Quote from micha1980de :

@nick

i was wondering what kind of hd you where using, because i always get a curve like this: (fast test of a Ide).
is it a S-Ata drive, or does your hd-infrastructure limit your bandwith to your drives, or are they just plain fast?
Because even the tests of the c't (well known magazine here in germany) get's results similar to mine...

regards

I changed those snapshots to .png's for easier viewing. You can see that my 1st HD is like yours.

HD1 is a Seagate ST380011A 80GB IDE
HD2 is a Hitachi HDS728080PLAT20 IDE

So not sure why my second HD's speed doesn't drop off.

I did a surface scan on it yesterday and it does have an error on it, but I don't think this is related to the freezing because I've had the freeze even with that HD unplugged.

EDIT: Actually, I just noticed something must be limiting the data transfer from HD2 because even at the end of the test on HD1, the transfer rate is higher than the rate on HD2.

There are test results charts on HDTune main site.
Performance seems to be a bit low for both HDs. 40 pin cables?
The computer is enough recent to support 80 pin cables. If it's so, I'd replace them and test again. If possible, I'd use a separate cable for each HD, at least during the test, setting both HDs as master on primary and secondary channel, and setting other devices (CDs, DVDs) as slave. Also check that HDs are properly set and wired: the master must be on one connector at the end of the cable, while the slave goes on the connector roughly in the middle of the cable.

Correction: one hard disk should have an 80 pin cable. However I consider a minimum throughput of 1.8 MB sec extremely low.

A good way to have more consistent results is (I guess I'm a monomaniac) testing from a preinstalled environment such as an UBCD 4 Windows, which already contains HDTune but runs from the CD, thus minimising interfering HD access from OS and apps. Sorry for the multiple edits to this post, it's been a hard day at work and I'm a bit confused

Thanks. I changed the power connection to HD 2 and re-ran the tests (see below). Not sure if it was that or something else but the output for HD2 looks much more normal now. Comparing with the results charts, the max. transfer rate and response time look right. I do always get those blips at the start but I did notice a couple of charts on the HDTune site with them too, so not too worried about that.

I checked and I do have an 80 pin cable which goes to the 2 HDs and a 40 pin cable going to the CDROM. I didn’t know that about the master/slave positions on the cable – I will check it tonight.

Now a general update:
I read on this forum somewhere that someone had problems because too many devices were connected to the same power line on the PSU (there are 2). I checked and had my 2 HDs and my GFX card on one, and only the CD and floppy drive on the other. So I rearranged that so the GFX card is on the second line, and removed the CD-ROM. Didn’t work though, I had a couple of freezes last night.

One thing I noticed is that since I’ve been trying to fix it the freezes have been less frequent so something I’ve done has affected it. But last night I returned some settings (hardware and software) to how they were and had 2 within half an hour, which actually is good because it’ll enable me to test things more quickly.

Anyway I did learn a couple of things from last night’s freezes. Firstly I looked at Process Explorer immediately after the freezes, and definitely there was no sign of any activity at the time of the freeze in any of the programs running (including LFS) so that tells me that it’s not another program hogging resources and is more likely to be hardware related. Secondly, on the second freeze I moved the wheel while the freeze occurred and saved the replay which shows that the wheel input was not registered. In the replay my car just looks like I’m not steering, you can’t see the freeze. The replay is attached. The freeze occurs at the end of lap 3, on the start/finish straight. I’m sure in previous replays of freezes I’ve sometimes seen my car disappear though, but that could have been an online replay –I’ll check tonight. Edit: Again, there was nothing reported in event viewer at the time of the freeze either.

One more thing:

I have some LEDs on the front of my machine (look like a fake cathode tube!) and they flicker slightly. I think they did this even with my old PSU, but it made me wonder whether the PSU output is not stable, but perhaps at a higher frequency than my voltmeter can detect?

So my current thoughts are:
Problem caused by my wheel
Problem caused by inconsistent PSU output
Problem caused by on-board sound (some people have reported freezing problems when both onboard and external sound are enabled. I only have onboard sound but I wondered if this could be the problem).

Now the HD charts look right.

Onboard sound: it could be, but generally setting the DX accel to basic solves it, so all you can do right now is borrow an external sound card, disable the inboard sound via BIOS (or via mobo jumper) and try. Get your hands on a soundcard with a different chipset, if you can. I don't really have hints about it, I'm no expert in low end commercial soundcards and I never tried LFS with onboard sound, which I disabled and which I will always disable (I use an Egosys WaveTerminal 192X on my PC, never had a single problem, but that's a semipro soundcard not exactly suited for gamers or general use. I just ordered a M-Audio Fast Track Pro USB for my laptop).

Inconsistent PSU output: it's another good option. Do you trust your electrical company, by the way? I'm using an APC UPS right now to stabilise the AC current. I need it because I have lots of spikes in AC current and I don't want my PSU to fail because of them. Besides that, I use a Nexus PSU, so it's not exactly the el cheapo PSU that you can't trust.

About the wheel, I couldn't say anything about it.

Have you tried running your rig with a traditional PS2 wired keyboard and mouse? If you got them it would be worth doing it. I may be a traditionalist, but I ruled them out in favour of a high quality wired USB keyboard (Enermax Aurora) and an OEM Logitech basic wired USB mouse. Reducing the number of USB connections to a minimum could be worth a try.

Hello Nick and all helping racers !

I have the same problem. As I checked your list of the pc's hw and sw components, i realized that the common is the Forceware 93.71. I have a much older gfx card: geforce 4 ti 4200, so maybe I will try to use an older driver. I really appreciate if anyone here can suggest me a suitable driver (w/ dl link ) for this card.
I have this problem since about a year too. A few months ago I reinstalled my Win, to check if it caused by any background process, but the same freezing occurred. (I just installed the hw drivers, and no anti virus sw, no firewall when I tried lfs) I thing my freezes are not as frequent as in Your case, but I usually get one freeze per lfs running. I dropped from one or two olfsl races due to this in the past . When it freezes, the sound is repeating, the HDD lights are on, so the same...
Another ~common thing is I have Athlon system (Barton 3200) on nForce mb.
shortly my system: Asus A7N8X, Barton 3200 (no oc), 2x512M ddr400, MSI nVidia Gef4 Ti4200 (AGP 4x), 2HDD, 1DVD RW, power supply: 400W.

I have another idea, maybe the IDE driver... Nick please check which version You have. Mine is: NVIDIA nForce2 ATA Controller, Version: 5.10.2600.518. Provide: NVIDIA. Maybe I will try to roll back to Microsoft one...

so for me:
- check with _old_ gfx driver
- roll back ide driver to MS drivers... (as I remember, i found this new ide drivers at nvidia about the same time when freezes started, but I was not playing every day lfs...)

I hope we found out the solution.

nope we didn't, since i got a Nvidia N-Force2 Ata Standard microsoft driver running for ages and i have the freezes too.
<<< hopes it's nothing with the memory N-Force drivers or the barton CPU, since you can forget a serious amount of processing speed without them.
And yes, i use the fastest Barton available (3200 Xp+) at nominal timing of DDR400

@Albieg:

To be protected by various net-side irregularyties, you need a PSU thats burst/surge resistant (german standard is 1 kilo-volt for home and office usage, while tests usually exceed 2 kv wich would be "cool" for industrial use) and as a high support time (bigger than 20ms) so bridge out net current-losses due to provider-net-switch acivities.
(notice a room-lighting drop and return to normal, in short timing).
let the search continue.

@kx_mole
how to find out (fast and convienient) what exact version of the N-forced ones i use?


regards
Micha

Hi Micha,

I hope I get it right, and you are looking for this info:
Control Panel/System/Hardware tab/Device Manager
IDE ATA/ATAPI Controllers/NVIDIA nForce2 ATA Controller - properties, Drivers, and here you can see the versions.

