Discussion in 'Overclocking & Cooling' started by Q9650, Aug 5, 2011.
Master Q9650, not victory. The shroud of the 2500k has fallen. Begun, the CPU War has!
He his still in denial that his more expensive P4 got beat by a cheaper Athlon back in the day lol
Yea i agree, he is dreamin.
I think it all goes back to how deep you want to stay invested in that era of technology. $200 can be a lot of money when you consider the amount you already have invested in the aging system. Not that it isnt capable but how much longer will it suffice you for another $200. If you figure you can live with it another two yrs thats a $100 a year and will you be happy with what you have during that time? Believe me I am with you on the cost of change. Add the cost of water blocks for the VGA to mine. But I am ready to do this I like the new features coming and wont be trying to get rid of 2 old chips for 1/8 of what I paid for them. Ill just have one to get rid of. Just wish we had some concrete on BD before I cant wait and get an i5.
On a totally off topic note - I went ahead and ordered my 6950 this morning! Boy I hope I get extra shaders in this crapshoot.
Just FYI, I was waiting on baited breath over a QX9770 on eBay that sold last night for $305. Still quite expensive, but damn near the best 775 chip you can buy.
Just pricing fyi.
Does staying in a platform because I am poor count? I have been using LGA775 for 5 years now. I knew the jig was up when the processor limited dead space with a GTX 380. Have only been playing new games on the xbox since then
Considering I use Phenom II X4s (which everyone s**ts on for sucking compared to Intel) and don't run into any CPU usage problems with any game I own with an AMD HD 5770, your 9650 should do you well for quite a while longer. Team Fortress 2 (which I think is probably the biggest CPU rapist in my game library) stays at 60 FPS for me (VSYNC) on an X4 @ 3.6 (and even my original Phenom X4 does well at 2.75). Your Q9650 would probably get higher min/max framerates with the same video card, but both CPUs are still decent nonetheless.
Well, a Q9650 is a lot faster than a Pentium D, for sure. It's similar to how the socket 939 Athlon X2 processors were a lot better than their single core counterparts, but 775 saw a ton of performance increases.
LGA 775 was a socket with quite a long life, and the last CPUs made for it are vastly superior to the CPUs that came out with the socket. Now, Intel seems to like to change sockets quite quickly, which isn't bad for 1366 which had a 3 year life (gets replaced by 2011 at the end of the year) but 1156 got the shaft with just a year long lifespan.
Wait, what? A GTX 380 doesn't exist, d'you mean a 480?
I do not want to get into any war here . But with my set up I can run crysis just fine one the highest settings they have . Both of them . As long as I can do this I find that this CPU is still way bad ass .
Thats because there are a lot of assumptions in this thread that aren't based in reality. There are a lot of claims that people cant back up with a single link. Here is one that shows how a Q9650 performs against Intel's last gen cpus, never mind a sandy. Keep in mind when looking at those benchmarks that i5 750 runs at 2.6ghz and will overclock to around where a Q9650 will.
Q9650 was a great cpu for its time but you would have to be delusional to think that it performs anywhere near where Intel's recent cpus perform.
Its not a viable option at all when a used i5 760, P55 board, and new 8GB DDR3 kit can be had for not much more and will wipe the floor with any Yorkfield.
umm that review has 2 benches one of which the phenom II 965 beats the q9650 by a fair amount, the other where the q9650 just edges out the phenom II 965.
if you use say the 1090 instead, you see that amd is able to beat 775 now.
it's 1156, 1366, and 1155 they can't seem to come close to.
Also that review is using DDR2 on the Q9650 vs DDR3 on the i5. There are many Socket 775 mobo's capable of running DDR3 and if the Q9650 was on a DDR3 mobo (x48) it would be beating that i5.
Even if socket 775 is EOL, the Q9650 is still a great performer and more than capable of running the latest modern games. My HTPC is a 4.0Ghz q9650, 6gb ddr2, and a gtx285. It playes Dirt 3, Batman, and Grid at max details 1920x1080.
