Do you worry about wear and tear on your system?

I was working on a heavy image today and started thinking more about the stress constant rendering puts on my system. In the past I usually upgraded my system every 3 years and it's past time for a new one now, but now with the crypto idiots, the poor state of supply chains, a war going on and other factors, prices are thru the roof. I priced some systems that would be an upgrade for what I have now and I just can't justify 4K for a new system..

My current rig is working great. I cleaned it up over a year ago, doubled the DDR, got a new GPU (2080ti), added new HDs. I just worry because I know between rendering, modeling, video work and the countless hours I spend in VR, it's talking it's toll and I don't want to wait till it breaks and then have no choice but to shell out that kind of money

Any others having the same thoughts on hardware.

«1

Comments

  • MattymanxMattymanx Posts: 6,949

    My current rig was built back in 2016 and runs great.  My one before that was built around 2008 or 2009.  The old one is still running, its played many games and done tons of renders.  Its not used as much as it use to be but it still works. 

  • xyer0xyer0 Posts: 6,033

    I've seen over the years that when it's time to replace something, I'll have what I need to replace it. I don't waste time worrying anymore because I get what I need WHEN I need it. Like Mark Twain said, "I've lived through some terrible things in my life, some of which actually happened."

  • IceCrMnIceCrMn Posts: 2,142

    I do.

    That's why I went overboard with the case fans on this build.

    I also went through every setting in the BIOS and OS level software to make certain nothing was over volted or over clocked from the factory(turbo nonsense).

  • Charlie JudgeCharlie Judge Posts: 12,875

    Mattymanx said:

    My current rig was built back in 2016 and runs great.  My one before that was built around 2008 or 2009.  The old one is still running, its played many games and done tons of renders.  Its not used as much as it use to be but it still works. 

    That is when my system also built and I upgraded the GPU to a GTX1080ti in 2017. While it is now getting a bit old it is still working well. I try to care for it but not let it interfer with what I want to to do. I just hope it lasts at least another 3 to 4 years until I have to get a new one to be compatible with Windows 11.

     

  • Unless there's a specific symptom (like overheating, a rattling cooling fan, a clicking sound from the hard drive, etc.), there's no sense worrying about wear and tear.  It's not that something can't go wrong; it's that it's too difficult to predict when the components might fail.  

     

  • marblemarble Posts: 7,500
    edited March 2022

    I upgraded about a year ago - replaced nearly everything except the case and the monitor. But yes, I do worry. Now I have a decent GPU (3090 thanks to a generous family contribution to my upgrade) I often sit here with IRay preview on constantly. My old 1070 GPU would have not allowed that so this new GPU gets worked a lot more. I'm told the GPU has a long warranty but it is a return to manufacturer warranty and I don't fancy sending it back to the USA from here in New Zealand just for them to tell me they can't find a fault. Occasionally my screen will just go black - I have no idea why and have had all the boards out to make sure the connectors are seated properly but it scares me that my GPU will just die on me one of these days. I will not get another generous contribution.

    Post edited by marble on
  • KrzysztofaKrzysztofa Posts: 226

    I once had a 7900GTX melt, it created a spectacularly weird show on my screen for a few seconds before everything went black. Since then, I don't fool around with cooling, even on my laptop I re-did its thermal paste and clean its cooling vents regularly.

    As long as you keep them from overheating, modern machines are quite resiliant. I have a decade old mac lying around that still runs like new, despite having thousands and thousands of hours of creative work logged on it. I just keep up with its care.

  • Matt_BrownMatt_Brown Posts: 177

    I dispalay and monitor the temperatures during big renders, but beyond that I don't particularly worry about my system.

    We are nothing like crypto-miners and probably using our kit close to how it was intended.

  • PerttiAPerttiA Posts: 10,024

    If the internals are kept clean and temperatures down, the only components that are in risk of failing are the ones with moving parts, ie. fans and spinning harddrives, but even with them the risk is small. Fans will usually start making noise before failing and good quality harddrives will usually become too small (in 8-10yrs) to keep in the system before the mechanism starts failing.

  • kyoto kidkyoto kid Posts: 41,212

    ...still on the same system built over a decade ago.  It's had a few upgrades: expanded memory (DDR3 1333) to 24 GB (maximum supported), added a 6 core Xeon 5660, two SSDs (boot and Daz library and topped it off with a Titan-X.  Also replaced the original PSU that blew a capacitor and added three more fans. However getting on in age as it still is Windows7 . which has all but been forsaken by software and hardware developers (Daz is about the last one that hasn't...yet, (though Nvidia could be spell the end soon if new driver requirements go beyond what they support for W7 effectively making Daz unusable as it would mean going back to glacial CPU rendering).  

