Support NeoGAF

[Gozer the Gozerian]

Disappointing that they apparently do little to flesh out Jensen as a character in this one. Seems like a waste to bring him back and not develop him further.

 

[IcyRhythms]

It really makes you wonder.

Then again, I'm recalling that shitty "UH OH Deus Ex on PC sure ain't lookin' good!" thread and that probably helped to color people's expectations



HOLY SHIT

Do you even play PC games? Real question.

Edit: I guess you do, so like you should know about ultra benchmarks n shit

Can't say I've played many PC games the recommended gave such poor results, outside of terrible ports of course.

 

[Huge Succeeded]

Contact hardening shadows by itself is a computationally expensive feature.

What is that? Is it essentially that super expensive shadow setting that GTA V featured that makes it so that shadows appear softer the further away the object casting them is?

 

[Movalpolos]

Too bad about the PC performance. Seems like more and more games coming out these days either broken or not living up to expectations for the PC player.

 

[dmshaposv]

Does this game look as good as TW3 and does ad much as well to guarantee those fps?

While I dont own the game, I think Witcher 3 is in a class of its own as an open world game.

The minute details and the dense but smaller environments in Deus Ex do have the more hand crafted feel of the Witcher 2 and this looks more detailed. But the scope of each game is different.

 

[Huge Succeeded]

Can't say I've played many PC games the recommended gave such poor results, outside of terrible ports of course.

I'm confused. Maybe I'm missing something.

What are these poor results you're referring to?

Because if you're saying what I think you're saying, and maybe I'm wrong, but if I'm not...

Well I'd like you to load up a copy of GTA V on recommended settings, and then boost the settings up. Is GTA V a terrible port because you can't hold 20 FPS on a 7870 at max?

 

[Rappy]

Thanks for posting this Nirolak!

I had been seeing that graph posted numerous times lately and always felt it was really dumb! It always felt like something someone who was too young to be critically aware of reviews in the late nineties and early-mid '00 had cooked up. Nice to see someone has actually looked at the data to prove it wrong. I can remember debates over review scores and ad-revenue from way back when.

Can someone tell me what conclusions this data is supposed to imply? Can't that data also be interpreted as backing up that scale chart? As in, as time goes on there are more and more games, and reviewers/critics have higher standards on what games are worth their time to review. On the other hand, users don't have to be vetted and don't even have to own the game to contribute to the user score. With the increased audience and internet users they can throw down 0's that greatly affect the User Score without actually reviewing the game.

It's dumb for a whole other reason too -- those two scales are the same. Logarithmic and power transformations are monotonic. Whether the space from 9 to 10 is bigger or smaller than the space from 7 to 8 is irrelevant. They still express information the same way. Go to the California DMV and you get a 50 question test where you fail if you get more than 4 wrong. This does not mean that you yell at the DMV worker that "Uh, excuse me, 45/50 is a 90%, fuck you for failing me", it means you engage brain cells and understand that the scale can be different depending on context. Even in the world where it is just absolute imperative to transform the scale from the current scale to one where games are uniformly distributed between a 1 and a 10 (this is a quantile transformation, basically), it's still monotonic--games that review better than other games would still review better than other games. So from a comparative perspective this would change absolutely nothing. The question is whether or not you can understand what the scale is trying to convey. Which you can do by, shock of shocks, reading the reviews... or the summaries... or comparing review averages between games... or engaging your brain for even a second.

But it's also dumb for a whole other other reason. People aren't randomly buying games. Hundreds, maybe thousands even excluding iOS, of games come out every year. Most people, even on GAF, buy a dozen or so at most. You're not randomly selecting from all those games. You wouldn't expect your collection to have some 1/10s, 2/10s, 3/10s, 4/10s, 5/10s, etc. The "average" game should be semantically interpreted as "not worth playing". Granted that there might be some subject matter or dev loyalty or whatever it is that puts it over the top for you, but most of your games should be tilted far towards the right. And the same selection bias applies to reviewers. Even if the average game produced was a 5, the average game reviewed would not be a 5, because reviewers don't randomly pick games off the shelf to review, they can only review so many and they prioritize games that are high profile or interesting. So if the average reviewed game was a 5, then the average game produced with be below a 5, thus causing the opposite worry of the one that the chart advances--with games being bunched up over the bottom half of the scale and sparse over the top end.

People complaining about review scales are basically people wearing Hot Topic "You laugh because I'm different, I laugh because you're all the same" t-shirts. They certainly aren't people who have completed high school math.

Context absolutely matters, but isn't that the point of the chart you're saying is dumb? That users nowadays interpret review scores as a game sux vs a good game and no inbetween? In a DMV written test, it's you either get this thing or you don't in the context of you're going to be handling something that can easily someone else's or your own life. In game reviews, there are various amounts of subjective areas where a game can have merits but a large amount of readers only care about the score.

This is the inherent problem with review scores and review aggregators. The end user only cares about the score, but just a score does not provide any context. OpenCritic was created to address the issues that Metacritic was causing to the industry. But one of the problems created is due to the whole existence of review aggregators. Why do you think that chart qualifies pre-2000 times?

