Question if you upgraded to a quad core for Crysis

[Indus]

How much of an increase did you see in minimum framerate in laggy areas from a dual core after upgrading to a quad core?

Reason is I think I'm severely CPU Limited. I get 25 FPS or lower in 1680x1050 and if I drop it down to 800 x 600, its the same exact framerate although the CPU meter on my g15 says 92% or so.

So I'm curious how have your experiences been after upgrading?

As a sidenote if you play aion too which is based on the same engine, how much difference did you have in dual core vs quad core minimum framerate?

I'm not looking to get framerate above 250 (which I might anyways) but I want to get minimum's above 30 so it doesn't feel like a slideshow in heavily populated areas.

 

[MrK6]

I haven't played Aion, but for Crysis, it's not so much about the cores as it is about frequency. Despite all the PR and marketing spread about the game, it's still largely dual core optimized. Running your 6550 under 3.0GHz is limiting your system, as you discerned. The P35-DS3L is a a good overclocking board, try squeezing some more out of that CPU. You could also crank the clocks on your GTX260 as well, that always helps.

 

[error8]

Crysis it's not quad core optimized. It only eats two cores and needs as much as GPU power you can give it. So going to a quad will probably give you nothing in plus for performance, at least nothing that you can really feel. As MrK6 said, squeeze as much as you can from your E6550 and from your videocard, that will improve your frame rate a bit, hopefully.

 

[bigriggg]

i notice a difference when switching from a dual core at 2.4ghz to a quad at 3.8ghz(q9550), people are saying crysis is only a dual core dependent program, but all four of my cores are at about 75-85 percent usage when playing cysis. I think at the time also, my cpu was a bottleneck (amd 4400+), so it wasnt letting my gpu perform to its max (8800gt), now the gpu is the bottleneck but NOW i got an even better gpu (5850) so my comp rips up any game I throw at it. SO..... I see you have a gtx 260, which probaly isnt getting it's full potential out because your cpu is holding back some of its power, so in theory, if you got a quad core, then YES, you would see a difference because it's unlocking your gpu's potential. It seems like a never ending battle, there's always something weaker in your system that limits everything overall.

 

[Ben90]

i notice a difference when switching from a dual core at 2.4ghz to a quad at 3.8ghz(q9550), people are saying crysis is only a dual core dependent program, but all four of my cores are at about 75-85 percent usage when playing cysis. I think at the time also, my cpu was a bottleneck (amd 4400+), so it wasnt letting my gpu perform to its max (8800gt), now the gpu is the bottleneck but NOW i got an even better gpu (5850) so my comp rips up any game I throw at it. SO..... I see you have a gtx 260, which probaly isnt getting it's full potential out because your cpu is holding back some of its power, so in theory, if you got a quad core, then YES, you would see a difference because it's unlocking your gpu's potential. It seems like a never ending battle, there's always something weaker in your system that limits everything overall.

http://www.anandtech.com/bench/default.aspx?b=50

The reason why you saw an increase is because the A64 architecture gets wiped by the core 2s. The absolute fastest A64 processor of all time (windsor, 3.2ghz) lost most battles with the e6600, which only ran at 2.4ghz. ~35% IPC improvement. Now lets run that same e6600 @ 3.8, and your talking over 2x the performance.

And on top of this, add the massive l2$ that the q9550 has (crysis gets a boner for cache) and its no wonder that you saw an increase. If you went from something like an e6600 to a q6600, you might get ~10% fps improvement, and i bet most of that is from having more cache.

 

[error8]

but all four of my cores are at about 75-85 percent usage when playing cysis.

That I don't believe. Crysis doesn't know what to do with the third and forth core and it's severely limited by the GPU.

 

[Indus]

http://www.anandtech.com/bench/default.aspx?b=50

The reason why you saw an increase is because the A64 architecture gets wiped by the core 2s. The absolute fastest A64 processor of all time (windsor, 3.2ghz) lost most battles with the e6600, which only ran at 2.4ghz. ~35% IPC improvement. Now lets run that same e6600 @ 3.8, and your talking over 2x the performance.

And on top of this, add the massive l2$ that the q9550 has (crysis gets a boner for cache) and its no wonder that you saw an increase. If you went from something like an e6600 to a q6600, you might get ~10% fps improvement, and i bet most of that is from having more cache.

Ok lets put this another way.. how much difference in MINIMUM framerate would I see with change from E6550 @ 2.93 vs say Q9550 @ 3.4ghz

 

[bigriggg]

That I don't believe.

ok, but who said you had to believe it? do you have a q9550? no

 

[bigriggg]

Ok lets put this another way.. how much difference in MINIMUM framerate would I see with change from E6550 @ 2.93 vs say Q9550 @ 3.4ghz

well im not sure, but when i put everything on max with the amd 4400+ it was unplayable, with everything cranked on the 8800gt is was somewhat playable but still not too much, untill i got my 5850... but like i said....by unlocking your gpu's potential, only you would know, unless you can find a benchmark the same or very simliar to your upgrade you're talking about.

 

[Ben90]

Ok lets put this another way.. how much difference in MINIMUM framerate would I see with change from E6550 @ 2.93 vs say Q9550 @ 3.4ghz

oops. I got mixed up, i thought u were coming from a A64 4400, cuz someone else posted that, so im all confused now...


