Aces High Bulletin Board
General Forums => Hardware and Software => Topic started by: MADe on November 14, 2016, 11:35:56 AM
-
so I'm flying last night, DX11, still trying to dial in new rig, using fast v-sync setting, pulling 150+ f/s. Screen freezes a beat, then moves on. This is the only real issue for me. No stuttering, just freezes and resumption of play.
I cannot tell if my system hardware was overloaded for a beat or if it got stuck waiting for packets from the server, or a software glitch burped.
still using only game options for graphics.
NVidia 373.xx driver
NVidia hd audio, hdmi hdtv speakes
realtek audio, headset/mic
I was not really keying mic much so no voice to freeze anomaly. It only happened 1 or twice but still.......many things have been disabled or de-auto'ed.
-
run the dx9 version of the game and try again. There is a bug HTC is having trouble finding that is hitting some players running the dx11 version. Just start the game using the dx9.exe instead of the dx11.exe.
If you still have the little freezes then it is something in your system that is the most likely culprit. If it doesn't, then it may be the bug and it would be helpfull if you posted your specs in the "petite freezes" thread.
-
I reckon it's HDMI audio related.
I use analog audio and never have the problem.
-
I reckon it's HDMI audio related.
I use analog audio and never have the problem.
QFT!
This is a known issue but there is a work around if you must have sound from the speakers on the TV. Use RCA cables from sound card audio output and plug those into TV. Crude yes but it eliminates the HDMI sound issue,just disable HDMI sound in your sound card settings.
:salute
-
I reckon it's HDMI audio related.
I use analog audio and never have the problem.
I dont use hdmi audio, dont even have the driver's installed. and I still get the stutters.
semp
-
I dont use hdmi audio, dont even have the driver's installed. and I still get the stutters.
semp
Have you also disabled the hdmi audio in Device Manager? Windows tends to install a generic HDaudio driver for it in case no "real" driver is present.
-
Why would HDMI drivers get freezes in DX11 when DX9 dont ?
-
Why would HDMI drivers get freezes in DX11 when DX9 dont ?
I don't know any exact answer, but logically thinking since DX11 can do a lot more than DX9, there's also a lot more possibilities for conflicts.
-
There is a substantial difference between the underlying architectures of DirectX9 and DirectX11.
-
Does posting up AH recorded sorties provide intel about actual experiences?
ie: will others viewing see the same stutters or freezes that I experience
I have been slowly doing sorties with different software/hardware configs to see if theres a specific cause to my little stutters.
I intended to post them with the config that was flown.
:salute
-
Typically, we have not seen the stutters show up in the films, so we have been asking for both films and a Fraps-like recording (thsy do show the stutter) of the same event, so we can accurately assemble the events of the stutter.
So far, there has been zero consistency in the few sets of films we have gotten.
Windows 10 seems to be a source of some it as we have noted complaints about other games having stutters as well, but there is not enough data to make an accurate assessment.
We are still speculating what it could be. The lack of data is really slowing the progress.
-
Right off let me say that nothing I did or tried had any positive result. I would still get minor stops and starts. IMO they are internet burps.....
Monitor is a SONY BRAVIA 48W600B, internet tests give it a 33ms input lag, 60Hz refresh, HDMI>HDMI.
picture settings> game original, all post processing disable, display area full pixel, advanced settings have color, white, black all full
Gigabyte GTX 1070G1, stock, no SLI, latest Nvidia driver with only driver and HD Audio installed.
ASPECT Scaling 16:9 with GPU selected and all override selected.
Global profile settings disabled, minimized ....DSR enabled at x1.78, res 2560x1440 @60Hz
AH3 profile, all settings MAXED, app overrides, 1 pre rendered frame, no 3 buffer, vsync fast, I thru in everything but the kitchen sink.
ACES HIGH, all detail sliders maxed, environment none, reflections disabled, post lighting enabled but NO options selected, v-sync disabled, 2560x1440 res
And its dam BEAUTIFUL!
-
I disabled Nvidia HD audio
I disabled RealTek Audio
I used native res of 1920x1080p@60Hz
I used an older driver
I disabled all nvidia profile settings and used only AH settings
I changed cpu affinity to 3 cores, then 2 core, 3&4.
I used nada but performance choices, now above I pushed the card hard.
I want to say that its internet related, that the machine is waiting for something. Crowded areas, as usual, were a dominating factor.
Ultimately its the same as its always been for me. There were always hiccups in a crowd.
I'm low and slow over a field, banging thru the views major attack, view comes forward and 6 enemy icons are all co-alt and.....burp.......its like it had to process all the new objects simultaneously and was not moving till it was all done. pure conjecture on my part, attempt at an illustration.......
I logged immediately and was sport'in a 300+ms ping times in the GUI.................
:salute
-
We are pretty sure the long pause (1/2 second) is a game issue. Well, possibly a shader compiler issue. It is very difficult to nail down as we have witnessed it once in the office and have never been able to duplicate it again.
-
We are pretty sure the long pause (1/2 second) is a game issue. Well, possibly a shader compiler issue. It is very difficult to nail down as we have witnessed it once in the office and have never been able to duplicate it again.
An chance of someone near by HTC office who is having this problem could let HTC see it in person?
-
An chance of someone near by HTC office who is having this problem could let HTC see it in person?
That is not going to help. We have to be able to trap it in a development environment with the debugger as it is happening on the video card and not in system RAM.
Debugging things happening on the video card is a bit of black magic mixed with a touch of luck.
-
So I just flew multiple sorties with the environment slider MAXed.
The other change was I plugged directly to modem. No router, 1 device.
The 1/2 sec freeze was much more active. I pulling 1024+/- vram. AVE 90+/- f/s.
I ride the views hard and flip around a lot. It seems like return view to the front, in a crowd, can trip a freeze. I'm not seeing much lag on the part of others as well.
