Thursday, July 3rd 2025

AMD Radeon RX 9070 XT Gains 9% Performance at 1440p with Latest Driver, Beats RTX 5070 Ti

In what could spell trouble for the upcoming GeForce RTX 5070 SUPER, Hardware Unboxed found that the latest Adrenalin drivers give the Radeon RX 9070 XT a 9% gain in raster performance at 1440p over review press drivers. This closes its performance gap with the current GeForce RTX 5070 Ti, which would make the card faster than any potential SKU NVIDIA intends to position between the RTX 5070 and RTX 5070 Ti. This testing is part of a feature by Hardware Unboxed where they compared current drivers to review press drivers for both the RX 9070 XT and RTX 5070 Ti.

When averaged across 16 games, the RX 9070 XT with latest drivers was found to gain 9% performance over review press drivers; whereas the RTX 5070 Ti with the latest drivers only gained 2.5% performance over its review press drivers. This 9% gain for the RX 9070 XT puts it ahead of the RTX 5070 Ti. Both the RTX 5070 Ti and RX 9070 XT are capable of gaming at 4K with some settings turned down. Here, averaged across the same 16 games, Hardware Unboxed found that the RTX 5070 Ti gained 3% performance, while the RX 9070 XT gained 4%, and the two GPUs were evenly matched. Helping AMD's average are specific game tests where the RX 9070 XT vastly improved its performance, with "Counter Strike 2" posting a 23% improvement, and "Spider-Man Remastered" posting a massive 27%. "Hogwarts Legacy" sees an 18% gain for AMD. There's more interesting findings and insights in the Hardware Unboxed video linked below.
Sources: Hardware Unboxed (YouTube), VideoCardz
Add your own comment

131 Comments on AMD Radeon RX 9070 XT Gains 9% Performance at 1440p with Latest Driver, Beats RTX 5070 Ti

#26
JustBenching
ChomiqLol, it's the 50-series that now has problems with stability.
Have you seen the megathread on amds reddit with thousands of users experiencing a variety of issues?
RinSenpaiMy main issue with AMD drivers is that we're playing lottery for anything using OpenGL. AMD, as far as my testing goes, can break OpenGL compatibility on Windows within a single update, and we (mostly Minecraft players) already know they just don't care about OpenGL on Windows and prefer focusing on the more modern Vulkan and Linux as a whole.
What im doing is I keep different versions of drivers around and clean installing them depending on what I want to play. It's not a great solution but it will have to do until I move to an nvidia laptop. Some older games though don't work at all no matter the drivers so there is that.
Posted on Reply
#27
Tomorrow
ChomiqLol, it's the 50-series that now has problems with stability.
I argued with him in another topic before seeing the pointlessness and ignored him. He's basing his opinion on a single reddit thread that from 9070 launch as only a few thousand replies and not all of those are exclusively AMD driver problems. And that's worldwide. If his best example is a few thousand posts divided by separate issues across four month's then id' say that very low bar to claim "see, AMD has driver issues too".
Of course they do have issues. No software is perfect. But there are no major issues. Drivers are stable.
RinSenpaiMy main issue with AMD drivers is that we're playing lottery for anything using OpenGL. AMD, as far as my testing goes, can break OpenGL compatibility on Windows within a single update, and we (mostly Minecraft players) already know they just don't care about OpenGL on Windows and prefer focusing on the more modern Vulkan and Linux as a whole.
OpenGL is old games. Considering the small amount of games using that i understand why it's not a priority. From my own testing i've found much bigger performance deficit in older DX11 titles. Not stability problems. Just performance that should be higher given the hardware performance in DX12/Vulkan.
Posted on Reply
#28
Daven
JustBenchingHave you seen the megathread on amds reddit with thousands of users experiencing a variety of issues?
Again, please stop with the misinformation. If there is any "megathread" for AMD driver problems as you say, then there is a "megathread" times 10 for Nvidia. Nvidia 50 series driver problems have been widely reported and confirmed by Nvidia themselves.

The reality for everyone here who isn't constantly trying to win the internet is that Nvidia and AMD drivers are about the same as far as bugs and 'stability' (whatever that means). Each launch of a new series can bring a variety of problems for all companies. While there were massive problems at launch, even Intel drivers are nominally the same as far as bugs and 'stability'.
Posted on Reply
#29
JustBenching
Daventhen there is a "megathread" times 10 for Nvidia.
I hope there is, since it has 10 times the marketshare, it would make sense

What's the misinformation though?
Posted on Reply
#30
RootinTootinPootin
I wouldn't fawn over this since its HWUB..lol..(fine wine in just months? we clearly know/knew NVIDIA has been under the rug recently)
Posted on Reply
#31
RinSenpai
JustBenchingHave you seen the megathread on amds reddit with thousands of users experiencing a variety of issues?