Honestly i dont think you will see that much of an improvement, but i dont have that much to do right now so im going to run my i7 @ 2.6 (dual core) and 3.2 (quad core) and see if my overall experience changes... be back in prolly an hour (if it does, ill try running dual @ 3.2 or 4)

If you have another configuration of anything you want me to test im up for it for a little while maybe

 

[error8]

ok, but who said you had to believe it? do you have a q9550? no

Can you read my sig? I have an overclocked quad core and if you can believe it or not, is not that far from yours.
I have played Crysis and I know that it doesn't even comes close to that load you are talking about. If you can prove that then I'll stand corrected, otherwise you are just talking nonsense.

 

[BTRY B 529th FA BN]

The free alternative is to overclock your cpu and find out for yourself

 

[Indus]

oops. I got mixed up, i thought u were coming from a A64 4400, cuz someone else posted that, so im all confused now...


Honestly i dont think you will see that much of an improvement, but i dont have that much to do right now so im going to run my i7 @ 2.6 (dual core) and 3.2 (quad core) and see if my overall experience changes... be back in prolly an hour (if it does, ill try running dual @ 3.2 or 4)

If you have another configuration of anything you want me to test im up for it for a little while maybe

Great I'm quite curious about your results.

Because this is my problem:

Normal Situation: http://img690.imageshack.us/img690/9149/aion0119a.jpg

Raid Situation: http://img527.imageshack.us/img527/4149/aion0149a.jpg

As you can see it becomes a slide show and the same thing happens with Crysis. They're both built on the same engine so 1 fix will work for both.

I upgraded to a GTX 260 a month ago and that helped a lot with stuttering I was experiencing earlier but still I get the minimum framerate very low. Now this happens even if my resolution is 800 x 600 or 1680 x 1050 so its not a GPU bottleneck.

 

[Indus]

The free alternative is to overclock your cpu and find out for yourself

If you had the time to read my rig link, you'll see it already is overclocked and is now limited by the RAM.

 

[BTRY B 529th FA BN]

If you had the time to read my rig link, you'll see it already is overclocked and is now limited by the RAM.

spend the time and drop your CPU clock back to stock and see what kind of difference you get

 

[Indus]

spend the time and drop your CPU clock back to stock and see what kind of difference you get

Not a bad idea. Will do when I'm home later tonight and post results.

 

[bigriggg]

Can you read my sig? I have an overclocked quad core and if you can believe it or not, is not that far from yours.
I have played Crysis and I know that it doesn't even comes close to that load you are talking about. If you can prove that then I'll stand corrected, otherwise you are just talking nonsense.

ok, well when i get home from my school session, then i will check...and if i'm wrong, i will also stand corrected. sometimes i am wrong though....just to make sure are you talking peak being the stock frequencies or my 3.8 oc?

 

[BTRY B 529th FA BN]

Not a bad idea. Will do when I'm home later tonight and post results.

It's a cost alternative idea and may prove helpful in your future upgrade decision

I did read a while back that an SSD drive will also help raise your minimum fps in Crysis

 

[error8]

ok, well when i get home from my school session, then i will check...and if i'm wrong, i will also stand corrected. sometimes i am wrong though....just to make sure are you talking peak being the stock frequencies or my 3.8 oc?

Sometimes I'm wrong too. It had happened before and it can also happen now Frequency is unimportant, if it uses all 4 cores at 90% at stock clocks, it will use them at 90% at 3.8 ghz. It's the same thing.
I'll probably test this in a couple of hours, but I have to install the darn Crysis again.

 

[Ben90]

It's a cost alternative idea and may prove helpful in your future upgrade decision

I did read a while back that an SSD drive will also help raise your minimum fps in Crysis

IIRC that was mostly in the flyby benchmarks when the game needed to load a metric crapload of textures, but in real world gameplay a SSD doesn't help much (i may be wrong though)

And i went from 2.6 dual core i7 to 3.8 quad i7.... honestly I don't think the framerate changed at all. I definitely didn't use the most scientific methods, but i can guarantee the experience will not change that much (vs in l4d2 having multicore doubles your framerate)

 

[error8]

This is what I get with Crysis, on the first level on the beach, everything put on Very High quality, 2X AA. The maximum I've seen was 30% cpu usage.

 

[bigriggg]

This is what I get with Crysis, on the first level on the beach, everything put on Very High quality, 2X AA. The maximum I've seen was 30% cpu usage.

so does this mean a dual core would be at about double because it isnt spreading the usage between the cores? so it'd be 60% max on a dual?

 

[error8]

so does this mean a dual core would be at about double because it isnt spreading the usage between the cores? so it'd be 60% max on a dual?

Yes, probably around that.

 

[error8]

And this is how a quad core optimized game loads the cpu ( Dragon Age ). In only seconds after loading the saved game, the usage spiked to 80% of all four cores.

 

[bigriggg]

That I don't believe. Crysis doesn't know what to do with the third and forth core and it's severely limited by the GPU.

http://profile.imageshack.us/user/bigriggg/images/detail/#692/crysis1.png