Far Cry 4 Video Card Showdown

[Colonel_Gerdauf]

-snip-

 

You still missed the info here. That new AMD driver is in beta form, and LMG refuses to use beta drivers, as they are unstable compared to WHQL. They have made this especially clear, and they are not making any exceptions, even if one particular beta driver works very well.

 

And the point of benchmarking is to get consistent results, and beta drivers do not cut it, because it defeats the purpose of benchmarking. It may get good results on your end, but it is still a beta driver, which, by definition of the term 'beta', means that it is an unfinished firmware.

If you want to call somebody out, do not insult them. It leaves a negative image of you, and people will ignore you as a result. And one rule of the internet is this; once you earn a bad reputation, that reputation sticks with you for a particularly long time.

[deathjester]

You still missed the info here. That new AMD driver is in beta form, and LMG refuses to use beta drivers, as they are unstable compared to WHQL. They have made this especially clear, and they are not making any exceptions, even if one particular beta driver works very well.

 

And the point of benchmarking is to get consistent results, and beta drivers do not cut it, because it defeats the purpose of benchmarking. It may get good results on your end, but it is still a beta driver, which, by definition of the term 'beta', means that it is an unfinished firmware.

If you want to call somebody out, do not insult them. It leaves a negative image of you, and people will ignore you as a result. And one rule of the internet is this; once you earn a bad reputation, that reputation sticks with you for a particularly long time.

 

Beta drivers post known issues if there is instability. There are no known issues on that driver. There was no problems on that driver with Far Cry 4. The game released a day early online and AMD users had the new driver before Nvidia users.

 

It is the same damn driver as the past driver with an update for Far Cry 4. As I said even on RECENT WHQL drivers a game ready driver has broken one game to fix another. You are comparing apples with apples with the newest WHQL and the AMD beta driver.

 

This is a complete cop out excuse and there is no one playing the game on an AMD card without the beta driver with MSAA. This is like telling people to intentionally cripple their game for no reason. You may as well tell people to wait a few weeks to use any WHQL drivers, because they have issues ALL THE TIME. Nvidia users constantly have to switch drivers between games. I did constantly. SWTOR would be one example when it released. AMD is saying the driver is in development. Nvidia is claiming the driver works perfectly for everything (which they rarely do). It is semantics.

 

In addition why were the tests done on MSAA which skewed the tests as much as possible? MSAA is simply not playable without stutter on any card with any cpu in existence. Slick used a scripted autodrive sequence in the game that has none of the problems as driving a vehicle yourself. The game chokes on CPU on one core. It chokes less with SMAA and STILL stutters. No one is playing this with high MSAA/TXAA unless they like stutter like crazy while driving or disabling mip mapping in the XML which defeats the whole purpose of Ultra settings and lowers textures and gives pop in.

 

You literally have 2 choices to get a enjoyable play experience on this game at high settings. Play with a XML edit which isn't Ultra to begin with. Lock the game at 30 FPS and push MSAA/TXAA downsampling as high as possible. Nvidia provides no bennefit other than TXAA which is debatable as a form of AA. Many people hate it and that includes people like Total Biscuit. I also hated it on my GTX 770. Too much blur. 

 

This benchmark was a sham performance wise and a sham as to what settings  people can actually play. It also failed to point out that performance problems were not related to AMD GPU's in any way and has everything to do with piss poor optimization on core usage for which Slick doesn't even provide us a benchmark for.

 

This was not the only BS benchmark we saw with Far Cry 4. Gamer Nexus did the same thing and that benchmark was posted in the tech news forum and stayed their for a week (while not being related to tech news whatsoever). Slick failed to even mention that a beta driver was out for AMD that would give much better performance. He then misinformed people saying maybe a driver would be out soon.

[Colonel_Gerdauf]

-shitpost-

Not only did you not address my points at all, you have been using the same, copy-paste argument for the third time now, where it was already debunked the first time. So your points have been discarded as irrelevant.

If you want to bring an argument, you need to provide some data and citations. You have provided nothing up to now. You have been treating you own words as absolute fact nonetheless, and discarding all of counter-arguments instead of addressing them.

And you should not insult others to get you point across. Doing that is counter-productive and self-defeating.

[deathjester]

Not only did you not address my points at all, you have been using the same, copy-paste argument for the third time now, where it was already debunked the first time. So your points have been discarded as irrelevant. If you want to bring an argument, you need to provide some data and citations. You have provided nothing up to now. You have been treating you own words as absolute fact nonetheless, and discarding all of counter-arguments instead of addressing them. And you should not insult others to get you point across. Doing that is counter-productive and self-defeating.

 

You  have made no points. AMD Beta driver = to rushed WHQL that is put out every week, that has not undergone extensive beta testing and can have the same issues as a AMD "beta driver". See WHQL Dragon's Age driver. When did they test that? Who were the testers? It was an AMD title. How about Tomb Raider? Another AMD title. Broken on Nvidia cards for a LONG time. I know, because I had a GTX 770. That wasn't even a closed library like Nvidia Game Works that AMD has to optimize. So are you saying people should have avoided new WHQL's when they finally fixed their performance in that game?

 

http://www.gamefaqs.com/boards/631230-dragon-age-inquisition/70631062

 

They can call the Dragon's Age Driver WHQL but it was BS. It hadn't been extensively tested. People are reporting problems all over the Bioware support forum. Again you are arguing semantics.

 

Are you claiming that the AMD beta driver broke something that would have effected this benchmark? List it. WHQL drivers also have issues on each driver. They list them in the PDF manual. There were no issues with the AMD driver in Far Cry 4. It was just a major upgrade in performance in Far Cry 4 in MSAA. That is why when you click update driver in Catalyst?

 

It links you a beta driver an an upgrade. This is what every AMD GPU user is running. That is the driver we had before the game even released. The first thing people do when a new game comes out is click update driver or their vendor specific program does it for them. They do the same thing with an untested WHQL driver. 

 

As far as self defeating? People called out this BS on day 1. I am late to the party. I was shocked when I recently watched the video because I had this game day 1 and my performance looked more like the GTX 980 on my Tri-x.  What shocked me more was the misinformation in the video that followed the benchmarks. When "white knights" argue for you (pitifully I might add) and those white knights have a friends list like yours whose post history is laughable as far as supporting certain companies and or products? AND you can't defend your own actions?

 

This site and Slick loses all credibility. You? Never had any. Neither has most of the people in your friends list. I LAUGHED when I clicked it. I was not disappointed. Same people I have been calling out for BS on this forum for months. What a surprise. 3/4 oi your friends list is like getting a MS post direct from Redmond. 

[Victorious Secret]

This site and Slick loses all credibility. You? Never had any. Neither has most of the people in your friends list. I LAUGHED when I clicked it. I was not disappointed. Same people I have been calling out for BS on this forum for months. What a surprise. 3/4 oi your friends list is like getting a MS post direct from Redmond. 

 

This post sums up everything wrong with your juvenile mind. If this site is so lacking in credibility, leave and let us stew in it. Why are you still here? Your kind of toxicity does no community any good and I think a large number of people have little issue with what I've just said. 

 

[deathjester]

This post sums up everything wrong with your juvenile mind. If this site is so lacking in credibility, leave and let us stew in it. Why are you still here? Your kind of toxicity does no community any good and I think a large number of people have little issue with what I've just said. 

 

 

No googling "Microsoft Astroturfing" sums up everything wrong with illegal advertising on forums. 

 

Having a brain and seeing who is a big sponsor to a site and whom may make donations, lets you know why some sites are strict with people like you and other sites are not. You need transparency for that though. 

 

But whatever. I know who you are, you know who you are, and there is no difference between you and the "cyber terrorists" who just attacked Xbox Live. You advertise illegally and terrorize innocent victims on a forum to promote, defend a brand. You can call it "reputation management" but it is still illegal without stating who you are. You are a criminal sir. Might be a "white crime" but it is still illegal. 

 

Not surprised to see you kissing Slick's butt here, because you would have been banned off any tech forum or gaming forum that doesn't put up with this crap like Neogaf... LTT seems to embrace their astroturfers and move threads calling them out or shut down a thread when proof is offered. 

 

The post history of you and Goodbytes is ALMOST as laughable as this benchmark. 

 

But anyways. Keep going. Tell people to look at your post history so they can have a laugh. Include the posts you made asking for silent bans on people etc. To be honest? I stopped wanting to come here after you and Goodbytes and Rhood were allowed to continue harassing the crap out of people who critique MS. Only reason I still do is to help people with problems in gaming etc.

 

This site has lost all credibility with me. Slick has lost all credibility with me.  You? Never had any. I looked at your post history and knew who you were immediately. To me you are just a petty criminal that probably makes the same amount of money as someone flipping burgers at McDonald's. At least those people have a sense of honor and do not resort to crime and harassment.

[Victorious Secret]

No googling "Microsoft Astroturfing" sums up everything wrong with illegal advertising on forums. 

 

Having a brain and seeing who is a big sponsor to a site and whom may make donations, lets you know why some sites are strict with people like you and other sites are not. You need transparency for that though. 

 

But whatever. I know who you are, you know who you are, and there is no difference between you and the "cyber terrorists" who just attacked Xbox Live. You advertise illegally and terrorize innocent victims on a forum to promote, defend a brand. You can call it "reputation management" but it is still illegal without stating who you are. You are a criminal sir. Might be a "white crime" but it is still illegal. 

 

Not surprised to see you kissing Slick's butt here, because you would have been banned off any tech forum or gaming forum that doesn't put up with this crap like Neogaf... LTT seems to embrace their astroturfers and move threads calling them out or shut down a thread when proof is offered. 

 

The post history of you and Goodbytes is ALMOST as laughable as this benchmark. 

 

But anyways. Keep going. Tell people to look at your post history so they can have a laugh. Include the posts you made asking for silent bans on people etc. To be honest? I stopped wanting to come here after you and Goodbytes and Rhood were allowed to continue harassing the crap out of people who critique MS. Only reason I still do is to help people with problems in gaming etc.

 

This site has lost all credibility with me. Slick has lost all credibility with me.  You? Never had any. I looked at your post history and knew who you were immediately. To me you are just a petty criminal that probably makes the same amount of money as someone flipping burgers at McDonald's. At least those people have a sense of honor and do not resort to crime and harassment.

 

 

You're assuming that I even work for MS, a subsidiary of MS, or otherwise was contracted by MS; none of which apply to me at all. I actually worked for Samsung, which I made the mods aware of when I joined this place. But nice try on trying to insult me or whatever it is you were trying to accomplish in that rant of yours. 

Thank you for showing everyone on this site just how off the rockers you are. I pity you, I truly do. 

[Colonel_Gerdauf]

-shitpost-

-shitpost-

 

You do realize that you are still being condescending for no reason, right? Some people have views different from your own; that does not make them astroturfers or sellouts.

 

Moreover, you can bring your facts all you want, but you are not going to change Slick's mind. It is a policy; WHQL or nothing, simple as that. You do not like it? Then do your own benchmarks. Instead of complaining all the time, you should offer your own solution and leave it at that.

[deathjester]

 

You do realize that you are still being condescending for no reason, right? Some people have views different from your own; that does not make them astroturfers or sellouts.

 

Moreover, you can bring your facts all you want, but you are not going to change Slick's mind. It is a policy, WHQL or nothing; simple as that. You do not like it? Then do your own benchmarks. Instead of complaining all the time, you should offer your own solution and leave it at that.

 

Every WHQL has issues listed in the PDF manual just like an AMD Beta. Slick is wrong. Slick also tested MSAA to sway the test heavily towards Nvidia without the driver. The only cards that could even run 2x MSAA anywhere close to playable (without a 30 FPS lock, in which case you could push 4xMSAA on some of these cards) are a heavily overclocked GTX 970, a R9 290/x, GTX 980, 780ti, heavily overclocked 780. So the only cards tested should have been those. End of story. As I asked before. Please list which issues were ok in WHQL that made them ok to use and issues from the AMD driver that made them not ok to use from AMD. There are issues with every new WHQL just like every new AMD beta and this AMD beta didn't effect single card setups on any game which is what Slick tested...

 

 

 

You're assuming that I even work for MS, a subsidiary of MS, or otherwise was contracted by MS; none of which apply to me at all. I actually worked for Samsung, which I made the mods aware of when I joined this place. But nice try on trying to insult me or whatever it is you were trying to accomplish in that rant of yours. 

Thank you for showing everyone on this site just how off the rockers you are. I pity you, I truly do. 

 

You work on the internet. Your post history reflects months of defending and promoting MS and attacking anyone critiquing it. Anyone can view it.  Please show me anything similar from you for Samsung which is bashed on this forum all the time. You are literally in every single MS thread and not to be found in a Samsung thread on the tech news front page. In fact the ONLY post you can be found in with Samsung is a post 8 minutes ago on a older thread which you avoided while selling/promoting and attacking anything pro/against MS in the same time frame. In fact you ONLY posted in that thread after the post in this thread. They are time stamped. You do realize this right? 

