ASUS STRIX R9 Fury DC3 Video Card Review @ [H]

The only game I wouldn't use in the benchmark is Dying light. I honestly don't know a single person, or anyone on my friends list who even owns or plays it.

I would rather see a FFXIV (since it has 5 million subs) and no i dont mean in builtin benchmark.......or an MMORPG or something other then that game.

Hell even World of Warcraft with 4xSSAA and 8XMSAA bits can make my 970 crawl.

Either way the review great, just Dying light is a head scratcher.
 
I think Dying Light is there for the VRAM. MMO's are generally going to be more CPU limited.
 
DASHlT said:
The only game I wouldn't use in the benchmark is Dying light. I honestly don't know a single person, or anyone on my friends list who even owns or plays it.

I would rather see a FFXIV (since it has 5 million subs) and no i dont mean in builtin benchmark.......or an MMORPG or something other then that game.

Hell even World of Warcraft with 4xSSAA and 8XMSAA bits can make my 970 crawl.

Either way the review great, just Dying light is a head scratcher.
Click to expand...

Dying Light was really popular when it was first released, but it's been dropping off pretty heavily since then. 30-day population average is only 1,618 players.

It's probably the only game I also disagree with using in benchmarks since I don't think it looks anywhere near as good enough to justify the performance.

If Batman: AK ever gets fixed I'm guessing [H] will replace Dying Light with that.
 
xIronCrossx said:
I think Dying Light is there for the VRAM. MMO's are generally going to be more CPU limited.
Click to expand...

Alright.....but why should it matter since only like 2000 people play it peak hours.

I understand the need to test Vram. But when 1 game that so happens to use a massive amount of Vram, while others don't.....Well that would tell me it was lazy devs....just my 0.02c.

The others I'm fine with. I mean I am not even here bitching about the review, it was well done lol.

Just that game needs replaced.

Edit: To a point it would be CPU limited. But even in world of warcraft you can test GPU horsepower very easily......I dunno i don't see an MMO being a horrible benchmark
 
durquavian said:
Actually being a forum it is expected from time to time to debate even age old dilemmas. Avoiding any such intercourse defeats the point of a forum. Unless this is more of a club, then the name is very misleading. At any rate, I always point out curious details and thought processes when they seem to be faulty or flimsy. Again I was just discussing some thoughts which at no point is it necessary to facilitate change nor required. But this dodge and weave every time it gets brought up is disheartening.

But let me point out one thing from this review which adds to the point I made about positives and negatives of benchmarking runs as done with [H]. Look at the Witcher3 the graph goes almost identical through peaks and valleys. This shows that the runs are nearly identical or in the least the same occurrences are evident in both runs at equal times during the gameplay. Now look at Grand Theft Auto 5, in this case the graph deviates toward the end (saw this in the fury review with Dying Light I think but vendors switched) with Nvidia dropping as AMD goes up. At this point it may be that the gameplay deviated and this is what can skew results, the negative of gameplay benchmarking.
Click to expand...

The problem with most of your writings is that you think people are somehow conspiring about AMD. I would just stop right there and ponder.
 
DASHlT said:
Alright.....but why should it matter since only like 2000 people play it peak hours.

I understand the need to test Vram. But when 1 game that so happens to use a massive amount of Vram, while others don't.....Well that would tell me it was lazy devs....just my 0.02c.
Click to expand...

This right here! The Fury X was made to look inferior because it did not have enough VRAM at 4GB for Dying Light at even 1440p resolution. Anyone in the market for a high-end video card would be running towards something with much higher VRAM if they wanted to even THINK about playing 4k. However, when tested with all other games at even 4k resolution, 4 GB of ram was NOT the bottleneck. Hell, these cards were tested at 4k SURROUND in crossfire with the same 4 GB vram "bottleneck," yet they were competitive to the Titans in SLI. Truth is, Dying Light developers did a very crappy job of allocating memory.

http://www.tweaktown.com/articles/7226/amd-radeon-r9-fury-video-cards-crossfire/index5.html

http://www.tweaktown.com/tweakipedi...re-triple-4k-eyefinity-11-520x2160/index.html
 
DASHlT said:
The only game I wouldn't use in the benchmark is Dying light. I honestly don't know a single person, or anyone on my friends list who even owns or plays it.

I would rather see a FFXIV (since it has 5 million subs) and no i dont mean in builtin benchmark.......or an MMORPG or something other then that game.

