• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Unreal Engine 4 Kite Demo, custom uncapped benchmarks (fraps)

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
would be willing to seedbox this if you SHA-1 your RAR, psolord

actually, probably going to unrar on my seedbox and 7zip it back up as i don't want to taint my system with WinRAR... (haven't had to use it in years)
 
Last edited:
By pressing "page up" when the demo loads, takes it to a non interactive, self executing, on a specific path and that's why I thought it should be uncapped in order to become a very decent gpu benchmark.

You can also use WASD and QE for height along with the mouse, to navigate it yourself, which is also pretty interesting.

As for the seedbox, I am not sure what you are saying. Not a big fan of p2p and so my knowledge is limited on the matter.
 
5820k @ 4.4ghz + 980 ti @ 1500mhz

1080p:
Frames, Time (ms), Min, Max, Avg
6866, 122578, 35, 174, 56.013
 
one more attempt!

anyone interested in a torrent of the archive 7zipped? (i upgraded to beta 15.xx in an attempt to decompress the RAR. hopefully the bitstream/archive format hasn't changed since 9.20 stable)

https://yourmom.likesbuttse.xxx/kite_dev.7z.torrent
(not sure if you'll need to add 107.161.22.123:55558 as peer)

you (or any other downloaders) able to confirm these EXE SHA-1s, psolord?

Code: 
< kite_dev\WindowsNoEditor\KiteDemo.exe >
  MD5: AB892C8E1D8E8F054F8FB08AEB0AC3A4
  SHA-1: C4833D126F280C800D1BF041BE37C77082195C05
  SHA-256: 3D57E717FE6BF1D30B22F70513894CE5A5F7A2B0B2326DD98D478968A9199717

< kite_dev\WindowsNoEditor\KiteDemo\Binaries\Win64\KiteDemo.exe >
  MD5: BD3740D144364FD88654D6216122B926
  SHA-1: 5C271566566D6FD047449E1F3A9885F9250AD0D4
  SHA-256: BF941A922BDE6C381751CF7D2FDF3A109C9801BD3871F0C2B086F00E68C36EC8

Code: 
< kite_dev.7z >
  MD5: 4ED027E77F1ADDA664DB40D71775CC9D
  SHA-1: 1854EEA8474BF814C09508F89A1F88908ED96234
  SHA-256: 39F1C28EF7D3F50BE32E557FA3A46A76D4438918ECE43DF3EFF6643B29C673A5

if anyone else is interested, i might be able to throw it on my faster seedbox, too.
 
.vodka said:
Amazing demo in video back then, even more amazing seeing your PC do it.


R9 290 @ 1150/1500, 2500k @ 4.5GHz, Catalyst 15.5 beta, 1080p

2015-06-28 13:57:45 - KiteDemo
Frames: 4361 - Time: 133406ms - Avg: 32.690 - Min: 15 - Max: 102
Click to expand...

R9 290X DirectCU II at defautl settings 1050/1350

But here the important part: i7-870 @ 3.4 Ghz

Frames Time (ms) Min Max Avg
3564 128170 3 80 27.807

So this is clearly highly CPU limited at least pre-Sandybridge. Time to upgrade.
 
okay, running on a factory i7-4770K and GTX 780 w/ 16GB and 840 EVO SSD

here are my benchies:

1080p
Max 31
Min 3
Avg 22.952

1050p
Max 79
Min 15
Avg 42.008
 
Alright, new rig, new bench!
i7-6700k Factory
EVGA GTX 1080 SC
32GB RAM
950 Pro 512GB

I tried 3440x1440, but it gave me black bars on the side. I'm guessing that's the 16:9 version of the 1440p...

Anyways ---

1440p
AVG 46.270
Min 28
Max 150

1080p
AVG 67.420
Min 41
Max 219
 
3x average improvement from a 780, excellent. Looking forward to upgrading mine to one of these soon.

What were each card clocked to?
 
Thanks finbarqs

I'd expect it to be even faster than the 980ti posted above, but still an awesome result.

Anyone with a 1070?
 
lol... don't think my case can handle OC... I'm using factory intel 95W cooler, in an inwin 901 case 😛

I'm pretty happy with it w/o any OC right now. I've been OC'ing before, and usually they fail after a couple of years.. making go back to factory clocks anyways... So I figure, I'm done with it... But maybe I can push my GTX 1080 a little more..
 
Wow thanks so much for uploading this unlocked version. Capped to 180fps I scored:

GTX 1080 ti, 4.4ghz 6600k, 3000Mhz 16GB memory (min,max) --- avgfps
2160p: (14,54) --- 29 fps
1440p: (35,121) --- 55 fps
1080p: (52,180) --- 81 fps
 
I remember when I first saw this. Was amazing. Really cool to see it running in real time at higher resolution and framerate.

The bench is basically 2 minutes on the nose so I set Fraps to bench for 120 seconds. Hitting the bench key the moment the UE4 loading icon goes away results in very consistent results. Ends right after the cave scene on the moment it goes to black.

Vsync off and no fps cap:

1080P
Frames, Time (ms), Min, Max, Avg
10585, 120000, 56, 236, 88.208

1440P
Frames, Time (ms), Min, Max, Avg
7507, 120000, 39, 204, 62.558

I tried to get it to run at 4K on my 1440P display, using Nvidia DSR, but it wouldn't take. Might try it on my TV later on.


MSI GTX 1080ti Gaming X Stock / 4.2 6850K / 32GB 2666
 
1440p OC results are (41,153) --- 63.708 fps, Zotac 1080 ti mini. I was accidentally in 90 percent power mode the first time i did it.

Beautiful, really looking forward to the next great UE4 demo 🙂
 
Alistair said:
Wow thanks so much for uploading this unlocked version. Capped to 180fps I scored:

GTX 1080 ti, 4.4ghz 6600k, 3000Mhz 16GB memory (min,max) --- avgfps
2160p: (14,54) --- 29 fps
1440p: (35,121) --- 55 fps
1080p: (52,180) --- 81 fps
Click to expand...

You are welcome mate. Thanks for sharing your results, as well as the rest of the guys.

Wow the 1080Ti is a beast. I take it, it was at stock?

Still a lot faster than my 1070@2000/9000.

Here are my 1070 results, compared with my 970 results. Same system, same demo version, different drivers however. Not that a new driver would help the 970 in any meaningful way.

DHKm4qqL.jpg


The 970's average is about the same a the 1070's minimum, lol.

Granted that is not the norm in most games I have benchmarked. The 970 still does an admirable job.
 
EVGA 1080 ti superclocked OC @2012/11908 (+70/+450 cpu/mem, mostly voltage and power limited) with a 6600k at 4.2ghz:

1080p: (61,236) --- 93.1 fps
1440p: (43,206) --- 66.4 fps

It's about 5 percent higher than the Zotac mini.

I did this with an ultrawide monitor but I assume the black bars were correct. I also like Unigine's Valley, and using the "A New Dawn" demo from nVidia for benchmarking and testing 🙂
 
Last edited:
3770k @ 4.2, Vega 56 , flash with 64 bios, power limit 50%, everything else dynamic.

2017-09-11 11:29:20 - KiteDemo
Frames: 7684 - Time: 138157ms - Avg: 55.618 - Min: 29 - Max: 164

edit: with HBCC just to see if it made a difference

2017-09-11 11:44:37 - KiteDemo
Frames: 7393 - Time: 132750ms - Avg: 55.691 - Min: 30 - Max: 167
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top