Jump to content

V-Sync with Ati Catalyst 10.3


Kankkis

Recommended Posts

Distiler, it not the same thing what V-Sych does, i know when put Vsync on you cannot override your monitor HZ, i'll tryed it in FC2.0. Max FPS60 mut no V-Sync without V-Sync


Edited by Kankkis
Link to comment
Share on other sites

  • 2 weeks later...

did you notice any FPS boost with D3Doverider vsync and its DX triple buffer option ? stutter improvement ?

 

Is D3Doverider included in MSI afterburner too?

.

 

i7 880 | HD 7870 | 8 Gb DDR3 1600 | ECS P55H-A | OCZ Vertex 2 180 | Intel 330 180 | WD 500 AAKS | 2x WD 2T Green | Enermax Liberty 620 | CH Combatstick & Throttle | TrackIR 3 | HP ZR24W | Windows 7 x64

Link to comment
Share on other sites

The triple buffering means that the V-sync no longer forces the frame rate to Division of 60Hz (if that is your refresh rate). So instead of only having 5,10,15,20,30,45, and 60hz as available frame rates Triple buffering allows every thing from 1-60hz. So frame rate changes are fluid and not done in big jumps as with Double buffering.

 

You will notice a big difference between V-sync only and V-sync with triple buffering.

 

The triple Buffering allows V-sync with no frame rate loss essentially.

 

Nate

Link to comment
Share on other sites

isn't the the triple buffering of CCC (refer to OGL?) actually working ? (with DX)

.

 

i7 880 | HD 7870 | 8 Gb DDR3 1600 | ECS P55H-A | OCZ Vertex 2 180 | Intel 330 180 | WD 500 AAKS | 2x WD 2T Green | Enermax Liberty 620 | CH Combatstick & Throttle | TrackIR 3 | HP ZR24W | Windows 7 x64

Link to comment
Share on other sites

The triple buffering means that the V-sync no longer forces the frame rate to Division of 60Hz (if that is your refresh rate). So instead of only having 5,10,15,20,30,45, and 60hz as available frame rates Triple buffering allows every thing from 1-60hz. So frame rate changes are fluid and not done in big jumps as with Double buffering.

 

You will notice a big difference between V-sync only and V-sync with triple buffering.

 

The triple Buffering allows V-sync with no frame rate loss essentially.

 

Nate

Are you completely sure it worked for you? Can you send me a text file from FRAPS benchmark with at least two different FPS values kicking in. Plus your monitor refresh reate of course. I'd appreciate that.

Link to comment
Share on other sites

Are you completely sure it worked for you? Can you send me a text file from FRAPS benchmark with at least two different FPS values kicking in. Plus your monitor refresh reate of course. I'd appreciate that.

 

I'll be honest, I just can't tell if it works or not, see the results and judge for yourself.

 

The track used is the Su-25 bomb delivery track, and the only changes were Vsync and Triple Buffering.

 

Refresh Rate is 60hz.

 

Nate

Link to comment
Share on other sites

I think its for ATI cards, I don't think nVidia cards suffer the same V-Sync issues

Of course it's not nVidia specific problem. see nHancer website articles (help).

 

I'll be honest, I just can't tell if it works or not, see the results and judge for yourself.

 

The track used is the Su-25 bomb delivery track, and the only changes were Vsync and Triple Buffering.

 

Refresh Rate is 60hz.

 

Nate

I really tried to make something out of this - the smallest number of data points I have ever seen in a benchmark :)

 

Vsynch_tripplebuffNate_wrong.png

 

There are two possibilities:

this bench is useless, make a longer one

Vsync+tripple works in both cases because ~50 is not equal 1/2 of 60

 

 

Again, with such a serie one can come up with more conclusions :P

 

Edit:

Please specify the test configuration. What was the test running time and how many FPS values were recorded per second. It looks like 1 FPS value per second, running time over a minute.

 

Edit:

isn't the the triple buffering of CCC (refer to OGL?) actually working ? (with DX)

Tripple buffering works normally only in OpenGL. That's why you need such quirky tools like D3Doverider. It will either work or not. It didn't work for me with DCS as far as I remember.

 

 

Here's my results for 75 Hz ref rate. nVidia control panel Tripple buffering setting will not have any effect so don't even try it.

 

Vsynch_weird_liketripplebuff-1.png

 

As you can clearly see in theory I should get 1/2 * 75 = ~30 FPS as soon as my rig is not able to maintain FPS above 75. This is not the case.

