![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
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.
GSYNC, FREESYNC, and the Consumer
- Thread starter PHeeNIxx
- Start date
Ha, you're funny.
When my controller vibrates I FEEL it, when the frames on my monitor drop from 120-90 to 60 I can SEE the difference without even having to look at an fps counter
G-sync is still better because it has no "operating window", its only limit is the monitor's max refresh rate. It handles lower framerates far better than Freesync and you can cap your framerate below the max refresh rate if needed to avoid the v-sync input lag.
False and false.
It indeed does have an "operating window."
It handles low frame-rates "better" when Free-Sync has V-sync on.
When G-Sync is below the "operating window" it supposedly duplicates frames which can actually increase latency and introduces judder, depending on the exact situation. It certainly looks better when looking at instruments and numbers but it still detracts from your gaming experience.
Why would duplicating frames increase latency or introduce judder when done properly? It also remains tearing and stutter free, unlike freesync with v-sync off.
And in practice we have, one freesync monitor where you can't have overdrive on while using freesync (wow) and one limited to 40-90hz and that skips frames at 144hz (wower). By comparison the first release of g-sync was working perfectly.
And in practice we have, one freesync monitor where you can't have overdrive on while using freesync (wow) and one limited to 40-90hz and that skips frames at 144hz (wower). By comparison the first release of g-sync was working perfectly.
Last edited:
35-90, not 40-90, and the frame skip at 144 is fixed.
also, here are all the available freesync/gsync monitors - theres a lot m ore than two.
http://www.pcgamer.com/the-g-sync-and-freesync-monitors-available-right-now/
also, here are all the available freesync/gsync monitors - theres a lot m ore than two.
http://www.pcgamer.com/the-g-sync-and-freesync-monitors-available-right-now/
NickJames
Supreme [H]ardness
- Joined
- Apr 28, 2009
- Messages
- 6,698
Oh god not this argument again..
- Joined
- Feb 4, 2003
- Messages
- 2,843
For what it's worth, here's my 2 cents as an Nvidia fanboy'...
I only buy nvidia graphics cards since the crappy ATI X800XL. I much prefer nvidia because I like to hard volt mod my graphics cards and watercool them, and frankly, nvidia hardware is just superior and less flaky in this regard. I also think their drivers are better and in general have better software support (DSR was possible long before it was made official, and SGSSAA is just awesome on old games).
That said, I don't really care about G-SYNC and certainly am not willing to pay for it. Refresh tearing has never bothered me but input lag (from v-sync) most definitely does. It's a real deal breaker for me. G-Sync was theoretically supposed to reduce input latency, but apparently tests show it actually increases it (albeit ever so slightly). Regardless, tearing really doesn't bother me, and certainly not enough to warrant paying a premium to remove it.
As an nvidia fanboy, if you offered me two identically specced monitors, one Freesync and the other G-Sync (but £100 more); I'm going with the Freesync one as it's cheaper.
I only buy nvidia graphics cards since the crappy ATI X800XL. I much prefer nvidia because I like to hard volt mod my graphics cards and watercool them, and frankly, nvidia hardware is just superior and less flaky in this regard. I also think their drivers are better and in general have better software support (DSR was possible long before it was made official, and SGSSAA is just awesome on old games).
That said, I don't really care about G-SYNC and certainly am not willing to pay for it. Refresh tearing has never bothered me but input lag (from v-sync) most definitely does. It's a real deal breaker for me. G-Sync was theoretically supposed to reduce input latency, but apparently tests show it actually increases it (albeit ever so slightly). Regardless, tearing really doesn't bother me, and certainly not enough to warrant paying a premium to remove it.
As an nvidia fanboy, if you offered me two identically specced monitors, one Freesync and the other G-Sync (but £100 more); I'm going with the Freesync one as it's cheaper.
- Joined
- Apr 10, 2003
- Messages
- 23,532
I don't need VSYNC lag. I always despised it. So when I see VSYNC touted as a feature when it's really a bandaid, I stay far, far, away. Now that DX12 is coming I'm hoping that more companies take advantage of technology like SFR to lower frame times.
