Witcher III PC requirements are here.

  • 74 results
  • 1
  • 2

This topic is locked from further discussion.

Avatar image for 04dcarraher
04dcarraher

23829

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#51  Edited By 04dcarraher
Member since 2004 • 23829 Posts

@Truth_Hurts_U said:

This is a test I did back when I had a 970.

MLL with everything max and PhysX = 51 FPS ave.

This guy got 47 FPS with PhysX off.

http://youtu.be/uGQ8fkb8Uzo?t=5m14s

Think that is concrete proof? , We have no idea what cpu he is running and at what clockrates, so many unknowns play a role in the results. Not only those facts but how do you explain the fact that that most his tests go against most of the benchmarks done from other sites..... Metro LL for example 970's maxed tend to get 48-51 fps average at 1440p vs his 29, or 1080 68 vs his 48.

Also 970's tend to average in unigine valley at 1080p around 58 fps and minimums are around 30 fps , vs his 52 average and 19 low.

Something is very off with his results.

Avatar image for Truth_Hurts_U
Truth_Hurts_U

9703

Forum Posts

0

Wiki Points

0

Followers

Reviews: 8

User Lists: 0

#52  Edited By Truth_Hurts_U
Member since 2006 • 9703 Posts

@04dcarraher:

You think you saying well a PhysX card doesn't do anything... is proof... I will take my proof every time. I ran tests I seen the difference first hand.

Here is SLI 970 Metro Last Light maxed out with PhysX...

https://www.youtube.com/watch?v=aYiA5dHipDI

Only 68 FPS vs my single card 51 with dedicated PhysX. Guy even has quad channel mem and X99. 34% better performance which is pathetic for the cost.

Even he says that with 1 card he gets 43 FPS. Gives me an 8 FPS edge and I'm running a lesser system.

Avatar image for 04dcarraher
04dcarraher

23829

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#53  Edited By 04dcarraher
Member since 2004 • 23829 Posts

@Truth_Hurts_U said:

You think you saying well a PhysX card doesn't do anything... is proof... I will take my proof every time. I ran tests I seen the difference first hand.

Here is SLI 970 Metro Last Light maxed out with PhysX...

Only 68 FPS vs my single card 51 with dedicated PhysX. Guy even has quad channel mem and X99. 34% better performance which is pathetic for the cost.

Even he says that with 1 card he gets 43 FPS. Gives me an 8 FPS edge and I'm running a lesser system.

No, your not understanding what im saying, With high end gpu's having a dedicated gpu for physx can hurt your minimum performance and only adding slight gains. Not saying it does not help just proving the point that it does not do that much. Also all the proof you posted means nothing because you have no common ground, it does not matter if someone has DDR4 or DDR3 though.

Here are some benchmarks showing the performance of metro with a 970 , and Metro LL with a single 680 with and without physx. That one tek guy youtube review is not the average. Here is a GTX 690 with multiple types of gpu's for physx as for minimums and averages it does little...

Metro LL is known to have issues with physx with certain cards.

Now back what I was saying, Witcher 3 does not look like its not using full blown advanced physx just the hair/fur, so having a dedicated physx card wont do anything.

Avatar image for DefconRave
DefconRave

806

Forum Posts

0

Wiki Points

0

Followers

Reviews: 40

User Lists: 0

#54 DefconRave
Member since 2013 • 806 Posts

I need an excuse to upgrade my 660, this game provides a good one. Will hold off for performance reviews 1st though before deciding.

Avatar image for Truth_Hurts_U
Truth_Hurts_U

9703

Forum Posts

0

Wiki Points

0

Followers

Reviews: 8

User Lists: 0

#55 Truth_Hurts_U
Member since 2006 • 9703 Posts

@04dcarraher:

95 FPS to about 115 FPS with card = 20 FPS. Just like I said... 20 FPS is not a little gain.

Then you post a bunch of pointless screens of Metro Last Light that are not even running the game maxed. lol

Avatar image for deactivated-583e460ca986b
deactivated-583e460ca986b

7240

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#56  Edited By deactivated-583e460ca986b
Member since 2004 • 7240 Posts

@Truth_Hurts_U: @04dcarraher: Here is a benchmark I ran in Arkham Origins with 2 Titan Blacks in SLI and a GTX 690 as a Physx card. Top image is with the 690 and bottom is without. 17fps average difference with a 31fps max difference. So there is a benfit with new cards and new games using a dedicated physx card.

Avatar image for 04dcarraher
04dcarraher

23829

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#57  Edited By 04dcarraher
Member since 2004 • 23829 Posts

@GoldenElementXL said:

Here is a benchmark I ran in Arkham Origins with 2 Titan Blacks in SLI and a GTX 690 as a Physx card. Top image is with the 690 and bottom is without. 17fps average difference with a 31fps max difference. So there is a benfit with new cards and new games using a dedicated physx card.

You prove my point, even with SLI titans + a GTX 690 gains you an average of 17 fps.

Other tests done with Batman Origins show titan +titan(physx) gets around 92 fps

Also shows that the game does not make use the 2nd card fully since titan + 650 vs sli titans was a whole 4-6 fps difference.

Metro LL SLI titans + gtx 650 gets 59 fps while SLI 51 fps

Mafia 2 SLI titans + 650 average 94 fps while a single titan with gtx 650 gets 92 fps avg which tells us the game does not make use the 2nd titan at all. And actually does not use the main gpu fully.

Dedicated physx cards do help but not enough to make it worthwhile and at times can actually lower your minimum performance. With Witcher 3 its looks like its not even using the advanced hardware physx but only using the Nvidia hair works, which AMD gpus could use it if Nvidia allows it. CD project has it ready to go if Nvidia lets them

Avatar image for deactivated-583e460ca986b
deactivated-583e460ca986b

7240

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#58 deactivated-583e460ca986b
Member since 2004 • 7240 Posts

@04dcarraher said:

@GoldenElementXL said:

Here is a benchmark I ran in Arkham Origins with 2 Titan Blacks in SLI and a GTX 690 as a Physx card. Top image is with the 690 and bottom is without. 17fps average difference with a 31fps max difference. So there is a benfit with new cards and new games using a dedicated physx card.

You prove my point, even with SLI titans + a GTX 690 gains you an average 17 fps.

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

Avatar image for 04dcarraher
04dcarraher

23829

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#59  Edited By 04dcarraher
Member since 2004 • 23829 Posts

@GoldenElementXL said:

@04dcarraher said:

@GoldenElementXL said:

Here is a benchmark I ran in Arkham Origins with 2 Titan Blacks in SLI and a GTX 690 as a Physx card. Top image is with the 690 and bottom is without. 17fps average difference with a 31fps max difference. So there is a benfit with new cards and new games using a dedicated physx card.

You prove my point, even with SLI titans + a GTX 690 gains you an average 17 fps.

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

guess you didnt read the rest of my post but

Other tests done with Batman Origins show titan +titan(physx) gets around 92 fps

Origins does not make use the 2nd card fully since titan + 650 vs sli titans was a whole 4-6 fps difference with the test.

Metro LL SLI titans + gtx 650 gets 59 fps while SLI 51 fps

Mafia 2 SLI titans + 650 average 94 fps while a single titan with gtx 650 gets 92 fps avg which tells us the game does not make use the 2nd titan at all. And actually does not use the main gpu fully.

I never said dedicated Physx cards dont give you gains just pointing out its not as large as some seem to think. average gains with mid ranged gpus as physx is usually under 10 fps. With your example 94 vs 111 is what a 16% gain and that is with a gtx 690.

Avatar image for Truth_Hurts_U
Truth_Hurts_U

9703

Forum Posts

0

Wiki Points

0

Followers

Reviews: 8

User Lists: 0

#60 Truth_Hurts_U
Member since 2006 • 9703 Posts

@GoldenElementXL:

Yep, those gains are big. Thanks for helping me out here.

I feel a diff in games that use PhysX... It feels a lot smoother and when the crap hits the fan there is a lot less slow down using a dedicated card.

Avatar image for alim298
alim298

2747

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#61  Edited By alim298
Member since 2012 • 2747 Posts

@FelipeInside said:

Avatar image for deactivated-579f651eab962
deactivated-579f651eab962

5404

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#62 deactivated-579f651eab962
Member since 2003 • 5404 Posts

@GoldenElementXL: Lol, love the idea of a 690 as a PhysX card. Hehe

Avatar image for mastershake575
mastershake575

8574

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#63  Edited By mastershake575
Member since 2007 • 8574 Posts

Like majority of requirements last year, they will be inflated big time and won't be close to the real requirements (an i5 thats 4-5x faster than the consoles CPU being listed as the minimum is a dead give away just by reading the first line)

Avatar image for deactivated-59d151f079814
deactivated-59d151f079814

47239

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#64 deactivated-59d151f079814
Member since 2003 • 47239 Posts

@GoldenElementXL said:

@04dcarraher said:

@GoldenElementXL said:

Here is a benchmark I ran in Arkham Origins with 2 Titan Blacks in SLI and a GTX 690 as a Physx card. Top image is with the 690 and bottom is without. 17fps average difference with a 31fps max difference. So there is a benfit with new cards and new games using a dedicated physx card.

You prove my point, even with SLI titans + a GTX 690 gains you an average 17 fps.

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

.. That is minimal when your PUSHING IN THE GAME 90+ fps.. Furthermore who the **** would do this to begin with?! Your SLI Two titans ON TOP of a 690? Your basically wasting a ton of energy, a ton of space, and generating a ton of heat for a feature that very few games even use.. And furthermore his point went completely over your head.. So let me speak plainly... THESE GAMES SUCK ASS with SLI with the ones your using! As he pointed out that SLI actually fell behind a single card + physx card that is much weaker, in no way should that ever happen inless the game sucks real bad for SLI.

Avatar image for pimphand_gamer
PimpHand_Gamer

3048

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 5

#65 PimpHand_Gamer
Member since 2014 • 3048 Posts

lol, wow even my previous PC meets the recommendation.

Avatar image for MK-Professor
MK-Professor

4214

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#66 MK-Professor
Member since 2009 • 4214 Posts

It is funny when you see really old pc's far exceeding Recommended System Requirements on current games.

  • my cpu 5 years old
  • my mobo 7 years old
  • my gpu 3 years old

and yet console gamers keep say you need to upgrade every year just to play the recent games....

Avatar image for deactivated-583e460ca986b
deactivated-583e460ca986b

7240

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#67 deactivated-583e460ca986b
Member since 2004 • 7240 Posts

@04dcarraher said:

@GoldenElementXL said:

@04dcarraher said:

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

guess you didnt read the rest of my post but

Other tests done with Batman Origins show titan +titan(physx) gets around 92 fps

Origins does not make use the 2nd card fully since titan + 650 vs sli titans was a whole 4-6 fps difference with the test.

Metro LL SLI titans + gtx 650 gets 59 fps while SLI 51 fps

Mafia 2 SLI titans + 650 average 94 fps while a single titan with gtx 650 gets 92 fps avg which tells us the game does not make use the 2nd titan at all. And actually does not use the main gpu fully.

I never said dedicated Physx cards dont give you gains just pointing out its not as large as some seem to think. average gains with mid ranged gpus as physx is usually under 10 fps. With your example 94 vs 111 is what a 16% gain and that is with a gtx 690.

Well a 690 being used a a physx card is actually a 2GB 680 since only one of the GPU's in the dual GPU card can be used for physx.

Second I would have to see what CPU and motherboard are being used in your examples. Also drivers play a big role.

@sSubZerOo said:

@GoldenElementXL said:

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

.. That is minimal when your PUSHING IN THE GAME 90+ fps.. Furthermore who the **** would do this to begin with?! Your SLI Two titans ON TOP of a 690? Your basically wasting a ton of energy, a ton of space, and generating a ton of heat for a feature that very few games even use.. And furthermore his point went completely over your head.. So let me speak plainly... THESE GAMES SUCK ASS with SLI with the ones your using! As he pointed out that SLI actually fell behind a single card + physx card that is much weaker, in no way should that ever happen inless the game sucks real bad for SLI.

I don't think you understand the point being made. The above poster is illustrating how small the gain with a dedicated physx card in addition to sli cards. You are saying sli sucks which is not what he is saying.

Maybe you should try reading plainly instead of trying to speak misinformation plainly.

Also I have HUGE gains with sli. I have no idea what the hell you are talking about.

Avatar image for 04dcarraher
04dcarraher

23829

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#68  Edited By 04dcarraher
Member since 2004 • 23829 Posts

@GoldenElementXL said:

Well a 690 being used a a physx card is actually a 2GB 680 since only one of the GPU's in the dual GPU card can be used for physx.

Second I would have to see what CPU and motherboard are being used in your examples. Also drivers play a big role.

I don't think you understand the point being made. The above poster is illustrating how small the gain with a dedicated physx card in addition to sli cards. You are saying sli sucks which is not what he is saying.

Maybe you should try reading plainly instead of trying to speak misinformation plainly.

Also I have HUGE gains with sli. I have no idea what the hell you are talking about.

I know this, fact is a 680 as a physx card only gains 17 average is not a large amount for the gpu power available . That titan setup was using i7 4770k @ 4.2 ghz with Z87 board.

Another prime example at 1080p is a single gtx 780ti with no ppu with Metro LL its gets 67 fps vs sli 680s with gtx 680 ppu. Now once you get into 1600p then it shifts some, 44 vs 49 fps in favor of the 680's.

The point im trying to make is that with a high end gpu at reasonable resolutions can handle physx and adding a physx card the gains can be small. The Witcher 3 as far as I can tell is not using the typical full blown hardware physx based engine since the devs are waiting for Nvidia for the go ahead to allow AMD gpu to use the hairworks feature. So the point I was trying to make to the other guy is that having a dedicated ppu for this game wont do anything.

Avatar image for RyviusARC
RyviusARC

5708

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#69 RyviusARC
Member since 2011 • 5708 Posts

My 2 GTX 970s will do fine rendering this game at ultra 2560x1440 settings.

Avatar image for red_chili
Red_Chili

94

Forum Posts

0

Wiki Points

0

Followers

Reviews: 67

User Lists: 1

#70 Red_Chili
Member since 2014 • 94 Posts
  • FX-6300
  • HD 7770
  • :(

I guess, i need a new Graphic Card and a new CPU, maybe my CPU is good enough to play this Game in Middle-Settings.

Avatar image for alucrd2009
Alucrd2009

787

Forum Posts

0

Wiki Points

0

Followers

Reviews: 7

User Lists: 0

#71 Alucrd2009
Member since 2007 • 787 Posts

I love my 8350 !

Avatar image for deactivated-59d151f079814
deactivated-59d151f079814

47239

Forum Posts

0

Wiki Points

0

Followers

Reviews: 2

User Lists: 0

#72  Edited By deactivated-59d151f079814
Member since 2003 • 47239 Posts

@GoldenElementXL said:

@04dcarraher said:

@GoldenElementXL said:

@04dcarraher said:

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

guess you didnt read the rest of my post but

Other tests done with Batman Origins show titan +titan(physx) gets around 92 fps

Origins does not make use the 2nd card fully since titan + 650 vs sli titans was a whole 4-6 fps difference with the test.

Metro LL SLI titans + gtx 650 gets 59 fps while SLI 51 fps

Mafia 2 SLI titans + 650 average 94 fps while a single titan with gtx 650 gets 92 fps avg which tells us the game does not make use the 2nd titan at all. And actually does not use the main gpu fully.

I never said dedicated Physx cards dont give you gains just pointing out its not as large as some seem to think. average gains with mid ranged gpus as physx is usually under 10 fps. With your example 94 vs 111 is what a 16% gain and that is with a gtx 690.

Well a 690 being used a a physx card is actually a 2GB 680 since only one of the GPU's in the dual GPU card can be used for physx.

Second I would have to see what CPU and motherboard are being used in your examples. Also drivers play a big role.

@sSubZerOo said:

@GoldenElementXL said:

But it doesn't hurt the performance. Isn't that what you claimed? A 17 fps average is the difference between generations of GPU's most of the time. If it was 2 or 3 frames then maybe I could get behind what you are saying. But a 17 fps average is more than minimal. That's like saying 30 fps is a slight gain over 13 fps.

.. That is minimal when your PUSHING IN THE GAME 90+ fps.. Furthermore who the **** would do this to begin with?! Your SLI Two titans ON TOP of a 690? Your basically wasting a ton of energy, a ton of space, and generating a ton of heat for a feature that very few games even use.. And furthermore his point went completely over your head.. So let me speak plainly... THESE GAMES SUCK ASS with SLI with the ones your using! As he pointed out that SLI actually fell behind a single card + physx card that is much weaker, in no way should that ever happen inless the game sucks real bad for SLI.

I don't think you understand the point being made. The above poster is illustrating how small the gain with a dedicated physx card in addition to sli cards. You are saying sli sucks which is not what he is saying.

Maybe you should try reading plainly instead of trying to speak misinformation plainly.

Also I have HUGE gains with sli. I have no idea what the hell you are talking about.

.........Please go back and read what he said.. a SINGLE Titan + a low end video card for Physx actually beat SLI Titans.. This isn't evidence to why you should have a dedicated physx card.. This is evidence how bad these games are with SLI.. In no way should a single Titan + a 650 for phyx outdo TWO Titans sli'd.. Yet in those benchmarks he commented on they did, or the SLI was a mere handful of fps ahead.. That is not a argument for why you should have a dedicated phyx card.. That is a argument that these games absolutely blow for SLI.

Avatar image for skull886
Skull886

25

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#73 Skull886
Member since 2015 • 25 Posts

Can anyone tell me If I will be able to run this with a Gtx 660 2gb? I know it won't run ultra.. I would like to run medium or high settings.

Windows 8.1

Intel core 4790 3.6 ghz

16gb ram

Avatar image for deactivated-583e460ca986b
deactivated-583e460ca986b

7240

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#74 deactivated-583e460ca986b
Member since 2004 • 7240 Posts

@sSubZerOo said:

@GoldenElementXL said:


I don't think you understand the point being made. The above poster is illustrating how small the gain with a dedicated physx card in addition to sli cards. You are saying sli sucks which is not what he is saying.

Maybe you should try reading plainly instead of trying to speak misinformation plainly.

Also I have HUGE gains with sli. I have no idea what the hell you are talking about.

.........Please go back and read what he said.. a SINGLE Titan + a low end video card for Physx actually beat SLI Titans.. This isn't evidence to why you should have a dedicated physx card.. This is evidence how bad these games are with SLI.. In no way should a single Titan + a 650 for phyx outdo TWO Titans sli'd.. Yet in those benchmarks he commented on they did, or the SLI was a mere handful of fps ahead.. That is not a argument for why you should have a dedicated phyx card.. That is a argument that these games absolutely blow for SLI.

Here is what he said.

"With average frame rates a dedicated gpu for physx is minor, and in some cases having a dedicated physx card with high end card/s can hurt performance... Batman Origins for example with SLI Titans and Physx

SLI:48 min fps, avg 78 fps

SLI+GTX 650: 39 min fps, avg 66 fps

Titan + Titan : 61 min fps,avg 94 fps

Titan +GTX 650: 39 min fps, avg 62 fps

Titan: 38 min fps, avg 66 fps."

I'm not sure where you are getting your information from, but it's not in this thread.

And lets not discount MY benchmark I ran with the game maxed at 1440P. 69fps min with a 94fps avg. That doesn't seem like poor SLI performance to me.