Q6600 Vs Q9450 - Is It Worth The Extra $$$'s

Message boards : Number crunching : Q6600 Vs Q9450 - Is It Worth The Extra $$$'s
Message board moderation

To post messages, you must log in.

AuthorMessage
Terror Australis
Volunteer tester

Send message
Joined: 14 Feb 04
Posts: 1817
Credit: 262,693,308
RAC: 44
Australia
Message 735644 - Posted: 7 Apr 2008, 2:55:50 UTC

Thinking of building a new machine and I'm wondering whether to go for a 9450 or just get a 6600 and OC the hell out of it.

Is the 9450 worth the extra $400 and what will it OC to ?

Which would be the better MoBo for SETI use a dfi x48 or a Gigabyte ep-ds3p ?

Regards
Brodo
(Trying to keep up with "David") :-)
ID: 735644 · Report as offensive
Profile JDWhale
Volunteer tester
Avatar

Send message
Joined: 6 Apr 99
Posts: 921
Credit: 21,935,817
RAC: 3
United States
Message 735648 - Posted: 7 Apr 2008, 3:02:21 UTC - in response to Message 735644.  

Thinking of building a new machine and I'm wondering whether to go for a 9450 or just get a 6600 and OC the hell out of it.

Is the 9450 worth the extra $400 and what will it OC to ?

Which would be the better MoBo for SETI use a dfi x48 or a Gigabyte ep-ds3p ?

Regards
Brodo
(Trying to keep up with "David") :-)



Don't know your prices down under, but here I can get the Q6600 for $200 (US)
can't beat it, IMO.
ID: 735648 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19048
Credit: 40,757,560
RAC: 67
United Kingdom
Message 735667 - Posted: 7 Apr 2008, 4:50:52 UTC

That seems to be a big difference in price, here in UK Q6600 is retail £140 ($US 280) which looks expensive , but Q9450 is £235 ($US470) only £95 ($190) more.

What about the Q9300, £195 retail £175 OEM, as a compromise.

For the Yorkfields I find mobo and RAM to match 1333 MHz bus more of a problem, when looking for cheap.
ID: 735667 · Report as offensive
Profile -= Vyper =-
Volunteer tester
Avatar

Send message
Joined: 5 Sep 99
Posts: 1652
Credit: 1,065,191,981
RAC: 2,537
Sweden
Message 735932 - Posted: 7 Apr 2008, 21:51:37 UTC

The Q9450 doesn't clock that well unfortunately, i had trouble with it when switching my OC'ed Q6600 to a Q9450 and thought this would work like charm but it didn't.

Seems like that the Q9450 doesn't like too high FSB frequencies for what i can tell and it only got 8x multiplyer.

But if you OC modest it's a true go for that one but don't expect much faster than 3500 Mhz on a Q9450 on aircooling atm.

Kind Regards Vyper

_________________________________________________________________________
Addicted to SETI crunching!
Founder of GPU Users Group
ID: 735932 · Report as offensive
Terror Australis
Volunteer tester

Send message
Joined: 14 Feb 04
Posts: 1817
Credit: 262,693,308
RAC: 44
Australia
Message 736004 - Posted: 8 Apr 2008, 1:10:00 UTC
Last modified: 8 Apr 2008, 1:18:04 UTC

Thanks Vyper
Thats the sort of info I was looking for. At current prices here in Oz I can buy 2 Q6600/MoBo combo's for only about 10% more than a single Q9450/MoBo setup. This would definitely give me more "bang per buck" I reckon.

(And if I put them in recycled cases the "Minister for War & Finance" wouldn't know. Hmmm, will have to think about this :-)

Much Respect
Brodo
ID: 736004 · Report as offensive
Profile Voyager
Volunteer tester
Avatar

Send message
Joined: 2 Nov 99
Posts: 602
Credit: 3,264,813
RAC: 0
United States
Message 736071 - Posted: 8 Apr 2008, 4:03:55 UTC - in response to Message 736004.  

(And if I put them in recycled cases the "Minister for War & Finance" wouldn't know. Hmmm, will have to think about this :-)

Make love not war! Hide your money.;-]
ID: 736071 · Report as offensive
Profile -= Vyper =-
Volunteer tester
Avatar

Send message
Joined: 5 Sep 99
Posts: 1652
Credit: 1,065,191,981
RAC: 2,537
Sweden
Message 736089 - Posted: 8 Apr 2008, 6:03:30 UTC

You're welcome.

But if i see this in my perspective its a win for me.

The improved SSE engine and cache amount helps me more than having my old Q6600.

I can run my Q9450 @ 3450 Mhz flawless and my old Q6600 i ran at 3550 Mhz flawlessly, after that the heat/voltage ratio couldn't be justified the small amount of extra performance with std. aircooling.

