• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Why no Vsync?

KenG10 said:
Try it and let me know how it goes. Do you have Half Life 2? Enable Vsync and then use the console command "FPS_max 57" or 58.

Or try the UT3 ini update to "MaxSmoothedFrameRate" = 59.9 if you have that game.
Click to expand...

oddly enough, source games are one of the few that give (for me at least) no input lag issues.
 
KenG10 said:
Try it and let me know how it goes. Do you have Half Life 2? Enable Vsync and then use the console command "FPS_max 57" or 58.

Or try the UT3 ini update to "MaxSmoothedFrameRate" = 59.9 if you have that game.
Click to expand...


well I dont notice the lag in those games to begin with so even if that worked I probably wouldnt notice it. still it goes against what I thought was a contributor to lag which is dropping below the 60fps with vsync on.
 
Violator said:
Maybe the system isn't quick enough to push more fps than the monitor can display ;)
Click to expand...
maybe so. lol. also there can still be mild tearing even when you dont actually exceed the refresh rate. I can see plenty of tearing in some games even if the fps is in the 40s or 30s.
 
cannondale06 said:
well I dont notice the lag in those games to begin with so even if that worked I probably wouldnt notice it. still it goes against what I thought was a contributor to lag which is dropping below the 60fps with vsync on.
Click to expand...

When I was researching this I was given the impression it is due to frames that would have normally been processed above the 60fps mark that are causing the lag. Something about buffers being full and waiting for something. I'm no video card pro, so I apologize for such a childish response, but in every case, limiting the FPS to just below 60Hz has eliminated all lag for me. *shrugs*

In HL2, I noticed it even on my machine. When I set FPS_max 59, movement became super crisp.
 
cannondale06 said:
so where do you get this special setup that has no tearing? lol
Click to expand...

I'm not sure. Maybe it's been there and I havn't noticed. I'm almost afraid to look for it now when I get home, but... /shrug. I'll whip out the digital cammera when I get home and we'll see.

The rig is in the sig and most games run 60+FPS.
 
vengence said:
I'm not sure. Maybe it's been there and I havn't noticed. I'm almost afraid to look for it now when I get home, but... /shrug. I'll whip out the digital cammera when I get home and we'll see.

The rig is in the sig and most games run 60+FPS.
Click to expand...
well every game is different and even a different monitor can make it more or either less noticeable. the original FEAR was one of the first games I remember that was really bad with vsync off. the flickering lights only amplified the screen tearing and made me almost go nuts. STALKER and Clear Sky have really bad tearing even with very low framerates.
 
Tearing don´t bother me and i hate mouse lag.

Vsync off

Taken from wikipedia:

Computer games often allow vertical synchronization as an option, because it delays the image update until the vertical blanking interval. This can cause lowered frame rates due to latency (the period of the refresh rate at maximum), which might be undesirable in games that require fast response (e.g. first person shooters), although in practise it is impossible to achieve a visible frame rate faster than the refresh rate of the monitor.
 
Last edited:
I've never noticed mouse lag with Vsync on. Granted I never looked for it either, so now I'll probably go play a game and notice it and start crying because I'll have to turn my beloved Vsync off. :p I do play FPS games online, and usually do quite well, but nothing competitive like clans and crap.

Ignorance is bliss sometimes. The more you learn the more fussy you become.
 
Tudz said:
I've never noticed mouse lag with Vsync on. Granted I never looked for it either, so now I'll probably go play a game and notice it and start crying because I'll have to turn my beloved Vsync off. :p I do play FPS games online, and usually do quite well, but nothing competitive like clans and crap.

Ignorance is bliss sometimes. The more you learn the more fussy you become.
Click to expand...
very true. I actually enjoyed gaming much more when I didnt know how or why all this stuff worked. pc gaming can really be quite a pain those of the anal retentive nature.
 
ok yea, i did turn off v-sync on fallout. there was too much grass, even when i lowered the grass count there was some input lag with all settings at max.

other than that. perma-vsync for me.
 
Ignorance is bliss sometimes. The more you learn the more fussy you become.
Click to expand...

I agree. I was completely convinced that LCDs were horrible awful things that looked like utter crap. Then I went and looked at the monitor I own in person at the store and realized it was very nice, and while I could see what people complained about, it wasn't anything like how bad they made it out to be.
 
oh vsync never on my computer. i do notice the tearing but all i play are fps, the input lag caused by vsync to too much to cope with
 
Always off for me. Can't stand input-lag, and tearing I barely notice anymore. Games like CS1.6 or, perhaps an even better example, Audiosurf, are night and day depending on how vsync is set. Would love a vsync mode without the input lag, though.

Does anyone know why vsync even affects the input-response time? Does it force the input down to 60 checks a second as well as the video?
 
