Borderlands 4 Launches To Mostly Negative Steam Reviews Over Performance Issues And Crashing

The recommended settings for 2160p (include DLSSP + FGx2), 140-150fps


Recommended settings without FGx2, 90-100fps


These recommended settings provide an extremely smooth and responsive experience (especially with correct mouse / gamepad settings shared on this forum), and image quality still look 4K like.

For comparison here's maxed out (badass) settings with DLSSP + FGx2. It seems recommended settings are 35-40fps faster compared to maxed out settings. I tried playing with badass settings for few minutes and it was still fairly good experience on my VRR monitor, even though framerate fluctuated between 57-63fps.


Badass with FGx2 was even better 102-110fps. I think I could play it like that without complaining about anything. However, very high or even high settings would look pretty much the same and run even better.





Randy-Tweet.jpg
There's nothing wrong with his reasoning. There's no way the PS4 would run this game in it's current scope and quality (real time lighting, big open levels with insane amount of grass all the way up to the horizon). Randy is also right about DLSS. If I can't tell the difference between true 4K and 140 fps, I don't care if the game uses AI to impress me. The real problem with Randy's tweet is that he's being honest. Nothing enrages people more than the truth. From developers perspective there's nothing you could do to please all people. If developers would use old technology (dating back to PS4 era) in their games people who build a modern high end PC will complain that developers dont use their hardware and still prebake lighting. However, if developers want to make full use of the high-end PC, gamers will complain even more (Crysis 1 or games using the UE5 engine).
 
Last edited:
The recommended settings for 2160p (include DLSSP + FGx2), 140-150fps


Recommended settings without FGx2, 90-100fps


These recommended settings provide an extremely smooth and responsive experience (especially with correct mouse / gamepad settings shared on this forum), and image quality still look 4K like.

For comparison here's maxed out (badass) settings with DLSSP + FGx2. It seems recommended settings are 35-40fps faster compared to maxed out settings. I tried playing with badass settings for few minutes and it was still fairly good experience on my VRR monitor, even though framerate fluctuated between 57-63fps.


Badass with FGx2 was even better 102-110fps. I think I could play it like that without complaining about anything. However, very high or even high settings would look pretty much the same and run even better.






There's nothing wrong with his reasoning. There's no way the PS4 would run this game in it's current scope and quality (real time lighting, big open levels with insane amount of grass all the way up to the horizon). Randy is also right about DLSS. If I can't tell the difference between true 4K and 140 fps, I don't care if the game uses AI to impress me. The real problem with Randy's tweet is that he's being honest. Nothing enrages people more than the truth. As a developer there's nothing you could do to please all people. If you use old technology people who build a high end PC will complain that developers dont use their hardware. However, if developers want to make full use of the high-end PC, gamers will complain even more (Crysis 1 or games using the UE5 engine).
I'm almost 40 hours into the game now and I'm running the game at 4K with performance DLSS. It looks good, but if you can't tell the difference between that and native 4K then I recommend new glasses lol. The most glaring difference to me is when you press Z to have the robot show the golden arrowed line to guide you to the next objective... it's very easy to tell it's not native 4K.

I agree that it's great to see games targeting high end hardware but Randy comes across as a manager with no technical knowledge who just shouts that everything they do is great and there are no faults.
 
Last edited:
From benchmarks the game seems to be designed around the medium preset with upscaling, most mainstream cards seem to hit and even well exceed 60fps. UE5s typical shader/stutter/hitching issues not withstanding. The textures are already ass and handicapped by 8GB cards, evidenced by the games file size. The biggest problem is the artstyle isn't conducive or noticably advanced by the extreme compute costs of lumen, nanite, and rt shadows/reflections.
Apparently game doesn't even have hardware lumen [according to df], only software with evident screen space reflections present. Nothing explains this abysmal performance other than totally incompetent devs.
 
Apparently game doesn't even have hardware lumen [according to df], only software with evident screen space reflections present. Nothing explains this abysmal performance other than totally incompetent devs.

Software lumen is still really expensive, especially for consoles and 60-class GPU's. The lowest lumen setting offered in the game Vs. the highest is like a 40% performance delta. A true "off" setting would probably add another 30%+.
 
Last edited:
I'm almost 40 hours into the game now and I'm running the game at 4K with performance DLSS. It looks good, but if you can't tell the difference between that and native 4K then I recommend new glasses lol. The most glaring difference to me is when you press Z to have the robot show the golden arrowed line to guide you to the next objective... it's very easy to tell it's not native 4K.

