FreeSync vs G-Sync

[Megahurt]

I'm not sure how you came to that conclusion for this video. Their test was done with a high speed camera using visual response time to measure the total input lag. pixel response time will add to that lag (and possible ghosting if changing from black to white, but irrelevant here).

Response time is the time a pixel takes to go from 1 color to another, they start changing at the same time, one just takes more time to achieve the desired result. Think of a very fat man versus an olympic runner in the 100m, both will start at the same time, they just won't finish at the same time.

[Briggsy]

Response time is the time a pixel takes to go from 1 color to another, they start changing at the same time, one just takes more time to achieve the desired result. Think of a very fat man versus an olympic runner in the 100m, both will start at the same time, they just won't finish at the same time.

 

and thereby adding to the total visual input lag. I don't know what point you are trying to make.

 

*edit* Linus is not using an instrument that tests when the pixels start changing, but when he visually sees them changing.

[Megahurt]

and thereby adding to the total visual input lag. I don't know what point you are trying to make.

You're gonna see the change before it reaches full color, especially that most of the change happens in the first half, so the number is meaningless. Besides even if it was meaningful, it's so small compared to actual input lag that the type of panel you use doesn't really matter, it's smaller than even the margin of error.

[Z3R0B4NG]

Yeah it was fairly recent.  Whenever they added window mode GSYNC.

 

ok... updated to the 353.60 just now, hope this one won't crash all the time like 353.06 did for me...

 

i can see the changes now...

 

 

....

so you are telling me that with V-Sync ON... on top of G-Sync there is more input lag because of V-Sync (frame buffer???), at lower than maximum refresh rates?

...

that is just retarded...

 

is the old driver version where they didn't do it like that also hit by this delay? (needs testing)  

 

 

conclusion... for now:

 

G-Sync ON, V-Sync OFF and i'll punch 

 

"GameTime.MaxVariableFps 140"

 

into the user.cfg of Battlefield 3, 4 and Hardline, then i should have minimum input lag at all times.... right?

 

and i guess other games with no custom FPS cap via config/console will just be screwed.

 

i hope Nvidia is going to work a bit more on this, V-Sync should not affect G-Sync at all when below max refresh rate and i was under the impression that it didn't until today.

[Shadow-R]

Terve!

For me the real question here is: does it matter?

I see the advantage of free- g-sync, but 10ms input lag, really?

I don't think anyone that is not a professional gamer does not notice 10ms or 30ms of difference.

The best example is: the wink of the eyelid is the fastest thing a human can do (according to some studied guy) and this takes about 300ms.

I'd like someone to tell me that he actually notices 30ms of difference.

Sayonara!

[Z3R0B4NG]

well... no of course you will not notice the difference while playing.

just as you will not notice that extra 200MHz that you squeezed out of your CPU by overclocking

or like you won't notice the difference between a 20 and a 50ms PING on the Server you just joined.

 

but all together it sums up to quite a bit of difference, in particular in competitive games.

[peanutsrevenge]

Good video, thanks guys. Amusing and entertaining (with a side order of DAFUQ?!!?!?!)

 

Regarding the vLog style, I like it, scratches my 'how/why does that happen' itch, however if I was new to the channel (having been linked by a friend maybe because it's something interesting (as I'm about to do to my millisecond or die, Twitch gamer aquintances, I think I'd either get bored really quick or think it was some skoddy bunch of newbs trying to make YT videos (which you are, but that's beside the point ).

Maybe putting the vLog style bits after the sponsor spot (which might help to get more sponsor spot views), though it would make the filming and editing a bit weirder and time consuming.
My Two Cents.

 

And yes, let the berk have a quad copter follow him. Given how much of a trolly bastage he is, perhaps it should be mandatory, likely to get some highly amusing/scary/vulgar/incriminating footage

[wdornbach]

Taran would have done this video better.

[decoy11]

I'm reading about how Nvidia's control panel has a settings called "Maximum pre-rendered frames" can affect the input latency. Any word on what setting Linus used for his test? Also the default is 3 but can go down to 1 which is the best for input latency.

[sgloux3470]

ok... updated to the 353.60 just now, hope this one won't crash all the time like 353.06 did for me...

 

i can see the changes now...

 

 

....

so you are telling me that with V-Sync ON... on top of G-Sync there is more input lag because of V-Sync (frame buffer???), at lower than maximum refresh rates?

...

that is just retarded...

