NX Gamer Discusses Game Tech *spin* [2015 - 2017]

Status
Not open for further replies.
AlNets said:
bandwidth for the analog sticks? There must be some practical limit to the poll rate.
Click to expand...
Agreed. I wonder if they just send over resistance values of the buttons/triggers and let the driver sort it out.

I assume there is back and forth on bandwidth due to rumble it's no longer one directional.

Triggers are analog but aren't the regular buttons as well? IIRC was playing GT2 with analog buttons.

Then there's touch screen for PS and the trigger rumble for XBO.

There's also sound/headset/speaker and possible chat pad accessories to take account for.

Could be quite a bit of bandwidth required at any moments time.
 
iroboto said:
I respectfully disagree :) when people ask between wired and wireless I assume they are trying to get an edge or at least limit disadvantage. if measured correctly there should be a consistent difference between the two, even if it's marginal. That consistency should be proven through a variety of titles and it's entirely possible that different controllers have different latencies as well.

For players looking to get the edge in competitive MP games this matters a lot.

Latency per game won't matter as much since it's consistent for all players.

That being said, the point isn't lost, but discussing frame latency (n-1,n-2,n-3) seems to fit the discussion when they compare graphical tech for the games. So as they are charting resolution and graphical details it's a good time to also mention frame buffering comparisons per title. It seems less fitting for controller discussion.
Click to expand...
In my first post I missed several things about this video, notably some improvements coming from the wired pads in some games.

The results for doom on PS4 are the same (within margin of errors etc.) using wireless and USB pads: mean of 104ms.

For XB1 the 150ms input lag is measured using wireless pad. Those numbers are measured in the same room so the difference between PS4 and XB1 (~50Ms) is not caused by framerate dips.

PC players with Doom have a better input lag, but not that much compared to PS4: 88ms / 82ms (Mouse, USB pad).

For BF1, there is a minor improvement using the USB pad for all 3 machines: NX game gives only the XB1 number: 112ms instead of 125ms.

So apparently the USB pads slightly improves input latency but only in some games, not all. Too bad he didn't give us all his numbers. Maybe he'll do it later in his website where he can write more information.

Using the OS, Wireless has the same latency as USB on both machines (within margin of error of 2-4ms). Interestingly latency on PC calibration tool is less consistent than on consoles.

http://i.imgur.com/K0o3vCV.png
 
Last edited:
NX doesn't always do the most concise or understandable videos and this is another case.

First off how does he remove out display lag from his calculations?

Second what EXACTLY is he measuring? He says different (game) actions I guess...but it's not really explained. Is he measuring until the first pixel moves or something (lets say a melee is a whole animation, that takes time)? I dont even know what I'm trying to say...

This stuff has always been pretty confusing for me. And how does frame rate play into it as well? If you are measuring a 30 FPS game, doesn't that limit the granularity with which you can possibly measure input lag?
 
Rangers said:
NX doesn't always do the most concise or understandable videos and this is another case.

First off how does he remove out display lag from his calculations?

Second what EXACTLY is he measuring? He says different (game) actions I guess...but it's not really explained. Is he measuring until the first pixel moves or something (lets say a melee is a whole animation, that takes time)? I dont even know what I'm trying to say...

This stuff has always been pretty confusing for me. And how does frame rate play into it as well? If you are measuring a 30 FPS game, doesn't that limit the granularity with which you can possibly measure input lag?
Click to expand...

This is my main concern with the video, I turned off after a while as it was clear he was not going to divulge his methodology which to makes the results pretty irrelevant for me as I don't have faith in them unfortunately, as it is an interesting subject.

Chris from BattleNonsense does similar videos on pc and his setup is probably what nx replicated

https://www.youtube.com/user/xFPxAUTh0r1ty

He modified his mouse to have an led that lights when he presses the button. He films with a high speed camera and counts the frames from led to movement 20 times and averages. He also does different movement types such as strafe, jump and shoot as these have different packet importance from the server and have different lag values.

Note these tests are usually server lag not local but he has done gsync and freesync and other rendering settings, well worth a watch and very well explained.
 
AlNets said:
There must be some practical limit to the poll rate.
Click to expand...
Its 16 bit not sure the frequency
i think the default xinput rate is the sticks are polled at 125 Hz, but the face and shoulder buttons are 20 Hz.
also found this
"USB 2.0 controllers poll devices at a maximum of 8kHz, so it is utterly pointless to have a polling period smaller than 0.125ms."

turkey said:
He modified his mouse to have an led that lights when he presses the button.
Click to expand...
There's a program called Mouse Rate Checker
http://www.softpedia.com/get/System/System-Miscellaneous/Mouse-Rate-Checker.shtml
 
Last edited:
Injustice 2 is the last game being compared by NX Gamer.

Once again he finds that the controller is a bit less responsive on XB1. Measured using the wireless pads with Flash (same stage, same punch) :

- PS4: 95 ms
- XB1: 116 ms


Rangers said:
NX doesn't always do the most concise or understandable videos and this is another case.

First off how does he remove out display lag from his calculations?
Click to expand...
Elsewhere he explains (in the description) that it's overall input latency, from pad input to TV change :

which is the time from an action being requested and it happening/being processed on-screen.
Click to expand...
 
I would be curious to see the controller response for Diablo 3 or even Destiny.
Wonder how the latency compares between fighting games (on the same platform) too.
 
How does this NXGamer guy measure input lag? His video never says. I'm also not sure why he wouldn't just pick specific movements in each game (strafe left, fire gun) and repeat them numerous times on each platform, wired and unwired until he gets the delay for the action to appear on screen.

Things I'd want checked are if there are different deadzone differences on the analog sticks between consoles, or whether the throw distance on the buttons are the same. I have no idea how he's testing, so no way to know if those things would impact the results. That said, deadzone and buttons are real lag. If you have a shorter button press, or less deadzone, your controller will be more responsive. Analog stick deadzone can be tweaked in some games, so it'd be interesting to see if he could remove some of the lag in Battlefield 1, for instance. Buttons are just on/off with a throw distance.

Found the full article: http://nxgamer.com/Technical-Analysis/how-we-test-input-latency

The way they measure should be very accurate. I'd be curious to see head-to-head button presses vs head-to-head analog stick movement tests for the same game on each platform. Then to see if in-game settings can affect stick latency.

EIther way, looks like Xbox One has some added latency. I don't think triple buffering would explain all of it. It is an issue that really should be sorted out. I'm not a fan of the modern trend in gaming where controllers are high latency, game engines are sometimes high latency, and most displays are high latency. It all adds up.
 
Last edited:
so....did he abandon the channel? I guess he overstimated his proposal value, which is amazing enough, maybe not for 100000+ subscribers when you have digital foundry or VGTech -super high quality, but also not many subscribers-, and then there is this El analista de bits channel https://www.youtube.com/user/ElAnalistaDeBits/videos which has more than 200.000 subscribers..

The later's approach is appealing because his comparisons are of high quality, plus he remains silent, it's just both games being compared, maybe that's what makes him different, and succesful, or maybe because there are more channels offering that in english, and the guy speaks spanish, so he has potential for more growth for being more exclusive in a given language, but anyways..
 
AlNets said:
Maybe he could join DF if he's still passionate about the actual work.
Click to expand...

NXG would be an asset to their videos because he comes across as enthusiastic and energetic which is something I think DF is lacking. Maybe DF are enthusiastic and energetic but they don't sound it; most of their videos have the tone of the average corporate motivational video.

It's not enough to be informative, you have to engage the viewer and keep their attention and that requires specific communications skills.
 
Status
Not open for further replies.
Back
Top