Since nobody's linked them yet, I'll link Tweakguides' pages on Vsync and Triplebuffering, which remain the most complete description of Vsync and Triplebuffering I've seen, and after careful reading will answer many of the questions asked in this thread:

http://www.tweakguides.com/Graphics_9.html (Vsync)
http://www.tweakguides.com/Graphics_10.html (Triplebuffering)

I'll also take this opportunity to point out that forcing Triplebuffering on in the nVidia or ATI drivers only works for OpenGL games; for DirectX games, you need to use other utilities to make Triplebuffering work. Nobody ever knows this because, at least in the nVidia Control Panel (I'm not familiar with ATi drivers), no indication is given of this important detail.
 
I play with it off since I am a ball to the walls [H]ard core gamer, no Im kidding, I just find that playing the games I play it works out better with it off.
 
Does anybody notice that tearing is less noticeable the higher the framerate? I find that when the fps is over 90+, tearing is minimal. When the fps is around 60-70, tearing is very very noticeable.

I used to play cod4 with vsync enabled and never noticed any input lag, but then I started to play CS:S without vysnc for awhile and when I went back to playing cod4, that's when I started to feel the input lag and had to disable vsync.
 
Skunt said:
Does anybody notice that tearing is less noticeable the higher the framerate? I find that when the fps is over 90+, tearing is minimal. When the fps is around 60-70, tearing is very very noticeable.

I used to play cod4 with vsync enabled and never noticed any input lag, but then I started to play CS:S without vysnc for awhile and when I went back to playing cod4, that's when I started to feel the input lag and had to disable vsync.
Click to expand...
that sounds backwards. the higher over the refresh rate you go the more tearing you should see.
 
vsync have a weird input lag, its not a big deal for some people, but it bugs me a lot in CS Source, it makes me feel like 1 sec delay....
 
No one ever mentions the mouse lag from vsync.
Am I the only person to ever notice it? Yes, even with "Flip Queue Size" (ATI) and "Pre-render Limit" (Nvidia) set to low values, it's still there.

Incredibly annoying.
However, I always use vsync anyway. Tearing is worse than a little lag.
For what it's worth, using Flip Queue Size value 0 on my 4870 is much less laggy than my 8800GTX was.

edit;
My prayers have been answered, someone mentioned mouse lag before me.
 
well, if you read this thread, you'd probably notice at least 10 replies with people talking about input lag.

seems to vary widely from system to system, regardless of specs.
 
I dont use vsync because I CAN feel the difference between 60 an 150 FPS. And I hardly notice any tearing what so ever.
Not to mention the aforementioned input lag (albeit small) which does not help its case what so ever in my opinion.
 
Vsync caused screen tearing a few games (bethesda) and drastically lowered the framerate.
 
with a 120hz lcd, and an array of different AA levels to choose from, you dont need vsync at all.

also, in a multiplayer fps, ppl get a competitive advantage from having more fps and vsync off.
 
grimreeferx09 said:
with a 120hz lcd, and an array of different AA levels to choose from, you dont need vsync at all.

also, in a multiplayer fps, ppl get a competitive advantage from having more fps and vsync off.
Click to expand...
framerate doesnt have to go above refresh rate to tear. its just much worse if it does.
 
My quick take:

Vsync on, refresh 60: Liquid smooth if framerate is 60, but if it drops under 60 than not only the framerate drops to 30 or 40 but the game lags/slows down for a split second, which is annoying to me.

Vsync off, any refresh: Screen tearing with super-high framerate, but very helpful on slower hardware (especially if 30, 40, or 60fps is barely out of reach)
 
geforce said:
My quick take:

Vsync on, refresh 60: Liquid smooth if framerate is 60, but if it drops under 60 than not only the framerate drops to 30 or 40 but the game lags/slows down for a split second, which is annoying to me.

Vsync off, any refresh: Screen tearing with super-high framerate, but very helpful on slower hardware (especially if 30, 40, or 60fps is barely out of reach)
Click to expand...
but it doesnt just drop to 30 form 60. look at my benchmarks I posted earlier. vsync on are off resulted in basically identical performance in Far Cry 2.
 
Apathy said:
well, if you read this thread, you'd probably notice at least 10 replies with people talking about input lag.

seems to vary widely from system to system, regardless of specs.
Click to expand...

I think it's more person to person when it comes to input lag honestly.

I felt my controls were super unresponsive with WoW on my old computer and yet my fiance felt they were lightning fast (at least faster than hers). Yet with Oblivion, I felt they were super quick and she felt the opposite.
 
ya I get input lag from enabling vsync. So I disable it. Or maybe vsync makes me have to set my sent higher. Maybe i will go mess with vsync again.
 
cannondale06 said:
that sounds backwards. the higher over the refresh rate you go the more tearing you should see.
Click to expand...

