Last post on stutter from me! - DX9 Vs DX11 differences explained!

[Vulcan]

Also, the connection has changed to Frontier. I've seen other ping issues with them besides what Steely describes. At times I have tracert the IP and it is smooth until it hits TX and then times-out and coughs blood, yet the game plays.
Then you have did any settings change?

See this is another problem. A ping is a pretty basic connectivity check. But it is not a truth.

Now first off I am not saying your pings are invalid. I just want to help people understand what a ping really is.

Ping operates on the ICMP protocol.
HTTP (web) operates on the TCP protocol.
AH operates mostly on the UDP protocol.

When you ping an HTC server the process that responds is complete different to the process that manages the game UDP connectivity.

A traceroute is basically pinging each hop in the route. Now here's where it gets interesting:
 - ping/icmp responses are not required to run an HTTP/web server, or even a UDP game
 - same goes for routers, they are not required to respond to ping/icmp packets (kinda)
 - the process that responds to ICMP (pings and traceroutes) can often be give very low processor priority AND low network priority. Some routers will drop pings if they are under load but will not drop TCP or UDP packets.
 - so sometimes you see packet loss or delays in a traceroute but there is no impact/problems on tcp/udp traffic

There is a method for doing a 'ping' over UDP, it used to be know as unix ping. I could be wrong but iirc the netstat chart in the game itself uses ping over UDP (I vaguely remember discussing this with skuzzy).

Also steely, I am not wanting to argue with you. All I am is offering my skills and networking knowledge to help you diagnose the issue.

[Shane]

If you change ISPs you don't typically change your equipment. Unless you have something to suggest that at the time HTC changed their route/firewall/switches as well as ISP.

It gets back to my previous comment about how their ISP and your ISP do peering, and the problem with peering is you can please most of the people all the time, but not all the people all the time.

Do you recall how many were having issues with general connectivity (I wasn't) after the switchover? HT even posted some suggested fixes for some of them, "if you're using yadda something, then something, something." I know it sat on the home page as an announcement for the longest time, lol, and now it's apparently archived, or at least inaccessible to us via the home page.

It wasn't related to microwarps, tho'  -  I think a fix of some sort may have created the microwarp situation.

HT did have this to say with the initial announcement on the forums. "The Ip's will also change for all our computers.  So it could take a while to propagate"

https://bbs.hitechcreations.com/smf/index.php/topic,409305.0.html 

But the microfreezes weren't present until maybe a few days, before I first remarked on it 6/7/24.   Something happened within that timeframe that affected many of us.

[steely07]

@Vulcan, I know you're not, your suggestions are useful, keep em coming

[Shane]

Shane, I know you're a supporter of the theory, you might have been one of the first to mention the 10 min timing, if you've got a spare 12 mins, grab a screenshot of your netstatus showing the spike, that'd get us further away from it just being my systems / in Australia etc etc.

I did the thing, was in tower the whole time, every 10mins on the :09 mark... too lazy to post the screenies here via postimages, but posting the raw .bmps in the films/screenshots forum here:

https://bbs.hitechcreations.com/smf/index.php/topic,412096.msg5465571.html#msg5465571

[Animl-AW]

We'll, you guys are more into IT than me. Back in band camp I knew quite a bit, but that's long forgotten stuff now. So I just take your word for it and see if it lines up on the PC end.

I think it may be hard to get others to try it because they can just pop up DX9 and not think about it. They have no interest in figuring it out, they are glad you do it instead.
Nerds like us just like solving issues and tend to get fixated.

DX11 is important tho, from what I understand, is what Steam uses. Seeing that kinda stuff as a noob can give a bad impression. And of course VR, but I rarely, if ever hear, someone with VR complaining. That's a curve ball.

I don't think there's any true argument here, it's hashing out slightly different input.

You should hear 3 sound guys go at it. We're all right, and each one thinks they are smarter than the one standing next to them. Except me, I stand back and watch because I've seen this repeat 10k times.
Thing is, debates usually end somewhere in the middle.

Dad says it worked for his system, but what's common between him and others ties it in. I did leave out Hardware Acceleration and added matching fps max to refresh rate. The rest is common between the two. his gaming settings are almost identical to mine.

The IT stuff absolutely makes sense, because even though vast differences from tweaks, we still haven't hit 100% And that leaves IT open and worth looking at.

[steely07]

Shane, thank you, at last some (more) data from a machine external to the bunker here , and more replication and confirmation, excellent!

Any interest in a tool to capture packets and create a connection graph?, you'd have to install wireshark and python, but it'll run in the background collecting packets, and then make a similar graph to what i've posted, I'm guessing not but throwing it out there

[steely07]

I had another thought last night while trying to fall asleep, which I sometimes wish would stop happening, there is a simple way to disprove my theory, so I call on Animl and Vulcan to assist!

Animl, your last post mentioned you'd not been having regular freezes, and Vulcan, you've said you're not getting any, but if this theory is correct, everyone using DX11 is getting them, because the theory says DX11 goes to a port that's configured with a timeout.

So I'm appealing to you both, do the 10 min test, log on, note time, watch net status for 11 mins, you'll either see one, and we can add more weight to this theory, or quickly prove it's incorrect.

As we saw yesterday, Shane (and I) can reproduce the issue at will, will you assist with this small test, pretty please?

Animl, you've been a champion of trying to get to the bottom of this for a long time, and have gathered a lot of excellent data, and done a lot of testing, please spare me 11 mins out of your day!, and Vulcan, you've made a lot of excellent points, I really appreciate your knowledge and experience, we can put this to rest today if you can run DX11 in the MA and not get a spike on netstatus between 10 and 11 mins after landing in the tower.

I invoke the spirit of ANZAC Vulcan, you have to help hehe

Salute!

Steely

[Animl-AW]

I had another thought last night while trying to fall asleep, which I sometimes wish would stop happening, there is a simple way to disprove my theory, so I call on Animl and Vulcan to assist!

Animl, your last post mentioned you'd not been having regular freezes, and Vulcan, you've said you're not getting any, but if this theory is correct, everyone using DX11 is getting them, because the theory says DX11 goes to a port that's configured with a timeout.

So I'm appealing to you both, do the 10 min test, log on, note time, watch net status for 11 mins, you'll either see one, and we can add more weight to this theory, or quickly prove it's incorrect.

As we saw yesterday, Shane (and I) can reproduce the issue at will, will you assist with this small test, pretty please?

Animl, you've been a champion of trying to get to the bottom of this for a long time, and have gathered a lot of excellent data, and done a lot of testing, please spare me 11 mins out of your day!, and Vulcan, you've made a lot of excellent points, I really appreciate your knowledge and experience, we can put this to rest today if you can run DX11 in the MA and not get a spike on netstatus between 10 and 11 mins after landing in the tower.

I invoke the spirit of ANZAC Vulcan, you have to help hehe

Salute!

Steely

Well, I did last night. I remembered  as soon as I got a stutter last night. Unfortunately, it didn’t happen again.

I think you may misunderstand, I am not discounting your findings at all. They make sense. I can’t change arena ports. It just may not be sole issue.

The Window settings may just be easing the burst. Did you try them? It doesn’t mean it’s the true fix, they could just be a work-around. But the would matter if the port issue isn’t changed.

I will say, the game is smoother with the settings.

Still HAVE to ask is why do other games have the same issue? Some of my verifying settings come from other game forums like MSFS.  I got the same reaction in AH.

This makes me stand back and think how many issues are we dealing with. Two different types of stutters and port issues?

One common center point between the two is DX. DX seems to be the base of both concepts?

At this point, you MIGHT just call HT, tell and show your findings and get input from him on it. I’m sure he’s as interested as me on finally killing the stutter issue.

I will try network again tonight

[steely07]

Yes, I did try all the changes suggested by Dadtallica, I still see the issue, you're a great candidate to test, as you've been working on it for ages, and have done some excellent troubleshooting, if you do the 10 min test, using DX11 in the MA, and no spikes between say 9:30 and 11:00 mins, I'd say I'm on the wrong track, and I'll just put up with it, or use DX9 (although I'll always be annoyed I'm not on the "latest version"

If you do see a spike in netstatus, and you have the time, getting a second one 10 mins later would be even better proof! (I hesitate to ask for 3, but if you can, that'd be incredible!)

If you're amenable, grab screenshots, but a post confirming or denying is just as good

As for other games, I can't say, but MSFS2024 is fine for me, I think going narrow on this and just testing one thing will shed more light that casting a wider net, if you and Vulcan confirm, then we could get others also, or if not, we're probably stuck with it.

I would say, the DX9 V DX11 exe issue really seals the deal for me, both exe's connect to the same physical server, using the same network path from your machine to HT's servers, the networking infra between your machine and HT doesn't know about the version of directx you're using, it just moves packets around, and we know for sure (through my testing, and from the players confirming and recommending that if you don't want stutter, use the DX9 version) that the DX9 exe works fine 100% of the time

I'm happy to be wrong, but it needs just a bit more confirmation, then we can say to HT, that all the steam users are affected, as there's no DX9 exe in the steam install (is what I've read anyway), and that every other player using DX11 exe is affected, if they know it or not.

Or not!, but we need more confirmation using a simple test anyone can do with their eyeballs and a stopwatch

[Animl-AW]

Otay.

Tested DX11, using AH Network window.

I'll use it because you confirmed it with an outside 3rd party app, which eliminates the Network window freeze concept.

I still want to see what Windows OS is doing at the same time, if it's a global stall, probably on performance monitor, which I have not done yet.

I got an upward spike on the 9-10 min mark, same time as the micro-stutter.

Basically my suspicions were confirmed.

I say 9-10 min because I was glancing.

These settings seem to minimize the graphic reaction. If you're not staring at the screen looking for it you will miss it, it's that fast. It's nowhere near the disturbance, but it does in fact exist on the 9-10 min mark.
When we say we just get a single stutter, it may be safe to say it's the one we noticed, because they are so fast now. Before they are 1-2 secs, now it's about .5 sec. It's digesting the data faster. And an illusion has taken place that it is gone.
 
It made no difference if the clipboard was open or not. Of course I had to close the CB to test that, so it was safe to assume on the 9-10 min mark I had that spike too. These are types of things that happen on a low-end machine.

Confirmed
The stutter happens same time as the spike, every 9-10 min. This is old news.
It makes no difference if Clipboard is open
The spike does not freeze in Network window. The window flows smooth right through the spike.
The settings are good for gaming, but not 100% on this issue. Basically, as I suspected, after reading your post, it's just masking the still existing issue. If some are good with it then kudos to them. There is a difference.

The settings are not wrong for a streamline running PC and minimizing the pain, they are not the fix. This is what I was referring to, not that your research is wrong.

You're now at a point where only one person can answer your remaining question. IF there is another port available, can can it be changed and test it to confirm what you found.

[steely07]

Fantastic!!, hoping Vulcan has a spare 20 mins or so , 2 spikes at 10 mins apart like Shane and I (and I suspect everyone using DX11 is) would confirm it in my mind

Thanks for spending the time!

If you notice one in normal gameplay, just peek at netstatus window, you have about 45 secs before it scrolls off, if you note the time, you'll get another one every 10 mins after that, if theory is correct, and I'm now 99% sure it is

PS : Anyone reading can play along, more reports the merrier!

[Animl-AW]

We had similar convo when the service was changed. It's been well established the stutter happens during the spike, last yr and now.

[Shane]

Shane, thank you, at last some (more) data from a machine external to the bunker here , and more replication and confirmation, excellent!

Any interest in a tool to capture packets and create a connection graph?, you'd have to install wireshark and python, but it'll run in the background collecting packets, and then make a similar graph to what i've posted, I'm guessing not but throwing it out there

If it's free and not too complicated, I don't giving it a go.

[Vulcan]

OK testing it now.

Yes a minor spike at 10 mins and 20 mins.

There were 3 ports operating to 47.190.119.207:
 - UDP 5110 (very tiny data size, tx/rx 1.6kb/1.6kb
 - UDP 5117 (looks like this is the game data) tx/rx 441kb/579kb
 - TCP 5111 tx/rx 224kb/100kb

I also see a connection to 47.190.119.209 on tcp 3011.

Firewall logs showed no dropped packets.

All packets are tiny in size.

So yes I do see your 10 min spike, but no I do not experience your stutter issues.

I'm running Windows 11 on a i7-7700k, GTX4060, an Intel i219-V ethernet adapter. I have 1Gbps fibre, and run a Sophos XG firewall.

What kind of ethernet adapter do you run and what is the driver?

edit: just had a middle of the night thought. I didn't see any ICMP traffic so the netstat test is encapsulated in one of the connections above. Secondly it's entirely possible the 10 min spike is resetting/restarted a UDP ping (i.e. the built in ping in netstat may have a 600 lifecycle).

[Shane]

Fantastic!!, hoping Vulcan has a spare 20 mins or so , 2 spikes at 10 mins apart like Shane and I (and I suspect everyone using DX11 is) would confirm it in my mind

If you notice one in normal gameplay, just peek at netstatus window, you have about 45 secs before it scrolls off, if you note the time, you'll get another one every 10 mins after that, if theory is correct, and I'm now 99% sure it is

PS : Anyone reading can play along, more reports the merrier!

And you just got confirmation from both Animl and Vulcan...