-
Hey Guest. Check out your NeoGAF Wrapped 2025 results here!
NXG - HZD 60fps Patch
- Thread starter N1tr0sOx1d3
- Start date
MightySquirrel
Banned
You're likely hitting bios set power limit and gpu downclocking itself on max load. Try undervolting. Heres quick and dirty guide: first 20 sec Just set 0,9v and something like 1950 and play from there.
Lysandros
Member
I was talking about the GPU polygon throughput, my point is not everything scales down with resolution in the graphics pipeline.lets see
a jaguar chip at 1.6 ghz scores a whopping 95 STR in cpu-z
CPU-Z Benchmark for AMD Athlon 5150 (1T) - CPU-Z VALIDATOR
Best CPU performance - 64-bit - February 2026valid.x86.fr
ryzen 3600 scores a solid 488 STR in the same test. said 3600 is probably operating at 4 ghz so if normalize it to 3.6-3.8 ghz, lets say its 475.
475 / 95 = a whopping 5 times IPC/single core horsepower right there. simple as that.
5 times...
mix SMT into the soup...
yamaci17
Member
oh sorry then
but then again we have series s though. only at 4 tflops, there's no games it cannot do 60 fps yet. even better, some devs managed to push 120 fps out of it. doom eternal for example.
i think this gen 30 fps will only be exclusive to those who want native 4k
Md Ray
Member
At 1080p, you can be CPU-limited. So I think it's not your GPU, maybe your CPU isn't boosting correctly, or perhaps XMP isn't enabled in the bios? What's the speed of your DDR4 kit?
Fess
Member
Never tried undervoltage before!
Unfortunately it didn't do much though, different figures at each run, sometimes slightly higher and sometimes lower, normal variations. I think it's something else going on. Annoying. It started crashing too now so I think I'm done, I've probably seen this benchmark sequence 20 times now lol
Lysandros
Member
No problem.
The thing that i particularly dislike about XSS is that it puts a hard cap to geometry complexity (at same FPS) of 'proper next-gen' third party titles for the entire generation having only around 30% of the geometry throughput (1 SE vs 2 SE, meaning half the number of rasterizers and prim units and running at significantly lower frequency) compared to PS5 for example.
assurdum
Banned
What exactly could have changed the dlc? Never played but I doubt it changes drastically how the game is designed.
Last edited:
Fess
Member
What's XMP? Lol I don't know this stuff.
Huge thanks though for making me check the bios, I had the DDR4 set to 2666MHz and it's Corsair 3600MHz. Whoops
The performance is still not where it should be but it's better! And thanks to
MightySquirrel
the card run about 6C cooler too while doing so!
Md Ray
Member
As I suspected, it's not your GPU. The D.O.C.P option you see there is the XMP (extreme memory profile) on Asus boards. You need to enable it to make your DDR4 kit run at its rated speed (which you say is 3600 MHz), this increases CPU performance on Ryzen particularly.What's XMP? Lol I don't know this stuff.
Huge thanks though for making me check the bios, I had the DDR4 set to 2666MHz and it's Corsair 3600MHz. Whoops
The performance is still not where it should be but it's better! And thanks to
EDIT: Looks like you've enabled it? 105fps to 114fps is a nice bump in perf.
Last edited:
drotahorror
Member
Here's CPU test on 10700K: [would love to see the results with latest Ryzen 5900X to see how far ahead that is]
I think we should give props to devs for the patches if anything this game perf got improved leaps and bounds from launch.
Man I need to upgrade! I'm straight outta 2016 with my computer and its starting to show. I still rock 120hz/1080p though.
Fess
Member
Yup it's activated, thanks again, I've learned so much today! This is the first PC I've built myself so I wouldn't be surprised if there are other wrong settings hidden somewhere. But it's stable now, and relatively cool, GPU is 63C max during the benchmark with the undervoltage tweak, was above 70C before with the Guru3D overclock.
Just tried Cyberpunk and I'm running it at like 75+ fps average now in RT Ultra wooo
AutumnLeaves
Member
According to steam survey, 1.76 % of individuals have AMD Radeon HD 6800 Series; 0.84% have NVIDIA GeForce RTX 3080; 0.16% have NVIDIA GeForce RTX 3080 Ti; 1.02% have AMD Radeon HD 6900 Series; and 0.37% have NVIDIA GeForce RTX 3090. Not really high numbers. And just because it is CAPABLE of 4K doesn't mean it's going to run that on high settings.
Source: https://store.steampowered.com/hwsurvey/videocard/
I just mentioned that might be the reason. I'm not denying that PC will run the game better and more prettier if you have the specs.
MightySquirrel
Banned
Sorry to hear that. That was one issue I had with my card [hitting power limit], so it was the first thing that popped in my head.
