• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

NVIDIA GeForce RTX 5060 8 GB

I'm not sure how I feel about the 5060; On the one hand it's mostly better than the 4060 for the same money. On the other hand it's 30W hungrier than the 4060 and it still only has 8GB, two years later.

The 4060's saving grace, IMO, was that it was good enough at just 115W, and 8GB despite being questionable back in 2023 was an acceptable amount of VRAM for an entry-level GPU. Not a good amount of VRAM, mind you, but it juuuuust scraped past the minimum bar for a trouble-free experience.

The 5060 has a lower performance/Watt than the 4060 and 8GB isn't enough to do all the things Nvidia is advertising in 2025 titles. In an ideal world, that makes the 5060 a $249 card, but Nvidia gotta Nvidia...
 
Sadly the current leader for fastest low profile card on the market.

No idea if the sparkle b570 low profile will move the needle there at all.
 
The problem is the majority of buyers are going to be people buying prebuilts who have no idea what they are actually getting only that it says Nvidia on the box so it must be good.

I think people in the diy market for the most part know better unless they can't spend a dime over 300 and need a gpu becuase their's died etc.
Exactly, prebuilts are where the configs and products we see as "not worth it for what it is" end up because the diy market knows better, and knows where corners could be cut.

But alas, the average joe just wants a PC, so they buy whatever prebuilt looks nice to them. And NVIDIA's branding rightfully built up weight over time, just now, it's cracked and most can't see it's being patched up with sly methods
 
"Excellent price/performance" "$300 low price point"
I don't understand how this card supposed to be excellent price to performance when it doesn't beat out the previous gen 4060Ti 8GB.
The card would be better price to performance at $200-250, not at $300, and the 5060 will probably remain close to MSRP because of how mediocre it is.

That's mostly a reflection of the 2025 gpu market more than this card being good.

I don't envy reviewers in 2025 though it's difficult treading the line and trying to show the pros/cons of a product without coming across as being biased for or against a specific brand.

Nvidia is definitely not helping them out lol.

I'm sure W1z put a ton of work into this review and probably knew regardless of how he felt about it there would be backlash for or against but I honestly think reviewers should just say what they feel about a product regardless of what the internet reaction will be and I feel tpu does a pretty good job of that.
 
100%

But in 2025 99% of their focus is on the AI market I honestly believe they fully believe releasing this gpu at 299 is a charity to gamers.

Certainly they don't need the revenue but without a lower end GPU for pre-builts and the bulk of the market, many prebuilt partners would be forced to move to AMD / Intel. Once companies make that move, it's probable that a significant number would choose to keep doing their business with AMD / Intel simply due to inertia and outstanding contracts. Plus bowing out of the market like that leads to instability and that's something companies will try to avoid at all cost. If Nvidia gives them a reason to think they'll leave a market whenever it isn't profitable, that will impact their decisions when designing Prebuilt SKUs. They are going to lean towards the more consistent partner, whoever that might be, thus the impact of such a move would have implications for years even if Nvidia only skipped a single xx60 class GPU.

The customer gets a cheaper GPU than a 12 or 16GB GPU would cost. That's the benefit to the customer.

A curious conclusion given that this bracket of the market has seen shrinking die sizes and the same 8GB of VRAM gen over gen despite the cost of that same 8GB plummeting over time.

You claim that lowered costs will get passed onto consumers, and yet at least the last 3 generations of Nvidia GPUs have done the very opposite of that.

It's entirely up to the customer to decide if it's worth it for them, Jensen isn't breaking into people's houses and forcing them to buy 8GB GPUs at gunpoint. People buy it because it's worth the money to them, simple as that. If it's not worth it, they won't buy it.

What percentage of purchases nowadays do you believe the average person thinks is actually "worth it"? Particularly in the GPU market where prices have exploded over the last several generations.

I'm willing to bet that if you did a poll on PC gamers regarding if GPUs were worth the price, a good number of them would have responses ranging from "disagree" to "heavily disagree".