 

The conclusions are:

1. The D3Doverrider did diddle squat for Nate :D

2. DCS appears to have some mysterious solution for Vsync limitations.

patrol_mis_1_edited2_bench_bucic.trk


Edited by Bucic
Link to comment
Share on other sites

I can say with 90% probability that DCS Black Shark uses special Vsync implementation (wait for Vsync ?) that allowes to bypass Vsync limitation re FPS being integer fractions of monitor refresh rate so you don't need any tools for forcing tripple buffering (like D3DOverrider).


Edited by Bucic
Link to comment
Share on other sites

It does look that way doesn't it? But don't forget that there is also the render ahead technique favoured by DirectX that could be at play here. It is very Difficult to spot input lag in DCS compared to a First Person Shooter for example.

 

Have a read of this (the Render Ahead stuff is at the end)-> http://www.anandtech.com/show/2794

 

Nate

 

Edit:- as for the Bench results, I've not done any Benchmarking before, sorry. The Datapoints were the FPS from a track about 70sec long. Fraps was not entirely accurate either, as can be seen from the FPS of 62 in some places.


Edited by Nate--IRL--
Link to comment
Share on other sites

Nate, to be honest I don't care what the method is. I don't see lag and there is absolutely no reason for BS users not to enable Vsync!

 

I know, sorry if I sounded grumpy/mocking or anything like that. The reason i asked for the benchmark parameters is I haven't done any for a long time myself and I didn't even know what am I looking at. Plus FRAPS doesn't feature any "benchmark parameters" :)

 

Just enjoy it people! Other simmers are not that lucky! :thumbup:

Link to comment
Share on other sites

Nate, to be honest I don't care what the method is. I don't see lag and there is absolutely no reason for BS users not to enable Vsync!

 

I know, sorry if I sounded grumpy/mocking or anything like that. The reason i asked for the benchmark parameters is I haven't done any for a long time myself and I didn't even know what am I looking at. Plus FRAPS doesn't feature any "benchmark parameters" :)

 

Just enjoy it people! Other simmers are not that lucky! :thumbup:

 

 

Since I got my new gaming rig I've been trying ALL types of settings. I DO have to use D3DOverrider for Vsync... But I have played around with Triple Buffering in my Catalyst Control Center and on D3D... I haven't noticed a dang difference at all with or without Triple Buffering. I just settled on having it off about 3 days ago. I also turned VSync off for a while and tried to deal with the tearing... But with it off I never saw a jump in framerates. It's good to have conformation from your benchmarks.

 

By the way, should I turn on Catalyst AI or leave it off? I thought I heard a mention that it could cause weird graphic anomolies and such if I left it on while playing BS.

Link to comment
Share on other sites

Since I got my new gaming rig I've been trying ALL types of settings. I DO have to use D3DOverrider for Vsync... But I have played around with Triple Buffering in my Catalyst Control Center and on D3D... I haven't noticed a dang difference at all with or without Triple Buffering. I just settled on having it off about 3 days ago. I also turned VSync off for a while and tried to deal with the tearing... But with it off I never saw a jump in framerates. It's good to have conformation from your benchmarks.

 

By the way, should I turn on Catalyst AI or leave it off? I thought I heard a mention that it could cause weird graphic anomolies and such if I left it on while playing BS.

Yes, DCS is a "Vsync ready TM" application and no tools are needed. Even nVidia control panel tripple buffering can be disabled. You just enable the Vsync there and there you go.

 

As for AI - I don't know. I have an nVidia card.

Link to comment
Share on other sites

 

By the way, should I turn on Catalyst AI or leave it off? I thought I heard a mention that it could cause weird graphic anomalies and such if I left it on while playing BS.

 

Have been doing exhaustive Benchmarks while looking for stutter/anomalies/popups.

 

 

 

 

 

For me:

  • A.I. off caused slight increase in stutter and tearing, a few extra % fps. Hi/Low made no decernalbe difference in FC2 (some other titles yes).
  • Vsync on/off no discernable difference
  • AA settings yes to stutters. I now use a 4x box or edge not 8x
  • Fullscreen on! Some connection to hi AA settings ( loss of 5-15 % FPS with "off"). Fullscreen off, low AA = no tearing, less stutter. yes a bit confusing.
  • Shadows - All Planar a few % FPS increase
  • Water as low as it will go, not much in FPS except High

Phenom11 x3 720 @ 3.5Gig

Win7 64

ATI 4850s 512mg x 2 (xfire) but still only total 512mg (limiting factor)

10.2 drivers

At 1920x1200 rez.

 

Shad

 

PS Vsync wont work with Full Screen enabled!!


Edited by ShadowVonChadwick

RyZen5 3600x, MSI GamingX RX 5700xt, AX-370-K7, 16 Gig G-Skil 3200 :thumbup:, Antec 650w (Still),Win10 on 256G 870 NVMe, 860+850 Evo for Apps, 2x1TB WD HDs for :music_whistling:, TR5 :detective:, Hog stick:joystick:, 3x TM MFD Bezels. a 32" AOC, @ 2560x1440, no floppy & a crappy chair :pain:. Its hard to find a chair that accepts you as you grow.:pilotfly:

Link to comment
Share on other sites

Have been doing exhaustive Benchmarks while looking for stutter/anomalies/popups.

