Value of Hardware Unboxed benchmarking

Maybe it started as that, but let's be honest, many commenters here have made their mind up about whether HUB is a trusted source or not a long time ago, even before this thread was created. This thread is largely a circle jerk of anti-fans hate-watching HUB content.
I refer you to my comment above:

(I don't know if this thread should be renamed "HUB Discussion" or what, as I still don't know or understand its/their contribution.)
People who have made their mind up HUB has no value - why are they posting here? People who have already decided it's trusted? Why do they post here? Is this just a thread to collect HUB content so it isn't elsewhere, as just a waste-bin to collect it? Should HUB be banned from B3D? Should we rate and approve/cancel content we don't like? ¯\_(ツ)_/¯
 
People who have made their mind up HUB has no value - why are they posting here? People who have already decided it's trusted? Why do they post here? Is this just a thread to collect HUB content so it isn't elsewhere, as just a waste-bin to collect it? Should HUB be banned from B3D? Should we rate and approve/cancel content we don't like? ¯\_(ツ)_/¯
I personally think a waste bin thread is good to have, not criticizing that. I know it works well in some other forums and helps keep other discussions cleaner and more constructive. Plus, I sort of enjoy arguing with super biased people sometimes, they can be impressively imaginative.

Just wanted to call a spade a spade, or rather a hate-thread a hate-thread.
 
Last edited:
Tim and Steve had an interesting exchange on their latest podcast on AMD pricing strategy. Seems they’re not quite on the same page on how to measure value. HUB’s standard measure is raster perf/$ with a cursory acknowledgement that other features also matter. They’re likely (or at least Tim is) feeling a little pressure to expand the definition to account for other things gamers seem to care about like RT, upscaling, encoding etc based on actual sales.

Steve’s take is that it’s fine to keep measuring value based on raster performance but AMD needs a bigger discount vs Nvidia on that metric in order to move units. Tim is leaning in a different direction where AMD needs performance and feature parity (RT included) plus a big discount. His example was that it’s awkward to talk about the 7900 XTX raster advantage over the 4070 Ti Super while ignoring the significant RT deficit vs the same card and declaring the AMD card the better value option.

This has been obvious for a while now but at least half of the HUB duo is coming around to facing reality.
 
Why do you think HUB's stance is that upscaling is not important for consumers? Where do you get the info from? Please provide quotes.
Maybe because it doesnt matter when they recommend GPUs. Here is their list from December 2024 with the $700+:

Let me quote "[...]The DLSS vs. FSR battle is less important if you play a game above 1440p. For example in 4K there is very little difference in visual quality between these upscaling methods.[...]"

Tim is just good cop, Steve the bad cop.
 
Pure raster performance metrics seem less relevant in Professional situations where Cuda and Optix software takes application results to new heights on NV products. Proconsumers endgame is to use the most performant option available. It is not to much of a stretch to apply similar acceptance to GPU game performance measurement situations.
 
Maybe it started as that, but let's be honest, many commenters here have made their mind up about whether HUB is a trusted source or not a long time ago, even before this thread was created. This thread is largely a circle jerk of anti-fans hate-watching HUB content.

Now, you could say that I'm just here to hate-watch the hate-watching. I won't deny that to some extent that is the case. But I also like arguing to separate shit from shinola and I get the feeling that many of the claims and accusations made about HUB's content are weak or outright baseless and when I challenge those claims or ask for clarification, I often get no response. That's lame, especially if the accusation/claim was delivered with a derisive tone as is sometimes the case in this thread.


Why do you think HUB's stance is that upscaling is not important for consumers? Where do you get the info from? Please provide quotes.

Quoting Tim's article from last July in which his writing clearly indicates he thinks upscaling provides value for the consumer:



Quoting Tim's article from April 2023:



Links to sources:

1) https://www.techspot.com/review/2860-amd-fsr-31-versus-dlss/
2) https://www.techspot.com/article/2665-dlss-vs-native-rendering/

PS. Could you clarify where Tim indicated has a motive to punish Nvidia by showering his audience with negativity? Or was it a strawman and he actually just wants to dispel unwarranted hype by the vendor?
Thank you breaking the Nvidia circlejerk and injecting some objectivity and sanity into this thread. It was becoming nauseating.

