Adding my LatencyMon into this mix.
ASRock Taichi X299 with Aorus 1080 Ti Extreme 11GB, Watercooled Intel X7740 (all-cores 5GHz overclock currently turned off at the moment), Samsung 960pro M.2 SSD and 16GB RAM, and gigabit cable Internet.
- 1.5 year old Windows 10 install with all its crud
- Default drivers
- Default autoupdate
- Latest NVIDIA driver
- Oculus Rift running
- Three Oculus sensors on USB 3.0 ports (75 megabyte/sec firehose each)
- Dozens of Chrome windows
- Dozen of system tray apps
- Visual Studio 2017 running
- Mostly unoptimized
Doesn't look too bad. VR games run deliciously smooth, and my FPS games appear to be 'normal' (no more stuttery than I've seen on other similar high end systems). Probably could get better with a fresher Windows 10 install and some driver optimizations, and not having so many damn apps running at the same time. So many things to do, so much to play, so much work to do, and I have to operate Blur Busters -- so I've ended up not having enough time to optimize everything. Nontheless, not terrible, all crud considered.
Frame times dont match with frame rate, causing stuttering
- Chief Blur Buster
- Site Admin
- Posts: 11653
- Joined: 05 Dec 2013, 15:44
- Location: Toronto / Hamilton, Ontario, Canada
- Contact:
Re: Frame times dont match with frame rate, causing stutteri
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter
Forum Rules wrote: 1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
2. Please report rule violations If you see a post that violates forum rules, then report the post.
3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!
- Chief Blur Buster
- Site Admin
- Posts: 11653
- Joined: 05 Dec 2013, 15:44
- Location: Toronto / Hamilton, Ontario, Canada
- Contact:
Re: Frame times dont match with frame rate, causing stutteri
My WORK system.
(I'll have to add another RGB at bottom to light up my frosted UFO better)
About motherboard
I still suggest motherboard replacement. I sometimes see peoples' LatencyMon change quite massively whenever motherboards a switched. This is almost your only unturned stone, so your best chance is probably a couple dice rolls in the Motherboard Lottery.
About power conditioner
A power conditioner isn't a shabby idea as it will protect your investment (and power-imperfections-induced freezes do exist, but almost impossible to confirm). Also, motherboards with superior (or simply less defective) power electronics can also filter imperfect power better and reduce odds of power-induced problems. Reduced noise in a PC reduces the number of errorcorrecting behaviours within all those errorcorrecting logic now throughout a modern computer, and cleaner power helps with that overall. All those are often nanoseconds and microseconds (except for the cascade effects into the human visible timescales, as described in earlier posts) This is a sub-1% chance type thing but the 99% "probably not factor" is still justified by the general protecting an expensive computer from lighting and power surges, as well as potentially UPS features (built into some, not all, power conditioners) to save your work from loss.
(I'll have to add another RGB at bottom to light up my frosted UFO better)
About motherboard
I still suggest motherboard replacement. I sometimes see peoples' LatencyMon change quite massively whenever motherboards a switched. This is almost your only unturned stone, so your best chance is probably a couple dice rolls in the Motherboard Lottery.
About power conditioner
A power conditioner isn't a shabby idea as it will protect your investment (and power-imperfections-induced freezes do exist, but almost impossible to confirm). Also, motherboards with superior (or simply less defective) power electronics can also filter imperfect power better and reduce odds of power-induced problems. Reduced noise in a PC reduces the number of errorcorrecting behaviours within all those errorcorrecting logic now throughout a modern computer, and cleaner power helps with that overall. All those are often nanoseconds and microseconds (except for the cascade effects into the human visible timescales, as described in earlier posts) This is a sub-1% chance type thing but the 99% "probably not factor" is still justified by the general protecting an expensive computer from lighting and power surges, as well as potentially UPS features (built into some, not all, power conditioners) to save your work from loss.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter
Forum Rules wrote: 1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
2. Please report rule violations If you see a post that violates forum rules, then report the post.
3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!
-
- Posts: 24
- Joined: 24 Mar 2019, 19:53
Re: Frame times dont match with frame rate, causing stutteri
Told my brother to test latencymon and he sent me this pic, he mentioned he only had chrome opened, and when he took the screenshot with ShareX, there was a spike in the high latency mon, not sure what he could do
After some minutes (he locked the session and signed in again)
After some minutes (he locked the session and signed in again)
- Chief Blur Buster
- Site Admin
- Posts: 11653
- Joined: 05 Dec 2013, 15:44
- Location: Toronto / Hamilton, Ontario, Canada
- Contact:
Re: Frame times dont match with frame rate, causing stutteri
(Was drinking coffee) *chokes* *splurt* *sprays coffee*
My crud-loaded system run circles around that. That's not an efficient computer.
THAT will cause human visible stutters with just a single event (not even a cascade!)
