Topher
Gold Member
Personally I liked VG Tech better. No commentary. Just data. For some reason he stopped making videos a few months ago though. A shame.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Serious discussion. Should DF be treated as biased source?
- Thread starter akira__
- Start date
Personally I liked VG Tech better. No commentary. Just data. For some reason he stopped making videos a few months ago though. A shame.
adamsapple
Or is it just one of Phil's balls in my throat?
VGT's videos are / were great just for the broad metrics like fps and resolution but he barely conveyed other graphical differences. I don't think he did PC content ever either.
IMO some commentary is needed in these kind of videos to show and explain to the average viewer what's being shown on screen.
Lens of Truth was the real OG tho.
Last edited:
Topher
Gold Member
That's fine, but end of the day folks who live and breathe these comparisons end up focusing cherry picked images of frame rates. VG Tech cut through all that bullshit.
Danny Dudekisser
I paid good money for this Dynex!
I like John and think he's genuinely enthusiastic about this stuff. I respect that.
Alex is a loser. I appreciate that he acknowledges that PC games have stuttering issues, but beyond that, he has nothing of value to offer.
It's an okay source. Watch the videos on mute.
Alex is a loser. I appreciate that he acknowledges that PC games have stuttering issues, but beyond that, he has nothing of value to offer.
It's an okay source. Watch the videos on mute.
Bulletbrain
Member
DF is just a tool to be used. Proper way to use it is for information on how a title is performing on various platforms, breakdown of visual features, or to understand the ”best" settings/mode to play on.
But, most here just use it to console warring smh.
Only major critique I have of them is that they are altogether a little too lenient on Devs and doesn't call Devs out with enough force when things aren't at acceptable levels. And this is applicable across all platforms.
But, most here just use it to console warring smh.
Only major critique I have of them is that they are altogether a little too lenient on Devs and doesn't call Devs out with enough force when things aren't at acceptable levels. And this is applicable across all platforms.
Feel Like I'm On 42
Member
That's still paid for bias. Just because these industries all have corrupt practices that are the status quo doesn't negate that fact. Why do you think the state of game reviews/journalism is so awful these days?
vg tech is good. But I like them explaining things. Sometimes I’m too dumb to know what I should be looking for.
Topher
Gold Member
SlimySnake
Flashless at the Golden Globes
I dont think its that awful when it comes to reviews. Plenty of big games these past few years have been massacred by reviewers. Concord, Suicide Squad, Skull and bones, Redfall in the last year alone. Most ubisoft games recently get marketed to hell and back on these websites and are still unable to get good review scores. now i think there is favortism in this industry and i will point to games like zelda, mario, rdr2, gta and tlou getting insanely high scores but people like the games they like and i cant fault them for that.
And if Sony is paying DF to fly them out and have them stay at hotels giving them early access to hardware just like MS did, then are they being paid for by Sony to be pro-Sony or by MS for being pro-MS? because the OP says they are biased against sony.
Hoddi
Member
I think they're mostly fine as a gamer channel and I don't really treat them a tech channel. If they were biased then they wouldn't be testing XSS on 4k screens for the simple reason that it's dumb.
I don't think anyone has damaged this console more than DF. Calling them biased for it seems misplaced, in my opinion.
I don't think anyone has damaged this console more than DF. Calling them biased for it seems misplaced, in my opinion.
Last edited:
kevboard
Member
we need gamers nexus to start doing tech analysis for games lol. steve would tear tear devs a new one I think
Bania Gold
Neo Member
I enjoy their videos and weekly podcast type things for the most part but don't watch anything with John in it.
He comes off as so egoistic and his attitude even seems to rub Rich and Alex and others the wrong way. If he tells you 1+1 = 3 you need to log off because he will rant and rave and never let you get your own opinion in.
Rich if fun and I remember reading some of his stuff in the 90s magazines if I remember right.
Alex knows his stuff and I don't get into his social media drama so that doesn't bother me.
I use their videos when I'm on the fence about a purchase if I'm concerned about performance.
But I agree with an earlier comment that they simply don't go hard enough on bad games or sugar coat issues. There is no way in hell they should have given Avatar on consoles the comments they did. It's image quality is absolutely disgraceful and they were too scared to say so.
Similar with Outlaws and some other recent AAA games where a bad looking game gets a quick 'can seem quite soft' comment and quickly glossed over.
He comes off as so egoistic and his attitude even seems to rub Rich and Alex and others the wrong way. If he tells you 1+1 = 3 you need to log off because he will rant and rave and never let you get your own opinion in.