There are many forces that can coerce a person's purchasing decision. You are assuming everything is a free and fair choice but corporations nowadays employ tactics specifically designed to rob customers of as much power and choice as possible. Nvidia's locked down proprietary tech is the perfect example of that as is their control over the AIBs and partners. It's about reducing people's choice in the market, gaining leverage, and blocking competition.

Maybe in 2025, more people decide 8GB for $300 isn't worth it for them. I am 100% open to that being a possibility. If that happens, Nvidia will lose money and and not sell GPUs. So Nvidia has a very, very strong incentive to sell what people want.

It's unlikely to happen, merely because the company that sells 90% of all GPUs doesn't have a GPU with more than 8GB of VRAM within the vast majority of people's price range.

In addition, you are under the impression that it's within the consumer's scope to change this but the fact of the matter is that most GPUs are sold as part of prebuilt PCs and Nvidia owns the contracts for the vast majority of them. As we've seen in the laptop market, it takes years to change things over. AMD hasn't done it yet over Intel and that with it being in the lead architecturally.

At the end of the day, most OEM PCs will have Nvidia GPUs and customers won't even have a choice. On top of that most customers who buy pre-built aren't even aware that AMD is a thing and will buy Nvidia by default. Customer awareness and perception takes years to cultivate. Plenty of normies still don't know about AMD in the CPU market even.

The only portion of the market that customers are likely to influence is the custom built, and there aren't any exact numbers on how much of the total market that represents other than it being smaller than OEM / prebuilt. It's also statistically unlikely that a significant number of those people will suddenly start dropping a ton of extra cash on their GPU.

Either way, I'm not bothered, because (a) it's not my money, I just buy what is worth it to me, and (b) I'm pretty sure a $3 trillion company has put at least a little bit of thought into what the market is willing to pay for GPUs (i.e. they've thought about what people want to buy).

You'd think but given how well this generation has gone there's some serious evidence to question this notion. We shouldn't just assume that the high dollar value of a company means they do everything perfectly for the same reason we shouldn't assume that an electrical engineer is always right. Expertise in a field isn't a shield against mistakes, concerns, or errors. In fact it often works in reverse and the only antitode to pride is humility.
 
small typo, one pcie 8-pin provides 150W (at least officially) not 300W as claimed.

@W1zzard
 
A curious conclusion given that this bracket of the market has seen shrinking die sizes and the same 8GB of VRAM gen over gen despite the cost of that same 8GB plummeting over time.
Die sizes are shrinking because nodes get smaller, and Nvidia has chosen to prioritize cost and power consumption instead of performance gains in this segment.

The 4060 sat at $300 (MSRP) for most of its lifecycle, while the 3060 was $400 or more for most of its lifecycle (with a $330 MSRP). The 5060 has only just launched, so I won't speculate about what the future holds, but at the moment, there's $300 MSRP models in stock. As I noted above, the cost of 8GB is not the cost of the VRAM chips. The cost is in the design of the GPU, the traces, the memory controller, etc, and of course the market pressure of a card with more VRAM being more valuable for AI purposes (which drives up cost from competition, see 4090 and 5090 launches). Nvidia has likely looked at the market and concluded that a 12GB 5060 would have a street price too high for the mass market, not to mention that there's not sufficient supply of 3GB GDDR7 chips yet (Nvidia is prioritizing their professional chips for those first, as they have much higher margins).
What percentage of purchases nowadays do you believe the average person thinks is actually "worth it"? Particularly in the GPU market where prices have exploded over the last several generations.
If they buy it, it's worth it. That's the definition of a purchase. If someone isn't holding a gun to your head, you exchange money for a product because the product is worth more to you than whatever else you could do with the money.
It's unlikely to happen, merely because the company that sells 90% of all GPUs doesn't have a GPU with more than 8GB of VRAM within the vast majority of people's price range.
You do realize that if people no longer want 8GB GPUs, that means Nvidia looses a lot of money, right? If they decide 8GB is no longer worth $300, Nvidia takes the hit. And I am perfectly fine with that. If Nvidia makes a product that won't sell, that's their fault.
You'd think but given how well this generation has gone there's some serious evidence to question this notion. We shouldn't just assume that the high dollar value of a company means they do everything perfectly for the same reason we shouldn't assume that an electrical engineer is always right. Expertise in a field isn't a shield against mistakes, concerns, or errors. In fact it often works in reverse and the only antitode to pride is humility.
Sure, sure. And again, if Nvidia has misjudged the market, they will lose money. And that is totally fine by me. If you don't make a product the market wants, you will lose money, that's how it works.

My point is that for the last few years, the general market has shown they are perfectly happy with 8GB GPUs, whether it's the 4060 or 4060-Ti, no matter how loudly enthusiasts screech and gnash their teeth and rend their garments. The TPU forums and Reddit are a tiny echo chamber. Maybe that changes this year, maybe it doesn't.
 
Considering this card has a decent performance improvement over the 4060, I really don't understand why nvidia went to such lengths to block proper reviews before launch day. It really seemed as if they were trying really hard to scam consumers using mfg to cover a null improvement over the 4060, when the card is actually quite a better product than the 4060, even without mfg.

Honestly, given how mediocre the rest of the blackwell cards have been, blocking reviews would have made more sense in almost any other blackwell card other than this one and the 5090.
 
Best price to performance at 1080p and if you ignore the fact that you could have gotten more for your money two years ago, that you can't run RT or PT, that you are going to have to lower graphics settings, and that the card won't last long.
I'm aware, it's just a stress on how competition either doesn't try (AMD) or can't try (Intel) to make a change.

5090: but ridiculously expensive + THIS connector.
5080: but 16 GB.
5070 Ti: but expensive.
5070: quack?
5060 series: we have all sorts of inherently wrongfully designed GPUs. But they're somehow the best value out here.
9070 XT: but MSRP is utterly fake.
9070: only exists to upsell the GPU above.
9060 series: should've been released forever ago, still don't exist. Also seem to offer very little over 5060 series.
B580: but sucks if paired with a relaxed CPU.
B570: but doesn't offer better value than 580.

Yahoo I guess.
 
Well it's small and low power which is great for many people, but no GPU should come with 8GB VRAM anymore.
 
no GPU should come with 8GB VRAM anymore
I'd be totally fine with a GT 1030-alike tiny card that only wants fiddy watts and takes a couple cubic inches of space that "only" has 8 GB on it. Unless they ask more than $180 for it.
 
I rarely write comments but yeah... this card is such a good joke. I had to log in and say something about it. I thought 4060 was horribe, but this is a new low. It's 2025 for god sakes. Can't even beat 3070 from like 500 years ago. Same VRAM too. Can barely beat last gen cards, and in fact... it doesn't even beat the 1 tier above it. JOKE JOKE JOKE.

When i hear the name Nvidia, i literally laugh now. Funny how this diamond of a company became pure trash. Look at that 5080 vs 4080 performance too. I still can't get over that 1 haha. People waited YEARS for that sweet 8 extra fps ;) Anyways, 5060 is a total skip and ofc it will become the number 1 most sold card :) So yay? This is why we can't have nice things. They literally give us horrible products and people eat them up.

So yeah, hate me all you want for speaking out my mind. It wont change anything. I don't see any value in 5060, 5070 and 5080.
 
Let the scalping begin.
 
That being said any card that can't beat a prior generation gpu one tier up by a noticeable margin is a joke while having less memory than 2 generations ago is just sad.
This is a fair point. The RTX5000 really seems to be an RTX4000 refresh like the GTX400->GTX500. A little different but more or less the same situation.

That said, I still think this is a decent value card.