"According to a Firaxis blog post, the reason for pushing SFR was to cut latency or frame times that are normally associated with AFR rendering. Instead, the developer claims that if you play Beyond Earth with Mantle, latency for a multi-GPU solution is no worse than on a single card."
So why would I intentionally add VSYNC to the game to increase the frame latency when the developer went through all the hassle to remove it? Makes no sense at all to me. I wish some of the Nvidia guys could have experienced what Mantle could do with games like Civilization Beyond Earth. Hopefully under DX12 they can. Then they would realize that tourniquets like VSYNC are crap.
"According to a Firaxis blog post, the reason for pushing SFR was to cut latency or frame times that are normally associated with AFR rendering. Instead, the developer claims that if you play Beyond Earth with Mantle, latency for a multi-GPU solution is no worse than on a single card."
So why would I intentionally add VSYNC to the game to increase the frame latency when the developer went through all the hassle to remove it? Makes no sense at all to me. I wish some of the Nvidia guys could have experienced what Mantle could do with games like Civilization Beyond Earth. Hopefully under DX12 they can. Then they would realize that tourniquets like VSYNC are crap.
What, no, absolutely not. What tests are you refering to? g-sync doesn't add any latency worth mentioning compared to v-sync OFF (it's less than 5ms) unless you're hitting the max refresh rate in which case it's similar to v-sync (and v-sync lag at 144hz isn't huge either). See blurbuster tests.
Ok, 35 but doesn't really change anything, g-sync still has no minimum rate. And I know the frame skip has been fixed but it doesn't matter, they released Freesync but it's not a real alternative to g-sync (for now anyway), that's my point. It doesn't even get close to the first g-sync monitors which all did 0-144hz flawlessly. The fault may not be Freesync itself here mind you, I don't know and don't pretend to know who's at fault. But until this gets sorted Freesync can't compete with g-sync.
Last edited:
Here's everything you need to know about Freesync and its various problems.
http://www.pcper.com/reviews/Displays/AMD-FreeSync-First-Impressions-and-Technical-Discussion
That's funny. That's like buying a sports car and limiting its speed to 60mph because it runs like shit any faster.
Adaptive sync was meant for power saving on a laptop, freesync using it to try and copy Gsync is just not going to work as well. Ever.
http://www.pcper.com/reviews/Displays/AMD-FreeSync-First-Impressions-and-Technical-Discussion
That's funny. That's like buying a sports car and limiting its speed to 60mph because it runs like shit any faster.
Adaptive sync was meant for power saving on a laptop, freesync using it to try and copy Gsync is just not going to work as well. Ever.
nope.
freesync is 9-240hz.
Gsync does not go to 0, once it reaches the bottom of its VRR it frame doubles, increasing lag and judder.
the earlier 144 moniors are all TN, freesync goes to 144 on TN as well.
35 absolutley changes things vserus 40, that extra 5hz could mean the difference in smoothness when frames drop below 40fps.
the monitors with a 90 limit are IPS panels and has to do with overdrive limitations of the scaler, not freesync itself. All TN panels go up to 144hz.
Those scalers will improve by the end of the year seeing 30-144 vrr on IPS panels.
There is currenly only a single advantage to gsync: Superior IPS performance. That will change shortly, giving GSYNC zero advantage.
Last edited:
Gsync has an image buffer and an image processor. Something freesync lacks and is why freesync has so many problems. Freesync will never get this additional hardware so it will always be inferior.
You either don't know anything about Gsync or you are intentionally leaving details out.
Just stop with that garbage. If humans were like that we would of died out long ago as a species. We see in terms of light, not frames. And yes, for those who've been using high refresh rate, Lightboost capable, or GSYNC monitors for years (not so much GSYNC, mostly a year or so), it's sadly pretty easy to tell the difference in frames. Oh, and Lightboost is capable on AMD cards if you did not know, thanks to ToastyX.
For those who think G-SYNC adds input lag, please read this site some more and inform yourself.
http://www.blurbusters.com/gsync/preview2/
Yeah, I think I won't bother with those guys. I would love to see FreeSync have at least some success though because competition is always good, just look at the CPU market the last few years :/ But it's a little tough to compete with nvidia's dedicated hardware module without making your own.
