting.

Though if you are willing to drop down to same resolution and details that consoles use it's not that far behind.
 
Pentium 4 has one 3.2GHz CPU core (two HW threads with Hyper Threading). Xbox 360 CPU has three 3.2GHz CPU cores (six HW threads with SMT). Xbox cores are slightly simpler (in-order execution), so the P4 would likely fare slightly better in single threaded (old) games, but the XBox CPU would outperform P4 by a wide margin in new multithreaded games (as it has three times more CPU cores).
 
That GPU is probably pretty close to RSX because RSX has only 8 ROPs and a 128bit bus. But game developers have worked around a lot of G7x issues with Cell.

Xenos performance is pretty much unknown AFAIK but in theory it's pretty neat with its edram, mem export and unified shader architecture. I see some poor texture filtering in 360 games sometimes though (bilinear!) so I wonder about the hardware on that front.
 
7900GS is really similar to RSX, and actually has more memory bandwidth I believe.
 
Pentium 4 has one 3.2GHz CPU core (two HW threads with Hyper Threading). Xbox 360 CPU has three 3.2GHz CPU cores (six HW threads with SMT). Xbox cores are slightly simpler (in-order execution), so the P4 would likely fare slightly better in single threaded (old) games, but the XBox CPU would outperform P4 by a wide margin in new multithreaded games (as it has three times more CPU cores).

How would a E6750 compare to Xenos? I've always assumed the E6750 would zoom zoom all around Xenos, but an informed answer would be nice. :smile:
 
Way slower than PS3/Xbox360 (even If run at the same settings ), the consoles have their programming potential unlocked , developers will code to the machine language directly if they have to , However for the PC , they have to put up with a crap-load of compilers and higher programming languages , wasting resources .
 
I mean the Core 2 Duo E6750. I know it's faster than the X360 CPU, I was just wondering by what margin.
 
I mean the Core 2 Duo E6750. I know it's faster than the X360 CPU, I was just wondering by what margin.

Faster at what? It's still slower with SP floats, say, if I remember correctly ... (50 vs 77?)
 
Faster at running videogames and whatever that entails. That's the CPU I have and it runs console ports at >console settings at >console framerates.
 
Faster at running videogames and whatever that entails. That's the CPU I have and it runs console ports at >console settings at >console framerates.

Well, not all console ports are actually console ports, but started on PC, never mind that then during development they'd go to (mostly) focus on 360 builds ... but that's a different discussion. What CPU with what GPU were you running again?
 
The Core 2 Duo E6750 (2.66GHz, 333MHz FSB) with a GTX260. The performance I get in multiplatform games suggests a tremendous IPC advantage for the Core 2 Duo over the XB360 CPU, since the C2D has one less core and runs at a lower clockspeed.

It is worth noting that my brother has my old computer which is a S939 Athlon X2 3800+ at stock speeds, and it runs console ports (at least the ones I've tried including Oblivion, Fallout 3, Skyrim, Mass Effect, and others I can't recall right now) about the same as the XB360 will. Except he can run at >console IQ settings since he has an 8800GT.
 
Last edited by a moderator:
I mean the Core 2 Duo E6750. I know it's faster than the X360 CPU, I was just wondering by what margin.

Your Core 2 Duo has a ~26.8% instruction advantage. How that figures in with PC OS and API overhead, I'm not exactly sure, I'd need to see some tests on the matter, tests that I don't think exist. However, its at a FPU disadvantage, as someone said above. I would imagine that it evens out to the C2D being faster in general purpose code that a CPU usually runs, with some rare specific cases where the Xenon can pull ahead.

The C2D is close to double the transistor count of the Xenon to do it though, in terms of perf/die area or perf/transistor it loses.

The GTX 260 is probably 3-4 times the raw performance of the Xenos GPU though, so thats probably the biggest factor of performance between them.
 
Back
Top