Not really, When I'm getting over 200fp is cs:s, there's almost no tearing at all. If im getting 60-70, the tearing is quite bad. Same with cod4.
 
Tearing gets worse the higher the FPS gets above the monitors refresh rate, you get more tears per screen on average.

Tearing occurs at all frame rates below, above and even equal to your monitors refresh rate the only thing that stops it is vsync.

I dont use it for 2 very important reasons, the first of which is input lag, input lag for a FPS game makes it near impossible to aim. Secondly frame rate inconsistency in a lot of games means smooth FPS in some parts and in others terrible FPS drop due to uneven map design, vsync then artificially drops your FPS in incriments to achieve some fraction of your refresh rate at which point it all becomes a stuttery jerky mess.

Trying to keep your min frame rate above 60 all the time requires an average frame rate of about 100fps, that means dumping an ass load of the graphics settings down.

Tearing really doesn't bother me that much, its one of those things you see if you're looking for it, and see if you're just admiring the scenery, but if you're actually playing the game and concentraiting on it then you dont see it, or at least I dont.

Example, last night playing mirror's edge, running through the rooftops I dont notice it, when resting in a lift with nothing to do but wait for it to arrive at the next floor, it looks awful.
 
I always keep it on... tearing bugs me terribly, and I haven't noticed anything resembling input lag.

The worst tearing I've ever seen is in Oblivion. Holy hell does that game tear like crazy without VSync enabled!
 
TheBlueChanell said:
I've played competitive CS for years and I've never used v-sync because, for me, it has always added huge amounts of mouse/input lag so I never use it. When in single player games I wish I could use it but the mouse/input lag drives me nuts. I guess I've just become accustomed to tearing and would rather deal with it in order to give myself more fluid input.
Click to expand...

This post makes me sad. I used to be one of those uber aggro vent CS 1.6 players, back in highschool, good times. Had my huge old CRT and radeon 9600. Things were good.

I used to be good at 1.6, but since grabbing my 22" LCD (10-20ms signal lag?) and triple buffering, (30ms lag) (and in vancouver BC seattle is the closest major server area and it clocks in at at least another 40ms, cali is 70), I've become so bad at all those games I used to be so good at. I'll look into this a little further. Do some studies all scientific like.

Vycka said:
I also don't like sacrificing my fps.
Click to expand...

this is a common misconception, toggling Vsync doesn't change your FPS. A teared frame is an incomplete frame (by definition), so if your FPS measuring program was smart enough to measure each frame by how much of it is actualy new data, and chart that as a fraction of a frame, your total FPS wouldn't change at all.

Yllon said:
Computer games often allow vertical synchronization as an option, because it delays the image update until the vertical blanking interval. This can cause lowered frame rates due to latency (the period of the refresh rate at maximum), which might be undesirable in games that require fast response (e.g. first person shooters), although in practise it is impossible to achieve a visible frame rate faster than the refresh rate of the monitor.
Click to expand...

Why does wikipedia always have to use robot speak? I don't get why these people writing for wikipedia always want to explain what is fundamentally a simple concept in words that you'd have to be a comp sci major to understand.

Imagine a buffer exactly 1 frame (as in 1600 X 1200 pixels, or whatever your resolution is) between your graphics cards output, and your monitors input. Your graphics card fills this buffer serially, almost like 1 pixel at a time. So it decides the top left pixel should be red, writes it to that spot on the buffer, then decides the next pixel should be orange writes it to the buffer, and the next one should be yellow, to the buffer, and so on. Every 60th of a second (or whatever your refresh rate is), your monitor turns to your graphics card and says "ok gemmi what you got". Now when it does this chances are that framebuffer isn't yet full, as in, some of the pixels in the bottom right still havn't been given a colour. Your graphics card has two options, it can borrow data from the previous frame (which is in another buffer), and use that to fill it in (this is v-sync off). As basically what your looking at is a small "hole" (for lack of a better word) on your screen, where most of the data is new data, but some of it is 1 frame old. The other option is that your graphics card can order your monitor to wait, so it displays your previous frame. 1/60th of a second later, when the monitor asks again, the graphics card will have finished that frame and can now hand over a frame with 100% new data. This causes no tearing, but a decrease in "FPS" (keeping in mind that in the previous example, as stated above, the displayed frame isnt a complete frame, its a fraction of one. Toggling V-sync doesn't change how many pixels are pumped by the card).