 

Anyways, as I am sick of conversing with criminals, and know Slick is not going to answer these questions? Have a great day. Every single person with an AMD card knows this benchmark is BS. So if LTT wants to alienate 3x percent of their viewers? I hope Nvidia spends a lot of money donating or advertising to this site. As I said I was not the first person to critique this benchmark. I just laughed at it and laughed at all the people claiming AMD cards ran the game badly seeing I had GTX 980 type FPS on a 290 dollar card that could probably be bought even cheaper now. I just built a machine with a GTX 970 in it. I posted pictures of it...Far Cry 4 runs quite a bit slower on it compared to my mild overclock on a Tri-x. I also praised the card for Nvidia DSR which can be nice on a SFF where you might want to use a controller for everything and don't want to deal with things like Gedosato.

 

I also laughed at an even bigger BS benchmark from "Gamer Nexus" (offered no proof of a benchmark even being done, and was basically just a sales pitch for Nvidia Game Works features) being on the Tech forum for a week and you not asking it to be removed as "non tech related news", when you have literally reported anything critiquing MS as non tech news and got it moved to a different forum. I repeatedly asked for it to be removed since it had benchmarks in no way close to reality. It stayed there a week.

 

But whatever. This forum has lost all credibility, the benchmarks are a fairy tale. I might as well post at ign.com or PCgamer. They are just as sold out as LTT. There are sites where AMD BS is almost as bad. I have called them out in the past (Tek Syndicate would be an example of that). 

 

http://www.pcgamer.com/far-cry-4-review-in-progress/

 

I see no difference from the lies there and the misinformation in Slick's benchmarks. That site is more up front with it's pro Nvidia articles though. At least they put a reference Nvidia card in damn near every article and flaunt it.  You can spin this however you want. Bottom line is the performance people have at home on their drivers is not what was in the benchmark and they had that performance before Nvidia users did. Other benchmark sites used the AMD driver. Why? Cus they are not corrupt/incompetent.

 

As this site sells out more? A new site will take it's place until that sites owners are overcome by greed. Just seems to be an endless cycle. This site will probably eventually be sold to a large media group like Tom's Hardware was. Whatever. Pointless to argue with sellouts and astroturfers. Don't think I will bother with the news section anymore. I will just post benchmarks that show LTT benchmarks are a joke when they choose to do this and help people with tech problems. 

 

Oh I will laugh as you MS slappies sell the subscription fee on Win 10, but if people are dumb enough to pay that or think you are random people when you are all on the same friends list? Let them. It is simply too much work arguing with people who do this for a living when people can be manipulated this easy by upvotes, and perceived public perception. On this forum Windows 8 is a massive hit. In reality it is a massive failure. The disconnect from reality on this forum has never been worse and I am ASHAMED to be a part of it.

[Colonel_Gerdauf]

-shitpost-

 

If you think LTT is such a sellout, then what are you even doing here? If you want to leave, then please do so! You would limit the damage to your public image, and we can instead focus our discussion with actual, grown men.

 

You are being a child right now; every single post you are making these days is going against you, and makes you look like a toll. If you want to actually make yourself the class clown (your username fits BTW), then we are not stopping you. Just do not come crying to us when nobody is listening!

[deathjester]

If you think LTT is such a sellout, then what are you even doing here? If you want to leave, then please do so! You would limit the damage to your public image, and we can instead focus our discussion with actual, grown men.

 

You are being a child right now; every single post you are making these days is going against you, and makes you look like a toll. If you want to actually make yourself the class clown (your username fits BTW), then we are not stopping you. Just do not come crying to us when nobody is listening!

 

Helping people with tech problems unlike you and most of the people in your friends list... You know...what this site was originally designed for.

 

You guys harass honest people for a living and advertise illegally. As I said before you are as much a cyber terrorist as the people who brought down Xbox Live. There is literally no difference.  Do not give me a lecture on morality or maturity. You are literally scum and criminals who serve no purpose on this site other than to promote certain things and attack genuine people in tandem. Slick? Also now confirmed scum. IF he made it clear that there was a beta driver (misinformed people in his video) and that he chose not to use it for "reasons" (would have loved to hear them), then all these BS posts you made would have some validity. He did not do any of those things.

 

Why am I still here? From now on? Just to help people. The Tech news forum is a joke. The benchmarks are a joke. Blaming only Ubisoft while ignoring API, and MS as far as "optimization" also a joke. 