I agree that it's great to see games targeting high end hardware, but Randy comes across as a manager with no technical knowledge who just shouts that everything they do is great and there are no faults.
I dont mind using DLSS, because this technology reconstructs real detail (rather than simply up-scaling the existing detail with bilinear filtering), so I always get razor sharp image.

However, the quality of certain effects, such as RT, lumen, nanite and SSR, is tied to the internal resolution, meaning there can be more noticeable difference if game use these effects. That difference isn't always obvious to me during normal gameplay (especially from normal viewing distance on my 4K monitor with insanely high pixel density), but I can definitely see these differences when I take screenshots and examine the details closely.

I've done a screenshot comparison in Borderlands 4 between DLSS and DLAA (native), and I can definitely see the difference now. In fact, the differences in this particular game are much bigger than in any other game. For example there's quite a big difference in the way grass is rendered. I wonder if borderlands 4 uses Nanite for foliage; that would explain such big difference, because at higher resolution nanite would render more detail. The size of the edge outline filter must also be different for internal 1080p.

1'st comparison

4K DLAA


4K DLSSQ


4K DLSSP


2'nd comparison, it shows that golden arrowed line

4K DLAA


4K DLSSP



Maybe I will play the game with DLSSQ and medium settings (I still get over 120fps thanks to FG) instead maxed out.
 
Last edited:
????



Randy trying to gaslight people isn't too surprising either. Everyone should refund it so they learn their lesson.
There will always be cases like that in any game.



The game is holding 60 FPS most of the time atleast in this area, which is also where all of the PC benchmarks of 4060's having similar performance but on medium at DLSS 1080p happen.

Also you can see how much worse low looks than the consoles.



A 3070 which is clearly more powerful than a PS5 can't reach the same performance even at 1080p with DLSS Quality at high settings and medium textures. PS5 looks and runs much better.
 
Last edited:
A 3070 which is clearly more powerful than a PS5 can't reach the same performance even at 1080p with DLSS Quality at high settings and medium textures. PS5 looks and runs much better.
I doubt the PS5 uses a high settings preset. It's most likely set to low or medium.
 
Last edited:
I doubt the PS5 uses a high settings preset. It's most likely set to low or medium.
Debatable between high and medium. However it's 100% not using low. Low gets rid of grass entirely and the lighting is very different.
mIjlfyjleniU2VPK.png


Even the Series S looks much better.



zu44MeZ1RH0cvLWQ.png


As you can see, the lighting and the grass is fully intact, unlike low settings.

Series S is probably running custom medium settings while PS5 and XSX are running custom high settings.

amF6Oq1iY8zR2UK0.png


These are pretty much low settings with a few tweaks Nvidia has prepared for PS5 class GPUs (3070, 4060, 2080). High is much closer to the consoles.

The PC performance of this title is completely broken.
 
Last edited:
Debatable between high and medium. However it's 100% not using low. Low gets rid of grass entirely and the lighting is very different.
mIjlfyjleniU2VPK.png


Even the Series S looks much better.



zu44MeZ1RH0cvLWQ.png


As you can see, the lighting and the grass is fully intact, unlike low settings.

Series S is probably running custom medium settings while PS5 and XSX are running custom high settings.

amF6Oq1iY8zR2UK0.png


These are pretty much low settings with a few tweaks Nvidia has prepared for PS5 class GPUs (3070, 4060, 2080). High is much closer to the consoles.

The PC performance of this title is completely broken.
The visual difference between medium and high settings is subtle in borderlands 4, yet performance difference is quite big, therefore I dont think gearbox would spend resources for marginal visual difference and especially when the game has performance problems on consoles.

Digital Foundry (12m40s) said in this video that borderlands 4 one the PS5 dips like crazy.

 
Last edited:
Debatable between high and medium. However it's 100% not using low. Low gets rid of grass entirely and the lighting is very different.
mIjlfyjleniU2VPK.png


Even the Series S looks much better.



zu44MeZ1RH0cvLWQ.png


As you can see, the lighting and the grass is fully intact, unlike low settings.

Series S is probably running custom medium settings while PS5 and XSX are running custom high settings.

amF6Oq1iY8zR2UK0.png


These are pretty much low settings with a few tweaks Nvidia has prepared for PS5 class GPUs (3070, 4060, 2080). High is much closer to the consoles.

The PC performance of this title is completely broken.

Yeah you need around rtx 3060 to get a ps5 experience in this game.

ps5


3060
00ftyskmoiws4555-Untitled.png
 
Last edited:
Buttttt, it does? One guy had an issue with the game and all of a sudden the shit runs bad on ps5? Is that what we are doing here? Just want to double check to make sure that's the bed you are making?
You have all sorts of videos reporting fps in the 40s when in comabat, DF claiming theres a ton of stutters and low fps on ps5 including on ps5 pro and you make this silly post?
 
Debatable between high and medium. However it's 100% not using low. Low gets rid of grass entirely and the lighting is very different.
mIjlfyjleniU2VPK.png


Even the Series S looks much better.



zu44MeZ1RH0cvLWQ.png


As you can see, the lighting and the grass is fully intact, unlike low settings.

Series S is probably running custom medium settings while PS5 and XSX are running custom high settings.

amF6Oq1iY8zR2UK0.png


These are pretty much low settings with a few tweaks Nvidia has prepared for PS5 class GPUs (3070, 4060, 2080). High is much closer to the consoles.

The PC performance of this title is completely broken.
It's clearly a large PC issue…and trolls and "-PC is the master race no matter what and everyone else is a peon!" types are trying to externalize the issues they are having onto other methods of play.
 
You have all sorts of videos reporting fps in the 40s when in comabat, DF claiming theres a ton of stutters and low fps on ps5 including on ps5 pro and you make this silly post?
Silly post? I've been playing the game for 30+ hours. Gtfoh with trying to make some minor subset of players as the whole. Shit is tiring here with people that do that, and it's getting old asf. A super vocal minority of people don't make up the large majority of ps5 players that are having no issues and actually enjoying the game. The PC version has some performance issues, we all get it, but don't try to bash the console versions because of it. And let me guess…you didn't watch the "all sorts of videos" with no issues did you? And you haven't played it yourself either, did you? Calling a post silly for expressing what I've experienced playing the game personally, is the whole problem with discussing video games anymore.
 
Last edited:
Silly post? I've been playing the game for 30+ hours. Gtfoh with trying to make some minor subset of players as the whole. Shit is tiring here with people that do that, and it's getting old asf. A super vocal minority of people don't make up the large majority of ps5 players that are having no issues and actually enjoying the game. The PC version has some performance issues, we all get it, but don't try to bash the console versions because of it. And let me guess…you didn't watch the "all sorts of videos" with no issues did you? And you haven't played it yourself either, did you? Calling a post silly for expressing what I've experienced playing the game personally, is the whole problem with discussing video games anymore.
How often do you restart the game?
 
I'm sad that they gave me this for free with my 5090 instead of something I actually wanted

Nvidia should change the free game bundle into "1 of your choice on Steam"
 
5090FE/9800X3D user here playing at 4k and getting constant Unreal Engine crashes to desktop every 20 minutes or so. Literally have crashed over 30 times since launch. Is there a workaround for this or am I just going to stop playing and wait for a patch for the crashes? Because this is unacceptable and driving me bonkers.
 
Guess I'm just lucky. Running 4070 on an ultra wide screen at 1440p. I don't have the settings maxed out, but I think mostly on High with vsync holding it at a steady 60 FPS.
 
5090FE/9800X3D user here playing at 4k and getting constant Unreal Engine crashes to desktop every 20 minutes or so. Literally have crashed over 30 times since launch. Is there a workaround for this or am I just going to stop playing and wait for a patch for the crashes? Because this is unacceptable and driving me bonkers.
Some people have reported that this fixed their issue with constant crashes. Maybe it will help you, too.

 
Debatable between high and medium. However it's 100% not using low. Low gets rid of grass entirely and the lighting is very different.
mIjlfyjleniU2VPK.png
could you share the link of the video from where you got this screen?

Here is a actual comparison between low and high, lighting doesnt seen "very different"
zVqXbWjiW1EvXZTm.png
 
Last edited:
I think whats the most annoying thing about this game is the emerging trend that people genuinely think 60fps with FG on to get around 100fps is acceptable.

If you're using a mouse and keyboard it feels like aiming through mud, its absolute shit. I hate FG and I'd rather do true 60fps with a controller than 40+ fake frames slapped on with a mouse and keyboard and even worse latency.
 
The recommended settings for 2160p (include DLSSP + FGx2), 140-150fps


Recommended settings without FGx2, 90-100fps


These recommended settings provide an extremely smooth and responsive experience (especially with correct mouse / gamepad settings shared on this forum), and image quality still look 4K like.

