Question Why is it lagging?

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
Hi all. So, I've been playing the same games for quite a while(mostly) and, since I got my GPU several months ago, I've never had a single hint of lag at full graphics on them all, until recently. I'll give details:

Games:

Dragon Age Inquisition
Mortal Kombat 11
Battlefield 1 & 4
(there are others, but those are the most graphically intensive of them)

Specs:

XFX Radeon R9 290x 8GB GDDR5 VRAM
Intel i5 4570 3.2GHz
16GB DDR3 RAM
750W PSU

More details on the issue:

Recently, one of the fans on my GPU stopped working, and that's roughly the time this started. However, as suspect as that may seem, I've extensively monitored the temperatures of every piece of hardware in my system, and there has been no significant change. My GPU is Hawaiian, which means it is built to run hotter. Good temperatures for my GPU are roughly 94-96C. I took a tower fan that has been detached from its mount for quite some time, and positioned it close to where the GPU fan was(I removed it after it stopped working). Also, I removed the side panel from my tower and am using a house fan to super-cool my entire board, temporarily. These measures have kept my GPU at an average of 96C, and it has raised no higher than 97C as a maximum with any given game. My CPU temperatures haven't changed at all. Because my GPU temperatures haven't increased, it leads me to believe that the missing fan isn't the problem here. But if it's not, then what could it be? Any ideas?
 
Last edited:
My GPU is Hawaiian, which means it is built to run hotter. Good temperatures for my GPU are roughly 94-96C.

No, ridiculous. Who told you that? Your GPU's codename is Hawaii and that has nothing to do with it 'being made to run hotter'. It's running hot because it's missing half its active cooling and because the cooler isn't very good to begin with.

Your issues are as follows:

You need to replace that fan.

You need to repaste your GPU as the XFX cooler is poor, repasting might help with temperatures.

This is why I avoid XFX graphics cards like the plague, terrible quality. They always cheap out on something.

Edit: Those temperatures you gave are terrible and will be killing your card and its performance.
 
Did you register the card with XFX for the lifetime warranty which was available at the time? If so open a ticket with them in the past they have replaced two cards for me HD4890 with fan issue replaced with HD6850 (better card) and HD 7950 with failing Vram with RX570 4GB (better card)
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
(Don't know how to quote on here)
@John Chesterfield Sorry, but you're wrong.

The XFX cards were built to run at hotter temperatures specifically for the climates of the region.
Citation: https://wccftech.com/amd-radeon-r9-...aii-gpus-gaming-synthetic-benchmarks-exposed/

The temperatures are fine. It has always run at those temperatures, and if you spend 5 minutes looking up the card and researching normal temperatures for it, you'll see that pretty much every review on the internet cites 93-96C, and you'll find people talking about the fact that it's Hawaiian as a cause. Unfortunately, AMD took their explanations for it down a long time ago, as it's not exactly a new card. Anyway, as I said in my OP, being that there have been no significant changes in temperature, I have no reason to believe replacing the fan will solve this issue. Of course I will, when I'm able, but I didn't create this post for that. I created it because my readings have given me legitimate reason to believe it's not a temp problem.
 
Last edited:

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
@rolli59
I just bought this card from a private dealer about 5 months ago, so it's not under warranty anymore, or at least I wouldn't think it is. I'll look into it, but I highly doubt it. Either way, that's doesn't answer my question. As I said, the card isn't running any hotter than it always has, nor are any of my other components, which begs the question: Why is it lagging?
 
(Don't know how to quote on here)
@John Chesterfield Sorry, but you're wrong.

The XFX cards were built to run at hotter temperatures specifically for the climates of the region.
Citation: https://wccftech.com/amd-radeon-r9-...aii-gpus-gaming-synthetic-benchmarks-exposed/

The temperatures are fine. It has always run at those temperatures, and if you spend 5 minutes looking up the card and researching normal temperatures for it, you'll see that pretty much every review on the internet cites 93-96C, and you'll find people talking about the fact that it's Hawaiian as a cause. Unfortunately, AMD took their explanations for it down a long time ago, as it's not exactly a new card. Anyway, as I said in my OP, being that there have been no significant changes in temperature, I have no reason to believe replacing the fan will solve this issue. Of course I will, when I'm able, but I didn't create this post for that. I created it because my readings have given me legitimate reason to believe it's not a temp problem.

