ever since i installed my new windows 7 ultimate x64 on my pc terragen is plagued by bucket errors and blue screens and just doesnt really function anymore im fed up:( i have made a dual boot and installed my old xp x64 pro on one hard drive but it has the same problems as well.
what should i do guys im really at my wits end here :'(
i have got about 6 scenes ready to render but they cant even manage the gi process without crashing and some of the scenes are just basic height fields no objects clouds or anything:(.
my last render i had to crop into 6 pieces and really dull down the settings and in my view that is not acceptable and if it persists i will be moving full time over to max and rely on it for rendering as well which is a shame because terragen is my favorite program by far
So you have no problems rendering in Max? Is CPU and memory utilization similar with both apps? Have you done a memory test?
- Oshyan
have no idea how to do a memory test lol ::)
and yes i can render fine in vray mentalray final render and maxwell
vue 8 has problems with the default render (blue screen) but runs fine through mentalray xstream
It sounds a lot more like a hardware or driver issue than a TG issue. I'd start with a memory test. Download and extract the ISO from the following URL, then burn it to a disc and use it to boot your computer:
http://www.memtest.org/download/4.00/memtest86+-4.00.iso.zip
When you start with the disc, it should go straight into a memory test. Let it run for at least an hour and see if any errors come up.
Also, are you doing any overclocking? Any other hardware or software changes recently?
- Oshyan
Quote from: Oshyan on March 21, 2010, 01:03:08 AM
It sounds a lot more like a hardware or driver issue than a TG issue. I'd start with a memory test. Download and extract the ISO from the following URL, then burn it to a disc and use it to boot your computer:
http://www.memtest.org/download/4.00/memtest86+-4.00.iso.zip
When you start with the disc, it should go straight into a memory test. Let it run for at least an hour and see if any errors come up.
Also, are you doing any overclocking? Any other hardware or software changes recently?
- Oshyan
cheers oshyan:) so does this tool fix errors or just finds out what is wrong? also do you have to mount it from a disk?
It just finds errors, does not fix them. If it does find errors, there are really only a few possible causes: bad memory, overclocking, or bad motherboard. Since it avoids all other software, drivers, etc. (by using its own boot environment), it eliminates the possibility of other components being a factor.
You need to burn the ISO to a CD as in "Burn image to disc" and then boot with it in the drive, and have your BIOS set to boot from CD. It will then auto-load itself. You can't use the computer while it's testing.
- Oshyan
mmmh i will try it tonight thanks id do not over clock my pc my mother boad is about 2 years old my memory 6gb is about a year old
Very good advice, Oshyan. I had to do this a couple of times, found faults in memory and had to replace the bars. After that; perfect. Hope it works for you, Wetbanana.
Quote from: Dune on March 21, 2010, 03:22:27 AM
Very good advice, Oshyan. I had to do this a couple of times, found faults in memory and had to replace the bars. After that; perfect. Hope it works for you, Wetbanana.
do you mean you had to actually replace the ram god i hope not i have no money atm:(
The manufacturers usually have a very long warranty against faults, so they'd replace for free if you're sure they're bad. Same with some hard disk manufacturers. The memory I found faulty came from a shop, so he replaced them, the hard disks were returned to Ireland, and I got sent a new one.
cool lets see if it is the ram though i have my suspisions they arnt slotted in right as sometimes it reads 2gb instead of 6gb or 4gb instead of 6gb
also do you think a simple cut polish of the machine would help abit the old rusty things getting a bit dusty lately
Quote from: wetbanana on March 21, 2010, 03:48:50 AM
also do you think a simple cut polish of the machine would help abit the old rusty things getting a bit dusty lately
I bought a cheap compressor just for this Wetbanana. Be careful with the cables and the other things in the case.
You say 6 gb ram... Do you have 3 pieces?. Try it with 2 pieces (with the same settings) and with different ram slots.
Read the manual of the motherboard for the best configuration.
There may be some settings in the bios that you must maybe enable (ı doubt it but just in case) .
You may have a bad HD (it happened to me) . A bad setup of Windows . Try these too .
I have seen strange problems in the past . There may be driver or graphic card issues too .
Kadri.
right i ran the memory test and it came up zero errors i ran 8 passes over night and still no errors so it isnt hardware i don't think. what now guys ???
It sounds to me like the memory controller is bad. These are now built into the CPU I believe.
Quote from: njeneb on March 21, 2010, 05:59:09 PM
It sounds to me like the memory controller is bad. These are now built into the CPU I believe.
mmmh how much would it cost for a new one?
im going to give it a full clean after school tonight and rearange the ram and make sure its sloted in right
The manufacturer warranty might cover the cost. On IBM computers, press F1 during the boot, before the disk starts loading. It varies by the ROM used by the manufacturer. Check their site for assistance.
I think memory controller problems would show up in the memory test.
Have you checked the CPU and System temperatures when rendering? It may be time for a cleanup -- it's good to check this often anyway.
Have you tried a fresh windows install? That way you could rule out the software problems and focus on hardware. The fact that this happens only with 2 programs might suggest such a thing. Also you might want to run some harddrive diagnostic tools. For example if you got seagate HD you might want to try Seatools. http://www.seagate.com/www/en-us/support/downloads/seatools Also if you have lots of dust it might just be the temperature exceeding shutdown temperature. You can see how high it is set when you got to bios
Also when having unstability issues and you suspect your motherboard you should look out for these little buggers. They tend to pop after a year or two in "cheap" motherboards. Ive had lot of these come up when I've fixed computers
(http://img395.imageshack.us/img395/5809/kzgexploded1hx9.jpg)
my cpu temp idols at 61c normally (NOT WHEN RENDERING) is that average?
also my hard drive is Western Digital is their any diagnostic tolls for a WD brand hard drive?
also with the memory tester it says my motherboard name then in brackets ECC diabled does anyone know what this means? sorry guys im an artist not a computer whizz ;D
I can't help with the EEC. I doubt this is a harddrive problem. You would have more problems with other software. You could use Windows Diskcheck. Open the properties of your harddrive. Under the Tools tab is an Error Checking button. This would do the trick.
The temperature sounds ok. 20c is 68 degrees, so (guessing) 61c would be about 180 degrees.
Wikapedia'd EEC. It's an error checking skeem. You may want to enable this. It will slow things down a little.
61c without rendering seems a little high to me. But it depends on the kind of cpu .
Check your fans (Case , mainboard , Cpu , Grapic card etc.) and test the temp with rendering.
I think it will on the 70-80 c hight and thus a little to high. But helping in such a situation from faraway isn't easy.
Do you have any friend who know this kind of things Wetbanana ? Or better ask the seller for help .
Maybe a little cleaning would help as you asked. I hope it will :)
yeah so im going to clean it out tonight ;D see how it goes.
61c at *idle* seems hot to me, but it does depend on the CPU. For a P4 that wouldn't be bad. ;)
Not having ECC enabled is fine as you probably don't have ECC RAM anyway.
Since it passed the memory test, that means it's not likely the memory (or memory controller for that matter), but it does *not* mean it's not a hardware issue. You've just eliminated one possible hardware culprit.
The next thing I'd test is the graphics card. First make sure you have up-to-date drivers, then run a benchmark/stress test app of some kind. TG2 uses OpenGL, so I'd try FurMark first: http://www.geeks3d.com/20100202/furmark-1-8-0-hot-like-hell-edition-available/ and maybe also 3DMark (DirectX).
- Oshyan
Quote from: Oshyan on March 22, 2010, 01:12:19 AM
61c at *idle* seems hot to me, but it does depend on the CPU. For a P4 that wouldn't be bad. ;)
Not having ECC enabled is fine as you probably don't have ECC RAM anyway.
Since it passed the memory test, that means it's not likely the memory (or memory controller for that matter), but it does *not* mean it's not a hardware issue. You've just eliminated one possible hardware culprit.
The next thing I'd test is the graphics card. First make sure you have up-to-date drivers, then run a benchmark/stress test app of some kind. TG2 uses OpenGL, so I'd try FurMark first: http://www.geeks3d.com/20100202/furmark-1-8-0-hot-like-hell-edition-available/ and maybe also 3DMark (DirectX).
- Oshyan
cool will test that out. could it be as simple as out of date drivers?
I recommend prime95 for spotting general stability errors with your CPU, PSU and memory. Recently caught some errors on a friend's comp which ran all the memory tests fine.
http://files.extremeoverclocking.com/file.php?f=103
There's a 64-bit version also in the "Related files" section.
Regarding your graphcis card, I've tried ATItool on testing and stressing GPU temps.
http://www.techpowerup.com/atitool/
If you want all your temp sensors in one place, CPUID HWMonitor is the best tool I've found for that.
http://www.cpuid.com/hwmonitor.php
Not saying it's impossible, but hardware is rarely the issue with these sorts of troubles.
Could be something simple, for example upgrading to a 64-bit os suddenly treats the RAM in a different way or a bad driver. It's always tough to tell what's causing these issues, even on your own-built machine, so the best bet is to run all manner of tests so you can narrow the list down.
Yes there are many variables . For example ; virus , spyware , malware , trojan kind of problems could make problems too.
I doubt the others (spyware ...) but if a virus changed some files it is possible .
I hate this kind of position .
Don't t misunderstand me Wetbanana , i would love to be near your pc and look for the problem . But from internet it isn't so easy.
Sometimes it is a very basic problem . Of course if you know or can see it :)
How much HD do you have Wetbanana? If you have more then 1 try to unplug the other....Check all the cables . Change the cables with others .
Be carefull with the temp of the HD's . They can be very hot too. Not a good thing of course. Unplug other hardware you don't need for testing.
Your PSU can be problematic too or did you buy anything new that is more power hungry ? The PSU could be not up to the demand.
Check the messages of the blue-screen sometimes it can help. For example ; if you see some things with graphic cards related it may be some kind of a card issue (driver or hardware ) .
There's a free program at http://crystalmark.info/?lang=en which can give you your HD temperatures (I believe you have to have SMART enabled -- this is usually done in your motherboard bios.)
Also, when you render, do you use more than one thread? Maybe trying a min/max of 1/1 will show a difference, since the CPU activity will be much less, hence less heat.
Check your cable routing to see if any are blocking fans. Is your power supply rated high enough for your computer?
Unless your case has excellent airflow characteristics, there's a good chance there are "hot spots" or "dead zones". This is where the air doesn't flow well. This can also cause problems. If you have a small desk fan, you can try opening your case and directing the fan in the general direction of the case airflow (like input air in front, exhaust air at back) so direct the fan so it tends to blow in the front-to-back direction. Also monitor the temperatures to see if the external fan has a significant effect.
My CPU is 38c at idle. A single threaded TG2 render takes it to 48c. A multi-threaded render goes to 57c. If I direct a small fan into the open case, the temperature of a multi-threaded render goes to 46c, so that external fan has a real impact. I use the external fan whenever I'm not able to monitor my computer during a long render (like sleeping.)
Of course the good/bad CPU temperature is dependent on the type (of CPU). I have older hardware. But even if your CPU can handle higher temperatures, the extra heat will effect other parts of your system (HD, Memory, GPU, etc.) And the ambient room temperature will make a big difference.
Anyway, with all the inputs from the participants of this thread, you should be able to safely and inexpensively rule out potential problem causes. And as mentioned, it's very difficult to troubleshoot over the Internet.
By the way did you update Windows 7 , Wetbanana ( i think you did, but just in case) ?
