Gamers Nexus pits the PS5 against a 4-year-old $500* PC. The results may surprise you. (Yes, clickbait title)

Woow to see some console player become so defensive is so hilarious. LOL

@drserigala:

To be clear; the PS5 has the equivalent of an rtx 2070 super and a zen2 8c/16t @3.5ghz inside of it. People are under the impression (or trolling) that a 1060 has it beat. The test is at 1080p, meaning cpu bound and the 3300x has higher boost clocks. Now the PS5 cpu has the potential to do better but you'd have to take advantage of the extra cores and threads and that's not going to happen on previous gen games unless the devs overhaul. If you want to compare a 1060 and the ps5 gpu you'd need to crank up the resolution to 4k.

Now obviously with the advent of RTX 30 series you can get more powerful hardware but a console is currently a really sweet deal. Sony and microsoft are eating a lot of the costs to get them out at the current price point.

@longestsprout: sorry but gamers nexus is pretty thorough with his testing and goes the extra mile to ensure a fair test. You can’t dispute the results, there is a serious bottleneck issue regardless. If you really believe the ps5 is running anything near a 2070 super then you haven’t gamed on a pc that is actually that spec. Like those that believe in flat earth seems console players aren’t willing to open up to the truth, even when the evidence is all around them.

Look consoles have their place for those that want to get into gaming that is more accessible, it is a good deal but it isn’t a deal that is getting you mid/high end of performance at a fraction of the cost like you believe

i mean i love to see other sources.

Not buying it. I watched the first couple minutes of that video and could immediately see higher texture resolution on the ps5 side with the naked eye. Even with compression artifacts and all.

That being said, the 1% and 0.1% lows are kinda concerning. I have to hope its just an issue with those specific games, or some sort of optimization issue.

In theory if the ps5 has a zen 2 at 3.5 ghz, it should be a bit faster than an 1800x, but a tad slower than a 2700x?

There are other factors though. There are a lot of bottlenecks, concerns about cooling which consoles tend to be very poor at. A lot of things can drag down performance. That's one of the reasons I will always be a PC gamer, because I get to pick and choose what goes inside it, and I have enough volume to stick a massive CPU cooler in there to not have to worry about the CPU or GPU throttling down due to heat.

Perhaps 3.5ghz is just the default clock speed but as the console heats up that throttles down as well?

Most likely memory limitation. v-ram is kinda shit for cpu operations and they probably run into this issue.

Yeah, if you're buying consoles because you want good hardware, you're wasting your money.

Don't half-ass it; build a PC if you want to do it right.

Indeed. My GTX 980 was running everything at 1080p (generally medium-high settings) perfectly well a year ago before I upgraded.

Yeah, and just shows how much Sony or even MS skimped on their CPUs. Those terrible 1% and .1% lows should've been easily handled by the "Zen 2" CPUs these shitboxes have.

Should also pit it from a $ 400 digital ver.

No, actually longestsprout is correct. This is stupid comparison since they are literally testing a GPU vs a CPU limitation. I haven't watched the video but I'm pretty confident that GN never compared the GTX 1060 vs the GPU in the PS5. Rather, only the performance of a older hardware vs a PS5 performance at 1080p. (Notice, it's not GPU vs GPU). Again, haven't watched it but I'm pretty sure GN and Steve Burke isn't stupid enough to claim GPU vs GPU performance here. Folks here shouldn't put words in GN's mouth by saying it's a GTX 1060 vs PS5. I'd actually be quite surprised if Gamers Nexus and Steve would do something this stupid if true. Whether it's GN or Hardware Unboxed or other YouTubers---they all know that at 1080p, the performance is CPU dependent.

It would be more fair to compare the CPU performances for this test. If you really wanted to to go GPU vs GPU--they would have set for GPU bound testing results at a higher resolution from 1440p to 4k---if there is actually a native 4k game on PS5 then match the settings. This way, the CPU workload is removed.

I think most of the poor 1% and 0.1% lows are because of the tested games favor single core/thread performance. And the PS5 uses a laptop Zen2 CPU. A game that's optimized for 4 threads will run considerably better on the Ryzen 3 3300X (desktop Zen2 CPU) than on the PS5.

I'm certain the PS5 will fare better than the Ryzen 3 3300X in heavily multi threaded games.

The PS5 GPU is about half the performance of the Radeon 6800XT (still well ahead of the the GTX 1060) but it's nothing to get overly excited about.

@cdragon_88: again that is your opinion, which is totally wrong. I think I will trust an expert when it comes to this stuff. It was the fairest test between the two he says why the comparison how you put couldn’t be achieved but even so the PS5 was incapable of keeping its settings at 1080p and had to drop to 720p plus no crowds in dirt 5 wow great power.

I have a pc with a i7 4770k processor a 7 year old cpu. And a 5700xt gpu and I can assure you the bottlenecks being seen aren’t their with my hardware even at 1080p! But those that want to hold onto a belief will be unwilling to see what is in front of them and make selective excuses for the results being wrong.

On BL3 the PC’s low settings still produced better and clearer graphics so they couldn’t even get the pc low enough to match the poor ps5 version

Here's what I have:

I can assure you that gaming at 1080p provides a CPU dependency AND that it is common knowledge.