What im doing is I keep different versions of drivers around and clean installing them depending on what I want to play. It's not a great solution but it will have to do until I move to an nvidia laptop. Some older games though don't work at all no matter the drivers so there is that.
That's what I used to do until chipset updates came to my laptop(s) (AMD CPU, Nvidia GPU for 2 of them) and considered these old drivers as Corrupted every single time I DDU'd latest versions for older ones.
Posted on Reply
#32
Daven
JustBenchingI hope there is, since it has 10 times the marketshare, it would make sense

What's the misinformation though?
As others and myself have constantly argued with you, there is no discernible difference on the whole between AMD and Nvidia drivers. We understand that you are an Nvidia fan and want to win arguments on the internet in their favor. We get it. Boy do we get it. But it's information to try and say that AMD is WORSE than Nvidia drivers when that is not the case.

Again for the rest of everyone here, there are drivers and there are features. AMD=Nvidia=Intel for drivers. Sometimes one of three might have more problems than the others relative to market share during the release of a new series. Nvidia has much better features such as super sampling and frame generation. These features have nothing to do with drivers as they require supporting hardware and game development.
Posted on Reply
#33
Chomiq
JustBenchingHave you seen the megathread on amds reddit with thousands of users experiencing a variety of issues?
Don't need to, I've had my buddy encounter black screen issue with a 5090 and myself yesterday having unstable undervolt on the latest driver with a 30 series.
Posted on Reply
#34
JustBenching
RinSenpaiThat's what I used to do until chipset updates came to my laptop(s) (AMD CPU, Nvidia GPU for 2 of them) and considered these old drivers as Corrupted every single time I DDU'd latest versions for older ones.
Rollback the chipset updates then :D

You don't need 10s of drivers though, I only have 3 released on the desktop ready to install.
ChomiqDon't need to, I've had my buddy encounter black screen issue with a 5090 and myself yesterday having unstable undervolt on the latest driver with a 30 series.
Unstable undervolts sounds like a driver issue. Wait for new drivers, im sure it will fix it.
Posted on Reply
#35
Daven
ThomasKAre you honestly still wasting your time with that kid? He's bitter 'cause he dislikes his Amd laptop, so he thinks he can speak on behalf of every Amd product out there, like he owned all of them.

Just hit the ignore button and move on. Rest assured no one here takes anything he says into consideration. He's beyond help.
What you say is true and very good advice. But there is still a need to make sure others don't fall for his misinformation from time to time. I'll take it on myself this time to make sure no one buys his crap.
Posted on Reply
#36
JustBenching
DavenAs others and myself have constantly argued with you, there is no discernible difference on the whole between AMD and Nvidia drivers. We understand that you are an Nvidia fan and want to win arguments on the internet in their favor. We get it. Boy do we get it. But it's information to try and say that AMD is WORSE than Nvidia drivers when that is not the case.

Again for the rest of everyone here, there are drivers and there are features. AMD=Nvidia=Intel for drivers. Sometimes one of three might have more problems than the others relative to market share during the release of a new series. Nvidia has much better features such as super sampling and frame generation. These features have nothing to do with drivers as they require supporting hardware and game development.
Did I say there is a discernible difference? Man what are you on about? I asked if amd has patched up their issues in an AMD thread. Then people start spamming me about nvidias driver issues as if that was part of the topic? What the hell man, seriously, I don't get what the hell are you quoting me for.

Honestly, do you actually go around and quote every person that says "nvidia drivers have issues" telling them they all have issues or are you just doing it to people saying amd drivers have issues? Cause clearly, you are the one that seems to be supportive of one brand over the other. Please, show me a similar reply that you made, quoting someone that said "nvidia has bad drivers" telling them how they all have bad drivers and ill eat my shoe
Posted on Reply
#37
Zendou
JustBenchingI hope there is, since it has 10 times the marketshare, it would make sense

What's the misinformation though?
The person is pointing out that going to something specific like that is likely to have implicit bias. That would then make the information less reliable and that leads to the statement of misinformation.
Posted on Reply
#38
JustBenching
ZendouThe person is pointing out that going to something specific like that is likely to have implicit bias. That would then make the information less reliable and that leads to the statement of misinformation.
Maybe, but then you can handwave any and all complaints about bugs and driver issues.
Posted on Reply
#39
Daven
JustBenchingDid I say there is a discernible difference? Man what are you on about? I asked if amd has patched up their issues in an AMD thread. Then people start spamming me about nvidias driver issues as if that was part of the topic? What the hell man, seriously, I don't get what the hell are you quoting me for.

Honestly, do you actually go around and quote every person that says "nvidia drivers have issues" telling them they all have issues or are you just doing it to people saying amd drivers have issues? Cause clearly, you are the one that seems to be supportive of one brand over the other. Please, show me a similar reply that you made, quoting someone that said "nvidia has bad drivers" telling them how they all have bad drivers and ill eat my shoe
This article is about improved RDNA4 performance, and you come out of nowhere trolling about 'stability' issues. So either link us to a professional article (not a reddit thread) that shows testing that proves stability issues or stop with the nonsense.

Here is an example. I started this thread showing improved performance from two different TPU reviews at two different times to corroborate what Hardware Unboxed found. Now you can post an article showing AMD stability problems as a counter to these performance improvements. I am willing to read or watch the link you send but again it can't be a reddit thread as there is no way to verify what's being said is true.
Posted on Reply
#40
JustBenching
DavenThis article is about improved RDNA4 performance, and you come out of nowhere trolling about 'stability' issues. So either link us to a professional article (not a reddit thread) that shows testing that proves stability issues or stop with the nonsense.