MightySquirrel
Banned
Aagh steam survery where high end users don't bother submitting it , but laptop owners rush to do so. Very reliable source. /s
MightySquirrel
Banned
Man I need to upgrade! I'm straight outta 2016 with my computer and its starting to show. I still rock 120hz/1080p though.
Look at the bright side you can't even tell what is bottlenecking what CPU or GPU, they're bang on even. With the rumors about RDNA 3 and 40 series cards being close to twice as powerful as their counterparts and with Zen4, DDR5, PCIe5.0, DirectStorage all coming next year, everyone will want to upgrade their whole systems. In a sense every pc gamer is in the same boat. [if you waited for so long it's prob better to upgrade 2nd half '22 so it will last long]
It's clearly the weak CPU on mine even at 4k:
Fess
Member
Don't be sorry it was very helpful, the card run a lot cooler now, so thanks for the advice there
And if I use the overclock Guru3D wrote about for my card it does hit the power and voltage limit, if I go down the graph list in Afterburner I can see spikes on those graphs. My main issue was just worse, I never thought RAM speed was that important, it's an eye epener for sure for me.Time to start playing the game now though. I'll just double check if some mods can sway me over but right now I think I'll still go with the PS5 version just because I don't want to start over. Both Sony and MS need to add cloud saves that work with Steam, it's a problem with every Steam release now.
Last edited:
Werewolfgrandma
Banned
If you experience crashes, your under volt might be a bit much. If you did it with an afterburner curve, just move ahead a notch for the MHz you picked. So if it's 1950 at .9 volts move it to 1950 at .915 volts or whatever the next one is. I forget the exact volt amount for each square.Don't be sorry it was very helpful, the card run a lot cooler now, so thanks for the advice there
Time to start playing the game now though. I'll just double check if some mods can sway me over but right now I think I'll still go with the PS5 version just because I don't want to start over. Both Sony and MS need to add cloud saves that work with Steam, it's a problem with every Steam release now.
DispatchNinja
Banned
I never had a problem with it years ago either, but I also felt the combat to be pretty rubbish after playing again for about 4-5 hours. When the map opened up revealing a shit load of icons it just overwhelmed me to the point of just deleting the game. I've beaten the game a couple of times already so I can wait for the sequel.
Shmunter
Member
They need to completely rework the melee. The bow is fine.
DispatchNinja
Banned
The bow is average imo, but felt bad after just finishing ghost of tsushima. I kept shooting over everything and kept missing shots that I felt should have been hit. I was also playing with aim assist off though. I'm sure I could have adjusted to it more but I really wasn't feeling it enough to continue. Curious to see how much they improve the combat in the sequel.
Kenpachii
Member
All his PC conclusion are wrong because the data behind it is wrong that's what i pointed out.
That its 1 big rambling mess i agree with, i ramble a lot i should have made it shorter and more to the point. But it is what it is. Its kinda useless to talk to the dude he's so stuck in his own delusion.
TL;DR
I think what he means is that Michael sucks, but I'm not sure because I didn't read it either.
Wonder what Michael did to infuriate all those guys that seem to hate him. I've been following and supporting him ever since he started making his comparisons and videos many years ago and he's a very nice guy and in fact I think he doesn't get enough credit for his excellent work, which is as good as DF and sometimes even better. Apparently Horizon runs now amazingly on PC and if you say otherwise you're a console warrior and a shill. I doesn' t matter that PC player have been complaining about this game's performance since it launched.
There is no hate here.
Dude is just a massive troll because he lacks arguments and comes with idiotic counter arguments that make him look even worse. His video's are fine and some of his other video's like his SSD comparison are absolutely god tier. The problem starts whenever PC gets involved the quality of his video starts to go down straight to console warrior level because either he's so clueless that he makes all the wrong outputs or he's straight up wanting to create them. Either way its misleading as shit.
The Bolded is VERY untrue, you are confusing what you own (and/or friends) to the Majority of the market, here check for yourself.
CPU by Core
GPU by Model
Most Popular resolutions, more people play sub 1080P than above 1200p.
The rest of your arguments works as well against it as for it, for example you state:-
"1) no clue what cpu they have to start with as steam doesn't say anything about architectures, they just say 4 cores / 6 cores / 8 cores ( far more then ps5's sold by the way ) and majority of them are intel. a 6 core of intel = dwarfs your cpu big time as its a minimum of 8700k at minimum in single core performance and even budget 4 cores already steamrol a 2700 for a while now in 95% of the games. Hell as its intel most of those 4 core chips steamrol a 2700 and we are talking here entirely in far far far more hardware then PS5's ever sold on top of it."
This is true for both sides, STOP SHOUTING AND BEING ALL ANGRY, about a test against PC specs that are not only at the higher end of the market they are in use. This was not and never was a PC v Console test, it was context for the BC quality of the PS5 I have stated this in the video and on here.