@Zotac
Props for getting W1zzard a sample for testing! Damn decent of you folks! :rockout:

@W1zzard
I noticed some MSI info in this review. Is there an MSI model review coming?
 
So the 5060 actually top the price to performance chart.
performance-per-dollar-1920-1080.png


Pretty funny when all the rage bait techtubers will find themselves becoming the game critics (where game get poor scores for not including certain things), while the audience will find the 5060 attractive
 
small typo, one pcie 8-pin provides 150W (at least officially) not 300W as claimed.

@W1zzard
Fixed

I love the fact that the first review of 5060 on TPU was a model with one fan. It speaks miles, more than numbers. Great job.
It wasn't specifically selected in any way. Zotac's card was first to arrive here, MSI's 5060 Gaming is the next review

I noticed some MSI info in this review. Is there an MSI model review coming?
See above
 
Why is this review so much more positive than the ones on youtube? Honest question....I haven't had the hour necessary to compare numbers, but if somebody has and would like to provide their take on it.....
I personally think based on the review it is because the reviewer has a positive opinion of DLSS and Frame Gen etc. Most other sites and YouTube channels are over it. It is like the Hardware Unboxed guys said if they could do all of the Frame Gen and DLSS stuff without any negative effects then it might be a slight benefit but when you already have a lower end system that is struggling adding more negatives such as latency just make things worse. This is from the clip they addressed the issue in their latest video. Video should start at timestamp for the section talking about Frame Gen etc.
 
Does the VRAM run out in Cyberpunk with RT, Balanced or Performance DLSS, Ray Reconstruction, and frame generation?

I got ~60 base fps with RT, Balanced DLSS (transformer), and ray reconstruction on my 4070 Super at 1440p. Then FG bumped me up to a smooth ~110-120 fps.

Looking at the benchmarks, the 5060 does better at 1080p RT than the 4070 Super does at 1440p RT. So, if VRAM doesn't run out, it should be a pretty good experience in that specific game when you add all of those features.
CP2077 can run out of VRAM for 8GB cards in that situation. I played on a 3070 with Digital Foundry recommended RT Ultra settings, at 1440p ultrawide with DLSS either Balanced or Performance. My 3070 performed acceptably for most of the game, but eventually the slowdowns got more severe, and by late game I would drop from 50 FPS to <20 FPS after exiting menus. The solution apparently is to drop textures to medium but I haven't tried this yet. (I gave up playing due to the slowdowns).

Digital Foundry's DF Weekly show recently discussed the challenge of coming up with optimized game settings for 8GB cards, because Alex has to play the entire game on an 8GB card in order to find the portions with the most VRAM pressure. If you own an 8GB card, expect to spend an extra 30-60 minutes with most AAA games researching which graphics settings you need to drop. TPU's performance per dollar chart doesn't lie: the 5060 is the best value card. As long as you value your time at $0/hr.
 
Last edited:
After seeing some of the comments of the review i just wanted to say that as a owner of a low 50 series card which i use mostly as a workstation and some gaming on the side (5060ti) my experience is actually exactly as the reviewer says, if i have 45-50 frames as base activating frame gen x2 is actually a good experience in all the titles i have played so far and DLSS quality even at 1080p would surprise you with the image quality it produces.

I can understand the dislike towards these technologies but for those not seeing any value i ask that you to check how the 5060 ti 16g card handles PT and RT on the new F1 game when benchmarks comes out, it may surprise you and realize that there was no bias from the reviewer.
 
Low quality post by Neo_Morpheus
Excellent price/performance

how in hell can this be real

In a better timeline this would be a bad release. It's all contextual. Case in point:
"Excellent price/performance" "$300 low price point"
I don't understand how this card is supposed to be excellent price to performance when it doesn't beat out the previous gen 4060Ti 8GB.
The card would be better price to performance at $200-250, not at $300, and the 5060 will probably remain close to MSRP because of how mediocre it is.
 
Back
Top