• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

GTX 1080 Ti with PCI-e 2.0

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
That's crazy on a X58 system. Last I checked, I was running 2 GTX 480s in SLI on 1080p and was hitting about 60% on average of load with a few stutters (100%) on the i7 950 CPU. Any GPUs from the last 3+ years would surely bottleneck the hell out of it at 1080p since you'd be holding back the GPUs. I see it all the time in EVGA forums. People on old systems unless its a 2600k+ type of system, you still might have a bit left in it for 1440p but anything less or 4K might not be enough as far as I've read and seen from others.
 
I thought the point was to see if the CPU was holding the fps back in BF1 at 1080p or 1440p that is what he is testing.
...and that is what testing at 1080p and higher is for. Testing lkwer doesnt show much of anything except for that res.

I had to install BF1 on the X58 system with Xeon X5660 @ 4.3Ghz

Here are my settings:
1080p
GPU Memory Restriction Off
Ultra preset
FPS are between 100 - 120 while playing the single player campaign where I am driving the tank. After that I am on foot scouting for the tank and I am getting 130 - 165 fps. Not sure why this is. I suck at this game. Ha!

Here is a Youtube video of a guy playing with a 6700K @ 4.7Ghz and a GTX 1080 Ti @ 2000MHz on ultra preset. He is getting as high as 180ish fps. It is a different scene but shows roughly a 10% difference on the high end.



I would like to point out that the Xeon X56xx series is not Gulftown, it is Westmere. I am still of the opinion that if you have an old 1366 system that the Xeon is a nice way to breath new life into it. Skip all of the i7's for this platform as they are cost prohibitive. I have a Nvidia reference GTX 980 that has been flashed to raise the power limits on the card and it runs very well. I have less than $200 in it. For those that still have one of these systems it can be upgraded nicely for well under $300 for a GPU and a CPU. I would *not* recommend buying a system like this today. If you don't already have one forget it.
thanks fkr your testing, but, we need to see the same scene to make a comparision. :)

There is no difference between a xeon amd regular cpu for all intents and purposes. :)
 
Lol, my spell check on my phone is borked. When i type and it autocorrects to amd..lololol
 
...and that is what testing at 1080p and higher is for. Testing lkwer doesnt show much of anything except for that res.

thanks fkr your testing, but, we need to see the same scene to make a comparision. :)

There is no difference between a xeon amd regular cpu for all intents and purposes. :)

You are somewhat correct. I should have pointed out that Westmere (32nm) is faster than Nehalem (45nm) at the same clock speed. i7-980x and i7-990X are Westmere as well but way more expensive to buy today than the X56xx series Xeon. Lower spec first gen i7s are Nehalem. Xeon X56xx is unlocked. The only real difference between the different X56xx CPUs is 4 core vs. 6 core and maximum multiplier. Higher mutli helps with system boards that don't go very high on BCLK (of course).

Lots of people have experienced the improved performance of Westmere CPU's when the Xeons got dumped on ebay a few years ago. I was kind of late to the party and would have never played with it at all except I was able to get some 1366 hardware very cheaply. I purchased an entire system including a 256GB SSD and 1TB HDD for $200 shipped. I ended up refurbing that system (mostly cleaning!) and giving to to a friend and he has had it a few months now. It was an upgrade from a P4. The only changes were the addition of a Xeon and Win 10. The GPU is GTX 275. He only plays WoW. He was going to buy a GTX 1050 Ti but hasn't done it yet.

This got me interested in this ancient X58 platform. I was able to snipe a few things off ebay (mostly untested stuff that turned out to be good), which ended up becoming this secondary system. I have very little money in it.

I have no idea what scene that BF1 clip is from. ;)
 
Would an Intel Core i7-4930K 6 core 3.9GHz CPU bottleneck a GTX 1070/1080 at 1080P gaming?
 
Yes, those minor differences i am aware of. It doesnt change the fact that a 1080ti is being held back by the platform. The goal for the pereon is 144hz gaming...to get there, all bottlenecks should be removed as much as possible... i dont buy a 700 gpu to have it run like a 600 gpu. ;)

Would an Intel Core i7-4930K 6 core 3.9GHz CPU bottleneck a GTX 1070/1080 at 1080P gaming?
perhpas in a few titles... cpu needs to be overclocked.
 
Last edited:
Would an Intel Core i7-4930K 6 core 3.9GHz CPU bottleneck a GTX 1070/1080 at 1080P gaming?

1070 and up is overkill for 1080p unless you are seeking FPS over 60. Neither of those CPUs will significantly hold back a 1070/1080 in games. If you are a numbers chaser then you need a faster CPU. It's going to be tough for me to justify an upgrade if I had the i7-4930K at 3.9GHz. All of this is a very personal decision.
 
Funny... a 1070 is a sweetspot gpu for 1080p to me. Considering one cant run ultra with AA on every title with a 1060... :)

Similar to the 700 gpu amd 600 gpu, i dont have a pc to have it look like a console either. If people are good with turning down some settings in some games, its solid. Bjt i prefer to run things on ultra with aa and over 60 fps.
 
Yes, those minor differences i am aware of. It doesnt change the fact that a 1080ti is being held back by the platform. The goal for the pereon is 144hz gaming...to get there, all bottlenecks should be removed as much as possible... i dont buy a 700 gpu to have it run like a 600 gpu. ;)

I bought a used GTX 980 for this old system. Got a good deal and with a BIOS flash it runs well. I will throw that card back in the X58 system tonight and post up my observations in BF1 since I already have that installed.
 
Funny... a 1070 is a sweetspot gpu for 1080p to me. Considering one cant run ultra with AA on every title with a 1060... :)

Similar to the 700 gpu amd 600 gpu, i dont have a pc to have it look like a console either. If people are good with turning down some settings in some games, its solid. Bjt i prefer to run things on ultra with aa and over 60 fps.

I think in many titles the difference between Ultra and High is almost indiscernible except performance can sometimes take a huge hit. But yeah, I like the eye candy too.

Still, I am pretty impressed with the X58 platform for what it is. It is not a goto system in today's world by any means. Although cheap 6 core performance was very enticing for many before the coming of Ryzen. Now, not so much.
 
Indeed, tough to discern. The AA is more of what i am talking about really. Just dont want to buy new to already have to turn things down. :)
 
Inno 3D GTX 1070 Airboss X3 was more a future-proofing decision. Apply OC to videocard and CPU monitor hoping that would provide a good UX well into the future. I think this constant upgrading is wasteful use of the Earth's resources, in the longterm this will lead to mankinds destruction at his own hands.
 
I thought the point was to see if the CPU was holding the fps back in BF1 at 1080p or 1440p that is what he is testing.

...and that is what testing at 1080p and higher is for. Testing lkwer doesnt show much of anything except for that res.
Review sites test at 720p, 1080p to see if there is more FPS to be gained from the CPU, that will show if the CPU is bottlenecking at 1080p, 1440p. When you lower the resolution to 720p it does not lower the CPU instructions needed for the GPU, it increases the instructions needed from the processor to keep the GPU busy.

When folks only test at one resolution there is know way to see if it is the CPU or GPU bottlenecking.
 
Last edited:
Review sites test at 720p, 1080p to see if there is more FPS to be gained from the CPU, that will show if the CPU is bottlenecking at 1080p, 1440p.
Yes.. and i disagree with that kind of testing being meaningful is my point.

People play at 1080p and higher. I dont find running a lower resolution, an exxageration of the actual playing environment, to tell anyone anything outside of what it does at that res. What does 720p giving a result of 15% improvement have to do with a 10% improvement at 1080p? People look at 720p testing thinking, hmm...im going to get the same/similar results at higher resos when clearly, that isnt true. You can see gains from 720p testing you wouldnt see at 1080p. It also varies by game how much or how little cpu power is needed.

When you lower the resolution to 720p it does not lower the CPU instructions needed for the GPU, it increases the instructions needed from the processor to keep the GPU busy.
well aware... this is part of my point. Is there something i said to make you believe i was saying otherwise???

When folks only test at one restitution[sic] there is know[sic] way to know...
If you want to see if a cpu is holding you back, overclock it!!


As far as what was said and what i actually disagreed with.....
Then he should test at 720p, 1080p, 1440p maximum, so we can see what performance he is leaving on the table in BF1.;)
 
Last edited:
Hi guys


I made another video with my older system but the new GTX 1080ti in Deus Ex: Mankind Divided.

Everything is on Ultra or Very High (where there is no Ultra option).
Only Motion Blur is not used, since I hate that.

Video:
 
Last edited:
Hi guys


I made another video with my older system but the new GTX 1080ti in Deus Ex: Mankind Divided. Everything is on Ultra or Very High (where there is no Ultra option).
Only Motion Blur is not used, since I hate that ****.

Video:

Looks like good FPS. What does the benchmark do at 1080p?
 
1080p vs 1440 - Deus Ex Mankind Divided

Looks like good FPS. What does the benchmark do at 1080p?


1080p
1080p.png


1440p
1440p.png

In both cases:
  • GPU utilization was at about 99%.
  • GPU kept a steady 2012 Mhz boost on core clock
  • CPU utilization was at about 60%.
  • CPU was overlocked to 4000 Mhz with HyperThreading on.
 
Back