 

The only thing worth coming here for now is to help genuine people with genuine problems. You know...like the people with GTX 900 cards they just spend quite a bit of money on, whose Far Cry 4 runs like @!$%^ and AC Unity runs like @$%@  and having to explain to them that it is the game and the game runs badly on anything and they were sold fairy tales from sellouts as to what kind of performance they would actually get. 

 

Want to see some of them? Scroll through the last few weeks. Also check the video card forum, cpu forum. People think their hardware is broke. It is the bench marker that is broke.

 

http://linustechtips.com/main/forum/11-pc-gaming/

 

Anyways have fun with the thread. Already said what I wanted to. Already showed why this benchmark was BS. I literally have nothing more to say. Like I said. Waste of time. I suggest bringing in Rhood and Goodbytes to blast me as well.   I literally don't care. I have helped countless people on this site and got private messages of thanks. From setting up Gedosato, texture packs, Dolphin emulator.  THAT is why I am here. Not to converse with you. Like I said. If idiots want to fall for forum manipulation and benchmark manipulation? Let them. Slick and you have to live with your actions. I don't. All I did was say the emperor has no clothes. He doesn't...

[Faa]

*

Benchmark rules were WHQL drivers only, but I agree if the beta driver would provide up to 50% more performance he should have used that one. And lets read the notes again; up to 50%. That 50% gain needs to be proven first before we're starting to accuse people.

Source#1

290x performs 13% worse than the 780 Ti with the AMD 14.11.2 beta driver. http://www.gamersnexus.net/game-bench/1701-far-cry-4-gpu-benchmark-amd-is-broken-again

In Slick his 1080p test the 290x performs 21% worse which is a 8% gain apparently. Slick can't take the blame AMD deserves.

Source#2

Another benchmark with 14.11.2. Same story.

Source#3 

http://www.sweclockers.com/artikel/19647-snabbtest-grafikprestanda-i-far-cry-4

Drivrutinerna är senast tillgängliga från respektive tillverkare, vilket innebär Geforce 344.75 WHQL för Nvidia samt Catalyst 14.11.2 beta för AMD.

Source#4 Directx Driver overhead again.

Thats a 33% improvement. If there was no driver overhead at all from AMD, the 760K would have had had the exact same result with both cards just like here;

I've proven you with a bunch of evidence from multiple sources confirming that AMD dx drivers are performing garbage which you didn't want to believe. Like I asked before; drop your Star Swarm 720p low settings result with Directx. If you don't want to post it then you'll be automatically wrong again. Posting or not comes to the same point -> AMD dx drivers are shit.

Even in their own Mantle games it's the same exact story we're seeing here, garbage performance without mantle suffering from a shitload of CPU overhead. Lets sum it quickly up;

- Who should be blamed? AMD.

- Who's driver are horrible? AMD.

- Who's lying about their performance improvements? AMD.

- Who's doing nothing about fixing their overhead issues? AMD.

- Who actually used their awful directx driver overhead as a reference to show how great Mantle is? AMD.

It's silly to blame Slick when those drivers you guys mentioned made almost no difference at all

[Victorious Secret]

-snip-

☐ rekt

☐ not rekt

☑Tyrannosaurus rekt

As if he needed to be shot down any more. Well written.

[Slick]

@deathjester Attack me all you want - that will not get you in trouble on this forum. I chose not to defend myself in order to not appear oppressive to the users of this website.

 

"Slick? Also now confirmed scum." <-- Sure it may make me sad but alas, I will allow and defend your right to do so.

 

Attacking other members of this forum however is NOT OKAY - this goes to everyone obviously, not just deathjester... despite using his quote... it was just easy to get because I was linked to the post through a report

"You are literally scum and criminals who serve no purpose on this site other than to promote certain things and attack genuine people in tandem."

 

^ This is not how we address each other on this forum. Honestly I have not had enough time to personally go through this thread. If you don't like something someone has said to you REPORT IT like that post was.

 

---------------------------------------------------

Now to explain some things.

 

1) AMD 14.11.2 beta driver

After FAA's post it may hurt to say this BUT I actually was running this driver. We did have the WHQL driver policy and honestly I still think that is how it should be done BUT this went out the window when it got to the point where I was testing games with AMD drivers that were MONTHS old. We are now, unfortunately, back to running beta drivers which I believe is bad for MANY reasons... but its positives do outweigh its negatives. because of rough driver update cycles that leave major performance increases and potential features off the WHQL releases for long periods of time it will have to work.