 

is the old driver version where they didn't do it like that also hit by this delay? (needs testing)  

 

 

conclusion... for now:

 

G-Sync ON, V-Sync OFF and i'll punch 

 

"GameTime.MaxVariableFps 140"

 

into the user.cfg of Battlefield 3, 4 and Hardline, then i should have minimum input lag at all times.... right?

 

and i guess other games with no custom FPS cap via config/console will just be screwed.

 

i hope Nvidia is going to work a bit more on this, V-Sync should not affect G-Sync at all when below max refresh rate and i was under the impression that it didn't until today.

 

Yeah what I do is I simply push "enable GSYNC" and let it do whatever it has to do.  But instead of saying GSYNC it does default to vsync now.  It's the same functionality as far as I can tell though.

[Nurmi]

I'd guess that the screens actually have "low" & "high" parts which keep on switching and cause that strange delay at the middle frames.

Secondly It'd be the best screen of screens to have G-Sync which turns off to plain FreeSync at higher than refresh rate numbers.

Thirdly, first reasoning would be the reason why second one isin't there as it'd cause some of that delay, ..or no one just thought it?

 

Thank you, *bows left & right*

[MAXXHEW]

Even with the best software and hardware running at optimum levels... the millisecond input lag I could save in game would take like 50 years to replace the fifteen and a half minutes I spent watching this video.

[DXMember]

@nicklmg can't really see the individual fan blades spinning..

please upload a video or two, that would be interesting to watch!

[kevin11189]

It would have been cool to see the values for each when G-Sync and FreeSync are both off. I know the input lags are reduced with variable refresh, but by how much would have been nice to know, or at least I'm curious about it.

 

Also, when talking about how "all over the place" something is, calculating some standard deviation values would be interesting. The small number of data sets might not have made them incredibly meaningful, but I think your viewers are keen enough to interpret those values. I know I would appreciate that.

 

Just a though. Thanks.

[Godlygamer23]

I think you mis-spelled the title of the video we want: Firepoles vs Firepoles

The fire pole is not happening.

[mr moose]

Terve!

For me the real question here is: does it matter?

I see the advantage of free- g-sync, but 10ms input lag, really?

I don't think anyone that is not a professional gamer does not notice 10ms or 30ms of difference.

The best example is: the wink of the eyelid is the fastest thing a human can do (according to some studied guy) and this takes about 300ms.

I'd like someone to tell me that he actually notices 30ms of difference.

Sayonara!

 

I don't know about how fast we blink but reaction time from visual stimulus makes a mockery of most peoples arguments regarding lag and FPS etc.

The human visual processing system as a whole is an adaptive one that creates information to fill the gaps from what is actually quite a poor quality visual input.

[Juanse]

FreeSync

NCIX: http://bit.ly/1HAMTHN

Amazon: http://geni.us/3qcT

 

G-Sync

NCIX: http://bit.ly/1Rtb1qV

Amazon: http://geni.us/2KaJ

 

Is there a final answer in the FreeSync vs G-Sync debate? We set out to find it...

 

I think this video style is more awesome than normal reviews

[Nonco]

The fire pole is not happening.

 

RIP DREAMS ;-;

[AMICLG]

The MCU of the makey makey has about a 3m delay from when it wakes up till the time in runs the code, which the code is another few milliseconds, then the code/api on the computer so this isn't very accurate. If you took apart a mouse and put a high speed linear actuator on the button you could see the button engages and great a better representation of what actually happens when you game with a mouse and adaptive sync.

[shananigans89]

i like this style of video guys. i say keep it up!

[mr moose]

The MCU of the makey makey has about a 3m delay from when it wakes up till the time in runs the code, which the code is another few milliseconds, then the code/api on the computer so this isn't very accurate. If you took apart a mouse and put a high speed linear actuator on the button you could see the button engages and great a better representation of what actually happens when you game with a mouse and adaptive sync.

 

The makey makey was used for both Gsync and Freesync so any inherent delay will be the same for both thus making it moot in the results.  Also the lag in the system as a whole is well beyond human perception so using a mouse or different types of mice will not change anything for the end user.

[ars2012]

So typing "fps_max 0" in the cs go console is too hard? 

[skogsvilden]

Didn't Wendell mention that different connection types have different latency on the same monitor when he did some panel testing? Was that taken into consideration when doing this test?

[mr moose]

Didn't Wendell mention that different connection types have different latency on the same monitor when he did some panel testing? Was that taken into consideration when doing this test?

Yes they can, but given they are testing technologies that can't even co exist in the same monitor for now I feel this is the best we can expect. 

[terror]

Nvm I got confused as to which monitor was for which