ScardyBobScardyBobJoin Date: 2009-11-25Member: 69528Forum Admins, Forum Moderators, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Shadow
<!--quoteo(post=1961096:date=Aug 11 2012, 07:08 AM:name=Bacillus)--><div class='quotetop'>QUOTE (Bacillus @ Aug 11 2012, 07:08 AM) <a href="index.php?act=findpost&pid=1961096"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I asked this on some other thread, but I don't think it ever got answered:
Has anyone noticed any inconsistencies on how the game functions on different FPSes? NS1 had jetpack tricks, jump timing, rate of fire to name a few. NS2 had some inconsistencies at movement at some point.
Has anyone noticed any similar thingies on the present NS2?<!--QuoteEnd--></div><!--QuoteEEnd--> In theory, there shouldn't be, in practice, such bugs have popped up. I remember there were issues with the rifle ROF that would actually slow down with lower FPS. If you find other issues, report them so they can get fixed.
<!--quoteo(post=1959801:date=Aug 9 2012, 11:56 AM:name=LPC)--><div class='quotetop'>QUOTE (LPC @ Aug 9 2012, 11:56 AM) <a href="index.php?act=findpost&pid=1959801"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->its a marketing trick, get over it. (sorry) but atleast these specs are for machines which might make the game somewhat playable... you should've seen the original pre-order specs... they advertised with a minimum of 256mb RAM.... seriously.<!--QuoteEnd--></div><!--QuoteEEnd-->
So.. unless the performance in 1.0 will be <b>a lot</b> better, those system requirements are heavily undersized.
To prevent loads of disappointed customers, I recommend changing the minimum required processor to at least <b>Intel Core i5 3.5 GHz</b> and the recommended processor to <b>Intel Core i5 4.5 GHz (OC)</b>.<!--QuoteEnd--></div><!--QuoteEEnd-->
I don't know how knowledgeable you are about computers but those specs are ridiculous... No game company would ever recommend an OC'd cpu to run the game... And maybe you are one of those people who think the game is still CPU locked, but I assure you it's not. Maybe your GPU isn't supported yet or maybe you are running it off an integrated GPU but right now my ATI 5850 is running NS2 smooth as silk and I could even live with a lower fps.
I'm running an Ivy Bridge i5 3570K 3.4G (Not OC'd) 8GB Ram, SSD (Helps more than you would think for map changes and game launching) and 2 ATI 5850's (but only 1 is being used for NS2 as crossfire is not supported yet). I run the game at med detail at 1600x900 with an FPS that fluctuates between 60 to 90. And a solid 50 to 60 at low detail in 1080p (native resolution).
I've tried the game with Intel HD4000 integrated graphics and I can manage around 20-30fps on low everything at 1366-768 which is decent but nothing special.
And of course the game is going to run smoother for 1.0... They wouldn't release right now without multi gpu support or without support for the majority of graphics cards....
To run NS I would estimate the required CPU to be well within the recommended stats posted in the screenshot of the OP but the GPU requirements should be boosted.
SecurityJoin Date: 2005-01-07Member: 33133Members, Constellation, Squad Five Blue
edited August 2012
<!--quoteo(post=1962781:date=Aug 14 2012, 06:09 PM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 14 2012, 06:09 PM) <a href="index.php?act=findpost&pid=1962781"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I don't know how knowledgeable you are about computers but those specs are ridiculous... No game company would ever recommend an OC'd cpu to run the game... And maybe you are one of those people who think the game is still CPU locked, but I assure you it's not. Maybe your GPU isn't supported yet or maybe you are running it off an integrated GPU but right now my ATI 5850 is running NS2 smooth as silk and I could even live with a lower fps.
I'm running an Ivy Bridge i5 3570K 3.4G (Not OC'd) 8GB Ram, SSD (Helps more than you would think for map changes and game launching) and 2 ATI 5850's (but only 1 is being used for NS2 as crossfire is not supported yet). I run the game at med detail at 1600x900 with an FPS that fluctuates between 60 to 90. And a solid 50 to 60 at low detail in 1080p (native resolution).
I've tried the game with Intel HD4000 integrated graphics and I can manage around 20-30fps on low everything at 1366-768 which is decent but nothing special.
And of course the game is going to run smoother for 1.0... They wouldn't release right now without multi gpu support or without support for the majority of graphics cards....
To run NS I would estimate the required CPU to be well within the recommended stats posted in the screenshot of the OP but the GPU requirements should be boosted.<!--QuoteEnd--></div><!--QuoteEEnd-->
Yeah, right.
Benchmark or it didn't happen. -> <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119506&st=20" target="_blank">http://www.unknownworlds.com/ns2/forums/in...19506&st=20</a>
<!--quoteo(post=1961090:date=Aug 11 2012, 08:53 AM:name=deathmonger87)--><div class='quotetop'>QUOTE (deathmonger87 @ Aug 11 2012, 08:53 AM) <a href="index.php?act=findpost&pid=1961090"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Seriously, I disregard everyone's answer when they don't respond with their FPS. I'll say the game is running great when I NEVER dip below 60 FPS.<!--QuoteEnd--></div><!--QuoteEEnd--> This is not how minimum or even recommended requirements work. 30fps average is perfectly acceptable for a recommended spec.
