![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Q9550 vs i5 i7 etc etc
- Thread starter vict0r
- Start date
i have both
Core i7 at 4.0 GHz on X58 (6 GB DDR3) and Q9550S at 4.0 GHz (4 GB DDR2) on a Rampage Formula X48 with Vista 64
if you are primarily a gamer - don't bother
houseofbugs
Weaksauce
- Joined
- Apr 5, 2007
- Messages
- 70
Yeah the Q9550 is a beast of a chip. Throw a nice cooling system on it and ramp it up to 3.4-3.6.
I have Q9550, what the best settings to 3.4-3.6?
Brackle
Old Timer
- Joined
- Jun 19, 2003
- Messages
- 8,568
I had a P2 940 o/c to 3.7ghz (which is around the same speed as a Q9550)
And I upgraded to an I7 920.
It is night and day how much faster then I7 920 is stock (let alone overclocked).
I7 > all
And I upgraded to an I7 920.
It is night and day how much faster then I7 920 is stock (let alone overclocked).
I7 > all
i also have a Ph II X4 955 and the i7 is NOT "much faster" in gaming - either at stock or at Max OC.
vsboxerboy
2[H]4U
- Joined
- Oct 17, 2005
- Messages
- 3,661
At stock volts I was able to just bump up the fsb to 400 and run at 3.4 with a fanless scythe ninja. I ended throwing a fan on it a month later, but it was easy as cake - literally took all of 20 seconds.
Brackle
Old Timer
- Joined
- Jun 19, 2003
- Messages
- 8,568
i also have a Ph II X4 955 and the i7 is NOT "much faster" in gaming - either at stock or at Max OC.
Uhhh yes it is
Trying playing at 1920X1200 and above, you will notice a difference.
And I had my p2 @ 3.7ghz.
Uhhh yes it is
Trying playing at 1920X1200 and above, you will notice a difference.
And I had my p2 @ 3.7ghz.
You are dreaming.
Besides comparing Q9550 directly against i7 920, i have also extensively tested Athlon X2 250, Ph II 550 X2, 720 X3 and 955 X4 with GTX 280, HD4870-X2 and 4870 Tri-Fire CF-X3
- that is with each CPU at stock, 3.5 GHz and also each CPU at their max OC (in the 3.8-4.0 GHz range).
i benchmark all my games at 19x12 with maxed out details and 4xAA/16XAF and i have the proof that there is almost no difference between these at high resolutions; sometimes i7 is slower, clock for clock than either Ph II or Penryn. Most games do not even take advantage of more than 2 cores.
Now if you drop the resolution below 14x9 and lower details, the core i7 will pull away from the crowd
Brackle
Old Timer
- Joined
- Jun 19, 2003
- Messages
- 8,568
I am not dreaming. I just sold my P2 940 build which had the best overclocking motherboard with 4gigs of ram bro.
Its night and day. Once you get an I7 920, then you can comment on which is faster. Otherwise you have no idea how fast an I7 920 is compared to an overclocked 955.
Stop being a fanboi bro....seriously....And this is coming from a LONG TIME AMD buyer.
P.S. The reason you dont see a difference is because you are Video card limited
. Get a new video card
Last edited:
I am not dreaming. I just sold my P2 940 build which had the best overclocking motherboard with 4gigs of ram bro.
Its night and day. Once you get an I7 920, then you can comment on which is faster. Otherwise you have no idea how fast an I7 920 is compared to an overclocked 955.
Stop being a fanboi bro....seriously....And this is coming from a LONG TIME AMD buyer.
P.S. The reason you dont see a difference is because you are Video card limited
Ridiculous. How is HD 4870 TriFire - 4870-X2 + 4870 in CF-X3 - limiting anything? Only 5870 CF or Tri-SLi is faster.
i have my i7 920 clocked to 3.8 GHz - effectively 3.97 GHz - and i have TESTED them - extensively - against e8600, Athlon II 250 X2, Ph II 550 X2, 720 X3 and 955 X4
- and i have published the results and my latest article goes up this weekend.
what testing have you done besides "feelings" ?
Don't call me an AMD fanboi
--i AM proudly an Intel fan; my first AMD rig was last year - and *in hi-res gaming* i7 is NOT faster than Ph II nor Penryn - at either 1) Clock for clock - or 2) Overclocked to the Max
Last edited:
FoxhoundOp
[H]ard|Gawd
- Joined
- Apr 26, 2003
- Messages
- 1,129
Why are you asking this question? Are the games you play not performing the way you want them to on your system?