 

 

 

For me:

  • A.I. off caused slight increase in stutter and tearing, a few extra % fps. Hi/Low made no decernalbe difference in FC2 (some other titles yes).
  • Vsync on/off no discernable difference
  • AA settings yes to stutters. I now use a 4x box or edge not 8x
  • Fullscreen on! Some connection to hi AA settings ( loss of 5-15 % FPS with "off")
  • Shadows - All Planar a few % FPS increase
  • Water as low as it will go, not much in FPS except High

Phenom11 x3 720 @ 3.5Gig

Win7 64

ATI 4850s 512mg x 2 (xfire) but still only total 512mg (limiting factor)

10.2 drivers

At 1920x1200 rez.

 

Shad

 

Spot on my friend. Those are almost my exact settings. I noticed a significant FPS drop when I cranked up the AA just like you. AF is about the same, I keep it fairly low also. I have to use D3DOveridder for Vsync, because our CCC Vsync simply does not work with it. The screen tearing for me gets pretty intense without it, especially noticeable with TrackIR.

 

I've turned Catalyst A.I. on to advanced and I haven't had any weird stuff happen just yet. But when I changed my sampling mode for AA I ran into some weird stuff with the cockpit. So it stays on the basic setting.

 

As far as water and all that stuff, on DCS I've tweaked all the LUA files for the best.

 

 

One thing the Goat forgot when building his machine was a danged sound card.... I think my lack of a true sound card is causing a good amount of FPS loss due to my processor and board having to deal with all of it. I'll wait until the patch comes out and see if there is a major difference there, if not, I'll be ordering a sound card ASAP. (I'll eventually end up getting one no matter what, but I'm trying to figure out if my current priority should be an extra 4 gigs of DDR3 RAM, taking my system up to 8... Or a sound card... Any suggestions would be nice.) :smilewink:

Link to comment
Share on other sites

Hi Goat

 

My impressions is that these days onboard sound is OK unless your a muso.

 

5/10 years back had many a fickle fight with sound drivers. And onboard was shit.

 

4 gigs is plenty or sufficient. My next toy will be a solid state drive, just a small one first.

RyZen5 3600x, MSI GamingX RX 5700xt, AX-370-K7, 16 Gig G-Skil 3200 :thumbup:, Antec 650w (Still),Win10 on 256G 870 NVMe, 860+850 Evo for Apps, 2x1TB WD HDs for :music_whistling:, TR5 :detective:, Hog stick:joystick:, 3x TM MFD Bezels. a 32" AOC, @ 2560x1440, no floppy & a crappy chair :pain:. Its hard to find a chair that accepts you as you grow.:pilotfly:

Link to comment
Share on other sites

Spot on my friend. Those are almost my exact settings. I noticed a significant FPS drop when I cranked up the AA just like you. AF is about the same, I keep it fairly low also. I have to use D3DOveridder for Vsync, because our CCC Vsync simply does not work with it. The screen tearing for me gets pretty intense without it, especially noticeable with TrackIR.

 

I've turned Catalyst A.I. on to advanced and I haven't had any weird stuff happen just yet. But when I changed my sampling mode for AA I ran into some weird stuff with the cockpit. So it stays on the basic setting.

 

As far as water and all that stuff, on DCS I've tweaked all the LUA files for the best.

 

 

One thing the Goat forgot when building his machine was a danged sound card.... I think my lack of a true sound card is causing a good amount of FPS loss due to my processor and board having to deal with all of it. I'll wait until the patch comes out and see if there is a major difference there, if not, I'll be ordering a sound card ASAP. (I'll eventually end up getting one no matter what, but I'm trying to figure out if my current priority should be an extra 4 gigs of DDR3 RAM, taking my system up to 8... Or a sound card... Any suggestions would be nice.) :smilewink:

Forget about extra RAM for DCS BS. 4 GB is more than enough for it and even RAMDISK will fit there in! I recommend affordable Asus Xonar DS or slightly more expensive DX. PCI-E version available of the latter one.

 

AA:

Did you know that certain 8x mode (CSAA) will provide almost the same performance as certain 4x mode (MSAA) and higher quality of course?

 

http://i17.photobucket.com/albums/b68/Bucic/A2Aforum/nVidia_settings_Bucic.jpg

 

In case you are using Frazer's Tweak Guide:

http://forums.eagle.ru/showpost.php?p=884734&postcount=62

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...