But if you use MS IDE drivers, than you don't see "NVIDIA nForce2 ATA Controller" under IDE ATA/ATAPI Controllers. (Yesterday I switched back to the Standard IDE Controllers (MS) (just uninstalled the nForce2 ones), and was testing about 2 hours, and no freezes, but I should test it much more, so I will play today too . )

it's freezing again
so it is not the IDE drivers...
i should check with an old geforce driver, but i think Micha has ATI, so I don't think that this will help...
Hmmm...

well i have Ati X700pro but i have the freezes too, most of the problems could be ruled out, but not the memory (mobo) drivers, yet.

so if anyone here (should) have another mobo vendor and N-Force chipset with the same problem, then the memory tought could be ruled out too...

regards

@kx_mole

I have a VIA IDE controller, and am using the Microsoft driver. Actually my previous GFX card was a Ti4200 - it was a pretty good card, but I can't remember if I had these freezes with it. Probably not but then that was with an earlier version of LFS before these freezes became more common.

Remember I have tried two different NVidia GFX cards, and they needed different drivers so I don't think the problem is caused by the GFX card.

Also, on further thinking from my earlier post I don't think it's my wheel because I've used it on a laptop with no problems.

kx_mole and micha1980de what sound card or integrated sound do you have? I was wondering whether the sound could be the problem because I think LFS uses synthetic sounds (mainly) where most games use only samples. Is that true? It might explain why only LFS seems to suffer from this problem.

It's interesting that we have quite different systems but the same problem.

I am still testing with/without sound at the moment, but the freezes have stopped happening so frequently, so it's really hard to tell if it makes nay difference.

My other suspicion is still the PSU or electicity supply, but I'm not sure why a problem like that that would affect LFS but nothing else.

@nick,

in my observations that freeze problem affects all directX apps.
As for sounds I know lfs uses synthetic sound (mainly).
I have a really old Creative AudioPCI (ES1371, ES1373) (Windows Driver Model) with MS-drivers running, because the Audio2k drivers (for NT 4.0) produce a regular stutter in all apps.
I remember that i've read an article about DX10 and their pro/conns in comparison with DX9, i'm no expert but they said that the sheduling with DX10 was been massively upgraded, while it was a bloody mess with DX9.
So i'd like to hear from a few Vista+ DX10 hardware equipped drivers if they have similar freezes.
And since the freeze occours only once per lfs session (shortly after joining a game, on/offline) i'd like to rule the PSU out because these freezes would be really often.

So long...

OK. So you have had the same freezes in games other than LFS?

I suggest you go down the list of things I tried at the start of this thread. I picked most of them up from other threads on this forum and some people had reported thay they fixed the problem. Most of them don't take long to do.

Then if you still have the problem we need to work out what our systems have in common, there must be something!

I have Creative Audigy2 audio - I disabled the integrated sound in bios - so it is different too.

And I have this 'freeze' only in lfs . Ok, mostly I am playing with age of kings if not with lfs, so not too much experience in other games. Ah and some Enemy Territory too, but far not as much as lfs and aok.

The only common thing I found is the Athlon XP system. But there must be something else I don't think that only we have this kind of CPU...

[ one good sideeffect of all this testings is that i improved my lap time on AS Nat with FZR , this is my 'testcombo'. ]

Quote from kx_mole :

The only common thing I found is the Athlon XP system. But there must be something else I don't think that only we have this kind of CPU...

In the attached file (safe html made with Aida32, rename its extension to .htm) you have my complete hardware info if you want to compare it, I have an Athlon XP 2400+. No problem at all with LFS and Trackmania United (the only stuff I play at the moment). No problem whatsoever with any other game I tried or played. I installed Win XP pro on my system more or less 3 years ago.

Additional info about PSU and fans: I use a Zalman passive cooler on my VGA adapter, and a Zalman fan on my CPU. My PSU is a Nexus NX-4090.

@kx_mole: Thanks. That's quite useful to know your soundcard. In which case I doubt it is my integrated sound that's the problem but I may continue testing it. Problem is it's really hard to drive properly without being able to hear the tyres squeel, so I don't gain anything from the testing!

@Albieg: That's fantastic! I will do the same tonight if I get a chance and post it here.

If anyone else with this problem could do the same that would be great. You can download Aida from the link in the post below.

At a quick glancer it looks like the thing we all have in common is AMD processors. Even then we have different speeds.

OK, I will make the analysis with Aida when I get home (+5 hours from now). than we will see... at least I hope...

Yes, I have nForce mobo. And other (but different) pairs are that Albieg and Micha has ATI, and Nick and me have nVidia... aaand as I see Albieg and me have VMWare (but I have just installed it about a month ago). But Albieg has no problem .

here is the link to aida

http://cgi.zdnet.de/c/?s=41&am ... 393.zip&dld_id=de0DX2

( another idea: maybe we should check the memory timings, when we have this hw info generated )

Quote from kx_mole :

( another idea: maybe we should check the memory timings, when we have this hw info generated )

I generally tend to be pretty conservative with memory timings. I could even be considered an underclocker sometimes, all for the sake of stability. So, if you find my timings to be low that's up to my motherboard being old, and to me being old

86 yrs nice :gandalf: :bowdown:

here is my report:

http://mole.hu/report.html

Quote from kx_mole :

86 yrs nice :gandalf: :bowdown:

here is my report:

http://mole.hu/report.html

Ooops, no, that's a practical joke because of a stupid post in this forum in which a user connected wisdom to age, which is true only to a certain extent. He got bashed by a wise 17 year old kid despite being older.
I'm 36 really. I'll watch the report later (maybe tomorrow) and edit this post if I find something I deem significant. And I'll correct my age

ok, so my hints are: memory timings, and agp aperture size. Maybe I should set it lower, I read somewhere today, that 64 is a good value for it... I will try this first.

OK, so here's mine:

http://members.aol.com/nickaaronson/Report.htm

I don't know anything about memory timings. I tried to change my AGP to 4X in the BIOS a while ago but the option was greyed out.

I'll compare the two and edit the post if I notice anything interesting. Only thing I noticed at first glance was that we have the same Seagate hard drive, but I have tried a new XP and LFS installation on a different one, so it isn't that.

I checked with apg aperture size 64, but I tested it only for 1 hour, so not too much, it was ok during this time. On the weekend I will test it more and more...

NO not the agp aperture size caused the stuttering

Hello Nick and all LFS drivers,

Did you solve the problem yet or not ? I'm encountering the same trouble since a few days although LFS runned perfectly before.

I use both W and X10 version but the above problem (graphic and sound freezing for 5 sec then LFS continued jumping in realtime image) happened with version W. I used to use both version because all replays I saved with version W cannot be opened with X10 version (another trouble).

Another information: I only bought and unlocked S1 from the S2W demo. That's enough for me who is a rallycross driver enthusiast.

I still use an old computer:

Hardware:
AMD Athlon 1600+ - 512mb RAM
NVIDIA GeForce2 MX400 - 64 MB

Software:
Windows 98SE
DirectX 9.0c

I didn't change any drivers but I only added some hardware:

-2nd HD Hitachi - IDE
-1 PCI firewire card for video capture
-1 PCI Ethernet card fir internet (I used USB connexion before)
-The power supply was updated from 400W to 470W.

If you found the solution to remove the graphic bug, please let me know and I will act accordingly with my PC. I'm in the nightmare at the moment and I'm not able to play in such condition.

Thank you in advance and awaiting your early reply ASAP

Mamynast

Hmmm, another Athlon...