 

These rough potentially problem plagued beta drivers from BOTH sides combined with 1st or 2nd day of release game testing results in issues ALL OVER THE PLACE. which is why I state very clearly that we test the games early because we know people want to know how it will perform RIGHT NOW. If you want the best possible result you should test it months down the line when everyone is done playing already and no one cares anymore because the game would probably be finally optimized by then... But as I'm sure you figured out by now that would be useless because "everyone is done playing already" sure you may be wondering how it performs because of a steam sale in 2017 but frankly most of our audience is interested in the now.

 

2) MSAA

We use this because it is mostly universally supported

---------------------------------------------------

We are ALWAYS open to constructive criticism and have changed many things over the course of our history due to public comment / suggestions from the audience feedback will be taken no matter what.

 

We are NEVER open to attacks on ANY our members however - I will be having a number of mods look over this thread and EVERYONE in it. Please remember that when it comes to moderative actions we do not discuss the results publicly

[Fate]

snip

Looks like I was completely wrong then, thanks for posting.

Possibly my results were so different due to a FC4 update and not a driver update? (I was using the beta drivers just like you and got significantly more FPS)

[Slick]

Looks like I was completely wrong then, thanks for posting.

Possibly my results were so different due to a FC4 update and not a driver update? (I was using the beta drivers just like you and got significantly more FPS)

Very possibly - I am already testing it again as we always do in a GPU showdown. This time amongst other games... and I can tell you already that the Nvidia numbers have changed quite a bit, I haven't tested AMD yet

[Victorious Secret]

Very possibly - I am already testing it again as we always do in a GPU showdown. This time amongst other games... and I can tell you already that the Nvidia numbers have changed quite a bit, I haven't tested AMD yet

 

Thanks for the updates on it Slick, I get why the WHQL v beta drivers is such a big issue for testing. Personally, beta drives cause headaches to no end, so I'm pretty happy to sit with WHQL drivers even if they are older. Hell, last time I did a Nvidia beta it wiped every single trace of the driver off my system entirely, I kept thinking I had broke Windows for crying out loud. 

You won't get many complaints from me on the testing methodology. You guys minimize as much as possible for variables so its hard to nitpick that the results aren't 1001003240120340123% perfection, all the time. 

[deathjester]

@deathjester Attack me all you want - that will not get you in trouble on this forum. I chose not to defend myself in order to not appear oppressive to the users of this website.

 

"Slick? Also now confirmed scum." <-- Sure it may make me sad but alas, I will allow and defend your right to do so.

 

Attacking other members of this forum however is NOT OKAY - this goes to everyone obviously, not just deathjester... despite using his quote... it was just easy to get because I was linked to the post through a report

"You are literally scum and criminals who serve no purpose on this site other than to promote certain things and attack genuine people in tandem."

 

^ This is not how we address each other on this forum. Honestly I have not had enough time to personally go through this thread. If you don't like something someone has said to you REPORT IT like that post was.

 

---------------------------------------------------

Now to explain some things.

 

1) AMD 14.11.2 beta driver

After FAA's post it may hurt to say this BUT I actually was running this driver. We did have the WHQL driver policy and honestly I still think that is how it should be done BUT this went out the window when it got to the point where I was testing games with AMD drivers that were MONTHS old. We are now, unfortunately, back to running beta drivers which I believe is bad for MANY reasons... but its positives do outweigh its negatives. because of rough driver update cycles that leave major performance increases and potential features off the WHQL releases for long periods of time it will have to work.

 

These rough potentially problem plagued beta drivers from BOTH sides combined with 1st or 2nd day of release game testing results in issues ALL OVER THE PLACE. which is why I state very clearly that we test the games early because we know people want to know how it will perform RIGHT NOW. If you want the best possible result you should test it months down the line when everyone is done playing already and no one cares anymore because the game would probably be finally optimized by then... But as I'm sure you figured out by now that would be useless because "everyone is done playing already" sure you may be wondering how it performs because of a steam sale in 2017 but frankly most of our audience is interested in the now.

 

2) MSAA

We use this because it is mostly universally supported

---------------------------------------------------

We are ALWAYS open to constructive criticism and have changed many things over the course of our history due to public comment / suggestions from the audience feedback will be taken no matter what.