BTW, I can see 1ms microstutters in motion-blur-reduction mode in fps=Hz synchronized motion of smoothpans like keyboard strafes -- metaphorically equivalent to a 4 pixel jump at 4000 pixels/second TestUFO.
Not an issue for 60Hz sample-and-hold displays obviously, but motion-optimized blur-reduced displays (e.g. NVIDIA ULMB Pulse Width 50% for sub-millisecond MPRTs) can actually expose 1ms stutter-visibility to human eyes when Time(persistence) is roughly equal or lower than Time(stutter), where Time(stutter) equals differences in (frame-emitting-photons-time) vs (real-world-time)
Now if even small cascades happen (10-event chain of 1ms piling up to a 10ms freeze between frames), nothing stops that stutter from becoming visible on plain full-blur 60Hz displays.
My crud-loaded system run circles around that. That's not an efficient computer.
THAT will cause human visible stutters with just a single event (not even a cascade!)
BTW, I can see 1ms microstutters in motion-blur-reduction mode in fps=Hz synchronized motion of smoothpans like keyboard strafes -- metaphorically equivalent to a 4 pixel jump at 4000 pixels/second TestUFO.
Not an issue for 60Hz sample-and-hold displays obviously, but motion-optimized blur-reduced displays (e.g. NVIDIA ULMB Pulse Width 50% for sub-millisecond MPRTs) can actually expose 1ms stutter-visibility to human eyes when Time(persistence) is roughly equal or lower than Time(stutter), where Time(stutter) equals differences in (frame-emitting-photons-time) vs (real-world-time)
Now if even small cascades happen (10-event chain of 1ms piling up to a 10ms freeze between frames), nothing stops that stutter from becoming visible on plain full-blur 60Hz displays.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter
Forum Rules wrote: 1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
2. Please report rule violations If you see a post that violates forum rules, then report the post.
3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!
-
- Posts: 24
- Joined: 24 Mar 2019, 19:53
Re: Frame times dont match with frame rate, causing stutteri
My brother opened and closed fortnite and this happened:Chief Blur Buster wrote:(Was drinking coffee) *chokes* *splurt* *sprays coffee*
My crud-loaded system run circles around that. That's not an efficient computer.
THAT will cause human visible stutters with just a single event (not even a cascade!)
BTW, I can see 1ms microstutters in motion-blur-reduction mode in fps=Hz synchronized motion of smoothpans like keyboard strafes -- metaphorically equivalent to a 4 pixel jump at 4000 pixels/second TestUFO.
Not an issue for 60Hz sample-and-hold displays obviously, but motion-optimized blur-reduced displays (e.g. NVIDIA ULMB Pulse Width 50% for sub-millisecond MPRTs) can actually expose 1ms stutter-visibility to human eyes when Time(persistence) is roughly equal or lower than Time(stutter), where Time(stutter) equals differences in (frame-emitting-photons-time):(real-world-time)
- Chief Blur Buster
- Site Admin
- Posts: 11653
- Joined: 05 Dec 2013, 15:44
- Location: Toronto / Hamilton, Ontario, Canada
- Contact:
Re: Frame times dont match with frame rate, causing stutteri
After only less than 3 minutes?
DPC of >16ms? Nosebleed stratosphere. Stutters from single events (non-cascades) of that will be visible on blurry full-persistence very old 60Hz LCD displays. Not even enough motion blur on those worst displays to hide stutters from those spikes.
The six thousand dollar question (my domestic cost of two high-end computers) is.... what's causing that abomination.
DPC of >16ms? Nosebleed stratosphere. Stutters from single events (non-cascades) of that will be visible on blurry full-persistence very old 60Hz LCD displays. Not even enough motion blur on those worst displays to hide stutters from those spikes.
The six thousand dollar question (my domestic cost of two high-end computers) is.... what's causing that abomination.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter
Forum Rules wrote: 1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
2. Please report rule violations If you see a post that violates forum rules, then report the post.
3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!
-
- Posts: 24
- Joined: 24 Mar 2019, 19:53
Re: Frame times dont match with frame rate, causing stutteri
So, lets say you open fortnite or any other videogame, this doesnt happen to you? Even opening and playing videogames your latency stays low? Im not sure what is causing this, Im hoping tonight to try to find the problem. Ive noticed something like if I minimized a screen or open a full screen game like Fortnite or the witcher, while the process of minimizing or opening the full screen program, and Im on discord, people wont listen to me, like theres a pause on the screenChief Blur Buster wrote:Nosebleed stratosphere. Stutters from that will be visible on blurry full-persistence very old 60Hz LCD displays. Not even enough motion blur on those worst displays to hide stutters from those spikes.
The six thousand dollar question (my domestic cost of two high-end computers) is.... what's causing that abomination.
Ive also noticed that when I turn on my pc, before I sign in to my windows account, theres a "pause" of the screen too for about 1 second.