Brackle
Old Timer
- Joined
- Jun 19, 2003
- Messages
- 8,568
Ridiculous. How is HD 4870 TriFire - 4870-X2 + 4870 in CF-X3 - limiting anything? Only 5870 CF or Tri-SLi is faster.
i have my i7 920 clocked to 3.8 GHz - effectively 3.97 GHz - and i have TESTED them - extensively - against e8600, Athlon II 250 X2, Ph II 550 X2, 720 X3 and 955 X4
- and i have published the results and my latest article goes up this weekend.
what testing have you done besides "feelings" ?
Don't call me an AMD fanboi
--i AM proudly an Intel fan; my first AMD rig was last year - and *in hi-res gaming* i7 is NOT faster than Ph II nor Penryn - at either 1) Clock for clock - or 2) Overclocked to the Max
Then your evaluation is wrong. Even hardocp proved it bro.
Link please to i7 beating Ph II or Penryn in high res gaming - if you can
- make sure they are compared clock for clock and also overclocked
- make sure the details are maxed out with 4xAA/16AF in DX10 and fast graphics is used
-- and don't just pick one game
FoxhoundOp
[H]ard|Gawd
- Joined
- Apr 26, 2003
- Messages
- 1,129
And you called him a fanboy?
Quote:
Originally Posted by DASHlT
Then your evaluation is wrong. Even hardocp proved it bro.
Link please to i7 beating Ph II or Penryn in high res gaming - if you can
- make sure they are compared clock for clock and also overclocked
- make sure the details are maxed out with 4xAA/16AF in DX10 and fast graphics is used
-- and don't just pick one game
http://hardforum.com/showthread.php?t=1420299
i7 is more powerful in high res gaming if [H] testing is anything to go by but it's definitely not a night-and-day difference.
Last edited:
Well, we are now comparing CPUs based on changing the setting of the Graphics to see "highest playable settings". So much is depending on the videocard.
What i asked for was Graphics at the same settings - only changing the CPUs
- and you will see much much closer results.
Even with the benches linked to, there is not much difference between i7, Penryn and Ph II - once they all get around 3.5 GHz; that is something that HardOCP also concluded.
The biggest visual difference will always be the graphics card - not the CPU (as long as it is capable)
FoxhoundOp
[H]ard|Gawd
- Joined
- Apr 26, 2003
- Messages
- 1,129
It doesn't really matter if graphics are at the same settings in the [H] test. The logical conclusion is that if the i7 performs better than X CPU at the same clocks despite the fact that it's using higher graphics settings, then that lead will only widen if you test the i7 at the same graphics settings, though not by enough that anyone using a fast quad should be looking to upgrade their processor, in my opinion.
I do agree with you though that the other guy is dreaming. There are very few cases in gaming where upgrading from a Phenom II 940 @ 3.7 is going to show a night-and-day difference. Hyperbole doesn't help anyone.
You're wrong in gaming. Check out any review sites newest Phenom II - 965 roundups. i7s dominate in benchmarks, but in regular high res games, i7/core2q/PhenomII all do very similarly. Anandtech even noted that the 965 was beating the i7's in some games....
Encoding, compressing, other heavily threaded apps, i7 will dominate. But games is PhII's strongest area. At equal or even unequal clocks, fast PhII X4's hang with i7's at real resolutions. If you run them at 640x480 so only the cpu is running the game, then i7 will dominate. But that is not whats in question here..
[H] uses testing methods unique to itself. My only issue with that is that you are now changing individual video settings that do not have the same or equal performance impact. Other tech sites show much the same thing - i7 is perhaps *marginally* faster than Ph II architecture or Penryn in gaming - and that will depend on the game; in some games, i7 is slower. And of course, we are talking hi-res/high detail gaming - not 10x7.
If someone has a dual or a quad in the 3.5 GHz range, the most benefit always comes from upgrading the graphics - not the CPU.
Now i will agree with the other poster that i7 is "night and day" faster than Penryn or Phenom II for video transcoding (where you are not primarily using the GPU), or where HT has a benefit. In other areas besides gaming, Intel got nice performance increases clock for clock over their older architecture.
I still wouldn't say night and day. I mean, any quad core in the 3.5+ range is massive overkill for 90% of what is done on a desktop PC. Gaming is the main and sometimes only thing people do that actually stresses the system and uses all its resources, and thats the thing that PhII does best. One of the reasons I love them so much!