A quick click on the calculator shows that the old Q6600 had 2.9% extra cpu power but if u add to the equation the improved cache and sse engine the Q9450 is a win. In my case that's a solid choice after all to the Q9450 plus improved wattage per s@h hour :-)

I think a Yorkfield vs. Kentfield is about 7-9% faster in average clock for clock spread across various applications and the boost mainly comes in games and for that matter i think s@h too!

Kind Regards Vyper

_________________________________________________________________________
Addicted to SETI crunching!
Founder of GPU Users Group
ID: 736089 · Report as offensive
Profile Daniel
Volunteer tester
Avatar

Send message
Joined: 21 May 07
Posts: 562
Credit: 437,494
RAC: 0
United States
Message 736112 - Posted: 8 Apr 2008, 8:28:11 UTC

What mobo and ram are you running Vyper?
Daniel

ID: 736112 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 736142 - Posted: 8 Apr 2008, 11:05:07 UTC - in response to Message 736112.  

What mobo and ram are you running Vyper?


Have a Q6600 on a ASUS P5E, running @ 3,2GHz, FSB 1385MHz.
Also a QX9650 on the same board, only 8 in stead off 4 GiG (800MHz)and run WIN XP64, on the latter.

Busy figuring out the Voltage settings on this one, cause it crashes if i go above 3,4GHz. I know it can be run @ about 4GHz. Watch the temps though.

But the QX9650 IS EXPENSIVE, about 500 Euro's !!!
With it's 12MByte L2 cache, it's fast for crunching anyway.

ID: 736142 · Report as offensive
Profile -= Vyper =-
Volunteer tester
Avatar

Send message
Joined: 5 Sep 99
Posts: 1652
Credit: 1,065,191,981
RAC: 2,537
Sweden
Message 736172 - Posted: 8 Apr 2008, 13:18:04 UTC - in response to Message 736112.  

What mobo and ram are you running Vyper?


I run on a Asus P5E.. Nice board indeed from my point of view.


_________________________________________________________________________
Addicted to SETI crunching!
Founder of GPU Users Group
ID: 736172 · Report as offensive
Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar

Send message
Joined: 20 Dec 05
Posts: 3187
Credit: 57,163,290
RAC: 0
United States
Message 736182 - Posted: 8 Apr 2008, 13:42:33 UTC
Last modified: 8 Apr 2008, 13:44:37 UTC

I concur with the suggestion of the q9300 - it's what I'm going to get for a Gigabyte GA-p35-s3g, with 4 Gb of DDR2 1066 onboard. FWIW, IMO the Q9300 has the better price/performance ratio, even though it costs about $50 us more than the q6600.

System is gradually building, as funds become available (have mb, memory, case, p/s, hd and video card, need CPU, CPU fan, OS, DVD, KB, and mouse - will eventually have a KVM switch for use with my other systems.)
.

Hello, from Albany, CA!...
ID: 736182 · Report as offensive
Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar

Send message
Joined: 20 Dec 05
Posts: 3187
Credit: 57,163,290
RAC: 0
United States
Message 746303 - Posted: 30 Apr 2008, 23:16:18 UTC
Last modified: 30 Apr 2008, 23:17:56 UTC

Further thoughts...

Umm, the most direct comparison for a Q6600 is the Q9300 - the direct comparison for the Q9450 is the Q6700...

Q9300 runs at 2.5 Ghz, Q6600 at 2.4Ghz, Q6700 and Q9450 at 2.66 Ghz - lets compare apples to apples, after all!
.

Hello, from Albany, CA!...
ID: 746303 · Report as offensive
Keith White
Avatar

Send message
Joined: 29 May 99
Posts: 392
Credit: 13,035,233
RAC: 22
United States
Message 746610 - Posted: 1 May 2008, 19:52:11 UTC

Now I know I'm a little late to this party but...

If you can use Folding@Home as a proxy for Seti then this chart in the TechReport's review of the Q9300 (scroll all the way to the bottom) shows a 25+% improvement in estimated credits per day over a Q6600.
"Life is just nature's way of keeping meat fresh." - The Doctor
ID: 746610 · Report as offensive
Profile Mad Max
Volunteer tester
Avatar

Send message
Joined: 16 Mar 00
Posts: 475
Credit: 213,231,775
RAC: 407
United States
Message 746675 - Posted: 1 May 2008, 22:44:58 UTC