Hell even World of Warcraft with 4xSSAA and 8XMSAA bits can make my 970 crawl.

Either way the review great, just Dying light is a head scratcher.
Click to expand...

MMO testing is going to be extremely problematic unless the reviewer is actually both an active player and can actually enlist a large competent group to help run such a test as really the only valid/interesting, representative, and somewhat reproducible results would be larger group "raid" scenarios.

Also FFXIV does not have 5 million subs in the sense you likely think it does. They phrased it that way to generate more publicity but they actually were referring to aggregate subs since release or basically the equivalent to copies sold.

Related to this I think one caveat of HardOCPs testing is that they need to be games the reviewers are at least (somewhat) inclined to actually play? They can correct me unless this is wrong. Otherwise you would not really get the representitive real world test they are trying to do since without actually being a player of said game how would you gauge what is actually representitive of what real players are doing?

Then there is the issue in which (and I think this is a problem with game reviews in general not just HardOCP) game review samples are often rather genre and type limited.

Another comment though I'd make is that there are plenty of hardware review sites out there. HardOCPs methodology is somewhat more unique which gives you an alternative look. I don't see the point in wanting them to change as that would just be one more review site out in the ocean, you can easily find the more generic large test suite results in anynumber of hardware sites, why add another one to the mix? I'd rather see more variety from different web sites.

Also I'm wondering if the people be critical today were also repeatedly being critical of the stretch run of Farcry 3, Crysis 3, Tomb Raider and Metro Last Light?
 
durquavian said:
Actually being a forum it is expected from time to time to debate even age old dilemmas. Avoiding any such intercourse defeats the point of a forum. Unless this is more of a club, then the name is very misleading. At any rate, I always point out curious details and thought processes when they seem to be faulty or flimsy. Again I was just discussing some thoughts which at no point is it necessary to facilitate change nor required. But this dodge and weave every time it gets brought up is disheartening.
Click to expand...

This is starting to sound like really skilled trolling.

durquavian said:
But let me point out one thing from this review which adds to the point I made about positives and negatives of benchmarking runs as done with [H]. Look at the Witcher3 the graph goes almost identical through peaks and valleys. This shows that the runs are nearly identical or in the least the same occurrences are evident in both runs at equal times during the gameplay.
Click to expand...

That just shows how good they have gotten at doing real world gameplay testing, and being as fair as they can, by recreating the run-thru so closely each time.

durquavian said:
Now look at Grand Theft Auto 5, in this case the graph deviates toward the end (saw this in the fury review with Dying Light I think but vendors switched) with Nvidia dropping as AMD goes up. At this point it may be that the gameplay deviated and this is what can skew results, the negative of gameplay benchmarking.
Click to expand...

Or, GPU X starts to get too hot on Game Z, and over time the performance takes a shit. Which is exactly what real-world testing shows, and canned benchmarks do not. You or someone else even pointed out that "it only takes 2 minutes to run a canned benchmark..." which is exactly why they are the poorest possible representation of the gameplay experience. Whatever in-game situation caused the performance shown on the graphs, it was something in the game, therefore something players will come across as they play the game. It's still valid. And as you pointed out "the play thru's are practically identical". I trust [H] enough to believe they are comparing the same run-though as best they can between the cards.

Other things that can only come from playing the game, are things like "Playing Game Z on the <Video card X> felt very choppy or stuttery. Even though the frame rates look great, the game play experience was bad." Just one more thing a canned benchmark can't show you. In fact I think it was [H]'s testing methodology which pointed out some glaring issues with AMD cards/drivers a few years back, which they have corrected with frame time monitoring in the drivers. This has lead to a much better gameplay experience for AMD users (finally).

What about "Graphics feature Y really makes this game look great..." More details that while subjective, are exactly the kind of thing I like about [H] reviews.

If none of the above makes sense to you, you shouldn't be reading [H] reviews.
 
So it sounds like drink the [H] review policy Kool Aid or GTFO. At this point, I take these reviews with a grain of salt. At best, they are a small piece of a larger pie that consists of reviews/editorials from a number of other sites I respect. Some here go out of their way to suck up to Kyle, Brent, whoever. Disgusting.
 
cageymaru said:
AMD needs to send Unwinder one of these cards so that he can unlock the voltage in MSI Afterburner. I'd love to see what one can do overclocked.
Click to expand...

hqdefault.jpg
 
Eradan said:
So it sounds like drink the [H] review policy Kool Aid or GTFO. Disgusting.
Click to expand...