- Joined
- Apr 10, 2003
- Messages
- 23,532
Not being mean or taking sides, but it adds latency to your games when you have an image buffer. Which is the opposite thing that you want in your gaming system. Might as well buy a 20ms monitor then.
Because they are polling the previous frames, to try and anticipate frame drop.
If you believe they are duplicating frames and are guessing when to do it, then you have instances where your frame-rate is fluctuating in and out of the GSync window. They can't interrupt the duplicated frames either, so that is where the additional latency comes from. The same with judder, sure you are refreshing the same frame 3-4 times in a row but that still has a judder to it because it is the same frame being repeated.
Yeah, the first release of GSync was working perfect, if you forget about the 2-3 issues it had...
What? Listening to PCPer, aka Nvidia's mouthpiece, is only something you would suggest Prime...Here's everything you need to know about Freesync and its various problems.
http://www.pcper.com/reviews/Displays/AMD-FreeSync-First-Impressions-and-Technical-Discussion
That's funny. That's like buying a sports car and limiting its speed to 60mph because it runs like shit any faster.
Adaptive sync was meant for power saving on a laptop, freesync using it to try and copy Gsync is just not going to work as well. Ever.
No, you aren't limiting its speed to 60mph, you are running it to the redline and making sure you don't go over it yourself. If you are going to do car comparisons at least make them somewhat comparable... You should try harder next time, I know it isn't your strong suit.
LoL, You seem to be the one that doesn't know what GSync is...
Last edited:
Freesync is not limited to 40Hz-90Hz nor is it exclusive of OD. You need to place the blame/fault where it belongs.
Pieter3dnow
Supreme [H]ardness
- Joined
- Jul 29, 2009
- Messages
- 6,784
More hardware does not mean inferior. It just means extra cost. The implementation is the same it is derived from the same VESA standard.
The solution Nvidia chose is just something they want to make money on, that is all. If they tell you it is superior because it has an image processor and you want to believe that sure.
In the end it is not about Nvidia or AMD it is about what you want as a gamer, in the end the next 5 to 10 years the people with freesync/g-sync monitors will be in the minority anyway. Since you can buy better hardware if you want better performance and the ability to push framerates will have more people buying faster hardware then buying better monitors which handle frames below 60 fps better or the cheapest variation just lower graphics settings
.The were both derived from eDP that was used on laptops.
durquavian
Gawd
- Joined
- Dec 13, 2014
- Messages
- 756
So you are saying you can discern a single frame @120Hz? When every test done to date proves that man can not discern a single frame beyond ~82 (cant remember exact and that number was from combat air pilots) generally most not beyond 60. And in case you aren't getting it: A single frame means that if that one frame was different than the two it is betwixt you would be able to tell with utter certainty what the difference was? Doubt it seeing that the test done showed one color light for 10ms then another directly after for another 10ms and every person saw their combined color because they could not discern the individual colors.
What I love most is every person that claims they CAN SEE the difference doesn't understand what that really means. I have no doubt they can feel the difference as the sub-conscious can in fact see things we cant directly see consciously. No one is saying that 120hz is the same as 60hz or that you cant see a difference. More aptly we are saying you aren't seeing it exactly. Remember the spinning wheel of a car on the road or did you not learn that in school?
I love how people quote "A human cannot discern X amount of fps from Y amount of fps" when there is absolutely no context from even the meaning of fps.
Like for example, when one says "When every test done to date proves that man can not discern a single frame beyond ~X, generally most not beyond Y", what tests are we talking about? Are we talking about fps when playing a movie when the frames are completely evenly spaced out with near zero variance? Or are we talking about real time graphic rendering? If it is the latter, are we talking about just an AVERAGE fps, or are we talking about rendering things that is so easy for the renderer in question that we must cap the frametime of each frame to a certain figure?
Have that information been provided in any of 'tests' you have quoted?
Because to me, the conclusion drawn from those claims are of dubious values at BEST, unreliable at worst, because real time rendering (such as in games) is dramatically different from playing a movie, and the rendering will also heavily depend on the load of the render.