Your HTPC is Bad Ass
Look, when you make claims like that you really should provide some benchmarks to back that up. The difference between DDR2 and DDR3 on X48 is minimal. Check some comparisons.
The reason that Nehalem has so much more memory bandwith is that the memory controller is built into the cpu, not what memory it uses. Secondly most desktop apps, especially gaming, aren't very memory intensive.
Its a different architecture. There are certain apps where a 965 will edge out something like an i7 860.
Yes, minimal... Like the difference between the i5 and the q9650 in that chart.
Do I really need to provide benchmarks to you or does common sense dictate that with DDR3 the gain, even though minimal, would be enough to place it above the i5, which just barely beat it. :shadedshu
Check out the 3dMark Vantage thread on TPU = Plenty of proof there. Many Q9650 and QX9770's beating the i5 750.
Which honestly answers the question of this entire thread. Is the Q9650 still a good and worth while chip. Yes!!
I love my Q9550 just wish I could hit 4GHz.
super PI run anyone with a AMD cpu?
Which chart? The one with i5 750 pumping a 20% higher framerate in Anno? The one with the 750 pumping a 35% higher framerate in GTA4?
If thats minimal, I'm not sure what to say. I've yet to see a memory comparison on 775, 1156, 1155, or 1366 where memory speed made more than a couple of an fps difference with gaming.
Yes, yes, you do and until you can dig up some evidence to support you claims outside of a forum dedicated to a synthetic benchmark you have no credibility. I'm talking real world testing.
Here is a good start
Why do you think that boards like the UD3P and P5Q Deluxe were so popular? I remember a time where the Tpower i45 ruled the super pi charts at HWbot.
Which real world application is super pi supposed to represent?
I would say no, you would be better off selling what you have and buying new stuff at $200. Maybe if you could get something used for $150 or less it might be worth it.
Also I think the synthetics are good enough. If anything, in games, it will be even less of a gap. Besides the thread isn't about your i5 750, it was asking if the q9650 is good enough. Which it is.
You can nit pick all you want about benchmarks and 18% better...blah blah. Fact is the Q9650 is still a great chip.
Above is a bench with a q9650 w/ddr3 and a i5 750.
Uh, how about crunching and doing mathimatical equations. You know, things computers were created for in the first place(contrary to popular belief, computers were not originally created to play games on). Tell you what, forget the superPI run. Show me a AMD quad-core cpu that has equaled my crunching score with a equal amount of run time(yes, a AMD 6 core can beat a Q9650 in applications that will use all six cores). I've built a PII 965 system with 4 gigs of the same speed RAM as I have in my Q9650 system. It's not as fast as my Q9650 in everyday use from what I saw.
I got over 10% on benchmarks when I changed to DDR3, so DDR3 does make a big difference, of course it would depend on the board/ram in particular. My ram is running a reasonably good speed/latency. I do agree that a Q9650 does not have a chance at the newer processors..... the bigger question is what the OP asked. Is a Q9650 a 4ghz still good, and to that I say yes.
I take it that you know very little about programing and how that can effect those computations. There are certain real world apps where a 965 can go toe to toe with Nehalem. Things aren't always so black and white.
This is one of the reasons why I don't care about synthetic benchmarks. Did you know that even my i5 760 threw out about the same amount of Gflops in Linpack and my Q9650? Did that translate to what I saw in real world usage, no.
I agree with you but is it worth $200? No, there is even someone with a Q9650 feeding a GTX580 in this thread. The same money spent could have bought him a GTX570 and 2500k setup and given better performance.
Yeah, but benchmarks don't always translate to what you see in real world applications. It usually doesn't.
I never said a thing about a Q9650 being able to beat a I5 or I7 cpu. I did say the Q9650 can still get the job done.
You do realise for me personally, I would cost me $400 to upgrade to a 2500k system. If I was still using say a q6600 or e8*00, it would be alot easier to afford a used $200 q9650 than it would be to afford a full system upgrade.
No, you said this.
Separate names with a comma.