SecurityJoin Date: 2005-01-07Member: 33133Members, Constellation, Squad Five Blue
<!--quoteo(post=1962949:date=Aug 14 2012, 09:56 PM:name=Zek)--><div class='quotetop'>QUOTE (Zek @ Aug 14 2012, 09:56 PM) <a href="index.php?act=findpost&pid=1962949"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->This is not how minimum or even recommended requirements work. 30fps average is perfectly acceptable for a recommended spec.<!--QuoteEnd--></div><!--QuoteEEnd-->
It's impossible to get 30 stable FPS throughout a whole game with the current recommended specs. Not even to speak of the minimum requirements.
It will go <b>far</b> below 30 FPS in a full lategame match. (12-18 Slots)
This quote from the benchmark thread sums it up very well. Note that his system is even better than the recommended requirements:
<!--quoteo(post=1953243:date=Jul 23 2012, 06:49 PM:name={GGs} Chicken)--><div class='quotetop'>QUOTE ({GGs} Chicken @ Jul 23 2012, 06:49 PM) <a href="index.php?act=findpost&pid=1953243"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->- AMD Phenom II x4 965 3.4GHz - HD 6950 - Windows 7 - 8GB - 1080p - 90-100 ping - Everything except AF turned off
40 fps on a freshly restarted All-In Server with 6v6 people. Mid-game 30fps, late game 18-25fps.
Seems like everyones getting badfps early game and badfps late game. Should just sticky this: <a href="http://www.quickmeme.com/meme/3q758x/" target="_blank">http://www.quickmeme.com/meme/3q758x/</a> and call it a day.<!--QuoteEnd--></div><!--QuoteEEnd-->
<!--quoteo(post=1962975:date=Aug 14 2012, 03:41 PM:name=Security)--><div class='quotetop'>QUOTE (Security @ Aug 14 2012, 03:41 PM) <a href="index.php?act=findpost&pid=1962975"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->It's impossible to get 30 stable FPS throughout a whole game with the current recommended specs. Not even to speak of the minimum requirements.
It will go <b>far</b> below 30 FPS in a full lategame match. (12-18 Slots)
This quote from the benchmark thread sums it up very well. Note that his system is even better than the recommended requirements:<!--QuoteEnd--></div><!--QuoteEEnd-->
I believe the recommended specs are for when 1.0 is released, not in its current state.
So you may not be able to get a stable 30fps throughout a game now, once 1.0 is released you "should" be able to. I believe this is what they are trying to convey.
On the twitch stream the other night I heard a quote of:
I heard a rumour that there was a laptop running NS2, and with the physics code improvements it went from 3/4 fps to 20...
So lets hope this patch that is apparently supposed to be coming out today does actually produce those sort of results. Plus, it would be great to know if there were some physics improvements as well... not just performance.
SecurityJoin Date: 2005-01-07Member: 33133Members, Constellation, Squad Five Blue
<!--quoteo(post=1962983:date=Aug 14 2012, 10:55 PM:name=MiniH0wie)--><div class='quotetop'>QUOTE (MiniH0wie @ Aug 14 2012, 10:55 PM) <a href="index.php?act=findpost&pid=1962983"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I believe the recommended specs are for when 1.0 is released, not in its current state.
So you may not be able to get a stable 30fps throughout a game now, once 1.0 is released you "should" be able to. I believe this is what they are trying to convey.<!--QuoteEnd--></div><!--QuoteEEnd-->
Well, I will be looking forward to that massive performance increase then. ;)
<!--quoteo(post=1962985:date=Aug 14 2012, 03:56 PM:name=Security)--><div class='quotetop'>QUOTE (Security @ Aug 14 2012, 03:56 PM) <a href="index.php?act=findpost&pid=1962985"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Well, I will be looking forward to that massive performance increase then. ;)<!--QuoteEnd--></div><!--QuoteEEnd-->
Benchmark or it didn't happen. -> <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119506&st=20" target="_blank">http://www.unknownworlds.com/ns2/forums/in...19506&st=20</a><!--QuoteEnd--></div><!--QuoteEEnd-->
Frames Time (ms) Min Max Avg 75291 1200000 24 101 62.743
<!--quoteo(post=1962781:date=Aug 14 2012, 04:09 PM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 14 2012, 04:09 PM) <a href="index.php?act=findpost&pid=1962781"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->And maybe you are one of those people who think the game is still CPU locked, but I assure you it's not.<!--QuoteEnd--></div><!--QuoteEEnd--> Nonsense. This game has been severely CPU-bottlenecked since day 1, and unless you manage to have an even tighter bottleneck elsewhere in your system (integrated graphics would qualify), improvement of your (single-threaded) CPU-performance directly correlates to a higher FPS.
ScardyBobScardyBobJoin Date: 2009-11-25Member: 69528Forum Admins, Forum Moderators, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Shadow
<!--quoteo(post=1963184:date=Aug 14 2012, 06:47 PM:name=player)--><div class='quotetop'>QUOTE (player @ Aug 14 2012, 06:47 PM) <a href="index.php?act=findpost&pid=1963184"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Nonsense. This game has been severely CPU-bottlenecked since day 1, and unless you manage to have an even tighter bottleneck elsewhere in your system (integrated graphics would qualify), improvement of your (single-threaded) CPU-performance directly correlates to a higher FPS.<!--QuoteEnd--></div><!--QuoteEEnd--> I'm actually wondering if this is still true after looking at <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119900&st=0&p=1961695&#entry1961695" target="_blank">this guys benchmark</a>. The only serious difference between the setups is the GPU.
Sounds like compatibility-problems to me, but the post is sorely lacking in detail (server-tickrate, CPU-clock, graphics-settings, et cetera) so no conclusion can be drawn from that either way. All the data that we have available to us clearly shows the CPU is the main driver in what FPS you are likely to have, and has been that way for quite some time and we have clear indication as to why that is (Lua). Simply, nothing in NS2 is taxing on your graphics-card, it is not that graphically heavy a game, so a current or previous gen gfx-card will very unlikely ever be a bottleneck for you.