Here is an example. I started this thread showing improved performance from two different TPU reviews at two different times. Now you can post an article showing AMD stability problems as a counter to these performance improvements. I am willing to read or watch the link you send but again it can't be a reddit thread.
So you are only playing defense when someone points out at the amd driver issues but ignore it when someone points out the nvidia driver issues. Yet you have the audacity to call others fanboys. Gotcha, im out, not gonna reply to this any further.
Posted on Reply
#41
TheGeekn°72
Maybe it's already done in which case I was just not aware so sorry if it's already the case but but how realistic could it be for TPU to bench cards regularly every now and then after a few driver releases to see if the Relative Performance Chart shifts ? I'm also curious to see if it could be possible to see a *per cards/series* Relative Performance Chart, with graduation say, every semester, showing the performance uplift

I know it would be quite the task to do something like that but I think there could be value in documenting and publishing this kind of information to help appropriately inform people about driver quality on top of card performance
Posted on Reply
#42
RinSenpai
JustBenchingRollback the chipset updates then :D

You don't need 10s of drivers though, I only have 3 released on the desktop ready to install.
I can't. Neither AMD nor Lenovo/HP provides them, and full reset to factory does nothing about that.
Posted on Reply
#43
JustBenching
TheGeekn°72Maybe it's already done in which case I was just not aware so sorry if it's already the case but but how realistic could it be for TPU to bench cards regularly every now and then after a few driver releases to see if the Relative Performance Chart shifts ? I'm also curious to see if it could be possible to see a *per cards/series* Relative Performance Chart, with graduation say, every semester, showing the performance uplift

I know it would be quite the task to do something like that but I think there could be value in documenting and publishing this kind of information to help appropriately inform people about driver quality on top of card performance
They are doing that already, when new cards release they rebench everything with newest drivers afaik. @W1zzard can shed some light

A 5050 review is upcoming, so he'll probably use newest drivers.
Posted on Reply
#44
Daven
JustBenchingSo you are only playing defense when someone points out at the amd driver issues but ignore it when someone points out the nvidia driver issues. Yet you have the audacity to call others fanboys. Gotcha, im out, not gonna reply to this any further.
Sorry we couldn't come to an understanding but I will repeat, drivers are basically good from all three companies (Nvidia, AMD, Intel). Complaints about drivers come up from time to time for ALL companies. It just happens that this time around there were complaints found by independent reviewers for the 50 series (black screens). Nvidia confirmed these complaints and started issuing driver fixes like any good company should do. If somehow I implied in this thread or past threads that one company has better drivers over another, I apologize, because that is just not the case.

Typically, my posts are about debunking the idea that one company's drivers are better than another's. Since the FUD surrounding AMD drivers was so venomous and false, some of this debunking could look like bias. Again I apologize if that seems the case to some of you here but I mainly get complaints everytime from the same three people: two TPU readers and one TPU staff member.
Posted on Reply
#45
jnv11
I am wondering if this is a result of AMD’s driver team learning how to take advantage of RDNA 4’s better scheduling hardware.

Apparently the scheduling hardware in RDNA 3 and earlier stall the pipeline to handle false dependencies, wasting the GPU compute hardware’s time. Nvidia and Intel use software instruction scheduling in the driver, and the software drivers do a better job scheduling instructions to not be stalled by false dependencies than RDNA 3 and earlier. This could be one reason why Nvidia and Intel could have higher driver overhead than AMD does, but the overhead pays off with less wasted GPU time. RDNA 4 fixes this scheduling problem.

Source: chipsandcheese.com/p/rdna-4s-out-of-order-memory-accesses
Posted on Reply
#46
JustBenching
jnv11I am wondering if this is a result of AMD’s driver team learning how to take advantage of RDNA 4’s better scheduling hardware.

Apparently the scheduling hardware in RDNA 3 and earlier stall the pipeline to handle false dependencies, wasting the GPU compute hardware’s time. Nvidia and Intel use software instruction scheduling in the driver, and the software drivers do a better job scheduling instructions to not be stalled by false dependencies than RDNA 3 and earlier. This could be one reason why Nvidia and Intel could have higher driver overhead than AMD does, but the overhead pays off with less wasted GPU time. RDNA 4 fixes this scheduling problem.

Source: chipsandcheese.com/p/rdna-4s-out-of-order-memory-accesses
Probably is the case, since the performance doesn't seem to carry over to 4k.
Posted on Reply
#48
Dr. Dro
I really dislike this clickbait stunt HUB pulled. We forgetting the 9070 XT was losing to the 4070 Super at CS?
Posted on Reply
#50
Zendou
JustBenchingMaybe, but then you can handwave any and all complaints about bugs and driver issues.
Not true, depends on the level of evidence and authenticity of what is provided. What you have stated is the slippery slope, which does not necessarily apply.
Posted on Reply
Add your own comment
Jul 24th, 2025 15:06 CDT change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts