Support NeoGAF

[thelastword]

PS4 uses a dynamic resolution with the lowest resolution found being approximately 1792x1008 and the highest resolution found being 1920x1080. PS4 Pro uses a dynamic resolution with the lowest resolution found being approximately 2368x1332 and the highest resolution found being 2880x1620, and this is downsampled when outputting at 1080p. Xbox One uses a dynamic resolution with the lowest resolution found being 960x1080 and the highest resolution found being 1440x1080. Xbox One X uses a dynamic resolution with the lowest resolution found being approximately 3136x1764 and the highest resolution found being 3840x2160. All four consoles often render at or near their maximum resolution, but the base PS4 in particular rarely ever deviates from its 1920x1080 peak resolution. Xbox One X does appear to stick closer to its maximum resolution, on average, than PS4 Pro does.

So dynamic resolution on all consoles.....Vgtech yet again delivers more accurate information.

Pretty much solid 30fps on all consoles, lots of tearing on XBOX=ONE-S, a few dropped frames, single digit dropped frame on PS4/PRO, none on XBONEX.

Torn frames; 156 = XBOX ONE, 2=PS4, 1=PRO, 6=XBONEX


All the evidence for the DR resolutions is there for persons to pixel count themselves......

 

[JimboJones]

So much for FP16 magic.

 

[DynamiteCop!]

Uh Far Cry 5 is a fixed native 4K on Xbox One X...

 

[Krappadizzle]

Even on PC I never thought the game was a looker. Still had a good time with it, but I don't expect it to blow me away compared to it's contemporaries like FC2 did in it's day.

 

[LostDonkey]

Uh Far Cry 5 is a fixed native 4K on Xbox One X...

It was left open for further testing by DF but if these results are indeed true it would prove DF wrong yet again on yet another game analysis.

 

[chilleverest]

It was left open for further testing by DF but if these results are indeed true it would prove DF wrong yet again on yet another game analysis.

Yet again? Have they been proved wrong before ?

 

[LostDonkey]

Yet again? Have they been proved wrong before ?

Yes. Redout most recently.

 

[Xiaoki]

Yet again? Have they been proved wrong before ?

Oh heck yeah

Early in the year they reviewed a XBX game, said it was locked at 1080p. Developer said it had dynamic resolution up to 1944p and threatened to sue DF. DF issued an apology and said they simply missed the dynamic resolution.

Also, last year they did a DF Retro of Need For Speed. They said the Saturn version didnt have music and that the speedometer and tachometer were missing in cockpit view. This is, of course, wrong if you actually played the real game, DF was playing it on an emulator.

 

[thelastword]

So much for FP16 magic.

FP16 is not magic, it's math it's data, it's real. Of course, there are different implementations, F1 2017 had a more substantial FP16 implementation and maybe FC5's was not as extensive, but still, there's less tearing on PS consoles, so maybe that's a small benefit.....Besides, FC5 is not gimping any platform, so I'm not even sure their FP16 implementation was said to be this huge boon to begin with.

FYI, FC5's FP16 implementation is mostly utilized on water surfaces or water simulations, so they save a couple of cycles there when they render water, but they're using a combination of async compute and RPM to achieve that. Also seeing that some of the simualations of prior games were dumbed down in this game, I'm really not sure it's that extensive anyway...

FC5 had to run on XBOX-ONE, I think they gave pretty good performance across the board seeing as how some vanilla XBOX one ports turn out these days. Both AMD and NV PC GPU's run it pretty much identical....So it's hardly a case of them going above and beyond on PRO or any platform vs the other.....Case in point, a very easy way for Ubi to have gained much more cycles on PRO would have been for them to do CB2160 instead of 1620p, but they didn't do that. OG PS4 seems to sit about 99% at native 1080p, so maybe it's the async compute on that, with only 1 drop to 50ms in that video and only two torn frames. Still they used it, but since PS4 could mostly handle the game at 1080p and PRO at 1620p, they did some minor compute and fp16 stuff but nothing substantial as F1...

Uh Far Cry 5 is a fixed native 4K on Xbox One X...

It's not, every version falls below their native render targets. OG PS4 sticks the closest to it's target...about 99%, then follows XBONEX, PRO (maybe about 85:80 respectively) and then XBONE-S....

Yet again? Have they been proved wrong before ?

Hmmmm...

DF said Redout was 1080p on XBONEX, Vgtech said it was dynamic...(they gave figures and screens to show, DF later apolgized and said they were wrong)

DF said RE7 was 1260 on PRO, Vgtech said it was 1800p CB (then DF came back and said it was 1800p)

DF said Sea of thieves was 1080p on XBONES in their first analysis of the game (Morgan+Bierton), Vgtech said it was 900p (DF's new video went on to say it was 900p)

Apart from that, DF always registers higher minimum resolutions than Vgtech in dynamic resolution games, especially on XBOX versions of said games. The same can be said of minimum framerates....

The above are more recent examples, but DF has been known to make many mistakes or push a narrative sometimes. I still remember Far Cry 4, DF used one non 1:1 scene to prove XBONES had a better framerate than PS4, till Nxgamer did a more thorough analysis and showed that every time you wade through deep foliage on the XBONES version, framerate would fall into the low 20's and the teens (as low as 15fps), compared to a much higher framerate on PS4. Such things would never have been known or reported or DF's errors corrected, had it not been for these other outlets...