I can think of one reason why 60fps gaming can feel different to 120 fps gaming: statistical frametime deviation. If that deviation is 0, then I'd partially agree there may be no difference. If there is any deviation, then it is more likely to be noticed on 60fps, because there is a greater chance of time between two frames to increase long enough to be detected by the human eye (such as, momentarily dip into 40 fps if the frametime was translated into fps), whereas 120fps it is much less likely to do so (if it momentarily dips into 80fps region, would one notice)?
Without what the tests actually consists of, it is as much scientific argument value as arguing that red cars go faster because it appears to move faster.
If one wants to use science to argue against something that may not be quantified by science (IE individual human eye and image intepretation by the brain), at least have sound science.
If one does not understand why this is important, then the fps argument will always end up at the same place: nowhere.
Like for example, when one says "When every test done to date proves that man can not discern a single frame beyond ~X, generally most not beyond Y", what tests are we talking about? Are we talking about fps when playing a movie when the frames are completely evenly spaced out with near zero variance? Or are we talking about real time graphic rendering? If it is the latter, are we talking about just an AVERAGE fps, or are we talking about rendering things that is so easy for the renderer in question that we must cap the frametime of each frame to a certain figure?
Have that information been provided in any of 'tests' you have quoted?
Because to me, the conclusion drawn from those claims are of dubious values at BEST, unreliable at worst, because real time rendering (such as in games) is dramatically different from playing a movie, and the rendering will also heavily depend on the load of the render.
I can think of one reason why 60fps gaming can feel different to 120 fps gaming: statistical frametime deviation. If that deviation is 0, then I'd partially agree there may be no difference. If there is any deviation, then it is more likely to be noticed on 60fps, because there is a greater chance of time between two frames to increase long enough to be detected by the human eye (such as, momentarily dip into 40 fps if the frametime was translated into fps), whereas 120fps it is much less likely to do so (if it momentarily dips into 80fps region, would one notice)?
Without what the tests actually consists of, it is as much scientific argument value as arguing that red cars go faster because it appears to move faster.
If one wants to use science to argue against something that may not be quantified by science (IE individual human eye and image intepretation by the brain), at least have sound science.
If one does not understand why this is important, then the fps argument will always end up at the same place: nowhere.
Wait! You're using fighter pilots instead of gamers? You obviously need to raise your standards!
Using software to try and emulate a hardware solution will always have drawbacks. Which is what we are seeing with Freesync. It has too many problems due to a lack of hardware and quality control. Don't be surprised if AMD abandons it like Mantle, Brook+, etc.
- Joined
- Feb 4, 2003
- Messages
- 2,843
Thankyou for providing a link with proof that G-Sync increases input lag. This was actually the article I'd read that led me to believe this.
Here's how it breaks down, averaging the 2 runs from each game they tested (lower is better):
BF4:
V-sync off: 73
G-sync: 75.5
Crysis 3:
V-sync off: 54
G-sync: 55.5
Counterstrike:
V-sync off: 24.5
G-sync: 33.83
The counterstrike difference was due to a bug, and was corrected when forced to cap at 120fps.
So yeah, it's laggier. Fact.
The difference wouldn't be perceivable mind you, but technically it is there.
Last edited:
The difference is so minor that I doubt it can be correctly attributed to G-Sync itself. It sounds like it's all within experimental error.
the difference in input lag in BF4 and Crysis 3 is barely above the experimental error (which should be around +/- 1ms, the maximum time variance possible for a 1ms/frame camera)
the difference in input lag in BF4 and Crysis 3 is barely above the experimental error (which should be around +/- 1ms, the maximum time variance possible for a 1ms/frame camera)
So you are just going to completely ignore what the article said?
" I found no perceptible input lag disadvantage to G-SYNC relative to VSYNC OFF"
At this point I think people are just baiting with all the misinformation they are posting. If this is what is needed to sell a product, it's no wonder AMD is on the verge of bankruptcy.
- Joined
- Feb 4, 2003
- Messages
- 2,843
Completely agree!
My original interest regarding input lag and G-sync (before this thread existed) was the spoken potential to reduce input lag to theoretically '0'.