Here is my 2 cents. To give an idea, I just put together a P5E with a q9450 and Patriot DDR2-1200. Wanted this one on the somewhat quiet/cheap side so after searching, I purchased a ZEROtherm 92mm cooler because of size restrictions, and a passive EVGA PCIe card (Total for the above: $782). Reused my Lian Li case, two Sata drives and P/S, replaced two stk 80mm case fans with Vantec SF8025L, so the thing makes practically no noise at all. Happy the thing is not a wind tunnel, and guessing at the rest of the prices I figure that there is a total of $1100. Currently running at a safe 3.0 with Alex Kan’s Optimized apps. Could have it running faster with better stuff but that wasn’t the idea. Times are at the 90 minute or less mark X 4! Feel free to check out my list to see what the averages are, my computers are not hidden. Just starting to work on details for “Seti Monster Experiment” after seeing what this thing can do. More to come…
IAS - Where Space Is Golden!
ID: 746675 · Report as offensive
Ingleside
Volunteer developer

Send message
Joined: 4 Feb 03
Posts: 1546
Credit: 15,832,022
RAC: 13
Norway
Message 746874 - Posted: 2 May 2008, 7:47:20 UTC - in response to Message 746610.  

Now I know I'm a little late to this party but...

If you can use Folding@Home as a proxy for Seti then this chart in the TechReport's review of the Q9300 (scroll all the way to the bottom) shows a 25+% improvement in estimated credits per day over a Q6600.

For anyone not familiar with Folding@home, Tinker and Amber is non-SSE, Gromac is SSE and Gromac-33 is most likely SSE2. Since Tinker and Amber is non-SSE, using their speed-up of 43% and 38% would not be a good indication of the performance of SSE-optimized SETI@home-applications, but of course for any non-SSE-optimized BOINC-projects these 2 would be a good indication.

For SETI@home, would guess more relevant is the 16% speed-up on SSE, and the 11% speed-up on AFAIK SSE2...

Hmm,maybe SSE4-optimized applications gives a significant extra speed-up compared to SSSE3 on the old cpu's?


BTW, it's a little strange that Tinker is still included in benchmarks, since FAH haven't had any Tinker-wu's for a couple years...


"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."
ID: 746874 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19048
Credit: 40,757,560
RAC: 67
United Kingdom
Message 746884 - Posted: 2 May 2008, 8:29:16 UTC

My observations with Q9450 at stock speed 2.66Ghz, and Q6600 OC'd to 2.88Ghz is that the Q9450 is about 4% quicker. That was before the latest optimised apps were introduced.
ID: 746884 · Report as offensive
Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar

Send message
Joined: 20 Dec 05
Posts: 3187
Credit: 57,163,290
RAC: 0
United States
Message 746960 - Posted: 2 May 2008, 14:14:26 UTC - in response to Message 746884.  

My observations with Q9450 at stock speed 2.66Ghz, and Q6600 OC'd to 2.88Ghz is that the Q9450 is about 4% quicker. That was before the latest optimised apps were introduced.



Again, let's compare apples to apples -as close as we can, at least... how does a Q9300 compare to a Q6600? - I'll even let the 6600 be OC'd to 2.5 Ghz for a more fair comparison

a Q9450 should be doing at least 2% better than a 6600, just because of the difference in clock speed, if both are left at stock speed (again 2.4 vs. 2.6 Ghz clock speed!)
.

Hello, from Albany, CA!...
ID: 746960 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19048
Credit: 40,757,560
RAC: 67
United Kingdom
Message 746971 - Posted: 2 May 2008, 14:56:28 UTC - in response to Message 746960.  
Last modified: 2 May 2008, 14:58:48 UTC

My observations with Q9450 at stock speed 2.66Ghz, and Q6600 OC'd to 2.88Ghz is that the Q9450 is about 4% quicker. That was before the latest optimised apps were introduced.



Again, let's compare apples to apples -as close as we can, at least... how does a Q9300 compare to a Q6600? - I'll even let the 6600 be OC'd to 2.5 Ghz for a more fair comparison

a Q9450 should be doing at least 2% better than a 6600, just because of the difference in clock speed, if both are left at stock speed (again 2.4 vs. 2.6 Ghz clock speed!)

But it is running 4% faster that overclocked Q6600 @2.88 when Q9450 is at stock 2.66GHz, on chicken soup.

Son has got the Q9450 to 3.18GHz stable running UT3, not crunching at moment.
ID: 746971 · Report as offensive
Profile John Clark
Volunteer tester
Avatar

Send message
Joined: 29 Sep 99
Posts: 16515
Credit: 4,418,829
RAC: 0
United Kingdom
Message 746986 - Posted: 2 May 2008, 15:35:10 UTC
Last modified: 2 May 2008, 15:35:31 UTC

I believe the Q9450 can be overclocked on air to 3.95GHz, as tis host suggests
It's good to be back amongst friends and colleagues



ID: 746986 · Report as offensive

Message boards : Number crunching : Q6600 Vs Q9450 - Is It Worth The Extra $$$'s


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.