It's Time To Revamp Our PC Testing - Alex vs PresentMon 2.0
Vložit
- čas přidán 11. 04. 2024
- ► Watch the FULL video here: • DF Direct Weekly #157:...
► Support us on Patreon! bit.ly/3jEGjvx
► Digital Foundry CZcams: / digitalfoundry
► Digital Foundry Merch: store.digitalfoundry.net
► Digital Foundry at Eurogamer: eurogamer.net/digitalfoundry
► Follow on Twitter: / digitalfoundry
Fantastic stuff! Now sort of want a video where you revisit a bunch of the more baffling PC perf issues you've hit over the years :)
Developer’s should be getting even more nervous when DF gets new tech to pull apart their games.
Pushing tech forward. I like this
🗣 SPECIAL K MENTIONED 🗣
I love that program, Kal is an absolute legend.
Thanks, John. Now I have to watch The Running Man again!
CPU Busy and GPU busy are the CPU time and GPU time, right?
These have been in the Oculus debug tool for 6-7 years and are super useful, because a stutter in VR is just unacceptable.
it's a more specific metric developed by intel engineers, look up gamers nexus presentmon
This is awsome
Alex, just to let you know. RTSS has an overlay template that shows every Reflex latency marker in a neat graph format. It is highly useful for PC performance analysis. You can also customize it to your heart's content.
RTSS overhead is worse than special K as it is.
Presentmon and special K are both better.
@@flingymingy What do you mean by "overhead" exactly?
@@JoshNautes Using the term 'overhead' here, they likely mean the performance loss from just running the overlay or system monitor itself. The term is used in many ways throughout tech, business, and elsewhere but usually means the same thing. Extra performance, money, or power lost by adding or using different things in your processes.
@@JoshNautes Overall I find that I get more issues when using RTSS. First, when using a Quest VR headset on a PC, it sometimes introduces issues with how smoothly the frames are displayed (which is much more noticeable in VR). Also when playing Minecraft, RTSS sometimes causes issues (like if you use the optimization mod sodium, it would purposely error out and tell the user to disable RTSS when launching the game due to performance issues caused by it). For the Minecraft scenario, it should be fine if running the latest version of RTSS now, but Presentmon and Special K are more modern and cause less issues (like @flingymingy said).
The main thing is presentmon has the new gpu busy and cpu busy metrics that rtss doesn't have. At least I don't see them in there now
The mentioned video: czcams.com/video/twEERkUyAXE/video.html
Has Alex had dental surgery?
you guys should really learn how to code and stuff. I think that would massively help you to improve your testing process and interpretation of results.
That's actually correct. Any game engine has unparalleled performance monitoring information output in debug mode.
Oh... I already did this by changing CPU % to graph on MSi afterburner... you were GUESSING THE WHOLE TIME?!
CPU busy is different from CPU Usage %. CPU Busy measure how many miliseconds the CPU spent on a given frame, not how much of the CPU is currently in use.
Are you polling CPU load in the real time?
@@JoshNautes Ah ok i like this answer thanks for the correction
@@yarost12 I think i am, its at the default 1000ms, if its the hardware polling we're talking about. I mistook busy with the utilization at the moment. I set every CPU's text % to graph %, so i get a more accurate reading on which core maxes out, or how the game is utilizing each core. I hope that answers the question correctly
@@JoshNautesCPU busy is good but i like CPU usage per thread a lot.
Just make sure no hot woman are involved, wouldn't want someone getting upset
Your testing is kind of outdated and not all that flattering I think.
😂😂😂😂
🤓