For comparison here's maxed out (badass) settings with DLSSP + FGx2. It seems recommended settings are 35-40fps faster compared to maxed out settings. I tried playing with badass settings for few minutes and it was still fairly good experience on my VRR monitor, even though framerate fluctuated between 57-63fps.


Badass with FGx2 was even better 102-110fps. I think I could play it like that without complaining about anything. However, very high or even high settings would look pretty much the same and run even better.






There's nothing wrong with his reasoning. There's no way the PS4 would run this game in it's current scope and quality (real time lighting, big open levels with insane amount of grass all the way up to the horizon). Randy is also right about DLSS. If I can't tell the difference between true 4K and 140 fps, I don't care if the game uses AI to impress me. The real problem with Randy's tweet is that he's being honest. Nothing enrages people more than the truth. From developers perspective there's nothing you could do to please all people. If developers would use old technology (dating back to PS4 era) in their games people who build a modern high end PC will complain that developers dont use their hardware and still prebake lighting. However, if developers want to make full use of the high-end PC, gamers will complain even more (Crysis 1 or games using the UE5 engine).
the ps4 reference i agree with but the game has performance issues on ps5/xsx as well
 
could you share the link of the video from where you got this screen?

Here is a actual comparison between low and high, lighting doesnt seen "very different"
zVqXbWjiW1EvXZTm.png
Depends on the scene, but it can look dramatically different. Here the difference isn't that big.

IgFwOUC.jpeg


But here, you're playing a last-gen version of the game.

kgftG9n.jpeg


It's important to note in this case that the weather is different too. One is rainy and the other is sunny, but the lighting is butchered on Low.
 
Last edited:
I've played other games Randy was involved in that ran like shit on my computer. The difference being, Duke Nukem 3D was still enjoyable at 15fps, as it's Duke Nukem 3D.

Borderlands is just Borderlands.

The man hasn't made a good game since 1997.
 
The devs had something really good going on and destroyed it by terrible optimisation/performance, hell tons of ppl were hyped and bought the game, just look at its steam ccu:
304k alltime peak from 2 days ago, 24h peak 229k.

I bet game did some crazy sales number, 3-4m sold copies first week at least worldwide, if not even 5m...
 
The devs had something really good going on and destroyed it by terrible optimisation/performance, hell tons of ppl were hyped and bought the game, just look at its steam ccu:
304k alltime peak from 2 days ago, 24h peak 229k.

I bet game did some crazy sales number, 3-4m sold copies first week at least worldwide, if not even 5m...
The peak was during the first weekend. 229K in the last 24 hours isn't much of a drop off.
 
Last edited:
Depends on the scene, but it can look dramatically different. Here the difference isn't that big.

IgFwOUC.jpeg


But here, you're playing a last-gen version of the game.

kgftG9n.jpeg


It's important to note in this case that the weather is different too. One is rainy and the other is sunny, but the lighting is butchered on Low.
I think the grass makes up most of the difference. It's non existent at low. So the difference is less pronounced in less grassy areas.

Lighting wise I've found out its hard to compare these shots as it seems like TOD or the weather is different.
 
Last edited:
The more I play Borderlands 4, the more impressed with its graphics. Initially, I thought the cell-shaded Borderlands art style wouldn't benefit from real-time lighting, but I was wrong. In most UE5 games TOD is static, so unless you know where to look the lighting doesnt really look that much different compared to raster games. That's not the case in borderlands 4, because of dynamic TOD. The lighting in this game is incredible, and I can't stop admiring the way the sun's rays illuminate the scene during the passing TOD. Borderlands 3 also had dynamic TOD but raster lighting always looked flat and never impressed me. What's more the volumetric lighting in borderlands 4 still beats the manually placed and pre rendered lighting in R&C Rift Apart. To get such results in real time is just incredible. I'm also impressed by the level design. Many locations caught my eye, especially those around the water. It's a strange mix of Assassin's Creed IV: Black Flag and Borderlands art style :P.

Based on what I saw, I no longer think the requirements are unjustified. It's also not true that Borderlands 4 requires similar GPU power to PT games. In Cyberpunk with PT, my RTX 4080 Super gets around 35–40 fps at native 1440p; even medium settings wouldn't improve the framerate that much. I had to use DLSS Q+FGx2 to get around 120 fps. In Borderlands 4, I get around 75 fps at native 1440p with high settings, so even without AI the game is fully playable. With the help of AI (DLSSQ+FGx2) I get close to 190 fps. The performance of Borderlands 4 is comparable to Ultra RT in Cyberpunk, but definitely not PT.
 
