DF and Leadbetter are different entities. Leadbetter write the article.
Where is exactly the hate to precise who wrote the article?
-
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.
Digital Foundry: GTA V PS4 and Xbox One compared in new frame-rate stress test.
- Thread starter Hanmik
- Start date
DeFiBkIlLeR
Banned
Leadbetter is the founder and owner of DF.
Leadbetter is a Microsoft plant, along with all the rest of video gaming publishing.
alexandros
Banned
And? Is he not part of the DF crew? Has he gone rogue?
krypt0nian
Banned
For quite good reasons. Let's not rely on the hater word, hun.
drotahorror
Member
150mhz is pretty negligible. You never hear about people OCing cpu's for a 150mhz gain. Maybe the garbage cpu's that ps4/xb1 use benefit greatly from tiny ass overclocks though, I am ignorant to that atleast.
He's right. The Scene on the PS4 has more cars.
http://youtu.be/5z6ONWZ7etA?t=40s
Start the video at 40 seconds, PS4 has more cars on the screen.
During the gun fight, the frame-rate drops more on the XB1
http://youtu.be/5z6ONWZ7etA?t=2m22s
This should be a better comparison considering there's the same amount of objects of the screen during this event.
Your mind sometimes is something of mystical to understand. It walks in ways beyond the human nature. I'm not sure what you want to say here, really.
alexandros
Banned
I'll make it simple then: why is the fact that Leadbetter wrote the article of importance? Someone above said that he is a Microsoft plant, a serious accusation. Is there any proof?
Yes, this ..
along with a whole other bunch of posts rightfully calling Leadbetter out on his BS.
But you can carry on with the 9/11 tinfoil hatting.
DeFiBkIlLeR
Banned
Yep..in the same way you can find proof that 9/11 was a government conspiracy.
Proof? I don't need proof when I have selective quoting and a funny feeling in my gut.
alexandros
Banned
nowheretorun
Member
Without a doubt.
Hopefully something good will come out of this soon - as in a Sony or another developer who really shows the power of GPCPU (?).
Perhaps Sony should have concentrated on a tools to make this easier access for Developers?
Or perhaps Sony didn't want this though too early so as to help along a continual graphical improvement through the generation (as in to combat the ease of 'to triangle per Mark Cerny' statement). Ie. It makes things interesting for Developers to push themselves to get better/more efficient results.
Windom Earle
Member
The gun fight starts earlier.
Also XB1 drops during the police chase scene after the 1st heist where PS4 has no issues.
Durante has said that not everything a CPU can do can be offloaded effectively to a GPU. It depends how parallelisable the task is.
So it's looking like the PS4 is more likely to drop in heavily populated areas while the XBO is likely to drop when alpha effects fill the screen? Makes sense considering the hardware, especially the XBO. I've noticed frame drops on the PS4 version but nothing to kill the experience.
No the reasons aren't good actually. It usually looks like people are bitter over him covering the lesser PS3 ports that plagued the system all those years. Many of the claims to support his hate have been from exaggerated examples or misunderstanding of what he says.
Not to mention how people sometimes just assumes there is a slant towards Microsoft when he isn't even the author of the article.
No the reasons aren't good actually. It usually looks like people are bitter over him covering the lesser PS3 ports that plagued the system all those years. Many of the claims to support his hate have been from exaggerated examples or misunderstanding of what he says.
Not to mention how people sometimes just assumes there is a slant towards Microsoft when he isn't even the author of the article.
It's not the same person who said these things, said that, it's the person of Leadbetter who wrote the article so if said bullshit it's not the whole redaction responsible of what write in the article. About the MS plant, well Leadbetter surely not hate them, but I don't judge a person for his love.
adamsapple
Or is it just one of Phil's balls in my throat?
Its a long standing tradition that whenever a DF article about something comes up, people bring up his name to try and discredit it as if he's a corporate shill or a sellout or something.
Its completely stupid, but still continues to happen in every Richard Leadbetter written article.
HORRORSHØW
Member
did you watch the video by any chance?
I think the point it's more ps4 not need to offload what xbone does with the juicy overheated cpu, because it's meaningless in the whole performance.
Killroyskbg
Member
Off course not but it does not need to do everything and it was never meant to. However unloading anything helps and there is always something that can be and the amount will only increase and its efficiency. I would much rather be CPU bound in the PS4 case with its hardware than GPU bound in the Xbox case.
No there is no real proof, just bitter tears.
You can see in the grass comparison thread many people claim that he made a big deal over the difference in foliage in the RDR comparison while downplaying the the significance of the difference in GTAV. However here is the original RDR quote surrounding the foliage:
Nothing in that quote is a lie, he even tries to educate the readers on why there are differences. Worse he even goes to say it's an "incidental detail" but that didn't stop people from exaggerating.
A flop is a flop is a flop. The CPU can handle certain code better than the GPU and vice versa but flops are flops and these are just a certain type of workload that is easy to calculate to form a comparison using actual, objective numbers rather than subjective reasoning.
If you think that cache allocations or OS CPU affinities will lead to a 20% FPS difference then that is a bit barmy tbh. That is more barmy than the idea that the CPU alone is the culprit, it may be a contributing factor but it is not the sole reason.
I agree in the concept that there are many unknowns and there are specific differences, such as the CPU - RAM bandwidth being 30GB/s on Xbox One and 20GB/s on PS4. Does that make a difference? Well I do not think so as even an i7 4770K only has 24GB/s of memory bandwidth when pared with DDR3 1600 and that is by far a more powerful CPU which could actually use the bandwidth.
All of these little things do make a difference but you need to always put it in the context of the overall performance picture and sometimes when looking at the finer details it can be hard to consider that it is only going to make a tiny difference overall.
The point I am making is that with what we know of the hardware CPU bottlenecking does not explain the differences in AC:U or the occasional differences in GTA:V. These differences are just too great for it to be CPU alone so for a website like DF to state that it points to CPU bottlenecking is just poor analysis. Even more so when you look at Kabini reviews which show a 28.1% clock speed increase giving 4% gains at best in gaming scenarios and 11.7% gains in CPU limited scenarios.
RurouniZel
Asks questions so Ezalc doesn't have to
Open-world game engines have been built to be very CPU-centric for some time. In games that don't stress the CPU the way Open-world games like AC: U and GTAV do the PS4's GPU and RAM win out. In CPU heavy-games we're seeing the XBO take the lead.
alexandros
Banned
It seems to me the poor guy is being blamed for nothing.
This poor guy has written a bunch of articles during 2014 which are...horrible for his reputation.
Gaf imploded in those days.
I think there are many unknown variables which *do* influence performance but we don't know exactly how much. And if many of these come into play things can even get worse. Proving points with different CPUs in a different context with different APIs and different OS's and schedulers will lead us to nohing but even more confusion. We need hard facts on how the new consoles really work and which variables in which way really make for the performance, currently. As we don't know this yet, any talk about power differences might be legit but doesn't tell us why game x performs worse/better on console 1 or 2.
But I agree that CPU clock itself shouldn't make a difference in this case but this doesn't mean it is playing a role in it.
Next.
dr. apocalipsis
Banned
150mhz is a 9,37% increase over 1,6Ghz for 8 cores on consoles CPU.
150mhz is a 3,95% increase over 3,8Ghz for 2/4 cores on average PC CPU.
Open world games are some of the hardest to optimize because of all of the unpredictable variables that result in having a free roam world filled with AI. I think it's a great port since the game doesn't look to struggle to hit 30fps like we saw with so many open world games last Gen.
Maninthemirror
Banned
First reply jn this thread and the ps4 seems more Consistent in all the videos than the xb1 version in Df's own videos
That's not what I saw in those videos posted on this very page. I saw the PS4 doing what you just described and the XBO version dropping a bit of frames, usually in the 2-4 FPS range quite a bit in action scenes.
Dont blame the poor guy, you 9/11 conspiracy theorist!
I doubt that tiny CPU tick is going to seriously do that much. Maybe, but not much. "Take the lead" is going to be at a lower resolution or compromised GPU-centric things most likely as well.
Also why in this CPU centric open world game does the PS4 outperform the XBO according to the videos posted on this page?
I don't know if I'd say that in this case.
There are scenes in the DF video where the XB1 is giving better performance and scenes where the PS4 is. In the car-chase example above both are stuttering, with a min-FPS gap of 4fps. In a scene later in the video XB1 is stuttering - with a min fps gap of 5fps - while PS4 remains locked at 30.
Either this game is not CPU bound enough, or XB1's CPU advantage is not substantial enough here, to offer a general overall lead at all.
I though it was explained with tons of examples in the pc rig comparison, that cpu has barely an advantage compared the Ps4 which not translate in something of really palpable.
travisbickle
Member
What's happening in the shoot-out scene that is taxing the XB1? It mentions effects in the article but even when nothing's happening the XB1 is still dropping frames?
well... I see a gunfight scene totally CPU-bound and the PS4 maintains a remarkable stable framerate whereas the XBO chugs along for the entire duration of the sequence, in the meantime the PS4 manages to do it with better gfx as well, also I see random driving scenes where the XBO drops frames too, we are not talking about an XBO "lead" here, more like about a "lead-better".
CPU, the effects come and go and the XBO still drops frames until the AI (gunfight) stops stressing the system.
dr. apocalipsis
Banned
Are you aware that your GFLOPS maths only cover CPU's FPU?
Talking about RISC and CISC was totally meaningless even 10 years ago.
RISC isn't a technology, just a CPU design philosophy. And CISC isn't even a thing, just how RISC apologists called everything else that wasn't RISC.
With current CPU technology there is no place for such distinctions.
Like I said with AC:U - chasing a few frames of difference down to a marginal CPU clock-speed differential without having any context about code, data setup or actual performance metrics of such frames is basically as good as random-guessing (and Gaf has banned certain sites for doing that with sales).gofreak said:
Eg. I could take a guess that high-speed scenes are related to streaming, and data-unpacking is where the differences come from. This could be due to dedicated LZO-units performing differently (which aren't the same across two consoles), issues with I/O async performance (one Console implementation just being more demanding then the other), using different compression alltogether, or maybe just the two tests not being 1:1 in terms of processing load (which they obviously aren't)...
Thing is that's already 4 variable guesses, all possible without any differences from "CPU" performance itself - and we're just scratching the surface here - could list a lot of other potential causes without knowing any better detail that we have.
travisbickle
Member
Too many variables.
I'm out of here until the inevitable post-stability-patch Digital Foundry frame-rate stress test.
Indeed. Outside of maybe two scenes in the DF video, there's blips on both consoles that could be pretty much anything or nothing of particular consequence.
And in the two scenes were things are a bit more suspect, it's impossible for us to really say for sure what's going on.
I'm just poking at the narrative that this is another 'CPU bound open world game' that does better on XBO. Even if one wanted to assign CPU perf as the factor where PS4 is doing worse in the DF video, I think you'd have to assume people haven't watched the video to try and pass that idea.
Krisprolls
Banned
I'm not sure why we care that much about Digital Foundry when it has shown time and time again it had a MS bias. Minimizing differences is now their business, the same way they used to magnify every 2 fps advantage the Xbox 360 used to have over the PS3.
XB1 and PS4 share a very similar architecture but PS4 is around 30 % more powerful, so of course in every game it'll show by running better on PS4, unless you really try hard to make the XB1 version runs the same (mostly by downgrading the PS4 version on purpose).
As far as I know, a slightly better XB1 version happened exactly once, with AC : Unity. But it had nothing to do with CPU and everything with Ubi wanting to force parity.
Analyzing made more sense when we had different architectures. Now ? Not so much... Of course the most powerful machine will get a slightly better version than a less powerful machine with the same architecture.
XB1 and PS4 share a very similar architecture but PS4 is around 30 % more powerful, so of course in every game it'll show by running better on PS4, unless you really try hard to make the XB1 version runs the same (mostly by downgrading the PS4 version on purpose).
As far as I know, a slightly better XB1 version happened exactly once, with AC : Unity. But it had nothing to do with CPU and everything with Ubi wanting to force parity.
Analyzing made more sense when we had different architectures. Now ? Not so much... Of course the most powerful machine will get a slightly better version than a less powerful machine with the same architecture.
No shit that the FPU deals with floating point operations, just like the shaders in a GPU deal with floating point operations and it has nothing to do with the TMUs, ROPS etc. It is just a single metric to show the scale difference between the CPU and the GPU.
Funny that the 3rd fastest super computer (Sequoia) is using a Power PC based CPU which is using a RISC type instruction set.
RISC vs CISC are different CPU architecture paradigms just like the Object Oriented paradigm vs the Functional paradigm. Going from one to the other is possible but it takes more work than going from different implementations of the same paradigm.
Exactly this, looking at the difference and just pointing at the CPU is lazy. If DF really wanted to earn their stripes they would see if they could speak to the devs to find out what is really going on. Sure the devs may not want to share but they could do a better analysis than 'points to a CPU bottleneck'.
Funny that the distinction doesn't exist in current x86 (or x64) cpus as they are RISC internally since a good amount of time.
ninjablade
Banned
I don't know, MGR had a 10fps average advantage, yet they called it a draw cause the ps3 had slightly better AA, which made them seem sony biased to me.