Even on transcoding and stuff, i7 is faster, but night and day? A i7-920 and a PhII X4 965 are both almost same price (amd still a bit cheaper). What would the max performance difference between the 2 be in say encoding, somewhere between 10 and 20%? Hardly night and day.
And compared to an i5 which lacks hyperthreading, which is still the same price or pricier than the 965, the difference is smaller.
So yeah its def true that intels chips are faster in certain applications, some more than others, but nothing crazy, overall much closer to hardly noticable than to night and day
Same goes for core2quads. Overclocked over 3ghz the performance differences between core2q's, i7's, and PhII's is quite small all around. Most notably in games.
And seriously, as for the apps where the i5/i7 really pull ahead like encoding or compressing, how much of your total pc time is really spent doing that stuff? Most people are not transcoding dvd's all day every day. Lets say 60% of your pc use time is gaming (accurate for me somewhat), and <10% (for me like 1%) doing things where i*'s are notably faster, it makes for a very small performance difference in your overall usage. Hardly worth the extra overall platform cost if you care about money.
As I always say, your overall system enjoyment will be better served by dumping extra money into a video card, more/faster storage, bigger monitor or better speakers, ect. Spend the money on something that will notably increase your enjoyment of your box, not a few percent that in the real world will never be noticed.
When I see people running i7 X58 systems with triple channel ddr3 on a little junky monitor it makes me want to cry. You could have put together a "budget" performance box, with say a 955BE and a 790GX board with 4gb cheap ddr3-1600, and put ~200+ dollars you save on those 3 components towards a good 24"+ or better. Newegg has a hanns-g 27" for 300 bucks! You could almost pay for that just with the savings from cpu and mobo!
Youll have nearly identical gaming performance, itll take you a few more minutes to transcode a dvd, os responsiveness and everything else will feel basically identical; but you will now have a huge beautiful monitor instead of a tiny little piece of poop! That hanns-g was just an example, my 24" (benq g2400wd) is IMO one of the best gaming TNs around, and it was only 300 shipped.
I just dont see why you would spend that much extra cash on something that you will rarely ever notice, vs putting that money towards something that will have a notable impact on your PC experience 100% of the time like a better monitor. Or as previously said, dump saved money into a better or second xfired video card, which will always give you a better gaming experience.
Even on transcoding and stuff, i7 is faster, but night and day? A i7-920 and a PhII X4 965 are both almost same price (amd still a bit cheaper). What would the max performance difference between the 2 be in say encoding, somewhere between 10 and 20%? Hardly night and day.
And compared to an i5 which lacks hyperthreading, which is still the same price or pricier than the 965, the difference is smaller.
So yeah its def true that intels chips are faster in certain applications, some more than others, but nothing crazy, overall much closer to hardly noticable than to night and day
Same goes for core2quads. Overclocked over 3ghz the performance differences between core2q's, i7's, and PhII's is quite small all around. Most notably in games.
And seriously, as for the apps where the i5/i7 really pull ahead like encoding or compressing, how much of your total pc time is really spent doing that stuff? Most people are not transcoding dvd's all day every day. Lets say 60% of your pc use time is gaming (accurate for me somewhat), and <10% (for me like 1%) doing things where i*'s are notably faster, it makes for a very small performance difference in your overall usage. Hardly worth the extra overall platform cost if you care about money.
As I always say, your overall system enjoyment will be better served by dumping extra money into a video card, more/faster storage, bigger monitor or better speakers, ect. Spend the money on something that will notably increase your enjoyment of your box, not a few percent that in the real world will never be noticed.
When I see people running i7 X58 systems with triple channel ddr3 on a little junky monitor it makes me want to cry. You could have put together a "budget" performance box, with say a 955BE and a 790GX board with 4gb cheap ddr3-1600, and put ~200+ dollars you save on those 3 components towards a good 24"+ or better. Newegg has a hanns-g 27" for 300 bucks! You could almost pay for that just with the savings from cpu and mobo!
Youll have nearly identical gaming performance, itll take you a few more minutes to transcode a dvd, os responsiveness and everything else will feel basically identical; but you will now have a huge beautiful monitor instead of a tiny little piece of poop! That hanns-g was just an example, my 24" (benq g2400wd) is IMO one of the best gaming TNs around, and it was only 300 shipped.
I just dont see why you would spend that much extra cash on something that you will rarely ever notice, vs putting that money towards something that will have a notable impact on your PC experience 100% of the time like a better monitor. Or as previously said, dump saved money into a better or second xfired video card, which will always give you a better gaming experience.