Yep. To be fair, this is not Kyle or Brent's fault, it is the nut-swinging totalitarians that roam the forums.

borg1.jpg


In all seriousness, I would still like Kyle or Brent to explain why they choose to run 4k benchmarks in their reviews of the GTX 780/780ti, 290x, GTX 970/980 and the 390x but not the Fury Pro.
 
DPI said:
Click to expand...

haha that's pretty funny...but sadly it could be true. I'm sure a few full-retard "overclockers" will fry their cards. I have no big delusions about massive overclocks for this card but there could be a decent bump there if done right
 
GoodBoy said:
This is starting to sound like really skilled trolling.



That just shows how good they have gotten at doing real world gameplay testing, and being as fair as they can, by recreating the run-thru so closely each time.



Or, GPU X starts to get too hot on Game Z, and over time the performance takes a shit. Which is exactly what real-world testing shows, and canned benchmarks do not. You or someone else even pointed out that "it only takes 2 minutes to run a canned benchmark..." which is exactly why they are the poorest possible representation of the gameplay experience. Whatever in-game situation caused the performance shown on the graphs, it was something in the game, therefore something players will come across as they play the game. It's still valid. And as you pointed out "the play thru's are practically identical". I trust [H] enough to believe they are comparing the same run-though as best they can between the cards.

Other things that can only come from playing the game, are things like "Playing Game Z on the <Video card X> felt very choppy or stuttery. Even though the frame rates look great, the game play experience was bad." Just one more thing a canned benchmark can't show you. In fact I think it was [H]'s testing methodology which pointed out some glaring issues with AMD cards/drivers a few years back, which they have corrected with frame time monitoring in the drivers. This has lead to a much better gameplay experience for AMD users (finally).

What about "Graphics feature Y really makes this game look great..." More details that while subjective, are exactly the kind of thing I like about [H] reviews.

If none of the above makes sense to you, you shouldn't be reading [H] reviews.
Click to expand...

I wasn't arguing that the way they do it was bad or inferior in any way. It is a good philosophy and a good testing strategy. I only asked if they considered adding for just the games they have the occasional Canned benchmark to SUPPLEMENT the existing gameplay bench. I gave the positives for both and the negatives. So wasn't pushing one over another at all, just seemed those that followed whether intentional or not misconstrued my debate/concern.

As far as the graphs and why they deviate from time to time: That was my question. They didn't seem to break it down or even mention it. Those deviations can make huge differences in the results as far as min/max/avg. I even mention that the issue I saw in this review was happening to the 980 so consider that before anyone claims I am just trying to make AMD look better. Saw once someone mention video of the gameplay. I think that would be good so even we can see what may have happened. Unfortunately that adds considerable time to a review and extra cost of equipment, and possibly having to use different software to do the gameplay recording if not using video capture with hardware sources. So that doesn't seem quite viable, depends on the resources.

As far as speaking to features and gameplay feel and the like, I don't disagree.

Unfortunately anytime you say anything that isn't in the popular view amongst the present posters you get labeled and misinterpreted and out right disrespected. I usually try to see the bigger picture and see both sides. A rather lost art and little recognized in many forums.
 
CPO_Sharky said:
DPI posted that pic in response to a comment about overclocking the Fury. It was kinda funny that one time..............
Click to expand...


Yes and I enjoyed in the revel, is that a problem for you? If you feel left out I'm sure we can make more jokes for you to fit in. I'm being sarcastic if you didn't get it.
 
CPO_Sharky said:
No, DPI made the joke, you just tried to swing from his nutsack......
Click to expand...


Well I didn't know you liked me like that :p, you like men swinging from other man's nut sacks, weird. Is that a personal fantasy for you?

OT, get back to the topic man.
 
CPO_Sharky said:
In all seriousness, I would still like Kyle or Brent to explain why they choose to run 4k benchmarks in their reviews of the GTX 780/780ti, 290x, GTX 970/980 and the 390x but not the Fury Pro.
Click to expand...

They did...read the thread on the Fury X to explain why. If it is important enough to you, you will find the information. If not...well, that explains something too.
 
CPO_Sharky said:
In all seriousness, I would still like Kyle or Brent to explain why they choose to run 4k benchmarks in their reviews of the GTX 780/780ti, 290x, GTX 970/980 and the 390x but not the Fury Pro.
Click to expand...