@cdragon_88: lol didn’t need to know what you have that doesn’t make you an expert unlike gamers nexus, I pointed out what I have because if the bottleneck is severe as what you are suggesting that when you go to 4k and RT you suddenly get high end performance then that is insane. That is what I’m pointing out, and I have a old cpu running on a fairly new gpu so have experience if this. I’m aware a lower res produces more heat on the cpu but not to the levels you are trying to suggest here. And if the cpu is that bad in the ps5 even at 4k it will struggle. They could have easily got more gpu dependant at 1080p if they used much higher settings other than low/medium but guess what the ps5 was incapable of pushing it.

I use Msi afterburner to record actually looking at these changes and what the settings do and how much the cpu or gpu is being used. So I have actually done the testing on my own pc to see what is being pushed. I have never been bottlenecked to the point where I have to turn settings from ultra to low/medium... so explain that

Anyway nice hardware there I have the money to build a new machine but can’t get hold of the cpu and gpu I want right now unfortunately

Yeah, I don't think the testing method is really accurate here.

Here's what Digital Foundry found when testing AC Valhalla:

Based on tests with a 2080 Ti, it looks like a 2080 Super or RTX 3060 Ti would be required to match or exceed PlayStation 5's output. However, based on my tests with a Navi-based RX 5700, I'd expect a 5700 XT to get within striking distance of the console's throughput.

So, looking at the PS5 box, it clearly says "4K120." People actually believed the box? They really thought they were getting an 8K ready, 4K120 gaming console? Are "gamers" really that easy?

A PS5 will beat any 1000 series card at 4k. I don't see your point .

*sigh* here you go. https://youtu.be/ko2KTKOcQQ8?t=456 Let the Hardware Unboxed expert tell you. You can look up CPU benchmarks from GN too if you want. . Notice how as the resolution goes up---the frame rates even out. Someone else has already stated the difference in CPU cores/threads and clock speed and what that equates too for less/more performance. Yes the results at 1080p on PS5 with GN's strict parameters test was disappointing---but to say that PS5 is weaker than the GTX 1060 is ludicrous.

@cdragon_88: I’m not interested in the tear down seen loads of them, the hard facts are the actual results in comparison. And those don’t bold well for the ps5, it has to run games in low/med I’m sorry but those are the facts. Lille I said I understand cpu is used more at lower res. Read what I said, keep believing you can get med/high pc performance from a ps5 for a fraction of the cost and cooling required. Not only that which you continue to ignore the ps5 was running at low/med settings not high/ultra either.

One thing we both agree on was the ridiculous comments that the ps5 had a 3600/3700 cpu, at least that has been put to bed

Damn can't wait for tormeltdowns and Quack.

At this point, I think you are just trolling. I'm just gonna leave it. LOL.

@silentchief: my point is a 1000 series card can play all games. Not just a sweeping statement like yours without anything to back it up

Ya know, playing through different games and checking out the quality and detail of different settings, There really isn't a ton of difference between medium and highest settings in most games. The amount of detail you lose to make games playable on older cards really ain't that much.

Ok go run DMC5 at 4k on max with RT on a 1000 series card and let me know how that turns out.

Or run AC Valhalla on max at 1440p and let me know how that turns out as well.

@cdragon_88: typical of someone who is unable to back up their claims to spout his a troll and then LOL... that doesn’t make you right , I have tons of proof to show the ps5 is no med/high end pc

The PS5 isn't running anything at low med unless maybe your talking about 120fps which nobody will use.

A PS5 so far is about on par with a 2070S better on some games worse on others.

@silentchief: you are making assumptions on what settings are actually being used and I’m certain you are comparing those results to what you see when everything is on ultra for pc. I await a proper test at 4k. And be interesting if it can outperform a 2060. As someone stated it is actually difficult to see much of a difference between medium/high and ultra when looking at a picture. You have to really inspect it, but the performance hit is pretty big. As I said consoles are good for affordable living room boxes, they aren’t gaming PCs.

I'm not assuming anything watch for yourself.

https://www.google.com/amp/s/www.eurogamer.net/amp/digitalfoundry-2020-assassins-creed-valhalla-ps5-vs-pc

The PS5 is mostly Ultra with some settings higher. They said it requires about a 2080 to match PS5 performance.

AC is an AMD sponsored game and hence runs better on AMD hardware. This is why in the article you posted they list the 5700XT as the AMD equivalent which you omitted. I'm going to assume you omitted it because you know the 2080 is the higher performing GPU in 95% of games therefore making your argument seem stronger since the PS5 being = to the 8700XT doesn't sound all that impressive. In reality its just another game in a long line of poorly optimized games from Ubi.

@silentchief: an Ubisoft game, badly optimised but also weren’t they renowned for gimping the pc version of their games to make consoles look better? Just can’t go on such a test when Gamers Nexus actually used a variety from different developers

Ubi games are not exactly great on consoles either.

Well will see as time goes on but all proof we have shows it beats the 1000 series of cards pretty handedly.

@silentchief: which pretty much makes the whole test mute in itself

@silentchief: you do know the 1080 ti can hold its own even against a 2070...

The leap between the 1000 to 2000 series was nowhere near as big of a leap between the 2000 and 3000 series. You do those cards an injustice

I guess if the PS5 gpu is equivalent to a 1060, then the series x gpu is 1050