doveman

It seems quite clear from LatencyMon that the network components are causing the high latency. ndis.sys - 1.23ms NETIO.sys - 0.56ms, tcpip.sys - 0.33ms. The next highest is storport.sys - 0.20ms I don't know if it will capture anything, as I wasn't running LatencyMon at the same time so couldn't see what was happening but I left the trace running for three minutes before capturing it, so maybe it will show something helpful if you'd be good enough to take a look. https://drive.google.com/file/d/0B1fDI89phEESSC1rZzlBMnFUeUE/edit?usp=sharing

Just made the move from AMD to Intel and it looks like I've got the same old problem with the network components of Windows causing latency problems. Not sure how to make a trace though, as I know I mustn't have LatencyMon running whilst I do that but without it I can't tell when it's spiked and I should stop the trace. In these screenshots, it spiked after about two minutes but I've restarted it and now it's been running for three and only hit 417us (which is still too high for my liking but LM says it's OK).

Yeah, my mistake. Although it sort of was 'hard coded' before I uninstalled/reinstalled the NIC, as I couldn't change it from .67

Thanks for that submix8c but I'm happy to say it appears to be fixed after uninstalling and reinstalling the NIC. I have to use a hardcoded IP address as my RPi XBMC box looks to 192.168.1.64 to populate it's library, so if one OS uses a different IP, XBMC can't find any of my movies. I can tell you that the hostname in both OS is set to X4 and I don't have any need for it to be different between the two OS. My router is rather crap (BT HomeHub3) and doesn't give me any useful information, in fact it doesn't even show my PC as connected at all and only seems to show devices using DHCP (i.e. my RPi), which isn't very helpful to say the least!

Yep, setting it to 192.168.1.254 and it starts working, albeit with IP .67, ignoring that it's set to .64. Whether I boot to OS2 again or OS1 first and then OS2, it's lost the gateway again each time I boot. I'll try uninstalling the NIC and reinstalling it, maybe that will fix it.

Yeah, both Windows have the Computer Name set to X4 but if anything, I would have thought this would make it use the same IP, not ignore the settings and use a different one

I have a dual-boot Win7/Win7 and whilst one is fine, every time I boot into the other (newer, cleaner) I have no network connection and checking the settings shows that the gateway is blank and I have to set it back to 192.168.1.254. It's also ignoring the IP setting, which is set to 192.168.1.64 and is currently using 192.168.1.67. Anyone got any ideas how to fix this weirdness?

Having tested some more, it seems I can boot from the ISO (and other ISOs) but not if the USB VMDK is connected to the VM as then, whether I'm booting from the ISO or the USB, it just gets stuck with the spinning circle animation forever.

I tried the setextradata command to try and test the VM with an ISO, instead of USB but it didn't help. Then I read that Vbox has to be restarted after using that command, so I tried that and got: "Callee RC: CO_E_SERVER_EXEC_FAILURE" and when I relaunched it still wouldn't boot from the ISO. So then I closed Vbox and ran the setextradata command again. I stil got that error when relaunching Vbox but the Win8 ISO actually boots now The USB doesn't give the C4 error anymore but just goes black after the loading circle animation shows up but that's not surprising as it doesn't work on baremetal either, so I'll have to experiment with some ways of building the USB.

Gotcha, thanks. I'll start a new thread for my brother's problem.

Thanks, I would try that but I've now tried to boot it on baremetal and it throws up much the same error, although it specifically refers to winload.exe as missing/corrupted. I have installed from the iso before though, so I know it works that way but I want to install it from USB now.

It didn't happen when I made this trace and he says it doesn't happen all the time but frequently enough to be annoying. I thought the trace might show something that's happening when resuming that could be causing it anyway. Is there a batch file I can make for him that will collect some useful files after he has to hard reboot? He's also finding his PC is on in the mornings after putting it into standby before going to bed (I got the impression that when this happens he is generally able to wake the monitor), so I've made him a batch file to run powercfg -lastwake >>log.txt, so he can send me the log so that we can try and find out what's doing that.

I'm trying to make a USB from my Win8AIO ISO but after creating a VMDK for the USB drive, adding that as a SATA HDD (port 0) to my VM, with a VDI as a SATA HDD (port 1), after the initial Win8 splash, it says: "Your PC needs to restart. Error code 0x000000C4. Parameters: 0x[lots of 0]91" I tried with QEMU and it does the same thing, albeit it takes a lot longer to get there! Any ideas what I need to change? It might even work OK on real hardware but it's a pain to test on that and have to keep re-doing it if it needs tweaking, particulary if any problems don't appear until half-way through the install, so if I can confirm it works in Vbox first it would be nice.

My brother's having a problem where his PC will wake from standby but the monitor won't come on, so he has to hard reboot. I recall having the same problem with my PC a while ago but I can't remember if I ever found a cause or fix or if I just reinstalled Windows in the end. He's up to date with Windows Updates except for IE11, which won't install properly and just ends up restoring/undoing the Update after a couple of reboots but that's another problem. I'd be grateful if you could have a look at his trace for me, so that we can hopefully fix this for him. https://drive.google.com/file/d/0B1fDI89phEESenRHQU41aWlHY2s/edit?usp=sharing

Hmm, seems to be OK (the clean HDD Windows install) tonight, no stuttering or audio cut-outs whilst playing the same game, although there was one cut-scene that played in it's entirety with no audio but that's a different problem I think. I'll keep an eye on it to see if it plays up again.

I've been having various problems with my system, even now I've got an SSD, so I've setup a clean Windows just for a few games (on the HDD as I don't have room on the SSD for two Windows installs as well as my games) and at least it doesn't crash like the SSD Windows but I'm having strange audio dropouts and stutters when playing games. I checked with LatencyMon and that showed some spikes for highest measured latency but DPC Latency Checker doesn't show any problems, so I'm not sure if the problem's latency related. Should I run the game and start an xperf trace and then stop it after I see/hear the problems?

and also from this: http://www.datapro.net/techinfo/usb_3_explained.html it seemed like they would work the same, but you raised an interesting point. According to the actual datasheets: http://www.mouser.com/pdfdocs/HiroseZX62Datasheet24200011.pdf http://www.mouser.com/pdfdocs/ZX360Flyer.pdf it is however confirmed that one half of the micro-USB B 3.0 socket/receptacle is compatible with the micro-USB B 2.0 plug. jaclaz Cool, thanks for looking into that and confirming it. I can at least get my brother to test with a micro-USB B 2.0 cable now, of which he already has a couple, so no need to buy a new USB 3 cable unnecessarily

Hmm, that's a bit confusing as the first link you posted http://support.usr.com/education/peripherals0.asp specifically states that the micro-B connectors are not compatible between USB3 and USB2 and the diagram comparing the two appears to show that the pins (or whatever the white space indicates) are on opposite sides.

I was thinking that maybe ArmA3 just stresses some component more than general apps or even other games such as DCS World but I'm not sure that the RAM would be susceptible to this, maybe just the CPU and GPU. I do lock the CPU to 3.5Ghz by switching to the High Performance power profile before launching games but that's not peculiar to ArmA3. I also sometimes overclock the GPU from the default 800/1250 to 850/1300 but did seem to have more issues/crashes (not total signal loss though) like that, so I generally play it safe and don't bother anymore when playing ArmA3 online, although I still do overlcok for DCS World which doesn't crash. As I mentioned before, I scrypt-mine at 880/1350 without any problems, so I don't know why 850/1300 should make ArmA3 more likely to crash but it did seem to. The only thing that's really unique to playing ArmA3 is running TeamSpeak3, with it's plugins and overlay but the other people I'm playing online with all use the exact same plugins and whilst not all of them use the overlay, some do. As TS3 is an audio-related program and the errors I found in the log relate to Razer Surround crashing maybe that's important. Even though I wasn't using Razer as the sound device in TS3 (as I couldn't hear anything when I tried), nor as the default device in Window, so ArmA3 was using the Realtek device, Razer was clearly still running to be able to crash, so I'm wondering if it crashing could have somehow brought down the system, resullting in the loss of video signal and incoming audio breakup and no outgoing audio. I was happily using Razer until quite recently though, so whether the version I'm trying to use now is newer and has issues or it's the same version but something else has changed which is causing it not to work anymore I don't know but it isn't working for me anymore, so I've uninstalled it as there's no point having it installed and it might be causing problems. I tried to contact support to fix the problem but their stupid website only offers a form with a 500-character limit, which is not enough to describe the problem. So it may be that uninstalling Razer and/or loosening the RAM timings will stop the problem reoccurring. My group only play ArmA3 on Thursdays, so I can only really recreate the same situation then. I could play online on another server that uses TS3 to test but they won't be using the same TS3 plugins and the signal loss crash only happened after 1-2 hours, so it's not a quick test I can do and whilst it's obviously annoying (for everyone else as well) to have it crash in the middle of a mission, at least I'm having fun with my friends until then and not just spending 2 hours trying to trigger the crash.