 

[IcyRhythms]

I'm confused. Maybe I'm missing something.

What are these poor results you're referring to?

The 970 performs poorly and is the recommend GPU, is it not?

 

[Sn4ke_911]

https://www.twitch.tv/gamestarde Pc version will be streamed there (german)

Game just crashed not even two minutes in.

 

[Huge Succeeded]

The 970 performs poorly and is the recommend GPU, is it not?

Performs poorly at what settings?

 

[DeathSparks]

This might seem a little off topic but someone found a Sonic The Hedgehog Easter Egg in Mankind Divided.

 

[KC Denton]

The 970 performs poorly and is the recommend GPU, is it not?

Recommended for not max settings. If I ran Witcher 3 on max settings with a 970 it would not perform very well, but I would not call the Witcher 3 a bad port because of that.

 

[IcyRhythms]

Performs poorly at what settings?

Have you not seen some of the numbers?

 

[Huge Succeeded]

Have you not seen some of the numbers?

Nope, that's actually why I'm asking. I'm aware that I could be wrong, but it seemed like the usual benchmark whining when I first dropped into this thread, wherein people complain about *fingerquotes and a derisive tone* poor PC ports because a specific GPU can't handle max settings with 8x MSAA

FYI those are at the absolute maximum settings. "Recommended" does not mean "run at max settings."

Oh christ so I was right

*bangs head on desk*

So IcyRhythms...

Well I'd like you to load up a copy of GTA V on recommended hardware, and then boost the settings up. Is GTA V a terrible port because you can't hold 20 FPS on a 7870 at maxed settings, extended distance scaling, 8x MSAA?

 

[KC Denton]

Have you not seen some of the numbers?

FYI those are at the absolute maximum settings. "Recommended" does not mean "run at max settings."

 

[Soltype]

https://www.youtube.com/watch?v=02uxM3vyClY

Found this video, look at the jump from ultra to high, almost double the framerate at the same spot.Something is chewing up performance at ultra, from what I've seen shadows seem to be a big part of it.

 

[BigTnaples]

The 970 performs poorly and is the recommend GPU, is it not?

What the hell are you talking about?



The 970 handles Ultra @1080p just fine.



Recommended does not mean it should handle ultra, but it is. That's actually more impressive than anything...

 

[Robobandit]

Can't say I've played many PC games the recommended gave such poor results, outside of terrible ports of course.

I think it's safe to say that the 'recommended' specs were not intended for playing the game maxed out on 'ultra' w/ MSAA.

I can't think of a game where the 'recommended' specs were acceptable for pushing all the sliders to the max.

 

[Foxxsoxx]

This might seem a little off topic but someone found a Sonic The Hedgehog Easter Egg in Mankind Divided.

I never asked for this.

 

[Carlius]

780ti sli has better scores than a 1070...wtf man

 

[Crimsonclaw111]

Is that Knuckles thing for real? Lol

 

[JudgmentJay]

I don't get the performance whining. The game looks to run great at high settings. As usual, ultra has some way over the top performance-hungry features. I'd wager a 970 will hit 60 with a mixture of high and ultra settings.

 

[Huge Succeeded]

It's like, sometimes I really wish this was GameFAQs where I could just requote something like ten times for emphasis without being banned. Because just LOL. I'm sorry to use this as an example, CertifiedFP, but I wish I could boost the font size on bolded and just plaster it across everyone's screen

Damnit, I knew this would be one of those games that wouldn't be hitting 60FPS.

Well, I mean, that benchmark is at max settings though.

 

[JayB1920]

The 970 performs poorly and is the recommend GPU, is it not?

The recommended card doesn't mean it will run maxed or even Ultra settings. Rise of the Tomb Raider recommends the High preset for recommended cards. The Ultra preset isn't even maxed in that game as some settings can be tweaked even further. Nothing wrong with Having settings that push even the best cards even if you don't necessarily see much of a difference.

 

[DeathSparks]

Is that Knuckles thing for real? Lol

Yes, it's no wonder why the Deus Ex twitter account are close friends with the Sonic twitter account.

 

[Movalpolos]

Performs poorly at what settings?

Game just crashed not even two minutes in.

Not a good look.

 

[Huge Succeeded]

Not a good look.

Sometimes PC games crash. Sometimes review code isn't perfect. Let me know when it's a widespread problem with the finished product and then maybe I'll concede that you have something resembling a point, because one crash doesn't signify a poor port to me, nor does it support the arguments made in this thread to that end.

 

[EdibleExplosives]

This might seem a little off topic but someone found a Sonic The Hedgehog Easter Egg in Mankind Divided.

lmao

 

[A-V-B]

This might seem a little off topic but someone found a Sonic The Hedgehog Easter Egg in Mankind Divided.

I didn't know Sega was affiliated with the game.

 

[jrcbandit]

I think it's safe to say that the 'recommended' specs were not intended for playing the game maxed out on 'ultra' w/ MSAA.