Look for the mainboard or-and graphic cards forum for the same problem and new drivers and such ...
right i cleaned all my computer last night pulled out a massive amount of clumped dust that was clogged in the cpu fan and made sure the ram was slotted the cpu temp idles at about 52c now with the computer temperature averaging about 32c does this sound a little better?
also i run two athlon dual core's 3.2ghz so is that the normal temp for this sort of processor?
I also cleared my hard drive and installed a fresh copy of W7 i have not tested it yet for rendering
Looks better for now! I hope you can render your scenes without problem , Wetbanana :)
Sounds better temp-wise, yeah. Are you saying though that you are running *dual* Athlon dual cores? (those would have to be Opterons I reckon) Definitely do some re-tests with TG and see if you get the same problem. If so, try some of the other suggestions in the thread.
- Oshyan
Quote from: wetbanana on March 22, 2010, 06:17:11 PMright i cleaned all my computer last night pulled out a massive amount of clumped dust that was clogged in the cpu fan and made sure the ram was slotted the cpu temp idles at about 52c now with the computer temperature averaging about 32c does this sound a little better?
Wow, that sounds like there was almost sentient life being born in there!
Quote from: tempaccount on March 23, 2010, 04:04:55 AM
...
Wow, that sounds like there was almost sentient life being born in there!
I saw 90c on my P4 ones. It was the fastest moment i reached to the off button on my pc .
The fan was faulty . But the same cpu still works after nearly 5(?) years and can play Crysis too , with an ati 3870 and 4 gb ram :)
( This isn't an attempt at the overdone Crysis joke ... Only for comparison )
I think i was lucky.
well big thanks guys for your help (as usual ::) !!!!)
hoping to get some renders done soonish:)
good news i rendered a picture that before hand crashed after 20min EVERY TIME!!! it rendered in 1 and a half hours with not a single error yay! ;D ;D ;D ;D ;D ;D
the banana is back lol!
Great! :)
Now where can I find it :P
Good News! 8) :)
:)
Great news. :D Now just file that away that you need to give your computer a good cleaning every so often. You can probably figure out when by watching your CPU and System temperatures.
One other thing that would be wise is to buy a static grounding wrist strap and use it anytime your inside your computer case. They are cheap compared to the cost of replacing electronic parts.
my granddad is an electrical engineer he says those wrist straps are useless
Quote from: wetbanana on March 25, 2010, 07:36:07 PMthe banana is back lol!
(http://www.terralights.de/terralightsII/images/smiles/banana.gif)
;)
Your Granddad is right. Touch the bottom aluminum of the case before doing anything. That will ground the static.
Quote from: wetbanana on March 26, 2010, 12:59:39 AM
my granddad is an electrical engineer he says those wrist straps are useless
Are you sure he's not talking about wireless or dissipative wrist straps? I've been away from micro-miniature repair for a few years, but you couldn't get into our facility without using ESD equipment. And if you can work with one hand (try replacing a CPU fan) then holding the case will work.
A mat on the floor which is grounded works much better. It's more expensive and can be hard to set up. But we had this in the computer shop I worked at. When I got my certification from Commodore back in the late 1980's, they recommended this. Commodore in many ways was a messed up company. The techs who worked there were pretty smart. In many ways their computers outclassed other computers in that era. Many of the technologies they implemented still influence computer design to this day.
Yes, EST mats and many other devices are needed, depending the type of equipment you work on. I think for general, at home pc service, a wrist strap is a safe way to go, but there will always be the "I've never used one and never had a problem..." A lot depends on the type of flooring and even the atmospheric environment.
True. I use extreme caution in winter. Static seems to be more prevalent during that time.
Would a length of copper/ironwire attached to wrist and pc case be good enough?
The wrist grounding gizmo has an advantage. It does not come off easily. The mat can't fall off, unless you turn off gravity.
Personally, I have always used touching the aluminum part of the case when I work on my machine. I have never had a problem. I would prefer the mat.