GTX 1070 Performance in Real world with i5 4590

[DayRider]

Games and FPS i have tested are as follows. The i5 is at max turbo clock (3.6Ghz)

Witcher 3 : - Max Settings with Hairworks kaer morhan 55 FPS Average

AC Unity - Max settings with all Nvidia stuff enabled 44 FPS average

Star Citezin - You know how this is... maxed out 20 fps .

Questions?

 

[king3pj]

You still didn't address the reason the numbers you reported in your video are about 25 FPS higher than the ones you reported in this thread.

The ones in the video are pretty comparable to professional reviews and when you factor in the stronger CPUs they are using it makes sense why they are a couple frames higher. The ones you reported in this thread for the same games with the same settings are much lower.

 

[DayRider]

You still didn't address the reason the numbers you reported in your video are about 25 FPS higher than the ones you reported in this thread.

The ones in the video are pretty comparable to professional reviews and when you factor in the stronger CPUs they are using it makes sense why they are a couple frames higher. The ones you reported in this thread for the same games with the same settings are much lower.

They were recorded inside the building.or house. Outside its terrible fps. ..

 

[DayRider]

You still didn't address the reason the numbers you reported in your video are about 25 FPS higher than the ones you reported in this thread.

The ones in the video are pretty comparable to professional reviews and when you factor in the stronger CPUs they are using it makes sense why they are a couple frames higher. The ones you reported in this thread for the same games with the same settings are much lower.

Ps thanks for the dislike. I took alot of effort into that video..

 

[Arch Angel]

@Dayrider No, I'm clearly giving you suggestions as to what your problem could be, and instead you are choosing to insult me and others.

Seeing your other posts it seems you are probably trolling trying to make the 1070 look bad anyways. Either way, I'm done helping you.

 

[king3pj]

You still didn't address the reason the numbers you reported in your video are about 25 FPS higher than the ones you reported in this thread.

The ones in the video are pretty comparable to professional reviews and when you factor in the stronger CPUs they are using it makes sense why they are a couple frames higher. The ones you reported in this thread for the same games with the same settings are much lower.

Ps thanks for the dislike. I took alot of effort into that video..

I didn't dislike your video. That was someone else. Believe it or not, I don't have a vendetta against you. I just don't think this thread is accurate as to what real world performance is with a 1070. That doesn't mean I took the time to go dislike your video on youtube.

 

[king3pj]

You still didn't address the reason the numbers you reported in your video are about 25 FPS higher than the ones you reported in this thread.

The ones in the video are pretty comparable to professional reviews and when you factor in the stronger CPUs they are using it makes sense why they are a couple frames higher. The ones you reported in this thread for the same games with the same settings are much lower.

They were recorded inside the building.or house. Outside its terrible fps. ..

By the way this is a direct quote from your video. "I was getting 65-70 FPS in the most dense and populated area in Witcher 3."

Then you go and make this thread saying it's 55 FPS average. I just don't understand why your video review seems to be pretty accurate and in line with the professional reviewers but your posts here tell a completely different story.

PS. I just went and gave your video a like to prove it wasn't me who did the dislike.

 

[DayRider]

You still didn't address the reason the numbers you reported in your video are about 25 FPS higher than the ones you reported in this thread.

The ones in the video are pretty comparable to professional reviews and when you factor in the stronger CPUs they are using it makes sense why they are a couple frames higher. The ones you reported in this thread for the same games with the same settings are much lower.

They were recorded inside the building.or house. Outside its terrible fps. ..

By the way this is a direct quote from your video. "I was getting 65-70 FPS in the most dense and populated area in Witcher 3."

Then you go and make this thread saying it's 55 FPS average. I just don't understand why your video review seems to be pretty accurate and in line with the professional reviewers but your posts here tell a completely different story.

PS. I just went and gave your video a like to prove it wasn't me who did the dislike.

Thanks 😀

I just found out that my CPU usage on witcher 3 in Novagrad is 100%... So i have a bottleneck. Never thought my CPU (i5 4590) would actaully bottleneck it. Aw well

 

[HEXiT]

not a surprise after hearing the other day that the 5960x @4ghz also bottlenecked a single 1080. the same chiped bumped to 4.5 gave an extra 15-20 fps which means the bottleneck is quite a severe 1.

i thought it might have just been the game (warhammer total war) but it really is starting to look like intel cpu's at 4.5 ghz are needed to get the most out of these cards.

this is bad because it will be the cpu's holding back gpu's for the next couple of years.
dx12 will help some, but saying as theres little in the way of dx12 games coming any time soon, we will literally be buying hardware that we cant get the max out of till intel pull there finger out... 🙁

amd zen wont be any better as there 40% IPC puts them on par with intel. so neither will be able to run these cards without bottlenecking or large oc's

 