Show me a low-clocked CPU with a good FPS, and we will pick up on this discussion (I just called out a guy claiming exactly that in another thread).
Making recommended an overclocked processor is a poor decision. It is ill-advised to promote overclocking when it's something that should not be taken lightly. It is very easy to destroy more than just your CPU, and I'm pretty sure people will be ultra pissed off if UWE said it was a good idea and they burn their $200+ proc.
This system actually isn't that hard on the CPU, your estimates are over-reaching.
I should know a thing or two about this, I'm a professional.
<!--quoteo(post=1963184:date=Aug 15 2012, 01:47 AM:name=player)--><div class='quotetop'>QUOTE (player @ Aug 15 2012, 01:47 AM) <a href="index.php?act=findpost&pid=1963184"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Nonsense. This game has been severely CPU-bottlenecked since day 1, and unless you manage to have an even tighter bottleneck elsewhere in your system (integrated graphics would qualify), improvement of your (single-threaded) CPU-performance directly correlates to a higher FPS.<!--QuoteEnd--></div><!--QuoteEEnd-->
That simply isn't true, are you saying because your computer does this it must happen for everyone? I'm an experienced computer technician and I'm flat out telling you my CPU barely hits 50% load. The entire game is being loaded onto my GPU and the rest is in memory. The only explanation I can offer as to why this isn't true for you and is for me is that the game currently does not support all graphics cards. When I load into a game and open console I'm given a message indicating that the game recognizes 1 GPU - my ATI 5850 and some other output I can give you if you are interested. I'm really getting bored of repeating this to everyone though because they all tend to react the same way you do... "It's not true because that's not what my game does..."
<!--quoteo(post=1963202:date=Aug 15 2012, 02:29 AM:name=player)--><div class='quotetop'>QUOTE (player @ Aug 15 2012, 02:29 AM) <a href="index.php?act=findpost&pid=1963202"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Simply, nothing in NS2 is taxing on your graphics-card, it is not that graphically heavy a game, so a current or previous gen gfx-card will very unlikely ever be a bottleneck for you.
Show me a low-clocked CPU with a good FPS, and we will pick up on this discussion (I just called out a guy claiming exactly that in another thread).<!--QuoteEnd--></div><!--QuoteEEnd-->
Proof: <a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a> (Note the console message indicating that my GPU is being selected as the render device (and not the cpu!?!?! Magic? Probably... what else could it be?)
<b>AND</b>
<a href="http://i.imgur.com/1Ejoo.jpg" target="_blank">http://i.imgur.com/1Ejoo.jpg</a> (Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1. I'm now in full screen-windowed mode to get the screenshot and dual monitor mode to have the Task Manager and Afterburner on the right. Pay special attention to the afterburner showing a rise in gpu usage and temperature AND last but not least, a clear lack of a bottleneck on my CPU.
<!--quoteo(post=1963233:date=Aug 15 2012, 12:34 AM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 15 2012, 12:34 AM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Proof: <a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a> (Note the console message indicating that my GPU is being selected as the render device (and not the cpu!?!?! Magic? Probably... what else could it be?)
<b>AND</b>
<a href="http://i.imgur.com/1Ejoo.jpg" target="_blank">http://i.imgur.com/1Ejoo.jpg</a> (Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1. I'm now in full screen-windowed mode to get the screenshot and dual monitor mode to have the Task Manager and Afterburner on the right. Pay special attention to the afterburner showing a rise in gpu usage and temperature AND last but not least, a clear lack of a bottleneck on my CPU.<!--QuoteEnd--></div><!--QuoteEEnd-->
nah man. the game only runs on 2 cores, so unless you've locked the process to a single core you won't see the effects. believe me when i say that the game is mostly CPU bound. what's happening is the computer is rapidly switching the threads from core to core so you'll never really see the usage go past around 35-40%. as for the gpu, i'm sure there are usage spikes (esp when there's a lot of particle effects and such on the screen, e.g. a shotgun to the face) but even a gpu at 50% load will raise the temp from idle.
<!--quoteo(post=1963233:date=Aug 14 2012, 11:34 PM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 14 2012, 11:34 PM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->That simply isn't true, are you saying because your computer does this it must happen for everyone? I'm an experienced computer technician and I'm flat out telling you my CPU barely hits 50% load.
That's exactly what you would expect if you were CPU bottlenecked. The game is almost single-threaded; 31% CPU usage means ~100% CPU usage on one core and a little bit of usage on your other 3 cores(graphics driver and a few other minor things).
If the game had set the thread affinity mask so that the main thread runs only on core 0 you would see one core running full tilt and the others doing very little.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->(Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1.<!--QuoteEnd--></div><!--QuoteEEnd-->
40 FPS in an empty corridor translates into god-awful FPS when it actually counts.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->Pay special attention to the afterburner showing a rise in gpu usage<!--QuoteEnd--></div><!--QuoteEEnd-->
Looks like it barely even bothered increasing the GPU clock and the GPU usage is well below 100%.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->and last but not least, a clear lack of a bottleneck on my CPU.<!--QuoteEnd--></div><!--QuoteEEnd-->
Which looks utterly CPU bound. In fact, CPU usage that high on a quad core is probably only explainable by you running in windowed mode, taking a screenshot and having other apps running alongside it.
Why don't you do the obvious, conclusive test? Benchmark, then set the resolution and texture resolution to something god-awful and benchmark again. If there is no major difference then you're utterly CPU bound.
ScardyBobScardyBobJoin Date: 2009-11-25Member: 69528Forum Admins, Forum Moderators, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Shadow
<!--quoteo(post=1963233:date=Aug 14 2012, 09:34 PM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 14 2012, 09:34 PM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->That simply isn't true, are you saying because your computer does this it must happen for everyone? I'm an experienced computer technician and I'm flat out telling you my CPU barely hits 50% load. The entire game is being loaded onto my GPU and the rest is in memory. The only explanation I can offer as to why this isn't true for you and is for me is that the game currently does not support all graphics cards. When I load into a game and open console I'm given a message indicating that the game recognizes 1 GPU - my ATI 5850 and some other output I can give you if you are interested. I'm really getting bored of repeating this to everyone though because they all tend to react the same way you do... "It's not true because that's not what my game does..."
Proof: <a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a> (Note the console message indicating that my GPU is being selected as the render device (and not the cpu!?!?! Magic? Probably... what else could it be?)
<b>AND</b>
<a href="http://i.imgur.com/1Ejoo.jpg" target="_blank">http://i.imgur.com/1Ejoo.jpg</a> (Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1. I'm now in full screen-windowed mode to get the screenshot and dual monitor mode to have the Task Manager and Afterburner on the right. Pay special attention to the afterburner showing a rise in gpu usage and temperature AND last but not least, a clear lack of a bottleneck on my CPU.<!--QuoteEnd--></div><!--QuoteEEnd--> Well, the funny thing about NS2 is that its mostly single-threaded (at least the main code that is the performance bottleneck), with a bit on a second thread. As such, Task Manager is only going to show up to 50% CPU usage for a quad-core (i.e. 100% of only 2 cores = 50% total usage). It still means it can be CPU bottlenecked (as most people have experienced).
<!--quoteo(post=1963233:date=Aug 15 2012, 04:34 AM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 15 2012, 04:34 AM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I'm an experienced computer technician<!--QuoteEnd--></div><!--QuoteEEnd--> That gave me a bit of a laugh.
<!--quoteo(post=1963233:date=Aug 15 2012, 04:34 AM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 15 2012, 04:34 AM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The only explanation I can offer as to why this isn't true for you and is for me is that the game currently does not support all graphics cards. When I load into a game and open console I'm given a message indicating that the game recognizes 1 GPU - my ATI 5850 and some other output I can give you if you are interested.<!--QuoteEnd--></div><!--QuoteEEnd--> And why do you keep saying this? You do realize that all of this is abstracted away by Direct3D and Windows, NS2 doesn't actually go into your graphics-card and do a read out of the name-string, it just makes a call to an API.
Anyway, the rest is fairly well covered by the previous posts.
<!--quoteo(post=1963241:date=Aug 15 2012, 05:14 AM:name=Soylent_green)--><div class='quotetop'>QUOTE (Soylent_green @ Aug 15 2012, 05:14 AM) <a href="index.php?act=findpost&pid=1963241"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->That's exactly what you would expect if you were CPU bottlenecked. The game is almost single-threaded; 31% CPU usage means ~100% CPU usage on one core and a little bit of usage on your other 3 cores(graphics driver and a few other minor things).
If the game had set the thread affinity mask so that the main thread runs only on core 0 you would see one core running full tilt and the others doing very little.
40 FPS in an empty corridor translates into god-awful FPS when it actually counts.
Looks like it barely even bothered increasing the GPU clock and the GPU usage is well below 100%.
Which looks utterly CPU bound. In fact, CPU usage that high on a quad core is probably only explainable by you running in windowed mode, taking a screenshot and having other apps running alongside it.
Why don't you do the obvious, conclusive test? Benchmark, then set the resolution and texture resolution to something god-awful and benchmark again. If there is no major difference then you're utterly CPU bound.<!--QuoteEnd--></div><!--QuoteEEnd-->
I did benchmark,
Frames Time (ms) Min Max Avg 75291 1200000 24 101 62.743
1080p Low settings. So 40 FPS is actually quite good (and what empty hallway, I had just killed 1 marine and another is in view shooting) when you consider dual display with other apps going.
<!--quoteo(post=1963252:date=Aug 15 2012, 06:04 AM:name=player)--><div class='quotetop'>QUOTE (player @ Aug 15 2012, 06:04 AM) <a href="index.php?act=findpost&pid=1963252"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->That gave me a bit of a laugh.<!--QuoteEnd--></div><!--QuoteEEnd-->
Classy guy. Odd that you choose to make this personal... I still think you are wrong and you still think I'm wrong so lets just drop this. My benchmarks prove this game is running fine and I'd hardly call 30% - 40% CPU across 4 cores a bottleneck. Also I had multi core rendering turned off in graphical options.
<b>Please follow these guidelines while capturing your benchmark: - Run FRAPS Benchmark for 20 Minutes (1200 Seconds) - DO NOT benchmark while in the readyroom. Wait until you have joined a team, and the round has started. - DO NOT benchmark loading screens. - ONLY benchmark on NS2_SUMMIT (This will help keep the benchmarks somewhat consistent.) - ONLY benchmark a single round. DO NOT benchmark multiple rounds which return you to the Readyroom or display the Loading screen. - ONLY benchmark while running the game FULLSCREEN. - ONLY benchmark on "High" Visual Detail Settings and ENABLE everything (Bloom, Atmospherics, etc.) </b><!--QuoteEnd--></div><!--QuoteEEnd-->
Following all of these instructions, you will absolutely not get 62 Avg Fps with your System.
Also, for the love of god:
<!--quoteo(post=1963230:date=Aug 15 2012, 06:19 AM:name=BloodyIron)--><div class='quotetop'>QUOTE (BloodyIron @ Aug 15 2012, 06:19 AM) <a href="index.php?act=findpost&pid=1963230"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I should know a thing or two about this, I'm a professional.<!--QuoteEnd--></div><!--QuoteEEnd-->
<!--quoteo(post=1963233:date=Aug 15 2012, 06:34 AM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 15 2012, 06:34 AM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I'm an experienced computer technician<!--QuoteEnd--></div><!--QuoteEEnd-->
Don't use such statements as argument in a discussion. It doesn't make you look very smart.
DghelneshiAims to surpass Fana in post edits.Join Date: 2011-11-01Member: 130634Members, Squad Five Blue, Reinforced - Shadow
edited August 2012
1. You should consider benchmarking with an FPS graph instead of min/avg/max numbers. You can do 80fps avg if you stay in the ready room for a bit and have a quick round or always try to die as fast as possible to spectate a lot, so those numbers don't tell much. 2. If the game is GPU bottlenecked, why do you not have lower fps than people with the same or similar processor, but a much better GPU? 3. If the game is GPU bottlenecked, why do you have 100 fps when spectating, but much less when actually playing?
SecurityJoin Date: 2005-01-07Member: 33133Members, Constellation, Squad Five Blue
<!--quoteo(post=1963590:date=Aug 15 2012, 08:49 PM:name=Dghelneshi)--><div class='quotetop'>QUOTE (Dghelneshi @ Aug 15 2012, 08:49 PM) <a href="index.php?act=findpost&pid=1963590"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->1. You should consider benchmarking with an FPS graph instead of min/avg/max numbers. You can do 80fps avg if you stay in the ready room for a bit and have a quick round or always try to die as fast as possible to spectate a lot, so those numbers don't tell much. 2. If the game is GPU bottlenecked, why do you not have lower fps than people with the same or similar processor, but a much better GPU? 3. If the game is GPU bottlenecked, why do you have 100 fps when spectating, but much less when actually playing?<!--QuoteEnd--></div><!--QuoteEEnd-->
<!--quoteo(post=1963470:date=Aug 15 2012, 09:25 AM:name=Security)--><div class='quotetop'>QUOTE (Security @ Aug 15 2012, 09:25 AM) <a href="index.php?act=findpost&pid=1963470"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->You seem to have a habit of not reading and understanding other peoples posts. You are supposed to benchmark with all visual details on High and On.
From the benchmark thread:
Following all of these instructions, you will absolutely not get 62 Avg Fps with your System.
Also, for the love of god:
Don't use such statements as argument in a discussion. It doesn't make you look very smart.
Especially when you are obviously wrong.<!--QuoteEnd--></div><!--QuoteEEnd-->
The argument is about minimum specs and recommended specs. Why would he have to test with all settings on max? That has nothing to do with the current conversation. That is not what th ey base these specs on.
My CPU has no problem with this game and my fps drops. Based on the previous comments, some of you don't seem to realize that there is more processing for your GPU in combat than there is in a static environment, like the ready room. Maybe this is what is confusing you?
"If you run a benchmark with all settings maxed, you will clearly see that the minimum system requirements are off!". How does that make sense to anyone?
SecurityJoin Date: 2005-01-07Member: 33133Members, Constellation, Squad Five Blue
edited August 2012
<!--quoteo(post=1963596:date=Aug 15 2012, 09:01 PM:name=lumina)--><div class='quotetop'>QUOTE (lumina @ Aug 15 2012, 09:01 PM) <a href="index.php?act=findpost&pid=1963596"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Why would he have to test with all settings on max?<!--QuoteEnd--></div><!--QuoteEEnd-->
Because there have to be consistant conditions for benchmarking, to produce useful and easy to compare results.
Thats the sole purpose of the <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119506" target="_blank">Benchmark Thread's</a> Guidelines, which I linked to him.
Comments
Has anyone noticed any inconsistencies on how the game functions on different FPSes? NS1 had jetpack tricks, jump timing, rate of fire to name a few. NS2 had some inconsistencies at movement at some point.
Has anyone noticed any similar thingies on the present NS2?<!--QuoteEnd--></div><!--QuoteEEnd-->
In theory, there shouldn't be, in practice, such bugs have popped up. I remember there were issues with the rifle ROF that would actually slow down with lower FPS. If you find other issues, report them so they can get fixed.
lolz
So.. unless the performance in 1.0 will be <b>a lot</b> better, those system requirements are heavily undersized.
To prevent loads of disappointed customers, I recommend changing the minimum required processor to at least <b>Intel Core i5 3.5 GHz</b> and the recommended processor to <b>Intel Core i5 4.5 GHz (OC)</b>.<!--QuoteEnd--></div><!--QuoteEEnd-->
I don't know how knowledgeable you are about computers but those specs are ridiculous... No game company would ever recommend an OC'd cpu to run the game... And maybe you are one of those people who think the game is still CPU locked, but I assure you it's not. Maybe your GPU isn't supported yet or maybe you are running it off an integrated GPU but right now my ATI 5850 is running NS2 smooth as silk and I could even live with a lower fps.
I'm running an Ivy Bridge i5 3570K 3.4G (Not OC'd) 8GB Ram, SSD (Helps more than you would think for map changes and game launching) and 2 ATI 5850's (but only 1 is being used for NS2 as crossfire is not supported yet). I run the game at med detail at 1600x900 with an FPS that fluctuates between 60 to 90. And a solid 50 to 60 at low detail in 1080p (native resolution).
I've tried the game with Intel HD4000 integrated graphics and I can manage around 20-30fps on low everything at 1366-768 which is decent but nothing special.
And of course the game is going to run smoother for 1.0... They wouldn't release right now without multi gpu support or without support for the majority of graphics cards....
To run NS I would estimate the required CPU to be well within the recommended stats posted in the screenshot of the OP but the GPU requirements should be boosted.
I'm running an Ivy Bridge i5 3570K 3.4G (Not OC'd) 8GB Ram, SSD (Helps more than you would think for map changes and game launching) and 2 ATI 5850's (but only 1 is being used for NS2 as crossfire is not supported yet). I run the game at med detail at 1600x900 with an FPS that fluctuates between 60 to 90. And a solid 50 to 60 at low detail in 1080p (native resolution).
I've tried the game with Intel HD4000 integrated graphics and I can manage around 20-30fps on low everything at 1366-768 which is decent but nothing special.
And of course the game is going to run smoother for 1.0... They wouldn't release right now without multi gpu support or without support for the majority of graphics cards....
To run NS I would estimate the required CPU to be well within the recommended stats posted in the screenshot of the OP but the GPU requirements should be boosted.<!--QuoteEnd--></div><!--QuoteEEnd-->
Yeah, right.
Benchmark or it didn't happen. -> <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119506&st=20" target="_blank">http://www.unknownworlds.com/ns2/forums/in...19506&st=20</a>
This is not how minimum or even recommended requirements work. 30fps average is perfectly acceptable for a recommended spec.
It's impossible to get 30 stable FPS throughout a whole game with the current recommended specs. Not even to speak of the minimum requirements.
It will go <b>far</b> below 30 FPS in a full lategame match. (12-18 Slots)
This quote from the benchmark thread sums it up very well. Note that his system is even better than the recommended requirements:
<!--quoteo(post=1953243:date=Jul 23 2012, 06:49 PM:name={GGs} Chicken)--><div class='quotetop'>QUOTE ({GGs} Chicken @ Jul 23 2012, 06:49 PM) <a href="index.php?act=findpost&pid=1953243"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->- AMD Phenom II x4 965 3.4GHz
- HD 6950
- Windows 7
- 8GB
- 1080p
- 90-100 ping
- Everything except AF turned off
40 fps on a freshly restarted All-In Server with 6v6 people. Mid-game 30fps, late game 18-25fps.
Seems like everyones getting badfps early game and badfps late game. Should just sticky this: <a href="http://www.quickmeme.com/meme/3q758x/" target="_blank">http://www.quickmeme.com/meme/3q758x/</a> and call it a day.<!--QuoteEnd--></div><!--QuoteEEnd-->
It will go <b>far</b> below 30 FPS in a full lategame match. (12-18 Slots)
This quote from the benchmark thread sums it up very well. Note that his system is even better than the recommended requirements:<!--QuoteEnd--></div><!--QuoteEEnd-->
I believe the recommended specs are for when 1.0 is released, not in its current state.
So you may not be able to get a stable 30fps throughout a game now, once 1.0 is released you "should" be able to. I believe this is what they are trying to convey.
I heard a rumour that there was a laptop running NS2, and with the physics code improvements it went from 3/4 fps to 20...
So lets hope this patch that is apparently supposed to be coming out today does actually produce those sort of results. Plus, it would be great to know if there were some physics improvements as well... not just performance.
So you may not be able to get a stable 30fps throughout a game now, once 1.0 is released you "should" be able to. I believe this is what they are trying to convey.<!--QuoteEnd--></div><!--QuoteEEnd-->
Well, I will be looking forward to that massive performance increase then. ;)
You and me both!
Benchmark or it didn't happen. -> <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119506&st=20" target="_blank">http://www.unknownworlds.com/ns2/forums/in...19506&st=20</a><!--QuoteEnd--></div><!--QuoteEEnd-->
Frames Time (ms) Min Max Avg
75291 1200000 24 101 62.743
Low detail 1080p
On summit, 9v9 ~75 ping, Marine side.
Nonsense. This game has been severely CPU-bottlenecked since day 1, and unless you manage to have an even tighter bottleneck elsewhere in your system (integrated graphics would qualify), improvement of your (single-threaded) CPU-performance directly correlates to a higher FPS.
I'm actually wondering if this is still true after looking at <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119900&st=0&p=1961695&#entry1961695" target="_blank">this guys benchmark</a>. The only serious difference between the setups is the GPU.
Show me a low-clocked CPU with a good FPS, and we will pick up on this discussion (I just called out a guy claiming exactly that in another thread).
This system actually isn't that hard on the CPU, your estimates are over-reaching.
I should know a thing or two about this, I'm a professional.
That simply isn't true, are you saying because your computer does this it must happen for everyone? I'm an experienced computer technician and I'm flat out telling you my CPU barely hits 50% load. The entire game is being loaded onto my GPU and the rest is in memory. The only explanation I can offer as to why this isn't true for you and is for me is that the game currently does not support all graphics cards. When I load into a game and open console I'm given a message indicating that the game recognizes 1 GPU - my ATI 5850 and some other output I can give you if you are interested. I'm really getting bored of repeating this to everyone though because they all tend to react the same way you do... "It's not true because that's not what my game does..."
<!--quoteo(post=1963202:date=Aug 15 2012, 02:29 AM:name=player)--><div class='quotetop'>QUOTE (player @ Aug 15 2012, 02:29 AM) <a href="index.php?act=findpost&pid=1963202"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Simply, nothing in NS2 is taxing on your graphics-card, it is not that graphically heavy a game, so a current or previous gen gfx-card will very unlikely ever be a bottleneck for you.
Show me a low-clocked CPU with a good FPS, and we will pick up on this discussion (I just called out a guy claiming exactly that in another thread).<!--QuoteEnd--></div><!--QuoteEEnd-->
Proof:
<a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a> (Note the console message indicating that my GPU is being selected as the render device (and not the cpu!?!?! Magic? Probably... what else could it be?)
<b>AND</b>
<a href="http://i.imgur.com/1Ejoo.jpg" target="_blank">http://i.imgur.com/1Ejoo.jpg</a> (Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1. I'm now in full screen-windowed mode to get the screenshot and dual monitor mode to have the Task Manager and Afterburner on the right. Pay special attention to the afterburner showing a rise in gpu usage and temperature AND last but not least, a clear lack of a bottleneck on my CPU.
<a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a> (Note the console message indicating that my GPU is being selected as the render device (and not the cpu!?!?! Magic? Probably... what else could it be?)
<b>AND</b>
<a href="http://i.imgur.com/1Ejoo.jpg" target="_blank">http://i.imgur.com/1Ejoo.jpg</a> (Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1. I'm now in full screen-windowed mode to get the screenshot and dual monitor mode to have the Task Manager and Afterburner on the right. Pay special attention to the afterburner showing a rise in gpu usage and temperature AND last but not least, a clear lack of a bottleneck on my CPU.<!--QuoteEnd--></div><!--QuoteEEnd-->
nah man. the game only runs on 2 cores, so unless you've locked the process to a single core you won't see the effects. believe me when i say that the game is mostly CPU bound. what's happening is the computer is rapidly switching the threads from core to core so you'll never really see the usage go past around 35-40%. as for the gpu, i'm sure there are usage spikes (esp when there's a lot of particle effects and such on the screen, e.g. a shotgun to the face) but even a gpu at 50% load will raise the temp from idle.
<a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a>[...]<!--QuoteEnd--></div><!--QuoteEEnd-->
That's exactly what you would expect if you were CPU bottlenecked. The game is almost single-threaded; 31% CPU usage means ~100% CPU usage on one core and a little bit of usage on your other 3 cores(graphics driver and a few other minor things).
If the game had set the thread affinity mask so that the main thread runs only on core 0 you would see one core running full tilt and the others doing very little.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->(Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1.<!--QuoteEnd--></div><!--QuoteEEnd-->
40 FPS in an empty corridor translates into god-awful FPS when it actually counts.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->Pay special attention to the afterburner showing a rise in gpu usage<!--QuoteEnd--></div><!--QuoteEEnd-->
Looks like it barely even bothered increasing the GPU clock and the GPU usage is well below 100%.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->and last but not least, a clear lack of a bottleneck on my CPU.<!--QuoteEnd--></div><!--QuoteEEnd-->
Which looks utterly CPU bound. In fact, CPU usage that high on a quad core is probably only explainable by you running in windowed mode, taking a screenshot and having other apps running alongside it.
Why don't you do the obvious, conclusive test? Benchmark, then set the resolution and texture resolution to something god-awful and benchmark again. If there is no major difference then you're utterly CPU bound.
Proof:
<a href="http://i.imgur.com/pHEN9.jpg" target="_blank">http://i.imgur.com/pHEN9.jpg</a> (Note the console message indicating that my GPU is being selected as the render device (and not the cpu!?!?! Magic? Probably... what else could it be?)
<b>AND</b>
<a href="http://i.imgur.com/1Ejoo.jpg" target="_blank">http://i.imgur.com/1Ejoo.jpg</a> (Note the FPS on the top left, which is obviously lower than the benchmarks I posted above because 1. I'm now in full screen-windowed mode to get the screenshot and dual monitor mode to have the Task Manager and Afterburner on the right. Pay special attention to the afterburner showing a rise in gpu usage and temperature AND last but not least, a clear lack of a bottleneck on my CPU.<!--QuoteEnd--></div><!--QuoteEEnd-->
Well, the funny thing about NS2 is that its mostly single-threaded (at least the main code that is the performance bottleneck), with a bit on a second thread. As such, Task Manager is only going to show up to 50% CPU usage for a quad-core (i.e. 100% of only 2 cores = 50% total usage). It still means it can be CPU bottlenecked (as most people have experienced).
That gave me a bit of a laugh.
<!--quoteo(post=1963233:date=Aug 15 2012, 04:34 AM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 15 2012, 04:34 AM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The only explanation I can offer as to why this isn't true for you and is for me is that the game currently does not support all graphics cards. When I load into a game and open console I'm given a message indicating that the game recognizes 1 GPU - my ATI 5850 and some other output I can give you if you are interested.<!--QuoteEnd--></div><!--QuoteEEnd-->
And why do you keep saying this? You do realize that all of this is abstracted away by Direct3D and Windows, NS2 doesn't actually go into your graphics-card and do a read out of the name-string, it just makes a call to an API.
Anyway, the rest is fairly well covered by the previous posts.
If the game had set the thread affinity mask so that the main thread runs only on core 0 you would see one core running full tilt and the others doing very little.
40 FPS in an empty corridor translates into god-awful FPS when it actually counts.
Looks like it barely even bothered increasing the GPU clock and the GPU usage is well below 100%.
Which looks utterly CPU bound. In fact, CPU usage that high on a quad core is probably only explainable by you running in windowed mode, taking a screenshot and having other apps running alongside it.
Why don't you do the obvious, conclusive test? Benchmark, then set the resolution and texture resolution to something god-awful and benchmark again. If there is no major difference then you're utterly CPU bound.<!--QuoteEnd--></div><!--QuoteEEnd-->
I did benchmark,
Frames Time (ms) Min Max Avg
75291 1200000 24 101 62.743
1080p Low settings. So 40 FPS is actually quite good (and what empty hallway, I had just killed 1 marine and another is in view shooting) when you consider dual display with other apps going.
<!--quoteo(post=1963252:date=Aug 15 2012, 06:04 AM:name=player)--><div class='quotetop'>QUOTE (player @ Aug 15 2012, 06:04 AM) <a href="index.php?act=findpost&pid=1963252"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->That gave me a bit of a laugh.<!--QuoteEnd--></div><!--QuoteEEnd-->
Classy guy. Odd that you choose to make this personal... I still think you are wrong and you still think I'm wrong so lets just drop this. My benchmarks prove this game is running fine and I'd hardly call 30% - 40% CPU across 4 cores a bottleneck. Also I had multi core rendering turned off in graphical options.
My NS2 still stutters/drops to low frames around action and one of the four cores is nearly always maxed out.
On a side note, in your screen shot you are standing in the ready room, where most of the games code is not being run, sigh.
Frames Time (ms) Min Max Avg
75291 1200000 24 101 62.743
1080p <b>Low settings</b>.<!--QuoteEnd--></div><!--QuoteEEnd-->
You seem to have a habit of not reading and understanding other peoples posts. You are supposed to benchmark with all visual details on High and On.
From the benchmark thread:
<!--quoteo(post=1952910:date=Jul 21 2012, 07:11 AM:name=DarkATi)--><div class='quotetop'>QUOTE (DarkATi @ Jul 21 2012, 07:11 AM) <a href="index.php?act=findpost&pid=1952910"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec--><!--coloro:#FFFF00--><span style="color:#FFFF00"><!--/coloro--><!--sizeo:5--><span style="font-size:18pt;line-height:100%"><!--/sizeo--><b>How to Properly Benchmark NS2</b><!--sizec--></span><!--/sizec--><!--colorc--></span><!--/colorc-->
<b>Please follow these guidelines while capturing your benchmark:
- Run FRAPS Benchmark for 20 Minutes (1200 Seconds)
- DO NOT benchmark while in the readyroom. Wait until you have joined a team, and the round has started.
- DO NOT benchmark loading screens.
- ONLY benchmark on NS2_SUMMIT (This will help keep the benchmarks somewhat consistent.)
- ONLY benchmark a single round. DO NOT benchmark multiple rounds which return you to the Readyroom or display the Loading screen.
- ONLY benchmark while running the game FULLSCREEN.
- ONLY benchmark on "High" Visual Detail Settings and ENABLE everything (Bloom, Atmospherics, etc.)
</b><!--QuoteEnd--></div><!--QuoteEEnd-->
Following all of these instructions, you will absolutely not get 62 Avg Fps with your System.
Also, for the love of god:
<!--quoteo(post=1963230:date=Aug 15 2012, 06:19 AM:name=BloodyIron)--><div class='quotetop'>QUOTE (BloodyIron @ Aug 15 2012, 06:19 AM) <a href="index.php?act=findpost&pid=1963230"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I should know a thing or two about this, I'm a professional.<!--QuoteEnd--></div><!--QuoteEEnd-->
<!--quoteo(post=1963233:date=Aug 15 2012, 06:34 AM:name=darkfiction)--><div class='quotetop'>QUOTE (darkfiction @ Aug 15 2012, 06:34 AM) <a href="index.php?act=findpost&pid=1963233"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I'm an experienced computer technician<!--QuoteEnd--></div><!--QuoteEEnd-->
Don't use such statements as argument in a discussion. It doesn't make you look very smart.
Especially when you are obviously wrong.
2. If the game is GPU bottlenecked, why do you not have lower fps than people with the same or similar processor, but a much better GPU?
3. If the game is GPU bottlenecked, why do you have 100 fps when spectating, but much less when actually playing?
2. If the game is GPU bottlenecked, why do you not have lower fps than people with the same or similar processor, but a much better GPU?
3. If the game is GPU bottlenecked, why do you have 100 fps when spectating, but much less when actually playing?<!--QuoteEnd--></div><!--QuoteEEnd-->
Indeed.
From the benchmark thread:
Following all of these instructions, you will absolutely not get 62 Avg Fps with your System.
Also, for the love of god:
Don't use such statements as argument in a discussion. It doesn't make you look very smart.
Especially when you are obviously wrong.<!--QuoteEnd--></div><!--QuoteEEnd-->
The argument is about minimum specs and recommended specs. Why would he have to test with all settings on max? That has nothing to do with the current conversation. That is not what th
ey base these specs on.
My CPU has no problem with this game and my fps drops. Based on the previous comments, some of you don't seem to realize that there is more processing for your GPU in combat than there is in a static environment, like the ready room. Maybe this is what is confusing you?
"If you run a benchmark with all settings maxed, you will clearly see that the minimum system requirements are off!". How does that make sense to anyone?
Because there have to be consistant conditions for benchmarking, to produce useful and easy to compare results.
Thats the sole purpose of the <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119506" target="_blank">Benchmark Thread's</a> Guidelines, which I linked to him.