We have had enough experience now with 4K display in-game testing and we do not think that any single GPU configuration truly lends itself to a great 4K gaming experience. We have stated this about both AMD and NVIDIA GPUs. We have not exactly been secretive about this. From today's review: "We do not feel as though either of these cards are suited for 4K resolution gaming as you simply have to sacrifice too many of the image quality settings in order to maintain an acceptable gaming framerate."
 
durquavian said:
Unfortunately anytime you say anything that isn't in the popular view amongst the present posters you get labeled and misinterpreted and out right disrespected. I usually try to see the bigger picture and see both sides. A rather lost art and little recognized in many forums.
Click to expand...

Maybe a little bit of that, and a little bit of you saying the same thing over and over and over again?
 
durquavian said:
Actually the repeats were to hopefully get those that were doing the first part:

Unfortunately anytime you say anything that isn't in the popular view amongst the present posters you get labeled and misinterpreted and out right disrespected. I usually try to see the bigger picture and see both sides. A rather lost art and little recognized in many forums.

to better understand my point.

What you can see quite clearly is that not one person debated the facts, referencing my exact statement or giving deliberate proof counter to my own. Even you just blew off any credence with vague statements and references to age old deliberations.

It is ok though, it just speaks to the fact that any counter argument that one may have had was probably too flimsy or lacking in ability to unseat mine.
Click to expand...

Dude all I can deduce is that you are trying to defend AMD as if it's sacred. The review gave us a perspective, why do we want to fight it out with the reviewer in order to get things done in a way we see fit.

You are going far beyond the line of simply suggesting something. What I can deduce is that you are trying to force feed your ideas onto the reviewer which is an unacceptable behavior.
 
durquavian said:
Actually the repeats were to hopefully get those that were doing the first part:

Unfortunately anytime you say anything that isn't in the popular view amongst the present posters you get labeled and misinterpreted and out right disrespected. I usually try to see the bigger picture and see both sides. A rather lost art and little recognized in many forums.

to better understand my point.

What you can see quite clearly is that not one person debated the facts, referencing my exact statement or giving deliberate proof counter to my own. Even you just blew off any credence with vague statements and references to age old deliberations.

It is ok though, it just speaks to the fact that any counter argument that one may have had was probably too flimsy or lacking in ability to unseat mine.
Click to expand...

If you are ever in the Dallas area, I want some of what you are smoking.
 
durquavian said:
Actually the repeats were to hopefully get those that were doing the first part:

Unfortunately anytime you say anything that isn't in the popular view amongst the present posters you get labeled and misinterpreted and out right disrespected. I usually try to see the bigger picture and see both sides. A rather lost art and little recognized in many forums.

to better understand my point.

What you can see quite clearly is that not one person debated the facts, referencing my exact statement or giving deliberate proof counter to my own. Even you just blew off any credence with vague statements and references to age old deliberations.

It is ok though, it just speaks to the fact that any counter argument that one may have had was probably too flimsy or lacking in ability to unseat mine.
Click to expand...

Except your arguments are not new nor are they creative and they bring nothing interesting to the table. As anyone that has been around here a few years can tell you this debate has popped up dozens of times (at least) and it's always the same points made ad nauseam. Kyle and co do not believe in canned benchmarks. They have done detailed studies on why they aren't reliable and even posted an article several years ago showing their reasons. Ever since I've been around (and likely long before that) the stance has been "if you wanted canned benchmarks there are plenty of other sites that do it". What you are asking [H] to do by adding canned benchmarks is to betray the very thing they believe in when comes to reviews. You are trying to play the victim card, yet you refuse to accept any answer given to you and just keep repeating the same crap over and over.
 
Derangel said:
Except your arguments are not new nor are they creative and they bring nothing interesting to the table. As anyone that has been around here a few years can tell you this debate has popped up dozens of times (at least) and it's always the same points made ad nauseam. Kyle and co do not believe in canned benchmarks. They have done detailed studies on why they aren't reliable and even posted an article several years ago showing their reasons. Ever since I've been around (and likely long before that) the stance has been "if you wanted canned benchmarks there are plenty of other sites that do it". What you are asking [H] to do by adding canned benchmarks is to betray the very thing they believe in when comes to reviews. You are trying to play the victim card, yet you refuse to accept any answer given to you and just keep repeating the same crap over and over.
Click to expand...