[DayRider]

not a surprise after hearing the other day that the 5960x @4ghz also bottlenecked a single 1080. the same chiped bumped to 4.5 gave an extra 15-20 fps which means the bottleneck is quite a severe 1.

i thought it might have just been the game (warhammer total war) but it really is starting to look like intel cpu's at 4.5 ghz are needed to get the most out of these cards.

this is bad because it will be the cpu's holding back gpu's for the next couple of years.
dx12 will help some, but saying as theres little in the way of dx12 games coming any time soon, we will literally be buying hardware that we cant get the max out of till intel pull there finger out... 🙁

amd zen wont be any better as there 40% IPC puts them on par with intel. so neither will be able to run these cards without bottlenecking or large oc's

Yeah, im getting 100% CPU usage and 99-100% usage from GPU with jumps to 80% usage of GPU in the Division AND Witcher 3.

The witcher 3 sticks to somehat 99-100% GPU usage tho. and 100% CPU usage in novagrad.. What is this..

 

[DayRider]

not a surprise after hearing the other day that the 5960x @4ghz also bottlenecked a single 1080. the same chiped bumped to 4.5 gave an extra 15-20 fps which means the bottleneck is quite a severe 1.

i thought it might have just been the game (warhammer total war) but it really is starting to look like intel cpu's at 4.5 ghz are needed to get the most out of these cards.

this is bad because it will be the cpu's holding back gpu's for the next couple of years.
dx12 will help some, but saying as theres little in the way of dx12 games coming any time soon, we will literally be buying hardware that we cant get the max out of till intel pull there finger out... 🙁

amd zen wont be any better as there 40% IPC puts them on par with intel. so neither will be able to run these cards without bottlenecking or large oc's

 

[HEXiT]

thats a cpu bottleneck mate. if the cpu hits 100% the gpu will be forced to wait till the cpu can respond, which results in the lower than expected fps.
if you have the cooling and the ability. oc your cpu to 4.5ghz from what ive seen it should be enough to alleviate the problems.

 

[DayRider]

thats a cpu bottleneck mate. if the cpu hits 100% rgw gpu will bw forced to wait till the cpu can respond, which results in the lower than expected fps.
if you have the cooling and the ability. oc your cpu to 4.5ghz from what ive seen it should be enough to alleviate the problems.

im kinda suprised because i was told multiple times my i5 4590 is more than enough for this setup. But anyway..

 

[HEXiT]

your surprised. lol your not the only 1. i was like, that cant be rite. it must be the game engine falling over (total war is well known fo being buggy)
but your own results seem to be bearing out what gamers nexus said. https://youtu.be/TkE08LM4dl8?t=3m20s

i twittered jayztwocents to see if he would be willing to give an opinion as he has multiple 1080's and a high end i7.
hopefully he will have enough time to answer and maybe do some testing.

 

[Guest]

Namely,i will have bottleneck in games on ultra with my i5 4590 and gtx 1070?

 

[DayRider]

your surprised. lol your not the only 1. i was like, that cant be rite. it must be the game engine falling over (total war is well known fo being buggy)
but your own results seem to be bearing out what gamers nexus said. https://youtu.be/TkE08LM4dl8?t=3m20s

i twittered jayztwocents to see if he would be willing to give an opinion as he has multiple 1080's and a high end i7.
hopefully he will have enough time to answer and maybe do some testing.

Thanks !

 

[jbc029]

Also, if you're looking for place to compare performance against, I'd steer clear of techpowerup for one simple reason: they don't disclose what settings they used to test with. The only say if they disabled on particular feature (Hairworks, for example) and then they just spew a number for a resolution. No notes about what settings preset they used, just chart after chart of meaningless gibberish. They used to be pretty good about including details (old Tomb Raider tests would say "we used ultra preset, 4xMSAA, no TressFX"), but lately they border on worthless.

Guru3d and HardOCP both do well to explain settings and how to replicate tests. They also illustrate how taxing specific settings can be. Guru3d doesn't use ultra godrays and shadows in their Fallout 4 testing at 1440p, while H does. This shows that just that bump alone drops a 1070/980ti from 85 fps to 58.

Guru does well to display several cards at common settings, no looking to optimize speicifically. Lots of different cards at several resolutions, all using a common preset. Good to see approximately what you can expect.

HardOCP focuses on taking the best of what a particular pricing tier of card is available and pushes it as far as they can. So if they review a 960, they'll put it against an R9 380, since they paired up well (around the $200 price range). Lots of 970 vs 390 ($300), 980 vs 390x vs Fury ($425-$500), 980ti vs Fury X ($600-$650). The best parts of the reviews are the Apples to Apples comparisons, where they take the 2-4 cards being reviewed, take the best performers most taxing settings that still held playable speed and then force the other cards to run at those settings. Give a true comparison of how well the rest of the pack held up.