The way vgtech works is that they go through a larger portion of the game, they show and upload every screen they used to determine their pixel counts, they give thorough stats on all their framerate tests. Frametimes, torn frames, average framerate, min-max frames and the list goes on. After they publish their stats and video, they still keep on testing the game and provide updates on their youtube page.....(even lower minimum rez discovered, or framerate, some anomalies etc...and they provide updates with technical notes and findings). In that sense, their data can always be accessed and it's all verifiable with proper data/stats..

 

[DynamiteCop!]

It's not, every version falls below their native render targets. OG PS4 sticks the closest to it's target...about 99%, then follows XBONEX, PRO (maybe about 85:80 respectively) and then XBONE-S....


Hmmmm...

DF said Redout was 1080p on XBONEX, Vgtech said it was dynamic...(they gave figures and screens to show, DF later apolgized and said they were wrong)

DF said RE7 was 1260 on PRO, Vgtech said it was 1800p CB (then DF came back and said it was 1800p)

DF said Sea of thieves was 1080p on XBONES in their first analysis of the game (Morgan+Bierton), Vgtech said it was 900p (DF's new video went on to say it was 900p)

Apart from that, DF always registers higher minimum resolutions than Vgtech in dynamic resolution games, especially on XBOX versions of said games. The same can be said of minimum framerates....

The above are more recent examples, but DF has been known to make many mistakes or push a narrative sometimes. I still remember Far Cry 4, DF used one non 1:1 scene to prove XBONES had a better framerate than PS4, till Nxgamer did a more thorough analysis and showed that every time you wade through deep foliage on the XBONES version, framerate would fall into the low 20's and the teens (as low as 15fps), compared to a much higher framerate on PS4. Such things would never have been known or reported or DF's errors corrected, had it not been for these other outlets...


The way vgtech works is that they go through a larger portion of the game, they show and upload every screen they used to determine their pixel counts, they give thorough stats on all their framerate tests. Frametimes, torn frames, average framerate, min-max frames and the list goes on. After they publish their stats and video, they still keep on testing the game and provide updates on their youtube page.....(even lower minimum rez discovered, or framerate, some anomalies etc...and they provide updates with technical notes and findings). In that sense, their data can always be accessed and it's all verifiable with proper data/stats..

Wrong, VGTech is not accounting for the heavy TAA implementation which is throwing their numbers. Digital Foundry specifically notated this as it creates a softer image, they accounted for it, VGTech did not.

Also Capcom told them Resident Evil 7 rendered at 2240x1260 on the Pro, they didn't find that result they simply reported it, They found the same 1800cb that VGtech did in their testing.

Sea of Thieves when they initially analyzed the beta they never gave an Xbox One resolution, they said the Xbox One X renders at 1080p and lacked its enhancements.

Their only mistake was Redout, that's it. This is all wrong and misinformation and you're peddling fanboy anti-Xbox propaganda.

"Apart from that, DF always registers higher minimum resolutions than Vgtech in dynamic resolution games, especially on XBOX versions of said games. The same can be said of minimum framerates...."

Give me a break...

 

[Koala Driving a Ferrari]

So, after taking everything into account, the Xbox One X version is objectively the best console version of the game by a large margin and it does with little to no effort and completely silently. Thanks TLW for the heads-up mate. You just sold me on a copy of the game for my X.

Boy am I glad that I held off on buying a Pro. Oof, that would have been a waste of money. Plus, I'd need to factor in the additional cost of industrial-tier hearing protection to preserve my ears from all of that screaming banshee fan noise. You're doing a fantastic job at selling MS products with all of your hilariously obvious and transparent Sony fanboyism. Keep it up mate.

Stock up on tissues and nappies for 2020; you're going to need them.

 

[Gavin Stevens]

TLW, you do know that FP16 and FP32 are two entirely different sets of instruction and used in different cases right...? I've been using fp16 and fp32 for years on different things where precision is either more or less needed. The thing is the Pro allows you to use two FP16 at a lower cost.

However...

You can't just say "let's make it all fp16 and get so hole frame rates!". It doesn't work that way.

For one, most of the stuff you do won't be using fp16. The precision is far too low. In fact, it's that low it would make people angry in some cases. For example, for some texture assets, it can look like the difference between compressed and none compressed textures (for an example, not a perfect one, but one you can put into visual terms at least).

Also, you can't just make a FP16 from a FP32 and be done with it. It requires its own written from the ground up and tested implementation.

So essentially what you are saying, is you are perfectly happy with games on the PlayStation looking much worse, and taking longer to make, and also more expensive... just so you may possibly get another 5fps at the end of it. And that 5fps is a rough average guess, as that could go up or down depending on all sorts of factors, such as the rendering capabilities of the engine in play, or even the basic type of game.

Real talk now... it's actually embarrassing how devoted you are to a plastic box. Just enjoy games. I do. Most do. This fan boy bullshit is so 2000.

 

[Gavin Stevens]

Oh and for the record, on my X at least, farcry 5 looked stunning, hardly ever to my knowledge (or at least in any large enough way) dropped below 4k, and was pretty much locked to 60fps.

The game was fantastic, and a blast to play. I had fun. Remember that word... fun.

 

[MMaRsu]

Its 30fps not 60

 

[Gavin Stevens]

Its 30fps not 60

You know I think that's right. The games I put delay is so sharp and controls so well it's haed to believe it's not 60. Damn fine effort.

 

[Mochilador]

So is Vgtech more reliable than DF?

 

[FranXico]

So is Vgtech more reliable than DF?

I would't put it like that. But you should never blindly trust only one source.

Each and every one of these tech analysts has made their own mistakes and found mistakes of others. If technical breakdowns are your thing, you should read up as many sources as you can.

People who just want to justify their latest hardware purchase, well, they never really cared about game engine optimization and really shouldn't need validation in the first place.

 

[Mochilador]

I would't put it like that. But you should never blindly trust only one source.

Each and every one of these tech analysts has made their own mistakes and found mistakes of others. If technical breakdowns are your thing, you should read up as many sources as you can.

People who just want to justify their latest hardware purchase, well, they never really cared about game engine optimization and really shouldn't need validation in the first place.

Good points. For me these analysis are more of a curiosity.

 

[GTR R35_Supra RZ]

So, after taking everything into account, the Xbox One X version is objectively the best console version of the game by a large margin and it does with little to no effort and completely silently. Thanks TLW for the heads-up mate. You just sold me on a copy of the game for my X.

Boy am I glad that I held off on buying a Pro. Oof, that would have been a waste of money. Plus, I'd need to factor in the additional cost of industrial-tier hearing protection to preserve my ears from all of that screaming banshee fan noise. You're doing a fantastic job at selling MS products with all of your hilariously obvious and transparent Sony fanboyism. Keep it up mate.

Stock up on tissues and nappies for 2020; you're going to need them.

Got em' easy with no sweat! Do you know how hard he's gathering material now..but can't find any..since entire thread like all of them backfired on his insecure ass..it's classic, I luv it..

 

[JimboJones]

FP16 is not magic, it's math it's data, it's real. Of course, there are different implementations, F1 2017 had a more substantial FP16 implementation and maybe FC5's was not as extensive, but still, there's less tearing on PS consoles, so maybe that's a small benefit.....Besides, FC5 is not gimping any platform, so I'm not even sure their FP16 implementation was said to be this huge boon to begin with.

Oh I know but you said back in feb in this post

Check out F1 2017, You can check Farcry 5 when it launches and you can be sure to check out some of the first party releases from March 2018 and beyond..... for your FP16 fix........

So here we are and it's another expected case of the X version being substantially better because we are all aware of the specs of both machines.

 

[Redneckerz]

It was left open for further testing by DF but if these results are indeed true it would prove DF wrong yet again on yet another game analysis.

I think that is a bit of an extreme take to make. Like NX Gamer/VG Tech, DF is subject to human analysis - They are not automatrons that get things right all the time. Dark1x has gone out of his way to explain this and that they consider NX/VG as their colleagues, having included their work in videos occassionally since then.

Its not a black and white scenario in my opinion, and there certainly aint no animosity between the three. So why should GAF?

Yes. Redout most recently.

Compare that to a lot of videos that see no scandals and the fact they pump out a lot of content, all relatively consistent and all. I feel these things should not be dismissed.

Oh heck yeah

Early in the year they reviewed a XBX game, said it was locked at 1080p. Developer said it had dynamic resolution up to 1944p and threatened to sue DF. DF issued an apology and said they simply missed the dynamic resolution.

Also, last year they did a DF Retro of Need For Speed. They said the Saturn version didnt have music and that the speedometer and tachometer were missing in cockpit view. This is, of course, wrong if you actually played the real game, DF was playing it on an emulator.

Again, in my eyes this kind of practice is really looking for faults and then assign blame. ''See? They make mistakes!'' I think that is a bit dramatic. Both DF, VG, and NX all have made their mistakes in the past and they all learn from it and occassionally work together (or reference) eachother's work. As it should.

Why look for a scapegoat when all three parties clearly aren't doing that among themselves and just try their best to report things? I dont understand that, but id love to know why that is so important for some. *

*speaking in a general sense here and not specifically you. ERA members also had similar opinions which is what i am referring at here.

So is Vgtech more reliable than DF?

I don't think it should be a case of who is the most reliable or who gets it right most of the times. In the end, all these analysts are still prone to human error, and the difference in grading them should be in how they handle these situations. For the Redout scenario, i cant say either VG or DF handled it unprofessionally.

 

[thelastword]

Wrong, VGTech is not accounting for the heavy TAA implementation which is throwing their numbers. Digital Foundry specifically notated this as it creates a softer image, they accounted for it, VGTech did not.

Also Capcom told them Resident Evil 7 rendered at 2240x1260 on the Pro, they didn't find that result they simply reported it, They found the same 1800cb that VGtech did in their testing.

Are you talking about FC5, that TAA is the reason why the image is softer in parts, are you serious? So evidence of dropped resolution from 2160p on XBONEX, from 1620p on PRO, from 1920 x 1080 on PS4 and 1440 x 1080 on XBONES are no rez drops just TAA...? ok got it? I'll take the site which has never got a resolution wrong over one which has on several occasions.....I'll take the site which has all the screens pixel counted, showing the actual lower resolutions. You keep on believing any fairy dust explanation you desire, but it's not relevant in this thread with actual stats and tangible data or evidence...

Sea of Thieves when they initially analyzed the beta they never gave an Xbox One resolution, they said the Xbox One X renders at 1080p and lacked its enhancements.

Right there at the mark.....

Their only mistake was Redout, that's it. This is all wrong and misinformation and you're peddling fanboy anti-Xbox propaganda.

"Apart from that, DF always registers higher minimum resolutions than Vgtech in dynamic resolution games, especially on XBOX versions of said games. The same can be said of minimum framerates...."

Give me a break...

The bolded is a fact, you look at any DR game really, from Wolfenstein to Doom etc etc...

....and I have no breaks to give, just stats. You might need to go elsewhere if you want those...

Oh I know but you said back in feb in this post


So here we are and it's another expected case of the X version being substantially better because we are all aware of the specs of both machines.

What does the specs of the machines have to do with this analysis...Did I say the XBONEX version had a lower rez than PRO, did I say it's framerate was worse. All you see here is the stats. It was thought that the XBONEX version was 4k native throughout according to digital foundry, that's clearly not the case....Yet, it's baffling that people don't want to take the hard evidence because it disrupts a pure 4k native game for X. So you would rather embrace wrong information because it sullies a statisitic for a box.....At the end of the day, it really does not matter anyway, because 4k is clearly not selling XBONEX's, so there's no reason to be preoccupied with one less true 4k game on the system....

As for my statements, the first party games are using FP16, like GOW and so is Farcry 5...FC5 may not have used it as extensively as F1-2017, but it's using it and developers are using it, third parties included.....No one ever said it was going to give you 60fps on 30fps games or anything like that, but it definitely saves some cycles in the rendering pipeline....

 

[demigod]

I don't know about VGtech but DF is terrible. I'm glad NXGamer didn't quit yet, i'll trust him over DF any day.

 

[DynamiteCop!]

Are you talking about FC5, that TAA is the reason why the image is softer in parts, are you serious? So evidence of dropped resolution from 2160p on XBONEX, from 1620p on PRO, from 1920 x 1080 on PS4 and 1440 x 1080 on XBONES are no rez drops just TAA...? ok got it? I'll take the site which has never got a resolution wrong over one which has on several occasions.....I'll take the site which has all the screens pixel counted, showing the actual lower resolutions. You keep on believing any fairy dust explanation you desire, but it's not relevant in this thread with actual stats and tangible data or evidence...


Right there at the mark.....



The bolded is a fact, you look at any DR game really, from Wolfenstein to Doom etc etc...

....and I have no breaks to give, just stats. You might need to go elsewhere if you want those...

What does the specs of the machines have to do with this analysis...Did I say the XBONEX version had a lower rez than PRO, did I say it's framerate was worse. All you see here is the stats. It was thought that the XBONEX version was 4k native throughout according to digital foundry, that's clearly not the case....Yet, it's baffling that people don't want to take the hard evidence because it disrupts a pure 4k native game for X. So you would rather embrace wrong information because it sullies a statisitic for a box.....At the end of the day, it really does not matter anyway, because 4k is clearly not selling XBONEX's, so there's no reason to be preoccupied with one less true 4k game on the system....

As for my statements, the first party games are using FP16, like GOW and so is Farcry 5...FC5 may not have used it as extensively as F1-2017, but it's using it and developers are using it, third parties included.....No one ever said it was going to give you 60fps on 30fps games or anything like that, but it definitely saves some cycles in the rendering pipeline....

Let me start off by saying lol.

Digital Foundry was wrong on one occasion and in the event of that they apologized, this is out of hundreds of testing scenarios. In terms of TAA it can throw counts as it does soften the image particularly in motion, I'm not here to debate that with you because it's just the reality of pixel counting and the function of that AA method; this is well known. None of this stuff from VG Tech is "hard evidence", it's literally raw screen data without TAA compensation, you're going to get inconsistent counts that alter perceived pixels. It's an aggressive AA method, why do you think the image becomes softened...

In terms of Sea of Thieves that was the beta, are you an authority to say that it was not 1080p during the beta? No you're absolutely not, the game could have very well have been 1080p prior to launch but given certain issue was scaled back to 1600x900. Can you say otherwise with supporting evidence? No.

As far as minimum resolutions are concerned this is never going to be a like for like, they all test different sections of games and it's not a static run between org's where they all agree to run the same test sequences. You're just being paranoid and delusional.

 

[dark10x]

I don't post here often these but I find some of these posts mean spirited. It's not really very nice to lash out at someone like this. I'm in a position no different than NX Gamer (who I consider a friend, by the way) and VGTech. We all work by ourselves from home and try our best. DF is not a large organization.

That said, in the case of Far Cry 5, I actually have a suspicion that the game has been patched. I spoke with a couple devs after and the 4K native thing was confirmed. I also pixel counted 30+ shots and had multiple people examine them as well. Never came back as anything less than native 4K. However, there were still performance dips AND the map system exhibited horrible tearing and slowdown. They have since patched the game multiple times and most of those issues are cleared up including the map. It runs great now!

However, I do believe they've either added dynamic resolution scaling or changed the threshold to be more aggressive. I think the version VG Tech is testing here has simply changed since we first looked at the game. Thus, I don't think it's fair to parade around as one of us being incorrect because we're not looking at the same version.

Oh heck yeah

Early in the year they reviewed a XBX game, said it was locked at 1080p. Developer said it had dynamic resolution up to 1944p and threatened to sue DF. DF issued an apology and said they simply missed the dynamic resolution.

Also, last year they did a DF Retro of Need For Speed. They said the Saturn version didnt have music and that the speedometer and tachometer were missing in cockpit view. This is, of course, wrong if you actually played the real game, DF was playing it on an emulator.

The issue with Need for Speed is not a typical mistake and I'll explain why. I do not use emulators for Saturn. So why was the music missing and the speedo not working?

Well, it turns out that there are variations in how the game runs on hardware depending on the mix of game + console region. I used the NTSC-U version of Need for Speed with an NTSC-J console which I booted using the Pseudo-Saturn cartridge. My NTSC-U system was non-functioning at the time (since fixed). Since then, I've also purchased a PAL copy of Need for Speed.

What I discovered is that the game behaves differently on different region Saturn consoles. This is known to happen when matching PAL and NTSC but I've never seen it with NTSC-J and U. If you play the US version on a Japanese system, the music and speedometer simply do not appear. Since I bought the US version specifically for the episode, I thought this was normal. Booting it on an NTSC-U system, however, corrects those issues. Now, if you boot the PAL version on either NTSC systems, you get OTHER weird glitches.

Basically, I discovered that you cannot mix and match regions in certain games and expect perfect results. There was no emulation used in that video (and you can tell based on the output quality of the capture - which is from an analog source) - the emulator would probably work properly, I'd imagine. This is something else entirely. It's my fault for mixing region codes like that but I've since started using the correct region console with the correct region game even though, 99% of the time, there isn't an issue when swapping between them.

Secondly, regarding Redout, Dave missed that because the game doesn't use dynamic resolution - the resolution changes are set at fixed intervals rather than dynamically scaling based on load. That was a HUGE mistake on his part that cost us dearly. I was on vacation when that hit and it caused so many headaches. Crazy that he missed it but, hey, he's no longer doing work for the site.

 

[Dibils2k]

I bet dynamic resolution came in when they fixed screen tear on the map and changed the HDR settings

 

[thelastword]

Also Capcom told them Resident Evil 7 rendered at 2240x1260 on the Pro, they didn't find that result they simply reported it, They found the same 1800cb that VGtech did in their testing.

No, Morgan revisited the analysis, yet again after vgtech posted a thorough analysis and specified 1800CB on a 4k screen and 1260p downsampled to 1080p on a FullHD screen on PRO. Vgtech gave a good breakdown why it was CB and mentioned more artifacts on the XBONEX version....which is obvious because PRO can do CB through hardware....

In Morgan's re-analysis, of course, he agreed that Vgtech's pixel count was right as usual, but then he said it was 1800p non cb and 2160p non CB on the platform, where the latter is incorrect.......RE7 is not doing 1800p native on PRO or 4k native on X. Keep in mind, this is the same DF that missed out on the 1080p mode in Blops 3 when it was pro patched....

As for the bit about Capcom told them?.....Ubisoft said Origins was 4k on XBONEX, The Ark guys said their game would be 4k 60fps on XBONEX....The "dev told them" has nothing to do with a tech breakdown, the job is to get the real answer without the devs, that's how it has always been done, since the days of the pixel counter on B3D, because many devs never divulged or were vague and this is how we got to know the real resolutions of games. This is how the tech outlets were born, to get details that the devs will not come out to you and specify.

Do you go to tech sites because they're just parroting what the dev said? I mean come on now, because in that case the tech sites would not be needed and would just be a news outlet more than anything.

Let me start off by saying lol.

So I give you the exact timestamp where both Bierton and Morgan say it's 1080p on base XBOX and all you have to say is lol, when you categorically stated they never said that.....ok, note taken...

Digital Foundry was wrong on one occasion and in the event of that they apologized, this is out of hundreds of testing scenarios. In terms of TAA it can throw counts as it does soften the image particularly in motion, I'm not here to debate that with you because it's just the reality of pixel counting and the function of that AA method; this is well known. None of this stuff from VG Tech is "hard evidence", it's literally raw screen data without TAA compensation, you're going to get inconsistent counts that alter perceived pixels. It's an aggressive AA method, why do you think the image becomes softened...

The last thing you want to do is to make excuses for non-thorough work, especially when it comes to information derived through tangible data and stats, checks and balances etc....You keep going on about TAA, but TAA does not prevent a proper pixel count, making excuses in that regard does not make any sense and only proves that other folk are doing more detailed and accurate work out there.

Do you know how many games have been pixel counted with TAA and a mud-load more PP effects than FC5??? a truckload load of them......and VGtech got them all right. It's the same principle behind Redout, it makes no sense to come in here and say it was hard to detect, it just looks bad on the person saying that. Hell, you don't even need pixelcounting to determine this was more than 1080p on the XBONEX, all you had to do was watch the screen and observe.....so making excuses that it was hard to notice because of the way DR was implemented, is just trying to save face and they look even sillier saying that, just accept you were wrong and ensure you do better next time. People are easier to move on from it if you take a humble and apologetic approach, but it's insulting to come in here and say this was hard to pick up. If this is so hard for the biggest console testing outlet, as opposed to one guy at Vgtech, then that pretty much says it all......proof is in the pudding really.

In terms of Sea of Thieves that was the beta, are you an authority to say that it was not 1080p during the beta? No you're absolutely not, the game could have very well have been 1080p prior to launch but given certain issue was scaled back to 1600x900. Can you say otherwise with supporting evidence? No.

As far as minimum resolutions are concerned this is never going to be a like for like, they all test different sections of games and it's not a static run between org's where they all agree to run the same test sequences. You're just being paranoid and delusional.

Tech threads are not about could have beens and would have beens. Tech threads are about data, how are you confirming what you're saying??? and it's an analysis from the same guy who did the redout analysis no less, with a confirmation by a senior editor in Morgan. From now on, it will be so easy to say games were patched and changed from beta and in patches with no corroborative evidence, no patch notes saying such, nothing really, against tangible data.....Yet , all we're seeing in Vgtech's video is more thorough and precise work, that is all...

As for min resolutions and framerates, you get a more accurate take of the range by playing much more of the game, getting deeper in chaotic scenes etc...Again, a more thorough approach nets you better and more accurate results......It's just like when DF could not be bothered to check the 1080p mode in BLOPS 3, but rather went on a continued narrative of vanilla PS4 having better performance at Pro's launch for clicks or whatever else.... instead of prioritizing a deeper take, making sure to get more basic checks in...1080p mode, 4k mode, getting a keener eye for detail and deciphering through console options.....That's the difference more than anything..

 

[Gavin Stevens]

Hey TLW, Sony recently said the sky was bright green.

Thoughts?

 

[dark10x]

People are easier to move on from it if you take a humble and apologetic approach, but it's insulting to come in here and say this was hard to pick up. If this is so hard for the biggest console testing outlet, as opposed to one guy at Vgtech

I've always tried to be humble and apologetic. I'm doing my best but I'm obviously not good enough. I know it. I hate myself enough already. That's fine. ...but you KEEP ATTACKING US over and over and over again.

You say VGtech is one guy? Guess what? So am I. I work from home by myself. I have no office to go to. I work alone. There is no difference between what I'm doing and someone like him besides the name on the site. That's literally it. I don't understand why you think this is different. Just because it's a known name? How does that actually matter when you're just a guy working from home on your own?

Show some compassion, man. Stop it. It's just video games. You have no idea how much it hurts. I wish you'd actually stop and consider this. You keep complaining about Bierton - he's not even a part of the site any more.

Anyways, I'm out. It's clear that there's no point in being here.

 

[Virex]

If all of you that are complaining here can do a better job than DF or VGtech. Then by all means start your own channel and show us how much better you are. DF is great in my opinion. They are a little bit in the Microsoft can do no wrong camp. But, they do a pretty good job with their videos and I enjoy Digital Foundry. There is a lot of work that goes into their videos and I appreciate it.

 

[Hendrick's]

I don't post here often these but I find some of these posts mean spirited. It's not really very nice to lash out at someone like this. I'm in a position no different than NX Gamer (who I consider a friend, by the way) and VGTech. We all work by ourselves from home and try our best. DF is not a large organization.

That said, in the case of Far Cry 5, I actually have a suspicion that the game has been patched. I spoke with a couple devs after and the 4K native thing was confirmed. I also pixel counted 30+ shots and had multiple people examine them as well. Never came back as anything less than native 4K. However, there were still performance dips AND the map system exhibited horrible tearing and slowdown. They have since patched the game multiple times and most of those issues are cleared up including the map. It runs great now!

However, I do believe they've either added dynamic resolution scaling or changed the threshold to be more aggressive. I think the version VG Tech is testing here has simply changed since we first looked at the game. Thus, I don't think it's fair to parade around as one of us being incorrect because we're not looking at the same version.

Thank you for bringing some common sense into this thread.

 

[Gavin Stevens]

Nice one TLW, another really good person doing a damn fine job annoyed by your blatant lack of self control and fanboyism. It's people like you that give sites a bad name. Just go to a Sony forum and sit among yourselves there, you clearly don't want to hear common sense or engage in actual discussion, you just want to spout the same lies as truth. Truly you are the Trump of the gaming world...

 

[thelastword]

No one is attacking anyone here, this is just a thread stating the facts. It's what I've been saying all this time, when there's evidence and more thorough work is done, we get better results.......that's it. So a tech site/mind should not be based on conjecture and opinions.....

To come in and say DR was patched in is wrong when anyone can test this pre-patch, this is not how you address this... Truth is, you can't make excuses everytime DF gets something wrong and it's also disingenuous of posters to dismiss this thorough analysis by talking about "they had fun", tech threads don't work that way. You can prove this was 4k native pro patch and make that point.....but not without evidence....

As I was saying, this is how you could have addressed this, the first thing to do is not to defend...You want to get to the bottomline of this and get the right information.

I don't think that the dynamic resolution on consoles was added in a patch as version 1.00 on PS4 Pro uses a dynamic resolution maxing out at 2880x1620. I haven't checked the game unpatched on the other consoles yet though.


I just checked the unpatched version on Xbox One X (0.1.0 in the game menu and 1.1.267.21 in the Xbox menu) and it also uses a dynamic resolution.

 

[Hendrick's]

No one is attacking anyone here, this is just a thread stating the facts. It's what I've been saying all this time, when there's evidence and more thorough work is done, we get better results.......that's it. So a tech site/mind should not be based on conjecture and opinions.....

To come in and say DR was patched in is wrong when anyone can test this pro-patch, this is not how you address this... Truth is, you can't make excuses everytime DF gets something wrong and it's also disingenuous of posters to dismiss this thorough analysis by talking about "they had fun", tech threads don't work that way. You can prove this was 4k native pro patch and make that point.....but not without evidence....

As I was saying, this is how you could have addressed this, the first thing to do is not to defend...You want to get to the bottomline of this and get the right information.

Just give it a rest, no one here besides you gives a crap.

 

[petrus91]

Nice one TLW, another really good person doing a damn fine job annoyed by your blatant lack of self control and fanboyism. It's people like you that give sites a bad name. Just go to a Sony forum and sit among yourselves there, you clearly don't want to hear common sense or engage in actual discussion, you just want to spout the same lies as truth. Truly you are the Trump of the gaming world...

Looking for common sense from that guys is pointless. I'm so sorry for John getting this treatment. He doesn't deserve at all. He's such a passionate guy, I followed all his E3 coverage and was amazed by the amount of work he's able to put on the table. It's so sad that this kind of subtle toxicity is still tolerated around here. John's totally right: there's no reason to stay in this place.

 

[LucidFeuer]

That's the most boring game I got to play recently. Oh really another american forest setting with bayous and rednecks...HOW ORIGINAL.

With a friend we quit the game after an hour so much we were bored and had seen what there was to be seen...for the 100th time.

 

[octiny]

I've always tried to be humble and apologetic. I'm doing my best but I'm obviously not good enough. I know it. I hate myself enough already. That's fine. ...but you KEEP ATTACKING US over and over and over again.

You say VGtech is one guy? Guess what? So am I. I work from home by myself. I have no office to go to. I work alone. There is no difference between what I'm doing and someone like him besides the name on the site. That's literally it. I don't understand why you think this is different. Just because it's a known name? How does that actually matter when you're just a guy working from home on your own?

Show some compassion, man. Stop it. It's just video games. You have no idea how much it hurts. I wish you'd actually stop and consider this. You keep complaining about Bierton - he's not even a part of the site any more.

Anyways, I'm out. It's clear that there's no point in being here.

Don't mind him, 99% of the people on here know exactly why he posts on this forum. I'm just surprised he hasn't been banned yet.

 

[Redneckerz]

I don't post here often these but I find some of these posts mean spirited. It's not really very nice to lash out at someone like this. I'm in a position no different than NX Gamer (who I consider a friend, by the way) and VGTech. We all work by ourselves from home and try our best. DF is not a large organization.

I am actually glad that you showed up here again John (You must remember me from ERA! Hiya!).
But.. yeah. This is why i made my initial posting in here. Clearly neither DF or VG or NX see themselves as competitors in bad standing, but rather, compatriots in good standing.

Either one of you provide a great service to a specific part of the gaming community wanting to know more. (In the past Lens of Truth did this, aswell). I'd say for all the content these 3 organizations collectively have pumped out, its okay to be off the mark on occassion. Especially with this generation introducing reconstruction techniques and dynamic framebuffers (And Redout even doing this per section of the track!) pixelcounting has only become more difficult for each of you.

That said, in the case of Far Cry 5, I actually have a suspicion that the game has been patched. I spoke with a couple devs after and the 4K native thing was confirmed. I also pixel counted 30+ shots and had multiple people examine them as well. Never came back as anything less than native 4K. However, there were still performance dips AND the map system exhibited horrible tearing and slowdown. They have since patched the game multiple times and most of those issues are cleared up including the map. It runs great now!

I am certain that this patch, given there is so little details about it (As you say, its a suspicion) is not enough to do a video upon. But, if you can get confirmation of this, would be out of this realm to add an update to the existing DF Analysis on FC5? If you think VG has a different version, perhaps the data can be cross-referenced and, when there is consensus that these are differing versions, providing an update to the existing DF article?

Secondly, regarding Redout, Dave missed that because the game doesn't use dynamic resolution - the resolution changes are set at fixed intervals rather than dynamically scaling based on load. That was a HUGE mistake on his part that cost us dearly. I was on vacation when that hit and it caused so many headaches. Crazy that he missed it but, hey, he's no longer doing work for the site.

If i recall correctly, this was mentioned in the DF video on Redout, or on a post by you on ERA since i just wrote the same thing above (And i had not yet read this part yet). Redout is, atleast when it comes to pixelcounting, an exception edge case. Where many games use reconstruction or dynamic framebuffers with vsync engaged (or not), Redout has what you describe. I reckon this game is definitely a benchmark game for testing if other games do it, aswell (I believe Redout was Unreal Engine 4 from memory.)

But i remember the meltdown over it on ERA and i can't say DF acted any way unprofessional on this. Mistakes happen, and the Redout devs eased up afterwards too, but a lot of the initial meltdown was on their tweets at first.

I've always tried to be humble and apologetic. I'm doing my best but I'm obviously not good enough. I know it. I hate myself enough already. That's fine.

Eh, John, i am going to assume you are being sarcastic here, but knowing your earlier posts when people took aim at your work, i have to doubt this assumption :/ And if this is geniune, than damn it man, that upsets me greatly. I should write something privately.

Show some compassion, man. Stop it. It's just video games. You have no idea how much it hurts. I wish you'd actually stop and consider this. You keep complaining about Bierton - he's not even a part of the site any more.

I know of TLW's history but i am not going to call him out - But i will say that stuff like this really upsets me a bit. We have talked on occassion at ERA, i appreciate your work very much - Because you have that shtick for useless and obscure knowledge, just like me. Except you do it for an audience and delegate myself to making halfassed posts on forums

Whilst you never should not cancel out critical voices, i do feel that you should zone out on this kind of nitpicking. Some users will never be pleased, will always have something to complain over. I for one am surprised that Alex isn't given more flak for his voice, for his ERA avatar, for his editing. The thing is, These folks, you won't win them over. That does not discredit TLW's view's - It just means that its better to quit giving in on that kind of nitpicking. Because clearly, it affects you on a more personal level than it should have. And it should not John.

Anyways, I'm out. It's clear that there's no point in being here.

John..... Stay for everyone else. Ill give Alex a shoutout to get in touch with you if you happen to not come here anymore.

Nice one TLW, another really good person doing a damn fine job annoyed by your blatant lack of self control and fanboyism. It's people like you that give sites a bad name. Just go to a Sony forum and sit among yourselves there, you clearly don't want to hear common sense or engage in actual discussion, you just want to spout the same lies as truth. Truly you are the Trump of the gaming world...

Hey Odium! We talked on ERA! Overdose, remember!

Don't mind him, 99% of the people on here know exactly why he posts on this forum. I'm just surprised he hasn't been banned yet.

Let's not advocate for his banning, but report - I do find his analysis amusing from time to time.

 

[c0de]

I don't post here often these but I find some of these posts mean spirited. It's not really very nice to lash out at someone like this. I'm in a position no different than NX Gamer (who I consider a friend, by the way) and VGTech. We all work by ourselves from home and try our best. DF is not a large organization.

That said, in the case of Far Cry 5, I actually have a suspicion that the game has been patched. I spoke with a couple devs after and the 4K native thing was confirmed. I also pixel counted 30+ shots and had multiple people examine them as well. Never came back as anything less than native 4K. However, there were still performance dips AND the map system exhibited horrible tearing and slowdown. They have since patched the game multiple times and most of those issues are cleared up including the map. It runs great now!

However, I do believe they've either added dynamic resolution scaling or changed the threshold to be more aggressive. I think the version VG Tech is testing here has simply changed since we first looked at the game. Thus, I don't think it's fair to parade around as one of us being incorrect because we're not looking at the same version.


The issue with Need for Speed is not a typical mistake and I'll explain why. I do not use emulators for Saturn. So why was the music missing and the speedo not working?

Well, it turns out that there are variations in how the game runs on hardware depending on the mix of game + console region. I used the NTSC-U version of Need for Speed with an NTSC-J console which I booted using the Pseudo-Saturn cartridge. My NTSC-U system was non-functioning at the time (since fixed). Since then, I've also purchased a PAL copy of Need for Speed.

What I discovered is that the game behaves differently on different region Saturn consoles. This is known to happen when matching PAL and NTSC but I've never seen it with NTSC-J and U. If you play the US version on a Japanese system, the music and speedometer simply do not appear. Since I bought the US version specifically for the episode, I thought this was normal. Booting it on an NTSC-U system, however, corrects those issues. Now, if you boot the PAL version on either NTSC systems, you get OTHER weird glitches.

Basically, I discovered that you cannot mix and match regions in certain games and expect perfect results. There was no emulation used in that video (and you can tell based on the output quality of the capture - which is from an analog source) - the emulator would probably work properly, I'd imagine. This is something else entirely. It's my fault for mixing region codes like that but I've since started using the correct region console with the correct region game even though, 99% of the time, there isn't an issue when swapping between them.

Secondly, regarding Redout, Dave missed that because the game doesn't use dynamic resolution - the resolution changes are set at fixed intervals rather than dynamically scaling based on load. That was a HUGE mistake on his part that cost us dearly. I was on vacation when that hit and it caused so many headaches. Crazy that he missed it but, hey, he's no longer doing work for the site.

Wait, Dave got fired or did he leave on his own?

 

[TheMikado]

This thread is incredibly interesting where you've got some technical analysis with guys popping in to take a victory lap because their race car of a console outperformed a minivan equivalent.

 

[Bryank75]

Just picked this up for 40 euro, enjoyable game... looks good but not really up to GOW or Sony first party visuals.
Like the characters and how it keeps it light though and the gameplay is decent... bow is satisfying.

 

[xStoyax]

What program are these sites using that show the FPS like that?

 

[Alexios]

Couldn't find a newer thread relevant enough with search not showing them by date. Good analysis of Far Cry 5's setting by RockPaperShotgun.

And that's what makes Far Cry 5's botched take on fascism in Montana so disappointing. Despite being an obviously Christian cult, the cultists sure never talk about Jesus. They weirdly aren't anti-Semitic either, not that Ubisoft included Montana's persecuted Jewish community in the game, and seeing as how every third cultist you shoot is black I guess we're supposed to assume they're not white nationalists either. Any time the game has an opportunity to offer the most basic critique of white America through representation of its worst actors, it backs off.

 

[Redneckerz]

Couldn't find a newer thread relevant enough with search not showing them by date. Good analysis of Far Cry 5's setting by RockPaperShotgun.

You might aswell make a seperate thread for that as this does not discuss tech, but more the setting.