I refuse to buy DENUVO games. It's too problematic and cannot be circunvented easely to prevent the game to give headaches. So I just will not buy a game that have DENUVO on PC. I'll preffer the console version for DENUVO games and with a large discount because these games DO NOT deserve money as they cause inconvenience and bad service.

View this comment - View article - View full comments

waiting for HDR.....

View this comment - View article - View full comments

Hum... the only thing that is keeping me away is: is nVidia already implemented VRR on wayland driver? This is taking a lot of time.

View this comment - View article - View full comments

finaly

View this comment - View article - View full comments

then there are the things I want in Linux for 2023:

- HDR
- VRR
- low latency focused development for everything (speed is everything)
- Proton being faster than native (because windows)
- the death of X.org
- better support for popular gamepads

What will be done in 2023:

- probably nothing because Linux isn't developed to atend the end user.

View this comment - View article - View full comments

As no gamepad reviewer measures actual latency and just limit itself to report one's own perception of the product I must say that this misleading type of non cientific review deceaved me greatly. I bought a SN30 pro+ 2 years ago. While I used at the time (and now) the awesome DS4 as main gamepad for everything that can be used with, I wanted a gamepad that was more friendly to remember which mode was connected to. The impact in the high latency was imediate so I went measure it and found astounishing bad results...

For a good gamepad review you need to measure at least these behaviors:

- Input delay (it's the polling rate). This measures the time between inputs. It's very important this does not have jitter as it'll ruin experience. This can't be done on consoles but as every gamepad nowadays can be connected to PC, this must be measured. (and no reviewer do this)
- overhall input lag (the total one). This measures time from button press to action on screen. This is not absolute like input delay as it depends on the user setup. So a comparison with something known like a largely used gamepad like DS4 or XBOX Controller must be used as reference.

My results:

Input delay: SN30 pro+ have horrible input delay. Over bt it have a lot of dummy packages around 3 to 5 ms and an actual data pacage around 33ms (this makes this gamepad updates it's input around 30 Hz. This is very low). Over USB all data updates twice as fast with dummy packages around 1 to 2ms and a real data package around 15ms (the actual update rate was calculed to be around 66 Hz... my God...). The update pattern is 2 dummy package and 1 real data package. I reported this many times to 8bitdo for them to just let my support ticket expire every single time without answer and never fix it. Some values for reference: XBOX Controller have a polling rate of 125 Hz but skips one entry every second (I don't know why but this shouldn't really be a problem). The Switch joycons also updates at 125 Hz but no skips. DS4 is the best one with a polling rate of 250 Hz (this is really amazing)

Input lag: I measured input lag for SN30 pro+ on Nintendo Switch in compasison with the Joycons and the DS4 using mission control to enable it on the console.

DS4 operates downclocked on Switch to save bt bandwith (it's a deliberated choice from Mission Control devs) so the actual measured diference using a high speed camera between DS4 and Joycons are within measure error (5 ms +-) (any cell phone can shoot 120fps nowadays. Mine can shoot 240 fps and it was used to measure time).

SN30 pro+ was astounishing 50 ms higher than the two when over bt. This is really bad results as it falls under human perception levels of lag.

I cannot in any sense or for any use recomend any 8bitdo controller. The "best" gamepad they had at marketing and reviews level is just a very bad, crap level of a product.

You SHOULD NOT buy anything from 8bitdo.

I also strongly recomend staying away from any 3rd party gamepad that don't have at least this type of data made public. Stick to the ones that have like XBOX and DS4/Dualsense

Edit: I forgot to mention that some windows games just give up on this gamepad as long as 5 minutes after beying started. Easy example: Trails in the sky trilogy.

View this comment - View article - View full comments

Quoting: slaapliedje

Quoting: tamodolo

Quoting: fagnerln

Quoting: tamodoloFedora 36 is the first Linux release that actually solves input delay to be on pair with windows. They did this unintentionaly as it was an issue related to touchpads. I am extremely happy by this fix as FINALY FINALY Linux is starting to move forward instead of just doing sidesteps and go anywhere.

I'm following this release and didn't read anything about this, looks like gnome improved the way that it handles input to be on par to the FPS and not to the refresh rate.

What fedora 36 have to make it different?

You need to search in the complete changelog. A friend pointed they did that after reading it and said probably it was it when I mentioned that DS4 was working on pair with windows now.

Is this the key thing? "DS4 was working on pair with windows now"? So your problem was that the input lag existed when using a Dual Shock 4 controller? I think I've only used it once or twice on mine. I'm guessing this is without a wired connection?

Bluetooth is... still pretty shit on Linux. Though they have miraculously tweaked / fixed it on the Steam Deck.

Yeah. Input lag and consistency with gamepads on Linux is very very bad. I'm waiting SteamOS 3 to be released for PC to test it.

View this comment - View article - View full comments

Quoting: fagnerln

Quoting: tamodoloFedora 36 is the first Linux release that actually solves input delay to be on pair with windows. They did this unintentionaly as it was an issue related to touchpads. I am extremely happy by this fix as FINALY FINALY Linux is starting to move forward instead of just doing sidesteps and go anywhere.

I'm following this release and didn't read anything about this, looks like gnome improved the way that it handles input to be on par to the FPS and not to the refresh rate.

What fedora 36 have to make it different?

It's a change on fedora's libinput. It isn't mentioned on keynotes.

View this comment - View article - View full comments

Quoting: fagnerln

Quoting: tamodoloFedora 36 is the first Linux release that actually solves input delay to be on pair with windows. They did this unintentionaly as it was an issue related to touchpads. I am extremely happy by this fix as FINALY FINALY Linux is starting to move forward instead of just doing sidesteps and go anywhere.

I'm following this release and didn't read anything about this, looks like gnome improved the way that it handles input to be on par to the FPS and not to the refresh rate.

What fedora 36 have to make it different?

You need to search in the complete changelog. A friend pointed they did that after reading it and said probably it was it when I mentioned that DS4 was working on pair with windows now.

View this comment - View article - View full comments

Quoting: TheSHEEEP

Quoting: tamodoloFedora 36 is the first Linux release that actually solves input delay to be on pair with windows. They did this unintentionaly as it was an issue related to touchpads. I am extremely happy by this fix as FINALY FINALY Linux is starting to move forward instead of just doing sidesteps and go anywhere.

Sounds more like a bug to me that you had there.
I've been gaming on Linux since years, including very fast-paced games (FPS) on both Ubuntu- & Arch-based distros and I've yet to notice any input lag (excluding lag that is caused by the CPU being overburdened).

I'm very sensitive to inputlag because my main system to play games is windows and DS4 have very little inputlag on it. When testing it on any Linux de first thing evident is that DS4 isn't as responsive. Fedora 36 bring parity to that.

Also, everytime I post about this problem I receave a lot of comments that nobudy have this problem or it's a bug in my side. I see this as the big problem Linux have because it needs someone big to make average user complams to reach devs. Otherwise it'll be ignored... This "fix" was a side effect, but the merit is still there.

View this comment - View article - View full comments