This game cannot be bottle necking but its waiting for something in a sudden crowd................windows is behind in thread allocation..................i f I'm playing my own game, the only thing I get from server is object updates, once object aqiured theres no issues, but newly added to your sphere causes overload..................... ....
Is it possible that the vid card is downclocking and upclocking itself?
I have max power perf selected but you can see its not running boosted in GPUz unless test rendered. Is it possible the card gets comfy and slows itself because of low demands?
-
So I just flew multiple sorties with the environment slider MAXed.
The other change was I plugged directly to modem. No router, 1 device.
The 1/2 sec freeze was much more active. I pulling 1024+/- vram. AVE 90+/- f/s.
I ride the views hard and flip around a lot. It seems like return view to the front in a crowd can trip a freeze.
The Environment effect is very demanding.
If you're running that at Max, that could be part of your problem.
Coogan
-
The Environment effect is very demanding.
If you're running that at Max, that could be part of your problem.
Coogan
naw I have seen it with everything off. Just trying to overburden the card to see, and basically I get beautiful with a glitch. Overburdening just makes the issue slightly more pronounced, hoping to isolate something, maybe. Overall this system is rock'in.
-
naw I have seen it with everything off. Just trying to overburden the card to see, and basically I get beautiful with a glitch. Overburdening just makes the issue slightly more pronounced, hoping to isolate something, maybe. Overall this system is rock'in.
Yea, mine runs pretty smooth too, (using a GTX 970) until I try and turn up the environment slider. That's when my FPS tank.
I know we're talking about stutters here. Just thought I throw this out here.
Coogan
-
a sourceforge test of my connection. pc>modem>net
the bufferbloat times??????????????????how can I lower this?
i mean 300+ms qued up on DL side and 100+ms on upload, with a 70ms ping time, with whtaever the tic rate is.............wheres the slow down, cable modem or pc NIC?
-
keeping what I had on listed before...
i totally disabled v-sync
disabled shader cache
put LoD on clamp
in 3 sorties in a crowd only a single 1/2 sec freeze. this is with everything pushing.
-
also had this problem with DX11 and each time you get a freeze if you look at the net status there is one large spike. everyone else in our squad that get the freezes sees the same spike in the net status. i went to 9 with no issues now i ran on 11 from release until about 1 week ago and i could not take the screen freezes while trying to shoot anymore
-
http://www.majorgeeks.com/news/story/nvidia_adds_telemetry_to_latest_drivers_heres_how_to_disable_it.html
not that this is the issue, but........................
-
can I ask,
what cores are AH using?
using Afterburner during game, showed the card at 113C for a long while then boom, fast drop.
Now this same dip on the graph for other stuff dipped. mem clock, shader clock, volts but at same time stamp it had cpu core 5 pegged as well as several others working, more than 3??????????????
My glitch is definitely connected to banging thru the views and the rapid screen changes in a busy area. Could be coincidental due to a heat issue!, I had the fan at 90% constant, still went to 113C.
I had to get gigabytes utility and drop msi. test tommorrow.
:salute
-
Applications do not have a choice as to which cores get assigned to which threads. That is handled in the operating system thread manager.
Natvely, the game will use up to 3 cores. That does not take into account all the other parts of DirectX which may be threaded, such as sound, data loading to the video card and so on.
-
Welcome to AH3 Made:)
I fly in 4k Rez, and DX9 gives me about 60% of my FPS as DX11 does, so I just fly Dx11.
The micro pause is still there, no matter what I try.
As of yesterday:
I currently have uninstalled Nvidia Audio, and disabled it.
I'm also running a batch file to turn off superfetch on the SSD when I play. Pauses were less, but I also found it dependent on maps.
I also turned off Shadows....even though I think they add to the games immersiveness, the rectangle shadow bug is ugly, and the loss of light on cockpit is irritating. So everything is brighter and not hidden. Of course I did these three things all at once, but it didn't matter since I still got the pause, albeit not as frequent.
Good luck in your efforts to find the culprit and thank you for the Nvidia telemetry/wireless controller find......have that removed as well now.
Oh and if you find a way to make the icons not mask the aircraft when dogfighting please share as well.
Cheers!
Dobs
-
So todays observation.
I got the gigabyte utility from gigabyte. Xtreme Gaming
I was not gonna OC, but that 113C temp from afterburner reading got me piqued. I mean dam card has 3 fans on it by design, case has mucho airflow...
So I got the more correct choice. set it to game OC, set fan to turbo, done.
45 min sorty, not 1 glitch! Now it was a rather empty area so I will need to wait on a real conclusion.
New build and I have learned to stay away from such apps, could this one actually help?
I also moved a case thermometer I have right to the cards backplate, can observe real time and fly. 50C max. Logging game and going directly to monitor app and its more acceptable than last nights. I wonder.
BTW the map in melee looks great! I wish I could take a pic that would convey what I see.
:salute
-
Applications do not have a choice as to which cores get assigned to which threads. That is handled in the operating system thread manager.
Natvely, the game will use up to 3 cores. That does not take into account all the other parts of DirectX which may be threaded, such as sound, data loading to the video card and so on.
tanx skuzzy
-
a sourceforge test of my connection. pc>modem>net
the bufferbloat times??????????????????how can I lower this?
i mean 300+ms qued up on DL side and 100+ms on upload, with a 70ms ping time, with whtaever the tic rate is.............wheres the slow down, cable modem or pc NIC?
I answered my own question here. So net tells you performance tweak this that the udder thing right.
anyways the test for bloat is about TCP and those protocols that will respond to shaping, unlike UDP.
I had done game tweaks with the NIC. Those tweaks created a George Washington Bridge traffic effect, retarded movement. DEfaulted most things and bufferbloat dropped considerably. Means nada to game.
-
couple of sorties, in a crowd, perfect!
so then my problem was a heat issue?????????
so then some of the basic profiles with the app, it what, tunes the card properly?????
if my issue was heat, then AH3 has NO glitch where I'm concerned.
v-sync disabled
all nvidia settings max, auto thread disabled, triple buffer disabled, pre-rendered 1, DSR 2560 x 1440
AH max all sliders, ALL. post light enabled no options selected, reflection disabled
ave 80f/s in a crowd, 1100+/- VRAM
I will add in the post light selections 1 at a time, I will leave games AA off and reflections off. I hate the reflection and nvidia doing the AA.
Its beautiful, Have I said that.
:salute
-
heat issue
solved
the vid card performs better with the Gigabyte Xtreme Game utility. It must work with the cooling better than nvidia driver alone.
I thought I'd be doing myself a favor by not junking up the pc with stuff like this utility. I had not been watching gpu temps cuz the pc is running in the low 30's and so new.
:salute
Now, At certain altitudes the trees flicker in the white , towns same, all very white shades seem to scintillate constantly.
Whats this?
:salute
-
Applications do not have a choice as to which cores get assigned to which threads. That is handled in the operating system thread manager.
Natvely, the game will use up to 3 cores. That does not take into account all the other parts of DirectX which may be threaded, such as sound, data loading to the video card and so on.
Thanks, Skuzzy.
That's good to know.
:salute
-
test 1 and 2 following Pudgie lead.
https://msdn.microsoft.com/en-US/library/ee377058(v=bts.10).aspx
these test done with the reg tweak describe for threads only.
done to see frametime diff if any, i think there is but...........frametimes seem more consistant with no affinity assignments
the OS thread is maxed.
-
test 3
-
I enabled vsync to adaptive in profile no other change.
I started up AH and assigned it to cores 7, 8, 9. I then went thru every other process listed on the list and disabled core 7, 8, 9. Now I was not able to adjust every process but I got most.
It froze up multiple times but kept going.
1st graph I use as a base line, I was just sitting a runway for 5 min.
2nd was the freeze, temp was acceptable but total system burp, cards, cpu. Frametime shot thru roof as the gpu completely went 0% usage.
You can really see AH using 3 cores. Crowd it starts to fluctuate.
for what its worth
Interestingly when windows left alone, the 1st core is pegged, frametime has a +10ms variance, with tonights setup cores 7, 8, 9, spread a load evening amongst themself, other cores did a little and frametimes were slightly steadier BUT IT FROZE.
-
today,
i disabled HT in bios. Now just 6 cores.
I adjusted additional worker thread setting to 6 from 12.
v sync disabled
AH given above normal process priority
Like yesterday, I selected 3 cores and isolated them as much as I could by adjusting affinity assignment options for every listed process. I chose 3, 4, 5 cores, removed affinity assignment from all stuff and allowed AH to run on only the 3. Major difference is that yesterday I chose all HT cores for assignments, very bad. Today will be all primary cpu cores!
I used ah with 3 cores, with 2 cores, with 1 core.
With each selection I went to same map, same area and sat the run way as a baseline reading. I went to arena crowded areas.
I dog fought, cruised, died, killed. I would stop and check the graphs at high point moments, kills, deaths, freeze glitches which were few.
By far and away the game runs best on 2 CORES only. The frametime responses appear to flatline the best, with variations occurring with view changes, objects entering vicinity, usual. What appears to me to happen is when the threads, dominant core shifts, it spikes the gpu usage to 0 and frametime 1000ms+.
I do not begin to understand the trigger threshold when 1 core gets maxed and a second is added but that crux point leads to a system glitch, or burp. Using 3 cores makes for more burps as windows tries to balance or pass things around I guess. Either way the burps are more pronounced because there is more core changing. Graphs show as a core dips its second core spikes as well as gpu taking hit, all at same time.
Now if I set up nada and just go with it it appears that the game latchs onto same core as windows OS, core 1. In game you can see the core constantly pegged so there some core handoffs happening and the frametime variation is doubled, with occassional system spikes.
I believe this was/is Pudgie's point. I roll with Nvidia so it seems to happen both.
It does feel better, the game, with it confined and all alone on 2 cores. They both get used more than 50% and the handoffs seem a lot less.
-
Hi MADe,
I read and saved the document that you linked then went into registry and found the 2 registry settings as follows:
AdditionalDelayedWorkerThread s was already set in registry at 0x10 (16) so nothing to do there.
AdditionalCriticalWorkerThrea ds was found in registry set at default (0x0 (0)) so I changed this to the recommended setting of 0x10 (16). This 1 I find critical as from understanding the importance of these worker threads to process time-critical work thus the stack is kept active in system mem for availability (with the fact that I possess a LOT of system mem onboard (16Gb in Quad Channel config) relative to my box's usage so this taking up mem space is a moot point to me (also checked all this in Task Manager and noted no significant mem usage amount change...less than 200Mb).
Rebooted my box and all came up well.....Windows shows to have picked up a little since doing this (little snappier and crisper in operation to me). While I was in the registry I also made a few cosmetic changes to give the appearance of snappier performance (reset MenuShowDelay from the default of 400 to 150 and set up NoLowDiskSpaceChecks then set to 1...disables this so Windows doesn't waste time checking my SSD's for disk space as I do this myself quite often).
Went flying under this "new" Windows OS config and all went very well using 3 CPU cores w\ AHIII Dx11 but the CPU core usage distribution was very low.....game ran well but this bothered me so I went and reset the CPU core affinity back to 2 CPU cores for AHIII Dx11 and retest.....now all was looking much better to me and the game ran very well on my box so I'm gonna stick w\ 2 CPU cores for AHIII Dx11 (3 CPU cores works very well w\ AHIII Dx9). Here is a snippet of my box running AHIII Dx11 using 2 CPU cores and reconfig'd OS provided below:
I encountered my 1st screen pause since updating to Patch 12 during this run (note the FT spike on left side of graph) but it didn't last long....less than .25 secs and resumed normal operations. Never saw another 1 since this 1 rest of the night. All ran well on my box.
Continuing to test.........................
:salute
-
All you are doing when you disable or enable cores is showing how efficient or inefficient the operating system thread manager is.
Also, when you reduce the core count, you reduce the likelihood of bus contention happening. Bus contention is the thing where multiple cores need external access (memory, I/O...) and they have to wait in turn as only one core at a time can access external resources.
Now, with reduced cores, also means threads have to be run in time slices which can cause all the threads to be run slower. A thread only needing to access the cache of the core it is running on can run full speed in parallel with everything else. That thread would be slower to run if it had to wait for a time slice to run.
-
I was not gonna OC, but that 113C temp from afterburner reading got me piqued. I mean dam card has 3 fans on it by design, case has mucho airflow...
I use AB to overclock and like it a lot. Plenty of tweaks and monitors. If you're gonna overclock at all you really have to be on top of the cooling otherwise you risk toasting your card. Did you try messing with the fan curve tab? There you can set up a curve of fan speed relative card temp. Just drag and drop the points as you like and check the box to enable it. AB will monitor your card's temp and ramp the fans up or down accordingly. I believe that curve is flat by default which may be why your temps went so high. 113 is pretty brutal! I get askeert if mine get to 80! But at the same time you don't need the fans howling away just to surf the net. The curve is handy for that too. Without looking at it I don't believe it goes below 30%.
I'm looking at an MSI card that will actually kill the fans completely under minimal loads for quieter operation. Really good coolers on the MSI cards.
Sent from my XT1585 using Tapatalk
-
All you are doing when you disable or enable cores is showing how efficient or inefficient the operating system thread manager is.
Also, when you reduce the core count, you reduce the likelihood of bus contention happening. Bus contention is the thing where multiple cores need external access (memory, I/O...) and they have to wait in turn as only one core at a time can access external resources.
Now, with reduced cores, also means threads have to be run in time slices which can cause all the threads to be run slower. A thread only needing to access the cache of the core it is running on can run full speed in parallel with everything else. That thread would be slower to run if it had to wait for a time slice to run.
Bus contention, this must be those spikes I noticed. The gpu would suffer at these contention points yes?
The game runs the best for me on 2 cores, it feels better. Maybe a placebo effect....
but the graphs bear out something..........I think getting it off core 1 and isolating it does help as well.
-
Bus contention occurs when one core screams, "I need the bus!" and another core yells, "I need it too!". It is then up to the hardware priority who wins. One gets the bus, the other goes into a wait state (consider it asleep as it literally has to stop everything) until it can get the bus.
-
let me ask this.
Are core assignments like a cascade effect?
ie: as demand increase things are diverted to next in line, so to speak.
I understand when you say I'm just moving things around. But I have also isolated the chosen cores from almost all other things.
I also wonder if a core takes precedence 1st, I know windows grabs the cpu 0. I know that when AH loads it uses cpu 0, pegging that core to 100%.
The graphs where I did no affinitization show pegged cores and lots of widely variated measurements. Other extreme which I used affinitization showed game cores not pegged and measurements more flat lined. Not perfect mind you but much less glitchy.
I really think its this thread contention is a possible issue for lesser pc's. Mine just bulls thru it, but I can recover, others????????????????
:salute
-
today
HT enabled
worker thread tweak to 12
affinity set for cores 3 and 9 for AH only.
If I understand correctly, I enabled core 3 and importantly its hyper thread companion 9 for AH only. I flew.
bus contention, skuzz new word to google you gave me...............
Unfortunately the arena was under 100 pilots by the time I got home but I got a baseline pic, sitting the tower. The other was right after getting waxed over an enemy field.
Note the lack of spikes, there is variation but no show stoppers, if it is thread contention making the graph spike, then on 1st impressions, mating a core with its HT core appears to have less thread contention.
Now I have to stick with this arrangement for a period of time to see if..........
graphs are best yet, very consistant, clean even.
-
Here ya go, MADe.......................
I finally got around to running your test according to your directions w\ a small deviation..............
When I went to kill CPU core 3 & 9 on the remaining running processes Win 7 SP1 said nada to that......denied all attempts.
I also forgot to go into registry and reset AdditionalDelayedWorkerThread s to 12 and AdditionalCriticalWorkerThrea ds to 12 (had reset these in my later on from the recommended 16 to 8 to give a couple more than the number of CPU cores available....6 at the time) so the results were using 8 instead of 12.
Just ran AHIII Patch 13 Dx11 w\ this configuration......snippet of MSI AB graph below.
Pretty much got the same results as you did in spite of the differences.............
FYI.......................... .
:salute
-
Update:
Have gone into registry and reset Executive settings as directed then ran test again......got the same results as before which were very good.
:aok
Here is a handy little link to an article on how to build a Windows shortcut to an app\game in which you can assign CPU core affinity in the target (or location) line so when you use it to run the app\game this is done automatically in the order you specify...................
https://www.eightforums.com/tutorials/40339-cpu-affinity-shortcut-program-create-windows.html
:salute
-
so i got into crowd and the graph pretty good. by good i mean no extremes, same consistant variability.
i enabled post lighting options
i backed all 3 tree/terrain/ sliders back to 80%.
the only thing not selected is reflections and AA
-
can someone tell me why I do not want these kinda of graphs.
I was buzzzin all over in a crowd, freakin got the butter................
Seriously why wouldn't I keep doing this?
HT my machine likes this. Fix it so I do not have to redo affinity assignments at reboots.
let me see...
I'll need a batch file at start up that remembers my choices, and tell AH to just use the cores I select option... and..............
just little scripts.
:x
-
Hi MADe,
This is why you want to use GPU frametiming graph lines to determine optimum system function as to achieve this you will need to focus on more than just a GPU driver's settings as the CPU operation is just as critical.......and our eyes can fool us just looking at a graphics scene alone as to the cause of what we see. It ain't always the GPU at fault but the evidence will show up thru the GPU frametiming graph line..............
As soon as I read the initial article on the advent of FCAT a few years ago I understood what the goal was........just needed a method of measuring this w\o having to buy & install a frame capture card (expensive). Unwinder stepped up to the plate and wrote a method into his code to provide this measurement & MSI licensed it into Afterburner (EVGA used to license his code until they parted ways & EVGA then wrote their own GPU control software). At the time MSI AB was the only 3rd party GPU control software that works equally well on both Nvidia and AMD graphics cards so in essence a 1 stop shop piece of software....which is why I (and a LOT of other users) use it & other users prefer it over the 3rd party GPU control software of the graphics card brand they bought......unless they too licensed Unwinder's code.
:salute
PS---I agree w\ your statement as well......................... .........
:salute
-
Here's where I'm settling at on all this.
I'm running w\ Hyperthreading disabled and running AHIII w\ CPU core affinity assigned as follows according to the binary numerical index:
101010 (right to left.....using CPU core1, 3, 5 staying off CPU core0, 2, 4 so most other processes outside of AHIII will be assigned to core0 which will pretty much give AHIII exclusive unshared access to core1 but especially core 3 & 5 and maintain CPU core temp leveling).
Since setting all this up I also discovered that w\ the new AMD Crimson ReLive 16.12.1 drivers came Radeon WattMan GPU power\clock control which helped this FuryX big time to harness the capable power delivery system built into these cards to handle the Fiji GPU that I was actually holding this GPU back using AMD's FRTC frame limiter software set @ 80 FPS. Reset FRTC to the FreeSynch upper FPS limit of 90 FPS (corresponds w\ the monitor Hz setting in Windows of 90Hz w\ Vsynch to keep the vid card from running RR above the FreeSynch upper FPS limit) and now this FuryX is running AHIII pretty much pegged at 88-90 FPS and doing it very smoothly as this graph provided below will show:
Note: I have AMD's new Chill smart frame rate limiter enabled but when I look at the MSI graphs it shows it isn't working (graph lines are too smooth) according to a video of Scott Wasson @ AMD giving a preview of AMD Chill's operation so we'll see how it shakes out later on.
:salute
-
been keeping all the above mentioned choices for games options and NVidia profiles options.
I now run the 3 terrain detail sliders at 75%. Having these maxed is unnecessary, when down low causes landscape stutters, when high causes whites to scintillate annoyingly. 75% provides best solution without loss. The maps really look good.
Using 12 added worker threads, HT enabled.
I am still confining AH to cpu 3 and its ht counterpart 9. Priority at above normal selection.
CPU 3 and 9 have been removed from all other apps or services.
I have been playing with v sync options, adaptive, fast and off. OFF has some minor freezes occasionally, but feels better game play wise. The other 2 provide butter smooth presentation but it feels stunted...................pla cebo effect???????..........................
Skuzzy said this was just testing MS's ability to multithread. I think MS's multithreading technique sux's. Strictly a crisis management approach. So much of windows is about management of mobile devices, power consumption, blah blah blah. A desktop PC suffers from this! Following the msi graphs shows where core handoffs occur and those resulted in a glitch. If ms is just gonna handoff when a core gets overloaded................... ..........day late and a dollar short...........
I have tried the affinity approach before but I never matched a core with its ht counterpart and I never isolated the cores from windows. This does appear more spontaneous for moi.
:salute
-
Hi MADe,
I finally figured out how to code into the cmd command line in my AHIII shortcuts the syntax to not only have Windows to assign the CPU core affinity selection I have chosen when AHIII process is executed but also to have Windows assign AHIII running process high priority on top of this w\ a single click of my trackball so the game operation will get pushed to the forefront of Windows process management.......I found this also makes a difference w\ how AHIII runs as now it's threads will be getting elevated priority from the OS vs the rest of the running processes (higher thread priority ranking)......once I did this I noticed that a lot of the cloud flickering in-game ceased on my box and the game performance smoothed out even more w\o any loss in other areas of operation. Maybe it's a placebo effect, maybe it ain't............. The results I'm seeing are definately real.
Have to say, the game is getting more enjoyable and exciting for me to play since doing all this...............
:aok
Skuzzy said this was just testing MS's ability to multithread. I think MS's multithreading technique sux's. Strictly a crisis management approach. So much of windows is about management of mobile devices, power consumption, blah blah blah. A desktop PC suffers from this! Following the msi graphs shows where core handoffs occur and those resulted in a glitch. If ms is just gonna handoff when a core gets overloaded................... ..........day late and a dollar short...........
Well after doing a lot of studying up on all this I can understand the reasoning to a certain degree as to why both MS and HTC have done what they have done........... The CPU cores aren't being overloaded per se, it's a side effect of the way in which they're being managed by the OS is what's causing the congestation to a few number of the CPU cores relative to the whole set of cores available.
For MS to write the logic into the OS to actually monitor then make these kinds of optimizations to optimize certain running processes to be used at the consumer level would be very tedious at best for little to no gain monetary wise (MS has already done some similar coding to do this.......it's active in the server-class versions of Windows OS's where it makes far more sense.....and money to boot.....to provide\use this complex of code algorithims) and for HTC to go thru hoops to write into the AHIII client all the coding to instruct the OS to optimize AHIII for each operational scenario, computer configuration being used\perceived to be used to run their software which would make the game client be overly complex and wouldn't run well on lesser systems due to the complexity causing issues so this was kept simple on the OS side to accomodate as many software developers as possible to achieve widespread adoption.....but w\ enough sophistication written in the OS for a software developer to make use of if they so desire.
So does it work as it is? The answer is yes. Can it work optimally as it is currently being used? Depending on the app\game being run, not always but it works well enough for the vast majority of uses at the consumer level of computing.
The issue is the definition of what is considered to be optimum regarding this topic...............
So all you & I are doing is making manual application(s) to take advantage of what is already there to be taken advantage of to help AHIII run better......
