[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology

[Advertise on 4chan]


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: GHZ.jpg (14 KB, 300x168)
14 KB
14 KB JPG
Which CPU would be more powerful: A 5Ghz Pentium 4 Extreme CPU, or a single core. 500Mhz Zen 3 CPU?
>>
>>83825548
extream
>>
>>83825563
Are you sure? It's like 16 years old; there hasn't been that much improvement?
>>
File: 1634073578951.jpg (358 KB, 1536x2048)
358 KB
358 KB JPG
p4 > zen3
>>
>>83825606
>>83825606
>>83825606
>>83825606
Simple math

p4 does 1/2 instruction per cycle, it does more but we look at worst case scenario.
amd lets say 3 ipc


amd 500 * 3 = 1500
intel 5000* 1/2 = 2500
>>
>>83825667
Where can I find stats on instructions per cycle?
>>
In what field? Multi core? Multi thread? In what usecase? Benchmarks? Browsers? Gaymes? Compilers?
>>
>>83825699
>two single core processors
>multi core
>>
>>83825679
Its a bit more difficult,not all x86 /x64 instructions are of the same length. Also modern processors have some vodoo magic inside then that can adjust some instructions to take up less clocks.

So it would depend on the program you are running
>>
>>83825679
ipc is not a static value. its a measurement for specific code. you can read them from cpu registers.

Intel and AMD have different kind of Model specifics Registers
Intel has vtune
amd has ┬ÁProf

if you care about the detail for amd look at
https://developer.amd.com/wp-content/resources/56255_3_03.PDF
u can have a look at way more little things like cache miss pipeline utilization how much a execution port is used.
>>
>>83825727
the best thing is some instruction can be zero time like move in some cases. stack operation can have zero time cost duo dedicated unit take care of them.
>>
File: cpuzp4.png (39 KB, 1795x336)
39 KB
39 KB PNG
>>83825548
Napkin math using the same benchmark
>P4@3.2GHz = 73 * (56% overclock) = 114
>My 5900X@4.85GHz (single core) = 675 * (0.103 = 500Mhz) = 69.5
Welp. I would say that's pretty conclusive even if zen scaled better than expected.
>>
>>83825727
>x64
Does this mean x86 refers to 86-bit CPUs? Where are x32 and x16?
>>
>>83825878
Hey, how's the weather in Australia ?
>>
>>83825548
>A 5Ghz Pentium 4 Extreme CPU
Powerful enough to function as a sauna heater.
>>
>>83825897
there is no x16 etc, if you want to specify a cpus memory bus width you say 32 bits or 64 bits for example. x86 is the name for intels 32 bit processors, x86_64 or amd64 is the name for intel/amds 64 bit processors or avr 8 is the name for 8 bit atmel mcus
>>
>>83825548
zen
the Pentium doesn't have enough bandwidth
>>
>>83827972
correction, amd64 has a 40/48bit memory bus, by 64 bits i actually meant that main registers are 64 bits wide
>>
>>83828001
>>83825878



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.