Rich if fun and I remember reading some of his stuff in the 90s magazines if I remember right.
Alex knows his stuff and I don't get into his social media drama so that doesn't bother me.
I use their videos when I'm on the fence about a purchase if I'm concerned about performance.
But I agree with an earlier comment that they simply don't go hard enough on bad games or sugar coat issues. There is no way in hell they should have given Avatar on consoles the comments they did. It's image quality is absolutely disgraceful and they were too scared to say so.
Similar with Outlaws and some other recent AAA games where a bad looking game gets a quick 'can seem quite soft' comment and quickly glossed over.
The Last Ninja
Member
I like John and if he did a podcast Gerstmann-style where he just riffs for 3 hours I'd listen..
Heisenberg007
Gold Journalism
1. I'm not necessarily saying that they are biased against Sony. I'm saying that they are "biased," and they let that bias creep into their analysis. They are rarely fair.
2. I'm also saying that they are kind of stupid. They do very rudimentary analysis and then stick by it, instead of correcting themselves. The famous "12 is greater than 10" is a prime example of that.
Non-technical folks like us - with simple logic and evidence - was saying that wins will be traded almost evenly this generation as the two consoles have their unique strengths and weaknesses. It took Digital Foundry 4 years to simply raise this question:
3. We have plenty of evidence that their bias perception is not unfounded. Folks like
HeisenbergFX4
have mentioned multiple times that Digital Foundry is paid by Microsoft for that bias. He even shared his receipts with NeoGAF mods. There is little room for ambiguity here.
Zathalus
Member
1. I'm not necessarily saying that they are biased against Sony. I'm saying that they are "biased," and they let that bias creep into their analysis. They are rarely fair.
2. I'm also saying that they are kind of stupid. They do very rudimentary analysis and then stick by it, instead of correcting themselves. The famous "12 is greater than 10" is a prime example of that.
Non-technical folks like us - with simple logic and evidence - was saying that wins will be traded almost evenly this generation as the two consoles have their unique strengths and weaknesses. It took Digital Foundry 4 years to simply raise this question:
3. We have plenty of evidence that their bias perception is not unfounded. Folks like
They were paid by Sony as well. It’s nothing more than a sponsored trip to an exclusive event. Sure the XSX reveal was a bit more of a win for DF, but it’s hardly the first or the last time a company will offer them exclusive access to something. Considering DF is now owned by IGN it will probably happen with greater frequency.
Last edited:
Heisenberg007
Gold Journalism
I believe
HeisenbergFX4
wasn't referring to just exclusive accesses. It's not the same what they got from Sony compared to Microsoft.
SKYF@ll
Member
Where did the developer say that the texture settings are the same?
There is a significant difference in image quality just from the difference in sharpness.
These images were captured by me previously. (I don't have the PC version.)
RCX
Member
Only if Jazz regrows the mullet
On topic though, I've noticed myself becoming less interested in DF videos over the last couple of months. Feels like the ign effect maybe making them go after the most commercially viable subjects. It all just seems a little bit samey these days.
Last edited:
Zathalus
Member
They said in a Reddit thread, the developer was posting there and talked about differences between platforms. And that’s not the issue I’m referring to, it’s the issue with FSR having a sharpening filter on the PS5 version and the PC did not, even with DLSS. So in still images FSR would appear to be resolving more detail compared to DLSS. You had to enable it manually with UUU, but I believe a patch added a sharpening slider.Where did the developer say that the texture settings are the same?
There is a significant difference in image quality just from the difference in sharpness.
These images were captured by me previously. (I don't have the PC version.)
Mownoc
Member
Everyone is biased, at digital foundry each individual has their own personal biases. One of them does some weird comparisons between console and PC sometimes and is definitely biased towards the latter. But do I think DF are so biased it corrupts all their work? No... They just called Astro Bot "Virtually Flawless", they can't be that Anti-PlayStation can they?
I've seen people say both they're biased in favour of Xbox and they're biased against Xbox over the years so... You just have to be sensible enough to determine if what they say is objective or subjective and understand that some of their takes are the personal opinion of the individual.
I've seen people say both they're biased in favour of Xbox and they're biased against Xbox over the years so... You just have to be sensible enough to determine if what they say is objective or subjective and understand that some of their takes are the personal opinion of the individual.
Last edited:
64bitmodels
Reverse groomer.
Same people complaining about bias would 500% insert their own opinions into DF if they were running the show
Last edited:
Senua
Member
Don't watch it then
Boss Mog
Member
Believe me I don't anymore.
JackMcGunns
Member
It's 10% less here, 10% more in some other game and identical in others.
6700 and PS5 GPU are not identical, plus you have different Api, drivers, operating systems etc. But 6700 is the closest match you can get.
Even in Wukong you see this:
Is that some major difference to you?
Same with XSX and PS5 comparisons, one game runs 5-10% better on one of the consoles and fanboys are like:
But when you have 100% better performance on PC:
Console fanboys are like...
So yeah, NO - they are not biased. Sony fanboys are just super insecure.
Mic drop.
PaintTinJr
Member
I don't have the source images, but these that he posted - cropped at just 500-600pixels wide and - without identifying labels do look virtually the same even to a trained eye, and the underlying settings might be identical to a point, but as the performance mode get less time to make inference from the PS5 specific motion vectors saying they have the same settings is false.
You are saying that the sharpening filter has no performance cost, and the reason for that is that it isn't a sharpening filter for the game, but a sharpening bias value for the FSR algorithm, meaning that the algorithm reconstructs with different displacement biases, so whether set at minimum or maximum the performance cost is already baked in, unlike a real sharpening filter such as Guassian blur filter using a kernel to multi-sample and average at sub pixel displacements, which has a performance cost that increase proportionally to the square of kernel width/height.
And that's how we can tell that the cinematic image quality difference between PS5 balance mode and RX 6700 is more than just an FSR sharpening bias value. The loss in quality on the PC version of the cinematic at 6:03 looks like the internal resolution on PC for Unreal 5 engine - prior UE5 scaling it to 1080p and generating motion vectors for FSR reconstruction - is lower than 1080p or that the Pc settings are lowering depth precision or cascades, making the objects beyond the foreground look like they are smeared in Vaseline(quincunx AA) and making the the fog equation start to end crush quickly, and making it obscure all the Vaselined models very quickly on PC, and making all the fog pixels clump, as would happen at lower source resolution, and be a problem FSR or DLSS couldn't repair because the obscured geometry is effectively deleting data to inference from.
Last edited:
adamsapple
Or is it just one of Phil's balls in my throat?
I don't have the source images, but these that he posted - cropped at just 500-600pixels wide and - without identifying labels do look virtually the same even to a trained eye, and the underlying settings might be identical to a point, but as the performance mode get less time to make inference from the PS5 specific motion vectors saying they have the same settings is false.
500x pixel cropped images from a YT video are a terrible metric to gauge something on, something that's likely being played on much bigger screens in real time when DF is recording it and pointing out the differences.
Last edited:
Gaiff
SBI’s Resident Gaslighter
They still did even with VGTech stuff. Had fanboys on Gaf battling it out over the SX or PS5 winning because the median was 58.1 on one and 59.2 on the other.
90% of the Gaf's discussion whether from DF, VGtech, NxGamer, or anyone else is about console wars. You got a few users (generally the same 5-6 people) genuinely interested in the differences and the reasons behind them, but they often get dragged into console wars to refute incorrect information or outright lies.
Last edited:
Gaiff
SBI’s Resident Gaslighter
PaintTinJr
Member
That was merely a side comment - to disagree with Zathalus, saying they were very different images. Me commenting I didn't have the (two PS5) source images was merely in case Zathalus had full uncompressed uncropped screen grabs unaltered in anyway via DF discord
that did in fact look different - but still not the substance of my comment which was about FSR 'sharpening' not being the cause of the massive image difference between the PS5 balance and DF's poor efforts to match it with an unpowered and unrepresentative substitute RX 6700
Last edited:
Zathalus
Member
The difference in IQ between the two images is from the one mode having frame generation applied and a stronger sharpening pass from FSR. I’ve tested this myself with FSR, and so have plenty of other people. You can alter the image from looking overly soft and blurry due to the aggressive TAA applied by FSR or you can add a huge amount of sharpening which gives the impression that textures look sharper at the cost of other visual artifacts, especially around hair and foliage. Which the PS5 version absolutely has, if you have actually played it.I don't have the source images, but these that he posted - cropped at just 500-600pixels wide and - without identifying labels do look virtually the same even to a trained eye, and the underlying settings might be identical to a point, but as the performance mode get less time to make inference from the PS5 specific motion vectors saying they have the same settings is false.
You are saying that the sharpening filter has no performance cost, and the reason for that is that it isn't a sharpening filter for the game, but a sharpening bias value for the FSR algorithm, meaning that the algorithm reconstructs with different displacement biases, so whether set at minimum or maximum the performance cost is already baked in, unlike a real sharpening filter such as Guassian blur filter using a kernel to multi-sample and average at sub pixel displacements, which has a performance cost that increase proportionally to the square of kernel width/height.
And that's how we can tell that the cinematic image quality difference between PS5 balance mode and RX 6700 is more than just an FSR sharpening bias value. The loss in quality on the PC version of the cinematic at 6:03 looks like the internal resolution on PC for Unreal 5 engine - prior UE5 scaling it to 1080p and generating motion vectors for FSR reconstruction - is lower than 1080p or that the Pc settings are lowering depth precision or cascades, making the objects beyond the foreground look like they are smeared in Vaseline(quincunx AA) and making the the fog equation start to end crush quickly, and making it obscure all the Vaselined models very quickly on PC, and making all the fog pixels clump, as would happen at lower source resolution, and be a problem FSR or DLSS couldn't repair because the obscured geometry is effectively deleting data to inference from.
But it really doesn’t matter what I say, you believe that the dynamic fog looks better on PS5 despite multiple sources, including NXGamer, indicating otherwise. If you refuse to believe NXGamer (who is in no way biased against the PS5, probably the opposite) then what chance do I have?
PaintTinJr
Member
You aren't following the conversation.
Forget that image of PS5 balance vs PS5 performance, and go look at time point 6:03 and slightly early in the video where the cinematic can be easily compared with major visual differences all favouring(in addition to the superior frame-rate) the PS5 balance mode vs RX 6700 PC by DF that they wrongly claim is a match.
The differences in the PC RX6700 cinematics can't be fixed even by ML AI techniques like PSSR, XeSS or DLSS - never mind simple FSR - because the fog at native has already destroyed pixel data in the PC's source image that no amount of upscaling or sharpening can improve it to match the PS5's far, far superior visuals and still with a 5-10fps performance advantage.
The DF analysis indicates they know nothing about signal processing, the limitations of ML AI upscaling and 3D rendering if that is their final expert opinion that the versions differ by an FSR sharpening bias value IMHO or it is intended bias to downplay the PS5 and falsely talk up mid PC.
marquimvfs
Member
I think they're biased. But I like John's videos.
www.neogaf.com
Hardware - DF Weekly: If Xbox Series X is more powerful, why do some PS5 games run better? We finally have some answers.
More people finally realizing that DF doesn't have that much technical knowledge, especially about hardware. They do some cool side by side comparisons. And if a devs tells them what is going on, they can write about it. But they don't make quality hardware analysis.
Zathalus
Member
The only difference in the video is that the PS5 image is more detailed, be that via sharpening or some other method. You’re the only one seeing this supposed difference in quality of the volumetrics. As for the sharpening difference, DF made no mention of that, it is posters in this thread that have been discussing it. I’ve tested it myself and the toggle via UUU makes a similar difference to the video. I’ve already addressed the dynamic nature of the volumetrics in a previous post, which you have completely ignored. But of course I’m the one not following the conversation.
As for the PS5 vs 6700? That’s already been settled, they perform quite close to each other in the far majority of games. Big surprise.
Bavarian_Sloth
Gold Member
You are all a bunch of bespoke blasphemers!
Since the beginning of DF, it has been a channel of gaming-focused tech amateurs who have learned and expanded their portfolio over time.
Their target group: Gaming-focused tech amateur trolls from the caves and cellars of neogaf and resetera.
I for one don't find any of their content biased. And yes, maybe they misinterpreted a pixel here and there. Can happen. After all, we're not talking about errors on the scale of a "Linus Tech Tips".
Furthermore, I personally find it interesting that everyone has their preferences ("RTX- and PC Alex", "Retro John", etc.) but this is also communicated quite openly.
What else happens or is said by the people in question outside of the videos is irrelevant to me. I just enjoy the videos.
Hail Richard! Hail RTX!
Since the beginning of DF, it has been a channel of gaming-focused tech amateurs who have learned and expanded their portfolio over time.
Their target group: Gaming-focused tech amateur trolls from the caves and cellars of neogaf and resetera.
I for one don't find any of their content biased. And yes, maybe they misinterpreted a pixel here and there. Can happen. After all, we're not talking about errors on the scale of a "Linus Tech Tips".
Furthermore, I personally find it interesting that everyone has their preferences ("RTX- and PC Alex", "Retro John", etc.) but this is also communicated quite openly.
What else happens or is said by the people in question outside of the videos is irrelevant to me. I just enjoy the videos.
Hail Richard! Hail RTX!
Last edited:
PaintTinJr
Member
Actually, I didn't ignore it, I just wanted to be sure you were prepared to double down on the volumetric claim so I could then get you to justify that claim and provide evidence of the exact opposite situation where the PS5 in the same cinematic point drops effective resolution so heavily with such basic fog, and in reversal the PC raises effective resolution and uses the sharpening bias to produce a similar quality image as the PS5 at time point 6:03, and if it can on that PC, observe the already deficient frame-rate dip even further. Does that seem like a pretty basic thing for you to prove your point that it isn't just coincidence the lighter volumetric workload favours the PC in this data?
And you keep saying the RX 6700 is accepted as being comparative, yet here we are with a new game at close to the mid gen stage running with an up-to-date version of the most taxing engine UE5 and the performance delta with the RX6700 is closer to the delta expected between a non-XT and an XT or xx50 XT AMD version going by techpowerups relative performance. And that's without even needing to assert that the RX6700 PC is getting a lighter workload.
I'm sure you'll point to much earlier results or a collection of results to counter that, but this game is probably the best example of getting close to using the PS5 well on a multi-plat PC game we have, or do you disagree with that?
Last edited:
akira__
Member
Took a while, but you are able to see it then. If you also read the counter of frames there is 10% difference.
But Richard said on top of that ps5 had lower textures...
These narratives have real effects on people working on these games. That was the reason of this thread.
rodrigolfp
Haptic Gamepads 4 Life
How dare they pointing that the better looking and running PC version looks and runs better.
rodrigolfp
Haptic Gamepads 4 Life
Because it has worse textures:
How many times it needs to be shown to you to understand???The image quality isn't the same because of the overly aggressive sharpening in the performance mode. You even have Youtube comments of people who were fooled by the sharpening like you.
When it's factual that the Quality Mode runs at a much higher resolution. 1440p vs 1080p.
As for the textures...
Last edited:
Zathalus
Member
I can’t provide evidence of what I can’t see. You say the volumetric looks much better on PS5. DF, and NXgamer all happen to disagree and have multiple analysis videos showing it is equal to the PC high preset.
Comparing the 6700 vs the PS5, some engines favour PS5, some the 6700. They don’t need to be perfectly identical in every game to be comparable. As for up to date version of Unreal? Not even close, it is 5.0. Not even 5.1.
I (and others) have already explained the difference in perceived image detail to you, as well as what the lower texture detail refers to. I’m not going to repeat myself if you can’t be bothered to read.Took a while, but you are able to see it then. If you also read the counter of frames there is 10% difference.
But Richard said on top of that ps5 had lower textures...
These narratives have real effects on people working on these games. That was the reason of this thread.
PaintTinJr
Member
I edited my post above to upload the comparative shot.
At 6:03 it clearly shows by that stage in the cinematic the Vaslined(quincunx-esq AA) objects on PC have been rapidly obscured by the fog equation having a narrow range and saturating quickly and isn't just a difference of volumetrics because there is a lack of precision in the fog between the frustum near and far plane, a control that can be clearly seen in the precision of even the heavier fogged parts of the PS5 balance image.
The PS5 image doesn't clump even at the far plane, suggesting the PS5 is rendering in a frustum cascade and is able to independently control the fog in each - of maybe 2 or 3 sub-frustum - and the PC version looks like it is done in just 1 frustum and exceeded the precision of its fog equation by the precision of the depth buffer values feeding it converging to full saturation too quickly.
Last edited:
Lagspike_exe
Member
Good tech analysis, but terribly biased in favor of MS.
Come On Tars
Member
Getting worked up over it shows the viewers bias. I really don’t give a shit, the most unbiased uncaring person on earth. I think they tell it how it is. I don’t even know what the general proclaimed biased people think they have?
The answer is nah
The answer is nah
Zathalus
Member
The fog is dynamic.
PS5:
PC:
I’m not seeing this major difference in fog quality. What I am seeing is differences in how the fog behaves in each video, even the PS5 ones.
Last edited:
FoxMcChief
Gold Member
I stopped watching their videos. Only because I can’t listen to others opinions on what mode is best. I just wish one day that someone out there knew what I cared about and know my point of view is the best.
Similar threads
- Poll