Last edited:
Silly post? I've been playing the game for 30+ hours. Gtfoh with trying to make some minor subset of players as the whole. Shit is tiring here with people that do that, and it's getting old asf. A super vocal minority of people don't make up the large majority of ps5 players that are having no issues and actually enjoying the game. The PC version has some performance issues, we all get it, but don't try to bash the console versions because of it. And let me guess…you didn't watch the "all sorts of videos" with no issues did you? And you haven't played it yourself either, did you? Calling a post silly for expressing what I've experienced playing the game personally, is the whole problem with discussing video games anymore.
If the PC version has performance issues, then so does the console version. We have enough videos to show the PS5 can drop well under 50 and sometimes 40fps. You probably don't notice them, but that doesn't mean the drops are not there. It is very, very far from perfect.
 
The only time I ever used FG was with Cyberpunk because of it's insanely demanding path tracing.

This game doesn't even look that good.
 
lol, I never liked borderlands, was tempted to give 4 a go, but wow what a cluster****, my rig would probably be fine (9800X3D and 9070XT) but not giving any money for this nonsense.
 
The RTX5090 has $2000 msrp and it's actually available at that price right now. I know that's still expensive, but it's nowhere near $5000 as you said.


ym18jbW2roZdM6ru.jpg


Moreover, the RTX 5090 certainly isn't struggling to run Borderlands 4 at 1080p60fps, as my 4080S can already achieve a locked 60fps with Badass (maxed out) settings at 1080p. I'm guessing you watched YouTubers who showed you Borderlands 4 performance while the game was still compiling its shades in the background. When I first started playing the game, the performance was much worse for around 5–10 minutes, but soon I was getting 95 fps instead of 45 fps in exactly the same place. Even if you just change the graphics settings you need to wait few minutes before performance will improve.


Here's 1080p DLAA (native) with badass (maxed out) settings on my RTX4080S. Keep in mind the RTX5090 is twice as fast (and two times as power hungry and three times as expenisve :P).


1080p-1.jpg


2.jpg


3.jpg


4.jpg



At 1440p with DLAA (native) and high settings, the game runs below 60 fps on my PC. However, I think the 5090 could handle these settings and maintain a consistent 60 fps, because even my 4080S isnt that far from that target,


1440p-DLAA-badass.jpg


1440p-DLAA-badass-2.jpg



With DLSS-Quality Borderlands 4 already runs at well over 60fps at 1440p and badass settings.


1440p-DLSSQ-badass-2.jpg


1440p-DLSSQ-badass.jpg



With FGx2 on top of that the game runs at high refreshrste and game feels smooth as butter. I measured 32-36 ms with FGx2. Without FGx2 latency was 28-32ms, so not big difference and you get much smoother and sharper image during motion. IMO DLSS FGx2 is a must in this game.

1440p-DLSSQ-4.jpg


1440p-DLSSQ-5.jpg

Borderlands4-2025-09-14-11-29-28-697.jpg


Here's comparison between badass vs high vs medium settings at 1440p DLSSQ + FGx2:


On my old 2560x1440 LCD I would probably play with badass settings, but on 4K OLED monitor I want sharper image, so I need to lower some settings.

I need to use medium settings at to get similar framerate at 4K DLSSQ (75-90fps)


With FGx2 I get around 130-150fps


With high settings preset and DLSSPerformance I get the same framerate ad medium settings with DLSSQ.


Here's 4K DLSS Ultra Performance with high settings and FGx2 (170-180fps).


4K Ultra Performance with Medium Settings and FGx2 190fps. I wouldn't be surprised if the PS5's image quality looked worse than that. Perhaps a kind PS5 owner will share a screenshot taken in exactly the same spot for a proper comparison. I didn't buy that awesome console, so I can't post my own comparison.


I dont want to defend Borderlands 4, becasue this game is much more demanding as a typical UE5 game and in my opinion, the graphics do not justify the requirements. Some of the assets in the game, especially the textures and trees reminds me Xbox Classic games, so not even X360.

That's it. It's settled. I'm buying a 5090 so I can play borderlands 4 at 1080p. I've been going back and forth for a while, as it is a huge purchase and if I get it I won't be able to make my rent payment... But borderlands 4 at 1080p is worth it.



/s
 
lol, I never liked borderlands, was tempted to give 4 a go, but wow what a cluster****, my rig would probably be fine (9800X3D and 9070XT) but not giving any money for this nonsense.
I bet in a month this disaster will be 30 on many key sites and I hope they finally manage to optimize at least to some degree.
 
Top Bottom