    So an upgrade plan is in place one that would support Windows11. Intel i7 11700K CPU, Nooctua D15S CPU cooler, ASUS Prime H570+ MB 64 GB DDR4 2666 Memory, and a 980500GB M.2 Boot drive.  Not an inexpensive prospect as the total comes to just over 1,000$ (including W11 Pro).  Already have a 3060 I hoped would work in my existing system  but the MB's BIOS is too old and doesn't recognise it so all I get is standard VGA output.. Still need to save up but may get a component or two at a time as financed permit.

  • LeatherGryphonLeatherGryphon Posts: 11,651
    edited March 2022

    Like an automobile, keep it on a safe road.  No off-roading or hill climbing or racing out of your class.  Keep it clean.  Don't tinker with the voltages.  Let it breathe.  Listen to its complaints. enlightened 

    Also monitor its drivers and utilities. 

    On that topic, I saw the video below a few days ago.  Which was coincidental because just hours before I accidentally ran across that video I had discovered the unknown file on my machine and was wondering why I had it.  Had I been hacked?  The video answers the question and also gives me pause.frown

    My own approach to perpetually oncoming doom(in addition to software backups), is to have spare parts available.  Perhaps not an entirely replicated system, but some sufficiently adequate replacements for power supply, RAM, & GPU.  My production might slow down for a while but it won't stop very long.  However, if anybody follows my posts, they'll remember that I'm a PC junkie and surround myself with their warm buzzy bodies so I always have a spare PC of some capability somewhere close by with readily excisable organs.devilblush

    Yes, I realize that the video is talking about RAM, not GPUs.  But anything that supports RGB lighting could be affected by this type of problem and the message is the same. Know what's being installed on your machines!indecision

    Post edited by LeatherGryphon on
  • Mark_e593e0a5Mark_e593e0a5 Posts: 1,598

    The biggest "wear" on any computer system is heat. Just pure silicon on a chip does not work. You have to add some other elements to it, selectively. That is what all the chip creation process is about. Unfortunately, these "impurities" have a tendency to migrate. Migration is a strictly physical process, based only on temperature. The hotter, the faster. 

    Each chip is designed with a given maximum heat it can tollerate. If you stay below that value, most of the time, the silicon chip can last forever. So, any setting reducing the heat on a given rig will help with its endurance. The best way of reducing lifetime of a given chip is the "tuning" that is often supplied via bios or software. Overvolting and overclocking are a good way of getting your rig to silicon heaven way too soon.

    As for SSD's, they do have a write limitation. A typical memory cell can be written about 1000 times. The SSD logic will ensure that every cell is equally used. So, with everyday use, an SSD should be safe for quite some time. Due to the nature of memory cells used in SSD's, they do only have a limied use for long term storage of data. They need to be connected to a power supply at least every three months. Otherwise, some bits will just flip and your date is at risk.

  • nicsttnicstt Posts: 11,715

    Worry about it?

    Nope.

    Accept it happens: yep.

     

  • rrwardrrward Posts: 556

    PerttiA said:

    If the internals are kept clean and temperatures down, the only components that are in risk of failing are the ones with moving parts, ie. fans and spinning harddrives, but even with them the risk is small. Fans will usually start making noise before failing and good quality harddrives will usually become too small (in 8-10yrs) to keep in the system before the mechanism starts failing.

    I once had two power supplies die within a week of each other. I had plugged what was advertised as a home audio amplifier into my computers. Turned out to be a PA system amp and when you powered it off it kicked enough voltage back at the two computers plugged into it to kill the PSUs. (Never, ever, ever, buy anything from Pyle electronics.) But other than that, it's been moving parts.

  • tsroemitsroemi Posts: 2,871

    Mark_e593e0a5 said:

    The biggest "wear" on any computer system is heat. Just pure silicon on a chip does not work. You have to add some other elements to it, selectively. That is what all the chip creation process is about. Unfortunately, these "impurities" have a tendency to migrate. Migration is a strictly physical process, based only on temperature. The hotter, the faster. 

    Each chip is designed with a given maximum heat it can tollerate. If you stay below that value, most of the time, the silicon chip can last forever. So, any setting reducing the heat on a given rig will help with its endurance. The best way of reducing lifetime of a given chip is the "tuning" that is often supplied via bios or software. Overvolting and overclocking are a good way of getting your rig to silicon heaven way too soon.

    As for SSD's, they do have a write limitation. A typical memory cell can be written about 1000 times. The SSD logic will ensure that every cell is equally used. So, with everyday use, an SSD should be safe for quite some time. Due to the nature of memory cells used in SSD's, they do only have a limied use for long term storage of data. They need to be connected to a power supply at least every three months. Otherwise, some bits will just flip and your date is at risk.

    Thanks for mentioning the SSDs needing power every three months at least, I didn't know that! 

  • nonesuch00nonesuch00 Posts: 18,293

    I never have problems with yet with this new hardware, built using 2019 & 2021 parts, but I really dislike that that lint gets in on the fans and vents. I clean it up on occasion but it seems eventually the lint starts causing the fans to get noisier eventually.

  • TorquinoxTorquinox Posts: 3,589

    That video was quite interesting, @LeatherGryphon . Apps that work at the bios level are uniquely empowered to cause problems, I think.

  • psfilipepsfilipe Posts: 164

    LeatherGryphon said:

    Like an automobile, keep it on a safe road.  No off-roading or hill climbing or racing out of your class.  Keep it clean.  Don't tinker with the voltages.  Let it breathe.  Listen to its complaints. enlightened 

    Also monitor its drivers and utilities. 

    On that topic, I saw the video below a few days ago.  Which was coincidental because just hours before I accidentally ran across that video I had discovered the unknown file on my machine and was wondering why I had it.  Had I been hacked?  The video answers the question and also gives me pause.frown

    My own approach to perpetually oncoming doom(in addition to software backups), is to have spare parts available.  Perhaps not an entirely replicated system, but some sufficiently adequate replacements for power supply, RAM, & GPU.  My production might slow down for a while but it won't stop very long.  However, if anybody follows my posts, they'll remember that I'm a PC junkie and surround myself with their warm buzzy bodies so I always have a spare PC of some capability somewhere close by with readily excisable organs.devilblush

    Yes, I realize that the video is talking about RAM, not GPUs.  But anything that supports RGB lighting could be affected by this type of problem and the message is the same. Know what's being installed on your machines!indecision


    Very interesting.

  • FSMCDesignsFSMCDesigns Posts: 12,775

    Appreciate all the comments, makes me feel better about some things, not as much about others, LOL. Think I'll break it down and do a good cleaning this weekend, been a few months.

  • prixatprixat Posts: 1,590

    Replaced an eleven year old system last year. I'd like to claim it was planned to fit the new Windows 11 requirements, but that was just luck.

    On the new system, I fitted more fans and fans and fans and the biggest cpu cooler that would fit in the case. It took much longer than it should have to remove the RGB settings, services and drivers.

  • FSMCDesignsFSMCDesigns Posts: 12,775

    Go figure, after I posted this I get an email from the PC place I got my last rig with a bday coupon for 20% off, LOL I spent an hour customizing systems and knowing what I need from experience, will still cost me around 3K. Time to fill out the income tax to see what I am getting back, LOL.

  • WandWWandW Posts: 2,845

    I built my render system ten years ago.  I have upgraded the graphics card twice and added an SSD boot drive.  No issues...

  • N-RArtsN-RArts Posts: 1,522

    My computer (still) suffers from random restarts. So I have anxiety before I turn the flipping thing on. 

    But yeah, I worry about temperature problems. CPU problems. Random restarts when rendering (and losing hours hours of work). If it needs cleaning. If the SSD or External Hardrive is failing (I don't trust that SSD)... Just... EVERYTHING. The damn thing is a mental minefield from switch on to shutdown. 

    Even if all of that stuff wasn't happening in the world, I STILL couldn't afford a replacement computer (my ideal replacement computer is £2k). 

  • kyoto kidkyoto kid Posts: 41,212

    LeatherGryphon said:

    Like an automobile, keep it on a safe road.  No off-roading or hill climbing or racing out of your class.  Keep it clean.  Don't tinker with the voltages.  Let it breathe.  Listen to its complaints. enlightened 

    Also monitor its drivers and utilities. 

    On that topic, I saw the video below a few days ago.  Which was coincidental because just hours before I accidentally ran across that video I had discovered the unknown file on my machine and was wondering why I had it.  Had I been hacked?  The video answers the question and also gives me pause.frown

    My own approach to perpetually oncoming doom(in addition to software backups), is to have spare parts available.  Perhaps not an entirely replicated system, but some sufficiently adequate replacements for power supply, RAM, & GPU.  My production might slow down for a while but it won't stop very long.  However, if anybody follows my posts, they'll remember that I'm a PC junkie and surround myself with their warm buzzy bodies so I always have a spare PC of some capability somewhere close by with readily excisable organs.devilblush

    Yes, I realize that the video is talking about RAM, not GPUs.  But anything that supports RGB lighting could be affected by this type of problem and the message is the same. Know what's being installed on your machines!indecision

    ...as I have no RGB hardware (since my case has no window and am not into digital "bling") fortunately won't have to worry about this.  . In years past I would get BSODs now and then, but that was usually due to a buggy Nvidia driver.