Last edited:
jeremyshaw
[H]F Junkie
- Joined
- Aug 26, 2009
- Messages
- 12,511
and a whole bunch better, too.
I remember the flak Ironclad drew when they make SoaSE such a huge game, yet it only really used one core. The second core was for 'texture work' which was to say (without modding), not much at all.
But then you have SupCom, WiC, and others (sorry, out of the loop here) that do leverage multicore to it's best.
OK, maybe there are 5 or 6 games where quads really make a performance difference over dual core
.. out of thousands of modern PC games
Can you name more?
EclipseRydr
Limp Gawd
- Joined
- Mar 7, 2009
- Messages
- 146
i mainly run FSX on a stock E8400, with regards to OP question, would i notice a difference if i put in a q9550 or start over with an i7 set up?
Flight Simulator is very CPU intensive. You might notice a difference if you *overclocked* your e8400. Try it, it costs you nothing (if you are even reasonably careful).
What is the rest of your system like (esp. the video card)?
crazyassazian
Limp Gawd
- Joined
- Dec 30, 2008
- Messages
- 250
If you encode, go for LGA 1366 core i7, it's worth it.
20% of an hour is 12 minutes. Food for thought.
20% of an hour is 12 minutes. Food for thought.
EclipseRydr
Limp Gawd
- Joined
- Mar 7, 2009
- Messages
- 146
i have gigabyte EP45-DS3LR
e8400
4gb ram
512 ATI HD4850
250gb hdd
it's not that i don't want to overclock, it's that i'm not 100% how to do it =/, especially given my mobo, read some reviews on newegg and it's not that great for OC =/
TheResidentEvil
[H]ard|Gawd
- Joined
- Jun 12, 2009
- Messages
- 1,943
How much better is the 9550 over a 9450 (which is what I got)?
GotNoRice
[H]F Junkie
- Joined
- Jul 11, 2001
- Messages
- 12,017
The Multiplier.
A Q9650 will be 333Mhz FSB (Quad pumped to 1333Mhz) with a 9x Multiplier for a stock speed of 3.00Ghz
A Q9550 will be 333Mhz FSB (Quad pumped to 1333Mhz) with a 8.5x Multiplier for a stock speed of 2.83Ghz
A Q9450 will be 333Mhz FSB (Quad pumped to 1333Mhz) with a 8x Multiplier for a stock speed of 2.66Ghz
This comes into play because when you are overclocking you are doing so by increasing the FSB. If your multiplier is too low, you are likely to hit a FSB limitation with your board before you hit the limits of your chip. Many boards start to become a bit flaky as you approach and potentially exceed 500Mhz FSB.
It is safe as long as you do not increase the voltage and watch your temps.
All you need to do is raise the FSB in your BIOS.
Go and get detailed instructions on overclocking. Do not bother with "reviews" on NewEgg
ivandurago
[H]ard|Gawd
- Joined
- Jun 5, 2008
- Messages
- 1,307
I think all of [H]s in depth testing concluded that if all chips are clocked to ~3.6Ghz+ that there is a minimal difference in fps between an i7 chip vs their slightly older quad core brethren.
Sorry BRO
I think I might just have to look up that test because you are so obnoxious.
Brent Justice said:
A i7 setup is slightly faster than c2d and phII when they are all clocked at 3.6Ghz+, however it is certainly not worth replacing an entire system for 3-10% performance gains, you are much better off upgrading the video card.
So BRO, you probably should read before you post, and before you upgrade your "P2 940 build which had the best overclocking motherboard with 4gigs of ram."
You will probably have trouble getting the q9450 over ~3.8, whereas the q9550 has been OC'ed to 4.4Ghz+. Also the vcore on the q9550 is generally much lower.
Last edited:
to tell you all the truth..i have an e8400 running at 3.6Ghz on my Asus Rampage X48 with 4 GB ... i got an upgrade bug thats why i asked....i played COD MW2 .... and i noticed that my pc even with a 9600gt with everything at high it worked flawlessly... maybe i shall get a new monitor or tv or videocard
kirbyrj
Fully [H]
- Joined
- Feb 1, 2005
- Messages
- 30,693
For me, I already had a decent board and memory, so I just dropped in a Q9550 that I got here for $170. OC'd it to 3.8Ghz. I know I have a 10-15% performance hit compared to a similarly clocked 920, but it's good enough for the price I paid for it.