Repeating requires saying the exact same thing, which I have not. I only went so far as to try and explain my point better to those that have apparent issues understanding. But obviously here again I get to do the same.

Adding Canned benchmarks for the games they test when available adds value in the fact that every other site uses different equipment thus not really reliable for comparing results and making sense of discrepancies. For instance running a 10 min session gives ample info, albeit not conclusive but reasonably so. However a Canned benchmark might highlight issues with either test. I get that Canned benchmarks don't generally correlate to Gameplay, but they do usually stress the system enough to get an idea of how the game might run with that particular setup. Tweaking is always necessary, as we generally always have our own preferences.

But again to compare any given site to reviews here will be difficult with any degree of certainty being they use a 3770K and others use considerably newer hardware. In general terms it wont make huge differences but it doesn't make conclusive ones either.

Adding the occasional Canned benchmark to existing benchmarks given the existing gaming suite would not add much time if any (not sure how many of the games in [H] test suite have Canned benchmarks) and therefore only adds to information without taking away any relevant impact of the review as usually wrote up.

As far as victim, well you are gonna need to provide some proof on that one buddy. I have seen both sides of the argument and generally have given both in each post of mine, hence positives and negatives. I know it is just easier to make such claims than to actually debate with even the basest of opinions or relevant facts, but it doesn't make your posts or any others posts more valid or mine less so.

But hey, be it beyond me to try and just ask questions or give an opinion on a forum. Exchanging ideas here must be taboo.
 
durquavian said:
Repeating requires saying the exact same thing, which I have not. I only went so far as to try and explain my point better to those that have apparent issues understanding. But obviously here again I get to do the same.

Adding Canned benchmarks for the games they test when available adds value in the fact that every other site uses different equipment thus not really reliable for comparing results and making sense of discrepancies. For instance running a 10 min session gives ample info, albeit not conclusive but reasonably so. However a Canned benchmark might highlight issues with either test. I get that Canned benchmarks don't generally correlate to Gameplay, but they do usually stress the system enough to get an idea of how the game might run with that particular setup. Tweaking is always necessary, as we generally always have our own preferences.

But again to compare any given site to reviews here will be difficult with any degree of certainty being they use a 3770K and others use considerably newer hardware. In general terms it wont make huge differences but it doesn't make conclusive ones either.

Adding the occasional Canned benchmark to existing benchmarks given the existing gaming suite would not add much time if any (not sure how many of the games in [H] test suite have Canned benchmarks) and therefore only adds to information without taking away any relevant impact of the review as usually wrote up.

As far as victim, well you are gonna need to provide some proof on that one buddy. I have seen both sides of the argument and generally have given both in each post of mine, hence positives and negatives. I know it is just easier to make such claims than to actually debate with even the basest of opinions or relevant facts, but it doesn't make your posts or any others posts more valid or mine less so.

But hey, be it beyond me to try and just ask questions or give an opinion on a forum. Exchanging ideas here must be taboo.
Click to expand...

We are not adding canned benchmarks. Question and thoughts asked and stated and answered....again.

I have grown tired of your veiled repeated trolling, oh I am sorry, I meant your "exchanging" the same "ideas" over and over again.
 
madgun said:
Dude all I can deduce is that you are trying to defend AMD as if it's sacred. The review gave us a perspective, why do we want to fight it out with the reviewer in order to get things done in a way we see fit.

You are going far beyond the line of simply suggesting something. What I can deduce is that you are trying to force feed your ideas onto the reviewer which is an unacceptable behavior.
Click to expand...

And where did I say they MUST do anything? Not once did I. Originally I asked why they didn't used Canned benchmarks in conjunction with the gameplay reviews. Next thing I know I am getting blasted with labels and disrespectful posts from posters that didn't take the time to read the posts fully and understand what I was ASKING not REQUIRING. I made valid points with more than enough ample information for others to reasonably understand what I was inquiring. If you take a moment and see it from a scientific perspective you see what it is I was wondering and why I thought it might be a good idea. Seriously still haven't seen one good counter point. All I have seen is that it was discussed some decade ago, which then brings the point of relevancy over time, or the sidestepping remarks to just avoid the topic altogether.
 
durquavian said:
If I were trolling I wouldn't be responding to posts. The only reason I am still posting is in defense of my original posts. Seems no one is taking the time to be decent and respectful at all.
Click to expand...


You need a mandatory break. Take a few days off.
 
You must log in or register to reply here.
Back
Top