As for whether or not this thread is worth having or not: Yeah, sure, at the very least it allows you to see which users have a clear axe to grind.
 
I was addressing the claims made about HUB specifically, so this is kind of besides my comment. But if "everything sucks" is a more common vibe in the PC hardware youtube, I'm kind of wondering why on this forum only HUB has a 80+ page thread with a mostly negative vibe. I see a thread about Linus with 7 pages... and that's it. Do other outlets get called names like AMD Unboxed here? I don't think so.

So there's gotta be something about HUB specifically that's wrong, no? Or maybe the HUB negativity is just an echo chamber thing on this forum?
I can't speak for anyone else but I feel that HUB is very much part of a wider trend on hardware Youtube (as well as other media) of spending a lot of time harping on things they don't like.
Most people who end up in the creator space initially start out of love for the game. But that game is always changing, and some will lean more naturally towards rejecting new developments, while others are much better at adapting.

I call out HUB in particular for a combination of reasons, nor least among them of course their undeniable success. They've established themselves as one of the most influential and widely referenced review channels out there, and it is well deserved. They have put in many years of hard work, and I personally think their reviews are generally thorough and reliable.

However, that omnipresence also makes it very hard to avoid remaining unaffected by the misery that seems to have taken hold at HUB. After all, their talking points pop up everywhere, echoed by a legion of fans. Their videos jump at you whereever you go, embedded all over the net. Which isn't their fault, but it is another way in which they stand out.

Speaking of misery. Don't you get that vibe? I think the sheer jadedness practically drips off the screen. The joy seems almost completely gone, and whatever comes Steve's way ends up being reduced to this value proposition that seems to have reached its zenith sometime in oh, the nineties or so, and has been unacceptable since. There's all these newfangled thingamajiggies that he has to deal with now, and he just sounds so.annoyed. What's with this power efficiency, hardware encoding, GPGPU, ray tracing, upscaling, frame pacing, driver overhead malarkey? Harumph. Don't they know I've been testing 50 motherboards this week? I am totally bored out of my mind by all of this.

It's pretty much a tone thing for me, and some of the negativity in this topic is mostly just a reflection.

Now, I know The Way to get engagement on the Internet is to go bold and avoid nuance at all cost, so let me just end here by adding that I don't enjoy Gamers Nexus either. That smirking garden gnome's shtick may be quite a bit more energetic than HUB Steve's, but the self-righteous causes and the fucking sarcasm is getting really tired in its own right.

As for Linus, who cares? He's barely in the same space.
 
Last edited:
Impressions of FSR4 were really good, but we'll see when it's launched. Assuming game adoption is good, FSR4 could be a really nice win for AMD in terms of equalizing with Nvidia. I do think reviews are missing the mark on upscaling. It's very popular and people know DLSS is significantly better. Personally, I wouldn't buy an AMD gpu until they could match DLSS and Reflex. They have Antilag+ now, but I don't think the adoption is up to the same level. There's just a lot more to a gpu than calculating $/frame for rasterization.
The problem is AMD creates competitive solutions, but only 4 years later. FSR4 is probably going to be on par with 2023-era DLSS2, but I bet the new transformer DLSS model is going to be better still. Granted, at this stage we are getting so close to native-tier graphics (native with TAA I should specify) that DLSS could improve another 20% and I honestly wouldn't notice a difference, both would look excellent, but that just means you can push resolution further down.

Feels like AMD is always fighting the last war. They are trying to bring their Reflex competitor up to par but that's been a feature since Maxwell. FSR4 will finally surpass what DLSS was doing in 2020 lol.

I usually disagree with this forum on their assessment of HUB but one thing I think you guys get right: HUB's take on 'value' is completely detached from how 99% of people buy hardware. Nobody is using 'raster fps/dollar' to buy hardware. A bunch of my friends are building or upgrading their PCs (including myself although I'm not upgrading the GPU). Not only are they not even concerned with fps/dollar or whatever, but they would gladly pay an extra 20% just to get a white motherboard lol. The builds we are coming up with spend 20% of the budget on mostly superfluous lights and fans (for example, those AIOs with an LCD screen, paired with a CPU that definitely does not need to be liquid cooled lol). Personally I don't use much RGB in my builds at all but idk how people can look at how people spend their budget and conclude that optimizing for cost is how anyone builds a PC. In fact, I'd go as far to say that almost nobody building a PC is doing so for any concept of value or budget, if they were they'd just go buy a console as that comes out ahead 9 times out of ten.

Really, all of this right here. When DLSS first came out, I had my reservations and at the same time saw FSR as a pretty reasonable (if not a bit behind) equivalent. Give them time, let them cook, and surely AMD can come up with additional IQ just like NVIDIA was doing. For what it's worth, FSR did get better over time, there were several good review comparisons made which showed it wasn't completely stagnant.

But holy hell, DLSS is so good now, especially with the latest transformer models. Sure, those of us on non-Blackwell hardware will pay a bit more tensor perf penalty to run it, however the tradeoff is absolutely worth it. I can't right now imagine how FSR is ever going to catch up; AMD is seemingly unable to compete with essentially any of NVIDIA's software advances as of late. I feel DLSS "K" has to be literally years ahead of what AMD could do if they right now started fully staffing and funding a renewed and reinvigorated FSR tech.

DLSS has become the defacto reason why we, as a hardware enthusiast community, really have to consider how to define and measure "performance" -- because IMO native raster just isn't good enough anymore.
Agreed, at this point I don't even recommend AMD to people. An Nvidia card can run the same game at 1/3rd the resolution of what the AMD card is running at and still look better with DLSS. There is very little you can do with hardware that will close that gap.
 
What's with this power efficiency, hardware encoding, GPGPU, ray tracing, upscaling, frame pacing, driver overhead malarkey?
I don't really get this? This is covered in all of their reviews. Hardware encoding and GPGPU aren't really relevant to like 90% of people and the rest are covered ad-nauseum.

In fact one of their latest pieces was on Arc driver overhead and everyone here hated it lol.
 
Why do you think HUB's stance is that upscaling is not important for consumers? Where do you get the info from? Please provide quotes.
They say that in reviews, but then they go into podcasts or when they make best GPU videos and deny all of that and make their recommendations based on raster performance per dollar alone. Here you go, none of the advantages materialized into any best GPU pick, they are just footnotes mentioned to deflect off criticism of bias.

In fact, they always do this, flip flop on everything to trim down any NVIDIA advantage as much as possible. You want to know why most people in this thread are critical of them? it's because they contradict themselves all the time. They are the only channel that does this.

First, they claimed Ray Tracing and ML upscaling are not important, DLSS1 is bad and RT has bad performance .. right? AMD cards that sport none of these is a better value.
Then, when DLSS2 became a thing and RT expanded it's support, they still claimed DLSS2 is not available in many games, and thus is irrelevant, oh and RT still has bad performance so it's irrelevant too.
Then when DLSS2 exploded in adoption, they still shrugged it off.
Then they claimed FSR1 is on equal footings with DLSS2. Thus DLSS2 is irrelevant again.
What about RT? oh look UE5 games will replace RT, and HWRT in UE5 has similar performance for NV/AMD in Fortnite, so NVIDIA's RT advantage is irrelevant guys!
Then they claimed FSR2 has leveled the playing field with DLSS2, so DLSS2 remains irrelevant.
Then when it became clear that neither FSR1 or FSR2 is equal to DLSS2/DLSS3 they started to admit that DLSS2 is superior but didn't factor that in any of their GPU evaluations.
What about RT? oh it still has bad performance, so it's irrelevant.
Then when DLSS FG got released they started shrugging it off again, fake frame people.
When people called them off on this they claimed it affected latency, HUB cares about latency and high refresh rate experiences!
But where is your Reflex coverage? where is your latency measurement with high frame rates on NVIDIA and AMD GPUs? crickets.
Oh and RT has bad performance .. so it's irrelevant again! But let's test the same game twice in the same review because fraaaames and high refresh rate gaming. This is what's relevant. Let's ignore RT testing altogether.
People want RT testing? let's test Riftbreakers with RT and test Cyberpunk without RT, let's pick games with the worst possible RT implementations and test them to spoil the results.
DLSS2 is superior? noo let's test ray traced games with FSR enabled on all GPUs. That's how fair it should be!
Games are releasing with RT on as a base and with Mesh Shaders? irrelevant again! The 2060/2070 can't play these games at good fps anyway, the 5700/5700XT are still better guys. We never picked the wrong choice ever!
Oh and RT still has bad performance, let's call RT an upsell, let's downplay it's importance even when games rely on RT acceleration to function or perform well (Indiana Jones, Doom Dark Ages, Metro Exodus Enhanced, Avatar, Star Wars Oulaws).

Here is what I said about them 2 years ago about their contradictions:
1-They don't test RT because they say it's useless, doesn't add much of visual enhancements and brings fps down (fps is king), yet they test Ultra settings, which is also useless and bring fps massively down (according to their own logic). You either test with Ultra, which includes RT at max settings. Or you don't test RT or Ultra settings, and test with the most visually optimized settings.
2-Even though they started testing with RT recently, they test it in the most superficial way possible, picking up many games with useless implementations that doesn't add much of visual enhancements (in contradiction with their logic), while excluding titles with more potent implementations, and actual visual quality upgrades.
3-They scolded GPUs with 8GB of VRAM based on testing that used Ultra settings, not suitable for such GPUs, and not in line with HUB's high fps mantra.

These points still hold true to this day. Their recent podcasts showed even more glaring flawed logic and face palm moments.

You still want to know why so many people are critical of HUB? well you have a full history right in front of you in this thread. Just search my name and you will find every misstep they made with links, and timestamped videos.
 
The negativity would be there from consumers regardless of what HUB or others say. The influence of these reviewers is being heavily over estimated here IMO. HUB is popular because they benchmark more games with more settings variations and more hardware combinations than anyone else. Their reviews always have up to date tests for all components being listed. They don’t recycle olds results. Their output is frankly insane.

People will by and large make their decisions based upon the benchmarks, features and prices; not Steve giving his personal views on RT in a few short sentences at the end of a video. It’s fair to say that given Nvidia’s marketshare, HUB isn’t “tricking” people into buying AMD GPUs.
 
Last edited:
People will by and large make their decisions based upon the benchmarks, features and prices; not Steve giving his personal views on RT in a few short sentences at the end of a video. It’s fair to say that given Nvidia’s marketshare, HUB isn’t “tricking” people into buying AMD GPUs.
The primary complaint I see though is that the benchmarking isn't fair. I don't watch any vids so can't comment, but if they are picking tests that favour one hardware over another, that's incorrect information for consumers. Ultimately consumers should engage in a wider amount of research, but they generally don't, with people, certainly a sizeable proportion of, finding one voice they like and just going along with everything they say, and then defending that voice because it's the one they've chosen and they are committed now.

Hence the concerns about influence. If the presentation was (perceived as) neutral, there wouldn't be this thread.

Edit: The full title of this thread is, "Value of Hardware Unboxed benchmarking" - the dispute is entirely about what data is being collected and presented upon which, as you say, "People will by and large make their decisions based upon the benchmarks..." If the benchmarks are no good, the channel is no good for informing people.
 
Last edited:
The primary complaint I see though is that the benchmarking isn't fair.
Huh? Aren't all cards tested running the same settings and test platforms?
Just because you don't agree with their choice of benchmarks and settings doesn't make it any less fair. Heck, one could bench everything at 640x480 lowest settings and it would still be fair.
 
The primary complaint I see though is that the benchmarking isn't fair. I don't watch any vids so can't comment, but if they are picking tests that favour one hardware over another, that's incorrect information for consumers. Ultimately consumers should engage in a wider amount of research, but they generally don't, with people, certainly a sizeable proportion of, finding one voice they like and just going along with everything they say, and then defending that voice because it's the one they've chosen and they are committed now.

Hence the concerns about influence. If the presentation was (perceived as) neutral, there wouldn't be this thread.

Edit: The full title of this thread is, "Value of Hardware Unboxed benchmarking" - the dispute is entirely about what data is being collected and presented upon which, as you say, "People will by and large make their decisions based upon the benchmarks..." If the benchmarks are no good, the channel is no good for informing people.
Their benchmarks have never been found to be anything other than absolutely reliable AFAIK. The issue people had WRT fairness was HUB not dedicating enough coverage to RT and DLSS.
 
Last edited:
The issue people had WRT fairness was HUB not dedicating enough coverage to RT and DLSS.

This is not an accurate representation at all. Please watch the latest HUB podcasts. To this day Steve insists RT and upscaling are useless “in his opinion” but “some people might like it”. What’s the point of saying that from his bully pulpit?

The fact that his choice of benchmarks in the past have reflected that stance are the real impact. Long term it doesn’t matter of course as the games they benchmark will increasingly sport RT by default.
 
This is not an accurate representation at all. Please watch the latest HUB podcasts. To this day Steve insists RT and upscaling are useless “in his opinion” but “some people might like it”. What’s the point of saying that from his bully pulpit?
Who is he bullying from way up there? Hes entitled to his opinions. For how he games, he doesn’t see value in them. Does his personal opinion affect the data he puts out?

As for why he might say it, they are features you are paying a premium for. He recommends anyone who values RT to not even consider an AMD GPU.
 
Last edited:
Who is he bullying from way up there?

Ha that’s not what bully means in bully pulpit.

Hes entitled to his opinions. For how he games, he doesn’t see value in them. Does his personal opinion affect the data he puts out?

Of course he’s entitled to his opinions and as a pretty large channel we’re entitled to comment on his content. If you think his very vocal personal opinions don’t influence his choice of benchmarks well you’re very generous. You can just look at his reviews and recommendations and know that they do.

When he says stuff like the 4070 is useless for RT and therefore that you HUB viewer shouldn’t factor RT into your decision to purchase one do you find that statement fair and useful to a potential 4070 buyer? Hint: there are tons of people enjoying RT on their 4070’s.
 
Even his own colleague Tim thinks that a 7900XTX should be priced at no more than 700$ to account for the ray tracing performance deficit and feature deficit. Yet we never saw this assessment reflected in any of their reviews, top picks, or anywhere else.
 
none of the advantages materialized into any best GPU pick, they are just footnotes mentioned to deflect off criticism of bias.
Conjecture.

In fact, they always do this, flip flop on everything to trim down any NVIDIA advantage as much as possible.
Conjecture.

they contradict themselves all the time.
Funny how if that is the case that from that noise you somehow get a clear signal that their stance is upscaling is not important to consumers? Or maybe on your part it's just... conjecture.

These points still hold true to this day.
Your conjecture holds true to you, I'm sure. To me, it looks bordering on delusional. But hey, I'm no expert and I have an anti-anti-fan bias, granted.

However, I asked for quotes, but it seems conjecture really is all you have as in your huge wall of text you fail to provide even a single quote from recent times where HUB says upscaling is not important to consumers. I will have to assume that you failed to provide that quote is because that in fact is not their stance and your original claim was a strawman.

PS. Oh and for the record, you also did not provide evidence that Tim wants to punish Nvidia in his recent MFG video. I will assume that this was a bad faith interpretation of his motive.
 
Last edited:
Ha that’s not what bully means in bully pulpit.



Of course he’s entitled to his opinions and as a pretty large channel we’re entitled to comment on his content. If you think his very vocal personal opinions don’t influence his choice of benchmarks well you’re very generous. You can just look at his reviews and recommendations and know that they do.

When he says stuff like the 4070 is useless for RT and therefore that you HUB viewer shouldn’t factor RT into your decision to purchase one do you find that statement fair and useful to a potential 4070 buyer? Hint: there are tons of people enjoying RT on their 4070’s.
I think HUB has portrayed AMD’s RT deficit quite accurately and in line with their peers. I find their game selection to be highly varied, fair and resulting in performance summaries quite similar to other respected outlets.

I don’t watch their podcasts but in his reviews he most commonly prefaces such statements with the qualifier of it being his opinion.
 
Back
Top