Sure, thousands or hundreds of thousands of people play that game on hardware EXACTLY identical to yours, with EXACTLY the same OS (and it's updates and tweaks and EXACTLY the SAME services running in background) using EXACTLY the SAME (outdated) videocard drivers , it would be very improbable that they don't suffer form EXACTLY the SAME issue. jaclaz Well EXACTLY It could be any one of those differences that causes a crash that others don't experience and not the fault of the game or app at all, which is all I said. Besides, there ARE lots of people using the same GPU, same CPU, same OS and updates, so I stand by my point that if there was a serious bug in the game that caused total loss of signal with my graphics card (or drivers), it wouldn't affect only me. It may be that I have a badly written other program/service running that triggers this failure when ArmA3 is running but that's not ArmA3's fault either. Anyway, what about the point I raised about it being somewhat unlikely that a bug in a game could cause the PC to fail to work after restarting or even a cold boot but only after removing power completely?

It might be my simplistic thinking, but if the same app/game is running whenever you have a problem then I would at least consider that it is the app/game that IS the problem. Cheers and Regards It was my first thought too but it seems hard to see how a game/app could cause the motherboard/GPU to fail to work on rebooting until after cutting the power from the board completely, not to mention that I was playing online wih other people using the same version of the game and apps and none of them experienced any similar problems and if the game had picked up such a serious bug in a recent update, I imagine there would be lots of reports about it

Yeah, thanks for the pointers, I'll do all that (or get my brother to anyway) before spending any money. I don't have another caddy though so can only try it directly connected to SATA, which won't allow me to eliminate problems with the PC's USB ports but I guess the fact that other devices do work in them rules that out.

I don't think my motherboard has the ability to beep, at least I've never heard it do so but it doesn't appear to be doing anything differently (like fans running at full speed which happens if I mess up some BIOS overclock setting and have to clear the CMOS). Hopefully my PSU is OK, as it's a good quality Antec CP850 which has hardly been stressed running my 300W system (that's probably a generous overestimate) since I bought it. Regarding the video card drivers, I was playing DCS World for some time the other day with no problems, so it would be rather strange if only ArmA3 (which has been running both times I've had this crash) had an issue with the drivers (Catalyst 13.10 for reference), which I haven't updated recently. I should also mention that it has no problem mining with cgminer, even overclocked to 880/1350 (still at default 1.1v), which I'm sure loads/stresses the GPU far more than ArmA3 does, as that's rather CPU limited and probably only uses about 60% of my GPU (and I tried running ArmA3 at stock 800/1250 after the first time it happened, only for it to happen again about an hour later.

Thanks. That's a shame as I was hoping he might be able to use a USB 2 micro-B cable to test the drive and confirm the cable is faulty but they're not interchangable (I used to have the same drive in a USB2 caddy, so I know it works fine without the extra power that USB3 can provide with the extra lines), so I'll have to spend £4 buying another cable to test if it's that, which probably means I'll end up having spent as much as I would have if I'd bought a slightly more expensive caddy I was considering but as this one had good reviews I figured there was no point spending more

Well it's hard to say if it POSTs as there's no signal to the monitor. I do hear the HDDs whirring up but I guess they'd do that even if it wasn't completing POST (I'm booting Windows from a SSD so I can't tell from drive activity if it's booting into Windows). I've only experienced this problem after the video signal has disappeared and the other problems have happened in Windows when gaming though (not had it happen so far when a game hasn't been loaded, at least in the background) and have found I have to completely turn off the PSU to clear it. I've never had any problems (touch wood) cold booting / restarting otherwise, so it's almost like the crashing sets a state that can only be reset be removing power from the board. I wondered if it might be a RAM problem, so I loosened the timings from the specced 9-9-9-24 (which I've been running at for several months without problems I should note) to the automatic 11-11-11-29 and then I ran memtest overnight and that had done 2 passes without any errors by the morning, so that seems to be fine, at these timings anyway. If you think it could have been the tighter timings that were the cause, I could just monitor it for a while running at the looser timings and hope it doesn't happen again but if you think that's an unlikely cause, I should continue trying to diagnose the problem.