[Gamegpu] GTA V PC Benchmark

[crashtech]

Edit: I do agree though that the i3 is somewhat overpriced compared to i5 on the higher end and celerons and pentiums for a cheap consumer box.

That seems to be the usual line of thinking, yet when I am speccing out builds the price delta between, for instance, the 4160 and 4460 is about $70 for those like me that have to buy online. And then the 4460 will be slightly slower in many office tasks anyway, so from my perspective on selling the maximum performance for the intended application while competing with cut-rate pre-builts, low-end i5s are a bit of a hard sell.

That's not to say I wouldn't plug a 4590 or better for those who can really use the power, though.

 

[ocre]

terrible post here!

a lot of uninformed pple on a tech forum.

----------------------------------------------------------------------------------------------------


[4 core ht off] 46,5

[4 core ht on] 56,1

[6 core ht off] 64,8

[6 core ht on] 74,9

haha
what a freaking joke!!!!

um......that was a joke, wasnt it?

There is nothing wrong with not knowing but a person should take the time (as much as needed) to try to understand what is actually being discussed before the chime in, especially if they are going to make bold proclamations such as "uniformed pple".

This was an opportunity for you to actually learn something but instead you kind of made a fool out of yourself. But do not distress, I will take extra time especially for you.

You seem to have missed the factual information posted directly from intel but I will not ask you to read it again.....well not just yet.

See, you might have missed the part about this being a GTA 5 thread. Knowing what the topic is will go a long way in understanding why hyperthreading was being discussed at all here.

I think the best way to aproach this is to show you some raw data and then I will ask you to explain the results to us. And remember, there is that factual data from intel you can look at as reference while you are trying to figure it out.

Interesting. It appears that the 4790k performs better with HT off. hmmm. Maybe this is an anomaly.

well, that is strange........
it performs better at these settings too. But, it must be the resolution, right?

Oh no.....
it is not the resolution. This is getting really weird. There is no way. What is going on here? surely 4k, the results will be different.

WHAT!!!!
how can this be?
The non hyperthreaded intel CPUs are flying past the 4790k. Even the ivy i5 3570k is faster, and it is clocked significantly lower than the 4790k stock.

Well it got it, GTA 5 must not be threaded past 4 cores. Yep that is it....

WHAT!!!!!

It does scale well past 4 cores.
There are also some test that show that haswell E performs better with hyperthreading off than with it on in GTA 5 as well, but i dont have time to look for it.

Anyway. There you have it. When you figure out what is going on, post back here and tell us "uninformed pple". I will give you a hint,
we were discussing it right before you made this post.


sources
http://www.gamersnexus.net/game-bench/1911-gta-v-cpu-benchmark-4790k-3570k-9590-more

http://gamegpu.ru/action-/-fps-/-tps/grand-theft-auto-v-test-gpu.html

 

[ZGR]

Haha, those Crysis 3 screens really threw me off for a second.

Anyone else really wanting to see some benchmarks with Broadwell K and how it compares to X99 and Devil's Canyon? (GTA V Bench)

I feel like the 128 MB L4 cache could help to smoothen frametimes out during intense scenes.

I would laugh to see the large cache do nothing in game though.

 

[TheELF]

haha
what a freaking joke!!!!

Interesting. It appears that the 4790k performs better with HT off. hmmm. Maybe this is an anomaly.

well, that is strange........
it performs better at these settings too. But, it must be the resolution, right?

Oh no.....
it is not the resolution. This is getting really weird. There is no way. What is going on here? surely 4k, the results will be different.

WHAT!!!!
how can this be?
The non hyperthreaded intel CPUs are flying past the 4790k. Even the ivy i5 3570k is faster, and it is clocked significantly lower than the 4790k stock.

Well it got it, GTA 5 must not be threaded past 4 cores. Yep that is it....

WHAT!!!!!

It does scale well past 4 cores.
There are also some test that show that haswell E performs better with hyperthreading off than with it on in GTA 5 as well, but i dont have time to look for it.

Anyway. There you have it. When you figure out what is going on, post back here and tell us "uninformed pple". I will give you a hint,
we were discussing it right before you made this post.

All tests were conducted using the benchmarking utility – we opted to record only the portion where the fighter jet travels through the valley. This valley showcases nearly every setting in action – grass, water, view distance and scaling, reflections, shadows, and real-time night/day cycling for dynamic lighting effects.

NOT actual gameplay.

Your rant is just like some peoples rant about hyperthreading not being any good because it wont double cinebench scores.

 

[jj109]

haha
what a freaking joke!!!!

um......that was a joke, wasnt it?

Anyway. There you have it. When you figure out what is going on, post back here and tell us "uninformed pple". I will give you a hint,
we were discussing it right before you made this post.

The GamersNexus benchmarks are GPU bottlenecked at the top end and the GameGPU benchmarks clearly show HT raising avg FPS by 10.

Yes, HT does lower single threaded performance because two threads are sharing execution resources and bad things happen if the OS schedules the DX11 submit thread together with other heavy weight threads. DX12's multi-threaded submission queues should help HT... The advantage of HT is that it does increase throughput and saves context switch time since two threads can run at the same time instead of spending thousands of cycles evicting each other.