So you use one of the least respected sources on the Internet and in said article, I can find nothing about the chip being designed for the climate in Hawaii? But it just so happens that Hawaii is the codename of the chip? No offence if English isn't your first language, but can you actually read English? Never mind that it's WCCF Tech.

Your patronising response won't get you very far on here, unhappy because you are given the truth. The XFX coolers for those cards were CRAP, period. Why don't YOU go and do some reading about that? I owned a 290x years ago long before you did, an MSI Gaming version and that did not run at 95C, it ran at around 83C, far less, because it didn't have a crap cooler like the XFX.

I won't be responding further after I answer your last post, you are not worth helping any further than that.
 
The issue is the cooling. If that GPU is running at 96C, it is throttling. Even when you got it, apparently. the GPU is slowing itself down, in order to reduce the amount of heat produced. The temps havent gone up when the cooler fan died, because the GPU controllers will not let the GPU get hotter. To prevent it from getting hotter when the fan died, it is slowing the GPU way down, in order to create less heat. If that, or any GPU Chip, gets above ~100C, it risks permeant damage. Damage is not guaranteed, but is not unlikely. I had an old AMD GPU, and it had a broken cooler fan. it only had one, so that was a bad issue. I tested it. The GPU ran still, and the temperatures climbed VERY fast under load. The temps hit 99C in seconds, and then stopped climbing. This is not because the GPU was able to run at 99C, but because when it hit that temperature, it started to run slower to prevent its own death. The amount of processing done by the GPU is directly related to how much heat is produced. If the heat cannot be taken away fast enough, the processing slows down to reduce the amount of heat produced. No GPU is designed to run at 95C, and if one is, it is only doing so as a last attempt to not overheat to the point of irreversible damage. The processing slowdown will be immediately visible to the user, in the form of lag and/or lowered FPS.
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
@John Chesterfield

"R9 290 is designed to run upto 95c. R9 280X is only safe for 85c. ideally shave off 5c from both those. so 90 for R9 290 and 80 for R9 280X is what would be safe for long term use. "
(from https://www.overclock.net/threads/whats-the-temp-range-for-r9-290-and-the-r9-280x.1450331/)

"65c is chilly for a GPU, my ASUS 290x hits 95c and the Powercolour hits 80c."
(from View: https://www.reddit.com/r/Amd/comments/3m3bck/whats_the_average_gpu_temp_of_r9_280x/
)

"70 is a good temp seeing that AMD says the chip can should run well at 95"
(from https://pcpartpicker.com/forums/topic/49773-xfx-r9-280x-double-dissipation-running-hot)

I could go on and on. Shall I? I'm not patronizing you. I'm responding to your "know it all" attitude attempting to tell me that my factual knowledge is wrong. the XFX 290x GPU is indeed intended to run at that temperature. If you believe otherwise, that's simply because you're uninformed on the topic. I appreciate the response, however,e if you were expecting rainbows and sunshine after the tone of your initial response, you're quite mistaken.
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
The issue is the cooling. If that GPU is running at 96C, it is throttling. Even when you got it, apparently. the GPU is slowing itself down, in order to reduce the amount of heat produced. The temps havent gone up when the cooler fan died, because the GPU controllers will not let the GPU get hotter. To prevent it from getting hotter when the fan died, it is slowing the GPU way down, in order to create less heat. If that, or any GPU Chip, gets above ~100C, it risks permeant damage. Damage is not guaranteed, but is not unlikely. I had an old AMD GPU, and it had a broken cooler fan. it only had one, so that was a bad issue. I tested it. The GPU ran still, and the temperatures climbed VERY fast under load. The temps hit 99C in seconds, and then stopped climbing. This is not because the GPU was able to run at 99C, but because when it hit that temperature, it started to run slower to prevent its own death. The amount of processing done by the GPU is directly related to how much heat is produced. If the heat cannot be taken away fast enough, the processing slows down to reduce the amount of heat produced. No GPU is designed to run at 95C, and if one is, it is only doing so as a last attempt to not overheat to the point of irreversible damage. The processing slowdown will be immediately visible to the user, in the form of lag and/or lowered FPS.