With regular monitors, regardless of refreshrate, after you make your mouse click, you obviously have to wait for the next game frame to be drawn in memory, and then wait for the next time the monitor is ready to show the next frame. With G-Sync, theoretically that frame could be drawn immediately and thus your input is more closely reflected in realtime.
The difference would be still a few milliseconds, but it gave me pause for thought that maybe the sheer directness of input would possibly make a noticable difference to twitch shooters to those who crave every last frame. This is what made me most personally excited in G-Sync, but real world testing shows no benefit in this regard.
So based on this data, G-Sync does not provide a gaming advantage over v-sync off in non g-sync mode.
- Joined
- Feb 4, 2003
- Messages
- 2,843
...and here's the final paragraph of my post you omitted in your quote:
The difference wouldn't be perceivable mind you, but technically it is there.
- Joined
- Apr 10, 2003
- Messages
- 23,532
Lots of people say that there is no perceptible latency difference in gaming between an IPS monitor, TV, or 1ms TN 144Hz monitor. Guess those are the same that believe that stacking frames in an image buffer isn't adding to the latency. Triple Buffering and VSYNC never added latency to games.
Yeah right....
Yeah right....
Pieter3dnow
Supreme [H]ardness
- Joined
- Jul 29, 2009
- Messages
- 6,784
The change is bigger that Nvidia will drop G-sync due to the costs involved and the benefits would only work for OEM if they keep on selling monitors with G-Sync. Unless Nvidia is paying OEM rather well for their "support" but if it does not sell enough monitors it will crater ....
AMD on the other hand does not involve itself just asks that the hardware meets certain parameters, there should be plenty of monitors out there that support adaptive sync which is the same as freesync but just not verified by AMD.
With all the problems Freesync has been having (including the Asus recall). Monitor makers may just shy away from it. Not to mention AMD's historically low market share which severely limits how many of these monitors will be sold.
Maybe Freesync 2.0 will offer a hardware solution if AMD decides to stay with it. Until then, it's just not up to par.
- Joined
- Feb 4, 2003
- Messages
- 2,843
Seems to me adaptive refreshrate is one of those technologies that'll be taken for granted as standard in the future, but right now needs to be carefully implemented.
The change needed is for all LCD panels produced to be capable of variable refresh. Right now, most aren't and the world will have to wait for LG, AUO, etc to deplete inventory of their older panels before the new tech can become ubiquitous.
I don't see much of a future for G-Sync, but I'm sure Nvidia would happily agree, and have released it anyway for those who care and say "Here it is, right now, but you pay for it". I applaud nvidia for this approach, it's how things improve; it made AMD respond with Freesync and will ultimately result in all panels fundamentally using adaptive refresh rates.
AMD may look like the good guys right now, but really it's nvidia who'll later be able to say "You're welcome".
The change needed is for all LCD panels produced to be capable of variable refresh. Right now, most aren't and the world will have to wait for LG, AUO, etc to deplete inventory of their older panels before the new tech can become ubiquitous.
I don't see much of a future for G-Sync, but I'm sure Nvidia would happily agree, and have released it anyway for those who care and say "Here it is, right now, but you pay for it". I applaud nvidia for this approach, it's how things improve; it made AMD respond with Freesync and will ultimately result in all panels fundamentally using adaptive refresh rates.
AMD may look like the good guys right now, but really it's nvidia who'll later be able to say "You're welcome".
Geforcepat
[H]ard|Gawd
- Joined
- Jun 2, 2012
- Messages
- 1,338
AMD may look like the good guys right now, but really it's nvidia who'll later be able to say "You're welcome".
Uh-huh..
durquavian
Gawd
- Joined
- Dec 13, 2014
- Messages
- 756
So you think Adaptive sync is only for AMD so therefore it will be dropped because of market share? So short sighted and .... Intel has great need for the tech using their own iGPU, would be great boon. Not to mention the eventual move to TVs as well doing away with 3:2 pulldown, again nothing to do with AMD. Maybe refrain from the hate and view the bigger picture Adaptive sync can be applied. Areas G-sync will never be.