If you haven't heard, Firefox Quantum supports multiple simultaneous processes. The question I have is whether an Intel CPU with Hyperthreading is able to take advantage of this new rendering engine. Specifically, how far back in Hyperthreading history will handle the additional process threads.
Okay, Firefox Quantum and later versions use a multiple process engine. Their default setting allows it to do 4 simultaneous processes. A program that can perform multiple simultaneous processes will benefit from multiple processors. What I am to know is if Intel CPUs that have Hyperthreding take advantage of this new capability in Firefox the same way a real processor would.
Typically one browser process isn't going to really load up more than one core. I'm not sure about the specifics of how Firefox handles how it chooses what to do with each process but I can't imagine it benifiting from the multiple processes in the way you're describing. They probably approach it more from a stability perspective meaning if one processes crashes you don't lose the others.
The performance benifit of SMT (hyperthreading) on Intel CPU's is very workload dependent. The performance hit is negligable for single threaded stuff (like web browsing) and you can get up to a 25% performance increase in heavily threaded stuff like video encoding.
WOW,how do people even come up with this stuff?It would be really hard to measure. HT works by assigning more work when a thread becomes stalled (usually loading stuff from memory). So for highly parallel tasks involving lots of memory (video encoding) there are clear benefits. On the other hand, a lot of localized memory parallel work won't see near the benefit and in some cases may even see performance losses (games).
WOW,how do people even come up with this stuff?
HT doubles your throughput (speed) as long as there is no thread that uses more then half the instructions available on a core.
Here, compared to the g4400 it's 97% faster in multicore ( ~200mhz difference)
In games the difference can be even more spectacular,depending on how crappy the game is made.
Here the game (BF1) completely kills the g4500 because the game doesn't adapt to core count,so the HT TRIPLES the performance from 30 to 90 FPS...
The 25-30% you get in video transcoding /3d rendering is actually the lowest possible benefit.
Just read up on how hyperthreading is implemented.
Here you go, I'll wait.
https://en.wikipedia.org/wiki/Hyper-threading
https://software.intel.com/en-us/ar...per-threading-technology-with-an-application/
Especially does not mean exclusively...When execution resources would not be used by the current task in a processor without hyper-threading, and especially when the processor is stalled, a hyper-threading equipped processor can use those execution resources to execute another scheduled task. (The processor may stall due to a cache miss, branch misprediction, or data dependency.)[citation needed]
This technology is transparent to operating systems and programs. The minimum that is required to take advantage of hyper-threading is symmetric multiprocessing (SMP) support in the operating system, as the logical processors appear as standard separate processors.
Server applications have little if anything to do with common everyday usage.Measured performance on the Intel® Xeon® processor MP with Hyper-Threading Technology shows performance gains of up to 30% on common server application benchmarks for this technology¹.
Set up the browser to reopen all tabs on startup.A question is how does one benchmark a browser? Once you decide that, the HT is (almost) trivial; simply run the benchmark with HT enabled and again HT disabled.
WOW,how do people even come up with this stuff?
HT doubles your throughput (speed) as long as there is no thread that uses more then half the instructions available on a core.
https://web.archive.org/web/2015112...insights-to-intel-hyper-threading-technology/Dude(tte), no offense but your second sentence in this quote is basically complete nonsense, along with several other statements you've made in this thread. You're obviously very excited about computers and performance benchmarking and so on, but on a technical level you really don't know what you're talking about here.
four instructions wideThe execution pipeline of processors based on Intel® Core™ microarchitecture is four instructions wide, meaning that it can execute up to four micro-operations per clock cycle. As shown in Figure 3, however, the software thread being executed often does not have four instructions eligible for simultaneous execution. Common reasons for fewer than four instructions per clock being retired include dependencies between the output of one instruction and the input of another, as well as latency waiting for data to be fetched from cache or memory.
Intel HT Technology improves performance through increased instruction level parallelism by having two threads with independent instruction streams, eliminating data dependencies between threads and increasing utilization of the available execution units. This effect typically increases the number of instructions executed in a given amount of time within a core, as shown in Figure 3. The impact of this greater efficiency is experienced by users as higher throughput (since more work gets completed per clock cycle) and higher performance per watt (since fewer idle execution units consume power without contributing to performance). In addition, when one thread has a cache miss, branch mispredict, or any other pipeline stall, the other thread continues processing instructions at nearly the same rate as a single thread running on the core.
So how does that change anything?Who talked about maximum throughput until now?No offense, @TheELF, but that glosses over a lot of specifics. The pipeline on Core, is not "four wide", for any kind of instruction, there are separate execution pipelines for different categories of instructions.
The only way to achieve maximal throughput, is if, for example, you have one thread doing integer ops, and one thread doing FP ops.