I appreciate the response. You're wrong in that my GPU is not supposed to run at this temperature, because in fact it is(I've just posted multiple pieces of evidence, but if it's not enough for you, call AMD and ask them, they'll confirm my words). However, what you said about throttling in order to maintain proper temperatures I hadn't thought about. That does make sense. Thanks for that.
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
To everyone insisting that the XFX Radeon R9 290x is not intended to run at 93-96C:

The entire 5 months I've had this card, I have closely monitored temperatures from my entire setup. This is due to me seeing the "high" temperatures when I first got the card, and researching the "issue"(it's not an issue, but thought it was at first) to see what I could do. Once I confirmed that this card is indeed supposed to run at these temperatures, I decided to keep a close eye on the rest of my rig to ensure the higher temperatures didn't cause damage or higher temperatures to the other parts of my setup. Until about 2 weeks ago, I had never experienced a single problem. No lag at all, no problems, yet my GPU temperatures have always been between 94C and 96C at full load. Why is it just now having lag problems, after the fan goes out, if these temperatures were a problem? Wouldn't it stand to reason that my GPU would've been throttling long before the fan went out if these temperatures were indeed too high? They're not. Call AMD and ask them if you don't believe me. The Sapphire version of the card runs around 80C, however, the ASUS and XFX both run 95C on average, and it is intended to be so.
 
Your responses are very hostile...

Aside from that, whether it was an issue before or not, your issue NOW, is that you have a broken cooler, and the card is not running at full power. Fix the cooler, and the issue should go away, or at least back to the way it was.
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
Your responses are very hostile...

Aside from that, whether it was an issue before or not, your issue NOW, is that you have a broken cooler, and the card is not running at full power. Fix the cooler, and the issue should go away, or at least back to the way it was.

Only toward John, because of what he said initially. I'm not trying to be hostile toward you. You've not said anything I find to be deserving of hostility. After you mentioning the throttling, I agree, the issue is the coolers. I decided to go into MSI Afterburner and max out the remaining fan's RPM. That should help until I can get a new one.
 
The issue is the cooling. If that GPU is running at 96C, it is throttling.

Indeed, as I pointed out just before you. I also pointed out that he would need to repair the one fan right from the very start, but he comes back with all this hostility. And then he shows his lack of knowledge by claiming that the chip was made to run in a Hawaiian climate and then linking to an article that mentions nothing of that, says I'm wrong and that it's actually FINE for it to run at peaks of 97C and averages of 95C. Stupid is as stupid does and he was stupid to buy that XFX card. Gets salty with me, should have checked with people on here before buying that hot mess of a card.

Like I say, not worth helping.
 

Fiorezy

Notable
Jul 3, 2020
376
86
890
Okay OP, you are right, everyone is wrong, your gpu is hawaiian, your temperature is safe and the fan is useless. Btw people in the middle east use sahara gpus that was designed to run up to 130c
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
Indeed, as I pointed out just before you. I also pointed out that he would need to repair the one fan right from the very start, but he comes back with all this hostility. And then he shows his lack of knowledge by claiming that the chip was made to run in a Hawaiian climate and then linking to an article that mentions nothing of that, says I'm wrong and that it's actually FINE for it to run at peaks of 97C and averages of 95C. Stupid is as stupid does and he was stupid to buy that XFX card. Gets salty with me, should have checked with people on here before buying that hot mess of a card.

Like I say, not worth helping.

I appreciate the help dude, I just generally don't respond well to people who act as if they know better than factual knowledge. It is a fact that the XFX R9 290 cards(as well as the ASUS ones of the same make) run at these temperatures purposefully. When someone attempts to tell me I'm wrong when what I said comes from factual knowledge, it annoys me, and I often become hostile. I'm sure most of you would too, or perhaps you'd just stop responding. Either way, nobody likes it. Like I said, I appreciate your help, as well as everyone else's.
 

robertbhart

Distinguished
Sep 12, 2012
77
3
18,545
How are you monitoring the temp?