I'm also taking advantage of a method of setting all this up in a manner that automates the application(s) process w\ a single click of the mouse................
Where I see all this becoming useful is when a CPU that is equipped w\ more than 4 physical CPU cores on die is being used. This is what I discovered on the side from all my testing to try to find the cause of the screen pauses\freezes. It is these type CPU's that the consumer versions of Windows is really gonna have issues properly optimizing all of these cores unless the OS is revamped to improve thread management\optimization\utilization across them.......
I have always believed that once you understand how something works, you can then exploit it to it's fullest potential or to the fullest potential of your level of understanding. It's there to be had.....just depends on whether it's worth the effort to get it.
The more I read, study & learn the more exploitation I will be able to take advantage of..................
Enjoy!
:salute
-
I see it differently.
MS OS's are really a forced generic environment that try to cover as many different variations of hardware and software as sanely possible. Lately they have tried to be somewhat more proprietary in there latest OS deployments as well..........
Latest games should and could be taking the bull by the horns and utilizing current available hardwares to their fullest but current game producers are not! I'm at a loss why they keep pushing multicore hardware with such poor software support in the industry.
AH is old software by modern standards, its not gonna take advantage, and its forced to operate under OS's bidding. Now AH itself could and should move itself off the standard OS operating core, 0, at minimum. By the same method you did. But theres small incentive.
I also wonder if it would help to have certain processes included with the games selected threads. ie: nvidia drivers
I should be able to run game, sound, nic, each on a dedicated core set.......
hmmm the pc has 41 processes running, i would need to understand what each does and if the game needs it besides the system
You see I'm all for manual, generic blah. You are using free sync so your 1 up on me, system wise. I'm using a 60Hz LED display and trying to over compensate. I really do like no vsync, I imagine its similar to freesync in feeling.
:salute
-
<snip>
Latest games should and could be taking the bull by the horns and utilizing current available hardwares to their fullest but current game producers are not! I'm at a loss why they keep pushing multicore hardware with such poor software support in the industry.
AH is old software by modern standards, its not gonna take advantage, and its forced to operate under OS's bidding. Now AH itself could and should move itself off the standard OS operating core, 0, at minimum. By the same method you did. But theres small incentive.
I also wonder if it would help to have certain processes included with the games selected threads. ie: nvidia drivers
I should be able to run game, sound, nic, each on a dedicated core set.......
hmmm the pc has 41 processes running, i would need to understand what each does and if the game needs it besides the system.
<snip>
First: Aces High III is a complete re-write of the game making use of the latest, greatest technology available to the game.
Second: No application can allocate itself to any given CPU core or cores. There is no API for that in Windows. That is the realm of the operating system and that is where it should be.
Third: The number of processes is certainly important, but take a look at the active number of threads running. That tells a more interesting story. My system, at this very second, only has 38 processes running on it, but there are 518 threads running. After I start the game and go online, the number of active threads jumps to 595.
The number of threads will vary, depending on what is going on in the game.
The CPU/core activity you see in graphs has a very course resolution of about 1/2 second. They are not going to visually be able to show all the activity.
Windows is a very bloated operating system. I mean, over 500 threads running (Windows 7 Pro), just for the operating system is absurd. If the operating system was really focused on running any given application, it would surprise you to see how fast it would be. Even AMD and Nvidia have been very vocal about the poor performance of DirectX. AMD got so fed up with it they invested in creating a new API.
-
Windows is a very bloated operating system. I mean, over 500 threads running (Windows 7 Pro), just for the operating system is absurd. If the operating system was really focused on running any given application, it would surprise you to see how fast it would be.
I can confirm that. Several years ago I reinstalled XP for a friend to be used for a USB effects/mixer platform at the training facilities of his band. So I just disabled most every process I figured would not be needed, including everything networking related. The speed boost was incredible!
-
Ran a little test on my box as follows:
Was at desktop w\ FF open this morning (reading this thread).....opened Task Manager to Performance tab to observe:
With FF open and all my stuff running in the notification tray came out to 76 processes running 1201 threads w\ 3%-5% CPU usage.
Shut down all 9 of my stuff in the tray (8 programs and Kapersky AV) and left FF open for a total of 14 processes removed so system is now at 62 processes running 888 threads w\ 0% CPU usage.
Then I shut down FF which removed 2 more processes from running for a total of 60 processes running 775 threads (desktop itself....Win 7 HP SP1).
Shut down my box to clear out the system mem then restarted box back to desktop fresh w\ all my stuff running but no FF............
Showed 80 processes running 1117 threads w\ 1%-3% CPU usage.
After 10 mins sitting w\o doing anything Win HP SP1 readjusted itself to 75 processes running 1086 threads w\ 0%-3% CPU usage..........
So all the stuff I've loaded is only creating 311-313 threads......the rest is Windows and drivers.......
I see your point......................
:salute
-
U know way more than moi.
U say bloated, I say generic.
If I can go in and manually assign affinity, I do not understand why AH cannot then do so as a matter of course?
Its all coding...........?thru task man you can do something, pudgie's link, and I will explore.
If windows allocates cores only as demand increases, if windows predominately uses cpu 0 1st, then why bother allowing the game to follow? The thread contention definitely diminishes when cpu 0 is not used for game.
Again I'm just a schmocuk but its seems that if MS is gonna play mr generic, AH could try a couple of end a rounds for better player experience. Core parking was a similiar stutter monster.
IE: AH2, I would run into this a lot. Be on a guys 6 getting into firing range and he would warp. I do not have such warps now. I still see latency warps, but its not close in any more. There is a difference in the games spontenaity with my setup and display.................
:salute
-
MADe, did I understand your post correctly, are you suggesting that AH should use the less used cores instead of core #0?
If so, knowing that AH can utilize three cores (IIRC) at the max, what would you suggest for a dual core cpu which can run AH just fine?
-
You guys keep drawing conclusions based on faulty data. You really think the CPU usage is really ever at zero? No, it is not.
Microsoft is not showing how much CPU it is taking to run the operating system. It does not show how much CPU activity it takes to just run all those threads it uses. There are some very active threads in the operating system. Just processing the message queues for all the GUI based threads is significant.
As far as taking over control, from the operating system and picking our own cores goes, there is nothing to be gained. What you might perceive as being better for your computer could be worse for another computer. The base assumption that what works for my computer must work for every other computer is not accurate. If it were true, then every computer would run Aces High III exactly the same.
You can take two identical computers and they will not act the same, based on what is installed on them. The operating system behavior is very easily altered by user software. It is why malware/spyware/viruses are so easy to inject into the system.
What you guys are doing is interesting, but it is almost impossible for you to draw conclusions, which can applied to any computer running Windows.
-
Through free software I've set my programs to permanently use higher cpu priority (in this case TrackIR, Teamspeak, and Aces High).
Specially TrackIR was sometimes lagging in operation before I started to use it.
-
You guys keep drawing conclusions based on faulty data. You really think the CPU usage is really ever at zero? No, it is not.
Microsoft is not showing how much CPU it is taking to run the operating system. It does not show how much CPU activity it takes to just run all those threads it uses. There are some very active threads in the operating system. Just processing the message queues for all the GUI based threads is significant.
As far as taking over control, from the operating system and picking our own cores goes, there is nothing to be gained. What you might perceive as being better for your computer could be worse for another computer. The base assumption that what works for my computer must work for every other computer is not accurate. If it were true, then every computer would run Aces High III exactly the same.
You can take two identical computers and they will not act the same, based on what is installed on them. The operating system behavior is very easily altered by user software. It is why malware/spyware/viruses are so easy to inject into the system.
What you guys are doing is interesting, but it is almost impossible for you to draw conclusions, which can applied to any computer running Windows.
absolutely no question
I am not sure I am drawing a conclusion, more an assumption. But I am still playing around and there is 1 recurring event, hiccup. The cpu cores usage drops, the frametime and framerate suffer accordingly. Its a minor thing but it happens no matter what my settings are. Can I ask, whats the server to client tick rate? just the positional sends is what I'd be interested in. My reasons are cuz I am playing around with vync and refresh rates on a 60Hz LED HDTV and well I'm playing...........
The game, with my system, runs better on 2 cores visually, less anomolies and hiccups. This is at its best, I complain 0 cuz its really nice looking and plane on plane is very good.
The game runs better when it is removed from cpu0. Of a 6 core cpu, I have run the gambit of core selection
TBH total understanding of the whys is not as important as a result.
For moi it is just 1 step to tuning my system for AH. I have no illusions/delusions. he he
:salute
-
MADe, did I understand your post correctly, are you suggesting that AH should use the less used cores instead of core #0?
If so, knowing that AH can utilize three cores (IIRC) at the max, what would you suggest for a dual core cpu which can run AH just fine?
I did hint that I feel AH/DX11 might benefit from a little aggressive handling with regards to the windows operating system. I do believe at minimum it should be coded not to use cpu0. My laymans reasoning being that the OS lives there 1, and 2 it appears that the hiccups I get coincide with thread handling. Windows appears to allocate cores by crisis management...............star t with cpu0 and spill down the line.
As far as dual core.....I have one and I have been considering checking this but its GHz is slow....
however I think with my current reasoning, AH runs better on 2 cores that it owns, so to speak, while the OS continues with cpu0. When running the game I have taken to just assigning AH just 2 cores. What happens is no single core is 100%, which is what happens if I do nada. Prolly minor thing.
But I have also taken to disabling vsync, in game and in nvidia. I use a 60Hz display, I now run it at 59Hz and use afterburners frame limiter, frames are capped at 59 now. I am slowly starting to convince myself that this is a big deal. I am now getting a spontenaity that was missing???????? Pudgie got me on this tac cuz he uses a freesync monitor and once I understood them well I just had to push my display to attain what it can't. My mind says 1:1 yes! way good so how can I get this. It appears through googling also that there is something to vsync causing input lag.
freesync/Gsync alleviate this.
:salute
-
I understand what you say, or at least I think so...
Just out of curiosity I opened Task Manager on the graph tab, I suppose the graph tells more about trends than the once in a second changing digits. There was very little going on, so I started AH. Immediately core #0 jumped to the top, #1 showing more modest action. So, it might make sense that setting #0 in a supporting role while favouring the other cores might balance the load. Then again, I know absolutely nothing about coding so I don't know if it were even possible. Also, as Skuzzy said, picking cores might not work for all computers. Plus the multitude of supported Windows versions not to mention what their updates might change in the operating system behaviour... Too many variables, IMO.
-
Through free software I've set my programs to permanently use higher cpu priority (in this case TrackIR, Teamspeak, and Aces High).
Specially TrackIR was sometimes lagging in operation before I started to use it.
Hi Save,
I've ran some tests running AHIII on my box using CPU affinity alone, then using CPU priority alone, then using both CPU affinity & CPU priority together by using cmd.exe......................
From all this testing it has shown me that the real goal is to get AHIII game client increased CPU core processing time to assist AHIII to operate better since it is a multi-threaded game client software that can take advantage of more CPU processing time assigned to it.
Setting CPU core affinity alone does accomplish this from a percentage standpoint but this also is limited to the CPU cores that are designated for Windows to assign the game threads to so when the game usage gets very heavy & the game thread counts were very high this tended to cause some CPU contention at certain points in the game scenes showing minor FPS drops but noticeable stuttering that even FreeSynch couldn't correct, depending on the number of CPU cores chosen at the time. After making a lot of game runs w\ various CPU core affinities set I found on my box using my Intel I7 5820K 6-core CPU w\ HT disabled the magic number of CPU cores needed to absorb all this activity was 4 CPU cores (which would equate somewhat to 2 CPU cores w\ HT enabled).
Setting CPU core priority alone also accomplishes this as well from a percentage standpoint but using priority also has the advantage of making this application across all available CPU cores as when CPU priority is used Windows is instructed to give the AHIII process a higher priority of CPU processor time regardless of which CPU core(s) AHIII's threads are assigned. I found this to be very useful and much more flexible w\ the game running, especially when the game usage got very heavy & the AHIII thread counts were very high & the performance as measured between running AHIII using CPU priority alone was overall better than running AHIII using CPU affinity alone so advantage goes to using CPU priority over CPU affinity w\ AHIII in either Dx API.
Setting both CPU core affinity & CPU priority will accomplish gaining CPU core processing time from a percentage standpoint but I found is really kinda counterproductive in nature as if using CPU priority w\ AHIII this kinda makes CPU core affinity somewhat a moot issue as I could set a lower number of CPU cores using affinity but since CPU priority will be assigned on all available CPU cores anyway that Windows assigns AHIII game threads to this will essentially cancel out CPU core affinity assignments anyway as the game will get a much higher CPU core processing time increase when averaged across all the CPU cores in use relative to the total percentage of CPU core processing time gained using the lesser number of CPU cores designated until you approach near the full number of available CPU cores using affinity so I didn't see any real advantage to continue to do this.
So I've settled on assigning CPU core priority alone w\ all 6 CPU cores in use for AHIII going forward. Running usage on my box still shows that none of the logical CPU cores are getting used w\ HT enabled (all 6 of them stay parked regardless of CPU load....which says that it isn't high enough for Windows to unpark any of them) and so I will keep HT disabled.
So if you don't mind, which software programs are you using to assign CPU priority and what's your take on any that you might recommend to try?
I'd appreciate any insight you could give..............
:salute
-
I understand what you say, or at least I think so...
Just out of curiosity I opened Task Manager on the graph tab, I suppose the graph tells more about trends than the once in a second changing digits. There was very little going on, so I started AH. Immediately core #0 jumped to the top, #1 showing more modest action. So, it might make sense that setting #0 in a supporting role while favouring the other cores might balance the load. Then again, I know absolutely nothing about coding so I don't know if it were even possible. Also, as Skuzzy said, picking cores might not work for all computers. Plus the multitude of supported Windows versions not to mention what their updates might change in the operating system behaviour... Too many variables, IMO.
Agreed, not every machine will pass code the same, even where they have identical hardware and OS.
Thing is on my other machine, I could not get what appeared to be help, by playing with core affinity. Here tho I am isolating AH from 90% of all else, setting priority to high and using a frame rate cap instead of v-sync. These 3 choices do seem to make a difference.
Since I am running a 60Hz monitor I feel this arrangement helps to kill display input lag and focuses my pc directly on AH. I would luv to try and tune pc to servers tck rate but I think HT does not want to give up intel, I did ask.
I do better in HO's and and slight turn shots. I can see the impacts now, whereas before it was iffy. My bullets may have been hitting but v= sync frame correction was cutting things off. The ability to see hits is important considering the enemy position is constantly being updated by server.
I have often said that offline mode is much more spontaneous than online. Offline, all is on your pc realtime, online tho always felt different, always waiting for the server update and v sync appears to make the game ethereal, like fighting a ghost. Now its diminished somewhat, not gone, but less ghosty.
If you listen to the game chatter, the biggest complaints, besides getting HO'ed, is peeps saying how can they be missing shots or getting killed with single hits. I believe some of this is like what I was suffering, I have lessened it enough that I wanted to share.
:salute
-
Hi Pudgie,
I use
https://www.prnwatch.com/prio/
I use the 64-bit version.
I works for me, but I use only prio as my problem have been trackIR lagging only.
You can set prio, affinity and also I/O (that I have not tested at all) permanently.
-
Hi Pudgie,
I use
https://www.prnwatch.com/prio/
I use the 64-bit version.
I works for me, but I use only prio as my problem have been trackIR lagging only.
You can set prio, affinity and also I/O (that I have not tested at all) permanently.
Thanks, Save.
What are your thoughts about Process Lasso as well?
Thru all my reading up on this Prio and Process Lasso were the 2 most recommended ones to use.
I understand that Prio mostly works in the background thru the Windows context menus and Process Lasso uses a GUI.
The reason I'm asking is that I was looking into a way to access the rest of the running processes that may have been written to report to Windows that they need normal CPU priority but realistically will operate at a lower CPU priority level w\ the intent to provide even more CPU processing time for the desired running process\processes to use. Now reading up on Process Lasso the creator of this utility states that they use an algorithim that will actually automatically do this to maintain 100% CPU responsiveness for any running process\processes that have high CPU priority assigned and do it on the fly as long as those processes need the extra CPU processing time then when the high priority process\processes are done then all is automatically set back to normal as initially reported to Windows.
This does indeed catch my attention and curiosity.....was wondering if you've already tried this utility as well and what was your assessment of it.
Thanks for the response!
:salute
-
Hi again,
I have not tested process lasso, so I can not compare.
Prio have a service running to be able to change the settings, it's very straight forward, right-click in the application and select the settings you want, check in permanent if you tested that the changes works.
-
Hi again,
I have not tested process lasso, so I can not compare.
Prio have a service running to be able to change the settings, it's very straight forward, right-click in the application and select the settings you want, check in permanent if you tested that the changes works.
Thanks for the feedback.
:salute
-
Dos AH3 use or benefit from these in the NVidia profile options?
Thread optimization
shader cache
-
so I added the games AA option into the mix with the cards forced maximums, wow it made a difference.
dx 11 version still hitches
dx 9 beautiful
for those using 60 hz refresh displays. turn off v-sync of any kind, game, card. use a frame limiter to cap your video cards frame rate to 60. ie: MSI Afterburner
I do 59Hz refresh rate and 59 frame rate.
This will prolly not benefit a 144HZ display, G sync or F sync but I noticed a difference with my 60Hz, HDTV.
:salute