I can't think of a game where the 'recommended' specs were acceptable for pushing all the sliders to the max.

Exactly, it's getting ridiculous that so many people freaking out that running the game on Ultra with max AA isn't giving 60+ FPS on 2 year old hardware or even new hardware for that matter. Games can have options meant for future hardware or SLI'ing of current hardware. The only caveat is that game developers really should tell us what specific settings affect performance the most to make it easy to know what to lower first (like the PC GoW 4 developers are doing). I'd imagine most people with high end video cards will still be running ultra, just with a few specific settings turned down.

 

[Crimsonclaw111]

Yes, it's no wonder why the Deus Ex twitter account are close friends with the Sonic twitter account.

Crazy. Augment my hedgehog.

 

[Chris_Rivera]

Also, just because MSGV performs amazing on minimal hardware, does not make every other game a "garbage" port. Some games go for lots of graphical complexity, which is a good thing for the PC side, we want that.

The console version looks pretty darn good too.

 

[JudgmentJay]

780ti sli has better scores than a 1070...wtf man

Why is that weird? 2 780Ti's are more powerful than a single 1070.

I think some people who have been predominantly console gamers are having a difficult time rectifying the fact that they can't play the game to it's maxed potential on a platform with their GPU.

Has this been linked yet? Should be required reading before posting about PC performance: http://www.neogaf.com/forum/showthread.php?t=885444

 

[WadeitOut]

I think some people who have been predominantly console gamers are having a difficult time rectifying the fact that they can't play the game to it's maxed potential on a platform with their GPU.

 

[Lionel Richie]

This might seem a little off topic but someone found a Sonic The Hedgehog Easter Egg in Mankind Divided.

Can't wait for the Knuckles DLC on that game.

 

[Lister]

Not a good look.

Polygon quoted bugs and issues with the consoles too.

This is all prior to day 1 patch so we shouldn't be judging it quite yet.

 

[XANDER CAGE]

It's pretty disappointing to see that pretty much across the board everyone is saying the story is "eh" at best. Human Revolution was hardly Shakespeare but still. Ah well, still pretty excited to play another one of these.

 

[MasterMenace]

PPolygon quoted bugs and issues with the consoles too.

This is all prior to day 1 patch so we shouldn't be judging it quite yet.

Isn't the day1 patch out already (patch 1.02 or something like that)?

It's pretty disappointing to see that pretty much across the board everyone is saying the story is "eh" at best. Human Revolution was hardly Shakespeare but still. Ah well, still pretty excited to play another one of these.

Story was forgettable for me in HR. But what stood out the most was the atmosphere the game puts you in. Eidos Montreal really nails that aspect of making you feel you are in a cyberpunk, futuristic, augmented world. Add on amazing, diverse gameplay...that's all I asked for. And that is what I got. Can't wait to play MD!

 

[injurai]

Wow.

High praise.

 

[Ronin Ray]

Disappointing that they apparently do little to flesh out Jensen as a character in this one. Seems like a waste to bring him back and not develop him further.

If that's the case let me create a character

 

[KillerMan91]

Too bad about the PC performance. Seems like more and more games coming out these days either broken or not living up to expectations for the PC player.

lol. Generally games are less broken on PC than they have ever been. Of course though there is always exceptions

 

[DJ Lushious]

(and may be on the short side?)

Well, to be fair, Human Revolution can be zipped through exceptionally quickly, too. That's hardly not the best way to experience the game, of course, but still.

 

[Kenzodielocke]

lol. Generally games are less broken on PC than they have ever been. Of course though there is always exceptions

Yeah, things get more blown up these days.

 

[tesqui]

Too bad about the PC performance. Seems like more and more games coming out these days either broken or not living up to expectations for the PC player.

Yeah it kind of seems like history is repeating itself. Early to mid last gen PC ports were a damn wreck. It got better towards the end of the gen.

Although, hopefully these benchmarks are just using rediculous graphics settings. I'd be happy if I can get a consistent 60 at medium and high settings.

 

[damstr]

Yeah, things get more blown up these days.

Maybe because of NMS and this are like back to back with poor PC performance.

 

[Nags]

Wow, so the PC version is bad?

 

[Shin-chan]

I was all in the hype for Human Revolution. It felt like a legit MGS game in the trailers and while it was great it did disappoint from that perspective.

Haven't felt the hype for this one yet. Maybe I'll pick it up but the Ni Oh demo is way too enticing.

 

[Kevin Shields]

Interesting that the CGM review gave it a 10 because of the narrative while everyone else had story as the negative point. I wonder if it's going to be really bad or just kinda there. If it's the latter then I'll be happy as gameplay has seemed to improve.

 

[Laurentius]

ZePEZ

 

[Carlius]

Wow, so the PC version is bad?

no. how bout you wait for real perfomance impressions? benchmarks are one thing, but i dont believe shit until i acutally play the game. am not gonna cancel a pre order cause of soemthing like this when it can all be fixed with a patch, specially considering its nixxes.