I've never found the PCs own readings in the software to be very reliable. Perhaps this is the source of confusion here.

It seems logical that, if you're seeing the same temperature with half the amount of cooling fans, that there is an issue with the reliability of that temp reading. This applies x10 if this is the case while playing 3d games.

It's a well known fact that temperature control is a massive factor with GPU and CPU performance. Most of us wouldn't even switch a PC on if there was a broken CPU or GPU cooling fan. It could cause permanent damage.

Nobody installs noisy annoying fans for no reason. If the issue started when the fan broke, it's probably not a coincidence....

I'd suggest fixing the fan and not using your PC for anything before that. If the GPU hasn't already been fried you can see if this fixes your issue. If not, go from there.

It would be silly to spend time troubleshooting other potential causes when such an obvious one (that can't be ignored anyway) is right in front of you.

That's my $0.02 anyway.
 

tecmo34

Administrator
Moderator
D

Deleted member 14196

Guest
You said the card is running at the same temps before and after the fan broke. Must be throttling to maintain the same temp despite having one less fan. You think that one fan was useless and had no effects whatsoever on temps?
Yes this. The fan was there to do cooling and now Itcan’t and that’s his problem no matter how much he don’t wanna believe it
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
How are you monitoring the temp?

I've never found the PCs own readings in the software to be very reliable. Perhaps this is the source of confusion here.

It seems logical that, if you're seeing the same temperature with half the amount of cooling fans, that there is an issue with the reliability of that temp reading. This applies x10 if this is the case while playing 3d games.

It's a well known fact that temperature control is a massive factor with GPU and CPU performance. Most of us wouldn't even switch a PC on if there was a broken CPU or GPU cooling fan. It could cause permanent damage.

Nobody installs noisy annoying fans for no reason. If the issue started when the fan broke, it's probably not a coincidence....

I'd suggest fixing the fan and not using your PC for anything before that. If the GPU hasn't already been fried you can see if this fixes your issue. If not, go from there.

It would be silly to spend time troubleshooting other potential causes when such an obvious one (that can't be ignored anyway) is right in front of you.

That's my $0.02 anyway.

I use 3 monitors: CPUID HWMonitor, and OpenHardwareMonitor to monitor everything connected to the board. MSI Afterburner to monitor more GPU specific attributes, but obviously the temperature of it as well. All 3 programs show the same temperature readings. Anyway, it's like I said, my graphical lag never existed until the fan went out, yet my GPU has always read between 94C and 96C, even when I first got it and it was running perfectly. What ttower2020 said is the most likely cause of the lag(he said the fan going out caused my GPU to start throttling in order to maintain proper temperatures). Once I replace the fan, it'll likely go back to normal. In the meantime I've added cooling measures to compensate, as well as lowering the graphical settings in all games, as well as in Adrenaline, and things seem stable.
 

Kultivater

Commendable
Aug 30, 2020
50
0
1,540
What are your GPU and Memory clock speeds in MSI Afterburner? This will show you if the temperatures are impacting the performance, as others have noted. My guess is the card is not running at stated speeds since it is throttling.

Here are two "old" reviews from Tom's Hardware

https://www.tomshardware.com/reviews/radeon-r9-290-and-290x,3728-2.html

https://www.tomshardware.com/reviews/r9-290x-case-performance,3710-2.html

They had reduced, but I gave it a temporary fix so that they run at full capacity(removed side panel of tower and placed house fan to super cool entire board, placed tower fan directly under card and pointed at missing fan, reduced graphical settings in games as well as in Adrenaline), though obviously not running at original load, since that would require better cooling. As a precaution, I also used it to reduce the clock limit so that it doesn't attempt to go to full capacity, even if it could, so I have it holding around 800MHz. Things seem stable, and should remain so until I can replace the fan.
 
Last edited: