<!--quoteo(post=2053973:date=Dec 31 2012, 08:12 PM:name=buhehe)--><div class='quotetop'>QUOTE (buhehe @ Dec 31 2012, 08:12 PM) <a href="index.php?act=findpost&pid=2053973"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->As of now, 2600K = 2500K in games. HT doesn't provide the tiniest benefit (which it does in video/audio encoding and many productivity tasks). I can't name a single game which benefits from more than 4 cores.
I get a steady 50ish FPS on 20-24 players, all maxed @ 1680*1050 except AO set to OFF. I rarely dip below 30 FPS when things get messy with many players during late game.
My bottleneck seems to be my GPU: r_stats 1 shows that the CPU usually waits 5ms for the GPU. With AO maxed it was 8ms.
Overall I'm satisfied with *my* performance, even if it's quite poor for the average PC out there.<!--QuoteEnd--></div><!--QuoteEEnd-->
I've actually tested the 3570k and 3770k in builds. You will get a performance advantage with the 3770k. A lot of the benchmarks that say otherwise are misleading, they are done in fixed-movie-scenes in single player modes, where the clock of the cpu is the main factor.
<!--quoteo(post=2053950:date=Dec 31 2012, 07:19 PM:name=DC_Darkling)--><div class='quotetop'>QUOTE (DC_Darkling @ Dec 31 2012, 07:19 PM) <a href="index.php?act=findpost&pid=2053950"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I am sorry but I have to correct you. Allow me to explain.
As you already said, my video card is better. But you went wrong. my CPU is also superior.
Yours is a i5-2500k. This is, according to the intel site, a 4 core with 4 threads, running at 3.7. (did you overclock?) Mine is a 4 core with 8 threads. Windows 7 + new games like NS2 can run multiple threads, so my CPU can run more threads for NS2 at one time. While its not as black and white as im gona say now, I can sort of run twice as much on my cpu as you can.
From the looks of it, both would be a bottleneck considering many stated its a CPU intensive task. (I can check next time I play myself) This may not feel strange to me, as many physics calculations in many games run directly on most CPUs instead the vid card.<!--QuoteEnd--></div><!--QuoteEEnd-->
Sorry I missed read your CPU the first time around but I am actually running at 4Ghz right now.
But looking at the benchmarks for a i7 940 (couldnt find the 930) vs i5 2500k at stock speeds, the i5 2500k pulls ahead for most tests. <a href="http://www.anandtech.com/bench/Product/46?vs=288" target="_blank">http://www.anandtech.com/bench/Product/46?vs=288</a>
<!--quoteo(post=2054220:date=Jan 1 2013, 11:38 AM:name=0xDECAFBAD)--><div class='quotetop'>QUOTE (0xDECAFBAD @ Jan 1 2013, 11:38 AM) <a href="index.php?act=findpost&pid=2054220"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->How do you guys get the graphs from Fraps benchmarks?<!--QuoteEnd--></div><!--QuoteEEnd-->
I just open the fps.csv file in Excel 07 and select insert 2-D line and it automatically creates the graphs I posted.
<!--quoteo(post=2054220:date=Jan 1 2013, 12:38 PM:name=0xDECAFBAD)--><div class='quotetop'>QUOTE (0xDECAFBAD @ Jan 1 2013, 12:38 PM) <a href="index.php?act=findpost&pid=2054220"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->How do you guys get the graphs from Fraps benchmarks?<!--QuoteEnd--></div><!--QuoteEEnd--> I use FRAFS Bench Viewer.
<!--quoteo(post=2054270:date=Jan 1 2013, 01:10 PM:name=MiniH0wie)--><div class='quotetop'>QUOTE (MiniH0wie @ Jan 1 2013, 01:10 PM) <a href="index.php?act=findpost&pid=2054270"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Sorry I missed read your CPU the first time around but I am actually running at 4Ghz right now.
But looking at the benchmarks for a i7 940 (couldnt find the 930) vs i5 2500k at stock speeds, the i5 2500k pulls ahead for most tests. <a href="http://www.anandtech.com/bench/Product/46?vs=288" target="_blank">http://www.anandtech.com/bench/Product/46?vs=288</a><!--QuoteEnd--></div><!--QuoteEEnd-->
Those benchmarks on that link when done in games are done in single player cut-movie-scenes. All that matters in those is clock speed. Extremely misleading unless all you care about is how good the movie scenes in games play.
<!--quoteo(post=2053970:date=Dec 31 2012, 06:01 PM:name=DC_Darkling)--><div class='quotetop'>QUOTE (DC_Darkling @ Dec 31 2012, 06:01 PM) <a href="index.php?act=findpost&pid=2053970"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->(on a side note, this is a cpu which model is not THAT old. And not all cpu are as good at certain calculations. so results may very)<!--QuoteEnd--></div><!--QuoteEEnd-->
An i7 9xx series is a first generation i7 and those came out in 2008, over 4 years is pretty old especially when you're 2 generations behind.
<!--quoteo(post=2054262:date=Jan 1 2013, 07:49 PM:name=Rich_)--><div class='quotetop'>QUOTE (Rich_ @ Jan 1 2013, 07:49 PM) <a href="index.php?act=findpost&pid=2054262"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I've actually tested the 3570k and 3770k in builds. You will get a performance advantage with the 3770k. A lot of the benchmarks that say otherwise are misleading, they are done in fixed-movie-scenes in single player modes, where the clock of the cpu is the main factor.<!--QuoteEnd--></div><!--QuoteEEnd-->
So, all the benchmarks on the internet are wrong and you're right? Everywhere you can see how the difference between the 2 is so tiny, often 0% - 3%, probably justified by the 100 more hertzes 3770k has over 3570k.
<!--quoteo(post=2054385:date=Jan 1 2013, 06:13 PM:name=buhehe)--><div class='quotetop'>QUOTE (buhehe @ Jan 1 2013, 06:13 PM) <a href="index.php?act=findpost&pid=2054385"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->So, all the benchmarks on the internet are wrong and you're right? Everywhere you can see how the difference between the 2 is so tiny, often 0% - 3%, probably justified by the 100 more hertzes 3770k has over 3570k.<!--QuoteEnd--></div><!--QuoteEEnd-->
No but your butthurt is showing. Only in the benchmarks when it is done in movie-scenes for consistency does the i5 show similar results to the i7. I said this, you are butthurt. Your i5 is weak sauce brah. Find the benchmarks with actual game play instead of movie cuts and you'll get mad butthurt cuz i7 does in fact kick i5's ass IN GAME. u wasted ur money brah, u fail.
THIS is an in game benchmark. Replace that i7 with an i5, and keep the 480. And guess what, you'll have much less frames. butthurt i5 fanboys weak sauce noobs.
btw nice rig ymicrazy502, your mannly processor makes me respect you.
<!--quoteo(post=2054409:date=Jan 1 2013, 08:20 PM:name=Rich_)--><div class='quotetop'>QUOTE (Rich_ @ Jan 1 2013, 08:20 PM) <a href="index.php?act=findpost&pid=2054409"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->THIS is an in game benchmark. Replace that i7 with an i5, and keep the 480. And guess what, you'll have much less frames. butthurt i5 fanboys weak sauce noobs.
btw nice rig ymicrazy502, your mannly processor makes me respect you.<!--QuoteEnd--></div><!--QuoteEEnd-->
kinda OT but you do realize that ns2 is probably one of the very few games that benefits heavily from HT right? In most games, the extra o/c you gain from disabling HT gives more performance than keeping it enabled.
<!--quoteo(post=2054411:date=Jan 1 2013, 07:24 PM:name=Makenshi)--><div class='quotetop'>QUOTE (Makenshi @ Jan 1 2013, 07:24 PM) <a href="index.php?act=findpost&pid=2054411"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->kinda OT but you do realize that ns2 is probably one of the very few games that benefits heavily from HT right? In most games, the extra o/c you gain from disabling HT gives more performance than keeping it enabled.<!--QuoteEnd--></div><!--QuoteEEnd-->
Games made over 2 years ago maybe. Or kids games like kawapooty maddern buttfare syclops 3. kids who cant afford hyperthreading with mommy's credit card.
Wow very intelligent conversation we are having here.... It's OT anyways so whatever, but since you brought it up: I am using an i7 that I bought myself
Interesting point though, I'll try 3.6ghz with HT on and see if NS2 runs better than what I have now (4ghz w/o HT)
<!--quoteo(post=2054420:date=Jan 1 2013, 07:35 PM:name=Makenshi)--><div class='quotetop'>QUOTE (Makenshi @ Jan 1 2013, 07:35 PM) <a href="index.php?act=findpost&pid=2054420"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Wow very intelligent conversation we are having here.... It's OT anyways so whatever, but since you brought it up: I am using an i7 that I bought myself
Interesting point though, I'll try 3.6ghz with HT on and see if NS2 runs better than what I have now (4ghz w/o HT)<!--QuoteEnd--></div><!--QuoteEEnd-->
To make it a true comparison you'd have to try without HT @ 3.6 to see the difference in practice. those 9xx intels dont overclock high?
<!--quoteo(post=2054421:date=Jan 1 2013, 08:44 PM:name=Rich_)--><div class='quotetop'>QUOTE (Rich_ @ Jan 1 2013, 08:44 PM) <a href="index.php?act=findpost&pid=2054421"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->To make it a true comparison you'd have to try without HT @ 3.6 to see the difference in practice. those 9xx intels dont overclock high?<!--QuoteEnd--></div><!--QuoteEEnd-->
It seems like you don't have much experience overclocking i7s. Guess there's no need to try after all. Enabling HT when you overclock generally results in much higher temp and takes more vcore to maintain a stable clock. It's basically a crapshoot after 4ghz on air without HT enabled and all but impossible with HT unless you are very lucky.
Comments
I can't name a single game which benefits from more than 4 cores.
On topic:
2500K @ 4.2 Ghz
Asus 6950 2GB
8GB DDR3 1600
SSD Samsung 830 128GB
I get a steady 50ish FPS on 20-24 players, all maxed @ 1680*1050 except AO set to OFF.
I rarely dip below 30 FPS when things get messy with many players during late game.
My bottleneck seems to be my GPU: r_stats 1 shows that the CPU usually waits 5ms for the GPU.
With AO maxed it was 8ms.
Overall I'm satisfied with *my* performance, even if it's quite poor for the average PC out there.<!--QuoteEnd--></div><!--QuoteEEnd-->
I've actually tested the 3570k and 3770k in builds. You will get a performance advantage with the 3770k. A lot of the benchmarks that say otherwise are misleading, they are done in fixed-movie-scenes in single player modes, where the clock of the cpu is the main factor.
As you already said, my video card is better. But you went wrong. my CPU is also superior.
Yours is a i5-2500k. This is, according to the intel site, a 4 core with 4 threads, running at 3.7. (did you overclock?)
Mine is a 4 core with 8 threads.
Windows 7 + new games like NS2 can run multiple threads, so my CPU can run more threads for NS2 at one time. While its not as black and white as im gona say now, I can sort of run twice as much on my cpu as you can.
From the looks of it, both would be a bottleneck considering many stated its a CPU intensive task. (I can check next time I play myself)
This may not feel strange to me, as many physics calculations in many games run directly on most CPUs instead the vid card.<!--QuoteEnd--></div><!--QuoteEEnd-->
Sorry I missed read your CPU the first time around but I am actually running at 4Ghz right now.
But looking at the benchmarks for a i7 940 (couldnt find the 930) vs i5 2500k at stock speeds, the i5 2500k pulls ahead for most tests.
<a href="http://www.anandtech.com/bench/Product/46?vs=288" target="_blank">http://www.anandtech.com/bench/Product/46?vs=288</a>
I just open the fps.csv file in Excel 07 and select insert 2-D line and it automatically creates the graphs I posted.
I use FRAFS Bench Viewer.
<a href="http://sourceforge.net/projects/frafsbenchview/files/" target="_blank">http://sourceforge.net/projects/frafsbenchview/files/</a>
But looking at the benchmarks for a i7 940 (couldnt find the 930) vs i5 2500k at stock speeds, the i5 2500k pulls ahead for most tests.
<a href="http://www.anandtech.com/bench/Product/46?vs=288" target="_blank">http://www.anandtech.com/bench/Product/46?vs=288</a><!--QuoteEnd--></div><!--QuoteEEnd-->
Those benchmarks on that link when done in games are done in single player cut-movie-scenes. All that matters in those is clock speed. Extremely misleading unless all you care about is how good the movie scenes in games play.
An i7 9xx series is a first generation i7 and those came out in 2008, over 4 years is pretty old especially when you're 2 generations behind.
So, all the benchmarks on the internet are wrong and you're right?
Everywhere you can see how the difference between the 2 is so tiny, often 0% - 3%, probably justified by the 100 more hertzes 3770k has over 3570k.
i7 920 @ 4.2 ghz
Gtx 480 Oced 900/2200
1920x1200 in 24 person server in refinery
2013-01-01 19:18:25 - ns2
Frames: 100234 - Time: 1200000ms - Avg: 83.528 - Min: 42 - Max: 135
Everywhere you can see how the difference between the 2 is so tiny, often 0% - 3%, probably justified by the 100 more hertzes 3770k has over 3570k.<!--QuoteEnd--></div><!--QuoteEEnd-->
No but your butthurt is showing. Only in the benchmarks when it is done in movie-scenes for consistency does the i5 show similar results to the i7. I said this, you are butthurt. Your i5 is weak sauce brah. Find the benchmarks with actual game play instead of movie cuts and you'll get mad butthurt cuz i7 does in fact kick i5's ass IN GAME. u wasted ur money brah, u fail.
i7 920 @ 4.2 ghz
Gtx 480 Oced 900/2200
1920x1200 in 24 person server in refinery
2013-01-01 19:18:25 - ns2
Frames: 100234 - Time: 1200000ms - Avg: 83.528 - Min: 42 - Max: 135<!--QuoteEnd--></div><!--QuoteEEnd-->
THIS is an in game benchmark. Replace that i7 with an i5, and keep the 480. And guess what, you'll have much less frames. butthurt i5 fanboys weak sauce noobs.
btw nice rig ymicrazy502, your mannly processor makes me respect you.
btw nice rig ymicrazy502, your mannly processor makes me respect you.<!--QuoteEnd--></div><!--QuoteEEnd-->
kinda OT but you do realize that ns2 is probably one of the very few games that benefits heavily from HT right? In most games, the extra o/c you gain from disabling HT gives more performance than keeping it enabled.
Games made over 2 years ago maybe. Or kids games like kawapooty maddern buttfare syclops 3. kids who cant afford hyperthreading with mommy's credit card.
It's OT anyways so whatever, but since you brought it up: I am using an i7 that I bought myself
Interesting point though, I'll try 3.6ghz with HT on and see if NS2 runs better than what I have now (4ghz w/o HT)
It's OT anyways so whatever, but since you brought it up: I am using an i7 that I bought myself
Interesting point though, I'll try 3.6ghz with HT on and see if NS2 runs better than what I have now (4ghz w/o HT)<!--QuoteEnd--></div><!--QuoteEEnd-->
To make it a true comparison you'd have to try without HT @ 3.6 to see the difference in practice. those 9xx intels dont overclock high?
It seems like you don't have much experience overclocking i7s. Guess there's no need to try after all. Enabling HT when you overclock generally results in much higher temp and takes more vcore to maintain a stable clock. It's basically a crapshoot after 4ghz on air without HT enabled and all but impossible with HT unless you are very lucky.
Gigabyte GA-970A-DS3 Rev 1.1 (F6)
Kingston HyperX 1866 DDR3 4GB
Gigabyte 7870 GHZ OC
Menue NO AA/ATMO/AMBIENT/BLOOM/Low infestation/particel all others on@high ~200FPS
Ingame early ~80 fps
midgame ~40-50fps
late (final fight/40-50 minutes) ~20-30 fps