The drops are not all related to CPU, the battle with the Sabretooth at the start for example is 100% GPU bound from a fill-rate perspective from the alpha and particles. I could have spent ages going through the various bits but my post above you quoted highlights, a game is never just limited by 1 thing. Hanging your hat on the 2700 being the sole reason is not only foolish but wrong.
Stop talking in silly, childish tones with your little soundbites "Idiotic, his main mistake, calling out this clowns" listen to yourself. The only Clown here is not me, Bozo!
Well i am a big boy so i will say sorry for the rambling and the rant tone the reaction had. I can see how this isn't taken well by people and frankly i should have spended more time on it even while its not worth it.
Anyway, you have no arguments worth of shit, and that's your problem. U have no clue how PC hardware functions if you compare a potato PC on settings its not build for. 1440p or even 4k with dynamic resolution a 2070 is laughable idiotic. a 1080ti i wouldnt'even call a 4k card unless u move to 30 fps. 4k cards are the card i mentioned and all of them need trade offs. I will go on this further down in my reaction why its so stupid to even do it even while i already explained it to you multiple times by now.
Most Popular resolutions, more people play sub 1080P than above 1200p
My point was 2070, isn't a 1800p or 4k or even 1440p card its a 1080p at best in current days for reasons i stated before. But it can do 1440p obviously which i also stated before. Still it never was a card build for high resolutions only a select few cards ever where because of the design of them, the same way a 3080 isn't a 1080p card. I will further on this a bit later.
CPU part i already adressed multiple times by now can't be arsed to explain it again.
Not really. The arguments i made where to showcase you that your arguments don't work. That was the entire goal of those arguments. U still don't seem to realize that.
It is when u compare it and draw conclusions like u did at 12 minutes. which result in people actually looking at your comparison and come to the conclusion its flawed.
You test hardware that is not made for the goal you are trying to archieve, u test against a product thats superior on multiple levels and a game designed around those specifics specifically on top of it. Then draw conclusions that are worth the title misleading and honestly just feels like u are trying your hardest to get a result which makes you a console warrior but on a far more sneakier level. and i call you out on it.
A good example is somebody doing the exact same thing but the other way around. Because he wants to see PC win here.
Instead of dragging PC down towards console levels like most of those youtubers do, they drag the console down towards PC levels and come to the conclusion the PS5 isn't faster then a 1060.
Is it true what he showcases here? sure. Is it the reality of what u can expect in games not at all. a 1060 gets steamrolled by the PS5 on every level.
The same way as u can't test a 3070 laptop with control, compare it with rt and dlss towards a PC amd card and come to the conclusion its a 6800 equally. Is it true, yes. Is it the reality in 99% of the other cases where the product is actually used for? no. ( DF )
Hardware unboxed also has a good trackrecord on this department.
Tests a AMD card, remove RT and DLSS out of it, make a sheet of 10 games with performance results and then make a best buy for dollar sheet after it, with at the end they cover a simple sheet on RT and DLSS and its not taken into account on any of those sheets. Then draw the conclusoin 6800 is better then a 3080. While if that guy buy sa 6800 and plays control his experience is nowhere near that of a nvidia user when RT gets activated.
I can go on and on, at the idiotic tech youtubers how they bend "correct" information into a delusion that fits there agenda and make it misleading all day long.
U do the same thing here, u ramble some numbers here and there like full rate and 85% of the market and fall flat on your face because u don't even realize non of that matters, what matters is you testing hardware on settings that it isn't builded for and drawing conclusions out of it that are simple misleading.
A 1060 will never perform good at 4k, its not builded for 4k, its a 1080p card, a PS4 pro was builded for high resolutions, not high framerates. Comparing it towards a 580 RTX card in counterstrike will make the PS4 Pro look incredible bad and conclusions drawn out of it are not worth much for multiple reasons. That's the reality.
The fact u miss the point with PC tech so much, u have no clue what u are doing and u make conclusions and come with the dumbest arguments to defend a already lost conclusion i either classify you as a troll or simple a console warrior at this point.
Anyway, i don't hate you or anything i don't really care who u are. And whenever i see quality i will give compliments without effort. the SSD for the PS5 ( i don't even own PS5 ) is mighty appriciated for the quality of footage that was.
However with PC, u simple just make statements after statements after statements and arguments after arguments that are honestly just beyond idiotic and that's the reality i see. And with the label of tech expert on the matter it becomes even worse.
All the other info, why something is wrong i provided in my earlier reactions, i hate to explain it twice just read through the ramble diarrhea and there you go. I have no interest in lecturing people that simple refuse to learn by repeating myself.
Like i said many times to you, learn and improve. Stop sticking your head in the sand. because it isn't helping you.
Last edited: