Ask HN: Is XBox One slower then XBox 360?
3 comments
Modern processors are much more complicated than old at-most-single-instruction-per-cycle processors. There are techniques collectively known as Instruction Level Parallelism that enable a processor to execute several instructions in each cycle on each core. That, and a more efficient memory system, that keeps the processors from stalling, waiting for memory, are significant factors in performance.
End of the day, what matters is the number of instructions executed in the unit of time, not the number of clock cycles. Clock frequency is just one factor in processor performance. A more efficient microarchitecture can easily beat a "faster" (clock speed-wise) processor, as it's been the case with Pentium 4's NetBurst vs Core.
End of the day, what matters is the number of instructions executed in the unit of time, not the number of clock cycles. Clock frequency is just one factor in processor performance. A more efficient microarchitecture can easily beat a "faster" (clock speed-wise) processor, as it's been the case with Pentium 4's NetBurst vs Core.
I don't think you measure system speed by summing up cores clock speed.
Also your question can not be answered anyway because nobody have access to XBOX One systems
Also your question can not be answered anyway because nobody have access to XBOX One systems
why not? assuming 100% paralalism, isn't it simply the amount of computations you can do in a certain period of time?
Well there are other things to consider like cache size which I'm sure is much larger.
Plus a lot of the load will be put onto the graphics processor.
Plus a lot of the load will be put onto the graphics processor.
It's worth noting that the Xbox 360 uses a PowerPC architecture while the Xbox One will use a x86 core. That will make just strictly comparing Ghz difficult.
And just to expand...
Let's look at the apple transition from powerpc to intel chips. Geekbench has a reasonable number that we can look at to compare against, so let's use that. http://browser.primatelabs.com/mac-benchmarks
We can see that the one of the late 2005 Power Mac G5 had 2 2.3 Ghz cores. This garnered a score of 2088.
About a year later, a mac mini was on the market using an Intel chip. In this case, the Intel Core Duo T2300 which runs at 1.66 Mhz with 2 cores. This received a very similar score of 2128, even though it is running considerably slower.
Later on in the years, I can come up with a Mac Mini (Mid 2011) that has an I5 running at 2.3 Ghz (The same as the original PPC we're comparing to). However, this gets a score of 5818, more than double the original PPC I picked.
---
My only point here is that it is VERY difficult to use processor speed as a tool to compare processors. It's even difficult to use it to compare a Xeon vs an I5, even though they both use the same architecture and are made by the same company.
Let's look at the apple transition from powerpc to intel chips. Geekbench has a reasonable number that we can look at to compare against, so let's use that. http://browser.primatelabs.com/mac-benchmarks
We can see that the one of the late 2005 Power Mac G5 had 2 2.3 Ghz cores. This garnered a score of 2088.
About a year later, a mac mini was on the market using an Intel chip. In this case, the Intel Core Duo T2300 which runs at 1.66 Mhz with 2 cores. This received a very similar score of 2128, even though it is running considerably slower.
Later on in the years, I can come up with a Mac Mini (Mid 2011) that has an I5 running at 2.3 Ghz (The same as the original PPC we're comparing to). However, this gets a score of 5818, more than double the original PPC I picked.
---
My only point here is that it is VERY difficult to use processor speed as a tool to compare processors. It's even difficult to use it to compare a Xeon vs an I5, even though they both use the same architecture and are made by the same company.
Is there something I'm missing here? Has the gaming industry overcome the hurdles of parallel execution?