• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Mantle Feedback/Bugs

After seeing pgi's frame latency testing on 780 SLI @1080p i decided to hunt down the program he used and test it on Mantle at 1440P instead of 1080P. I also tried each of the three different Mantle frame pacing implementations to see if i could feel a difference. I couldn't.

2 = Extreme
1 = Standard
0 = Disabled

Frame Pacing Method 2
9lBdznB.jpg


Frame Pacing Method 1
3CuNIYV.jpg


Frame Pacing Method 0
jAWbA7T.jpg
 
Last edited:
Was that using a fraps output or the direct bf4 output? I have no idea if the internal engine based tests would be more accurate or not, but they do lack timed bench options.
 
Was that using a fraps output or the direct bf4 output? I have no idea if the internal engine based tests would be more accurate or not, but they do lack timed bench options.

BF4 output. That was using Perfoverlay.framefilelogenable 1 once i was loaded into the map and started benching and Perfoverlay.framefilelogenable 0 when i completed the 60 seconds going on a stopwatch on my phone. Not an exact science. :p
 
After seeing pgi's frame latency testing on 780 SLI @1080P i decided to hunt down the program he used and test it on Mantle at 1440P instead of 1080P. I also tried each of the three different Mantle frame pacing implementations to see if i could feel a difference. I couldn't.

2 = Extreme
1 = Standard
0 = Disabled

Frame Pacing Method 2
9lBdznB.jpg


Frame Pacing Method 1
3CuNIYV.jpg


Frame Pacing Method 0
jAWbA7T.jpg

Oh crap. I just realised i did this testing with my pc in power saving mode, as i left it on last night downloading games. That means my cpu was locked at 3.9ghz and my ram was running at 1600mhz 9-9-9-27-2 and maximum power savings were set in windows lol. :D

Will run these tests again now with my cpu at 5ghz and my ram at 2133mhz.
 
After seeing pgi's frame latency testing on 780 SLI @1080P i decided to hunt down the program he used and test it on Mantle at 1440P instead of 1080P. I also tried each of the three different Mantle frame pacing implementations to see if i could feel a difference. I couldn't.

2 = Extreme
1 = Standard
0 = Disabled

Frame Pacing Method 2
9lBdznB.jpg


Frame Pacing Method 1
3CuNIYV.jpg


Frame Pacing Method 0
jAWbA7T.jpg

Nice testing matt, I using pacing 1 and find it superb.
 
After seeing pgi's frame latency testing on 780 SLI @1080P i decided to hunt down the program he used and test it on Mantle at 1440P instead of 1080P. I also tried each of the three different Mantle frame pacing implementations to see if i could feel a difference. I couldn't.

2 = Extreme
1 = Standard
0 = Disabled

EDIT - REALISED I WAS RUNNING THESE TESTS IN POWER SAVING MODE. CPU @3.9GHZ LOCKED AND DRAM AT 1600MHZ.

Frame Pacing Method 2
9lBdznB.jpg


Frame Pacing Method 1
3CuNIYV.jpg


Frame Pacing Method 0
jAWbA7T.jpg

Will have results of 5.2ghz vs 3.9ghz Mantle shortly.

This time tested at 5.2ghz and 2133mhz ram speed with tight timings. There is virtually no difference. :eek:

2 = Extreme
1 = Standard
0 = Disabled

Frame Pacing Method 2
0icfpkR.jpg


Frame Pacing Method 1
FxEeLsG.jpg


Frame Pacing Method 0
IrQ0Ise.jpg
 
Hi Matt,

I had my hand in my pocket for a 4k monitor but this article has changed my mind. Bit risky on expecting smooth performance on a 7990 and a 4k monitor as essentailly all monitors are seen as 2 screens and frametime smoothness still suffers under multimonitor setups.

Bit annoying as I was looking forward to when these £500 Sammys went 'In stock' or the £600 asus.

What to do?
 
Hi Matt,

I had my hand in my pocket for a 4k monitor but this article has changed my mind. Bit risky on expecting smooth performance on a 7990 and a 4k monitor as essentailly all monitors are seen as 2 screens and frametime smoothness still suffers under multimonitor setups.

Bit annoying as I was looking forward to when these £500 Sammys went 'In stock' or the £600 asus.

What to do?

I woudn't bother.

Get a 1440/1600p monitor with/without a 120/144hz refresh rate.
 
Hi Matt,

I had my hand in my pocket for a 4k monitor but this article has changed my mind. Bit risky on expecting smooth performance on a 7990 and a 4k monitor as essentailly all monitors are seen as 2 screens and frametime smoothness still suffers under multimonitor setups.

Bit annoying as I was looking forward to when these £500 Sammys went 'In stock' or the £600 asus.

What to do?

I'd get a 1440P/1600P screen as 3gb vram might be cutting it a bit fine at 4K. One 7990 probably would be a bit lacking grunt wise as well. I guess it all depends on what settings you want to play with and what fps you consider playable. For me id want around 60 fps at high/highest settings, or close to highest settings with sacrificing AA to be happy at 4k. To do that you'll need 2x290's or better imo. However if you don't mind lowering the details then im sure a 7990 would be fine at 4k.
 
I'd be prepaered to drop settings until I could add a card to my 7990 or until I upgrade the gfx to 4k maxed out solution in the future.

And I would have had a 1440p 120hz months ago if you can get one in the UK. But all 120hz 1440p's AFAIK are obtainable only from importing. Fair play to those that have.

Only the ASUS ROG 1440 120hz is touted to appear at some point. But they'll go out of stock quicker than £10 notes for £a fiverand they'll be £800ish.

With Gregsters honest and positive review of his 4k tn panel, I was tempted with the lure of cheap (ish) 4k. I dont do image editing or photography so tn panel for me is fine. I only game with my PC.


The framepacing on GCN 1.0 cards highlighted in that article pushes me back towards 1440 and 120hz but you can't get one in the UK unless 2nd hand can you?
 
Back
Top Bottom