 

We are NEVER open to attacks on ANY our members however - I will be having a number of mods look over this thread and EVERYONE in it. Please remember that when it comes to moderative actions we do not discuss the results publicly

 

You are saying that your test was on 11.2? Even though you said there was no new AMD driver in the video? That is interesting. Also I must have a magic GPU then.

 

So did all the other bench markers. Your results are literally not close and to be frank impossible.

 

Also as far as Faa. Faa sells Intel/Nvidia on this forum in almost every post. That is literally all he does. His post history speaks for itself. Clicking on his profile will reveal the same benchmarks reposted about 1000 times in the last year. You have a major problem with astroturfing on this site and it is to the point where it is outright laughable. Those benchmarks often include reference R9 290's/x's that throttled. You said as much when you had 290's beating 290x's in the past.  The reference cooling simply sucked. Everyone in the world knows that. 

 

Also you would expect the Nvidia cards to run tree tessellation better.  As you know Slick. Tree Tesselation (tree relief) was literally added to this game at the last second on release day. The pre release that legal customers had the day before did not even contain the settings. On Nvidia's own site under the optimization article they tell you that the setting screws up the shadows on trees (it does). It is literally a sucker setting that no one should be running (and will just cause more stutter than is already there)..It looks worse on than off and was a big hit to AMD cards. It was added LATE for one reason. To put Nvidia ahead in benchmarks (but not close to the distance your benchmarks show).

 

http://gamegpu.ru/action-/-fps-/-tps/far-cry-4-test-gpu.html

 

Tessellation on the trees and off was a big swing in performance but still nothing close to what you benchmarked. The other Game works features don't even work. God rays works kind of. Hairworks a no go. Tessellation was the only thing that could put Nvidia ahead. 

 

It had nothing to do with CPU and Faa's posts are simply hilarious. It had to do with a tessellation setting that literally shouldn't have been in the game in the first place.

 

Please record the same test on all cards with the 11.2 driver. Try it without tree tessellation and with it. Also include shots of tree tessellation on/off. Your benchmarks are literally impossible unless the card was throttling. My Trix X runs in the low 70C with a overclock and Linus has two sapphire vapor x's sitting there you can play with. I suggest testing with one of them and setting the clocks to stock....

 

Your benchmarks aren't matching what people are seeing on their own systems. They aren't even close, even with tessellation on trees that screws up shadows and makes them look distorted...

 

Bottom line, this was a major release, your benchmarks do not match other benchmarks, and are not even close. You tested settings that are not playable on anything, and this will effect sales/stocks of the companies you tested the cards for. I have now seen this game in person run worse and stutter worse on a 780ti and a GTX 970 than my Sapphire Tri-x on a LIGHT overclock of 1100/1450. That was at SMAA. Not MSAA. MSAA is even worse.

[Slick]

Far Cry 4 Release date - Nov 18

http://puu.sh/dg4ot/0fe1cc078d.png

 

Nvidia Driver update - Nov 18

http://puu.sh/dg4mK/c5225bddaf.png

 

AMD Beta Driver update - Nov 16

http://puu.sh/dg4nE/f5edf87765.png

 

In regards of the release of FC4 especially because FC4 was an Nvidia backed title, it wasn't really a new driver

[Slick]

tree relief is in this game

tressFX was in tomb raider

 

I ran them both on and will continue to do so - these videos are showcasing technology like this, if it doesn't run very well on a certain card then so be it, people will know

 

780Ti had(has) massive issues with Tomb Raider, we called that out - this is how these things work

[mr.skyrim_pc]

tree relief is in this game

tressFX was in tomb raider

 

I ran them both on and will continue to do so - these videos are showcasing technology like this, if it doesn't run very well on a certain card then so be it, people will know

 

780Ti had(has) massive issues with Tomb Raider, we called that out - this is how these things work

Yay Slick has a Santa hat!

[Faa]

 

We are ALWAYS open to constructive criticism and have changed many things over the course of our history due to public comment / suggestions from the audience feedback will be taken no matter what.

I actually would love that you guys slightly change your benchmarking. We're only getting FPS as information, no information regarding the GPU usage at all which should be at 99% preferably for a fair comparison. AMD drivers cause significantly more CPU overhead that might hold the GPU usage back affecting performance. I've explained everything about AMD's driver overhead in this thread; http://linustechtips.com/main/topic/249881-amd-releasing-public-mantle-sdk-this-year-encourages-nvidia-and-intel-to-use-it-for-free/

Not really fair if a 780 runs at 99% load and a 290x at 70% completely getting outperformed. If you guys manage to show us a 50% difference and lets say you are suffering from that issue in 1080p only, nobody will take this serious, so why not leave 1080p out and only provide 4K which would most likely lift the CPU bottleneck up? Or you get criticism like Deathjester is dealing out here.