The problem of max fps being higher than the refresh rate causing massive tearing is one of collisions, and good data being thrown away. If you tell source that your max FPS is 200, when your refresh rate is 60, your graphics card will fill this buffer up, and then 1/200th of a second later presume that the data has been taken, and start filling it again. This is what causes tearing. What your getting isn't so much a frame full of old data superimposed on new data, so much as it is a frame of new data superimposed on newer data. The effect is the same: a big horizontal "tare" across the screen.

the problem of input lag isn't that some people have it and some don't, its that some people notice it and some don't. Human reaction time is 200ms. There's suprisingly little spread in that: professional athletes can get as low as 150ms, and fat slow ugly people as high as 250ms. 1/60th of a second is 17ms. Less than a tenth of your reaction time. When you enable V-sync, you add a 2nd buffer to my discription above. When your graphics card is finished filling one buffer, it pushes that one toward the monitor and starts work on the next. When your monitor next asks for a frame, it takes this frame that was previously completed. This is your mouse lag, and its (at worst) 17ms long. Tripple buffering adds another 17ms into the mix, making for 34ms, or ~20% of your reaction time. Some notice that, some don't.

Sorry for the wall of text but I've looked into this pretty extensively :p
 
MrWizard6600 said:
This post makes me sad. I used to be one of those uber aggro vent CS 1.6 players, back in highschool, good times. Had my huge old CRT and radeon 9600. Things were good.

I used to be good at 1.6, but since grabbing my 22" LCD (10-20ms signal lag?) and triple buffering, (30ms lag) (and in vancouver BC seattle is the closest major server area and it clocks in at at least another 40ms, cali is 70), I've become so bad at all those games I used to be so good at. I'll look into this a little further. Do some studies all scientific like.



this is a common misconception, toggling Vsync doesn't change your FPS. A teared frame is an incomplete frame (by definition), so if your FPS measuring program was smart enough to measure each frame by how much of it is actualy new data, and chart that as a fraction of a frame, your total FPS wouldn't change at all.



Why does wikipedia always have to use robot speak? I don't get why these people writing for wikipedia always want to explain what is fundamentally a simple concept in words that you'd have to be a comp sci major to understand.

Imagine a buffer exactly 1 frame (as in 1600 X 1200 pixels, or whatever your resolution is) between your graphics cards output, and your monitors input. Your graphics card fills this buffer serially, almost like 1 pixel at a time. So it decides the top left pixel should be red, writes it to that spot on the buffer, then decides the next pixel should be orange writes it to the buffer, and the next one should be yellow, to the buffer, and so on. Every 60th of a second (or whatever your refresh rate is), your monitor turns to your graphics card and says "ok gemmi what you got". Now when it does this chances are that framebuffer isn't yet full, as in, some of the pixels in the bottom right still havn't been given a colour. Your graphics card has two options, it can borrow data from the previous frame (which is in another buffer), and use that to fill it in (this is v-sync off). As basically what your looking at is a small "hole" (for lack of a better word) on your screen, where most of the data is new data, but some of it is 1 frame old. The other option is that your graphics card can order your monitor to wait, so it displays your previous frame. 1/60th of a second later, when the monitor asks again, the graphics card will have finished that frame and can now hand over a frame with 100% new data. This causes no tearing, but a decrease in "FPS" (keeping in mind that in the previous example, as stated above, the displayed frame isnt a complete frame, its a fraction of one. Toggling V-sync doesn't change how many pixels are pumped by the card).

The problem of max fps being higher than the refresh rate causing massive tearing is one of collisions, and good data being thrown away. If you tell source that your max FPS is 200, when your refresh rate is 60, your graphics card will fill this buffer up, and then 1/200th of a second later presume that the data has been taken, and start filling it again. This is what causes tearing. What your getting isn't so much a frame full of old data superimposed on new data, so much as it is a frame of new data superimposed on newer data. The effect is the same: a big horizontal "tare" across the screen.

the problem of input lag isn't that some people have it and some don't, its that some people notice it and some don't. Human reaction time is 200ms. There's suprisingly little spread in that: professional athletes can get as low as 150ms, and fat slow ugly people as high as 250ms. 1/60th of a second is 17ms. Less than a tenth of your reaction time. When you enable V-sync, you add a 2nd buffer to my discription above. When your graphics card is finished filling one buffer, it pushes that one toward the monitor and starts work on the next. When your monitor next asks for a frame, it takes this frame that was previously completed. This is your mouse lag, and its (at worst) 17ms long. Tripple buffering adds another 17ms into the mix, making for 34ms, or ~20% of your reaction time. Some notice that, some don't.

Sorry for the wall of text but I've looked into this pretty extensively :p
Click to expand...

You forgot the lag that an USB mouse adds too...PS/2 is way more less laggy that USB for input devices ;)
 
You must log in or register to reply here.
Back
Top