Starcraft 2’s menu screens would like a word with you.
Blizzard left the FPS uncapped on those screens at launch, and boy was it a nightmare for some people.
Ok, so, yes menu’s should be capped.
But an uncapped menu screen should not be able to fry your card, that’s an issue with the card itself. Cards have things in place to prevent that kind of thing, it’s when those things fail that a card fries.
Every instance of cards frying has been related to poor manufacturing and issues with cards themselves.
1 Like
Many people wish that were the case. But when your GPU is rendering many multi hundreds of complex-ish frames a second, it can cause your GPU to overheat, especially flagship GPUs and that’s what was happening. Worse even if you had poor ventilation.
Fortunately I didn’t suffer from this.
So the game itself didn’t fry the cards but the very nature of an uncapped menu screen pushing the GPU to its limits with no kinetic feedback outside of roaring fans will.
1 Like
Your gpu has safeties against overheating, it’ll throttle to prevent it.
This is entirely a manufacturer problem. If you look at the New World problem, those cards frying were actually going beyond voltage safeties.
A good card would handle uncapped framerates with zero issues and zero overheating.
Let’s not forget them not honoring warranties on claims being made after bios updates, they released fail and cause the pcs to fail outright. For mor info look up Jaystwocents on youyube he made video on it either yesterday or the day before that. Better yet here is the video on it.
You can set a frame limit for your graphics card and avoid it. The problem was shoddy work and uncapped framerates that caused it. Same as with New World IIRC
I don’t care to watch his videos. He’s proven twice already he’s a fearmonger. First he claimed that 2000 series cards (or maybe it was 3000 series cards) were dying because of the type of capacitors on the back of the GPU, making people worried if they had the “bad” one. Then it turned out to be poor soldering of mosfets if I remember correctly.
Second, more recently, when the 4000 series was announced, with the new 12+4 pin connectors, he said everyone needs to go and get ATX 3.0 PSUs, in order to use the 4-pin sens wires, or they’re going to have current problems. Then when reports came in of melted connectors, he made a video saying “I told you so.” And then after ACTUAL proper investigation by Gamers Nexus and Nvidia, it was determined that it was incredibly likely these connectors melted due to not being inserted properly, to where they are actually seated and retention mechanism clicks into place. Once this data was published, melted connector reports practically stopped appearing overnight.
That’s two occasions of fear mongering by jayztwocents claiming he knows what he’s talking about. He jumps to early conclusions for views, nothing more. Many youtubers are a waste of time but ones who just ignorantly spread misinformation and fear, so they can get views is absolutely ridiculous.
Maybe one day he will make an accurate video on hardware failures and perhaps that video is 100% accurate but his past offences of fear mongering mean I will never watch a video of his again.
Correct. EVERY gpu is designed to throttle itself before it gets damaged (and has been that way for a very, very long time) be it from heat or current draw. Uncapped fps wont ever force a card to not do that. It cant.
When it fails to do that, because, its defective, you get what happened here.
0, zero, nada, nothing, at all, to do with the game.
Those cards were going to melt down one way or another eventually, no matter how much time its users tried to buy them.
The uneducated, and even one of the forum mvps is going around saying othetwise (which is sad and he should stop), but THATS how it works.
Everyone take note here.
Delusion is not a replacement for knowledge.
Stop posting on these forums like you have any idea about what you are talking about.
I’m not here to educate you on subjects that I actually know about.
I will introduce you to a little treat though!
Thermal throttling only prevents the GPU from passing a temperature threshold. It does not however stop the graphics card from attempting to pump out higher frame rates. Whatever the new roof becomes is what frame rates the graphics card will render and without an FPS cap in Starcraft 2’s menu screen, the GPU continued as if nothing were wrong.
So instead of rendering 900 FPS, it throttles back and is now pushing… 200 FPS! Golly gee, those numbers are still incredibly high! Almost as if the GPU is still putting in a lot of reps for those gains.
For the uninitiated (I’m looking at you Corpius-chan), Thermal Throttling is only set to one value. That’s typically 90C nowadays, I’m not sure what they were back in 2010.
900 FPS will push that value instantly. 200 FPS will maintain that value. There is no second throttle point, instead what will now happen is the GPU will begin to degrade in performance which starts to kill off the performance of the processors onboard the graphics card which manifests in still maintained temps and lower frame rates, and we repeat this process until the graphics card kills itself or if you’re lucky, voltages drop too low that the GPU shuts off and your computer BSOD’s.
By the way, 900 and 200 FPS are arbitrary values I chose for demonstration, actual numbers may vary. Starcraft 2’s menu screen was brutally unkind to graphics cards back in the day. Fortunately most people didn’t have to worry about it because it affected primarily FLAGSHIPS.
This isn’t a problem in most scenarios because people typically do not forget to include FPS caps in their games where necessary but this scenario CAN and WILL happen if you do forget to include one.
As you kindly provided so I don’t need to type it out myself;
Lovingly signed,
Muyu
You miss the point that a proper video card won’t fry even if kept at thermal limit, thermal limit is well below thermal failure.
The cards that were frying in all recent cases had major flaws.
I can’t speak for SC2 as that was a long time ago and limits and safeties have improved in that time. But no MODERN card should ever be capable of being killed by a game that forgot to cap FPS in menus.
That isn’t to say that devs should forget to cap things.
1 Like
Agreed 1000%. Modern devices are much more capable than they were 13 years ago. The probability of a card frying the way they did back then is very very very low. Typically only occurring on faulty boards.
1 Like
4000+ cards is safe.
Honestly though, i have a 144hz monitor, i globally cap at 144 fps as… what would be the point of going higher lol.
I sincerely wish everyone did this because VSync will maintain your refresh rate for you, so any potential overhead you have will just dump into the value your monitor is set to.
Or as you said, setting your FPS cap to a limit.
Yea, and IIRC your cpu still has to do something with all that extra unused frames, so it has to do extra work, but I read that ages ago and it may not be true or fully correct
You should be able to avoid a problem by capping your frames and having updated drivers. However, if the “game” fried your card then it’s probably a good thing. It would have exposed a defect in the card while its under warranty. As much of a headache as RMA is, it’s better to happen now than down the road when you have to pay for a new card instead.