Besides I would love that you guys would do some driver overhead tests, has to be done with the CPU being the bottleneck, AMD seems to be suffering heavily from this with Directx as API. Seen it happening in like 20 games orsomething and newer ones are suffering from it as well. Link I provided has a bunch of them. AMD doesn't seem to be aware of it or at least not enough people are reporting it, if LTT could do some benchmarks to pull AMD's attention that would most likely force AMD to fix their issues. Just like Frame latency tests did.

Atm it's not even clear why AMD's performance is so bad in FC4, if you provided the gpu usage like "290x ran at 80% and the 780ti at 99%" then it would have been clear that we are again dealing with driver overhead. If both ran at 99% then it's just something else. Looking at PClab.pl's fc4 comparison between a 970 & 290x with a 760K we see a 33% improvement which is way too much.

 

Also as far as Faa. Faa sells Intel/Nvidia on this forum in almost every post. That is literally all he does. His post history speaks for itself. Clicking on his profile will reveal the same benchmarks reposted about 1000 times in the last year. You have a major problem with astroturfing on this site and it is to the point where it is outright laughable. Those benchmarks often include reference R9 290's/x's that throttled. You said as much when you had 290's beating 290x's in the past.  The reference cooling simply sucked. Everyone in the world knows that. 

I actually never recommended anyone here a nvidia card yet. I'm not even active in the GPU forums. 290's being throttled is what you made up, never they mentioned they were reference cards, in all of the benchmarks I provided where AMD managed to lift the huge CPU bottleneck with Mantle up they outperformed the 780/Ti. A 290, thats outperforming a 780 with mantle is throttling makes no sense at all. 

 

It had nothing to do with CPU and Faa's posts are simply hilarious.

Let me remind you;
 

 

I've proven you with a bunch of evidence from multiple sources confirming that AMD dx drivers are performing garbage which you didn't want to believe. Like I asked before; drop your Star Swarm 720p low settings result with Directx. If you don't want to post it then you'll be automatically wrong again. Posting or not comes to the same point -> AMD dx drivers are shit.

Get me this information I asked for, doesn't matter so much anymore since you proved yourself wrong. Then tell me it had nothing to do with driver overhead in an AMD benchmark. Yelling here things like Gameworks crippling AMD's performance is what I would call hilarious especially if you're completely on your own and when the same performance gap happens in their OWN Mantle advertised titles.
 

How can I be astroturfing when you're the only one here defending his brand? I'm pointing an issue out thats holding AMD's performance back, never Ive been claiming their GPU's performance is bad/crap. I never said Nvidia cards are faster, simply said they have less driver overhead having an advantage in cpu bound scenario's that doesn't translate to "290x is a shitty performing card". I provided data for both brands suffering from overhead, not only AMD. When I'm discussing who reduces CPU bottlenecks the best between AMD & Nvidia, you are thinking I'm discussing which GPU is faster >.> 

[hi_tech_lowlife]

A chart showing the average fps should be showing an average of *all* of the fps measurements taken throughout a given run.  For example, if you have a 2 minute run you are doing, and your measurement tool samples the fps every second, then you'll have a total of 120 measurements taken during the run.  An "average" should be the average of *all* of those 120 measurements.  For consistency, it's best to do a run about 3 times, then average together those three averages.

 

I might be misunderstanding your question, though...

 

 

What you're describing is showing the average of the average FPS. And yes, that's what a chart labeled 'average FPS' should be showing, what I'm asking is if that's actually what LTT's charts are showing. The reason I ask this is because in the video for the R9 290 unboxing they threw up an FPS chart labeled 'Tomb Raider  - Ultimate, 1080p' (which I assume means running Tomb Raider using the 'Ultimate' quality preset, which includes TressFX) and didn't really say if it was max FPS or average FPS. I'd like to know because in their chart the HD 7950 is listed as getting 70FPS, and my 7950 doesn't reach anywhere near an average of 70FPS, even when my OC matches the one they have on their doc. So I'm wanting to know if there's something wrong with my setup or if that's a list of an average taken from max FPS readings.

[zappian]

Runs fine on my 270x .

 

Much better than most ubi games.