  • RL_MediaRL_Media Posts: 339

    Not really. Every time I build a new one or replace a part, I put the old parts into a giant PC organ donor box just in case. Worse comes to worse I can frankenstien a whole new system or ten out of thos if needed lol. 99% of the time when I replace a part, it's still working fine, just replaced with something faster or better and keep the part I replaced. On one hand, I find the transparent window on my giant tower annoying, because everything has to have LED on them these days so I have to block it off when having my PC doing something over night. It's great for taking a peek inside now and then and seeing if I have dust bunnies multiplying in there lol. Dust is a lot less of an issue now that I built my massive desk, and keep the tower on the top instead of on or near the floor. Also since I switched from smoking to vaping years ago, I no longer have clingy tarry dust buildup issues. I can just hit it with a vacume and the dust mostly just comes off no problem. Cleaning the PC is a lot less of an ordeal now.

  • ed3Ded3D Posts: 2,280
    edited March 2022

    And then -

    one of these  -  bought at Staples

    The MetroVac DataVac Electric Duster

    super  powerful _ 

    Metro-Vac Duster 1.jpeg
    612 x 612 - 31K
    Post edited by ed3D on
  • FSMCDesigns said:

    I was working on a heavy image today and started thinking more about the stress constant rendering puts on my system. In the past I usually upgraded my system every 3 years and it's past time for a new one now, but now with the crypto idiots, the poor state of supply chains, a war going on and other factors, prices are thru the roof. I priced some systems that would be an upgrade for what I have now and I just can't justify 4K for a new system..

    My current rig is working great. I cleaned it up over a year ago, doubled the DDR, got a new GPU (2080ti), added new HDs. I just worry because I know between rendering, modeling, video work and the countless hours I spend in VR, it's talking it's toll and I don't want to wait till it breaks and then have no choice but to shell out that kind of money

    Any others having the same thoughts on hardware.

    Keeping it clean, properly cooled, and ambient temperatures down to a reasonable value should go a long way in extending the life of your system. Depending on how quickly dust builds up in your house or office, it might be worth investing in a small air purifier that will suck out some of those particles floating about in the air. If you're frequently doing long renders, it might be a good idea to keep the ambient temperature of the room between 65-68 F, assuming that is feasible. I know the enterprise systems they have at the one manufacturing site that I was working at were kept in a server room where the ambient temp was constantly kept at 65 F, which are constantly running 24/7. Don't do anything stupid like burning scented candles anywhere near the system. After a couple of weeks, I ended up having to clean all of the filters. While it wasn't having any effect on cooling at the time, I could only imagine what kind of a mess it would make inside of the case had it went on for several months or longer.

    During winter, the small room(roughly a 12' x 12' carpeted + drywalled attic type room) where I have my system can get up to 81 F when doing frequent renders on an RTX A6000(exhausts the hot air out the back with the card hitting 80 C ) plus I have the RTX 3090 Kingpin(VRMs 1-5 stay between 40-56 C ) that is continually expunging warm air out of the top of the case through the 360mm radiator. The system is a literal space heater. I'm guesstimating that I'll need a 10,000 BTU AC unit to keep the ambient temps down without needing to have the unit running constantly. I can't imagine how some are running server farms filled with 3090s keep the area cool. They must have one hell of an electric bill for the AC.

  • UHFUHF Posts: 515

    Well.... If anyone here ever did CPU rendering with Reality, then you kinda know that running your PC at full tilt for 5 days isn't a big deal.  I did that for years.

    However...

    You should ask yourself just how badly you need the last 10-20% of your performace.  Did you need that render done in 8 minutes or 10?

    With my current PC set up, I could only put my Asus RTX 3080TI right next to the CPU, and there it generates lots of heat if I run it hard.  I wasn't that worried about all that, but this PC has an M.2 SSD right in there and I didn't want to overheat that.  (M.2 SSDs are very temperature sensative.) So I configured MSI Afterburner to max my GPU at 90% output and throttle it to a lower temperature.

    I verified all my configurations with an IR camera, the M.2 was pushing 100C when I was rendering, and now it rarely goes over 70C.  I really don't notice the performance loss.

  • Sfariah DSfariah D Posts: 26,525

    Sometimes but I'm also happy to have a computer.

  • frank0314frank0314 Posts: 14,263

    ed3D said:

    And then -

    one of these  -  bought at Staples

    The MetroVac DataVac Electric Duster

    super  powerful _ 

    That thing is a god send. I love mine. We bought one probably 10 yrs ago and still runs strong and I use it for a lot of things besides cleaning out the computer.

Sign In or Register to comment.