- Chief Blur Buster
- Site Admin
- Posts: 11653
- Joined: 05 Dec 2013, 15:44
- Location: Toronto / Hamilton, Ontario, Canada
- Contact:
Re: Frame times dont match with frame rate, causing stutteri
Startup pauses is not something I normally bother to troubleshoot -- what important is what happens when all the startup settles down.
Now.... if your typing freezes in a different window for 1 second while another application is opening, then you have a problem somewhere. Typing freezing in one of your windows for 1 second when a different app launches -- that's not good.
Have you scanned your Samsung EVO SSD for errors -- via Samsung Magican software? I've seen barely failing SSDs start inserting random pauses in computer systems. I've got the Pro and you only got the EVO, and the EVO is known to fail more often than the Pros. But EVO is really, really good compared to a lot of the SSD competition. Longshot, but worth a try.
You already said you've scanned for malware.... right? A fresh reinstall (the complete wipe type) is sometimes the way to go too and that is something you could do simultaneously with a motherboard replacement.
On my system, Application launches doesn't even noticeably slow down my other windows at all, most of the time. It's as if I'm not even launching any application at all -- my system feels really fast even with background launching. Even a background installer app that is downloading & installing software gives nearly zero stutter to my foreground game or VR software, as most installers are single dominant threads, and most games don't need all the threads of my CPU. Although I'm not Threadrippered to the wazoo, I at least have sufficient threads and SSD power to make that kind of overhead invisible, so if your overhead is not sufficiently invisibilized then you've possibly got an underperformance or problem somewhere.
Now.... if your typing freezes in a different window for 1 second while another application is opening, then you have a problem somewhere. Typing freezing in one of your windows for 1 second when a different app launches -- that's not good.
Have you scanned your Samsung EVO SSD for errors -- via Samsung Magican software? I've seen barely failing SSDs start inserting random pauses in computer systems. I've got the Pro and you only got the EVO, and the EVO is known to fail more often than the Pros. But EVO is really, really good compared to a lot of the SSD competition. Longshot, but worth a try.
You already said you've scanned for malware.... right? A fresh reinstall (the complete wipe type) is sometimes the way to go too and that is something you could do simultaneously with a motherboard replacement.
On my system, Application launches doesn't even noticeably slow down my other windows at all, most of the time. It's as if I'm not even launching any application at all -- my system feels really fast even with background launching. Even a background installer app that is downloading & installing software gives nearly zero stutter to my foreground game or VR software, as most installers are single dominant threads, and most games don't need all the threads of my CPU. Although I'm not Threadrippered to the wazoo, I at least have sufficient threads and SSD power to make that kind of overhead invisible, so if your overhead is not sufficiently invisibilized then you've possibly got an underperformance or problem somewhere.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter
Forum Rules wrote: 1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
2. Please report rule violations If you see a post that violates forum rules, then report the post.
3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!
-
- Posts: 24
- Joined: 24 Mar 2019, 19:53
Re: Frame times dont match with frame rate, causing stutteri
It seems my problem right now is with this:Chief Blur Buster wrote:Startup pauses is not something I normally bother to troubleshoot -- what important is what happens when all the startup settles down.
Now.... if your typing freezes in a different window for 1 second while another application is opening, then you have a problem somewhere. Typing freezing in one of your windows for 1 second when a different app launches -- that's not good.
Have you scanned your Samsung EVO SSD for errors -- via Samsung Magican software? I've seen barely failing SSDs start inserting random pauses in computer systems. I've got the Pro and you only got the EVO, and the EVO is known to fail more often than the Pros. But EVO is really, really good compared to a lot of the SSD competition. Longshot, but worth a try.
You already said you've scanned for malware.... right? A fresh reinstall (the complete wipe type) is sometimes the way to go too and that is something you could do simultaneously with a motherboard replacement.
On my system, Application launches doesn't even noticeably slow down my other windows at all, most of the time. It's as if I'm not even launching any application at all -- my system feels really fast even with background launching. Even a background installer app that is downloading & installing software gives nearly zero stutter to my foreground game or VR software, as most installers are single dominant threads, and most games don't need all the threads of my CPU. Although I'm not Threadrippered to the wazoo, I at least have sufficient threads and SSD power to make that kind of overhead invisible, so if your overhead is not sufficiently invisibilized then you've possibly got an underperformance or problem somewhere.
I think I've managed to somehow avoid the ones on the other images, at least so far, but that one doesnt matter, it still reaches that amount easily.
I dont know if the other values are normal so far:
-
- Posts: 25
- Joined: 21 Dec 2017, 21:20
Re: Frame times dont match with frame rate, causing stutteri
I can suggest something that has helped me in the past if it wasn't already mentioned in the thread as i just skimmed through most of it.
Try getting msi utility v2 and turning on msi mode interrupts for your gpu driver and then see if it has any improvements to the stuttering.
Try getting msi utility v2 and turning on msi mode interrupts for your gpu driver and then see if it has any improvements to the stuttering.